]> jfr.im git - yt-dlp.git/blame - yt_dlp/YoutubeDL.py
[extractor/pornez] Handle relative URLs in iframe (#6171)
[yt-dlp.git] / yt_dlp / YoutubeDL.py
CommitLineData
26e63931 1import collections
31bd3925 2import contextlib
9d2ecdbc 3import datetime
c1c9a79c 4import errno
31bd3925 5import fileinput
b5ae35ee 6import functools
8222d8de 7import io
b82f815f 8import itertools
8694c600 9import json
62fec3b2 10import locale
083c9df9 11import operator
8222d8de 12import os
f8271158 13import random
8222d8de
JMF
14import re
15import shutil
dca08720 16import subprocess
8222d8de 17import sys
21cd8fae 18import tempfile
8222d8de 19import time
67134eab 20import tokenize
8222d8de 21import traceback
524e2e4f 22import unicodedata
f9934b96 23import urllib.request
961ea474
S
24from string import ascii_letters
25
f8271158 26from .cache import Cache
14f25df2 27from .compat import compat_os_name, compat_shlex_quote
982ee69a 28from .cookies import load_cookies
f8271158 29from .downloader import FFmpegFD, get_suitable_downloader, shorten_protocol_name
30from .downloader.rtmp import rtmpdump_version
f8271158 31from .extractor import gen_extractor_classes, get_info_extractor
fe7866d0 32from .extractor.common import UnsupportedURLIE
f8271158 33from .extractor.openload import PhantomJSwrapper
34from .minicurses import format_text
8e40b9d1 35from .plugins import directories as plugin_directories
e756f45b 36from .postprocessor import _PLUGIN_CLASSES as plugin_pps
f8271158 37from .postprocessor import (
38 EmbedThumbnailPP,
39 FFmpegFixupDuplicateMoovPP,
40 FFmpegFixupDurationPP,
41 FFmpegFixupM3u8PP,
42 FFmpegFixupM4aPP,
43 FFmpegFixupStretchedPP,
44 FFmpegFixupTimestampPP,
45 FFmpegMergerPP,
46 FFmpegPostProcessor,
ca9def71 47 FFmpegVideoConvertorPP,
f8271158 48 MoveFilesAfterDownloadPP,
49 get_postprocessor,
50)
ca9def71 51from .postprocessor.ffmpeg import resolve_mapping as resolve_recode_mapping
b5e7a2e6 52from .update import REPOSITORY, current_git_head, detect_variant
8c25f81b 53from .utils import (
f8271158 54 DEFAULT_OUTTMPL,
7b2c3f47 55 IDENTITY,
f8271158 56 LINK_TEMPLATES,
8dc59305 57 MEDIA_EXTENSIONS,
f8271158 58 NO_DEFAULT,
1d485a1a 59 NUMBER_RE,
f8271158 60 OUTTMPL_TYPES,
61 POSTPROCESS_WHEN,
62 STR_FORMAT_RE_TMPL,
63 STR_FORMAT_TYPES,
64 ContentTooShortError,
65 DateRange,
66 DownloadCancelled,
67 DownloadError,
68 EntryNotInPlaylist,
69 ExistingVideoReached,
70 ExtractorError,
784320c9 71 FormatSorter,
f8271158 72 GeoRestrictedError,
73 HEADRequest,
f8271158 74 ISO3166Utils,
75 LazyList,
76 MaxDownloadsReached,
19a03940 77 Namespace,
f8271158 78 PagedList,
79 PerRequestProxyHandler,
7e88d7d7 80 PlaylistEntries,
f8271158 81 Popen,
82 PostProcessingError,
83 ReExtractInfo,
84 RejectedVideoReached,
85 SameFileError,
86 UnavailableVideoError,
693f0600 87 UserNotLive,
f8271158 88 YoutubeDLCookieProcessor,
89 YoutubeDLHandler,
90 YoutubeDLRedirectHandler,
eedb7ba5
S
91 age_restricted,
92 args_to_str,
cb794ee0 93 bug_reports_message,
ce02ed60 94 date_from_str,
da4db748 95 deprecation_warning,
ce02ed60 96 determine_ext,
b5559424 97 determine_protocol,
c0384f22 98 encode_compat_str,
ce02ed60 99 encodeFilename,
a06916d9 100 error_to_compat_str,
47cdc68e 101 escapeHTML,
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,
fc61aff4 109 get_compatible_ext,
0bb322b9 110 get_domain,
c9969434 111 int_or_none,
732044af 112 iri_to_uri,
941e881e 113 is_path_like,
34921b43 114 join_nonempty,
ce02ed60 115 locked_file,
0647d925 116 make_archive_id,
0202b52a 117 make_dir,
dca08720 118 make_HTTPS_handler,
8b7539d2 119 merge_headers,
3158150c 120 network_exceptions,
ec11a9f4 121 number_of_digits,
cd6fc19e 122 orderedSet,
5314b521 123 orderedSet_from_options,
083c9df9 124 parse_filesize,
ce02ed60 125 preferredencoding,
eedb7ba5 126 prepend_extension,
51fb4995 127 register_socks_protocols,
3efb96a6 128 remove_terminal_sequences,
cfb56d1a 129 render_table,
eedb7ba5 130 replace_extension,
ce02ed60 131 sanitize_filename,
1bb5c511 132 sanitize_path,
dcf77cf1 133 sanitize_url,
67dda517 134 sanitized_Request,
e5660ee6 135 std_headers,
1211bb6d 136 str_or_none,
e29663c6 137 strftime_or_none,
ce02ed60 138 subtitles_filename,
819e0531 139 supports_terminal_sequences,
b1f94422 140 system_identifier,
f2ebc5c7 141 timetuple_from_msec,
732044af 142 to_high_limit_path,
324ad820 143 traverse_obj,
fc61aff4 144 try_call,
6033d980 145 try_get,
29eb5174 146 url_basename,
7d1eb38a 147 variadic,
58b1f00d 148 version_tuple,
53973b4d 149 windows_enable_vt_mode,
ce02ed60
PH
150 write_json_file,
151 write_string,
4f026faf 152)
70b23409 153from .version import RELEASE_GIT_HEAD, VARIANT, __version__
8222d8de 154
e9c0cdd3
YCH
155if compat_os_name == 'nt':
156 import ctypes
157
2459b6e1 158
86e5f3ed 159class YoutubeDL:
8222d8de
JMF
160 """YoutubeDL class.
161
162 YoutubeDL objects are the ones responsible of downloading the
163 actual video file and writing it to disk if the user has requested
164 it, among some other tasks. In most cases there should be one per
165 program. As, given a video URL, the downloader doesn't know how to
166 extract all the needed information, task that InfoExtractors do, it
167 has to pass the URL to one of them.
168
169 For this, YoutubeDL objects have a method that allows
170 InfoExtractors to be registered in a given order. When it is passed
171 a URL, the YoutubeDL object handles it to the first InfoExtractor it
172 finds that reports being able to handle it. The InfoExtractor extracts
173 all the information about the video or videos the URL refers to, and
174 YoutubeDL process the extracted information, possibly using a File
175 Downloader to download the video.
176
177 YoutubeDL objects accept a lot of parameters. In order not to saturate
178 the object constructor with arguments, it receives a dictionary of
179 options instead. These options are available through the params
180 attribute for the InfoExtractors to use. The YoutubeDL also
181 registers itself as the downloader in charge for the InfoExtractors
182 that are added to it, so this is a "mutual registration".
183
184 Available options:
185
186 username: Username for authentication purposes.
187 password: Password for authentication purposes.
180940e0 188 videopassword: Password for accessing a video.
1da50aa3
S
189 ap_mso: Adobe Pass multiple-system operator identifier.
190 ap_username: Multiple-system operator account username.
191 ap_password: Multiple-system operator account password.
8222d8de
JMF
192 usenetrc: Use netrc for authentication instead.
193 verbose: Print additional info to stdout.
194 quiet: Do not print messages to stdout.
ad8915b7 195 no_warnings: Do not print out anything for warnings.
bb66c247 196 forceprint: A dict with keys WHEN mapped to a list of templates to
197 print to stdout. The allowed keys are video or any of the
198 items in utils.POSTPROCESS_WHEN.
ca30f449 199 For compatibility, a single list is also accepted
bb66c247 200 print_to_file: A dict with keys WHEN (same as forceprint) mapped to
201 a list of tuples with (template, filename)
8694c600 202 forcejson: Force printing info_dict as JSON.
63e0be34
PH
203 dump_single_json: Force printing the info_dict of the whole playlist
204 (or video) as a single JSON line.
c25228e5 205 force_write_download_archive: Force writing download archive regardless
206 of 'skip_download' or 'simulate'.
b7b04c78 207 simulate: Do not download the video files. If unset (or None),
208 simulate only if listsubtitles, listformats or list_thumbnails is used
eb8a4433 209 format: Video format code. see "FORMAT SELECTION" for more details.
093a1710 210 You can also pass a function. The function takes 'ctx' as
211 argument and returns the formats to download.
212 See "build_format_selector" for an implementation
63ad4d43 213 allow_unplayable_formats: Allow unplayable formats to be extracted and downloaded.
b7da73eb 214 ignore_no_formats_error: Ignore "No video formats" error. Usefull for
215 extracting metadata even if the video is not actually
216 available for download (experimental)
0930b11f 217 format_sort: A list of fields by which to sort the video formats.
218 See "Sorting Formats" for more details.
c25228e5 219 format_sort_force: Force the given format_sort. see "Sorting Formats"
220 for more details.
08d30158 221 prefer_free_formats: Whether to prefer video formats with free containers
222 over non-free ones of same quality.
c25228e5 223 allow_multiple_video_streams: Allow multiple video streams to be merged
224 into a single file
225 allow_multiple_audio_streams: Allow multiple audio streams to be merged
226 into a single file
0ba692ac 227 check_formats Whether to test if the formats are downloadable.
9f1a1c36 228 Can be True (check all), False (check none),
229 'selected' (check selected formats),
0ba692ac 230 or None (check only if requested by extractor)
4524baf0 231 paths: Dictionary of output paths. The allowed keys are 'home'
232 'temp' and the keys of OUTTMPL_TYPES (in utils.py)
de6000d9 233 outtmpl: Dictionary of templates for output names. Allowed keys
4524baf0 234 are 'default' and the keys of OUTTMPL_TYPES (in utils.py).
34488702 235 For compatibility with youtube-dl, a single string can also be used
a820dc72
RA
236 outtmpl_na_placeholder: Placeholder for unavailable meta fields.
237 restrictfilenames: Do not allow "&" and spaces in file names
238 trim_file_name: Limit length of filename (extension excluded)
4524baf0 239 windowsfilenames: Force the filenames to be windows compatible
b1940459 240 ignoreerrors: Do not stop on download/postprocessing errors.
241 Can be 'only_download' to ignore only download errors.
242 Default is 'only_download' for CLI, but False for API
26e2805c 243 skip_playlist_after_errors: Number of allowed failures until the rest of
244 the playlist is skipped
fe7866d0 245 allowed_extractors: List of regexes to match against extractor names that are allowed
0c3d0f51 246 overwrites: Overwrite all video and metadata files if True,
247 overwrite only non-video files if None
248 and don't overwrite any file if False
34488702 249 For compatibility with youtube-dl,
250 "nooverwrites" may also be used instead
c14e88f0 251 playlist_items: Specific indices of playlist to download.
75822ca7 252 playlistrandom: Download playlist items in random order.
7e9a6125 253 lazy_playlist: Process playlist entries as they are received.
8222d8de
JMF
254 matchtitle: Download only matching titles.
255 rejecttitle: Reject downloads for matching titles.
8bf9319e 256 logger: Log messages to a logging.Logger instance.
17ffed18 257 logtostderr: Print everything to stderr instead of stdout.
258 consoletitle: Display progress in console window's titlebar.
8222d8de
JMF
259 writedescription: Write the video description to a .description file
260 writeinfojson: Write the video description to a .info.json file
75d43ca0 261 clean_infojson: Remove private fields from the infojson
34488702 262 getcomments: Extract video comments. This will not be written to disk
06167fbb 263 unless writeinfojson is also given
1fb07d10 264 writeannotations: Write the video annotations to a .annotations.xml file
8222d8de 265 writethumbnail: Write the thumbnail image to a file
c25228e5 266 allow_playlist_files: Whether to write playlists' description, infojson etc
267 also to disk when using the 'write*' options
ec82d85a 268 write_all_thumbnails: Write all thumbnail formats to files
732044af 269 writelink: Write an internet shortcut file, depending on the
270 current platform (.url/.webloc/.desktop)
271 writeurllink: Write a Windows internet shortcut file (.url)
272 writewebloclink: Write a macOS internet shortcut file (.webloc)
273 writedesktoplink: Write a Linux internet shortcut file (.desktop)
8222d8de 274 writesubtitles: Write the video subtitles to a file
741dd8ea 275 writeautomaticsub: Write the automatically generated subtitles to a file
8222d8de 276 listsubtitles: Lists all available subtitles for the video
a504ced0 277 subtitlesformat: The format code for subtitles
c32b0aab 278 subtitleslangs: List of languages of the subtitles to download (can be regex).
279 The list may contain "all" to refer to all the available
280 subtitles. The language can be prefixed with a "-" to
62b58c09 281 exclude it from the requested languages, e.g. ['all', '-live_chat']
8222d8de
JMF
282 keepvideo: Keep the video file after post-processing
283 daterange: A DateRange object, download only if the upload_date is in the range.
284 skip_download: Skip the actual download of the video file
c35f9e72 285 cachedir: Location of the cache files in the filesystem.
a0e07d31 286 False to disable filesystem cache.
47192f92 287 noplaylist: Download single video instead of a playlist if in doubt.
8dbe9899
PH
288 age_limit: An integer representing the user's age in years.
289 Unsuitable videos for the given age are skipped.
5fe18bdb
PH
290 min_views: An integer representing the minimum view count the video
291 must have in order to not be skipped.
292 Videos without view count information are always
293 downloaded. None for no limit.
294 max_views: An integer representing the maximum view count.
295 Videos that are more popular than that are not
296 downloaded.
297 Videos without view count information are always
298 downloaded. None for no limit.
ae103564 299 download_archive: A set, or the name of a file where all downloads are recorded.
300 Videos already present in the file are not downloaded again.
8a51f564 301 break_on_existing: Stop the download process after attempting to download a
302 file that is in the archive.
303 break_on_reject: Stop the download process when encountering a video that
304 has been filtered out.
b222c271 305 break_per_url: Whether break_on_reject and break_on_existing
306 should act on each input URL as opposed to for the entire queue
d76fa1f3 307 cookiefile: File name or text stream from where cookies should be read and dumped to
f59f5ef8 308 cookiesfrombrowser: A tuple containing the name of the browser, the profile
9bd13fe5 309 name/path from where cookies are loaded, the name of the keyring,
310 and the container name, e.g. ('chrome', ) or
311 ('vivaldi', 'default', 'BASICTEXT') or ('firefox', 'default', None, 'Meta')
f81c62a6 312 legacyserverconnect: Explicitly allow HTTPS connection to servers that do not
313 support RFC 5746 secure renegotiation
f59f5ef8 314 nocheckcertificate: Do not verify SSL certificates
bb58c9ed 315 client_certificate: Path to client certificate file in PEM format. May include the private key
316 client_certificate_key: Path to private key file for client certificate
317 client_certificate_password: Password for client certificate private key, if encrypted.
318 If not provided and the key is encrypted, yt-dlp will ask interactively
7e8c0af0 319 prefer_insecure: Use HTTP instead of HTTPS to retrieve information.
c6e07cf1 320 (Only supported by some extractors)
8300774c 321 enable_file_urls: Enable file:// URLs. This is disabled by default for security reasons.
8b7539d2 322 http_headers: A dictionary of custom headers to be used for all requests
a1ee09e8 323 proxy: URL of the proxy server to use
38cce791 324 geo_verification_proxy: URL of the proxy to use for IP address verification
504f20dd 325 on geo-restricted sites.
e344693b 326 socket_timeout: Time to wait for unresponsive hosts, in seconds
0783b09b
PH
327 bidi_workaround: Work around buggy terminals without bidirectional text
328 support, using fridibi
a0ddb8a2 329 debug_printtraffic:Print out sent and received HTTP traffic
04b4d394
PH
330 default_search: Prepend this string if an input url is not valid.
331 'auto' for elaborate guessing
62fec3b2 332 encoding: Use this encoding instead of the system-specified.
134c913c 333 extract_flat: Whether to resolve and process url_results further
334 * False: Always process (default)
335 * True: Never process
336 * 'in_playlist': Do not process inside playlist/multi_video
337 * 'discard': Always process, but don't return the result
338 from inside playlist/multi_video
339 * 'discard_in_playlist': Same as "discard", but only for
340 playlists (not multi_video)
f2ebc5c7 341 wait_for_video: If given, wait for scheduled streams to become available.
342 The value should be a tuple containing the range
343 (min_secs, max_secs) to wait between retries
4f026faf 344 postprocessors: A list of dictionaries, each with an entry
71b640cc 345 * key: The name of the postprocessor. See
7a5c1cfe 346 yt_dlp/postprocessor/__init__.py for a list.
bb66c247 347 * when: When to run the postprocessor. Allowed values are
348 the entries of utils.POSTPROCESS_WHEN
56d868db 349 Assumed to be 'post_process' if not given
71b640cc
PH
350 progress_hooks: A list of functions that get called on download
351 progress, with a dictionary with the entries
5cda4eda 352 * status: One of "downloading", "error", or "finished".
ee69b99a 353 Check this first and ignore unknown values.
3ba7740d 354 * info_dict: The extracted info_dict
71b640cc 355
5cda4eda 356 If status is one of "downloading", or "finished", the
ee69b99a
PH
357 following properties may also be present:
358 * filename: The final filename (always present)
5cda4eda 359 * tmpfilename: The filename we're currently writing to
71b640cc
PH
360 * downloaded_bytes: Bytes on disk
361 * total_bytes: Size of the whole file, None if unknown
5cda4eda
PH
362 * total_bytes_estimate: Guess of the eventual file size,
363 None if unavailable.
364 * elapsed: The number of seconds since download started.
71b640cc
PH
365 * eta: The estimated time in seconds, None if unknown
366 * speed: The download speed in bytes/second, None if
367 unknown
5cda4eda
PH
368 * fragment_index: The counter of the currently
369 downloaded video fragment.
370 * fragment_count: The number of fragments (= individual
371 files that will be merged)
71b640cc
PH
372
373 Progress hooks are guaranteed to be called at least once
374 (with status "finished") if the download is successful.
819e0531 375 postprocessor_hooks: A list of functions that get called on postprocessing
376 progress, with a dictionary with the entries
377 * status: One of "started", "processing", or "finished".
378 Check this first and ignore unknown values.
379 * postprocessor: Name of the postprocessor
380 * info_dict: The extracted info_dict
381
382 Progress hooks are guaranteed to be called at least twice
383 (with status "started" and "finished") if the processing is successful.
fc61aff4 384 merge_output_format: "/" separated list of extensions to use when merging formats.
6b591b29 385 final_ext: Expected final extension; used to detect when the file was
59a7a13e 386 already downloaded and converted
6271f1ca
PH
387 fixup: Automatically correct known faults of the file.
388 One of:
389 - "never": do nothing
390 - "warn": only emit a warning
391 - "detect_or_warn": check whether we can do anything
62cd676c 392 about it, warn otherwise (default)
504f20dd 393 source_address: Client-side IP address to bind to.
1cf376f5 394 sleep_interval_requests: Number of seconds to sleep between requests
395 during extraction
7aa589a5
S
396 sleep_interval: Number of seconds to sleep before each download when
397 used alone or a lower bound of a range for randomized
398 sleep before each download (minimum possible number
399 of seconds to sleep) when used along with
400 max_sleep_interval.
401 max_sleep_interval:Upper bound of a range for randomized sleep before each
402 download (maximum possible number of seconds to sleep).
403 Must only be used along with sleep_interval.
404 Actual sleep time will be a random float from range
405 [sleep_interval; max_sleep_interval].
1cf376f5 406 sleep_interval_subtitles: Number of seconds to sleep before each subtitle download
cfb56d1a
PH
407 listformats: Print an overview of available video formats and exit.
408 list_thumbnails: Print a table of all thumbnails and exit.
0a41f331 409 match_filter: A function that gets called for every video with the signature
410 (info_dict, *, incomplete: bool) -> Optional[str]
411 For backward compatibility with youtube-dl, the signature
412 (info_dict) -> Optional[str] is also allowed.
413 - If it returns a message, the video is ignored.
414 - If it returns None, the video is downloaded.
415 - If it returns utils.NO_DEFAULT, the user is interactively
416 asked whether to download the video.
347de493 417 match_filter_func in utils.py is one example for this.
7e5db8c9 418 no_color: Do not emit color codes in output.
0a840f58 419 geo_bypass: Bypass geographic restriction via faking X-Forwarded-For
504f20dd 420 HTTP header
0a840f58 421 geo_bypass_country:
773f291d
S
422 Two-letter ISO 3166-2 country code that will be used for
423 explicit geographic restriction bypassing via faking
504f20dd 424 X-Forwarded-For HTTP header
5f95927a
S
425 geo_bypass_ip_block:
426 IP range in CIDR notation that will be used similarly to
504f20dd 427 geo_bypass_country
52a8a1e1 428 external_downloader: A dictionary of protocol keys and the executable of the
429 external downloader to use for it. The allowed protocols
430 are default|http|ftp|m3u8|dash|rtsp|rtmp|mms.
431 Set the value to 'native' to use the native downloader
53ed7066 432 compat_opts: Compatibility options. See "Differences in default behavior".
3acf6d38 433 The following options do not work when used through the API:
b5ae35ee 434 filename, abort-on-error, multistreams, no-live-chat, format-sort
dac5df5a 435 no-clean-infojson, no-playlist-metafiles, no-keep-subs, no-attach-info-json.
e4f02757 436 Refer __init__.py for their implementation
819e0531 437 progress_template: Dictionary of templates for progress outputs.
438 Allowed keys are 'download', 'postprocess',
439 'download-title' (console title) and 'postprocess-title'.
440 The template is mapped on a dictionary with keys 'progress' and 'info'
23326151 441 retry_sleep_functions: Dictionary of functions that takes the number of attempts
442 as argument and returns the time to sleep in seconds.
443 Allowed keys are 'http', 'fragment', 'file_access'
0f446365
SW
444 download_ranges: A callback function that gets called for every video with
445 the signature (info_dict, ydl) -> Iterable[Section].
446 Only the returned sections will be downloaded.
447 Each Section is a dict with the following keys:
5ec1b6b7 448 * start_time: Start time of the section in seconds
449 * end_time: End time of the section in seconds
450 * title: Section title (Optional)
451 * index: Section number (Optional)
0f446365 452 force_keyframes_at_cuts: Re-encode the video when downloading ranges to get precise cuts
a7dc6a89 453 noprogress: Do not print the progress bar
a831c2ea 454 live_from_start: Whether to download livestreams videos from the start
fe7e0c98 455
8222d8de 456 The following parameters are not used by YoutubeDL itself, they are used by
7a5c1cfe 457 the downloader (see yt_dlp/downloader/common.py):
51d9739f 458 nopart, updatetime, buffersize, ratelimit, throttledratelimit, min_filesize,
205a0654 459 max_filesize, test, noresizebuffer, retries, file_access_retries, fragment_retries,
a7dc6a89 460 continuedl, xattr_set_filesize, hls_use_mpegts, http_chunk_size,
59a7a13e 461 external_downloader_args, concurrent_fragment_downloads.
76b1bd67
JMF
462
463 The following options are used by the post processors:
c0b7d117
S
464 ffmpeg_location: Location of the ffmpeg/avconv binary; either the path
465 to the binary or its containing directory.
43820c03 466 postprocessor_args: A dictionary of postprocessor/executable keys (in lower case)
34488702 467 and a list of additional command-line arguments for the
468 postprocessor/executable. The dict can also have "PP+EXE" keys
469 which are used when the given exe is used by the given PP.
470 Use 'default' as the name for arguments to passed to all PP
471 For compatibility with youtube-dl, a single list of args
472 can also be used
e409895f 473
474 The following options are used by the extractors:
62bff2c1 475 extractor_retries: Number of times to retry for known errors
476 dynamic_mpd: Whether to process dynamic DASH manifests (default: True)
e409895f 477 hls_split_discontinuity: Split HLS playlists to different formats at
62bff2c1 478 discontinuities such as ad breaks (default: False)
5d3a0e79 479 extractor_args: A dictionary of arguments to be passed to the extractors.
480 See "EXTRACTOR ARGUMENTS" for details.
62b58c09 481 E.g. {'youtube': {'skip': ['dash', 'hls']}}
88f23a18 482 mark_watched: Mark videos watched (even with --simulate). Only for YouTube
1890fc63 483
484 The following options are deprecated and may be removed in the future:
485
fe7866d0 486 force_generic_extractor: Force downloader to use the generic extractor
487 - Use allowed_extractors = ['generic', 'default']
7e9a6125 488 playliststart: - Use playlist_items
489 Playlist item to start at.
490 playlistend: - Use playlist_items
491 Playlist item to end at.
492 playlistreverse: - Use playlist_items
493 Download playlist items in reverse order.
1890fc63 494 forceurl: - Use forceprint
495 Force printing final URL.
496 forcetitle: - Use forceprint
497 Force printing title.
498 forceid: - Use forceprint
499 Force printing ID.
500 forcethumbnail: - Use forceprint
501 Force printing thumbnail URL.
502 forcedescription: - Use forceprint
503 Force printing description.
504 forcefilename: - Use forceprint
505 Force printing final filename.
506 forceduration: - Use forceprint
507 Force printing duration.
508 allsubtitles: - Use subtitleslangs = ['all']
509 Downloads all the subtitles of the video
510 (requires writesubtitles or writeautomaticsub)
511 include_ads: - Doesn't work
512 Download ads as well
513 call_home: - Not implemented
514 Boolean, true iff we are allowed to contact the
515 yt-dlp servers for debugging.
516 post_hooks: - Register a custom postprocessor
517 A list of functions that get called as the final step
518 for each video file, after all postprocessors have been
519 called. The filename will be passed as the only argument.
520 hls_prefer_native: - Use external_downloader = {'m3u8': 'native'} or {'m3u8': 'ffmpeg'}.
