]> jfr.im git - yt-dlp.git/blame - yt_dlp/YoutubeDL.py
[extractor] Add a way to distinguish IEs that returns only videos
[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
35from .postprocessor import _PLUGIN_CLASSES as plugin_postprocessors
36from .postprocessor import (
37 EmbedThumbnailPP,
38 FFmpegFixupDuplicateMoovPP,
39 FFmpegFixupDurationPP,
40 FFmpegFixupM3u8PP,
41 FFmpegFixupM4aPP,
42 FFmpegFixupStretchedPP,
43 FFmpegFixupTimestampPP,
44 FFmpegMergerPP,
45 FFmpegPostProcessor,
ca9def71 46 FFmpegVideoConvertorPP,
f8271158 47 MoveFilesAfterDownloadPP,
48 get_postprocessor,
49)
ca9def71 50from .postprocessor.ffmpeg import resolve_mapping as resolve_recode_mapping
b5e7a2e6 51from .update import REPOSITORY, current_git_head, detect_variant
8c25f81b 52from .utils import (
f8271158 53 DEFAULT_OUTTMPL,
7b2c3f47 54 IDENTITY,
f8271158 55 LINK_TEMPLATES,
8dc59305 56 MEDIA_EXTENSIONS,
f8271158 57 NO_DEFAULT,
1d485a1a 58 NUMBER_RE,
f8271158 59 OUTTMPL_TYPES,
60 POSTPROCESS_WHEN,
61 STR_FORMAT_RE_TMPL,
62 STR_FORMAT_TYPES,
63 ContentTooShortError,
64 DateRange,
65 DownloadCancelled,
66 DownloadError,
67 EntryNotInPlaylist,
68 ExistingVideoReached,
69 ExtractorError,
70 GeoRestrictedError,
71 HEADRequest,
f8271158 72 ISO3166Utils,
73 LazyList,
74 MaxDownloadsReached,
19a03940 75 Namespace,
f8271158 76 PagedList,
77 PerRequestProxyHandler,
7e88d7d7 78 PlaylistEntries,
f8271158 79 Popen,
80 PostProcessingError,
81 ReExtractInfo,
82 RejectedVideoReached,
83 SameFileError,
84 UnavailableVideoError,
693f0600 85 UserNotLive,
f8271158 86 YoutubeDLCookieProcessor,
87 YoutubeDLHandler,
88 YoutubeDLRedirectHandler,
eedb7ba5
S
89 age_restricted,
90 args_to_str,
cb794ee0 91 bug_reports_message,
ce02ed60 92 date_from_str,
da4db748 93 deprecation_warning,
ce02ed60 94 determine_ext,
b5559424 95 determine_protocol,
c0384f22 96 encode_compat_str,
ce02ed60 97 encodeFilename,
a06916d9 98 error_to_compat_str,
47cdc68e 99 escapeHTML,
590bc6f6 100 expand_path,
90137ca4 101 filter_dict,
e29663c6 102 float_or_none,
02dbf93f 103 format_bytes,
e0fd9573 104 format_decimal_suffix,
f8271158 105 format_field,
525ef922 106 formatSeconds,
fc61aff4 107 get_compatible_ext,
0bb322b9 108 get_domain,
c9969434 109 int_or_none,
732044af 110 iri_to_uri,
941e881e 111 is_path_like,
34921b43 112 join_nonempty,
ce02ed60 113 locked_file,
0647d925 114 make_archive_id,
0202b52a 115 make_dir,
dca08720 116 make_HTTPS_handler,
8b7539d2 117 merge_headers,
3158150c 118 network_exceptions,
ec11a9f4 119 number_of_digits,
cd6fc19e 120 orderedSet,
5314b521 121 orderedSet_from_options,
083c9df9 122 parse_filesize,
ce02ed60 123 preferredencoding,
eedb7ba5 124 prepend_extension,
51fb4995 125 register_socks_protocols,
3efb96a6 126 remove_terminal_sequences,
cfb56d1a 127 render_table,
eedb7ba5 128 replace_extension,
ce02ed60 129 sanitize_filename,
1bb5c511 130 sanitize_path,
dcf77cf1 131 sanitize_url,
67dda517 132 sanitized_Request,
e5660ee6 133 std_headers,
1211bb6d 134 str_or_none,
e29663c6 135 strftime_or_none,
ce02ed60 136 subtitles_filename,
819e0531 137 supports_terminal_sequences,
b1f94422 138 system_identifier,
f2ebc5c7 139 timetuple_from_msec,
732044af 140 to_high_limit_path,
324ad820 141 traverse_obj,
fc61aff4 142 try_call,
6033d980 143 try_get,
29eb5174 144 url_basename,
7d1eb38a 145 variadic,
58b1f00d 146 version_tuple,
53973b4d 147 windows_enable_vt_mode,
ce02ed60
PH
148 write_json_file,
149 write_string,
4f026faf 150)
70b23409 151from .version import RELEASE_GIT_HEAD, VARIANT, __version__
8222d8de 152
e9c0cdd3
YCH
153if compat_os_name == 'nt':
154 import ctypes
155
2459b6e1 156
86e5f3ed 157class YoutubeDL:
8222d8de
JMF
158 """YoutubeDL class.
159
160 YoutubeDL objects are the ones responsible of downloading the
161 actual video file and writing it to disk if the user has requested
162 it, among some other tasks. In most cases there should be one per
163 program. As, given a video URL, the downloader doesn't know how to
164 extract all the needed information, task that InfoExtractors do, it
165 has to pass the URL to one of them.
166
167 For this, YoutubeDL objects have a method that allows
168 InfoExtractors to be registered in a given order. When it is passed
169 a URL, the YoutubeDL object handles it to the first InfoExtractor it
170 finds that reports being able to handle it. The InfoExtractor extracts
171 all the information about the video or videos the URL refers to, and
172 YoutubeDL process the extracted information, possibly using a File
173 Downloader to download the video.
174
175 YoutubeDL objects accept a lot of parameters. In order not to saturate
176 the object constructor with arguments, it receives a dictionary of
177 options instead. These options are available through the params
178 attribute for the InfoExtractors to use. The YoutubeDL also
179 registers itself as the downloader in charge for the InfoExtractors
180 that are added to it, so this is a "mutual registration".
181
182 Available options:
183
184 username: Username for authentication purposes.
185 password: Password for authentication purposes.
180940e0 186 videopassword: Password for accessing a video.
1da50aa3
S
187 ap_mso: Adobe Pass multiple-system operator identifier.
188 ap_username: Multiple-system operator account username.
189 ap_password: Multiple-system operator account password.
8222d8de
JMF
190 usenetrc: Use netrc for authentication instead.
191 verbose: Print additional info to stdout.
192 quiet: Do not print messages to stdout.
ad8915b7 193 no_warnings: Do not print out anything for warnings.
bb66c247 194 forceprint: A dict with keys WHEN mapped to a list of templates to
195 print to stdout. The allowed keys are video or any of the
196 items in utils.POSTPROCESS_WHEN.
ca30f449 197 For compatibility, a single list is also accepted
bb66c247 198 print_to_file: A dict with keys WHEN (same as forceprint) mapped to
199 a list of tuples with (template, filename)
8694c600 200 forcejson: Force printing info_dict as JSON.
63e0be34
PH
201 dump_single_json: Force printing the info_dict of the whole playlist
202 (or video) as a single JSON line.
c25228e5 203 force_write_download_archive: Force writing download archive regardless
204 of 'skip_download' or 'simulate'.
b7b04c78 205 simulate: Do not download the video files. If unset (or None),
206 simulate only if listsubtitles, listformats or list_thumbnails is used
eb8a4433 207 format: Video format code. see "FORMAT SELECTION" for more details.
093a1710 208 You can also pass a function. The function takes 'ctx' as
209 argument and returns the formats to download.
210 See "build_format_selector" for an implementation
63ad4d43 211 allow_unplayable_formats: Allow unplayable formats to be extracted and downloaded.
b7da73eb 212 ignore_no_formats_error: Ignore "No video formats" error. Usefull for
213 extracting metadata even if the video is not actually
214 available for download (experimental)
0930b11f 215 format_sort: A list of fields by which to sort the video formats.
216 See "Sorting Formats" for more details.
c25228e5 217 format_sort_force: Force the given format_sort. see "Sorting Formats"
218 for more details.
08d30158 219 prefer_free_formats: Whether to prefer video formats with free containers
220 over non-free ones of same quality.
c25228e5 221 allow_multiple_video_streams: Allow multiple video streams to be merged
222 into a single file
223 allow_multiple_audio_streams: Allow multiple audio streams to be merged
224 into a single file
0ba692ac 225 check_formats Whether to test if the formats are downloadable.
9f1a1c36 226 Can be True (check all), False (check none),
227 'selected' (check selected formats),
0ba692ac 228 or None (check only if requested by extractor)
4524baf0 229 paths: Dictionary of output paths. The allowed keys are 'home'
230 'temp' and the keys of OUTTMPL_TYPES (in utils.py)
de6000d9 231 outtmpl: Dictionary of templates for output names. Allowed keys
4524baf0 232 are 'default' and the keys of OUTTMPL_TYPES (in utils.py).
34488702 233 For compatibility with youtube-dl, a single string can also be used
a820dc72
RA
234 outtmpl_na_placeholder: Placeholder for unavailable meta fields.
235 restrictfilenames: Do not allow "&" and spaces in file names
236 trim_file_name: Limit length of filename (extension excluded)
4524baf0 237 windowsfilenames: Force the filenames to be windows compatible
b1940459 238 ignoreerrors: Do not stop on download/postprocessing errors.
239 Can be 'only_download' to ignore only download errors.
240 Default is 'only_download' for CLI, but False for API
26e2805c 241 skip_playlist_after_errors: Number of allowed failures until the rest of
242 the playlist is skipped
fe7866d0 243 allowed_extractors: List of regexes to match against extractor names that are allowed
0c3d0f51 244 overwrites: Overwrite all video and metadata files if True,
245 overwrite only non-video files if None
246 and don't overwrite any file if False
34488702 247 For compatibility with youtube-dl,
248 "nooverwrites" may also be used instead
c14e88f0 249 playlist_items: Specific indices of playlist to download.
75822ca7 250 playlistrandom: Download playlist items in random order.
7e9a6125 251 lazy_playlist: Process playlist entries as they are received.
8222d8de
JMF
252 matchtitle: Download only matching titles.
253 rejecttitle: Reject downloads for matching titles.
8bf9319e 254 logger: Log messages to a logging.Logger instance.
17ffed18 255 logtostderr: Print everything to stderr instead of stdout.
256 consoletitle: Display progress in console window's titlebar.