521 Use the native HLS downloader instead of ffmpeg/avconv
522 if True, otherwise use ffmpeg/avconv if False, otherwise
523 use downloader suggested by extractor if None.
524 prefer_ffmpeg: - avconv support is deprecated
525 If False, use avconv instead of ffmpeg if both are available,
526 otherwise prefer ffmpeg.
527 youtube_include_dash_manifest: - Use extractor_args
5d3a0e79 528 If True (default), DASH manifests and related
62bff2c1 529 data will be downloaded and processed by extractor.
530 You can reduce network I/O by disabling it if you don't
531 care about DASH. (only for youtube)
1890fc63 532 youtube_include_hls_manifest: - Use extractor_args
5d3a0e79 533 If True (default), HLS manifests and related
62bff2c1 534 data will be downloaded and processed by extractor.
535 You can reduce network I/O by disabling it if you don't
536 care about HLS. (only for youtube)
8222d8de
JMF
537 """
538
86e5f3ed 539 _NUMERIC_FIELDS = {
b8ed0f15 540 'width', 'height', 'asr', 'audio_channels', 'fps',
541 'tbr', 'abr', 'vbr', 'filesize', 'filesize_approx',
e6f21b3d 542 'timestamp', 'release_timestamp',
c9969434
S
543 'duration', 'view_count', 'like_count', 'dislike_count', 'repost_count',
544 'average_rating', 'comment_count', 'age_limit',
545 'start_time', 'end_time',
546 'chapter_number', 'season_number', 'episode_number',
547 'track_number', 'disc_number', 'release_year',
86e5f3ed 548 }
c9969434 549
6db9c4d5 550 _format_fields = {
551 # NB: Keep in sync with the docstring of extractor/common.py
a44ca5a4 552 'url', 'manifest_url', 'manifest_stream_number', 'ext', 'format', 'format_id', 'format_note',
105bfd90 553 'width', 'height', 'aspect_ratio', 'resolution', 'dynamic_range', 'tbr', 'abr', 'acodec', 'asr', 'audio_channels',
d5d1df8a 554 'vbr', 'fps', 'vcodec', 'container', 'filesize', 'filesize_approx', 'rows', 'columns',
6db9c4d5 555 'player_url', 'protocol', 'fragment_base_url', 'fragments', 'is_from_start',
556 'preference', 'language', 'language_preference', 'quality', 'source_preference',
557 'http_headers', 'stretched_ratio', 'no_resume', 'has_drm', 'downloader_options',
558 'page_url', 'app', 'play_path', 'tc_url', 'flash_version', 'rtmp_live', 'rtmp_conn', 'rtmp_protocol', 'rtmp_real_time'
559 }
48ee10ee 560 _format_selection_exts = {
8dc59305 561 'audio': set(MEDIA_EXTENSIONS.common_audio),
562 'video': set(MEDIA_EXTENSIONS.common_video + ('3gp', )),
563 'storyboards': set(MEDIA_EXTENSIONS.storyboards),
48ee10ee 564 }
565
3511266b 566 def __init__(self, params=None, auto_init=True):
883d4b1e 567 """Create a FileDownloader object with the given options.
568 @param auto_init Whether to load the default extractors and print header (if verbose).
49a57e70 569 Set to 'no_verbose_header' to not print the header
883d4b1e 570 """
e9f9a10f
JMF
571 if params is None:
572 params = {}
592b7485 573 self.params = params
8b7491c8 574 self._ies = {}
56c73665 575 self._ies_instances = {}
1e43a6f7 576 self._pps = {k: [] for k in POSTPROCESS_WHEN}
b35496d8 577 self._printed_messages = set()
1cf376f5 578 self._first_webpage_request = True
ab8e5e51 579 self._post_hooks = []
933605d7 580 self._progress_hooks = []
819e0531 581 self._postprocessor_hooks = []
8222d8de
JMF
582 self._download_retcode = 0
583 self._num_downloads = 0
9c906919 584 self._num_videos = 0
592b7485 585 self._playlist_level = 0
586 self._playlist_urls = set()
a0e07d31 587 self.cache = Cache(self)
34308b30 588
591bb9d3 589 stdout = sys.stderr if self.params.get('logtostderr') else sys.stdout
590 self._out_files = Namespace(
591 out=stdout,
592 error=sys.stderr,
593 screen=sys.stderr if self.params.get('quiet') else stdout,
594 console=None if compat_os_name == 'nt' else next(
cf4f42cb 595 filter(supports_terminal_sequences, (sys.stderr, sys.stdout)), None)
591bb9d3 596 )
f0795149 597
598 try:
599 windows_enable_vt_mode()
600 except Exception as e:
601 self.write_debug(f'Failed to enable VT mode: {e}')
602
591bb9d3 603 self._allow_colors = Namespace(**{
604 type_: not self.params.get('no_color') and supports_terminal_sequences(stream)
64fa820c 605 for type_, stream in self._out_files.items_ if type_ != 'console'
591bb9d3 606 })
819e0531 607
6929b41a 608 # The code is left like this to be reused for future deprecations
609 MIN_SUPPORTED, MIN_RECOMMENDED = (3, 7), (3, 7)
eff42759 610 current_version = sys.version_info[:2]
611 if current_version < MIN_RECOMMENDED:
9d339c41 612 msg = ('Support for Python version %d.%d has been deprecated. '
24093d52 613 'See https://github.com/yt-dlp/yt-dlp/issues/3764 for more details.'
c6e07cf1 614 '\n You will no longer receive updates on this version')
eff42759 615 if current_version < MIN_SUPPORTED:
616 msg = 'Python version %d.%d is no longer supported'
617 self.deprecation_warning(
618 f'{msg}! Please update to Python %d.%d or above' % (*current_version, *MIN_RECOMMENDED))
a61f4b28 619
88acdbc2 620 if self.params.get('allow_unplayable_formats'):
621 self.report_warning(
ec11a9f4 622 f'You have asked for {self._format_err("UNPLAYABLE", self.Styles.EMPHASIS)} formats to be listed/downloaded. '
819e0531 623 'This is a developer option intended for debugging. \n'
624 ' If you experience any issues while using this option, '
ec11a9f4 625 f'{self._format_err("DO NOT", self.Styles.ERROR)} open a bug report')
88acdbc2 626
497074f0 627 if self.params.get('bidi_workaround', False):
628 try:
629 import pty
630 master, slave = pty.openpty()
631 width = shutil.get_terminal_size().columns
632 width_args = [] if width is None else ['-w', str(width)]
633 sp_kwargs = {'stdin': subprocess.PIPE, 'stdout': slave, 'stderr': self._out_files.error}
634 try:
635 self._output_process = Popen(['bidiv'] + width_args, **sp_kwargs)
636 except OSError:
637 self._output_process = Popen(['fribidi', '-c', 'UTF-8'] + width_args, **sp_kwargs)
638 self._output_channel = os.fdopen(master, 'rb')
639 except OSError as ose:
640 if ose.errno == errno.ENOENT:
641 self.report_warning(
642 'Could not find fribidi executable, ignoring --bidi-workaround. '
643 'Make sure that fribidi is an executable file in one of the directories in your $PATH.')
644 else:
645 raise
646
647 self.params['compat_opts'] = set(self.params.get('compat_opts', ()))
648 if auto_init and auto_init != 'no_verbose_header':
649 self.print_debug_header()
650
be5df5ee
S
651 def check_deprecated(param, option, suggestion):
652 if self.params.get(param) is not None:
86e5f3ed 653 self.report_warning(f'{option} is deprecated. Use {suggestion} instead')
be5df5ee
S
654 return True
655 return False
656
657 if check_deprecated('cn_verification_proxy', '--cn-verification-proxy', '--geo-verification-proxy'):
38cce791
YCH
658 if self.params.get('geo_verification_proxy') is None:
659 self.params['geo_verification_proxy'] = self.params['cn_verification_proxy']
660
0d1bb027 661 check_deprecated('autonumber', '--auto-number', '-o "%(autonumber)s-%(title)s.%(ext)s"')
662 check_deprecated('usetitle', '--title', '-o "%(title)s-%(id)s.%(ext)s"')
53ed7066 663 check_deprecated('useid', '--id', '-o "%(id)s.%(ext)s"')
0d1bb027 664
49a57e70 665 for msg in self.params.get('_warnings', []):
0d1bb027 666 self.report_warning(msg)
ee8dd27a 667 for msg in self.params.get('_deprecation_warnings', []):
da4db748 668 self.deprecated_feature(msg)
0d1bb027 669
8a82af35 670 if 'list-formats' in self.params['compat_opts']:
ec11a9f4 671 self.params['listformats_table'] = False
672
b5ae35ee 673 if 'overwrites' not in self.params and self.params.get('nooverwrites') is not None:
b868936c 674 # nooverwrites was unnecessarily changed to overwrites
675 # in 0c3d0f51778b153f65c21906031c2e091fcfb641
676 # This ensures compatibility with both keys
677 self.params['overwrites'] = not self.params['nooverwrites']
b5ae35ee 678 elif self.params.get('overwrites') is None:
679 self.params.pop('overwrites', None)
b868936c 680 else:
681 self.params['nooverwrites'] = not self.params['overwrites']
b9d973be 682
e4221b70 683 if self.params.get('simulate') is None and any((
684 self.params.get('list_thumbnails'),
685 self.params.get('listformats'),
686 self.params.get('listsubtitles'),
687 )):
688 self.params['simulate'] = 'list_only'
689
455a15e2 690 self.params.setdefault('forceprint', {})
691 self.params.setdefault('print_to_file', {})
bb66c247 692
693 # Compatibility with older syntax
ca30f449 694 if not isinstance(params['forceprint'], dict):
455a15e2 695 self.params['forceprint'] = {'video': params['forceprint']}
ca30f449 696
97ec5bc5 697 if auto_init:
97ec5bc5 698 self.add_default_info_extractors()
699
3089bc74
S
700 if (sys.platform != 'win32'
701 and sys.getfilesystemencoding() in ['ascii', 'ANSI_X3.4-1968']
455a15e2 702 and not self.params.get('restrictfilenames', False)):
e9137224 703 # Unicode filesystem API will throw errors (#1474, #13027)
34308b30 704 self.report_warning(
6febd1c1 705 'Assuming --restrict-filenames since file system encoding '
1b725173 706 'cannot encode all characters. '
6febd1c1 707 'Set the LC_ALL environment variable to fix this.')
4a98cdbf 708 self.params['restrictfilenames'] = True
34308b30 709
bf1824b3 710 self._parse_outtmpl()
486dd09e 711
187986a8 712 # Creating format selector here allows us to catch syntax errors before the extraction
713 self.format_selector = (
fa9f30b8 714 self.params.get('format') if self.params.get('format') in (None, '-')
093a1710 715 else self.params['format'] if callable(self.params['format'])
187986a8 716 else self.build_format_selector(self.params['format']))
717
8b7539d2 718 # Set http_headers defaults according to std_headers
719 self.params['http_headers'] = merge_headers(std_headers, self.params.get('http_headers', {}))
720
013b50b7 721 hooks = {
722 'post_hooks': self.add_post_hook,
723 'progress_hooks': self.add_progress_hook,
724 'postprocessor_hooks': self.add_postprocessor_hook,
725 }
726 for opt, fn in hooks.items():
727 for ph in self.params.get(opt, []):
728 fn(ph)
71b640cc 729
5bfc8bee 730 for pp_def_raw in self.params.get('postprocessors', []):
731 pp_def = dict(pp_def_raw)
732 when = pp_def.pop('when', 'post_process')
733 self.add_post_processor(
f9934b96 734 get_postprocessor(pp_def.pop('key'))(self, **pp_def),
5bfc8bee 735 when=when)
736
97ec5bc5 737 self._setup_opener()
51fb4995
YCH
738 register_socks_protocols()
739
ed39cac5 740 def preload_download_archive(fn):
741 """Preload the archive, if any is specified"""
ae103564 742 archive = set()
ed39cac5 743 if fn is None:
ae103564 744 return archive
941e881e 745 elif not is_path_like(fn):
ae103564 746 return fn
747
49a57e70 748 self.write_debug(f'Loading archive file {fn!r}')
ed39cac5 749 try:
750 with locked_file(fn, 'r', encoding='utf-8') as archive_file:
751 for line in archive_file:
ae103564 752 archive.add(line.strip())
86e5f3ed 753 except OSError as ioe:
ed39cac5 754 if ioe.errno != errno.ENOENT:
755 raise
ae103564 756 return archive
ed39cac5 757
ae103564 758 self.archive = preload_download_archive(self.params.get('download_archive'))
ed39cac5 759
7d4111ed
PH
760 def warn_if_short_id(self, argv):
761 # short YouTube ID starting with dash?
762 idxs = [
763 i for i, a in enumerate(argv)
764 if re.match(r'^-[0-9A-Za-z_-]{10}$', a)]
765 if idxs:
766 correct_argv = (
7a5c1cfe 767 ['yt-dlp']
3089bc74
S
768 + [a for i, a in enumerate(argv) if i not in idxs]
769 + ['--'] + [argv[i] for i in idxs]
7d4111ed
PH
770 )
771 self.report_warning(
772 'Long argument string detected. '
49a57e70 773 'Use -- to separate parameters and URLs, like this:\n%s' %
7d4111ed
PH
774 args_to_str(correct_argv))
775
8222d8de
JMF
776 def add_info_extractor(self, ie):
777 """Add an InfoExtractor object to the end of the list."""
8b7491c8 778 ie_key = ie.ie_key()
779 self._ies[ie_key] = ie
e52d7f85 780 if not isinstance(ie, type):
8b7491c8 781 self._ies_instances[ie_key] = ie
e52d7f85 782 ie.set_downloader(self)
8222d8de 783
56c73665
JMF
784 def get_info_extractor(self, ie_key):
785 """
786 Get an instance of an IE with name ie_key, it will try to get one from
787 the _ies list, if there's no instance it will create a new one and add
788 it to the extractor list.
789 """
790 ie = self._ies_instances.get(ie_key)
791 if ie is None:
792 ie = get_info_extractor(ie_key)()
793 self.add_info_extractor(ie)
794 return ie
795
023fa8c4
JMF
796 def add_default_info_extractors(self):
797 """
798 Add the InfoExtractors returned by gen_extractors to the end of the list
799 """
fe7866d0 800 all_ies = {ie.IE_NAME.lower(): ie for ie in gen_extractor_classes()}
801 all_ies['end'] = UnsupportedURLIE()
802 try:
803 ie_names = orderedSet_from_options(
804 self.params.get('allowed_extractors', ['default']), {
805 'all': list(all_ies),
806 'default': [name for name, ie in all_ies.items() if ie._ENABLED],
807 }, use_regex=True)
808 except re.error as e:
809 raise ValueError(f'Wrong regex for allowed_extractors: {e.pattern}')
810 for name in ie_names:
811 self.add_info_extractor(all_ies[name])
812 self.write_debug(f'Loaded {len(ie_names)} extractors')
023fa8c4 813
56d868db 814 def add_post_processor(self, pp, when='post_process'):
8222d8de 815 """Add a PostProcessor object to the end of the chain."""
8aa0e7cd 816 assert when in POSTPROCESS_WHEN, f'Invalid when={when}'
5bfa4862 817 self._pps[when].append(pp)
8222d8de
JMF
818 pp.set_downloader(self)
819
ab8e5e51
AM
820 def add_post_hook(self, ph):
821 """Add the post hook"""
822 self._post_hooks.append(ph)
823
933605d7 824 def add_progress_hook(self, ph):
819e0531 825 """Add the download progress hook"""
933605d7 826 self._progress_hooks.append(ph)
8ab470f1 827
819e0531 828 def add_postprocessor_hook(self, ph):
829 """Add the postprocessing progress hook"""
830 self._postprocessor_hooks.append(ph)
5bfc8bee 831 for pps in self._pps.values():
832 for pp in pps:
833 pp.add_progress_hook(ph)
819e0531 834
1c088fa8 835 def _bidi_workaround(self, message):
5d681e96 836 if not hasattr(self, '_output_channel'):
1c088fa8
PH
837 return message
838
5d681e96 839 assert hasattr(self, '_output_process')
14f25df2 840 assert isinstance(message, str)
6febd1c1 841 line_count = message.count('\n') + 1
0f06bcd7 842 self._output_process.stdin.write((message + '\n').encode())
5d681e96 843 self._output_process.stdin.flush()
0f06bcd7 844 res = ''.join(self._output_channel.readline().decode()
9e1a5b84 845 for _ in range(line_count))
6febd1c1 846 return res[:-len('\n')]
1c088fa8 847
b35496d8 848 def _write_string(self, message, out=None, only_once=False):
849 if only_once:
850 if message in self._printed_messages:
851 return
852 self._printed_messages.add(message)
853 write_string(message, out=out, encoding=self.params.get('encoding'))
734f90bb 854
cf4f42cb 855 def to_stdout(self, message, skip_eol=False, quiet=None):
0760b0a7 856 """Print message to stdout"""
cf4f42cb 857 if quiet is not None:
da4db748 858 self.deprecation_warning('"YoutubeDL.to_stdout" no longer accepts the argument quiet. '
859 'Use "YoutubeDL.to_screen" instead')
8a82af35 860 if skip_eol is not False:
da4db748 861 self.deprecation_warning('"YoutubeDL.to_stdout" no longer accepts the argument skip_eol. '
862 'Use "YoutubeDL.to_screen" instead')
0bf9dc1e 863 self._write_string(f'{self._bidi_workaround(message)}\n', self._out_files.out)
cf4f42cb 864
dfea94f8 865 def to_screen(self, message, skip_eol=False, quiet=None, only_once=False):
cf4f42cb 866 """Print message to screen if not in quiet mode"""
8bf9319e 867 if self.params.get('logger'):
43afe285 868 self.params['logger'].debug(message)
cf4f42cb 869 return
870 if (self.params.get('quiet') if quiet is None else quiet) and not self.params.get('verbose'):
871 return
872 self._write_string(
873 '%s%s' % (self._bidi_workaround(message), ('' if skip_eol else '\n')),
dfea94f8 874 self._out_files.screen, only_once=only_once)
8222d8de 875
b35496d8 876 def to_stderr(self, message, only_once=False):
0760b0a7 877 """Print message to stderr"""
14f25df2 878 assert isinstance(message, str)
8bf9319e 879 if self.params.get('logger'):
43afe285
IB
880 self.params['logger'].error(message)
881 else:
5792c950 882 self._write_string(f'{self._bidi_workaround(message)}\n', self._out_files.error, only_once=only_once)
cf4f42cb 883
884 def _send_console_code(self, code):
591bb9d3 885 if compat_os_name == 'nt' or not self._out_files.console:
cf4f42cb 886 return
591bb9d3 887 self._write_string(code, self._out_files.console)
8222d8de 888
1e5b9a95
PH
889 def to_console_title(self, message):
890 if not self.params.get('consoletitle', False):
891 return
3efb96a6 892 message = remove_terminal_sequences(message)
4bede0d8
C
893 if compat_os_name == 'nt':
894 if ctypes.windll.kernel32.GetConsoleWindow():
895 # c_wchar_p() might not be necessary if `message` is
896 # already of type unicode()
897 ctypes.windll.kernel32.SetConsoleTitleW(ctypes.c_wchar_p(message))
cf4f42cb 898 else:
899 self._send_console_code(f'\033]0;{message}\007')
1e5b9a95 900
bdde425c 901 def save_console_title(self):
cf4f42cb 902 if not self.params.get('consoletitle') or self.params.get('simulate'):
bdde425c 903 return
592b7485 904 self._send_console_code('\033[22;0t') # Save the title on stack
bdde425c
PH
905
906 def restore_console_title(self):
cf4f42cb 907 if not self.params.get('consoletitle') or self.params.get('simulate'):
bdde425c 908 return
592b7485 909 self._send_console_code('\033[23;0t') # Restore the title from stack
bdde425c
PH
910
911 def __enter__(self):
912 self.save_console_title()
913 return self
914
915 def __exit__(self, *args):
916 self.restore_console_title()
f89197d7 917
dca08720 918 if self.params.get('cookiefile') is not None:
1bab3437 919 self.cookiejar.save(ignore_discard=True, ignore_expires=True)
bdde425c 920
fa9f30b8 921 def trouble(self, message=None, tb=None, is_error=True):
8222d8de
JMF
922 """Determine action to take when a download problem appears.
923
924 Depending on if the downloader has been configured to ignore
925 download errors or not, this method may throw an exception or
926 not when errors are found, after printing the message.
927
fa9f30b8 928 @param tb If given, is additional traceback information
929 @param is_error Whether to raise error according to ignorerrors
8222d8de
JMF
930 """
931 if message is not None:
932 self.to_stderr(message)
933 if self.params.get('verbose'):
934 if tb is None:
935 if sys.exc_info()[0]: # if .trouble has been called from an except block
6febd1c1 936 tb = ''
8222d8de 937 if hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
6febd1c1 938 tb += ''.join(traceback.format_exception(*sys.exc_info()[1].exc_info))
c0384f22 939 tb += encode_compat_str(traceback.format_exc())
8222d8de
JMF
940 else:
941 tb_data = traceback.format_list(traceback.extract_stack())
6febd1c1 942 tb = ''.join(tb_data)
c19bc311 943 if tb:
944 self.to_stderr(tb)
fa9f30b8 945 if not is_error:
946 return
b1940459 947 if not self.params.get('ignoreerrors'):
8222d8de
JMF
948 if sys.exc_info()[0] and hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
949 exc_info = sys.exc_info()[1].exc_info
950 else:
951 exc_info = sys.exc_info()
952 raise DownloadError(message, exc_info)
953 self._download_retcode = 1
954
19a03940 955 Styles = Namespace(
956 HEADERS='yellow',
957 EMPHASIS='light blue',
492272fe 958 FILENAME='green',
19a03940 959 ID='green',
960 DELIM='blue',
961 ERROR='red',
962 WARNING='yellow',
963 SUPPRESS='light black',
964 )
ec11a9f4 965
7578d77d 966 def _format_text(self, handle, allow_colors, text, f, fallback=None, *, test_encoding=False):
e5a998f3 967 text = str(text)
ec11a9f4 968 if test_encoding:
969 original_text = text
5c104538 970 # handle.encoding can be None. See https://github.com/yt-dlp/yt-dlp/issues/2711
971 encoding = self.params.get('encoding') or getattr(handle, 'encoding', None) or 'ascii'
ec11a9f4 972 text = text.encode(encoding, 'ignore').decode(encoding)
973 if fallback is not None and text != original_text:
974 text = fallback
7578d77d 975 return format_text(text, f) if allow_colors else text if fallback is None else fallback
ec11a9f4 976
591bb9d3 977 def _format_out(self, *args, **kwargs):
978 return self._format_text(self._out_files.out, self._allow_colors.out, *args, **kwargs)
979
ec11a9f4 980 def _format_screen(self, *args, **kwargs):
591bb9d3 981 return self._format_text(self._out_files.screen, self._allow_colors.screen, *args, **kwargs)
ec11a9f4 982
983 def _format_err(self, *args, **kwargs):
591bb9d3 984 return self._format_text(self._out_files.error, self._allow_colors.error, *args, **kwargs)
819e0531 985
c84aeac6 986 def report_warning(self, message, only_once=False):
8222d8de
JMF
987 '''
988 Print the message to stderr, it will be prefixed with 'WARNING:'
989 If stderr is a tty file the 'WARNING:' will be colored
990 '''
6d07ce01
JMF
991 if self.params.get('logger') is not None:
992 self.params['logger'].warning(message)
8222d8de 993 else:
ad8915b7
PH
994 if self.params.get('no_warnings'):
995 return
ec11a9f4 996 self.to_stderr(f'{self._format_err("WARNING:", self.Styles.WARNING)} {message}', only_once)
8222d8de 997
da4db748 998 def deprecation_warning(self, message, *, stacklevel=0):
999 deprecation_warning(
1000 message, stacklevel=stacklevel + 1, printer=self.report_error, is_error=False)
1001
1002 def deprecated_feature(self, message):
ee8dd27a 1003 if self.params.get('logger') is not None:
da4db748 1004 self.params['logger'].warning(f'Deprecated Feature: {message}')
1005 self.to_stderr(f'{self._format_err("Deprecated Feature:", self.Styles.ERROR)} {message}', True)
ee8dd27a 1006
fa9f30b8 1007 def report_error(self, message, *args, **kwargs):
8222d8de
JMF
1008 '''
1009 Do the same as trouble, but prefixes the message with 'ERROR:', colored
1010 in red if stderr is a tty file.
1011 '''
fa9f30b8 1012 self.trouble(f'{self._format_err("ERROR:", self.Styles.ERROR)} {message}', *args, **kwargs)
8222d8de 1013
b35496d8 1014 def write_debug(self, message, only_once=False):
0760b0a7 1015 '''Log debug message or Print message to stderr'''
1016 if not self.params.get('verbose', False):
1017 return
8a82af35 1018 message = f'[debug] {message}'
0760b0a7 1019 if self.params.get('logger'):
1020 self.params['logger'].debug(message)
1021 else:
b35496d8 1022 self.to_stderr(message, only_once)
0760b0a7 1023
8222d8de
JMF
1024 def report_file_already_downloaded(self, file_name):
1025 """Report file has already been fully downloaded."""
1026 try:
6febd1c1 1027 self.to_screen('[download] %s has already been downloaded' % file_name)
ce02ed60 1028 except UnicodeEncodeError:
6febd1c1 1029 self.to_screen('[download] The file has already been downloaded')
8222d8de 1030
0c3d0f51 1031 def report_file_delete(self, file_name):
1032 """Report that existing file will be deleted."""
1033 try:
c25228e5 1034 self.to_screen('Deleting existing file %s' % file_name)
0c3d0f51 1035 except UnicodeEncodeError:
c25228e5 1036 self.to_screen('Deleting existing file')
0c3d0f51 1037
319b6059 1038 def raise_no_formats(self, info, forced=False, *, msg=None):
0a5a191a 1039 has_drm = info.get('_has_drm')
319b6059 1040 ignored, expected = self.params.get('ignore_no_formats_error'), bool(msg)
1041 msg = msg or has_drm and 'This video is DRM protected' or 'No video formats found!'