8222d8de
JMF
257 writedescription: Write the video description to a .description file
258 writeinfojson: Write the video description to a .info.json file
75d43ca0 259 clean_infojson: Remove private fields from the infojson
34488702 260 getcomments: Extract video comments. This will not be written to disk
06167fbb 261 unless writeinfojson is also given
1fb07d10 262 writeannotations: Write the video annotations to a .annotations.xml file
8222d8de 263 writethumbnail: Write the thumbnail image to a file
c25228e5 264 allow_playlist_files: Whether to write playlists' description, infojson etc
265 also to disk when using the 'write*' options
ec82d85a 266 write_all_thumbnails: Write all thumbnail formats to files
732044af 267 writelink: Write an internet shortcut file, depending on the
268 current platform (.url/.webloc/.desktop)
269 writeurllink: Write a Windows internet shortcut file (.url)
270 writewebloclink: Write a macOS internet shortcut file (.webloc)
271 writedesktoplink: Write a Linux internet shortcut file (.desktop)
8222d8de 272 writesubtitles: Write the video subtitles to a file
741dd8ea 273 writeautomaticsub: Write the automatically generated subtitles to a file
8222d8de 274 listsubtitles: Lists all available subtitles for the video
a504ced0 275 subtitlesformat: The format code for subtitles
c32b0aab 276 subtitleslangs: List of languages of the subtitles to download (can be regex).
277 The list may contain "all" to refer to all the available
278 subtitles. The language can be prefixed with a "-" to
62b58c09 279 exclude it from the requested languages, e.g. ['all', '-live_chat']
8222d8de
JMF
280 keepvideo: Keep the video file after post-processing
281 daterange: A DateRange object, download only if the upload_date is in the range.
282 skip_download: Skip the actual download of the video file
c35f9e72 283 cachedir: Location of the cache files in the filesystem.
a0e07d31 284 False to disable filesystem cache.
47192f92 285 noplaylist: Download single video instead of a playlist if in doubt.
8dbe9899
PH
286 age_limit: An integer representing the user's age in years.
287 Unsuitable videos for the given age are skipped.
5fe18bdb
PH
288 min_views: An integer representing the minimum view count the video
289 must have in order to not be skipped.
290 Videos without view count information are always
291 downloaded. None for no limit.
292 max_views: An integer representing the maximum view count.
293 Videos that are more popular than that are not
294 downloaded.
295 Videos without view count information are always
296 downloaded. None for no limit.
ae103564 297 download_archive: A set, or the name of a file where all downloads are recorded.
298 Videos already present in the file are not downloaded again.
8a51f564 299 break_on_existing: Stop the download process after attempting to download a
300 file that is in the archive.
301 break_on_reject: Stop the download process when encountering a video that
302 has been filtered out.
b222c271 303 break_per_url: Whether break_on_reject and break_on_existing
304 should act on each input URL as opposed to for the entire queue
d76fa1f3 305 cookiefile: File name or text stream from where cookies should be read and dumped to
f59f5ef8 306 cookiesfrombrowser: A tuple containing the name of the browser, the profile
9bd13fe5 307 name/path from where cookies are loaded, the name of the keyring,
308 and the container name, e.g. ('chrome', ) or
309 ('vivaldi', 'default', 'BASICTEXT') or ('firefox', 'default', None, 'Meta')
f81c62a6 310 legacyserverconnect: Explicitly allow HTTPS connection to servers that do not
311 support RFC 5746 secure renegotiation
f59f5ef8 312 nocheckcertificate: Do not verify SSL certificates
bb58c9ed 313 client_certificate: Path to client certificate file in PEM format. May include the private key
314 client_certificate_key: Path to private key file for client certificate
315 client_certificate_password: Password for client certificate private key, if encrypted.
316 If not provided and the key is encrypted, yt-dlp will ask interactively
7e8c0af0 317 prefer_insecure: Use HTTP instead of HTTPS to retrieve information.
c6e07cf1 318 (Only supported by some extractors)
8b7539d2 319 http_headers: A dictionary of custom headers to be used for all requests
a1ee09e8 320 proxy: URL of the proxy server to use
38cce791 321 geo_verification_proxy: URL of the proxy to use for IP address verification
504f20dd 322 on geo-restricted sites.
e344693b 323 socket_timeout: Time to wait for unresponsive hosts, in seconds
0783b09b
PH
324 bidi_workaround: Work around buggy terminals without bidirectional text
325 support, using fridibi
a0ddb8a2 326 debug_printtraffic:Print out sent and received HTTP traffic
04b4d394
PH
327 default_search: Prepend this string if an input url is not valid.
328 'auto' for elaborate guessing
62fec3b2 329 encoding: Use this encoding instead of the system-specified.
134c913c 330 extract_flat: Whether to resolve and process url_results further
331 * False: Always process (default)
332 * True: Never process
333 * 'in_playlist': Do not process inside playlist/multi_video
334 * 'discard': Always process, but don't return the result
335 from inside playlist/multi_video
336 * 'discard_in_playlist': Same as "discard", but only for
337 playlists (not multi_video)
f2ebc5c7 338 wait_for_video: If given, wait for scheduled streams to become available.
339 The value should be a tuple containing the range
340 (min_secs, max_secs) to wait between retries
4f026faf 341 postprocessors: A list of dictionaries, each with an entry
71b640cc 342 * key: The name of the postprocessor. See
7a5c1cfe 343 yt_dlp/postprocessor/__init__.py for a list.
bb66c247 344 * when: When to run the postprocessor. Allowed values are
345 the entries of utils.POSTPROCESS_WHEN
56d868db 346 Assumed to be 'post_process' if not given
71b640cc
PH
347 progress_hooks: A list of functions that get called on download
348 progress, with a dictionary with the entries
5cda4eda 349 * status: One of "downloading", "error", or "finished".
ee69b99a 350 Check this first and ignore unknown values.
3ba7740d 351 * info_dict: The extracted info_dict
71b640cc 352
5cda4eda 353 If status is one of "downloading", or "finished", the
ee69b99a
PH
354 following properties may also be present:
355 * filename: The final filename (always present)
5cda4eda 356 * tmpfilename: The filename we're currently writing to
71b640cc
PH
357 * downloaded_bytes: Bytes on disk
358 * total_bytes: Size of the whole file, None if unknown
5cda4eda
PH
359 * total_bytes_estimate: Guess of the eventual file size,
360 None if unavailable.
361 * elapsed: The number of seconds since download started.
71b640cc
PH
362 * eta: The estimated time in seconds, None if unknown
363 * speed: The download speed in bytes/second, None if
364 unknown
5cda4eda
PH
365 * fragment_index: The counter of the currently
366 downloaded video fragment.
367 * fragment_count: The number of fragments (= individual
368 files that will be merged)
71b640cc
PH
369
370 Progress hooks are guaranteed to be called at least once
371 (with status "finished") if the download is successful.
819e0531 372 postprocessor_hooks: A list of functions that get called on postprocessing
373 progress, with a dictionary with the entries
374 * status: One of "started", "processing", or "finished".
375 Check this first and ignore unknown values.
376 * postprocessor: Name of the postprocessor
377 * info_dict: The extracted info_dict
378
379 Progress hooks are guaranteed to be called at least twice
380 (with status "started" and "finished") if the processing is successful.
fc61aff4 381 merge_output_format: "/" separated list of extensions to use when merging formats.
6b591b29 382 final_ext: Expected final extension; used to detect when the file was
59a7a13e 383 already downloaded and converted
6271f1ca
PH
384 fixup: Automatically correct known faults of the file.
385 One of:
386 - "never": do nothing
387 - "warn": only emit a warning
388 - "detect_or_warn": check whether we can do anything
62cd676c 389 about it, warn otherwise (default)
504f20dd 390 source_address: Client-side IP address to bind to.
1cf376f5 391 sleep_interval_requests: Number of seconds to sleep between requests
392 during extraction
7aa589a5
S
393 sleep_interval: Number of seconds to sleep before each download when
394 used alone or a lower bound of a range for randomized
395 sleep before each download (minimum possible number
396 of seconds to sleep) when used along with
397 max_sleep_interval.
398 max_sleep_interval:Upper bound of a range for randomized sleep before each
399 download (maximum possible number of seconds to sleep).
400 Must only be used along with sleep_interval.
401 Actual sleep time will be a random float from range
402 [sleep_interval; max_sleep_interval].
1cf376f5 403 sleep_interval_subtitles: Number of seconds to sleep before each subtitle download
cfb56d1a
PH
404 listformats: Print an overview of available video formats and exit.
405 list_thumbnails: Print a table of all thumbnails and exit.
0a41f331 406 match_filter: A function that gets called for every video with the signature
407 (info_dict, *, incomplete: bool) -> Optional[str]
408 For backward compatibility with youtube-dl, the signature
409 (info_dict) -> Optional[str] is also allowed.
410 - If it returns a message, the video is ignored.
411 - If it returns None, the video is downloaded.
412 - If it returns utils.NO_DEFAULT, the user is interactively
413 asked whether to download the video.
347de493 414 match_filter_func in utils.py is one example for this.
7e5db8c9 415 no_color: Do not emit color codes in output.
0a840f58 416 geo_bypass: Bypass geographic restriction via faking X-Forwarded-For
504f20dd 417 HTTP header
0a840f58 418 geo_bypass_country:
773f291d
S
419 Two-letter ISO 3166-2 country code that will be used for
420 explicit geographic restriction bypassing via faking
504f20dd 421 X-Forwarded-For HTTP header
5f95927a
S
422 geo_bypass_ip_block:
423 IP range in CIDR notation that will be used similarly to
504f20dd 424 geo_bypass_country
52a8a1e1 425 external_downloader: A dictionary of protocol keys and the executable of the
426 external downloader to use for it. The allowed protocols
427 are default|http|ftp|m3u8|dash|rtsp|rtmp|mms.
428 Set the value to 'native' to use the native downloader
53ed7066 429 compat_opts: Compatibility options. See "Differences in default behavior".
3acf6d38 430 The following options do not work when used through the API:
b5ae35ee 431 filename, abort-on-error, multistreams, no-live-chat, format-sort
dac5df5a 432 no-clean-infojson, no-playlist-metafiles, no-keep-subs, no-attach-info-json.
e4f02757 433 Refer __init__.py for their implementation
819e0531 434 progress_template: Dictionary of templates for progress outputs.
435 Allowed keys are 'download', 'postprocess',
436 'download-title' (console title) and 'postprocess-title'.
437 The template is mapped on a dictionary with keys 'progress' and 'info'
23326151 438 retry_sleep_functions: Dictionary of functions that takes the number of attempts
439 as argument and returns the time to sleep in seconds.