1042 if forced or not ignored:
1151c407 1043 raise ExtractorError(msg, video_id=info['id'], ie=info['extractor'],
319b6059 1044 expected=has_drm or ignored or expected)
88acdbc2 1045 else:
1046 self.report_warning(msg)
1047
de6000d9 1048 def parse_outtmpl(self):
bf1824b3 1049 self.deprecation_warning('"YoutubeDL.parse_outtmpl" is deprecated and may be removed in a future version')
1050 self._parse_outtmpl()
1051 return self.params['outtmpl']
1052
1053 def _parse_outtmpl(self):
7b2c3f47 1054 sanitize = IDENTITY
bf1824b3 1055 if self.params.get('restrictfilenames'): # Remove spaces in the default template
71ce444a 1056 sanitize = lambda x: x.replace(' - ', ' ').replace(' ', '-')
bf1824b3 1057
1058 outtmpl = self.params.setdefault('outtmpl', {})
1059 if not isinstance(outtmpl, dict):
1060 self.params['outtmpl'] = outtmpl = {'default': outtmpl}
1061 outtmpl.update({k: sanitize(v) for k, v in DEFAULT_OUTTMPL.items() if outtmpl.get(k) is None})
de6000d9 1062
21cd8fae 1063 def get_output_path(self, dir_type='', filename=None):
1064 paths = self.params.get('paths', {})
d2c8aadf 1065 assert isinstance(paths, dict), '"paths" parameter must be a dictionary'
21cd8fae 1066 path = os.path.join(
1067 expand_path(paths.get('home', '').strip()),
1068 expand_path(paths.get(dir_type, '').strip()) if dir_type else '',
1069 filename or '')
21cd8fae 1070 return sanitize_path(path, force=self.params.get('windowsfilenames'))
1071
76a264ac 1072 @staticmethod
901130bb 1073 def _outtmpl_expandpath(outtmpl):
1074 # expand_path translates '%%' into '%' and '$$' into '$'
1075 # correspondingly that is not what we want since we need to keep
1076 # '%%' intact for template dict substitution step. Working around
1077 # with boundary-alike separator hack.
efa944f4 1078 sep = ''.join(random.choices(ascii_letters, k=32))
86e5f3ed 1079 outtmpl = outtmpl.replace('%%', f'%{sep}%').replace('$$', f'${sep}$')
901130bb 1080
1081 # outtmpl should be expand_path'ed before template dict substitution
1082 # because meta fields may contain env variables we don't want to
62b58c09 1083 # be expanded. E.g. for outtmpl "%(title)s.%(ext)s" and
901130bb 1084 # title "Hello $PATH", we don't want `$PATH` to be expanded.
1085 return expand_path(outtmpl).replace(sep, '')
1086
1087 @staticmethod
1088 def escape_outtmpl(outtmpl):
1089 ''' Escape any remaining strings like %s, %abc% etc. '''
1090 return re.sub(
1091 STR_FORMAT_RE_TMPL.format('', '(?![%(\0])'),
1092 lambda mobj: ('' if mobj.group('has_key') else '%') + mobj.group(0),
1093 outtmpl)
1094
1095 @classmethod
1096 def validate_outtmpl(cls, outtmpl):
76a264ac 1097 ''' @return None or Exception object '''
7d1eb38a 1098 outtmpl = re.sub(
47cdc68e 1099 STR_FORMAT_RE_TMPL.format('[^)]*', '[ljhqBUDS]'),
7d1eb38a 1100 lambda mobj: f'{mobj.group(0)[:-1]}s',
1101 cls._outtmpl_expandpath(outtmpl))
76a264ac 1102 try:
7d1eb38a 1103 cls.escape_outtmpl(outtmpl) % collections.defaultdict(int)
76a264ac 1104 return None
1105 except ValueError as err:
1106 return err
1107
03b4de72 1108 @staticmethod
1109 def _copy_infodict(info_dict):
1110 info_dict = dict(info_dict)
09b49e1f 1111 info_dict.pop('__postprocessors', None)
415f8d51 1112 info_dict.pop('__pending_error', None)
03b4de72 1113 return info_dict
1114
e0fd9573 1115 def prepare_outtmpl(self, outtmpl, info_dict, sanitize=False):
1116 """ Make the outtmpl and info_dict suitable for substitution: ydl.escape_outtmpl(outtmpl) % info_dict
1117 @param sanitize Whether to sanitize the output as a filename.
1118 For backward compatibility, a function can also be passed
1119 """
1120
6e84b215 1121 info_dict.setdefault('epoch', int(time.time())) # keep epoch consistent once set
143db31d 1122
03b4de72 1123 info_dict = self._copy_infodict(info_dict)
752cda38 1124 info_dict['duration_string'] = ( # %(duration>%H-%M-%S)s is wrong if duration > 24hrs
53c18592 1125 formatSeconds(info_dict['duration'], '-' if sanitize else ':')
143db31d 1126 if info_dict.get('duration', None) is not None
1127 else None)
1d485a1a 1128 info_dict['autonumber'] = int(self.params.get('autonumber_start', 1) - 1 + self._num_downloads)
9c906919 1129 info_dict['video_autonumber'] = self._num_videos
752cda38 1130 if info_dict.get('resolution') is None:
1131 info_dict['resolution'] = self.format_resolution(info_dict, default=None)
143db31d 1132
e6f21b3d 1133 # For fields playlist_index, playlist_autonumber and autonumber convert all occurrences
143db31d 1134 # of %(field)s to %(field)0Nd for backward compatibility
1135 field_size_compat_map = {
0a5a191a 1136 'playlist_index': number_of_digits(info_dict.get('__last_playlist_index') or 0),
ec11a9f4 1137 'playlist_autonumber': number_of_digits(info_dict.get('n_entries') or 0),
752cda38 1138 'autonumber': self.params.get('autonumber_size') or 5,
143db31d 1139 }
752cda38 1140
385a27fa 1141 TMPL_DICT = {}
47cdc68e 1142 EXTERNAL_FORMAT_RE = re.compile(STR_FORMAT_RE_TMPL.format('[^)]*', f'[{STR_FORMAT_TYPES}ljhqBUDS]'))
385a27fa 1143 MATH_FUNCTIONS = {
1144 '+': float.__add__,
1145 '-': float.__sub__,
1146 }
e625be0d 1147 # Field is of the form key1.key2...
07a1250e 1148 # where keys (except first) can be string, int, slice or "{field, ...}"
1149 FIELD_INNER_RE = r'(?:\w+|%(num)s|%(num)s?(?::%(num)s?){1,2})' % {'num': r'(?:-?\d+)'}
1150 FIELD_RE = r'\w*(?:\.(?:%(inner)s|{%(field)s(?:,%(field)s)*}))*' % {
1151 'inner': FIELD_INNER_RE,
1152 'field': rf'\w*(?:\.{FIELD_INNER_RE})*'
1153 }
1d485a1a 1154 MATH_FIELD_RE = rf'(?:{FIELD_RE}|-?{NUMBER_RE})'
385a27fa 1155 MATH_OPERATORS_RE = r'(?:%s)' % '|'.join(map(re.escape, MATH_FUNCTIONS.keys()))
1d485a1a 1156 INTERNAL_FORMAT_RE = re.compile(rf'''(?x)
e625be0d 1157 (?P<negate>-)?
1d485a1a 1158 (?P<fields>{FIELD_RE})
1159 (?P<maths>(?:{MATH_OPERATORS_RE}{MATH_FIELD_RE})*)
e625be0d 1160 (?:>(?P<strf_format>.+?))?
34baa9fd 1161 (?P<remaining>
1162 (?P<alternate>(?<!\\),[^|&)]+)?
1163 (?:&(?P<replacement>.*?))?
1164 (?:\|(?P<default>.*?))?
1d485a1a 1165 )$''')
752cda38 1166
07a1250e 1167 def _traverse_infodict(fields):
1168 fields = [f for x in re.split(r'\.({.+?})\.?', fields)
1169 for f in ([x] if x.startswith('{') else x.split('.'))]
1170 for i in (0, -1):
1171 if fields and not fields[i]:
1172 fields.pop(i)
1173
1174 for i, f in enumerate(fields):
1175 if not f.startswith('{'):
1176 continue
1177 assert f.endswith('}'), f'No closing brace for {f} in {fields}'
1178 fields[i] = {k: k.split('.') for k in f[1:-1].split(',')}
1179
1180 return traverse_obj(info_dict, fields, is_user_input=True, traverse_string=True)
76a264ac 1181
752cda38 1182 def get_value(mdict):
1183 # Object traversal
2b8a2973 1184 value = _traverse_infodict(mdict['fields'])
752cda38 1185 # Negative
1186 if mdict['negate']:
1187 value = float_or_none(value)
1188 if value is not None:
1189 value *= -1
1190 # Do maths
385a27fa 1191 offset_key = mdict['maths']
1192 if offset_key:
752cda38 1193 value = float_or_none(value)
1194 operator = None
385a27fa 1195 while offset_key:
1196 item = re.match(
1197 MATH_FIELD_RE if operator else MATH_OPERATORS_RE,
1198 offset_key).group(0)
1199 offset_key = offset_key[len(item):]
1200 if operator is None:
752cda38 1201 operator = MATH_FUNCTIONS[item]
385a27fa 1202 continue
1203 item, multiplier = (item[1:], -1) if item[0] == '-' else (item, 1)
1204 offset = float_or_none(item)
1205 if offset is None:
2b8a2973 1206 offset = float_or_none(_traverse_infodict(item))
385a27fa 1207 try:
1208 value = operator(value, multiplier * offset)
1209 except (TypeError, ZeroDivisionError):
1210 return None
1211 operator = None
752cda38 1212 # Datetime formatting
1213 if mdict['strf_format']:
7c37ff97 1214 value = strftime_or_none(value, mdict['strf_format'].replace('\\,', ','))
752cda38 1215
a6bcaf71 1216 # XXX: Workaround for https://github.com/yt-dlp/yt-dlp/issues/4485
1217 if sanitize and value == '':
1218 value = None
752cda38 1219 return value
1220
b868936c 1221 na = self.params.get('outtmpl_na_placeholder', 'NA')
1222
e0fd9573 1223 def filename_sanitizer(key, value, restricted=self.params.get('restrictfilenames')):
5c3895ff 1224 return sanitize_filename(str(value), restricted=restricted, is_id=(
1225 bool(re.search(r'(^|[_.])id(\.|$)', key))
8a82af35 1226 if 'filename-sanitization' in self.params['compat_opts']
5c3895ff 1227 else NO_DEFAULT))
e0fd9573 1228
1229 sanitizer = sanitize if callable(sanitize) else filename_sanitizer
1230 sanitize = bool(sanitize)
1231
6e84b215 1232 def _dumpjson_default(obj):
1233 if isinstance(obj, (set, LazyList)):
1234 return list(obj)
adbc4ec4 1235 return repr(obj)
6e84b215 1236
752cda38 1237 def create_key(outer_mobj):
1238 if not outer_mobj.group('has_key'):
b836dc94 1239 return outer_mobj.group(0)
752cda38 1240 key = outer_mobj.group('key')
752cda38 1241 mobj = re.match(INTERNAL_FORMAT_RE, key)
e0fd9573 1242 initial_field = mobj.group('fields') if mobj else ''
e978789f 1243 value, replacement, default = None, None, na
7c37ff97 1244 while mobj:
e625be0d 1245 mobj = mobj.groupdict()
7c37ff97 1246 default = mobj['default'] if mobj['default'] is not None else default
752cda38 1247 value = get_value(mobj)
e978789f 1248 replacement = mobj['replacement']
7c37ff97 1249 if value is None and mobj['alternate']:
34baa9fd 1250 mobj = re.match(INTERNAL_FORMAT_RE, mobj['remaining'][1:])
7c37ff97 1251 else:
1252 break
752cda38 1253
b868936c 1254 fmt = outer_mobj.group('format')
752cda38 1255 if fmt == 's' and value is not None and key in field_size_compat_map.keys():
86e5f3ed 1256 fmt = f'0{field_size_compat_map[key]:d}d'
752cda38 1257
e978789f 1258 value = default if value is None else value if replacement is None else replacement
752cda38 1259
4476d2c7 1260 flags = outer_mobj.group('conversion') or ''
7d1eb38a 1261 str_fmt = f'{fmt[:-1]}s'
524e2e4f 1262 if fmt[-1] == 'l': # list
4476d2c7 1263 delim = '\n' if '#' in flags else ', '
9e907ebd 1264 value, fmt = delim.join(map(str, variadic(value, allowed_types=(str, bytes)))), str_fmt
524e2e4f 1265 elif fmt[-1] == 'j': # json
deae7c17 1266 value, fmt = json.dumps(
1267 value, default=_dumpjson_default,
9b9dad11 1268 indent=4 if '#' in flags else None, ensure_ascii='+' not in flags), str_fmt
47cdc68e 1269 elif fmt[-1] == 'h': # html
deae7c17 1270 value, fmt = escapeHTML(str(value)), str_fmt
524e2e4f 1271 elif fmt[-1] == 'q': # quoted
4476d2c7 1272 value = map(str, variadic(value) if '#' in flags else [value])
1273 value, fmt = ' '.join(map(compat_shlex_quote, value)), str_fmt
524e2e4f 1274 elif fmt[-1] == 'B': # bytes
0f06bcd7 1275 value = f'%{str_fmt}'.encode() % str(value).encode()
f5aa5cfb 1276 value, fmt = value.decode('utf-8', 'ignore'), 's'
524e2e4f 1277 elif fmt[-1] == 'U': # unicode normalized
524e2e4f 1278 value, fmt = unicodedata.normalize(
1279 # "+" = compatibility equivalence, "#" = NFD
4476d2c7 1280 'NF%s%s' % ('K' if '+' in flags else '', 'D' if '#' in flags else 'C'),
524e2e4f 1281 value), str_fmt
e0fd9573 1282 elif fmt[-1] == 'D': # decimal suffix
abbeeebc 1283 num_fmt, fmt = fmt[:-1].replace('#', ''), 's'
1284 value = format_decimal_suffix(value, f'%{num_fmt}f%s' if num_fmt else '%d%s',
1285 factor=1024 if '#' in flags else 1000)
37893bb0 1286 elif fmt[-1] == 'S': # filename sanitization
e0fd9573 1287 value, fmt = filename_sanitizer(initial_field, value, restricted='#' in flags), str_fmt
7d1eb38a 1288 elif fmt[-1] == 'c':
524e2e4f 1289 if value:
1290 value = str(value)[0]
76a264ac 1291 else:
524e2e4f 1292 fmt = str_fmt
76a264ac 1293 elif fmt[-1] not in 'rs': # numeric
a439a3a4 1294 value = float_or_none(value)
752cda38 1295 if value is None:
1296 value, fmt = default, 's'
901130bb 1297
752cda38 1298 if sanitize:
1299 if fmt[-1] == 'r':
1300 # If value is an object, sanitize might convert it to a string
1301 # So we convert it to repr first
7d1eb38a 1302 value, fmt = repr(value), str_fmt
639f1cea 1303 if fmt[-1] in 'csr':
e0fd9573 1304 value = sanitizer(initial_field, value)
901130bb 1305
b868936c 1306 key = '%s\0%s' % (key.replace('%', '%\0'), outer_mobj.group('format'))
385a27fa 1307 TMPL_DICT[key] = value
b868936c 1308 return '{prefix}%({key}){fmt}'.format(key=key, fmt=fmt, prefix=outer_mobj.group('prefix'))
752cda38 1309
385a27fa 1310 return EXTERNAL_FORMAT_RE.sub(create_key, outtmpl), TMPL_DICT
143db31d 1311
819e0531 1312 def evaluate_outtmpl(self, outtmpl, info_dict, *args, **kwargs):
1313 outtmpl, info_dict = self.prepare_outtmpl(outtmpl, info_dict, *args, **kwargs)
1314 return self.escape_outtmpl(outtmpl) % info_dict
1315
5127e92a 1316 def _prepare_filename(self, info_dict, *, outtmpl=None, tmpl_type=None):
1317 assert None in (outtmpl, tmpl_type), 'outtmpl and tmpl_type are mutually exclusive'
1318 if outtmpl is None:
bf1824b3 1319 outtmpl = self.params['outtmpl'].get(tmpl_type or 'default', self.params['outtmpl']['default'])
8222d8de 1320 try:
5127e92a 1321 outtmpl = self._outtmpl_expandpath(outtmpl)
e0fd9573 1322 filename = self.evaluate_outtmpl(outtmpl, info_dict, True)
6a0546e3 1323 if not filename:
1324 return None
15da37c7 1325
5127e92a 1326 if tmpl_type in ('', 'temp'):
6a0546e3 1327 final_ext, ext = self.params.get('final_ext'), info_dict.get('ext')
1328 if final_ext and ext and final_ext != ext and filename.endswith(f'.{final_ext}'):
1329 filename = replace_extension(filename, ext, final_ext)
5127e92a 1330 elif tmpl_type:
6a0546e3 1331 force_ext = OUTTMPL_TYPES[tmpl_type]
1332 if force_ext:
1333 filename = replace_extension(filename, force_ext, info_dict.get('ext'))
de6000d9 1334
bdc3fd2f
U
1335 # https://github.com/blackjack4494/youtube-dlc/issues/85
1336 trim_file_name = self.params.get('trim_file_name', False)
1337 if trim_file_name:
5c22c63d 1338 no_ext, *ext = filename.rsplit('.', 2)
1339 filename = join_nonempty(no_ext[:trim_file_name], *ext, delim='.')
bdc3fd2f 1340
0202b52a 1341 return filename
8222d8de 1342 except ValueError as err:
6febd1c1 1343 self.report_error('Error in output template: ' + str(err) + ' (encoding: ' + repr(preferredencoding()) + ')')
8222d8de
JMF
1344 return None
1345
5127e92a 1346 def prepare_filename(self, info_dict, dir_type='', *, outtmpl=None, warn=False):
1347 """Generate the output filename"""
1348 if outtmpl:
1349 assert not dir_type, 'outtmpl and dir_type are mutually exclusive'
1350 dir_type = None
1351 filename = self._prepare_filename(info_dict, tmpl_type=dir_type, outtmpl=outtmpl)
80c03fa9 1352 if not filename and dir_type not in ('', 'temp'):
1353 return ''
de6000d9 1354
c84aeac6 1355 if warn:
21cd8fae 1356 if not self.params.get('paths'):
de6000d9 1357 pass
1358 elif filename == '-':
c84aeac6 1359 self.report_warning('--paths is ignored when an outputting to stdout', only_once=True)
de6000d9 1360 elif os.path.isabs(filename):
c84aeac6 1361 self.report_warning('--paths is ignored since an absolute path is given in output template', only_once=True)
de6000d9 1362 if filename == '-' or not filename:
1363 return filename
1364
21cd8fae 1365 return self.get_output_path(dir_type, filename)
0202b52a 1366
120fe513 1367 def _match_entry(self, info_dict, incomplete=False, silent=False):
6368e2e6 1368 """Returns None if the file should be downloaded"""
d7b460d0 1369 _type = info_dict.get('_type', 'video')
1370 assert incomplete or _type == 'video', 'Only video result can be considered complete'
8222d8de 1371
3bec830a 1372 video_title = info_dict.get('title', info_dict.get('id', 'entry'))
c77495e3 1373
8b0d7497 1374 def check_filter():
d7b460d0 1375 if _type in ('playlist', 'multi_video'):
1376 return
1377 elif _type in ('url', 'url_transparent') and not try_call(
1378 lambda: self.get_info_extractor(info_dict['ie_key']).is_single_video(info_dict['url'])):
1379 return
1380
8b0d7497 1381 if 'title' in info_dict:
1382 # This can happen when we're just evaluating the playlist
1383 title = info_dict['title']
1384 matchtitle = self.params.get('matchtitle', False)
1385 if matchtitle:
1386 if not re.search(matchtitle, title, re.IGNORECASE):
1387 return '"' + title + '" title did not match pattern "' + matchtitle + '"'
1388 rejecttitle = self.params.get('rejecttitle', False)
1389 if rejecttitle:
1390 if re.search(rejecttitle, title, re.IGNORECASE):
1391 return '"' + title + '" title matched reject pattern "' + rejecttitle + '"'
6368e2e6 1392
8b0d7497 1393 date = info_dict.get('upload_date')
1394 if date is not None:
1395 dateRange = self.params.get('daterange', DateRange())
1396 if date not in dateRange:
86e5f3ed 1397 return f'{date_from_str(date).isoformat()} upload date is not in range {dateRange}'
8b0d7497 1398 view_count = info_dict.get('view_count')
1399 if view_count is not None:
1400 min_views = self.params.get('min_views')
1401 if min_views is not None and view_count < min_views:
1402 return 'Skipping %s, because it has not reached minimum view count (%d/%d)' % (video_title, view_count, min_views)
1403 max_views = self.params.get('max_views')
1404 if max_views is not None and view_count > max_views:
1405 return 'Skipping %s, because it has exceeded the maximum view count (%d/%d)' % (video_title, view_count, max_views)
1406 if age_restricted(info_dict.get('age_limit'), self.params.get('age_limit')):
1407 return 'Skipping "%s" because it is age restricted' % video_title
8b0d7497 1408
8f18aca8 1409 match_filter = self.params.get('match_filter')
1410 if match_filter is not None:
1411 try:
1412 ret = match_filter(info_dict, incomplete=incomplete)
1413 except TypeError:
1414 # For backward compatibility
1415 ret = None if incomplete else match_filter(info_dict)
492272fe 1416 if ret is NO_DEFAULT:
1417 while True:
1418 filename = self._format_screen(self.prepare_filename(info_dict), self.Styles.FILENAME)
1419 reply = input(self._format_screen(
1420 f'Download "{filename}"? (Y/n): ', self.Styles.EMPHASIS)).lower().strip()
1421 if reply in {'y', ''}:
1422 return None
1423 elif reply == 'n':
1424 return f'Skipping {video_title}'
492272fe 1425 elif ret is not None:
8f18aca8 1426 return ret
8b0d7497 1427 return None
1428
c77495e3 1429 if self.in_download_archive(info_dict):
1430 reason = '%s has already been recorded in the archive' % video_title
1431 break_opt, break_err = 'break_on_existing', ExistingVideoReached
1432 else:
1433 reason = check_filter()
1434 break_opt, break_err = 'break_on_reject', RejectedVideoReached
8b0d7497 1435 if reason is not None:
120fe513 1436 if not silent:
1437 self.to_screen('[download] ' + reason)
c77495e3 1438 if self.params.get(break_opt, False):
1439 raise break_err()
8b0d7497 1440 return reason
fe7e0c98 1441
b6c45014
JMF
1442 @staticmethod
1443 def add_extra_info(info_dict, extra_info):
1444 '''Set the keys from extra_info in info dict if they are missing'''
1445 for key, value in extra_info.items():
1446 info_dict.setdefault(key, value)
1447
409e1828 1448 def extract_info(self, url, download=True, ie_key=None, extra_info=None,
61aa5ba3 1449 process=True, force_generic_extractor=False):
41d1cca3 1450 """
17ffed18 1451 Extract and return the information dictionary of the URL
41d1cca3 1452
1453 Arguments:
17ffed18 1454 @param url URL to extract
41d1cca3 1455
1456 Keyword arguments:
17ffed18 1457 @param download Whether to download videos
1458 @param process Whether to resolve all unresolved references (URLs, playlist items).
1459 Must be True for download to work
1460 @param ie_key Use only the extractor with this key
1461
1462 @param extra_info Dictionary containing the extra values to add to the info (For internal use only)
1463 @force_generic_extractor Force using the generic extractor (Deprecated; use ie_key='Generic')
41d1cca3 1464 """
fe7e0c98 1465
409e1828 1466 if extra_info is None:
1467 extra_info = {}
1468
61aa5ba3 1469 if not ie_key and force_generic_extractor:
d22dec74
S
1470 ie_key = 'Generic'
1471
8222d8de 1472 if ie_key:
fe7866d0 1473 ies = {ie_key: self._ies[ie_key]} if ie_key in self._ies else {}
8222d8de
JMF
1474 else:
1475 ies = self._ies
1476
fe7866d0 1477 for key, ie in ies.items():
8222d8de
JMF
1478 if not ie.suitable(url):
1479 continue
1480
1481 if not ie.working():
6febd1c1
PH
1482 self.report_warning('The program functionality for this site has been marked as broken, '
1483 'and will probably not work.')
8222d8de 1484
1151c407 1485 temp_id = ie.get_temp_id(url)
fe7866d0 1486 if temp_id is not None and self.in_download_archive({'id': temp_id, 'ie_key': key}):
1487 self.to_screen(f'[{key}] {temp_id}: has already been recorded in the archive')
5e5be0c0 1488 if self.params.get('break_on_existing', False):
1489 raise ExistingVideoReached()
a0566bbf 1490 break
fe7866d0 1491 return self.__extract_info(url, self.get_info_extractor(key), download, extra_info, process)
a0566bbf 1492 else:
fe7866d0 1493 extractors_restricted = self.params.get('allowed_extractors') not in (None, ['default'])
1494 self.report_error(f'No suitable extractor{format_field(ie_key, None, " (%s)")} found for URL {url}',
1495 tb=False if extractors_restricted else None)
a0566bbf 1496
7e88d7d7 1497 def _handle_extraction_exceptions(func):
b5ae35ee 1498 @functools.wraps(func)
a0566bbf 1499 def wrapper(self, *args, **kwargs):
6da22e7d 1500 while True:
1501 try:
1502 return func(self, *args, **kwargs)
1503 except (DownloadCancelled, LazyList.IndexError, PagedList.IndexError):
8222d8de 1504 raise
6da22e7d 1505 except ReExtractInfo as e:
1506 if e.expected:
1507 self.to_screen(f'{e}; Re-extracting data')
1508 else:
1509 self.to_stderr('\r')
1510 self.report_warning(f'{e}; Re-extracting data')
1511 continue
1512 except GeoRestrictedError as e:
1513 msg = e.msg
1514 if e.countries:
1515 msg += '\nThis video is available in %s.' % ', '.join(
1516 map(ISO3166Utils.short2full, e.countries))
1517 msg += '\nYou might want to use a VPN or a proxy server (with --proxy) to workaround.'