440 Allowed keys are 'http', 'fragment', 'file_access'
0f446365
SW
441 download_ranges: A callback function that gets called for every video with
442 the signature (info_dict, ydl) -> Iterable[Section].
443 Only the returned sections will be downloaded.
444 Each Section is a dict with the following keys:
5ec1b6b7 445 * start_time: Start time of the section in seconds
446 * end_time: End time of the section in seconds
447 * title: Section title (Optional)
448 * index: Section number (Optional)
0f446365 449 force_keyframes_at_cuts: Re-encode the video when downloading ranges to get precise cuts
a7dc6a89 450 noprogress: Do not print the progress bar
a831c2ea 451 live_from_start: Whether to download livestreams videos from the start
fe7e0c98 452
8222d8de 453 The following parameters are not used by YoutubeDL itself, they are used by
7a5c1cfe 454 the downloader (see yt_dlp/downloader/common.py):
51d9739f 455 nopart, updatetime, buffersize, ratelimit, throttledratelimit, min_filesize,
205a0654 456 max_filesize, test, noresizebuffer, retries, file_access_retries, fragment_retries,
a7dc6a89 457 continuedl, xattr_set_filesize, hls_use_mpegts, http_chunk_size,
59a7a13e 458 external_downloader_args, concurrent_fragment_downloads.
76b1bd67
JMF
459
460 The following options are used by the post processors:
c0b7d117
S
461 ffmpeg_location: Location of the ffmpeg/avconv binary; either the path
462 to the binary or its containing directory.
43820c03 463 postprocessor_args: A dictionary of postprocessor/executable keys (in lower case)
34488702 464 and a list of additional command-line arguments for the
465 postprocessor/executable. The dict can also have "PP+EXE" keys
466 which are used when the given exe is used by the given PP.
467 Use 'default' as the name for arguments to passed to all PP
468 For compatibility with youtube-dl, a single list of args
469 can also be used
e409895f 470
471 The following options are used by the extractors:
62bff2c1 472 extractor_retries: Number of times to retry for known errors
473 dynamic_mpd: Whether to process dynamic DASH manifests (default: True)
e409895f 474 hls_split_discontinuity: Split HLS playlists to different formats at
62bff2c1 475 discontinuities such as ad breaks (default: False)
5d3a0e79 476 extractor_args: A dictionary of arguments to be passed to the extractors.
477 See "EXTRACTOR ARGUMENTS" for details.
62b58c09 478 E.g. {'youtube': {'skip': ['dash', 'hls']}}
88f23a18 479 mark_watched: Mark videos watched (even with --simulate). Only for YouTube
1890fc63 480
481 The following options are deprecated and may be removed in the future:
482
fe7866d0 483 force_generic_extractor: Force downloader to use the generic extractor
484 - Use allowed_extractors = ['generic', 'default']
7e9a6125 485 playliststart: - Use playlist_items
486 Playlist item to start at.
487 playlistend: - Use playlist_items
488 Playlist item to end at.
489 playlistreverse: - Use playlist_items
490 Download playlist items in reverse order.
1890fc63 491 forceurl: - Use forceprint
492 Force printing final URL.
493 forcetitle: - Use forceprint
494 Force printing title.
495 forceid: - Use forceprint
496 Force printing ID.
497 forcethumbnail: - Use forceprint
498 Force printing thumbnail URL.
499 forcedescription: - Use forceprint
500 Force printing description.
501 forcefilename: - Use forceprint
502 Force printing final filename.
503 forceduration: - Use forceprint
504 Force printing duration.
505 allsubtitles: - Use subtitleslangs = ['all']
506 Downloads all the subtitles of the video
507 (requires writesubtitles or writeautomaticsub)
508 include_ads: - Doesn't work
509 Download ads as well
510 call_home: - Not implemented
511 Boolean, true iff we are allowed to contact the
512 yt-dlp servers for debugging.
513 post_hooks: - Register a custom postprocessor
514 A list of functions that get called as the final step
515 for each video file, after all postprocessors have been
516 called. The filename will be passed as the only argument.
517 hls_prefer_native: - Use external_downloader = {'m3u8': 'native'} or {'m3u8': 'ffmpeg'}.
518 Use the native HLS downloader instead of ffmpeg/avconv
519 if True, otherwise use ffmpeg/avconv if False, otherwise
520 use downloader suggested by extractor if None.
521 prefer_ffmpeg: - avconv support is deprecated
522 If False, use avconv instead of ffmpeg if both are available,
523 otherwise prefer ffmpeg.
524 youtube_include_dash_manifest: - Use extractor_args
5d3a0e79 525 If True (default), DASH manifests and related
62bff2c1 526 data will be downloaded and processed by extractor.
527 You can reduce network I/O by disabling it if you don't
528 care about DASH. (only for youtube)
1890fc63 529 youtube_include_hls_manifest: - Use extractor_args
5d3a0e79 530 If True (default), HLS manifests and related
62bff2c1 531 data will be downloaded and processed by extractor.
532 You can reduce network I/O by disabling it if you don't
533 care about HLS. (only for youtube)
8222d8de
JMF
534 """
535
86e5f3ed 536 _NUMERIC_FIELDS = {
b8ed0f15 537 'width', 'height', 'asr', 'audio_channels', 'fps',
538 'tbr', 'abr', 'vbr', 'filesize', 'filesize_approx',
e6f21b3d 539 'timestamp', 'release_timestamp',
c9969434
S
540 'duration', 'view_count', 'like_count', 'dislike_count', 'repost_count',
541 'average_rating', 'comment_count', 'age_limit',
542 'start_time', 'end_time',
543 'chapter_number', 'season_number', 'episode_number',
544 'track_number', 'disc_number', 'release_year',
86e5f3ed 545 }
c9969434 546
6db9c4d5 547 _format_fields = {
548 # NB: Keep in sync with the docstring of extractor/common.py
a44ca5a4 549 'url', 'manifest_url', 'manifest_stream_number', 'ext', 'format', 'format_id', 'format_note',
b8ed0f15 550 'width', 'height', 'resolution', 'dynamic_range', 'tbr', 'abr', 'acodec', 'asr', 'audio_channels',
d5d1df8a 551 'vbr', 'fps', 'vcodec', 'container', 'filesize', 'filesize_approx', 'rows', 'columns',
6db9c4d5 552 'player_url', 'protocol', 'fragment_base_url', 'fragments', 'is_from_start',
553 'preference', 'language', 'language_preference', 'quality', 'source_preference',
554 'http_headers', 'stretched_ratio', 'no_resume', 'has_drm', 'downloader_options',
555 'page_url', 'app', 'play_path', 'tc_url', 'flash_version', 'rtmp_live', 'rtmp_conn', 'rtmp_protocol', 'rtmp_real_time'
556 }
48ee10ee 557 _format_selection_exts = {
8dc59305 558 'audio': set(MEDIA_EXTENSIONS.common_audio),
559 'video': set(MEDIA_EXTENSIONS.common_video + ('3gp', )),
560 'storyboards': set(MEDIA_EXTENSIONS.storyboards),
48ee10ee 561 }
562
3511266b 563 def __init__(self, params=None, auto_init=True):
883d4b1e 564 """Create a FileDownloader object with the given options.
565 @param auto_init Whether to load the default extractors and print header (if verbose).
49a57e70 566 Set to 'no_verbose_header' to not print the header
883d4b1e 567 """
e9f9a10f
JMF
568 if params is None:
569 params = {}
592b7485 570 self.params = params
8b7491c8 571 self._ies = {}
56c73665 572 self._ies_instances = {}
1e43a6f7 573 self._pps = {k: [] for k in POSTPROCESS_WHEN}
b35496d8 574 self._printed_messages = set()
1cf376f5 575 self._first_webpage_request = True
ab8e5e51 576 self._post_hooks = []
933605d7 577 self._progress_hooks = []
819e0531 578 self._postprocessor_hooks = []
8222d8de
JMF
579 self._download_retcode = 0
580 self._num_downloads = 0
9c906919 581 self._num_videos = 0
592b7485 582 self._playlist_level = 0
583 self._playlist_urls = set()
a0e07d31 584 self.cache = Cache(self)
34308b30 585
819e0531 586 windows_enable_vt_mode()
591bb9d3 587 stdout = sys.stderr if self.params.get('logtostderr') else sys.stdout
588 self._out_files = Namespace(
589 out=stdout,
590 error=sys.stderr,
591 screen=sys.stderr if self.params.get('quiet') else stdout,
592 console=None if compat_os_name == 'nt' else next(
cf4f42cb 593 filter(supports_terminal_sequences, (sys.stderr, sys.stdout)), None)
591bb9d3 594 )
595 self._allow_colors = Namespace(**{
596 type_: not self.params.get('no_color') and supports_terminal_sequences(stream)
64fa820c 597 for type_, stream in self._out_files.items_ if type_ != 'console'
591bb9d3 598 })
819e0531 599
6929b41a 600 # The code is left like this to be reused for future deprecations
601 MIN_SUPPORTED, MIN_RECOMMENDED = (3, 7), (3, 7)
eff42759 602 current_version = sys.version_info[:2]
603 if current_version < MIN_RECOMMENDED:
9d339c41 604 msg = ('Support for Python version %d.%d has been deprecated. '
24093d52 605 'See https://github.com/yt-dlp/yt-dlp/issues/3764 for more details.'
c6e07cf1 606 '\n You will no longer receive updates on this version')
eff42759 607 if current_version < MIN_SUPPORTED:
608 msg = 'Python version %d.%d is no longer supported'
609 self.deprecation_warning(
610 f'{msg}! Please update to Python %d.%d or above' % (*current_version, *MIN_RECOMMENDED))
a61f4b28 611
88acdbc2 612 if self.params.get('allow_unplayable_formats'):
613 self.report_warning(
ec11a9f4 614 f'You have asked for {self._format_err("UNPLAYABLE", self.Styles.EMPHASIS)} formats to be listed/downloaded. '
819e0531 615 'This is a developer option intended for debugging. \n'
616 ' If you experience any issues while using this option, '
ec11a9f4 617 f'{self._format_err("DO NOT", self.Styles.ERROR)} open a bug report')
88acdbc2 618
497074f0 619 if self.params.get('bidi_workaround', False):
620 try:
621 import pty
622 master, slave = pty.openpty()
623 width = shutil.get_terminal_size().columns
624 width_args = [] if width is None else ['-w', str(width)]
625 sp_kwargs = {'stdin': subprocess.PIPE, 'stdout': slave, 'stderr': self._out_files.error}
626 try:
627 self._output_process = Popen(['bidiv'] + width_args, **sp_kwargs)
628 except OSError:
629 self._output_process = Popen(['fribidi', '-c', 'UTF-8'] + width_args, **sp_kwargs)
630 self._output_channel = os.fdopen(master, 'rb')
631 except OSError as ose:
632 if ose.errno == errno.ENOENT:
633 self.report_warning(
634 'Could not find fribidi executable, ignoring --bidi-workaround. '
635 'Make sure that fribidi is an executable file in one of the directories in your $PATH.')
636 else:
637 raise
638
639 self.params['compat_opts'] = set(self.params.get('compat_opts', ()))
640 if auto_init and auto_init != 'no_verbose_header':
641 self.print_debug_header()
642
be5df5ee
S
643 def check_deprecated(param, option, suggestion):
644 if self.params.get(param) is not None:
86e5f3ed 645 self.report_warning(f'{option} is deprecated. Use {suggestion} instead')
be5df5ee
S
646 return True
647 return False
648
649 if check_deprecated('cn_verification_proxy', '--cn-verification-proxy', '--geo-verification-proxy'):
38cce791
YCH
650 if self.params.get('geo_verification_proxy') is None:
651 self.params['geo_verification_proxy'] = self.params['cn_verification_proxy']
652
0d1bb027 653 check_deprecated('autonumber', '--auto-number', '-o "%(autonumber)s-%(title)s.%(ext)s"')
654 check_deprecated('usetitle', '--title', '-o "%(title)s-%(id)s.%(ext)s"')
53ed7066 655 check_deprecated('useid', '--id', '-o "%(id)s.%(ext)s"')
0d1bb027 656
49a57e70 657 for msg in self.params.get('_warnings', []):
0d1bb027 658 self.report_warning(msg)
ee8dd27a 659 for msg in self.params.get('_deprecation_warnings', []):
da4db748 660 self.deprecated_feature(msg)
0d1bb027 661
8a82af35 662 if 'list-formats' in self.params['compat_opts']:
ec11a9f4 663 self.params['listformats_table'] = False
664
b5ae35ee 665 if 'overwrites' not in self.params and self.params.get('nooverwrites') is not None:
b868936c 666 # nooverwrites was unnecessarily changed to overwrites
667 # in 0c3d0f51778b153f65c21906031c2e091fcfb641
668 # This ensures compatibility with both keys
669 self.params['overwrites'] = not self.params['nooverwrites']
b5ae35ee 670 elif self.params.get('overwrites') is None:
671 self.params.pop('overwrites', None)
b868936c 672 else:
673 self.params['nooverwrites'] = not self.params['overwrites']
b9d973be 674
e4221b70 675 if self.params.get('simulate') is None and any((
676 self.params.get('list_thumbnails'),
677 self.params.get('listformats'),
678 self.params.get('listsubtitles'),
679 )):
680 self.params['simulate'] = 'list_only'
681
455a15e2 682 self.params.setdefault('forceprint', {})
683 self.params.setdefault('print_to_file', {})
bb66c247 684
685 # Compatibility with older syntax
ca30f449 686 if not isinstance(params['forceprint'], dict):
455a15e2 687 self.params['forceprint'] = {'video': params['forceprint']}
ca30f449 688
97ec5bc5 689 if auto_init:
97ec5bc5 690 self.add_default_info_extractors()
691
3089bc74
S
692 if (sys.platform != 'win32'
693 and sys.getfilesystemencoding() in ['ascii', 'ANSI_X3.4-1968']
455a15e2 694 and not self.params.get('restrictfilenames', False)):
e9137224 695 # Unicode filesystem API will throw errors (#1474, #13027)
34308b30 696 self.report_warning(
6febd1c1 697 'Assuming --restrict-filenames since file system encoding '
1b725173 698 'cannot encode all characters. '
6febd1c1 699 'Set the LC_ALL environment variable to fix this.')
4a98cdbf 700 self.params['restrictfilenames'] = True
34308b30 701
bf1824b3 702 self._parse_outtmpl()
486dd09e 703
187986a8 704 # Creating format selector here allows us to catch syntax errors before the extraction
705 self.format_selector = (
fa9f30b8 706 self.params.get('format') if self.params.get('format') in (None, '-')
093a1710 707 else self.params['format'] if callable(self.params['format'])
187986a8 708 else self.build_format_selector(self.params['format']))
709
8b7539d2 710 # Set http_headers defaults according to std_headers
711 self.params['http_headers'] = merge_headers(std_headers, self.params.get('http_headers', {}))
712
013b50b7 713 hooks = {
714 'post_hooks': self.add_post_hook,
715 'progress_hooks': self.add_progress_hook,
716 'postprocessor_hooks': self.add_postprocessor_hook,
717 }
718 for opt, fn in hooks.items():
719 for ph in self.params.get(opt, []):
720 fn(ph)
71b640cc 721
5bfc8bee 722 for pp_def_raw in self.params.get('postprocessors', []):
723 pp_def = dict(pp_def_raw)
724 when = pp_def.pop('when', 'post_process')
725 self.add_post_processor(
f9934b96 726 get_postprocessor(pp_def.pop('key'))(self, **pp_def),
5bfc8bee 727 when=when)
728
97ec5bc5 729 self._setup_opener()
51fb4995
YCH
730 register_socks_protocols()
731
ed39cac5 732 def preload_download_archive(fn):
733 """Preload the archive, if any is specified"""
ae103564 734 archive = set()
ed39cac5 735 if fn is None:
ae103564 736 return archive
941e881e 737 elif not is_path_like(fn):
ae103564 738 return fn
739
49a57e70 740 self.write_debug(f'Loading archive file {fn!r}')
ed39cac5 741 try:
742 with locked_file(fn, 'r', encoding='utf-8') as archive_file:
743 for line in archive_file:
ae103564 744 archive.add(line.strip())
86e5f3ed 745 except OSError as ioe:
ed39cac5 746 if ioe.errno != errno.ENOENT:
747 raise
ae103564 748 return archive
ed39cac5 749
ae103564 750 self.archive = preload_download_archive(self.params.get('download_archive'))
ed39cac5 751
7d4111ed
PH
752 def warn_if_short_id(self, argv):
753 # short YouTube ID starting with dash?
754 idxs = [
755 i for i, a in enumerate(argv)
756 if re.match(r'^-[0-9A-Za-z_-]{10}$', a)]
757 if idxs:
758 correct_argv = (
7a5c1cfe 759 ['yt-dlp']
3089bc74
S
760 + [a for i, a in enumerate(argv) if i not in idxs]
761 + ['--'] + [argv[i] for i in idxs]
7d4111ed
PH
762 )
763 self.report_warning(
764 'Long argument string detected. '
49a57e70 765 'Use -- to separate parameters and URLs, like this:\n%s' %
7d4111ed
PH
766 args_to_str(correct_argv))
767
8222d8de
JMF
768 def add_info_extractor(self, ie):
769 """Add an InfoExtractor object to the end of the list."""
8b7491c8 770 ie_key = ie.ie_key()
771 self._ies[ie_key] = ie
e52d7f85 772 if not isinstance(ie, type):
8b7491c8 773 self._ies_instances[ie_key] = ie
e52d7f85 774 ie.set_downloader(self)
8222d8de 775
56c73665
JMF
776 def get_info_extractor(self, ie_key):
777 """
778 Get an instance of an IE with name ie_key, it will try to get one from
779 the _ies list, if there's no instance it will create a new one and add
780 it to the extractor list.
781 """
782 ie = self._ies_instances.get(ie_key)
783 if ie is None:
784 ie = get_info_extractor(ie_key)()
785 self.add_info_extractor(ie)
786 return ie
787
023fa8c4
JMF
788 def add_default_info_extractors(self):
789 """
790 Add the InfoExtractors returned by gen_extractors to the end of the list
791 """
fe7866d0 792 all_ies = {ie.IE_NAME.lower(): ie for ie in gen_extractor_classes()}
793 all_ies['end'] = UnsupportedURLIE()
794 try:
795 ie_names = orderedSet_from_options(
796 self.params.get('allowed_extractors', ['default']), {
797 'all': list(all_ies),
798 'default': [name for name, ie in all_ies.items() if ie._ENABLED],
799 }, use_regex=True)
800 except re.error as e:
801 raise ValueError(f'Wrong regex for allowed_extractors: {e.pattern}')
802 for name in ie_names:
803 self.add_info_extractor(all_ies[name])
804 self.write_debug(f'Loaded {len(ie_names)} extractors')
023fa8c4 805
56d868db 806 def add_post_processor(self, pp, when='post_process'):
8222d8de 807 """Add a PostProcessor object to the end of the chain."""