1518 self.report_error(msg)
1519 except ExtractorError as e: # An error we somewhat expected
1520 self.report_error(str(e), e.format_traceback())
1521 except Exception as e:
1522 if self.params.get('ignoreerrors'):
1523 self.report_error(str(e), tb=encode_compat_str(traceback.format_exc()))
1524 else:
1525 raise
1526 break
a0566bbf 1527 return wrapper
1528
693f0600 1529 def _wait_for_video(self, ie_result={}):
f2ebc5c7 1530 if (not self.params.get('wait_for_video')
1531 or ie_result.get('_type', 'video') != 'video'
1532 or ie_result.get('formats') or ie_result.get('url')):
1533 return
1534
1535 format_dur = lambda dur: '%02d:%02d:%02d' % timetuple_from_msec(dur * 1000)[:-1]
1536 last_msg = ''
1537
1538 def progress(msg):
1539 nonlocal last_msg
a7dc6a89 1540 full_msg = f'{msg}\n'
1541 if not self.params.get('noprogress'):
1542 full_msg = msg + ' ' * (len(last_msg) - len(msg)) + '\r'
1543 elif last_msg:
1544 return
1545 self.to_screen(full_msg, skip_eol=True)
f2ebc5c7 1546 last_msg = msg
1547
1548 min_wait, max_wait = self.params.get('wait_for_video')
1549 diff = try_get(ie_result, lambda x: x['release_timestamp'] - time.time())
1550 if diff is None and ie_result.get('live_status') == 'is_upcoming':
16c620bc 1551 diff = round(random.uniform(min_wait, max_wait) if (max_wait and min_wait) else (max_wait or min_wait), 0)
f2ebc5c7 1552 self.report_warning('Release time of video is not known')
693f0600 1553 elif ie_result and (diff or 0) <= 0:
f2ebc5c7 1554 self.report_warning('Video should already be available according to extracted info')
38d79fd1 1555 diff = min(max(diff or 0, min_wait or 0), max_wait or float('inf'))
f2ebc5c7 1556 self.to_screen(f'[wait] Waiting for {format_dur(diff)} - Press Ctrl+C to try now')
1557
1558 wait_till = time.time() + diff
1559 try:
1560 while True:
1561 diff = wait_till - time.time()
1562 if diff <= 0:
1563 progress('')
1564 raise ReExtractInfo('[wait] Wait period ended', expected=True)
1565 progress(f'[wait] Remaining time until next attempt: {self._format_screen(format_dur(diff), self.Styles.EMPHASIS)}')
1566 time.sleep(1)
1567 except KeyboardInterrupt:
1568 progress('')
1569 raise ReExtractInfo('[wait] Interrupted by user', expected=True)
1570 except BaseException as e:
1571 if not isinstance(e, ReExtractInfo):
1572 self.to_screen('')
1573 raise
1574
7e88d7d7 1575 @_handle_extraction_exceptions
58f197b7 1576 def __extract_info(self, url, ie, download, extra_info, process):
693f0600 1577 try:
1578 ie_result = ie.extract(url)
1579 except UserNotLive as e:
1580 if process:
1581 if self.params.get('wait_for_video'):
1582 self.report_warning(e)
1583 self._wait_for_video()
1584 raise
a0566bbf 1585 if ie_result is None: # Finished already (backwards compatibility; listformats and friends should be moved here)
cb794ee0 1586 self.report_warning(f'Extractor {ie.IE_NAME} returned nothing{bug_reports_message()}')
a0566bbf 1587 return
1588 if isinstance(ie_result, list):
1589 # Backwards compatibility: old IE result format
1590 ie_result = {
1591 '_type': 'compat_list',
1592 'entries': ie_result,
1593 }
e37d0efb 1594 if extra_info.get('original_url'):
1595 ie_result.setdefault('original_url', extra_info['original_url'])
a0566bbf 1596 self.add_default_extra_info(ie_result, ie, url)
1597 if process:
f2ebc5c7 1598 self._wait_for_video(ie_result)
a0566bbf 1599 return self.process_ie_result(ie_result, download, extra_info)
8222d8de 1600 else:
a0566bbf 1601 return ie_result
fe7e0c98 1602
ea38e55f 1603 def add_default_extra_info(self, ie_result, ie, url):
6033d980 1604 if url is not None:
1605 self.add_extra_info(ie_result, {
1606 'webpage_url': url,
1607 'original_url': url,
57ebfca3 1608 })
1609 webpage_url = ie_result.get('webpage_url')
1610 if webpage_url:
1611 self.add_extra_info(ie_result, {
1612 'webpage_url_basename': url_basename(webpage_url),
1613 'webpage_url_domain': get_domain(webpage_url),
6033d980 1614 })
1615 if ie is not None:
1616 self.add_extra_info(ie_result, {
1617 'extractor': ie.IE_NAME,
1618 'extractor_key': ie.ie_key(),
1619 })
ea38e55f 1620
58adec46 1621 def process_ie_result(self, ie_result, download=True, extra_info=None):
8222d8de
JMF
1622 """
1623 Take the result of the ie(may be modified) and resolve all unresolved
1624 references (URLs, playlist items).
1625
1626 It will also download the videos if 'download'.
1627 Returns the resolved ie_result.
1628 """
58adec46 1629 if extra_info is None:
1630 extra_info = {}
e8ee972c
PH
1631 result_type = ie_result.get('_type', 'video')
1632
057a5206 1633 if result_type in ('url', 'url_transparent'):
8f97a15d 1634 ie_result['url'] = sanitize_url(
1635 ie_result['url'], scheme='http' if self.params.get('prefer_insecure') else 'https')
8791e78c 1636 if ie_result.get('original_url') and not extra_info.get('original_url'):
1637 extra_info = {'original_url': ie_result['original_url'], **extra_info}
e37d0efb 1638
057a5206 1639 extract_flat = self.params.get('extract_flat', False)
3089bc74
S
1640 if ((extract_flat == 'in_playlist' and 'playlist' in extra_info)
1641 or extract_flat is True):
ecb54191 1642 info_copy = ie_result.copy()
6033d980 1643 ie = try_get(ie_result.get('ie_key'), self.get_info_extractor)
360167b9 1644 if ie and not ie_result.get('id'):
4614bc22 1645 info_copy['id'] = ie.get_temp_id(ie_result['url'])
6033d980 1646 self.add_default_extra_info(info_copy, ie, ie_result['url'])
4614bc22 1647 self.add_extra_info(info_copy, extra_info)
b5475f11 1648 info_copy, _ = self.pre_process(info_copy)
94dc8604 1649 self._fill_common_fields(info_copy, False)
ecb54191 1650 self.__forced_printings(info_copy, self.prepare_filename(info_copy), incomplete=True)
415f8d51 1651 self._raise_pending_errors(info_copy)
4614bc22 1652 if self.params.get('force_write_download_archive', False):
1653 self.record_download_archive(info_copy)
e8ee972c
PH
1654 return ie_result
1655
8222d8de 1656 if result_type == 'video':
b6c45014 1657 self.add_extra_info(ie_result, extra_info)
9c2b75b5 1658 ie_result = self.process_video_result(ie_result, download=download)
415f8d51 1659 self._raise_pending_errors(ie_result)
28b0eb0f 1660 additional_urls = (ie_result or {}).get('additional_urls')
9c2b75b5 1661 if additional_urls:
e9f4ccd1 1662 # TODO: Improve MetadataParserPP to allow setting a list
14f25df2 1663 if isinstance(additional_urls, str):
9c2b75b5 1664 additional_urls = [additional_urls]
1665 self.to_screen(
1666 '[info] %s: %d additional URL(s) requested' % (ie_result['id'], len(additional_urls)))
1667 self.write_debug('Additional URLs: "%s"' % '", "'.join(additional_urls))
1668 ie_result['additional_entries'] = [
1669 self.extract_info(
b69fd25c 1670 url, download, extra_info=extra_info,
9c2b75b5 1671 force_generic_extractor=self.params.get('force_generic_extractor'))
1672 for url in additional_urls
1673 ]
1674 return ie_result
8222d8de
JMF
1675 elif result_type == 'url':
1676 # We have to add extra_info to the results because it may be
1677 # contained in a playlist
07cce701 1678 return self.extract_info(
1679 ie_result['url'], download,
1680 ie_key=ie_result.get('ie_key'),
1681 extra_info=extra_info)
7fc3fa05
PH
1682 elif result_type == 'url_transparent':
1683 # Use the information from the embedding page
1684 info = self.extract_info(
1685 ie_result['url'], ie_key=ie_result.get('ie_key'),
1686 extra_info=extra_info, download=False, process=False)
1687
1640eb09
S
1688 # extract_info may return None when ignoreerrors is enabled and
1689 # extraction failed with an error, don't crash and return early
1690 # in this case
1691 if not info:
1692 return info
1693
3975b4d2 1694 exempted_fields = {'_type', 'url', 'ie_key'}
1695 if not ie_result.get('section_end') and ie_result.get('section_start') is None:
1696 # For video clips, the id etc of the clip extractor should be used
1697 exempted_fields |= {'id', 'extractor', 'extractor_key'}
1698
412c617d 1699 new_result = info.copy()
3975b4d2 1700 new_result.update(filter_dict(ie_result, lambda k, v: v is not None and k not in exempted_fields))
7fc3fa05 1701
0563f7ac
S
1702 # Extracted info may not be a video result (i.e.
1703 # info.get('_type', 'video') != video) but rather an url or
1704 # url_transparent. In such cases outer metadata (from ie_result)
1705 # should be propagated to inner one (info). For this to happen
1706 # _type of info should be overridden with url_transparent. This
067aa17e 1707 # fixes issue from https://github.com/ytdl-org/youtube-dl/pull/11163.
0563f7ac
S
1708 if new_result.get('_type') == 'url':
1709 new_result['_type'] = 'url_transparent'
7fc3fa05
PH
1710
1711 return self.process_ie_result(
1712 new_result, download=download, extra_info=extra_info)
40fcba5e 1713 elif result_type in ('playlist', 'multi_video'):
30a074c2 1714 # Protect from infinite recursion due to recursively nested playlists
1715 # (see https://github.com/ytdl-org/youtube-dl/issues/27833)
0bd5a039 1716 webpage_url = ie_result.get('webpage_url') # Playlists maynot have webpage_url
1717 if webpage_url and webpage_url in self._playlist_urls:
7e85e872 1718 self.to_screen(
30a074c2 1719 '[download] Skipping already downloaded playlist: %s'
1720 % ie_result.get('title') or ie_result.get('id'))
1721 return
7e85e872 1722
30a074c2 1723 self._playlist_level += 1
1724 self._playlist_urls.add(webpage_url)
03f83004 1725 self._fill_common_fields(ie_result, False)
bc516a3f 1726 self._sanitize_thumbnails(ie_result)
30a074c2 1727 try:
1728 return self.__process_playlist(ie_result, download)
1729 finally:
1730 self._playlist_level -= 1
1731 if not self._playlist_level:
1732 self._playlist_urls.clear()
8222d8de 1733 elif result_type == 'compat_list':
c9bf4114
PH
1734 self.report_warning(
1735 'Extractor %s returned a compat_list result. '
1736 'It needs to be updated.' % ie_result.get('extractor'))
5f6a1245 1737
8222d8de 1738 def _fixup(r):
b868936c 1739 self.add_extra_info(r, {
1740 'extractor': ie_result['extractor'],
1741 'webpage_url': ie_result['webpage_url'],
1742 'webpage_url_basename': url_basename(ie_result['webpage_url']),
0bb322b9 1743 'webpage_url_domain': get_domain(ie_result['webpage_url']),
b868936c 1744 'extractor_key': ie_result['extractor_key'],
1745 })
8222d8de
JMF
1746 return r
1747 ie_result['entries'] = [
b6c45014 1748 self.process_ie_result(_fixup(r), download, extra_info)
8222d8de
JMF
1749 for r in ie_result['entries']
1750 ]
1751 return ie_result
1752 else:
1753 raise Exception('Invalid result type: %s' % result_type)
1754
e92caff5 1755 def _ensure_dir_exists(self, path):
1756 return make_dir(path, self.report_error)
1757
3b603dbd 1758 @staticmethod
3bec830a 1759 def _playlist_infodict(ie_result, strict=False, **kwargs):
1760 info = {
1761 'playlist_count': ie_result.get('playlist_count'),
3b603dbd 1762 'playlist': ie_result.get('title') or ie_result.get('id'),
1763 'playlist_id': ie_result.get('id'),
1764 'playlist_title': ie_result.get('title'),
1765 'playlist_uploader': ie_result.get('uploader'),
1766 'playlist_uploader_id': ie_result.get('uploader_id'),
3b603dbd 1767 **kwargs,
1768 }
3bec830a 1769 if strict:
1770 return info
0bd5a039 1771 if ie_result.get('webpage_url'):
1772 info.update({
1773 'webpage_url': ie_result['webpage_url'],
1774 'webpage_url_basename': url_basename(ie_result['webpage_url']),
1775 'webpage_url_domain': get_domain(ie_result['webpage_url']),
1776 })
3bec830a 1777 return {
1778 **info,
1779 'playlist_index': 0,
59d7de0d 1780 '__last_playlist_index': max(ie_result.get('requested_entries') or (0, 0)),
3bec830a 1781 'extractor': ie_result['extractor'],
3bec830a 1782 'extractor_key': ie_result['extractor_key'],
1783 }
3b603dbd 1784
30a074c2 1785 def __process_playlist(self, ie_result, download):
7e88d7d7 1786 """Process each entry in the playlist"""
f5ea4748 1787 assert ie_result['_type'] in ('playlist', 'multi_video')
1788
3bec830a 1789 common_info = self._playlist_infodict(ie_result, strict=True)
3955b207 1790 title = common_info.get('playlist') or '<Untitled>'
3bec830a 1791 if self._match_entry(common_info, incomplete=True) is not None:
1792 return
c6e07cf1 1793 self.to_screen(f'[download] Downloading {ie_result["_type"]}: {title}')
f0d785d3 1794
7e88d7d7 1795 all_entries = PlaylistEntries(self, ie_result)
7e9a6125 1796 entries = orderedSet(all_entries.get_requested_items(), lazy=True)
1797
1798 lazy = self.params.get('lazy_playlist')
1799 if lazy:
1800 resolved_entries, n_entries = [], 'N/A'
1801 ie_result['requested_entries'], ie_result['entries'] = None, None
1802 else:
1803 entries = resolved_entries = list(entries)
1804 n_entries = len(resolved_entries)
1805 ie_result['requested_entries'], ie_result['entries'] = tuple(zip(*resolved_entries)) or ([], [])
1806 if not ie_result.get('playlist_count'):
1807 # Better to do this after potentially exhausting entries
1808 ie_result['playlist_count'] = all_entries.get_full_count()
498f5606 1809
0647d925 1810 extra = self._playlist_infodict(ie_result, n_entries=int_or_none(n_entries))
1811 ie_copy = collections.ChainMap(ie_result, extra)
3bec830a 1812
e08a85d8 1813 _infojson_written = False
0bfc53d0 1814 write_playlist_files = self.params.get('allow_playlist_files', True)
1815 if write_playlist_files and self.params.get('list_thumbnails'):
1816 self.list_thumbnails(ie_result)
1817 if write_playlist_files and not self.params.get('simulate'):
e08a85d8 1818 _infojson_written = self._write_info_json(
1819 'playlist', ie_result, self.prepare_filename(ie_copy, 'pl_infojson'))
1820 if _infojson_written is None:
80c03fa9 1821 return
1822 if self._write_description('playlist', ie_result,
1823 self.prepare_filename(ie_copy, 'pl_description')) is None:
1824 return
681de68e 1825 # TODO: This should be passed to ThumbnailsConvertor if necessary
3bec830a 1826 self._write_thumbnails('playlist', ie_result, self.prepare_filename(ie_copy, 'pl_thumbnail'))
30a074c2 1827
7e9a6125 1828 if lazy:
1829 if self.params.get('playlistreverse') or self.params.get('playlistrandom'):
1830 self.report_warning('playlistreverse and playlistrandom are not supported with lazy_playlist', only_once=True)
1831 elif self.params.get('playlistreverse'):
1832 entries.reverse()
1833 elif self.params.get('playlistrandom'):
30a074c2 1834 random.shuffle(entries)
1835
bc5c2f8a 1836 self.to_screen(f'[{ie_result["extractor"]}] Playlist {title}: Downloading {n_entries} items'
7e88d7d7 1837 f'{format_field(ie_result, "playlist_count", " of %s")}')
30a074c2 1838
134c913c 1839 keep_resolved_entries = self.params.get('extract_flat') != 'discard'
1840 if self.params.get('extract_flat') == 'discard_in_playlist':
1841 keep_resolved_entries = ie_result['_type'] != 'playlist'
1842 if keep_resolved_entries:
1843 self.write_debug('The information of all playlist entries will be held in memory')
1844
26e2805c 1845 failures = 0
1846 max_failures = self.params.get('skip_playlist_after_errors') or float('inf')
7e9a6125 1847 for i, (playlist_index, entry) in enumerate(entries):
1848 if lazy:
1849 resolved_entries.append((playlist_index, entry))
3bec830a 1850 if not entry:
7e88d7d7 1851 continue
1852
7e88d7d7 1853 entry['__x_forwarded_for_ip'] = ie_result.get('__x_forwarded_for_ip')
7e9a6125 1854 if not lazy and 'playlist-index' in self.params.get('compat_opts', []):
1855 playlist_index = ie_result['requested_entries'][i]
1856
0647d925 1857 entry_copy = collections.ChainMap(entry, {
3bec830a 1858 **common_info,
3955b207 1859 'n_entries': int_or_none(n_entries),
71729754 1860 'playlist_index': playlist_index,
7e9a6125 1861 'playlist_autonumber': i + 1,
0647d925 1862 })
3bec830a 1863
0647d925 1864 if self._match_entry(entry_copy, incomplete=True) is not None:
f0ad6f8c 1865 # For compatabilty with youtube-dl. See https://github.com/yt-dlp/yt-dlp/issues/4369
1866 resolved_entries[i] = (playlist_index, NO_DEFAULT)
3bec830a 1867 continue
1868
bc5c2f8a 1869 self.to_screen('[download] Downloading item %s of %s' % (
3bec830a 1870 self._format_screen(i + 1, self.Styles.ID), self._format_screen(n_entries, self.Styles.EMPHASIS)))
1871
ec54bd43 1872 entry_result = self.__process_iterable_entry(entry, download, collections.ChainMap({
a6ca61d4 1873 'playlist_index': playlist_index,
1874 'playlist_autonumber': i + 1,
ec54bd43 1875 }, extra))
26e2805c 1876 if not entry_result:
1877 failures += 1
1878 if failures >= max_failures:
1879 self.report_error(
7e88d7d7 1880 f'Skipping the remaining entries in playlist "{title}" since {failures} items failed extraction')
26e2805c 1881 break
134c913c 1882 if keep_resolved_entries:
1883 resolved_entries[i] = (playlist_index, entry_result)
7e88d7d7 1884
1885 # Update with processed data
f0ad6f8c 1886 ie_result['entries'] = [e for _, e in resolved_entries if e is not NO_DEFAULT]
bc5c2f8a 1887 ie_result['requested_entries'] = [i for i, e in resolved_entries if e is not NO_DEFAULT]
1888 if ie_result['requested_entries'] == try_call(lambda: list(range(1, ie_result['playlist_count'] + 1))):
1889 # Do not set for full playlist
1890 ie_result.pop('requested_entries')
e08a85d8 1891
1892 # Write the updated info to json
cb96c5be 1893 if _infojson_written is True and self._write_info_json(
e08a85d8 1894 'updated playlist', ie_result,
1895 self.prepare_filename(ie_copy, 'pl_infojson'), overwrite=True) is None:
1896 return
ca30f449 1897
ed5835b4 1898 ie_result = self.run_all_pps('playlist', ie_result)
7e88d7d7 1899 self.to_screen(f'[download] Finished downloading playlist: {title}')
30a074c2 1900 return ie_result
1901
7e88d7d7 1902 @_handle_extraction_exceptions
a0566bbf 1903 def __process_iterable_entry(self, entry, download, extra_info):
1904 return self.process_ie_result(
1905 entry, download=download, extra_info=extra_info)
1906
67134eab
JMF
1907 def _build_format_filter(self, filter_spec):
1908 " Returns a function to filter the formats according to the filter_spec "
083c9df9
PH
1909
1910 OPERATORS = {
1911 '<': operator.lt,
1912 '<=': operator.le,
1913 '>': operator.gt,
1914 '>=': operator.ge,
1915 '=': operator.eq,
1916 '!=': operator.ne,
1917 }
67134eab 1918 operator_rex = re.compile(r'''(?x)\s*
187986a8 1919 (?P<key>width|height|tbr|abr|vbr|asr|filesize|filesize_approx|fps)\s*
1920 (?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
1921 (?P<value>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?)\s*
083c9df9 1922 ''' % '|'.join(map(re.escape, OPERATORS.keys())))
187986a8 1923 m = operator_rex.fullmatch(filter_spec)
9ddb6925
S
1924 if m:
1925 try:
1926 comparison_value = int(m.group('value'))
1927 except ValueError:
1928 comparison_value = parse_filesize(m.group('value'))
1929 if comparison_value is None:
1930 comparison_value = parse_filesize(m.group('value') + 'B')
1931 if comparison_value is None:
1932 raise ValueError(
1933 'Invalid value %r in format specification %r' % (
67134eab 1934 m.group('value'), filter_spec))
9ddb6925
S
1935 op = OPERATORS[m.group('op')]
1936
083c9df9 1937 if not m:
9ddb6925
S
1938 STR_OPERATORS = {
1939 '=': operator.eq,
10d33b34
YCH
1940 '^=': lambda attr, value: attr.startswith(value),
1941 '$=': lambda attr, value: attr.endswith(value),
1942 '*=': lambda attr, value: value in attr,
1ce9a3cb 1943 '~=': lambda attr, value: value.search(attr) is not None
9ddb6925 1944 }
187986a8 1945 str_operator_rex = re.compile(r'''(?x)\s*
1946 (?P<key>[a-zA-Z0-9._-]+)\s*
1ce9a3cb
LF
1947 (?P<negation>!\s*)?(?P<op>%s)\s*(?P<none_inclusive>\?\s*)?
1948 (?P<quote>["'])?
1949 (?P<value>(?(quote)(?:(?!(?P=quote))[^\\]|\\.)+|[\w.-]+))
1950 (?(quote)(?P=quote))\s*
9ddb6925 1951 ''' % '|'.join(map(re.escape, STR_OPERATORS.keys())))
187986a8 1952 m = str_operator_rex.fullmatch(filter_spec)
9ddb6925 1953 if m:
1ce9a3cb
LF
1954 if m.group('op') == '~=':
1955 comparison_value = re.compile(m.group('value'))
1956 else:
1957 comparison_value = re.sub(r'''\\([\\"'])''', r'\1', m.group('value'))
2cc779f4
S
1958 str_op = STR_OPERATORS[m.group('op')]
1959 if m.group('negation'):
e118a879 1960 op = lambda attr, value: not str_op(attr, value)
2cc779f4
S
1961 else:
1962 op = str_op
083c9df9 1963
9ddb6925 1964 if not m:
187986a8 1965 raise SyntaxError('Invalid filter specification %r' % filter_spec)
083c9df9
PH
1966
1967 def _filter(f):
1968 actual_value = f.get(m.group('key'))
1969 if actual_value is None:
1970 return m.group('none_inclusive')
1971 return op(actual_value, comparison_value)
67134eab
JMF
1972 return _filter
1973
9f1a1c36 1974 def _check_formats(self, formats):
1975 for f in formats:
1976 self.to_screen('[info] Testing format %s' % f['format_id'])
75689fe5 1977 path = self.get_output_path('temp')
1978 if not self._ensure_dir_exists(f'{path}/'):
1979 continue
1980 temp_file = tempfile.NamedTemporaryFile(suffix='.tmp', delete=False, dir=path or None)
9f1a1c36 1981 temp_file.close()
1982 try:
1983 success, _ = self.dl(temp_file.name, f, test=True)
8a82af35 1984 except (DownloadError, OSError, ValueError) + network_exceptions:
9f1a1c36 1985 success = False
1986 finally:
1987 if os.path.exists(temp_file.name):
1988 try:
1989 os.remove(temp_file.name)
1990 except OSError:
1991 self.report_warning('Unable to delete temporary file "%s"' % temp_file.name)
1992 if success:
1993 yield f
1994 else:
1995 self.to_screen('[info] Unable to download format %s. Skipping...' % f['format_id'])
1996
0017d9ad 1997 def _default_format_spec(self, info_dict, download=True):
0017d9ad 1998
af0f7428
S
1999 def can_merge():
2000 merger = FFmpegMergerPP(self)
2001 return merger.available and merger.can_merge()
2002
91ebc640 2003 prefer_best = (
b7b04c78 2004 not self.params.get('simulate')
91ebc640 2005 and download
2006 and (
2007 not can_merge()
21633673 2008 or info_dict.get('is_live') and not self.params.get('live_from_start')
bf1824b3 2009 or self.params['outtmpl']['default'] == '-'))
53ed7066 2010 compat = (
2011 prefer_best
2012 or self.params.get('allow_multiple_audio_streams', False)
8a82af35 2013 or 'format-spec' in self.params['compat_opts'])
91ebc640 2014
2015 return (
53ed7066 2016 'best/bestvideo+bestaudio' if prefer_best
2017 else 'bestvideo*+bestaudio/best' if not compat
91ebc640 2018 else 'bestvideo+bestaudio/best')
0017d9ad 2019
67134eab
JMF
2020 def build_format_selector(self, format_spec):
2021 def syntax_error(note, start):
2022 message = (
2023 'Invalid format specification: '
86e5f3ed 2024 '{}\n\t{}\n\t{}^'.format(note, format_spec, ' ' * start[1]))
67134eab
JMF
2025 return SyntaxError(message)
2026
2027 PICKFIRST = 'PICKFIRST'
2028 MERGE = 'MERGE'
2029 SINGLE = 'SINGLE'
0130afb7 2030 GROUP = 'GROUP'
67134eab
JMF
2031 FormatSelector = collections.namedtuple('FormatSelector', ['type', 'selector', 'filters'])
2032
91ebc640 2033 allow_multiple_streams = {'audio': self.params.get('allow_multiple_audio_streams', False),
2034 'video': self.params.get('allow_multiple_video_streams', False)}
909d24dd 2035
9f1a1c36 2036 check_formats = self.params.get('check_formats') == 'selected'
e8e73840 2037
67134eab
JMF
2038 def _parse_filter(tokens):
2039 filter_parts = []
2040 for type, string, start, _, _ in tokens:
2041 if type == tokenize.OP and string == ']':
2042 return ''.join(filter_parts)
2043 else:
2044 filter_parts.append(string)
2045
232541df 2046 def _remove_unused_ops(tokens):
62b58c09
L
2047 # Remove operators that we don't use and join them with the surrounding strings.
2048 # E.g. 'mp4' '-' 'baseline' '-' '16x9' is converted to 'mp4-baseline-16x9'
232541df
JMF
2049 ALLOWED_OPS = ('/', '+', ',', '(', ')')
2050 last_string, last_start, last_end, last_line = None, None, None, None
2051 for type, string, start, end, line in tokens:
2052 if type == tokenize.OP and string == '[':
2053 if last_string:
2054 yield tokenize.NAME, last_string, last_start, last_end, last_line
2055 last_string = None
2056 yield type, string, start, end, line
2057 # everything inside brackets will be handled by _parse_filter
2058 for type, string, start, end, line in tokens:
2059 yield type, string, start, end, line
2060 if type == tokenize.OP and string == ']':
2061 break
2062 elif type == tokenize.OP and string in ALLOWED_OPS:
2063 if last_string:
2064 yield tokenize.NAME, last_string, last_start, last_end, last_line
2065 last_string = None
2066 yield type, string, start, end, line
2067 elif type in [tokenize.NAME, tokenize.NUMBER, tokenize.OP]:
2068 if not last_string:
2069 last_string = string
2070 last_start = start
2071 last_end = end
2072 else:
2073 last_string += string
2074 if last_string:
2075 yield tokenize.NAME, last_string, last_start, last_end, last_line
2076
cf2ac6df 2077 def _parse_format_selection(tokens, inside_merge=False, inside_choice=False, inside_group=False):
67134eab
JMF
2078 selectors = []
2079 current_selector = None
2080 for type, string, start, _, _ in tokens:
2081 # ENCODING is only defined in python 3.x
2082 if type == getattr(tokenize, 'ENCODING', None):
2083 continue
2084 elif type in [tokenize.NAME, tokenize.NUMBER]:
2085 current_selector = FormatSelector(SINGLE, string, [])
2086 elif type == tokenize.OP:
cf2ac6df
JMF
2087 if string == ')':
2088 if not inside_group:
2089 # ')' will be handled by the parentheses group
2090 tokens.restore_last_token()
67134eab 2091 break
cf2ac6df 2092 elif inside_merge and string in ['/', ',']:
0130afb7
JMF
2093 tokens.restore_last_token()
2094 break
cf2ac6df
JMF
2095 elif inside_choice and string == ',':
2096 tokens.restore_last_token()
2097 break
2098 elif string == ',':
0a31a350
JMF
2099 if not current_selector:
2100 raise syntax_error('"," must follow a format selector', start)
67134eab
JMF
2101 selectors.append(current_selector)
2102 current_selector = None
2103 elif string == '/':
d96d604e
JMF
2104 if not current_selector:
2105 raise syntax_error('"/" must follow a format selector', start)
67134eab 2106 first_choice = current_selector
cf2ac6df 2107 second_choice = _parse_format_selection(tokens, inside_choice=True)
f5f4a27a 2108 current_selector = FormatSelector(PICKFIRST, (first_choice, second_choice), [])
67134eab
JMF
2109 elif string == '[':
2110 if not current_selector:
2111 current_selector = FormatSelector(SINGLE, 'best', [])
2112 format_filter = _parse_filter(tokens)
2113 current_selector.filters.append(format_filter)
0130afb7
JMF
2114 elif string == '(':
2115 if current_selector:
2116 raise syntax_error('Unexpected "("', start)
cf2ac6df
JMF
2117 group = _parse_format_selection(tokens, inside_group=True)
2118 current_selector = FormatSelector(GROUP, group, [])
67134eab 2119 elif string == '+':
d03cfdce 2120 if not current_selector:
2121 raise syntax_error('Unexpected "+"', start)
2122 selector_1 = current_selector
2123 selector_2 = _parse_format_selection(tokens, inside_merge=True)
2124 if not selector_2:
2125 raise syntax_error('Expected a selector', start)
2126 current_selector = FormatSelector(MERGE, (selector_1, selector_2), [])
67134eab 2127 else:
86e5f3ed 2128 raise syntax_error(f'Operator not recognized: "{string}"', start)
67134eab
JMF
2129 elif type == tokenize.ENDMARKER:
2130 break
2131 if current_selector:
2132 selectors.append(current_selector)
2133 return selectors
2134
f8d4ad9a 2135 def _merge(formats_pair):
2136 format_1, format_2 = formats_pair
2137
2138 formats_info = []
2139 formats_info.extend(format_1.get('requested_formats', (format_1,)))
2140 formats_info.extend(format_2.get('requested_formats', (format_2,)))
2141
2142 if not allow_multiple_streams['video'] or not allow_multiple_streams['audio']:
551f9388 2143 get_no_more = {'video': False, 'audio': False}
f8d4ad9a 2144 for (i, fmt_info) in enumerate(formats_info):
551f9388 2145 if fmt_info.get('acodec') == fmt_info.get('vcodec') == 'none':
2146 formats_info.pop(i)
2147 continue
2148 for aud_vid in ['audio', 'video']:
f8d4ad9a 2149 if not allow_multiple_streams[aud_vid] and fmt_info.get(aud_vid[0] + 'codec') != 'none':
2150 if get_no_more[aud_vid]:
2151 formats_info.pop(i)
f5510afe 2152 break
f8d4ad9a 2153 get_no_more[aud_vid] = True
2154
2155 if len(formats_info) == 1:
2156 return formats_info[0]
2157
2158 video_fmts = [fmt_info for fmt_info in formats_info if fmt_info.get('vcodec') != 'none']
2159 audio_fmts = [fmt_info for fmt_info in formats_info if fmt_info.get('acodec') != 'none']
2160
2161 the_only_video = video_fmts[0] if len(video_fmts) == 1 else None
2162 the_only_audio = audio_fmts[0] if len(audio_fmts) == 1 else None
2163
fc61aff4
LL
2164 output_ext = get_compatible_ext(
2165 vcodecs=[f.get('vcodec') for f in video_fmts],
2166 acodecs=[f.get('acodec') for f in audio_fmts],
2167 vexts=[f['ext'] for f in video_fmts],
2168 aexts=[f['ext'] for f in audio_fmts],
2169 preferences=(try_call(lambda: self.params['merge_output_format'].split('/'))
2170 or self.params.get('prefer_free_formats') and ('webm', 'mkv')))
f8d4ad9a 2171
975a0d0d 2172 filtered = lambda *keys: filter(None, (traverse_obj(fmt, *keys) for fmt in formats_info))
2173
f8d4ad9a 2174 new_dict = {
2175 'requested_formats': formats_info,
975a0d0d 2176 'format': '+'.join(filtered('format')),
2177 'format_id': '+'.join(filtered('format_id')),
f8d4ad9a 2178 'ext': output_ext,
975a0d0d 2179 'protocol': '+'.join(map(determine_protocol, formats_info)),
093a1710 2180 'language': '+'.join(orderedSet(filtered('language'))) or None,
2181 'format_note': '+'.join(orderedSet(filtered('format_note'))) or None,
2182 'filesize_approx': sum(filtered('filesize', 'filesize_approx')) or None,
975a0d0d 2183 'tbr': sum(filtered('tbr', 'vbr', 'abr')),
f8d4ad9a 2184 }
2185
2186 if the_only_video:
2187 new_dict.update({
2188 'width': the_only_video.get('width'),
2189 'height': the_only_video.get('height'),
2190 'resolution': the_only_video.get('resolution') or self.format_resolution(the_only_video),
2191 'fps': the_only_video.get('fps'),
49a57e70 2192 'dynamic_range': the_only_video.get('dynamic_range'),
f8d4ad9a 2193 'vcodec': the_only_video.get('vcodec'),
2194 'vbr': the_only_video.get('vbr'),
2195 'stretched_ratio': the_only_video.get('stretched_ratio'),
105bfd90 2196 'aspect_ratio': the_only_video.get('aspect_ratio'),
f8d4ad9a 2197 })
2198
2199 if the_only_audio:
2200 new_dict.update({
2201 'acodec': the_only_audio.get('acodec'),
2202 'abr': the_only_audio.get('abr'),
975a0d0d 2203 'asr': the_only_audio.get('asr'),
b8ed0f15 2204 'audio_channels': the_only_audio.get('audio_channels')
f8d4ad9a 2205 })
2206
2207 return new_dict
2208
e8e73840 2209 def _check_formats(formats):
981052c9 2210 if not check_formats:
2211 yield from formats
b5ac45b1 2212 return
9f1a1c36 2213 yield from self._check_formats(formats)
e8e73840 2214
67134eab 2215 def _build_selector_function(selector):
909d24dd 2216 if isinstance(selector, list): # ,
67134eab
JMF
2217 fs = [_build_selector_function(s) for s in selector]
2218
317f7ab6 2219 def selector_function(ctx):
67134eab 2220 for f in fs:
981052c9 2221 yield from f(ctx)
67134eab 2222 return selector_function
909d24dd 2223
2224 elif selector.type == GROUP: # ()
0130afb7 2225 selector_function = _build_selector_function(selector.selector)
909d24dd 2226
2227 elif selector.type == PICKFIRST: # /
67134eab
JMF
2228 fs = [_build_selector_function(s) for s in selector.selector]
2229
317f7ab6 2230 def selector_function(ctx):
67134eab 2231 for f in fs:
317f7ab6 2232 picked_formats = list(f(ctx))
67134eab
JMF
2233 if picked_formats:
2234 return picked_formats
2235 return []
67134eab 2236
981052c9 2237 elif selector.type == MERGE: # +
2238 selector_1, selector_2 = map(_build_selector_function, selector.selector)
2239
2240 def selector_function(ctx):
adbc4ec4 2241 for pair in itertools.product(selector_1(ctx), selector_2(ctx)):
981052c9 2242 yield _merge(pair)
2243
909d24dd 2244 elif selector.type == SINGLE: # atom
598d185d 2245 format_spec = selector.selector or 'best'
909d24dd 2246
f8d4ad9a 2247 # TODO: Add allvideo, allaudio etc by generalizing the code with best/worst selector
909d24dd 2248 if format_spec == 'all':
2249 def selector_function(ctx):
9222c381 2250 yield from _check_formats(ctx['formats'][::-1])
f8d4ad9a 2251 elif format_spec == 'mergeall':
2252 def selector_function(ctx):
316f2650 2253 formats = list(_check_formats(
2254 f for f in ctx['formats'] if f.get('vcodec') != 'none' or f.get('acodec') != 'none'))
e01d6aa4 2255 if not formats:
2256 return
921b76ca 2257 merged_format = formats[-1]
2258 for f in formats[-2::-1]:
f8d4ad9a 2259 merged_format = _merge((merged_format, f))
2260 yield merged_format
909d24dd 2261
2262 else:
85e801a9 2263 format_fallback, seperate_fallback, format_reverse, format_idx = False, None, True, 1
eff63539 2264 mobj = re.match(
2265 r'(?P<bw>best|worst|b|w)(?P<type>video|audio|v|a)?(?P<mod>\*)?(?:\.(?P<n>[1-9]\d*))?$',
2266 format_spec)
2267 if mobj is not None:
2268 format_idx = int_or_none(mobj.group('n'), default=1)
e8e73840 2269 format_reverse = mobj.group('bw')[0] == 'b'
eff63539 2270 format_type = (mobj.group('type') or [None])[0]
2271 not_format_type = {'v': 'a', 'a': 'v'}.get(format_type)
2272 format_modified = mobj.group('mod') is not None
909d24dd 2273
2274 format_fallback = not format_type and not format_modified # for b, w
8326b00a 2275 _filter_f = (
eff63539 2276 (lambda f: f.get('%scodec' % format_type) != 'none')
2277 if format_type and format_modified # bv*, ba*, wv*, wa*
2278 else (lambda f: f.get('%scodec' % not_format_type) == 'none')
2279 if format_type # bv, ba, wv, wa
2280 else (lambda f: f.get('vcodec') != 'none' and f.get('acodec') != 'none')
2281 if not format_modified # b, w
8326b00a 2282 else lambda f: True) # b*, w*
2283 filter_f = lambda f: _filter_f(f) and (
2284 f.get('vcodec') != 'none' or f.get('acodec') != 'none')
67134eab 2285 else:
48ee10ee 2286 if format_spec in self._format_selection_exts['audio']:
b11c04a8 2287 filter_f = lambda f: f.get('ext') == format_spec and f.get('acodec') != 'none'
48ee10ee 2288 elif format_spec in self._format_selection_exts['video']:
b11c04a8 2289 filter_f = lambda f: f.get('ext') == format_spec and f.get('acodec') != 'none' and f.get('vcodec') != 'none'
85e801a9 2290 seperate_fallback = lambda f: f.get('ext') == format_spec and f.get('vcodec') != 'none'
48ee10ee 2291 elif format_spec in self._format_selection_exts['storyboards']:
b11c04a8 2292 filter_f = lambda f: f.get('ext') == format_spec and f.get('acodec') == 'none' and f.get('vcodec') == 'none'
2293 else:
b5ae35ee 2294 filter_f = lambda f: f.get('format_id') == format_spec # id
909d24dd 2295
2296 def selector_function(ctx):
2297 formats = list(ctx['formats'])
909d24dd 2298 matches = list(filter(filter_f, formats)) if filter_f is not None else formats
85e801a9 2299 if not matches:
2300 if format_fallback and ctx['incomplete_formats']:
2301 # for extractors with incomplete formats (audio only (soundcloud)
2302 # or video only (imgur)) best/worst will fallback to
2303 # best/worst {video,audio}-only format
2304 matches = formats
2305 elif seperate_fallback and not ctx['has_merged_format']:
2306 # for compatibility with youtube-dl when there is no pre-merged format
2307 matches = list(filter(seperate_fallback, formats))
981052c9 2308 matches = LazyList(_check_formats(matches[::-1 if format_reverse else 1]))
2309 try:
e8e73840 2310 yield matches[format_idx - 1]
4abea8ca 2311 except LazyList.IndexError:
981052c9 2312 return
083c9df9 2313
67134eab 2314 filters = [self._build_format_filter(f) for f in selector.filters]
083c9df9 2315
317f7ab6 2316 def final_selector(ctx):
adbc4ec4 2317 ctx_copy = dict(ctx)
67134eab 2318 for _filter in filters:
317f7ab6
S
2319 ctx_copy['formats'] = list(filter(_filter, ctx_copy['formats']))
2320 return selector_function(ctx_copy)
67134eab 2321 return final_selector
083c9df9 2322
0f06bcd7 2323 stream = io.BytesIO(format_spec.encode())
0130afb7 2324 try:
f9934b96 2325 tokens = list(_remove_unused_ops(tokenize.tokenize(stream.readline)))
0130afb7
JMF
2326 except tokenize.TokenError:
2327 raise syntax_error('Missing closing/opening brackets or parenthesis', (0, len(format_spec)))
2328
86e5f3ed 2329 class TokenIterator:
0130afb7
JMF
2330 def __init__(self, tokens):
2331 self.tokens = tokens
2332 self.counter = 0
2333
2334 def __iter__(self):
2335 return self
2336
2337 def __next__(self):
2338 if self.counter >= len(self.tokens):
2339 raise StopIteration()
2340 value = self.tokens[self.counter]
2341 self.counter += 1
2342 return value
2343
2344 next = __next__
2345
2346 def restore_last_token(self):
2347 self.counter -= 1
2348
2349 parsed_selector = _parse_format_selection(iter(TokenIterator(tokens)))
67134eab 2350 return _build_selector_function(parsed_selector)
a9c58ad9 2351
e5660ee6 2352 def _calc_headers(self, info_dict):
8b7539d2 2353 res = merge_headers(self.params['http_headers'], info_dict.get('http_headers') or {})
e5660ee6 2354
c487cf00 2355 cookies = self._calc_cookies(info_dict['url'])
e5660ee6
JMF
2356 if cookies:
2357 res['Cookie'] = cookies
2358
0016b84e
S
2359 if 'X-Forwarded-For' not in res:
2360 x_forwarded_for_ip = info_dict.get('__x_forwarded_for_ip')
2361 if x_forwarded_for_ip:
2362 res['X-Forwarded-For'] = x_forwarded_for_ip
2363
e5660ee6
JMF
2364 return res
2365
c487cf00 2366 def _calc_cookies(self, url):
2367 pr = sanitized_Request(url)
e5660ee6 2368 self.cookiejar.add_cookie_header(pr)
662435f7 2369 return pr.get_header('Cookie')
e5660ee6 2370
9f1a1c36 2371 def _sort_thumbnails(self, thumbnails):
2372 thumbnails.sort(key=lambda t: (
2373 t.get('preference') if t.get('preference') is not None else -1,
2374 t.get('width') if t.get('width') is not None else -1,
2375 t.get('height') if t.get('height') is not None else -1,
2376 t.get('id') if t.get('id') is not None else '',
2377 t.get('url')))
2378
b0249bca 2379 def _sanitize_thumbnails(self, info_dict):
bc516a3f 2380 thumbnails = info_dict.get('thumbnails')
2381 if thumbnails is None:
2382 thumbnail = info_dict.get('thumbnail')
2383 if thumbnail:
2384 info_dict['thumbnails'] = thumbnails = [{'url': thumbnail}]
9f1a1c36 2385 if not thumbnails:
2386 return
2387
2388 def check_thumbnails(thumbnails):
2389 for t in thumbnails:
2390 self.to_screen(f'[info] Testing thumbnail {t["id"]}')
2391 try:
2392 self.urlopen(HEADRequest(t['url']))
2393 except network_exceptions as err:
2394 self.to_screen(f'[info] Unable to connect to thumbnail {t["id"]} URL {t["url"]!r} - {err}. Skipping...')
2395 continue
2396 yield t
2397
2398 self._sort_thumbnails(thumbnails)
2399 for i, t in enumerate(thumbnails):
2400 if t.get('id') is None:
2401 t['id'] = '%d' % i
2402 if t.get('width') and t.get('height'):
2403 t['resolution'] = '%dx%d' % (t['width'], t['height'])
2404 t['url'] = sanitize_url(t['url'])
2405
2406 if self.params.get('check_formats') is True:
282f5709 2407 info_dict['thumbnails'] = LazyList(check_thumbnails(thumbnails[::-1]), reverse=True)
9f1a1c36 2408 else:
2409 info_dict['thumbnails'] = thumbnails
bc516a3f 2410
94dc8604 2411 def _fill_common_fields(self, info_dict, final=True):
03f83004 2412 # TODO: move sanitization here
94dc8604 2413 if final:
7aefd19a 2414 title = info_dict['fulltitle'] = info_dict.get('title')
d4736fdb 2415 if not title:
2416 if title == '':
2417 self.write_debug('Extractor gave empty title. Creating a generic title')
2418 else:
2419 self.report_warning('Extractor failed to obtain "title". Creating a generic title instead')
1d485a1a 2420 info_dict['title'] = f'{info_dict["extractor"].replace(":", "-")} video #{info_dict["id"]}'
03f83004
LNO
2421
2422 if info_dict.get('duration') is not None:
2423 info_dict['duration_string'] = formatSeconds(info_dict['duration'])
2424
2425 for ts_key, date_key in (
2426 ('timestamp', 'upload_date'),
2427 ('release_timestamp', 'release_date'),
2428 ('modified_timestamp', 'modified_date'),
2429 ):
2430 if info_dict.get(date_key) is None and info_dict.get(ts_key) is not None:
2431 # Working around out-of-range timestamp values (e.g. negative ones on Windows,
2432 # see http://bugs.python.org/issue1646728)
19a03940 2433 with contextlib.suppress(ValueError, OverflowError, OSError):
03f83004
LNO
2434 upload_date = datetime.datetime.utcfromtimestamp(info_dict[ts_key])
2435 info_dict[date_key] = upload_date.strftime('%Y%m%d')
03f83004
LNO
2436
2437 live_keys = ('is_live', 'was_live')
2438 live_status = info_dict.get('live_status')
2439 if live_status is None:
2440 for key in live_keys:
2441 if info_dict.get(key) is False:
2442 continue
2443 if info_dict.get(key):
2444 live_status = key
2445 break
2446 if all(info_dict.get(key) is False for key in live_keys):
2447 live_status = 'not_live'
2448 if live_status:
2449 info_dict['live_status'] = live_status
2450 for key in live_keys:
2451 if info_dict.get(key) is None:
2452 info_dict[key] = (live_status == key)
a057779d 2453 if live_status == 'post_live':
2454 info_dict['was_live'] = True
03f83004
LNO
2455
2456 # Auto generate title fields corresponding to the *_number fields when missing
2457 # in order to always have clean titles. This is very common for TV series.