8aa0e7cd 808 assert when in POSTPROCESS_WHEN, f'Invalid when={when}'
5bfa4862 809 self._pps[when].append(pp)
8222d8de
JMF
810 pp.set_downloader(self)
811
ab8e5e51
AM
812 def add_post_hook(self, ph):
813 """Add the post hook"""
814 self._post_hooks.append(ph)
815
933605d7 816 def add_progress_hook(self, ph):
819e0531 817 """Add the download progress hook"""
933605d7 818 self._progress_hooks.append(ph)
8ab470f1 819
819e0531 820 def add_postprocessor_hook(self, ph):
821 """Add the postprocessing progress hook"""
822 self._postprocessor_hooks.append(ph)
5bfc8bee 823 for pps in self._pps.values():
824 for pp in pps:
825 pp.add_progress_hook(ph)
819e0531 826
1c088fa8 827 def _bidi_workaround(self, message):
5d681e96 828 if not hasattr(self, '_output_channel'):
1c088fa8
PH
829 return message
830
5d681e96 831 assert hasattr(self, '_output_process')
14f25df2 832 assert isinstance(message, str)
6febd1c1 833 line_count = message.count('\n') + 1
0f06bcd7 834 self._output_process.stdin.write((message + '\n').encode())
5d681e96 835 self._output_process.stdin.flush()
0f06bcd7 836 res = ''.join(self._output_channel.readline().decode()
9e1a5b84 837 for _ in range(line_count))
6febd1c1 838 return res[:-len('\n')]
1c088fa8 839
b35496d8 840 def _write_string(self, message, out=None, only_once=False):
841 if only_once:
842 if message in self._printed_messages:
843 return
844 self._printed_messages.add(message)
845 write_string(message, out=out, encoding=self.params.get('encoding'))
734f90bb 846
cf4f42cb 847 def to_stdout(self, message, skip_eol=False, quiet=None):
0760b0a7 848 """Print message to stdout"""
cf4f42cb 849 if quiet is not None:
da4db748 850 self.deprecation_warning('"YoutubeDL.to_stdout" no longer accepts the argument quiet. '
851 'Use "YoutubeDL.to_screen" instead')
8a82af35 852 if skip_eol is not False:
da4db748 853 self.deprecation_warning('"YoutubeDL.to_stdout" no longer accepts the argument skip_eol. '
854 'Use "YoutubeDL.to_screen" instead')
0bf9dc1e 855 self._write_string(f'{self._bidi_workaround(message)}\n', self._out_files.out)
cf4f42cb 856
dfea94f8 857 def to_screen(self, message, skip_eol=False, quiet=None, only_once=False):
cf4f42cb 858 """Print message to screen if not in quiet mode"""
8bf9319e 859 if self.params.get('logger'):
43afe285 860 self.params['logger'].debug(message)
cf4f42cb 861 return
862 if (self.params.get('quiet') if quiet is None else quiet) and not self.params.get('verbose'):
863 return
864 self._write_string(
865 '%s%s' % (self._bidi_workaround(message), ('' if skip_eol else '\n')),
dfea94f8 866 self._out_files.screen, only_once=only_once)
8222d8de 867
b35496d8 868 def to_stderr(self, message, only_once=False):
0760b0a7 869 """Print message to stderr"""
14f25df2 870 assert isinstance(message, str)
8bf9319e 871 if self.params.get('logger'):
43afe285
IB
872 self.params['logger'].error(message)
873 else:
5792c950 874 self._write_string(f'{self._bidi_workaround(message)}\n', self._out_files.error, only_once=only_once)
cf4f42cb 875
876 def _send_console_code(self, code):
591bb9d3 877 if compat_os_name == 'nt' or not self._out_files.console:
cf4f42cb 878 return
591bb9d3 879 self._write_string(code, self._out_files.console)
8222d8de 880
1e5b9a95
PH
881 def to_console_title(self, message):
882 if not self.params.get('consoletitle', False):
883 return
3efb96a6 884 message = remove_terminal_sequences(message)
4bede0d8
C
885 if compat_os_name == 'nt':
886 if ctypes.windll.kernel32.GetConsoleWindow():
887 # c_wchar_p() might not be necessary if `message` is
888 # already of type unicode()
889 ctypes.windll.kernel32.SetConsoleTitleW(ctypes.c_wchar_p(message))
cf4f42cb 890 else:
891 self._send_console_code(f'\033]0;{message}\007')
1e5b9a95 892
bdde425c 893 def save_console_title(self):
cf4f42cb 894 if not self.params.get('consoletitle') or self.params.get('simulate'):
bdde425c 895 return
592b7485 896 self._send_console_code('\033[22;0t') # Save the title on stack
bdde425c
PH
897
898 def restore_console_title(self):
cf4f42cb 899 if not self.params.get('consoletitle') or self.params.get('simulate'):
bdde425c 900 return
592b7485 901 self._send_console_code('\033[23;0t') # Restore the title from stack
bdde425c
PH
902
903 def __enter__(self):
904 self.save_console_title()
905 return self
906
907 def __exit__(self, *args):
908 self.restore_console_title()
f89197d7 909
dca08720 910 if self.params.get('cookiefile') is not None:
1bab3437 911 self.cookiejar.save(ignore_discard=True, ignore_expires=True)
bdde425c 912
fa9f30b8 913 def trouble(self, message=None, tb=None, is_error=True):
8222d8de
JMF
914 """Determine action to take when a download problem appears.
915
916 Depending on if the downloader has been configured to ignore
917 download errors or not, this method may throw an exception or
918 not when errors are found, after printing the message.
919
fa9f30b8 920 @param tb If given, is additional traceback information
921 @param is_error Whether to raise error according to ignorerrors
8222d8de
JMF
922 """
923 if message is not None:
924 self.to_stderr(message)
925 if self.params.get('verbose'):
926 if tb is None:
927 if sys.exc_info()[0]: # if .trouble has been called from an except block
6febd1c1 928 tb = ''
8222d8de 929 if hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
6febd1c1 930 tb += ''.join(traceback.format_exception(*sys.exc_info()[1].exc_info))
c0384f22 931 tb += encode_compat_str(traceback.format_exc())
8222d8de
JMF
932 else:
933 tb_data = traceback.format_list(traceback.extract_stack())
6febd1c1 934 tb = ''.join(tb_data)
c19bc311 935 if tb:
936 self.to_stderr(tb)
fa9f30b8 937 if not is_error:
938 return
b1940459 939 if not self.params.get('ignoreerrors'):
8222d8de
JMF
940 if sys.exc_info()[0] and hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
941 exc_info = sys.exc_info()[1].exc_info
942 else:
943 exc_info = sys.exc_info()
944 raise DownloadError(message, exc_info)
945 self._download_retcode = 1
946
19a03940 947 Styles = Namespace(
948 HEADERS='yellow',
949 EMPHASIS='light blue',
492272fe 950 FILENAME='green',
19a03940 951 ID='green',
952 DELIM='blue',
953 ERROR='red',
954 WARNING='yellow',
955 SUPPRESS='light black',
956 )
ec11a9f4 957
7578d77d 958 def _format_text(self, handle, allow_colors, text, f, fallback=None, *, test_encoding=False):
e5a998f3 959 text = str(text)
ec11a9f4 960 if test_encoding:
961 original_text = text
5c104538 962 # handle.encoding can be None. See https://github.com/yt-dlp/yt-dlp/issues/2711
963 encoding = self.params.get('encoding') or getattr(handle, 'encoding', None) or 'ascii'
ec11a9f4 964 text = text.encode(encoding, 'ignore').decode(encoding)
965 if fallback is not None and text != original_text:
966 text = fallback
7578d77d 967 return format_text(text, f) if allow_colors else text if fallback is None else fallback
ec11a9f4 968
591bb9d3 969 def _format_out(self, *args, **kwargs):
970 return self._format_text(self._out_files.out, self._allow_colors.out, *args, **kwargs)
971
ec11a9f4 972 def _format_screen(self, *args, **kwargs):
591bb9d3 973 return self._format_text(self._out_files.screen, self._allow_colors.screen, *args, **kwargs)
ec11a9f4 974
975 def _format_err(self, *args, **kwargs):
591bb9d3 976 return self._format_text(self._out_files.error, self._allow_colors.error, *args, **kwargs)
819e0531 977
c84aeac6 978 def report_warning(self, message, only_once=False):
8222d8de
JMF
979 '''
980 Print the message to stderr, it will be prefixed with 'WARNING:'
981 If stderr is a tty file the 'WARNING:' will be colored
982 '''
6d07ce01
JMF
983 if self.params.get('logger') is not None:
984 self.params['logger'].warning(message)
8222d8de 985 else:
ad8915b7
PH
986 if self.params.get('no_warnings'):
987 return
ec11a9f4 988 self.to_stderr(f'{self._format_err("WARNING:", self.Styles.WARNING)} {message}', only_once)
8222d8de 989
da4db748 990 def deprecation_warning(self, message, *, stacklevel=0):
991 deprecation_warning(
992 message, stacklevel=stacklevel + 1, printer=self.report_error, is_error=False)
993
994 def deprecated_feature(self, message):
ee8dd27a 995 if self.params.get('logger') is not None:
da4db748 996 self.params['logger'].warning(f'Deprecated Feature: {message}')
997 self.to_stderr(f'{self._format_err("Deprecated Feature:", self.Styles.ERROR)} {message}', True)
ee8dd27a 998
fa9f30b8 999 def report_error(self, message, *args, **kwargs):
8222d8de
JMF
1000 '''
1001 Do the same as trouble, but prefixes the message with 'ERROR:', colored
1002 in red if stderr is a tty file.
1003 '''
fa9f30b8 1004 self.trouble(f'{self._format_err("ERROR:", self.Styles.ERROR)} {message}', *args, **kwargs)
8222d8de 1005
b35496d8 1006 def write_debug(self, message, only_once=False):
0760b0a7 1007 '''Log debug message or Print message to stderr'''
1008 if not self.params.get('verbose', False):
1009 return
8a82af35 1010 message = f'[debug] {message}'
0760b0a7 1011 if self.params.get('logger'):
1012 self.params['logger'].debug(message)
1013 else:
b35496d8 1014 self.to_stderr(message, only_once)
0760b0a7 1015
8222d8de
JMF
1016 def report_file_already_downloaded(self, file_name):
1017 """Report file has already been fully downloaded."""
1018 try:
6febd1c1 1019 self.to_screen('[download] %s has already been downloaded' % file_name)
ce02ed60 1020 except UnicodeEncodeError:
6febd1c1 1021 self.to_screen('[download] The file has already been downloaded')
8222d8de 1022
0c3d0f51 1023 def report_file_delete(self, file_name):
1024 """Report that existing file will be deleted."""
1025 try:
c25228e5 1026 self.to_screen('Deleting existing file %s' % file_name)
0c3d0f51 1027 except UnicodeEncodeError:
c25228e5 1028 self.to_screen('Deleting existing file')
0c3d0f51 1029
319b6059 1030 def raise_no_formats(self, info, forced=False, *, msg=None):
0a5a191a 1031 has_drm = info.get('_has_drm')
319b6059 1032 ignored, expected = self.params.get('ignore_no_formats_error'), bool(msg)
1033 msg = msg or has_drm and 'This video is DRM protected' or 'No video formats found!'
1034 if forced or not ignored:
1151c407 1035 raise ExtractorError(msg, video_id=info['id'], ie=info['extractor'],
319b6059 1036 expected=has_drm or ignored or expected)
88acdbc2 1037 else:
1038 self.report_warning(msg)
1039
de6000d9 1040 def parse_outtmpl(self):
bf1824b3 1041 self.deprecation_warning('"YoutubeDL.parse_outtmpl" is deprecated and may be removed in a future version')
1042 self._parse_outtmpl()
1043 return self.params['outtmpl']
1044
1045 def _parse_outtmpl(self):
7b2c3f47 1046 sanitize = IDENTITY
bf1824b3 1047 if self.params.get('restrictfilenames'): # Remove spaces in the default template
71ce444a 1048 sanitize = lambda x: x.replace(' - ', ' ').replace(' ', '-')
bf1824b3 1049
1050 outtmpl = self.params.setdefault('outtmpl', {})
1051 if not isinstance(outtmpl, dict):
1052 self.params['outtmpl'] = outtmpl = {'default': outtmpl}
1053 outtmpl.update({k: sanitize(v) for k, v in DEFAULT_OUTTMPL.items() if outtmpl.get(k) is None})
de6000d9 1054
21cd8fae 1055 def get_output_path(self, dir_type='', filename=None):
1056 paths = self.params.get('paths', {})
d2c8aadf 1057 assert isinstance(paths, dict), '"paths" parameter must be a dictionary'
21cd8fae 1058 path = os.path.join(
1059 expand_path(paths.get('home', '').strip()),
1060 expand_path(paths.get(dir_type, '').strip()) if dir_type else '',
1061 filename or '')
21cd8fae 1062 return sanitize_path(path, force=self.params.get('windowsfilenames'))
1063
76a264ac 1064 @staticmethod
901130bb 1065 def _outtmpl_expandpath(outtmpl):
1066 # expand_path translates '%%' into '%' and '$$' into '$'
1067 # correspondingly that is not what we want since we need to keep
1068 # '%%' intact for template dict substitution step. Working around
1069 # with boundary-alike separator hack.
1070 sep = ''.join([random.choice(ascii_letters) for _ in range(32)])
86e5f3ed 1071 outtmpl = outtmpl.replace('%%', f'%{sep}%').replace('$$', f'${sep}$')
901130bb 1072
1073 # outtmpl should be expand_path'ed before template dict substitution
1074 # because meta fields may contain env variables we don't want to
62b58c09 1075 # be expanded. E.g. for outtmpl "%(title)s.%(ext)s" and
901130bb 1076 # title "Hello $PATH", we don't want `$PATH` to be expanded.
1077 return expand_path(outtmpl).replace(sep, '')
1078
1079 @staticmethod
1080 def escape_outtmpl(outtmpl):
1081 ''' Escape any remaining strings like %s, %abc% etc. '''
1082 return re.sub(
1083 STR_FORMAT_RE_TMPL.format('', '(?![%(\0])'),
1084 lambda mobj: ('' if mobj.group('has_key') else '%') + mobj.group(0),
1085 outtmpl)
1086
1087 @classmethod
1088 def validate_outtmpl(cls, outtmpl):
76a264ac 1089 ''' @return None or Exception object '''
7d1eb38a 1090 outtmpl = re.sub(
47cdc68e 1091 STR_FORMAT_RE_TMPL.format('[^)]*', '[ljhqBUDS]'),
7d1eb38a 1092 lambda mobj: f'{mobj.group(0)[:-1]}s',
1093 cls._outtmpl_expandpath(outtmpl))
76a264ac 1094 try:
7d1eb38a 1095 cls.escape_outtmpl(outtmpl) % collections.defaultdict(int)
76a264ac 1096 return None
1097 except ValueError as err:
1098 return err
1099
03b4de72 1100 @staticmethod
1101 def _copy_infodict(info_dict):
1102 info_dict = dict(info_dict)
09b49e1f 1103 info_dict.pop('__postprocessors', None)
415f8d51 1104 info_dict.pop('__pending_error', None)
03b4de72 1105 return info_dict
1106
e0fd9573 1107 def prepare_outtmpl(self, outtmpl, info_dict, sanitize=False):
1108 """ Make the outtmpl and info_dict suitable for substitution: ydl.escape_outtmpl(outtmpl) % info_dict
1109 @param sanitize Whether to sanitize the output as a filename.
1110 For backward compatibility, a function can also be passed
1111 """
1112
6e84b215 1113 info_dict.setdefault('epoch', int(time.time())) # keep epoch consistent once set
143db31d 1114
03b4de72 1115 info_dict = self._copy_infodict(info_dict)
752cda38 1116 info_dict['duration_string'] = ( # %(duration>%H-%M-%S)s is wrong if duration > 24hrs
53c18592 1117 formatSeconds(info_dict['duration'], '-' if sanitize else ':')
143db31d 1118 if info_dict.get('duration', None) is not None
1119 else None)
1d485a1a 1120 info_dict['autonumber'] = int(self.params.get('autonumber_start', 1) - 1 + self._num_downloads)
9c906919 1121 info_dict['video_autonumber'] = self._num_videos
752cda38 1122 if info_dict.get('resolution') is None:
1123 info_dict['resolution'] = self.format_resolution(info_dict, default=None)
143db31d 1124
e6f21b3d 1125 # For fields playlist_index, playlist_autonumber and autonumber convert all occurrences
143db31d 1126 # of %(field)s to %(field)0Nd for backward compatibility
1127 field_size_compat_map = {
0a5a191a 1128 'playlist_index': number_of_digits(info_dict.get('__last_playlist_index') or 0),
ec11a9f4 1129 'playlist_autonumber': number_of_digits(info_dict.get('n_entries') or 0),
752cda38 1130 'autonumber': self.params.get('autonumber_size') or 5,
143db31d 1131 }
752cda38 1132
385a27fa 1133 TMPL_DICT = {}
47cdc68e 1134 EXTERNAL_FORMAT_RE = re.compile(STR_FORMAT_RE_TMPL.format('[^)]*', f'[{STR_FORMAT_TYPES}ljhqBUDS]'))
385a27fa 1135 MATH_FUNCTIONS = {
1136 '+': float.__add__,
1137 '-': float.__sub__,
1138 }
e625be0d 1139 # Field is of the form key1.key2...
07a1250e 1140 # where keys (except first) can be string, int, slice or "{field, ...}"
1141 FIELD_INNER_RE = r'(?:\w+|%(num)s|%(num)s?(?::%(num)s?){1,2})' % {'num': r'(?:-?\d+)'}
1142 FIELD_RE = r'\w*(?:\.(?:%(inner)s|{%(field)s(?:,%(field)s)*}))*' % {
1143 'inner': FIELD_INNER_RE,
1144 'field': rf'\w*(?:\.{FIELD_INNER_RE})*'
1145 }
1d485a1a 1146 MATH_FIELD_RE = rf'(?:{FIELD_RE}|-?{NUMBER_RE})'
385a27fa 1147 MATH_OPERATORS_RE = r'(?:%s)' % '|'.join(map(re.escape, MATH_FUNCTIONS.keys()))
1d485a1a 1148 INTERNAL_FORMAT_RE = re.compile(rf'''(?x)
e625be0d 1149 (?P<negate>-)?
1d485a1a 1150 (?P<fields>{FIELD_RE})
1151 (?P<maths>(?:{MATH_OPERATORS_RE}{MATH_FIELD_RE})*)
e625be0d 1152 (?:>(?P<strf_format>.+?))?
34baa9fd 1153 (?P<remaining>
1154 (?P<alternate>(?<!\\),[^|&)]+)?
1155 (?:&(?P<replacement>.*?))?
1156 (?:\|(?P<default>.*?))?
1d485a1a 1157 )$''')
752cda38 1158
07a1250e 1159 def _traverse_infodict(fields):
1160 fields = [f for x in re.split(r'\.({.+?})\.?', fields)
1161 for f in ([x] if x.startswith('{') else x.split('.'))]
1162 for i in (0, -1):
1163 if fields and not fields[i]:
1164 fields.pop(i)
1165
1166 for i, f in enumerate(fields):
1167 if not f.startswith('{'):
1168 continue
1169 assert f.endswith('}'), f'No closing brace for {f} in {fields}'
1170 fields[i] = {k: k.split('.') for k in f[1:-1].split(',')}
1171
1172 return traverse_obj(info_dict, fields, is_user_input=True, traverse_string=True)
76a264ac 1173
752cda38 1174 def get_value(mdict):
1175 # Object traversal
2b8a2973 1176 value = _traverse_infodict(mdict['fields'])
752cda38 1177 # Negative
1178 if mdict['negate']:
1179 value = float_or_none(value)
1180 if value is not None:
1181 value *= -1
1182 # Do maths
385a27fa 1183 offset_key = mdict['maths']
1184 if offset_key:
752cda38 1185 value = float_or_none(value)
1186 operator = None
385a27fa 1187 while offset_key:
1188 item = re.match(
1189 MATH_FIELD_RE if operator else MATH_OPERATORS_RE,
1190 offset_key).group(0)
1191 offset_key = offset_key[len(item):]
1192 if operator is None:
752cda38 1193 operator = MATH_FUNCTIONS[item]
385a27fa 1194 continue
1195 item, multiplier = (item[1:], -1) if item[0] == '-' else (item, 1)
1196 offset = float_or_none(item)
1197 if offset is None:
2b8a2973 1198 offset = float_or_none(_traverse_infodict(item))
385a27fa 1199 try:
1200 value = operator(value, multiplier * offset)
1201 except (TypeError, ZeroDivisionError):
1202 return None
1203 operator = None
752cda38 1204 # Datetime formatting
1205 if mdict['strf_format']:
7c37ff97 1206 value = strftime_or_none(value, mdict['strf_format'].replace('\\,', ','))
752cda38 1207
a6bcaf71 1208 # XXX: Workaround for https://github.com/yt-dlp/yt-dlp/issues/4485
1209 if sanitize and value == '':
1210 value = None
752cda38 1211 return value
1212
b868936c 1213 na = self.params.get('outtmpl_na_placeholder', 'NA')
1214
e0fd9573 1215 def filename_sanitizer(key, value, restricted=self.params.get('restrictfilenames')):
5c3895ff 1216 return sanitize_filename(str(value), restricted=restricted, is_id=(
1217 bool(re.search(r'(^|[_.])id(\.|$)', key))
8a82af35 1218 if 'filename-sanitization' in self.params['compat_opts']
5c3895ff 1219 else NO_DEFAULT))
e0fd9573 1220
1221 sanitizer = sanitize if callable(sanitize) else filename_sanitizer
1222 sanitize = bool(sanitize)
1223
6e84b215 1224 def _dumpjson_default(obj):
1225 if isinstance(obj, (set, LazyList)):
1226 return list(obj)
adbc4ec4 1227 return repr(obj)
6e84b215 1228
752cda38 1229 def create_key(outer_mobj):
1230 if not outer_mobj.group('has_key'):
b836dc94 1231 return outer_mobj.group(0)
752cda38 1232 key = outer_mobj.group('key')
752cda38 1233 mobj = re.match(INTERNAL_FORMAT_RE, key)
e0fd9573 1234 initial_field = mobj.group('fields') if mobj else ''
e978789f 1235 value, replacement, default = None, None, na
7c37ff97 1236 while mobj:
e625be0d 1237 mobj = mobj.groupdict()
7c37ff97 1238 default = mobj['default'] if mobj['default'] is not None else default
752cda38 1239 value = get_value(mobj)
e978789f 1240 replacement = mobj['replacement']
7c37ff97 1241 if value is None and mobj['alternate']:
34baa9fd 1242 mobj = re.match(INTERNAL_FORMAT_RE, mobj['remaining'][1:])
7c37ff97 1243 else:
1244 break
752cda38 1245
b868936c 1246 fmt = outer_mobj.group('format')
752cda38 1247 if fmt == 's' and value is not None and key in field_size_compat_map.keys():
86e5f3ed 1248 fmt = f'0{field_size_compat_map[key]:d}d'
752cda38 1249
e978789f 1250 value = default if value is None else value if replacement is None else replacement
752cda38 1251
4476d2c7 1252 flags = outer_mobj.group('conversion') or ''
7d1eb38a 1253 str_fmt = f'{fmt[:-1]}s'
524e2e4f 1254 if fmt[-1] == 'l': # list
4476d2c7 1255 delim = '\n' if '#' in flags else ', '
9e907ebd 1256 value, fmt = delim.join(map(str, variadic(value, allowed_types=(str, bytes)))), str_fmt
524e2e4f 1257 elif fmt[-1] == 'j': # json
deae7c17 1258 value, fmt = json.dumps(
1259 value, default=_dumpjson_default,
9b9dad11 1260 indent=4 if '#' in flags else None, ensure_ascii='+' not in flags), str_fmt
47cdc68e 1261 elif fmt[-1] == 'h': # html
deae7c17 1262 value, fmt = escapeHTML(str(value)), str_fmt