2458 for field in ('chapter', 'season', 'episode'):
94dc8604 2459 if final and info_dict.get('%s_number' % field) is not None and not info_dict.get(field):
03f83004
LNO
2460 info_dict[field] = '%s %d' % (field.capitalize(), info_dict['%s_number' % field])
2461
415f8d51 2462 def _raise_pending_errors(self, info):
2463 err = info.pop('__pending_error', None)
2464 if err:
2465 self.report_error(err, tb=False)
2466
784320c9 2467 def sort_formats(self, info_dict):
2468 formats = self._get_formats(info_dict)
2469 if not formats:
2470 return
2471 # Backward compatibility with InfoExtractor._sort_formats
2472 field_preference = formats[0].pop('__sort_fields', None)
2473 if field_preference:
2474 info_dict['_format_sort_fields'] = field_preference
2475
2476 formats.sort(key=FormatSorter(
2477 self, info_dict.get('_format_sort_fields', [])).calculate_preference)
2478
dd82ffea
JMF
2479 def process_video_result(self, info_dict, download=True):
2480 assert info_dict.get('_type', 'video') == 'video'
9c906919 2481 self._num_videos += 1
dd82ffea 2482
bec1fad2 2483 if 'id' not in info_dict:
fc08bdd6 2484 raise ExtractorError('Missing "id" field in extractor result', ie=info_dict['extractor'])
2485 elif not info_dict.get('id'):
2486 raise ExtractorError('Extractor failed to obtain "id"', ie=info_dict['extractor'])
455a15e2 2487
c9969434
S
2488 def report_force_conversion(field, field_not, conversion):
2489 self.report_warning(
2490 '"%s" field is not %s - forcing %s conversion, there is an error in extractor'
2491 % (field, field_not, conversion))
2492
2493 def sanitize_string_field(info, string_field):
2494 field = info.get(string_field)
14f25df2 2495 if field is None or isinstance(field, str):
c9969434
S
2496 return
2497 report_force_conversion(string_field, 'a string', 'string')
14f25df2 2498 info[string_field] = str(field)
c9969434
S
2499
2500 def sanitize_numeric_fields(info):
2501 for numeric_field in self._NUMERIC_FIELDS:
2502 field = info.get(numeric_field)
f9934b96 2503 if field is None or isinstance(field, (int, float)):
c9969434
S
2504 continue
2505 report_force_conversion(numeric_field, 'numeric', 'int')
2506 info[numeric_field] = int_or_none(field)
2507
2508 sanitize_string_field(info_dict, 'id')
2509 sanitize_numeric_fields(info_dict)
3975b4d2 2510 if info_dict.get('section_end') and info_dict.get('section_start') is not None:
2511 info_dict['duration'] = round(info_dict['section_end'] - info_dict['section_start'], 3)
4c3f8c3f 2512 if (info_dict.get('duration') or 0) <= 0 and info_dict.pop('duration', None):
50e93e03 2513 self.report_warning('"duration" field is negative, there is an error in extractor')
be6217b2 2514
9eef7c4e 2515 chapters = info_dict.get('chapters') or []
a3976e07 2516 if chapters and chapters[0].get('start_time'):
2517 chapters.insert(0, {'start_time': 0})
2518
9eef7c4e 2519 dummy_chapter = {'end_time': 0, 'start_time': info_dict.get('duration')}
a3976e07 2520 for idx, (prev, current, next_) in enumerate(zip(
2521 (dummy_chapter, *chapters), chapters, (*chapters[1:], dummy_chapter)), 1):
9eef7c4e 2522 if current.get('start_time') is None:
2523 current['start_time'] = prev.get('end_time')
2524 if not current.get('end_time'):
2525 current['end_time'] = next_.get('start_time')
a3976e07 2526 if not current.get('title'):
2527 current['title'] = f'<Untitled Chapter {idx}>'
9eef7c4e 2528
dd82ffea
JMF
2529 if 'playlist' not in info_dict:
2530 # It isn't part of a playlist
2531 info_dict['playlist'] = None
2532 info_dict['playlist_index'] = None
2533
bc516a3f 2534 self._sanitize_thumbnails(info_dict)
d5519808 2535
536a55da 2536 thumbnail = info_dict.get('thumbnail')
bc516a3f 2537 thumbnails = info_dict.get('thumbnails')
536a55da
S
2538 if thumbnail:
2539 info_dict['thumbnail'] = sanitize_url(thumbnail)
2540 elif thumbnails:
d5519808
PH
2541 info_dict['thumbnail'] = thumbnails[-1]['url']
2542
ae30b840 2543 if info_dict.get('display_id') is None and 'id' in info_dict:
0afef30b
PH
2544 info_dict['display_id'] = info_dict['id']
2545
03f83004 2546 self._fill_common_fields(info_dict)
33d2fc2f 2547
05108a49
S
2548 for cc_kind in ('subtitles', 'automatic_captions'):
2549 cc = info_dict.get(cc_kind)
2550 if cc:
2551 for _, subtitle in cc.items():
2552 for subtitle_format in subtitle:
2553 if subtitle_format.get('url'):
2554 subtitle_format['url'] = sanitize_url(subtitle_format['url'])
2555 if subtitle_format.get('ext') is None:
2556 subtitle_format['ext'] = determine_ext(subtitle_format['url']).lower()
2557
2558 automatic_captions = info_dict.get('automatic_captions')
4bba3716 2559 subtitles = info_dict.get('subtitles')
4bba3716 2560
360e1ca5 2561 info_dict['requested_subtitles'] = self.process_subtitles(
05108a49 2562 info_dict['id'], subtitles, automatic_captions)
a504ced0 2563
784320c9 2564 self.sort_formats(info_dict)
aebb4f4b 2565 formats = self._get_formats(info_dict)
dd82ffea 2566
0a5a191a 2567 # or None ensures --clean-infojson removes it
2568 info_dict['_has_drm'] = any(f.get('has_drm') for f in formats) or None
88acdbc2 2569 if not self.params.get('allow_unplayable_formats'):
2570 formats = [f for f in formats if not f.get('has_drm')]
17ffed18 2571
2572 if formats and all(f.get('acodec') == f.get('vcodec') == 'none' for f in formats):
2573 self.report_warning(
2574 f'{"This video is DRM protected and " if info_dict["_has_drm"] else ""}'
2575 'only images are available for download. Use --list-formats to see them'.capitalize())
88acdbc2 2576
319b6059 2577 get_from_start = not info_dict.get('is_live') or bool(self.params.get('live_from_start'))
2578 if not get_from_start:
2579 info_dict['title'] += ' ' + datetime.datetime.now().strftime('%Y-%m-%d %H:%M')
2580 if info_dict.get('is_live') and formats:
adbc4ec4 2581 formats = [f for f in formats if bool(f.get('is_from_start')) == get_from_start]
319b6059 2582 if get_from_start and not formats:
a44ca5a4 2583 self.raise_no_formats(info_dict, msg=(
2584 '--live-from-start is passed, but there are no formats that can be downloaded from the start. '
2585 'If you want to download from the current time, use --no-live-from-start'))
adbc4ec4 2586
73af5cc8
S
2587 def is_wellformed(f):
2588 url = f.get('url')
a5ac0c47 2589 if not url:
73af5cc8
S
2590 self.report_warning(
2591 '"url" field is missing or empty - skipping format, '
2592 'there is an error in extractor')
a5ac0c47
S
2593 return False
2594 if isinstance(url, bytes):
2595 sanitize_string_field(f, 'url')
2596 return True
73af5cc8
S
2597
2598 # Filter out malformed formats for better extraction robustness
1ac7f461 2599 formats = list(filter(is_wellformed, formats or []))
2600
2601 if not formats:
2602 self.raise_no_formats(info_dict)
73af5cc8 2603
181c7053
S
2604 formats_dict = {}
2605
dd82ffea 2606 # We check that all the formats have the format and format_id fields
db95dc13 2607 for i, format in enumerate(formats):
c9969434
S
2608 sanitize_string_field(format, 'format_id')
2609 sanitize_numeric_fields(format)
dcf77cf1 2610 format['url'] = sanitize_url(format['url'])
e74e3b63 2611 if not format.get('format_id'):
14f25df2 2612 format['format_id'] = str(i)
e2effb08
S
2613 else:
2614 # Sanitize format_id from characters used in format selector expression
ec85ded8 2615 format['format_id'] = re.sub(r'[\s,/+\[\]()]', '_', format['format_id'])
181c7053
S
2616 format_id = format['format_id']
2617 if format_id not in formats_dict:
2618 formats_dict[format_id] = []
2619 formats_dict[format_id].append(format)
2620
2621 # Make sure all formats have unique format_id
03b4de72 2622 common_exts = set(itertools.chain(*self._format_selection_exts.values()))
181c7053 2623 for format_id, ambiguous_formats in formats_dict.items():
48ee10ee 2624 ambigious_id = len(ambiguous_formats) > 1
2625 for i, format in enumerate(ambiguous_formats):
2626 if ambigious_id:
181c7053 2627 format['format_id'] = '%s-%d' % (format_id, i)
48ee10ee 2628 if format.get('ext') is None:
2629 format['ext'] = determine_ext(format['url']).lower()
2630 # Ensure there is no conflict between id and ext in format selection
2631 # See https://github.com/yt-dlp/yt-dlp/issues/1282
2632 if format['format_id'] != format['ext'] and format['format_id'] in common_exts:
2633 format['format_id'] = 'f%s' % format['format_id']
181c7053
S
2634
2635 for i, format in enumerate(formats):
8c51aa65 2636 if format.get('format') is None:
6febd1c1 2637 format['format'] = '{id} - {res}{note}'.format(
8c51aa65
JMF
2638 id=format['format_id'],
2639 res=self.format_resolution(format),
b868936c 2640 note=format_field(format, 'format_note', ' (%s)'),
8c51aa65 2641 )
6f0be937 2642 if format.get('protocol') is None:
b5559424 2643 format['protocol'] = determine_protocol(format)
239df021 2644 if format.get('resolution') is None:
2645 format['resolution'] = self.format_resolution(format, default=None)
176f1866 2646 if format.get('dynamic_range') is None and format.get('vcodec') != 'none':
2647 format['dynamic_range'] = 'SDR'
105bfd90 2648 if format.get('aspect_ratio') is None:
2649 format['aspect_ratio'] = try_call(lambda: round(format['width'] / format['height'], 2))
f2fe69c7 2650 if (info_dict.get('duration') and format.get('tbr')
2651 and not format.get('filesize') and not format.get('filesize_approx')):
56ba69e4 2652 format['filesize_approx'] = int(info_dict['duration'] * format['tbr'] * (1024 / 8))
f2fe69c7 2653
e5660ee6
JMF
2654 # Add HTTP headers, so that external programs can use them from the
2655 # json output
2656 full_format_info = info_dict.copy()
2657 full_format_info.update(format)
2658 format['http_headers'] = self._calc_headers(full_format_info)
0016b84e
S
2659 # Remove private housekeeping stuff
2660 if '__x_forwarded_for_ip' in info_dict:
2661 del info_dict['__x_forwarded_for_ip']
dd82ffea 2662
9f1a1c36 2663 if self.params.get('check_formats') is True:
282f5709 2664 formats = LazyList(self._check_formats(formats[::-1]), reverse=True)
9f1a1c36 2665
88acdbc2 2666 if not formats or formats[0] is not info_dict:
b3d9ef88
JMF
2667 # only set the 'formats' fields if the original info_dict list them
2668 # otherwise we end up with a circular reference, the first (and unique)
f89197d7 2669 # element in the 'formats' field in info_dict is info_dict itself,
dfb1b146 2670 # which can't be exported to json
b3d9ef88 2671 info_dict['formats'] = formats
4ec82a72 2672
2673 info_dict, _ = self.pre_process(info_dict)
2674
6db9c4d5 2675 if self._match_entry(info_dict, incomplete=self._format_fields) is not None:
09b49e1f 2676 return info_dict
2677
2678 self.post_extract(info_dict)
2679 info_dict, _ = self.pre_process(info_dict, 'after_filter')
2680
093a1710 2681 # The pre-processors may have modified the formats
aebb4f4b 2682 formats = self._get_formats(info_dict)
093a1710 2683
e4221b70 2684 list_only = self.params.get('simulate') == 'list_only'
fa9f30b8 2685 interactive_format_selection = not list_only and self.format_selector == '-'
b7b04c78 2686 if self.params.get('list_thumbnails'):
2687 self.list_thumbnails(info_dict)
b7b04c78 2688 if self.params.get('listsubtitles'):
2689 if 'automatic_captions' in info_dict:
2690 self.list_subtitles(
2691 info_dict['id'], automatic_captions, 'automatic captions')
2692 self.list_subtitles(info_dict['id'], subtitles, 'subtitles')
fa9f30b8 2693 if self.params.get('listformats') or interactive_format_selection:
b69fd25c 2694 self.list_formats(info_dict)
169dbde9 2695 if list_only:
b7b04c78 2696 # Without this printing, -F --print-json will not work
169dbde9 2697 self.__forced_printings(info_dict, self.prepare_filename(info_dict), incomplete=True)
c487cf00 2698 return info_dict
bfaae0a7 2699
187986a8 2700 format_selector = self.format_selector
2701 if format_selector is None:
0017d9ad 2702 req_format = self._default_format_spec(info_dict, download=download)
0760b0a7 2703 self.write_debug('Default format spec: %s' % req_format)
187986a8 2704 format_selector = self.build_format_selector(req_format)
317f7ab6 2705
fa9f30b8 2706 while True:
2707 if interactive_format_selection:
2708 req_format = input(
2709 self._format_screen('\nEnter format selector: ', self.Styles.EMPHASIS))
2710 try:
2711 format_selector = self.build_format_selector(req_format)
2712 except SyntaxError as err:
2713 self.report_error(err, tb=False, is_error=False)
2714 continue
2715
85e801a9 2716 formats_to_download = list(format_selector({
fa9f30b8 2717 'formats': formats,
85e801a9 2718 'has_merged_format': any('none' not in (f.get('acodec'), f.get('vcodec')) for f in formats),
2719 'incomplete_formats': (
2720 # All formats are video-only or
2721 all(f.get('vcodec') != 'none' and f.get('acodec') == 'none' for f in formats)
2722 # all formats are audio-only
2723 or all(f.get('vcodec') == 'none' and f.get('acodec') != 'none' for f in formats)),
2724 }))
fa9f30b8 2725 if interactive_format_selection and not formats_to_download:
2726 self.report_error('Requested format is not available', tb=False, is_error=False)
2727 continue
2728 break
317f7ab6 2729
dd82ffea 2730 if not formats_to_download:
b7da73eb 2731 if not self.params.get('ignore_no_formats_error'):
c0b6e5c7 2732 raise ExtractorError(
2733 'Requested format is not available. Use --list-formats for a list of available formats',
2734 expected=True, video_id=info_dict['id'], ie=info_dict['extractor'])
b62fa6d7 2735 self.report_warning('Requested format is not available')
2736 # Process what we can, even without any available formats.
2737 formats_to_download = [{}]
a13e6848 2738
0500ee3d 2739 requested_ranges = tuple(self.params.get('download_ranges', lambda *_: [{}])(info_dict, self))
5ec1b6b7 2740 best_format, downloaded_formats = formats_to_download[-1], []
b62fa6d7 2741 if download:
0500ee3d 2742 if best_format and requested_ranges:
5ec1b6b7 2743 def to_screen(*msg):
2744 self.to_screen(f'[info] {info_dict["id"]}: {" ".join(", ".join(variadic(m)) for m in msg)}')
2745
2746 to_screen(f'Downloading {len(formats_to_download)} format(s):',
2747 (f['format_id'] for f in formats_to_download))
0500ee3d 2748 if requested_ranges != ({}, ):
5ec1b6b7 2749 to_screen(f'Downloading {len(requested_ranges)} time ranges:',
fc2ba496 2750 (f'{c["start_time"]:.1f}-{c["end_time"]:.1f}' for c in requested_ranges))
a13e6848 2751 max_downloads_reached = False
5ec1b6b7 2752
0500ee3d 2753 for fmt, chapter in itertools.product(formats_to_download, requested_ranges):
5ec1b6b7 2754 new_info = self._copy_infodict(info_dict)
b7da73eb 2755 new_info.update(fmt)
3975b4d2 2756 offset, duration = info_dict.get('section_start') or 0, info_dict.get('duration') or float('inf')
fc2ba496 2757 end_time = offset + min(chapter.get('end_time', duration), duration)
3975b4d2 2758 if chapter or offset:
5ec1b6b7 2759 new_info.update({
3975b4d2 2760 'section_start': offset + chapter.get('start_time', 0),
2576d53a 2761 # duration may not be accurate. So allow deviations <1sec
2762 'section_end': end_time if end_time <= offset + duration + 1 else None,
5ec1b6b7 2763 'section_title': chapter.get('title'),
2764 'section_number': chapter.get('index'),
2765 })
2766 downloaded_formats.append(new_info)
a13e6848 2767 try:
2768 self.process_info(new_info)
2769 except MaxDownloadsReached:
2770 max_downloads_reached = True
415f8d51 2771 self._raise_pending_errors(new_info)
f46e2f9d 2772 # Remove copied info
2773 for key, val in tuple(new_info.items()):
2774 if info_dict.get(key) == val:
2775 new_info.pop(key)
a13e6848 2776 if max_downloads_reached:
2777 break
ebed8b37 2778
5ec1b6b7 2779 write_archive = {f.get('__write_download_archive', False) for f in downloaded_formats}
a13e6848 2780 assert write_archive.issubset({True, False, 'ignore'})
2781 if True in write_archive and False not in write_archive:
2782 self.record_download_archive(info_dict)
be72c624 2783
5ec1b6b7 2784 info_dict['requested_downloads'] = downloaded_formats
ed5835b4 2785 info_dict = self.run_all_pps('after_video', info_dict)
a13e6848 2786 if max_downloads_reached:
2787 raise MaxDownloadsReached()
ebed8b37 2788
49a57e70 2789 # We update the info dict with the selected best quality format (backwards compatibility)
be72c624 2790 info_dict.update(best_format)
dd82ffea
JMF
2791 return info_dict
2792
98c70d6f 2793 def process_subtitles(self, video_id, normal_subtitles, automatic_captions):
a504ced0 2794 """Select the requested subtitles and their format"""
d8a58ddc 2795 available_subs, normal_sub_langs = {}, []
98c70d6f
JMF
2796 if normal_subtitles and self.params.get('writesubtitles'):
2797 available_subs.update(normal_subtitles)
d8a58ddc 2798 normal_sub_langs = tuple(normal_subtitles.keys())
98c70d6f
JMF
2799 if automatic_captions and self.params.get('writeautomaticsub'):
2800 for lang, cap_info in automatic_captions.items():
360e1ca5
JMF
2801 if lang not in available_subs:
2802 available_subs[lang] = cap_info
2803
d2c8aadf 2804 if not available_subs or (
2805 not self.params.get('writesubtitles')
2806 and not self.params.get('writeautomaticsub')):
4d171848 2807 return None
a504ced0 2808
d8a58ddc 2809 all_sub_langs = tuple(available_subs.keys())
a504ced0 2810 if self.params.get('allsubtitles', False):
c32b0aab 2811 requested_langs = all_sub_langs
2812 elif self.params.get('subtitleslangs', False):
5314b521 2813 try:
2814 requested_langs = orderedSet_from_options(
2815 self.params.get('subtitleslangs'), {'all': all_sub_langs}, use_regex=True)
2816 except re.error as e:
2817 raise ValueError(f'Wrong regex for subtitlelangs: {e.pattern}')
d8a58ddc 2818 elif normal_sub_langs:
2819 requested_langs = ['en'] if 'en' in normal_sub_langs else normal_sub_langs[:1]
a504ced0 2820 else:
d8a58ddc 2821 requested_langs = ['en'] if 'en' in all_sub_langs else all_sub_langs[:1]
ad3dc496 2822 if requested_langs:
d2c8aadf 2823 self.to_screen(f'[info] {video_id}: Downloading subtitles: {", ".join(requested_langs)}')
a504ced0
JMF
2824
2825 formats_query = self.params.get('subtitlesformat', 'best')
2826 formats_preference = formats_query.split('/') if formats_query else []
2827 subs = {}
2828 for lang in requested_langs:
2829 formats = available_subs.get(lang)
2830 if formats is None:
86e5f3ed 2831 self.report_warning(f'{lang} subtitles not available for {video_id}')
a504ced0 2832 continue
a504ced0
JMF
2833 for ext in formats_preference:
2834 if ext == 'best':
2835 f = formats[-1]
2836 break
2837 matches = list(filter(lambda f: f['ext'] == ext, formats))
2838 if matches:
2839 f = matches[-1]
2840 break
2841 else:
2842 f = formats[-1]
2843 self.report_warning(
2844 'No subtitle format found matching "%s" for language %s, '
2845 'using %s' % (formats_query, lang, f['ext']))
2846 subs[lang] = f
2847 return subs
2848
bb66c247 2849 def _forceprint(self, key, info_dict):
2850 if info_dict is None:
2851 return
2852 info_copy = info_dict.copy()
2853 info_copy['formats_table'] = self.render_formats_table(info_dict)
2854 info_copy['thumbnails_table'] = self.render_thumbnails_table(info_dict)
2855 info_copy['subtitles_table'] = self.render_subtitles_table(info_dict.get('id'), info_dict.get('subtitles'))
2856 info_copy['automatic_captions_table'] = self.render_subtitles_table(info_dict.get('id'), info_dict.get('automatic_captions'))
2857
2858 def format_tmpl(tmpl):
48c8424b 2859 mobj = re.fullmatch(r'([\w.:,]|-\d|(?P<dict>{([\w.:,]|-\d)+}))+=?', tmpl)
07a1250e 2860 if not mobj:
2861 return tmpl
48c8424b 2862
2863 fmt = '%({})s'
2864 if tmpl.startswith('{'):
2865 tmpl = f'.{tmpl}'
2866 if tmpl.endswith('='):
2867 tmpl, fmt = tmpl[:-1], '{0} = %({0})#j'
2868 return '\n'.join(map(fmt.format, [tmpl] if mobj.group('dict') else tmpl.split(',')))
8130779d 2869
bb66c247 2870 for tmpl in self.params['forceprint'].get(key, []):
2871 self.to_stdout(self.evaluate_outtmpl(format_tmpl(tmpl), info_copy))
2872
2873 for tmpl, file_tmpl in self.params['print_to_file'].get(key, []):
5127e92a 2874 filename = self.prepare_filename(info_dict, outtmpl=file_tmpl)
bb66c247 2875 tmpl = format_tmpl(tmpl)
2876 self.to_screen(f'[info] Writing {tmpl!r} to: {filename}')
8d93e69d 2877 if self._ensure_dir_exists(filename):
86e5f3ed 2878 with open(filename, 'a', encoding='utf-8') as f:
8d93e69d 2879 f.write(self.evaluate_outtmpl(tmpl, info_copy) + '\n')
ca30f449 2880
d06daf23 2881 def __forced_printings(self, info_dict, filename, incomplete):
53c18592 2882 def print_mandatory(field, actual_field=None):
2883 if actual_field is None:
2884 actual_field = field
d06daf23 2885 if (self.params.get('force%s' % field, False)
53c18592 2886 and (not incomplete or info_dict.get(actual_field) is not None)):
2887 self.to_stdout(info_dict[actual_field])
d06daf23
S
2888
2889 def print_optional(field):
2890 if (self.params.get('force%s' % field, False)
2891 and info_dict.get(field) is not None):
2892 self.to_stdout(info_dict[field])
2893
53c18592 2894 info_dict = info_dict.copy()
2895 if filename is not None:
2896 info_dict['filename'] = filename
2897 if info_dict.get('requested_formats') is not None:
2898 # For RTMP URLs, also include the playpath
2899 info_dict['urls'] = '\n'.join(f['url'] + f.get('play_path', '') for f in info_dict['requested_formats'])
10331a26 2900 elif info_dict.get('url'):
53c18592 2901 info_dict['urls'] = info_dict['url'] + info_dict.get('play_path', '')
2902
bb66c247 2903 if (self.params.get('forcejson')
2904 or self.params['forceprint'].get('video')
2905 or self.params['print_to_file'].get('video')):
2b8a2973 2906 self.post_extract(info_dict)
bb66c247 2907 self._forceprint('video', info_dict)
53c18592 2908
d06daf23
S
2909 print_mandatory('title')
2910 print_mandatory('id')
53c18592 2911 print_mandatory('url', 'urls')
d06daf23
S
2912 print_optional('thumbnail')
2913 print_optional('description')
53c18592 2914 print_optional('filename')
b868936c 2915 if self.params.get('forceduration') and info_dict.get('duration') is not None:
d06daf23
S
2916 self.to_stdout(formatSeconds(info_dict['duration']))
2917 print_mandatory('format')
53c18592 2918
2b8a2973 2919 if self.params.get('forcejson'):
6e84b215 2920 self.to_stdout(json.dumps(self.sanitize_info(info_dict)))
d06daf23 2921
e8e73840 2922 def dl(self, name, info, subtitle=False, test=False):
88acdbc2 2923 if not info.get('url'):
1151c407 2924 self.raise_no_formats(info, True)
e8e73840 2925
2926 if test:
2927 verbose = self.params.get('verbose')
2928 params = {
2929 'test': True,
a169858f 2930 'quiet': self.params.get('quiet') or not verbose,
e8e73840 2931 'verbose': verbose,
2932 'noprogress': not verbose,
2933 'nopart': True,
2934 'skip_unavailable_fragments': False,
2935 'keep_fragments': False,
2936 'overwrites': True,
2937 '_no_ytdl_file': True,
2938 }
2939 else:
2940 params = self.params
96fccc10 2941 fd = get_suitable_downloader(info, params, to_stdout=(name == '-'))(self, params)
e8e73840 2942 if not test:
2943 for ph in self._progress_hooks:
2944 fd.add_progress_hook(ph)
42676437
M
2945 urls = '", "'.join(
2946 (f['url'].split(',')[0] + ',<data>' if f['url'].startswith('data:') else f['url'])
2947 for f in info.get('requested_formats', []) or [info])
3a408f9d 2948 self.write_debug(f'Invoking {fd.FD_NAME} downloader on "{urls}"')
03b4de72 2949
adbc4ec4
THD
2950 # Note: Ideally info should be a deep-copied so that hooks cannot modify it.
2951 # But it may contain objects that are not deep-copyable
2952 new_info = self._copy_infodict(info)
e8e73840 2953 if new_info.get('http_headers') is None:
2954 new_info['http_headers'] = self._calc_headers(new_info)
2955 return fd.download(name, new_info, subtitle)
2956
e04938ab 2957 def existing_file(self, filepaths, *, default_overwrite=True):
2958 existing_files = list(filter(os.path.exists, orderedSet(filepaths)))
2959 if existing_files and not self.params.get('overwrites', default_overwrite):
2960 return existing_files[0]
2961
2962 for file in existing_files:
2963 self.report_file_delete(file)
2964 os.remove(file)
2965 return None
2966
8222d8de 2967 def process_info(self, info_dict):
09b49e1f 2968 """Process a single resolved IE result. (Modifies it in-place)"""
8222d8de
JMF
2969
2970 assert info_dict.get('_type', 'video') == 'video'
f46e2f9d 2971 original_infodict = info_dict
fd288278 2972
4513a41a 2973 if 'format' not in info_dict and 'ext' in info_dict:
8222d8de
JMF
2974 info_dict['format'] = info_dict['ext']
2975
c77495e3 2976 if self._match_entry(info_dict) is not None:
9e907ebd 2977 info_dict['__write_download_archive'] = 'ignore'
8222d8de
JMF
2978 return
2979
09b49e1f 2980 # Does nothing under normal operation - for backward compatibility of process_info
277d6ff5 2981 self.post_extract(info_dict)
119e40ef 2982
2983 def replace_info_dict(new_info):
2984 nonlocal info_dict
2985 if new_info == info_dict:
2986 return
2987 info_dict.clear()
2988 info_dict.update(new_info)
2989
2990 new_info, _ = self.pre_process(info_dict, 'video')
2991 replace_info_dict(new_info)
0c14d66a 2992 self._num_downloads += 1
8222d8de 2993
dcf64d43 2994 # info_dict['_filename'] needs to be set for backward compatibility
de6000d9 2995 info_dict['_filename'] = full_filename = self.prepare_filename(info_dict, warn=True)
2996 temp_filename = self.prepare_filename(info_dict, 'temp')
0202b52a 2997 files_to_move = {}
8222d8de
JMF
2998
2999 # Forced printings
4513a41a 3000 self.__forced_printings(info_dict, full_filename, incomplete=('format' not in info_dict))
8222d8de 3001
ca6d59d2 3002 def check_max_downloads():
3003 if self._num_downloads >= float(self.params.get('max_downloads') or 'inf'):
3004 raise MaxDownloadsReached()
3005
b7b04c78 3006 if self.params.get('simulate'):
9e907ebd 3007 info_dict['__write_download_archive'] = self.params.get('force_write_download_archive')
ca6d59d2 3008 check_max_downloads()
8222d8de
JMF
3009 return
3010
de6000d9 3011 if full_filename is None:
8222d8de 3012 return
e92caff5 3013 if not self._ensure_dir_exists(encodeFilename(full_filename)):
0202b52a 3014 return
e92caff5 3015 if not self._ensure_dir_exists(encodeFilename(temp_filename)):
8222d8de
JMF
3016 return
3017
80c03fa9 3018 if self._write_description('video', info_dict,
3019 self.prepare_filename(info_dict, 'description')) is None:
3020 return
3021
3022 sub_files = self._write_subtitles(info_dict, temp_filename)
3023 if sub_files is None:
3024 return
3025 files_to_move.update(dict(sub_files))
3026
3027 thumb_files = self._write_thumbnails(
3028 'video', info_dict, temp_filename, self.prepare_filename(info_dict, 'thumbnail'))
3029 if thumb_files is None:
3030 return
3031 files_to_move.update(dict(thumb_files))
8222d8de 3032
80c03fa9 3033 infofn = self.prepare_filename(info_dict, 'infojson')
3034 _infojson_written = self._write_info_json('video', info_dict, infofn)
3035 if _infojson_written:
dac5df5a 3036 info_dict['infojson_filename'] = infofn
e75bb0d6 3037 # For backward compatibility, even though it was a private field
80c03fa9 3038 info_dict['__infojson_filename'] = infofn
3039 elif _infojson_written is None:
3040 return
3041
3042 # Note: Annotations are deprecated
3043 annofn = None
1fb07d10 3044 if self.params.get('writeannotations', False):
de6000d9 3045 annofn = self.prepare_filename(info_dict, 'annotation')
80c03fa9 3046 if annofn:
e92caff5 3047 if not self._ensure_dir_exists(encodeFilename(annofn)):
0202b52a 3048 return
0c3d0f51 3049 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(annofn)):
6febd1c1 3050 self.to_screen('[info] Video annotations are already present')
ffddb112
RA
3051 elif not info_dict.get('annotations'):
3052 self.report_warning('There are no annotations to write.')
7b6fefc9
PH
3053 else:
3054 try:
6febd1c1 3055 self.to_screen('[info] Writing video annotations to: ' + annofn)
86e5f3ed 3056 with open(encodeFilename(annofn), 'w', encoding='utf-8') as annofile:
7b6fefc9
PH
3057 annofile.write(info_dict['annotations'])
3058 except (KeyError, TypeError):
6febd1c1 3059 self.report_warning('There are no annotations to write.')