524e2e4f 1263 elif fmt[-1] == 'q': # quoted
4476d2c7 1264 value = map(str, variadic(value) if '#' in flags else [value])
1265 value, fmt = ' '.join(map(compat_shlex_quote, value)), str_fmt
524e2e4f 1266 elif fmt[-1] == 'B': # bytes
0f06bcd7 1267 value = f'%{str_fmt}'.encode() % str(value).encode()
f5aa5cfb 1268 value, fmt = value.decode('utf-8', 'ignore'), 's'
524e2e4f 1269 elif fmt[-1] == 'U': # unicode normalized
524e2e4f 1270 value, fmt = unicodedata.normalize(
1271 # "+" = compatibility equivalence, "#" = NFD
4476d2c7 1272 'NF%s%s' % ('K' if '+' in flags else '', 'D' if '#' in flags else 'C'),
524e2e4f 1273 value), str_fmt
e0fd9573 1274 elif fmt[-1] == 'D': # decimal suffix
abbeeebc 1275 num_fmt, fmt = fmt[:-1].replace('#', ''), 's'
1276 value = format_decimal_suffix(value, f'%{num_fmt}f%s' if num_fmt else '%d%s',
1277 factor=1024 if '#' in flags else 1000)
37893bb0 1278 elif fmt[-1] == 'S': # filename sanitization
e0fd9573 1279 value, fmt = filename_sanitizer(initial_field, value, restricted='#' in flags), str_fmt
7d1eb38a 1280 elif fmt[-1] == 'c':
524e2e4f 1281 if value:
1282 value = str(value)[0]
76a264ac 1283 else:
524e2e4f 1284 fmt = str_fmt
76a264ac 1285 elif fmt[-1] not in 'rs': # numeric
a439a3a4 1286 value = float_or_none(value)
752cda38 1287 if value is None:
1288 value, fmt = default, 's'
901130bb 1289
752cda38 1290 if sanitize:
1291 if fmt[-1] == 'r':
1292 # If value is an object, sanitize might convert it to a string
1293 # So we convert it to repr first
7d1eb38a 1294 value, fmt = repr(value), str_fmt
639f1cea 1295 if fmt[-1] in 'csr':
e0fd9573 1296 value = sanitizer(initial_field, value)
901130bb 1297
b868936c 1298 key = '%s\0%s' % (key.replace('%', '%\0'), outer_mobj.group('format'))
385a27fa 1299 TMPL_DICT[key] = value
b868936c 1300 return '{prefix}%({key}){fmt}'.format(key=key, fmt=fmt, prefix=outer_mobj.group('prefix'))
752cda38 1301
385a27fa 1302 return EXTERNAL_FORMAT_RE.sub(create_key, outtmpl), TMPL_DICT
143db31d 1303
819e0531 1304 def evaluate_outtmpl(self, outtmpl, info_dict, *args, **kwargs):
1305 outtmpl, info_dict = self.prepare_outtmpl(outtmpl, info_dict, *args, **kwargs)
1306 return self.escape_outtmpl(outtmpl) % info_dict
1307
5127e92a 1308 def _prepare_filename(self, info_dict, *, outtmpl=None, tmpl_type=None):
1309 assert None in (outtmpl, tmpl_type), 'outtmpl and tmpl_type are mutually exclusive'
1310 if outtmpl is None:
bf1824b3 1311 outtmpl = self.params['outtmpl'].get(tmpl_type or 'default', self.params['outtmpl']['default'])
8222d8de 1312 try:
5127e92a 1313 outtmpl = self._outtmpl_expandpath(outtmpl)
e0fd9573 1314 filename = self.evaluate_outtmpl(outtmpl, info_dict, True)
6a0546e3 1315 if not filename:
1316 return None
15da37c7 1317
5127e92a 1318 if tmpl_type in ('', 'temp'):
6a0546e3 1319 final_ext, ext = self.params.get('final_ext'), info_dict.get('ext')
1320 if final_ext and ext and final_ext != ext and filename.endswith(f'.{final_ext}'):
1321 filename = replace_extension(filename, ext, final_ext)
5127e92a 1322 elif tmpl_type:
6a0546e3 1323 force_ext = OUTTMPL_TYPES[tmpl_type]
1324 if force_ext:
1325 filename = replace_extension(filename, force_ext, info_dict.get('ext'))
de6000d9 1326
bdc3fd2f
U
1327 # https://github.com/blackjack4494/youtube-dlc/issues/85
1328 trim_file_name = self.params.get('trim_file_name', False)
1329 if trim_file_name:
5c22c63d 1330 no_ext, *ext = filename.rsplit('.', 2)
1331 filename = join_nonempty(no_ext[:trim_file_name], *ext, delim='.')
bdc3fd2f 1332
0202b52a 1333 return filename
8222d8de 1334 except ValueError as err:
6febd1c1 1335 self.report_error('Error in output template: ' + str(err) + ' (encoding: ' + repr(preferredencoding()) + ')')
8222d8de
JMF
1336 return None
1337
5127e92a 1338 def prepare_filename(self, info_dict, dir_type='', *, outtmpl=None, warn=False):
1339 """Generate the output filename"""
1340 if outtmpl:
1341 assert not dir_type, 'outtmpl and dir_type are mutually exclusive'
1342 dir_type = None
1343 filename = self._prepare_filename(info_dict, tmpl_type=dir_type, outtmpl=outtmpl)
80c03fa9 1344 if not filename and dir_type not in ('', 'temp'):
1345 return ''
de6000d9 1346
c84aeac6 1347 if warn:
21cd8fae 1348 if not self.params.get('paths'):
de6000d9 1349 pass
1350 elif filename == '-':
c84aeac6 1351 self.report_warning('--paths is ignored when an outputting to stdout', only_once=True)
de6000d9 1352 elif os.path.isabs(filename):
c84aeac6 1353 self.report_warning('--paths is ignored since an absolute path is given in output template', only_once=True)
de6000d9 1354 if filename == '-' or not filename:
1355 return filename
1356
21cd8fae 1357 return self.get_output_path(dir_type, filename)
0202b52a 1358
120fe513 1359 def _match_entry(self, info_dict, incomplete=False, silent=False):
ecdec191 1360 """ Returns None if the file should be downloaded """
8222d8de 1361
3bec830a 1362 video_title = info_dict.get('title', info_dict.get('id', 'entry'))
c77495e3 1363
8b0d7497 1364 def check_filter():
8b0d7497 1365 if 'title' in info_dict:
1366 # This can happen when we're just evaluating the playlist
1367 title = info_dict['title']
1368 matchtitle = self.params.get('matchtitle', False)
1369 if matchtitle:
1370 if not re.search(matchtitle, title, re.IGNORECASE):
1371 return '"' + title + '" title did not match pattern "' + matchtitle + '"'
1372 rejecttitle = self.params.get('rejecttitle', False)
1373 if rejecttitle:
1374 if re.search(rejecttitle, title, re.IGNORECASE):
1375 return '"' + title + '" title matched reject pattern "' + rejecttitle + '"'
1376 date = info_dict.get('upload_date')
1377 if date is not None:
1378 dateRange = self.params.get('daterange', DateRange())
1379 if date not in dateRange:
86e5f3ed 1380 return f'{date_from_str(date).isoformat()} upload date is not in range {dateRange}'
8b0d7497 1381 view_count = info_dict.get('view_count')
1382 if view_count is not None:
1383 min_views = self.params.get('min_views')
1384 if min_views is not None and view_count < min_views:
1385 return 'Skipping %s, because it has not reached minimum view count (%d/%d)' % (video_title, view_count, min_views)
1386 max_views = self.params.get('max_views')
1387 if max_views is not None and view_count > max_views:
1388 return 'Skipping %s, because it has exceeded the maximum view count (%d/%d)' % (video_title, view_count, max_views)
1389 if age_restricted(info_dict.get('age_limit'), self.params.get('age_limit')):
1390 return 'Skipping "%s" because it is age restricted' % video_title
8b0d7497 1391
8f18aca8 1392 match_filter = self.params.get('match_filter')
1393 if match_filter is not None:
1394 try:
1395 ret = match_filter(info_dict, incomplete=incomplete)
1396 except TypeError:
1397 # For backward compatibility
1398 ret = None if incomplete else match_filter(info_dict)
492272fe 1399 if ret is NO_DEFAULT:
1400 while True:
1401 filename = self._format_screen(self.prepare_filename(info_dict), self.Styles.FILENAME)
1402 reply = input(self._format_screen(
1403 f'Download "{filename}"? (Y/n): ', self.Styles.EMPHASIS)).lower().strip()
1404 if reply in {'y', ''}:
1405 return None
1406 elif reply == 'n':
1407 return f'Skipping {video_title}'
492272fe 1408 elif ret is not None:
8f18aca8 1409 return ret
8b0d7497 1410 return None
1411
c77495e3 1412 if self.in_download_archive(info_dict):
1413 reason = '%s has already been recorded in the archive' % video_title
1414 break_opt, break_err = 'break_on_existing', ExistingVideoReached
1415 else:
1416 reason = check_filter()
1417 break_opt, break_err = 'break_on_reject', RejectedVideoReached
8b0d7497 1418 if reason is not None:
120fe513 1419 if not silent:
1420 self.to_screen('[download] ' + reason)
c77495e3 1421 if self.params.get(break_opt, False):
1422 raise break_err()
8b0d7497 1423 return reason
fe7e0c98 1424
b6c45014
JMF
1425 @staticmethod
1426 def add_extra_info(info_dict, extra_info):
1427 '''Set the keys from extra_info in info dict if they are missing'''
1428 for key, value in extra_info.items():
1429 info_dict.setdefault(key, value)
1430
409e1828 1431 def extract_info(self, url, download=True, ie_key=None, extra_info=None,
61aa5ba3 1432 process=True, force_generic_extractor=False):
41d1cca3 1433 """
17ffed18 1434 Extract and return the information dictionary of the URL
41d1cca3 1435
1436 Arguments:
17ffed18 1437 @param url URL to extract
41d1cca3 1438
1439 Keyword arguments:
17ffed18 1440 @param download Whether to download videos
1441 @param process Whether to resolve all unresolved references (URLs, playlist items).