86e5f3ed 3060 except OSError:
6febd1c1 3061 self.report_error('Cannot write annotations file: ' + annofn)
7b6fefc9 3062 return
1fb07d10 3063
732044af 3064 # Write internet shortcut files
08438d2c 3065 def _write_link_file(link_type):
60f3e995 3066 url = try_get(info_dict['webpage_url'], iri_to_uri)
3067 if not url:
3068 self.report_warning(
3069 f'Cannot write internet shortcut file because the actual URL of "{info_dict["webpage_url"]}" is unknown')
3070 return True
08438d2c 3071 linkfn = replace_extension(self.prepare_filename(info_dict, 'link'), link_type, info_dict.get('ext'))
0e6b018a
Z
3072 if not self._ensure_dir_exists(encodeFilename(linkfn)):
3073 return False
10e3742e 3074 if self.params.get('overwrites', True) and os.path.exists(encodeFilename(linkfn)):
08438d2c 3075 self.to_screen(f'[info] Internet shortcut (.{link_type}) is already present')
3076 return True
3077 try:
3078 self.to_screen(f'[info] Writing internet shortcut (.{link_type}) to: {linkfn}')
86e5f3ed 3079 with open(encodeFilename(to_high_limit_path(linkfn)), 'w', encoding='utf-8',
3080 newline='\r\n' if link_type == 'url' else '\n') as linkfile:
60f3e995 3081 template_vars = {'url': url}
08438d2c 3082 if link_type == 'desktop':
3083 template_vars['filename'] = linkfn[:-(len(link_type) + 1)]
3084 linkfile.write(LINK_TEMPLATES[link_type] % template_vars)
86e5f3ed 3085 except OSError:
08438d2c 3086 self.report_error(f'Cannot write internet shortcut {linkfn}')
3087 return False
732044af 3088 return True
3089
08438d2c 3090 write_links = {
3091 'url': self.params.get('writeurllink'),
3092 'webloc': self.params.get('writewebloclink'),
3093 'desktop': self.params.get('writedesktoplink'),
3094 }
3095 if self.params.get('writelink'):
3096 link_type = ('webloc' if sys.platform == 'darwin'
3097 else 'desktop' if sys.platform.startswith('linux')
3098 else 'url')
3099 write_links[link_type] = True
3100
3101 if any(should_write and not _write_link_file(link_type)
3102 for link_type, should_write in write_links.items()):
3103 return
732044af 3104
415f8d51 3105 new_info, files_to_move = self.pre_process(info_dict, 'before_dl', files_to_move)
3106 replace_info_dict(new_info)
56d868db 3107
a13e6848 3108 if self.params.get('skip_download'):
56d868db 3109 info_dict['filepath'] = temp_filename
3110 info_dict['__finaldir'] = os.path.dirname(os.path.abspath(encodeFilename(full_filename)))
3111 info_dict['__files_to_move'] = files_to_move
f46e2f9d 3112 replace_info_dict(self.run_pp(MoveFilesAfterDownloadPP(self, False), info_dict))
9e907ebd 3113 info_dict['__write_download_archive'] = self.params.get('force_write_download_archive')
56d868db 3114 else:
3115 # Download
b868936c 3116 info_dict.setdefault('__postprocessors', [])
4340deca 3117 try:
0202b52a 3118
e04938ab 3119 def existing_video_file(*filepaths):
6b591b29 3120 ext = info_dict.get('ext')
e04938ab 3121 converted = lambda file: replace_extension(file, self.params.get('final_ext') or ext, ext)
3122 file = self.existing_file(itertools.chain(*zip(map(converted, filepaths), filepaths)),
3123 default_overwrite=False)
3124 if file:
3125 info_dict['ext'] = os.path.splitext(file)[1][1:]
3126 return file
0202b52a 3127
7b2c3f47 3128 fd, success = None, True
fccf90e7 3129 if info_dict.get('protocol') or info_dict.get('url'):
56ba69e4 3130 fd = get_suitable_downloader(info_dict, self.params, to_stdout=temp_filename == '-')
71df9b7f 3131 if fd is not FFmpegFD and 'no-direct-merge' not in self.params['compat_opts'] and (
56ba69e4 3132 info_dict.get('section_start') or info_dict.get('section_end')):
7b2c3f47 3133 msg = ('This format cannot be partially downloaded' if FFmpegFD.available()
56ba69e4 3134 else 'You have requested downloading the video partially, but ffmpeg is not installed')
3135 self.report_error(f'{msg}. Aborting')
5ec1b6b7 3136 return
5ec1b6b7 3137
4340deca 3138 if info_dict.get('requested_formats') is not None:
81cd954a 3139 requested_formats = info_dict['requested_formats']
0202b52a 3140 old_ext = info_dict['ext']
4e3b637d 3141 if self.params.get('merge_output_format') is None:
4e3b637d 3142 if (info_dict['ext'] == 'webm'
3143 and info_dict.get('thumbnails')
3144 # check with type instead of pp_key, __name__, or isinstance
3145 # since we dont want any custom PPs to trigger this
c487cf00 3146 and any(type(pp) == EmbedThumbnailPP for pp in self._pps['post_process'])): # noqa: E721
4e3b637d 3147 info_dict['ext'] = 'mkv'
3148 self.report_warning(
3149 'webm doesn\'t support embedding a thumbnail, mkv will be used')
124bc071 3150 new_ext = info_dict['ext']
0202b52a 3151
124bc071 3152 def correct_ext(filename, ext=new_ext):
96fccc10 3153 if filename == '-':
3154 return filename
0202b52a 3155 filename_real_ext = os.path.splitext(filename)[1][1:]
3156 filename_wo_ext = (
3157 os.path.splitext(filename)[0]
124bc071 3158 if filename_real_ext in (old_ext, new_ext)
0202b52a 3159 else filename)
86e5f3ed 3160 return f'{filename_wo_ext}.{ext}'
0202b52a 3161
38c6902b 3162 # Ensure filename always has a correct extension for successful merge
0202b52a 3163 full_filename = correct_ext(full_filename)
3164 temp_filename = correct_ext(temp_filename)
e04938ab 3165 dl_filename = existing_video_file(full_filename, temp_filename)
1ea24129 3166 info_dict['__real_download'] = False
18e674b4 3167
7b2c3f47 3168 merger = FFmpegMergerPP(self)
adbc4ec4 3169 downloaded = []
dbf5416a 3170 if dl_filename is not None:
6c7274ec 3171 self.report_file_already_downloaded(dl_filename)
adbc4ec4
THD
3172 elif fd:
3173 for f in requested_formats if fd != FFmpegFD else []:
3174 f['filepath'] = fname = prepend_extension(
3175 correct_ext(temp_filename, info_dict['ext']),
3176 'f%s' % f['format_id'], info_dict['ext'])
3177 downloaded.append(fname)
dbf5416a 3178 info_dict['url'] = '\n'.join(f['url'] for f in requested_formats)
3179 success, real_download = self.dl(temp_filename, info_dict)
3180 info_dict['__real_download'] = real_download
18e674b4 3181 else:
18e674b4 3182 if self.params.get('allow_unplayable_formats'):
3183 self.report_warning(
3184 'You have requested merging of multiple formats '
3185 'while also allowing unplayable formats to be downloaded. '
3186 'The formats won\'t be merged to prevent data corruption.')
3187 elif not merger.available:
e8969bda 3188 msg = 'You have requested merging of multiple formats but ffmpeg is not installed'
3189 if not self.params.get('ignoreerrors'):
3190 self.report_error(f'{msg}. Aborting due to --abort-on-error')
3191 return
3192 self.report_warning(f'{msg}. The formats won\'t be merged')
18e674b4 3193
96fccc10 3194 if temp_filename == '-':
adbc4ec4 3195 reason = ('using a downloader other than ffmpeg' if FFmpegFD.can_merge_formats(info_dict, self.params)
96fccc10 3196 else 'but the formats are incompatible for simultaneous download' if merger.available
3197 else 'but ffmpeg is not installed')
3198 self.report_warning(
3199 f'You have requested downloading multiple formats to stdout {reason}. '
3200 'The formats will be streamed one after the other')
3201 fname = temp_filename
dbf5416a 3202 for f in requested_formats:
3203 new_info = dict(info_dict)
3204 del new_info['requested_formats']
3205 new_info.update(f)
96fccc10 3206 if temp_filename != '-':
124bc071 3207 fname = prepend_extension(
3208 correct_ext(temp_filename, new_info['ext']),
3209 'f%s' % f['format_id'], new_info['ext'])
96fccc10 3210 if not self._ensure_dir_exists(fname):
3211 return
a21e0ab1 3212 f['filepath'] = fname
96fccc10 3213 downloaded.append(fname)
dbf5416a 3214 partial_success, real_download = self.dl(fname, new_info)
3215 info_dict['__real_download'] = info_dict['__real_download'] or real_download
3216 success = success and partial_success
adbc4ec4
THD
3217
3218 if downloaded and merger.available and not self.params.get('allow_unplayable_formats'):
3219 info_dict['__postprocessors'].append(merger)
3220 info_dict['__files_to_merge'] = downloaded
3221 # Even if there were no downloads, it is being merged only now
3222 info_dict['__real_download'] = True
3223 else:
3224 for file in downloaded:
3225 files_to_move[file] = None
4340deca
P
3226 else:
3227 # Just a single file
e04938ab 3228 dl_filename = existing_video_file(full_filename, temp_filename)
6c7274ec 3229 if dl_filename is None or dl_filename == temp_filename:
3230 # dl_filename == temp_filename could mean that the file was partially downloaded with --no-part.
3231 # So we should try to resume the download
e8e73840 3232 success, real_download = self.dl(temp_filename, info_dict)
0202b52a 3233 info_dict['__real_download'] = real_download
6c7274ec 3234 else:
3235 self.report_file_already_downloaded(dl_filename)
0202b52a 3236
0202b52a 3237 dl_filename = dl_filename or temp_filename
c571435f 3238 info_dict['__finaldir'] = os.path.dirname(os.path.abspath(encodeFilename(full_filename)))
0202b52a 3239
3158150c 3240 except network_exceptions as err:
7960b056 3241 self.report_error('unable to download video data: %s' % error_to_compat_str(err))
4340deca 3242 return
86e5f3ed 3243 except OSError as err:
4340deca
P
3244 raise UnavailableVideoError(err)
3245 except (ContentTooShortError, ) as err:
86e5f3ed 3246 self.report_error(f'content too short (expected {err.expected} bytes and served {err.downloaded})')
4340deca 3247 return
8222d8de 3248
415f8d51 3249 self._raise_pending_errors(info_dict)
de6000d9 3250 if success and full_filename != '-':
f17f8651 3251
fd7cfb64 3252 def fixup():
3253 do_fixup = True
3254 fixup_policy = self.params.get('fixup')
3255 vid = info_dict['id']
3256
3257 if fixup_policy in ('ignore', 'never'):
3258 return
3259 elif fixup_policy == 'warn':
3fe75fdc 3260 do_fixup = 'warn'
f89b3e2d 3261 elif fixup_policy != 'force':
3262 assert fixup_policy in ('detect_or_warn', None)
3263 if not info_dict.get('__real_download'):
3264 do_fixup = False
fd7cfb64 3265
3266 def ffmpeg_fixup(cndn, msg, cls):
3fe75fdc 3267 if not (do_fixup and cndn):
fd7cfb64 3268 return
3fe75fdc 3269 elif do_fixup == 'warn':
fd7cfb64 3270 self.report_warning(f'{vid}: {msg}')
3271 return
3272 pp = cls(self)
3273 if pp.available:
3274 info_dict['__postprocessors'].append(pp)
3275 else:
3276 self.report_warning(f'{vid}: {msg}. Install ffmpeg to fix this automatically')
3277
3278 stretched_ratio = info_dict.get('stretched_ratio')
ca9def71
LNO
3279 ffmpeg_fixup(stretched_ratio not in (1, None),
3280 f'Non-uniform pixel ratio {stretched_ratio}',
3281 FFmpegFixupStretchedPP)
fd7cfb64 3282
993191c0 3283 downloader = get_suitable_downloader(info_dict, self.params) if 'protocol' in info_dict else None
24146491 3284 downloader = downloader.FD_NAME if downloader else None
adbc4ec4 3285
ca9def71
LNO
3286 ext = info_dict.get('ext')
3287 postprocessed_by_ffmpeg = info_dict.get('requested_formats') or any((
3288 isinstance(pp, FFmpegVideoConvertorPP)
3289 and resolve_recode_mapping(ext, pp.mapping)[0] not in (ext, None)
3290 ) for pp in self._pps['post_process'])
3291
3292 if not postprocessed_by_ffmpeg:
3293 ffmpeg_fixup(ext == 'm4a' and info_dict.get('container') == 'm4a_dash',
f2df4071 3294 'writing DASH m4a. Only some players support this container',
3295 FFmpegFixupM4aPP)
24146491 3296 ffmpeg_fixup(downloader == 'hlsnative' and not self.params.get('hls_use_mpegts')
494f5230 3297 or info_dict.get('is_live') and self.params.get('hls_use_mpegts') is None,
adbc4ec4
THD
3298 'Possible MPEG-TS in MP4 container or malformed AAC timestamps',
3299 FFmpegFixupM3u8PP)
3300 ffmpeg_fixup(info_dict.get('is_live') and downloader == 'DashSegmentsFD',
3301 'Possible duplicate MOOV atoms', FFmpegFixupDuplicateMoovPP)
3302
24146491 3303 ffmpeg_fixup(downloader == 'web_socket_fragment', 'Malformed timestamps detected', FFmpegFixupTimestampPP)
3304 ffmpeg_fixup(downloader == 'web_socket_fragment', 'Malformed duration detected', FFmpegFixupDurationPP)
fd7cfb64 3305
3306 fixup()
8222d8de 3307 try:
f46e2f9d 3308 replace_info_dict(self.post_process(dl_filename, info_dict, files_to_move))
af819c21 3309 except PostProcessingError as err:
3310 self.report_error('Postprocessing: %s' % str(err))
8222d8de 3311 return
ab8e5e51
AM
3312 try:
3313 for ph in self._post_hooks:
23c1a667 3314 ph(info_dict['filepath'])
ab8e5e51
AM
3315 except Exception as err:
3316 self.report_error('post hooks: %s' % str(err))
3317 return
9e907ebd 3318 info_dict['__write_download_archive'] = True
2d30509f 3319
c487cf00 3320 assert info_dict is original_infodict # Make sure the info_dict was modified in-place
a13e6848 3321 if self.params.get('force_write_download_archive'):
9e907ebd 3322 info_dict['__write_download_archive'] = True
ca6d59d2 3323 check_max_downloads()
8222d8de 3324
aa9369a2 3325 def __download_wrapper(self, func):
3326 @functools.wraps(func)
3327 def wrapper(*args, **kwargs):
3328 try:
3329 res = func(*args, **kwargs)
3330 except UnavailableVideoError as e:
3331 self.report_error(e)
b222c271 3332 except DownloadCancelled as e:
3333 self.to_screen(f'[info] {e}')
3334 if not self.params.get('break_per_url'):
3335 raise
fd404bec 3336 self._num_downloads = 0
aa9369a2 3337 else:
3338 if self.params.get('dump_single_json', False):
3339 self.post_extract(res)
3340 self.to_stdout(json.dumps(self.sanitize_info(res)))
3341 return wrapper
3342
8222d8de
JMF
3343 def download(self, url_list):
3344 """Download a given list of URLs."""
aa9369a2 3345 url_list = variadic(url_list) # Passing a single URL is a common mistake
bf1824b3 3346 outtmpl = self.params['outtmpl']['default']
3089bc74
S
3347 if (len(url_list) > 1
3348 and outtmpl != '-'
3349 and '%' not in outtmpl
3350 and self.params.get('max_downloads') != 1):
acd69589 3351 raise SameFileError(outtmpl)
8222d8de
JMF
3352
3353 for url in url_list:
aa9369a2 3354 self.__download_wrapper(self.extract_info)(
3355 url, force_generic_extractor=self.params.get('force_generic_extractor', False))
8222d8de
JMF
3356
3357 return self._download_retcode
3358
1dcc4c0c 3359 def download_with_info_file(self, info_filename):
31bd3925
JMF
3360 with contextlib.closing(fileinput.FileInput(
3361 [info_filename], mode='r',
3362 openhook=fileinput.hook_encoded('utf-8'))) as f:
3363 # FileInput doesn't have a read method, we can't call json.load
8012d892 3364 info = self.sanitize_info(json.loads('\n'.join(f)), self.params.get('clean_infojson', True))
d4943898 3365 try:
aa9369a2 3366 self.__download_wrapper(self.process_ie_result)(info, download=True)
f2ebc5c7 3367 except (DownloadError, EntryNotInPlaylist, ReExtractInfo) as e:
bf5f605e 3368 if not isinstance(e, EntryNotInPlaylist):
3369 self.to_stderr('\r')
d4943898
JMF
3370 webpage_url = info.get('webpage_url')
3371 if webpage_url is not None:
aa9369a2 3372 self.report_warning(f'The info failed to download: {e}; trying with URL {webpage_url}')
d4943898
JMF
3373 return self.download([webpage_url])
3374 else:
3375 raise
3376 return self._download_retcode
1dcc4c0c 3377
cb202fd2 3378 @staticmethod
8012d892 3379 def sanitize_info(info_dict, remove_private_keys=False):
3380 ''' Sanitize the infodict for converting to json '''
3ad56b42 3381 if info_dict is None:
3382 return info_dict
6e84b215 3383 info_dict.setdefault('epoch', int(time.time()))
6a5a30f9 3384 info_dict.setdefault('_type', 'video')
b5e7a2e6 3385 info_dict.setdefault('_version', {
3386 'version': __version__,
3387 'current_git_head': current_git_head(),
3388 'release_git_head': RELEASE_GIT_HEAD,
3389 'repository': REPOSITORY,
3390 })
09b49e1f 3391
8012d892 3392 if remove_private_keys:
0a5a191a 3393 reject = lambda k, v: v is None or k.startswith('__') or k in {
f46e2f9d 3394 'requested_downloads', 'requested_formats', 'requested_subtitles', 'requested_entries',
0a5a191a 3395 'entries', 'filepath', '_filename', 'infojson_filename', 'original_url', 'playlist_autonumber',
08e29b9f 3396 '_format_sort_fields',
6e84b215 3397 }
ae8f99e6 3398 else:
09b49e1f 3399 reject = lambda k, v: False
adbc4ec4
THD
3400
3401 def filter_fn(obj):
3402 if isinstance(obj, dict):
3403 return {k: filter_fn(v) for k, v in obj.items() if not reject(k, v)}
3404 elif isinstance(obj, (list, tuple, set, LazyList)):
3405 return list(map(filter_fn, obj))
3406 elif obj is None or isinstance(obj, (str, int, float, bool)):
3407 return obj
3408 else:
3409 return repr(obj)
3410
5226731e 3411 return filter_fn(info_dict)
cb202fd2 3412
8012d892 3413 @staticmethod
3414 def filter_requested_info(info_dict, actually_filter=True):
3415 ''' Alias of sanitize_info for backward compatibility '''
3416 return YoutubeDL.sanitize_info(info_dict, actually_filter)
3417
43d7f5a5 3418 def _delete_downloaded_files(self, *files_to_delete, info={}, msg=None):
3419 for filename in set(filter(None, files_to_delete)):
3420 if msg:
3421 self.to_screen(msg % filename)
3422 try:
3423 os.remove(filename)
3424 except OSError:
3425 self.report_warning(f'Unable to delete file {filename}')
3426 if filename in info.get('__files_to_move', []): # NB: Delete even if None
3427 del info['__files_to_move'][filename]
3428
ed5835b4 3429 @staticmethod
3430 def post_extract(info_dict):
3431 def actual_post_extract(info_dict):
3432 if info_dict.get('_type') in ('playlist', 'multi_video'):
3433 for video_dict in info_dict.get('entries', {}):
3434 actual_post_extract(video_dict or {})
3435 return
3436
09b49e1f 3437 post_extractor = info_dict.pop('__post_extractor', None) or (lambda: {})
3438 info_dict.update(post_extractor())
ed5835b4 3439
3440 actual_post_extract(info_dict or {})
3441
dcf64d43 3442 def run_pp(self, pp, infodict):
5bfa4862 3443 files_to_delete = []
dcf64d43 3444 if '__files_to_move' not in infodict:
3445 infodict['__files_to_move'] = {}
b1940459 3446 try:
3447 files_to_delete, infodict = pp.run(infodict)
3448 except PostProcessingError as e:
3449 # Must be True and not 'only_download'
3450 if self.params.get('ignoreerrors') is True:
3451 self.report_error(e)
3452 return infodict
3453 raise
3454
5bfa4862 3455 if not files_to_delete:
dcf64d43 3456 return infodict
5bfa4862 3457 if self.params.get('keepvideo', False):
3458 for f in files_to_delete:
dcf64d43 3459 infodict['__files_to_move'].setdefault(f, '')
5bfa4862 3460 else:
43d7f5a5 3461 self._delete_downloaded_files(
3462 *files_to_delete, info=infodict, msg='Deleting original file %s (pass -k to keep)')
dcf64d43 3463 return infodict
5bfa4862 3464
ed5835b4 3465 def run_all_pps(self, key, info, *, additional_pps=None):
193fb150 3466 if key != 'video':
3467 self._forceprint(key, info)
ed5835b4 3468 for pp in (additional_pps or []) + self._pps[key]:
dc5f409c 3469 info = self.run_pp(pp, info)
ed5835b4 3470 return info
277d6ff5 3471
56d868db 3472 def pre_process(self, ie_info, key='pre_process', files_to_move=None):
5bfa4862 3473 info = dict(ie_info)
56d868db 3474 info['__files_to_move'] = files_to_move or {}
415f8d51 3475 try:
3476 info = self.run_all_pps(key, info)
3477 except PostProcessingError as err:
3478 msg = f'Preprocessing: {err}'
3479 info.setdefault('__pending_error', msg)
3480 self.report_error(msg, is_error=False)
56d868db 3481 return info, info.pop('__files_to_move', None)
5bfa4862 3482
f46e2f9d 3483 def post_process(self, filename, info, files_to_move=None):
8222d8de 3484 """Run all the postprocessors on the given file."""
8222d8de 3485 info['filepath'] = filename
dcf64d43 3486 info['__files_to_move'] = files_to_move or {}
ed5835b4 3487 info = self.run_all_pps('post_process', info, additional_pps=info.get('__postprocessors'))
dcf64d43 3488 info = self.run_pp(MoveFilesAfterDownloadPP(self), info)
3489 del info['__files_to_move']
ed5835b4 3490 return self.run_all_pps('after_move', info)
c1c9a79c 3491
5db07df6 3492 def _make_archive_id(self, info_dict):
e9fef7ee
S
3493 video_id = info_dict.get('id')
3494 if not video_id:
3495 return
5db07df6
PH
3496 # Future-proof against any change in case
3497 # and backwards compatibility with prior versions
e9fef7ee 3498 extractor = info_dict.get('extractor_key') or info_dict.get('ie_key') # key in a playlist
7012b23c 3499 if extractor is None:
1211bb6d
S
3500 url = str_or_none(info_dict.get('url'))
3501 if not url:
3502 return
e9fef7ee 3503 # Try to find matching extractor for the URL and take its ie_key
8b7491c8 3504 for ie_key, ie in self._ies.items():
1211bb6d 3505 if ie.suitable(url):
8b7491c8 3506 extractor = ie_key
e9fef7ee
S
3507 break
3508 else:
3509 return
0647d925 3510 return make_archive_id(extractor, video_id)
5db07df6
PH
3511
3512 def in_download_archive(self, info_dict):
ae103564 3513 if not self.archive:
5db07df6
PH
3514 return False
3515
1e8fe57e 3516 vid_ids = [self._make_archive_id(info_dict)]
c200096c 3517 vid_ids.extend(info_dict.get('_old_archive_ids') or [])
1e8fe57e 3518 return any(id_ in self.archive for id_ in vid_ids)
c1c9a79c
PH
3519
3520 def record_download_archive(self, info_dict):
3521 fn = self.params.get('download_archive')
3522 if fn is None:
3523 return
5db07df6
PH
3524 vid_id = self._make_archive_id(info_dict)
3525 assert vid_id
ae103564 3526
a13e6848 3527 self.write_debug(f'Adding to archive: {vid_id}')
9c935fbc 3528 if is_path_like(fn):
ae103564 3529 with locked_file(fn, 'a', encoding='utf-8') as archive_file:
3530 archive_file.write(vid_id + '\n')
a45e8619 3531 self.archive.add(vid_id)
dd82ffea 3532
8c51aa65 3533 @staticmethod
8abeeb94 3534 def format_resolution(format, default='unknown'):
9359f3d4 3535 if format.get('vcodec') == 'none' and format.get('acodec') != 'none':
fb04e403 3536 return 'audio only'
f49d89ee
PH
3537 if format.get('resolution') is not None:
3538 return format['resolution']
35615307 3539 if format.get('width') and format.get('height'):
ff51ed58 3540 return '%dx%d' % (format['width'], format['height'])
35615307 3541 elif format.get('height'):
ff51ed58 3542 return '%sp' % format['height']
35615307 3543 elif format.get('width'):
ff51ed58 3544 return '%dx?' % format['width']
3545 return default
8c51aa65 3546
8130779d 3547 def _list_format_headers(self, *headers):
3548 if self.params.get('listformats_table', True) is not False:
591bb9d3 3549 return [self._format_out(header, self.Styles.HEADERS) for header in headers]
8130779d 3550 return headers
3551
c57f7757
PH
3552 def _format_note(self, fdict):
3553 res = ''
3554 if fdict.get('ext') in ['f4f', 'f4m']:
f304da8a 3555 res += '(unsupported)'
32f90364
PH
3556 if fdict.get('language'):
3557 if res:
3558 res += ' '
f304da8a 3559 res += '[%s]' % fdict['language']
c57f7757 3560 if fdict.get('format_note') is not None:
f304da8a 3561 if res:
3562 res += ' '
3563 res += fdict['format_note']
c57f7757 3564 if fdict.get('tbr') is not None:
f304da8a 3565 if res:
3566 res += ', '
3567 res += '%4dk' % fdict['tbr']
c57f7757
PH
3568 if fdict.get('container') is not None:
3569 if res:
3570 res += ', '
3571 res += '%s container' % fdict['container']
3089bc74
S
3572 if (fdict.get('vcodec') is not None
3573 and fdict.get('vcodec') != 'none'):
c57f7757
PH
3574 if res:
3575 res += ', '
3576 res += fdict['vcodec']
91c7271a 3577 if fdict.get('vbr') is not None:
c57f7757
PH
3578 res += '@'
3579 elif fdict.get('vbr') is not None and fdict.get('abr') is not None:
3580 res += 'video@'
3581 if fdict.get('vbr') is not None:
3582 res += '%4dk' % fdict['vbr']
fbb21cf5 3583 if fdict.get('fps') is not None:
5d583bdf
S
3584 if res:
3585 res += ', '
3586 res += '%sfps' % fdict['fps']
c57f7757
PH
3587 if fdict.get('acodec') is not None:
3588 if res:
3589 res += ', '
3590 if fdict['acodec'] == 'none':
3591 res += 'video only'
3592 else:
3593 res += '%-5s' % fdict['acodec']
3594 elif fdict.get('abr') is not None:
3595 if res:
3596 res += ', '
3597 res += 'audio'
3598 if fdict.get('abr') is not None:
3599 res += '@%3dk' % fdict['abr']
3600 if fdict.get('asr') is not None:
3601 res += ' (%5dHz)' % fdict['asr']
3602 if fdict.get('filesize') is not None:
3603 if res:
3604 res += ', '
3605 res += format_bytes(fdict['filesize'])
9732d77e
PH
3606 elif fdict.get('filesize_approx') is not None:
3607 if res:
3608 res += ', '
3609 res += '~' + format_bytes(fdict['filesize_approx'])
c57f7757 3610 return res
91c7271a 3611
aebb4f4b 3612 def _get_formats(self, info_dict):
3613 if info_dict.get('formats') is None:
3614 if info_dict.get('url') and info_dict.get('_type', 'video') == 'video':
3615 return [info_dict]
3616 return []
3617 return info_dict['formats']
b69fd25c 3618
aebb4f4b 3619 def render_formats_table(self, info_dict):
3620 formats = self._get_formats(info_dict)
3621 if not formats:
3622 return
8130779d 3623 if not self.params.get('listformats_table', True) is not False:
76d321f6 3624 table = [
3625 [
3626 format_field(f, 'format_id'),
3627 format_field(f, 'ext'),
3628 self.format_resolution(f),
8130779d 3629 self._format_note(f)
d5d1df8a 3630 ] for f in formats if (f.get('preference') or 0) >= -1000]
8130779d 3631 return render_table(['format code', 'extension', 'resolution', 'note'], table, extra_gap=1)
3632
d816f61f 3633 def simplified_codec(f, field):
3634 assert field in ('acodec', 'vcodec')
3635 codec = f.get(field, 'unknown')
f5ea4748 3636 if not codec:
3637 return 'unknown'
3638 elif codec != 'none':
d816f61f 3639 return '.'.join(codec.split('.')[:4])
3640
3641 if field == 'vcodec' and f.get('acodec') == 'none':
3642 return 'images'
3643 elif field == 'acodec' and f.get('vcodec') == 'none':
3644 return ''
3645 return self._format_out('audio only' if field == 'vcodec' else 'video only',
3646 self.Styles.SUPPRESS)
3647
591bb9d3 3648 delim = self._format_out('\u2502', self.Styles.DELIM, '|', test_encoding=True)
8130779d 3649 table = [
3650 [
591bb9d3 3651 self._format_out(format_field(f, 'format_id'), self.Styles.ID),
8130779d 3652 format_field(f, 'ext'),
3653 format_field(f, func=self.format_resolution, ignore=('audio only', 'images')),
563e0bf8 3654 format_field(f, 'fps', '\t%d', func=round),
8130779d 3655 format_field(f, 'dynamic_range', '%s', ignore=(None, 'SDR')).replace('HDR', ''),
b8ed0f15 3656 format_field(f, 'audio_channels', '\t%s'),
8130779d 3657 delim,
3658 format_field(f, 'filesize', ' \t%s', func=format_bytes) + format_field(f, 'filesize_approx', '~\t%s', func=format_bytes),
563e0bf8 3659 format_field(f, 'tbr', '\t%dk', func=round),
8130779d 3660 shorten_protocol_name(f.get('protocol', '')),
3661 delim,
d816f61f 3662 simplified_codec(f, 'vcodec'),
563e0bf8 3663 format_field(f, 'vbr', '\t%dk', func=round),
d816f61f 3664 simplified_codec(f, 'acodec'),
563e0bf8 3665 format_field(f, 'abr', '\t%dk', func=round),
ae61d108 3666 format_field(f, 'asr', '\t%s', func=format_decimal_suffix),
8130779d 3667 join_nonempty(
591bb9d3 3668 self._format_out('UNSUPPORTED', 'light red') if f.get('ext') in ('f4f', 'f4m') else None,
8130779d 3669 format_field(f, 'language', '[%s]'),
3670 join_nonempty(format_field(f, 'format_note'),
3671 format_field(f, 'container', ignore=(None, f.get('ext'))),
3672 delim=', '),
3673 delim=' '),
3674 ] for f in formats if f.get('preference') is None or f['preference'] >= -1000]
3675 header_line = self._list_format_headers(
b8ed0f15 3676 'ID', 'EXT', 'RESOLUTION', '\tFPS', 'HDR', 'CH', delim, '\tFILESIZE', '\tTBR', 'PROTO',
8130779d 3677 delim, 'VCODEC', '\tVBR', 'ACODEC', '\tABR', '\tASR', 'MORE INFO')
3678
3679 return render_table(
3680 header_line, table, hide_empty=True,
591bb9d3 3681 delim=self._format_out('\u2500', self.Styles.DELIM, '-', test_encoding=True))
8130779d 3682
3683 def render_thumbnails_table(self, info_dict):
88f23a18 3684 thumbnails = list(info_dict.get('thumbnails') or [])
cfb56d1a 3685 if not thumbnails:
8130779d 3686 return None
3687 return render_table(
ec11a9f4 3688 self._list_format_headers('ID', 'Width', 'Height', 'URL'),
177662e0 3689 [[t.get('id'), t.get('width') or 'unknown', t.get('height') or 'unknown', t['url']] for t in thumbnails])
2412044c 3690
8130779d 3691 def render_subtitles_table(self, video_id, subtitles):
2412044c 3692 def _row(lang, formats):
49c258e1 3693 exts, names = zip(*((f['ext'], f.get('name') or 'unknown') for f in reversed(formats)))
2412044c 3694 if len(set(names)) == 1:
7aee40c1 3695 names = [] if names[0] == 'unknown' else names[:1]
2412044c 3696 return [lang, ', '.join(names), ', '.join(exts)]
3697
8130779d 3698 if not subtitles:
3699 return None
3700 return render_table(
ec11a9f4 3701 self._list_format_headers('Language', 'Name', 'Formats'),
2412044c 3702 [_row(lang, formats) for lang, formats in subtitles.items()],
8130779d 3703 hide_empty=True)
3704
3705 def __list_table(self, video_id, name, func, *args):
3706 table = func(*args)
3707 if not table:
3708 self.to_screen(f'{video_id} has no {name}')
3709 return
3710 self.to_screen(f'[info] Available {name} for {video_id}:')
3711 self.to_stdout(table)
3712
3713 def list_formats(self, info_dict):
3714 self.__list_table(info_dict['id'], 'formats', self.render_formats_table, info_dict)
3715
3716 def list_thumbnails(self, info_dict):
3717 self.__list_table(info_dict['id'], 'thumbnails', self.render_thumbnails_table, info_dict)
3718
3719 def list_subtitles(self, video_id, subtitles, name='subtitles'):
3720 self.__list_table(video_id, name, self.render_subtitles_table, video_id, subtitles)
a504ced0 3721
dca08720
PH
3722 def urlopen(self, req):
3723 """ Start an HTTP download """
f9934b96 3724 if isinstance(req, str):
67dda517 3725 req = sanitized_Request(req)
19a41fc6 3726 return self._opener.open(req, timeout=self._socket_timeout)
dca08720
PH
3727
3728 def print_debug_header(self):
3729 if not self.params.get('verbose'):
3730 return
49a57e70 3731
a057779d 3732 from . import _IN_CLI # Must be delayed import
3733
560738f3 3734 # These imports can be slow. So import them only as needed
3735 from .extractor.extractors import _LAZY_LOADER
e756f45b
M
3736 from .extractor.extractors import (
3737 _PLUGIN_CLASSES as plugin_ies,
3738 _PLUGIN_OVERRIDES as plugin_ie_overrides
3739 )
560738f3 3740
49a57e70 3741 def get_encoding(stream):
2a938746 3742 ret = str(getattr(stream, 'encoding', 'missing (%s)' % type(stream).__name__))
49a57e70 3743 if not supports_terminal_sequences(stream):
53973b4d 3744 from .utils import WINDOWS_VT_MODE # Must be imported locally
e3c7d495 3745 ret += ' (No VT)' if WINDOWS_VT_MODE is False else ' (No ANSI)'
49a57e70 3746 return ret
3747
591bb9d3 3748 encoding_str = 'Encodings: locale %s, fs %s, pref %s, %s' % (
49a57e70 3749 locale.getpreferredencoding(),
3750 sys.getfilesystemencoding(),
591bb9d3 3751 self.get_encoding(),
3752 ', '.join(
64fa820c 3753 f'{key} {get_encoding(stream)}' for key, stream in self._out_files.items_
591bb9d3 3754 if stream is not None and key != 'console')
3755 )
883d4b1e 3756
3757 logger = self.params.get('logger')
3758 if logger:
3759 write_debug = lambda msg: logger.debug(f'[debug] {msg}')
3760 write_debug(encoding_str)
3761 else:
96565c7e 3762 write_string(f'[debug] {encoding_str}\n', encoding=None)
49a57e70 3763 write_debug = lambda msg: self._write_string(f'[debug] {msg}\n')
734f90bb 3764
4c88ff87 3765 source = detect_variant()
70b23409 3766 if VARIANT not in (None, 'pip'):
3767 source += '*'
36eaf303 3768 write_debug(join_nonempty(
b5e7a2e6 3769 f'{"yt-dlp" if REPOSITORY == "yt-dlp/yt-dlp" else REPOSITORY} version',
3770 __version__,
36eaf303 3771 f'[{RELEASE_GIT_HEAD}]' if RELEASE_GIT_HEAD else '',
3772 '' if source == 'unknown' else f'({source})',
a057779d 3773 '' if _IN_CLI else 'API',
36eaf303 3774 delim=' '))
497074f0 3775
3776 if not _IN_CLI:
3777 write_debug(f'params: {self.params}')
3778
6e21fdd2 3779 if not _LAZY_LOADER:
3780 if os.environ.get('YTDLP_NO_LAZY_EXTRACTORS'):
49a57e70 3781 write_debug('Lazy loading extractors is forcibly disabled')
6e21fdd2 3782 else:
49a57e70 3783 write_debug('Lazy loading extractors is disabled')
8a82af35 3784 if self.params['compat_opts']:
3785 write_debug('Compatibility options: %s' % ', '.join(self.params['compat_opts']))
36eaf303 3786
b5e7a2e6 3787 if current_git_head():
3788 write_debug(f'Git HEAD: {current_git_head()}')
b1f94422 3789 write_debug(system_identifier())
d28b5171 3790
8913ef74 3791 exe_versions, ffmpeg_features = FFmpegPostProcessor.get_versions_and_features(self)
3792 ffmpeg_features = {key for key, val in ffmpeg_features.items() if val}
3793 if ffmpeg_features:
19a03940 3794 exe_versions['ffmpeg'] += ' (%s)' % ','.join(sorted(ffmpeg_features))
8913ef74 3795
4c83c967 3796 exe_versions['rtmpdump'] = rtmpdump_version()
feee8d32 3797 exe_versions['phantomjs'] = PhantomJSwrapper._version()
d28b5171 3798 exe_str = ', '.join(
2831b468 3799 f'{exe} {v}' for exe, v in sorted(exe_versions.items()) if v
3800 ) or 'none'
49a57e70 3801 write_debug('exe versions: %s' % exe_str)
dca08720 3802
1d485a1a 3803 from .compat.compat_utils import get_package_info
9b8ee23b 3804 from .dependencies import available_dependencies
3805
3806 write_debug('Optional libraries: %s' % (', '.join(sorted({
1d485a1a 3807 join_nonempty(*get_package_info(m)) for m in available_dependencies.values()
9b8ee23b 3808 })) or 'none'))
2831b468 3809
97ec5bc5 3810 self._setup_opener()
dca08720
PH
3811 proxy_map = {}
3812 for handler in self._opener.handlers:
3813 if hasattr(handler, 'proxies'):
3814 proxy_map.update(handler.proxies)
49a57e70 3815 write_debug(f'Proxy map: {proxy_map}')
dca08720 3816
e756f45b
M
3817 for plugin_type, plugins in {'Extractor': plugin_ies, 'Post-Processor': plugin_pps}.items():
3818 display_list = ['%s%s' % (
8e40b9d1 3819 klass.__name__, '' if klass.__name__ == name else f' as {name}')
e756f45b
M
3820 for name, klass in plugins.items()]
3821 if plugin_type == 'Extractor':
3822 display_list.extend(f'{plugins[-1].IE_NAME.partition("+")[2]} ({parent.__name__})'
3823 for parent, plugins in plugin_ie_overrides.items())
3824 if not display_list:
3825 continue
3826 write_debug(f'{plugin_type} Plugins: {", ".join(sorted(display_list))}')
3827
8e40b9d1
M
3828 plugin_dirs = plugin_directories()
3829 if plugin_dirs:
3830 write_debug(f'Plugin directories: {plugin_dirs}')
3831
49a57e70 3832 # Not implemented
3833 if False and self.params.get('call_home'):
0f06bcd7 3834 ipaddr = self.urlopen('https://yt-dl.org/ip').read().decode()
49a57e70 3835 write_debug('Public IP address: %s' % ipaddr)
58b1f00d 3836 latest_version = self.urlopen(
0f06bcd7 3837 'https://yt-dl.org/latest/version').read().decode()
58b1f00d
PH
3838 if version_tuple(latest_version) > version_tuple(__version__):
3839 self.report_warning(
3840 'You are using an outdated version (newest version: %s)! '
3841 'See https://yt-dl.org/update if you need help updating.' %
3842 latest_version)
3843
e344693b 3844 def _setup_opener(self):
97ec5bc5 3845 if hasattr(self, '_opener'):
3846 return
6ad14cab 3847 timeout_val = self.params.get('socket_timeout')
17bddf3e 3848 self._socket_timeout = 20 if timeout_val is None else float(timeout_val)
6ad14cab 3849
982ee69a 3850 opts_cookiesfrombrowser = self.params.get('cookiesfrombrowser')
dca08720
PH
3851 opts_cookiefile = self.params.get('cookiefile')
3852 opts_proxy = self.params.get('proxy')
3853
982ee69a 3854 self.cookiejar = load_cookies(opts_cookiefile, opts_cookiesfrombrowser, self)
dca08720 3855
6a3f4c3f 3856 cookie_processor = YoutubeDLCookieProcessor(self.cookiejar)
dca08720
PH
3857 if opts_proxy is not None:
3858 if opts_proxy == '':
3859 proxies = {}
3860 else:
3861 proxies = {'http': opts_proxy, 'https': opts_proxy}
3862 else:
ac668111 3863 proxies = urllib.request.getproxies()
067aa17e 3864 # Set HTTPS proxy to HTTP one if given (https://github.com/ytdl-org/youtube-dl/issues/805)
dca08720
PH
3865 if 'http' in proxies and 'https' not in proxies:
3866 proxies['https'] = proxies['http']
91410c9b 3867 proxy_handler = PerRequestProxyHandler(proxies)
a0ddb8a2
PH
3868
3869 debuglevel = 1 if self.params.get('debug_printtraffic') else 0
be4a824d
PH
3870 https_handler = make_HTTPS_handler(self.params, debuglevel=debuglevel)
3871 ydlh = YoutubeDLHandler(self.params, debuglevel=debuglevel)
fca6dba8 3872 redirect_handler = YoutubeDLRedirectHandler()
f9934b96 3873 data_handler = urllib.request.DataHandler()
6240b0a2
JMF
3874
3875 # When passing our own FileHandler instance, build_opener won't add the
3876 # default FileHandler and allows us to disable the file protocol, which
3877 # can be used for malicious purposes (see
067aa17e 3878 # https://github.com/ytdl-org/youtube-dl/issues/8227)
ac668111 3879 file_handler = urllib.request.FileHandler()
6240b0a2 3880
8300774c
M
3881 if not self.params.get('enable_file_urls'):
3882 def file_open(*args, **kwargs):
3883 raise urllib.error.URLError(
3884 'file:// URLs are explicitly disabled in yt-dlp for security reasons. '
3885 'Use --enable-file-urls to enable at your own risk.')
3886 file_handler.file_open = file_open
6240b0a2 3887
ac668111 3888 opener = urllib.request.build_opener(
fca6dba8 3889 proxy_handler, https_handler, cookie_processor, ydlh, redirect_handler, data_handler, file_handler)
2461f79d 3890
dca08720
PH
3891 # Delete the default user-agent header, which would otherwise apply in
3892 # cases where our custom HTTP handler doesn't come into play
067aa17e 3893 # (See https://github.com/ytdl-org/youtube-dl/issues/1309 for details)
dca08720
PH
3894 opener.addheaders = []
3895 self._opener = opener
62fec3b2
PH
3896
3897 def encode(self, s):
3898 if isinstance(s, bytes):
3899 return s # Already encoded
3900
3901 try:
3902 return s.encode(self.get_encoding())
3903 except UnicodeEncodeError as err:
3904 err.reason = err.reason + '. Check your system encoding configuration or use the --encoding option.'
3905 raise
3906
3907 def get_encoding(self):
3908 encoding = self.params.get('encoding')
3909 if encoding is None:
3910 encoding = preferredencoding()
3911 return encoding
ec82d85a 3912
e08a85d8 3913 def _write_info_json(self, label, ie_result, infofn, overwrite=None):
cb96c5be 3914 ''' Write infojson and returns True = written, 'exists' = Already exists, False = skip, None = error '''
e08a85d8 3915 if overwrite is None:
3916 overwrite = self.params.get('overwrites', True)
80c03fa9 3917 if not self.params.get('writeinfojson'):
3918 return False
3919 elif not infofn:
3920 self.write_debug(f'Skipping writing {label} infojson')
3921 return False
3922 elif not self._ensure_dir_exists(infofn):
3923 return None
e08a85d8 3924 elif not overwrite and os.path.exists(infofn):
80c03fa9 3925 self.to_screen(f'[info] {label.title()} metadata is already present')
cb96c5be 3926 return 'exists'
3927
3928 self.to_screen(f'[info] Writing {label} metadata as JSON to: {infofn}')
3929 try:
3930 write_json_file(self.sanitize_info(ie_result, self.params.get('clean_infojson', True)), infofn)
3931 return True
86e5f3ed 3932 except OSError:
cb96c5be 3933 self.report_error(f'Cannot write {label} metadata to JSON file {infofn}')
3934 return None
80c03fa9 3935
3936 def _write_description(self, label, ie_result, descfn):
3937 ''' Write description and returns True = written, False = skip, None = error '''
3938 if not self.params.get('writedescription'):
3939 return False
3940 elif not descfn:
3941 self.write_debug(f'Skipping writing {label} description')
3942 return False
3943 elif not self._ensure_dir_exists(descfn):
3944 return None
3945 elif not self.params.get('overwrites', True) and os.path.exists(descfn):
3946 self.to_screen(f'[info] {label.title()} description is already present')
3947 elif ie_result.get('description') is None:
88fb9425 3948 self.to_screen(f'[info] There\'s no {label} description to write')
80c03fa9 3949 return False
3950 else:
3951 try:
3952 self.to_screen(f'[info] Writing {label} description to: {descfn}')
86e5f3ed 3953 with open(encodeFilename(descfn), 'w', encoding='utf-8') as descfile:
80c03fa9 3954 descfile.write(ie_result['description'])
86e5f3ed 3955 except OSError:
80c03fa9 3956 self.report_error(f'Cannot write {label} description file {descfn}')
3957 return None
3958 return True
3959
3960 def _write_subtitles(self, info_dict, filename):
3961 ''' Write subtitles to file and return list of (sub_filename, final_sub_filename); or None if error'''
3962 ret = []
3963 subtitles = info_dict.get('requested_subtitles')
88fb9425 3964 if not (self.params.get('writesubtitles') or self.params.get('writeautomaticsub')):
80c03fa9 3965 # subtitles download errors are already managed as troubles in relevant IE
3966 # that way it will silently go on when used with unsupporting IE
3967 return ret
88fb9425 3968 elif not subtitles:
3969 self.to_screen('[info] There\'s no subtitles for the requested languages')
3970 return ret
80c03fa9 3971 sub_filename_base = self.prepare_filename(info_dict, 'subtitle')
3972 if not sub_filename_base:
3973 self.to_screen('[info] Skipping writing video subtitles')
3974 return ret
88fb9425 3975
80c03fa9 3976 for sub_lang, sub_info in subtitles.items():
3977 sub_format = sub_info['ext']
3978 sub_filename = subtitles_filename(filename, sub_lang, sub_format, info_dict.get('ext'))
3979 sub_filename_final = subtitles_filename(sub_filename_base, sub_lang, sub_format, info_dict.get('ext'))
e04938ab 3980 existing_sub = self.existing_file((sub_filename_final, sub_filename))
3981 if existing_sub:
80c03fa9 3982 self.to_screen(f'[info] Video subtitle {sub_lang}.{sub_format} is already present')
e04938ab 3983 sub_info['filepath'] = existing_sub
3984 ret.append((existing_sub, sub_filename_final))
80c03fa9 3985 continue
3986
3987 self.to_screen(f'[info] Writing video subtitles to: {sub_filename}')
3988 if sub_info.get('data') is not None:
3989 try:
3990 # Use newline='' to prevent conversion of newline characters
3991 # See https://github.com/ytdl-org/youtube-dl/issues/10268
86e5f3ed 3992 with open(sub_filename, 'w', encoding='utf-8', newline='') as subfile:
80c03fa9 3993 subfile.write(sub_info['data'])
3994 sub_info['filepath'] = sub_filename
3995 ret.append((sub_filename, sub_filename_final))
3996 continue
86e5f3ed 3997 except OSError:
80c03fa9 3998 self.report_error(f'Cannot write video subtitles file {sub_filename}')
3999 return None
4000
4001 try:
4002 sub_copy = sub_info.copy()
4003 sub_copy.setdefault('http_headers', info_dict.get('http_headers'))
4004 self.dl(sub_filename, sub_copy, subtitle=True)
4005 sub_info['filepath'] = sub_filename
4006 ret.append((sub_filename, sub_filename_final))
6020e05d 4007 except (DownloadError, ExtractorError, IOError, OSError, ValueError) + network_exceptions as err:
c70c418d 4008 msg = f'Unable to download video subtitles for {sub_lang!r}: {err}'
6020e05d 4009 if self.params.get('ignoreerrors') is not True: # False or 'only_download'
c70c418d 4010 if not self.params.get('ignoreerrors'):
4011 self.report_error(msg)
4012 raise DownloadError(msg)
4013 self.report_warning(msg)
519804a9 4014 return ret
80c03fa9 4015
4016 def _write_thumbnails(self, label, info_dict, filename, thumb_filename_base=None):
4017 ''' Write thumbnails to file and return list of (thumb_filename, final_thumb_filename) '''
6c4fd172 4018 write_all = self.params.get('write_all_thumbnails', False)
80c03fa9 4019 thumbnails, ret = [], []
6c4fd172 4020 if write_all or self.params.get('writethumbnail', False):
0202b52a 4021 thumbnails = info_dict.get('thumbnails') or []
88fb9425 4022 if not thumbnails:
4023 self.to_screen(f'[info] There\'s no {label} thumbnails to download')
4024 return ret
6c4fd172 4025 multiple = write_all and len(thumbnails) > 1
ec82d85a 4026
80c03fa9 4027 if thumb_filename_base is None:
4028 thumb_filename_base = filename
4029 if thumbnails and not thumb_filename_base:
4030 self.write_debug(f'Skipping writing {label} thumbnail')
4031 return ret
4032
dd0228ce 4033 for idx, t in list(enumerate(thumbnails))[::-1]:
80c03fa9 4034 thumb_ext = (f'{t["id"]}.' if multiple else '') + determine_ext(t['url'], 'jpg')
aa9369a2 4035 thumb_display_id = f'{label} thumbnail {t["id"]}'
80c03fa9 4036 thumb_filename = replace_extension(filename, thumb_ext, info_dict.get('ext'))
4037 thumb_filename_final = replace_extension(thumb_filename_base, thumb_ext, info_dict.get('ext'))
ec82d85a 4038
e04938ab 4039 existing_thumb = self.existing_file((thumb_filename_final, thumb_filename))
4040 if existing_thumb:
aa9369a2 4041 self.to_screen('[info] %s is already present' % (
4042 thumb_display_id if multiple else f'{label} thumbnail').capitalize())
e04938ab 4043 t['filepath'] = existing_thumb
4044 ret.append((existing_thumb, thumb_filename_final))
ec82d85a 4045 else:
80c03fa9 4046 self.to_screen(f'[info] Downloading {thumb_display_id} ...')
ec82d85a 4047 try:
297e9952 4048 uf = self.urlopen(sanitized_Request(t['url'], headers=t.get('http_headers', {})))
80c03fa9 4049 self.to_screen(f'[info] Writing {thumb_display_id} to: {thumb_filename}')
d3d89c32 4050 with open(encodeFilename(thumb_filename), 'wb') as thumbf:
ec82d85a 4051 shutil.copyfileobj(uf, thumbf)
80c03fa9 4052 ret.append((thumb_filename, thumb_filename_final))
885cc0b7 4053 t['filepath'] = thumb_filename
3158150c 4054 except network_exceptions as err:
dd0228ce 4055 thumbnails.pop(idx)
80c03fa9 4056 self.report_warning(f'Unable to download {thumb_display_id}: {err}')
6c4fd172 4057 if ret and not write_all:
4058 break
0202b52a 4059 return ret