1442 Must be True for download to work
1443 @param ie_key Use only the extractor with this key
1444
1445 @param extra_info Dictionary containing the extra values to add to the info (For internal use only)
1446 @force_generic_extractor Force using the generic extractor (Deprecated; use ie_key='Generic')
41d1cca3 1447 """
fe7e0c98 1448
409e1828 1449 if extra_info is None:
1450 extra_info = {}
1451
61aa5ba3 1452 if not ie_key and force_generic_extractor:
d22dec74
S
1453 ie_key = 'Generic'
1454
8222d8de 1455 if ie_key:
fe7866d0 1456 ies = {ie_key: self._ies[ie_key]} if ie_key in self._ies else {}
8222d8de
JMF
1457 else:
1458 ies = self._ies
1459
fe7866d0 1460 for key, ie in ies.items():
8222d8de
JMF
1461 if not ie.suitable(url):
1462 continue
1463
1464 if not ie.working():
6febd1c1
PH
1465 self.report_warning('The program functionality for this site has been marked as broken, '
1466 'and will probably not work.')
8222d8de 1467
1151c407 1468 temp_id = ie.get_temp_id(url)
fe7866d0 1469 if temp_id is not None and self.in_download_archive({'id': temp_id, 'ie_key': key}):
1470 self.to_screen(f'[{key}] {temp_id}: has already been recorded in the archive')
5e5be0c0 1471 if self.params.get('break_on_existing', False):
1472 raise ExistingVideoReached()
a0566bbf 1473 break
fe7866d0 1474 return self.__extract_info(url, self.get_info_extractor(key), download, extra_info, process)
a0566bbf 1475 else:
fe7866d0 1476 extractors_restricted = self.params.get('allowed_extractors') not in (None, ['default'])
1477 self.report_error(f'No suitable extractor{format_field(ie_key, None, " (%s)")} found for URL {url}',
1478 tb=False if extractors_restricted else None)
a0566bbf 1479
7e88d7d7 1480 def _handle_extraction_exceptions(func):
b5ae35ee 1481 @functools.wraps(func)
a0566bbf 1482 def wrapper(self, *args, **kwargs):
6da22e7d 1483 while True:
1484 try:
1485 return func(self, *args, **kwargs)
1486 except (DownloadCancelled, LazyList.IndexError, PagedList.IndexError):
8222d8de 1487 raise
6da22e7d 1488 except ReExtractInfo as e:
1489 if e.expected:
1490 self.to_screen(f'{e}; Re-extracting data')
1491 else:
1492 self.to_stderr('\r')
1493 self.report_warning(f'{e}; Re-extracting data')
1494 continue
1495 except GeoRestrictedError as e:
1496 msg = e.msg
1497 if e.countries:
1498 msg += '\nThis video is available in %s.' % ', '.join(
1499 map(ISO3166Utils.short2full, e.countries))
1500 msg += '\nYou might want to use a VPN or a proxy server (with --proxy) to workaround.'
1501 self.report_error(msg)
1502 except ExtractorError as e: # An error we somewhat expected
1503 self.report_error(str(e), e.format_traceback())
1504 except Exception as e:
1505 if self.params.get('ignoreerrors'):
1506 self.report_error(str(e), tb=encode_compat_str(traceback.format_exc()))
1507 else:
1508 raise
1509 break
a0566bbf 1510 return wrapper
1511
693f0600 1512 def _wait_for_video(self, ie_result={}):
f2ebc5c7 1513 if (not self.params.get('wait_for_video')
1514 or ie_result.get('_type', 'video') != 'video'
1515 or ie_result.get('formats') or ie_result.get('url')):
1516 return
1517
1518 format_dur = lambda dur: '%02d:%02d:%02d' % timetuple_from_msec(dur * 1000)[:-1]
1519 last_msg = ''
1520
1521 def progress(msg):
1522 nonlocal last_msg
a7dc6a89 1523 full_msg = f'{msg}\n'
1524 if not self.params.get('noprogress'):
1525 full_msg = msg + ' ' * (len(last_msg) - len(msg)) + '\r'
1526 elif last_msg:
1527 return
1528 self.to_screen(full_msg, skip_eol=True)
f2ebc5c7 1529 last_msg = msg
1530
1531 min_wait, max_wait = self.params.get('wait_for_video')
1532 diff = try_get(ie_result, lambda x: x['release_timestamp'] - time.time())
1533 if diff is None and ie_result.get('live_status') == 'is_upcoming':
16c620bc 1534 diff = round(random.uniform(min_wait, max_wait) if (max_wait and min_wait) else (max_wait or min_wait), 0)
f2ebc5c7 1535 self.report_warning('Release time of video is not known')
693f0600 1536 elif ie_result and (diff or 0) <= 0:
f2ebc5c7 1537 self.report_warning('Video should already be available according to extracted info')
38d79fd1 1538 diff = min(max(diff or 0, min_wait or 0), max_wait or float('inf'))
f2ebc5c7 1539 self.to_screen(f'[wait] Waiting for {format_dur(diff)} - Press Ctrl+C to try now')
1540
1541 wait_till = time.time() + diff
1542 try:
1543 while True:
1544 diff = wait_till - time.time()
1545 if diff <= 0:
1546 progress('')
1547 raise ReExtractInfo('[wait] Wait period ended', expected=True)
1548 progress(f'[wait] Remaining time until next attempt: {self._format_screen(format_dur(diff), self.Styles.EMPHASIS)}')
1549 time.sleep(1)
1550 except KeyboardInterrupt:
1551 progress('')
1552 raise ReExtractInfo('[wait] Interrupted by user', expected=True)
1553 except BaseException as e:
1554 if not isinstance(e, ReExtractInfo):
1555 self.to_screen('')
1556 raise
1557
7e88d7d7 1558 @_handle_extraction_exceptions
58f197b7 1559 def __extract_info(self, url, ie, download, extra_info, process):
693f0600 1560 try:
1561 ie_result = ie.extract(url)
1562 except UserNotLive as e:
1563 if process:
1564 if self.params.get('wait_for_video'):
1565 self.report_warning(e)
1566 self._wait_for_video()
1567 raise
a0566bbf 1568 if ie_result is None: # Finished already (backwards compatibility; listformats and friends should be moved here)
cb794ee0 1569 self.report_warning(f'Extractor {ie.IE_NAME} returned nothing{bug_reports_message()}')
a0566bbf 1570 return
1571 if isinstance(ie_result, list):
1572 # Backwards compatibility: old IE result format
1573 ie_result = {
1574 '_type': 'compat_list',
1575 'entries': ie_result,
1576 }
e37d0efb 1577 if extra_info.get('original_url'):
1578 ie_result.setdefault('original_url', extra_info['original_url'])
a0566bbf 1579 self.add_default_extra_info(ie_result, ie, url)
1580 if process:
f2ebc5c7 1581 self._wait_for_video(ie_result)
a0566bbf 1582 return self.process_ie_result(ie_result, download, extra_info)
8222d8de 1583 else:
a0566bbf 1584 return ie_result
fe7e0c98 1585
ea38e55f 1586 def add_default_extra_info(self, ie_result, ie, url):
6033d980 1587 if url is not None:
1588 self.add_extra_info(ie_result, {
1589 'webpage_url': url,
1590 'original_url': url,
57ebfca3 1591 })
1592 webpage_url = ie_result.get('webpage_url')
1593 if webpage_url:
1594 self.add_extra_info(ie_result, {
1595 'webpage_url_basename': url_basename(webpage_url),
1596 'webpage_url_domain': get_domain(webpage_url),
6033d980 1597 })
1598 if ie is not None:
1599 self.add_extra_info(ie_result, {
1600 'extractor': ie.IE_NAME,
1601 'extractor_key': ie.ie_key(),
1602 })
ea38e55f 1603
58adec46 1604 def process_ie_result(self, ie_result, download=True, extra_info=None):
8222d8de
JMF
1605 """
1606 Take the result of the ie(may be modified) and resolve all unresolved
1607 references (URLs, playlist items).
1608
1609 It will also download the videos if 'download'.
1610 Returns the resolved ie_result.
1611 """
58adec46 1612 if extra_info is None:
1613 extra_info = {}
e8ee972c
PH
1614 result_type = ie_result.get('_type', 'video')
1615
057a5206 1616 if result_type in ('url', 'url_transparent'):
8f97a15d 1617 ie_result['url'] = sanitize_url(
1618 ie_result['url'], scheme='http' if self.params.get('prefer_insecure') else 'https')
e37d0efb 1619 if ie_result.get('original_url'):
1620 extra_info.setdefault('original_url', ie_result['original_url'])
1621
057a5206 1622 extract_flat = self.params.get('extract_flat', False)
3089bc74
S
1623 if ((extract_flat == 'in_playlist' and 'playlist' in extra_info)
1624 or extract_flat is True):
ecb54191 1625 info_copy = ie_result.copy()
6033d980 1626 ie = try_get(ie_result.get('ie_key'), self.get_info_extractor)
360167b9 1627 if ie and not ie_result.get('id'):
4614bc22 1628 info_copy['id'] = ie.get_temp_id(ie_result['url'])
6033d980 1629 self.add_default_extra_info(info_copy, ie, ie_result['url'])
4614bc22 1630 self.add_extra_info(info_copy, extra_info)
b5475f11 1631 info_copy, _ = self.pre_process(info_copy)
94dc8604 1632 self._fill_common_fields(info_copy, False)
ecb54191 1633 self.__forced_printings(info_copy, self.prepare_filename(info_copy), incomplete=True)
415f8d51 1634 self._raise_pending_errors(info_copy)
4614bc22 1635 if self.params.get('force_write_download_archive', False):
1636 self.record_download_archive(info_copy)
e8ee972c
PH
1637 return ie_result
1638
8222d8de 1639 if result_type == 'video':
b6c45014 1640 self.add_extra_info(ie_result, extra_info)
9c2b75b5 1641 ie_result = self.process_video_result(ie_result, download=download)
415f8d51 1642 self._raise_pending_errors(ie_result)
28b0eb0f 1643 additional_urls = (ie_result or {}).get('additional_urls')
9c2b75b5 1644 if additional_urls:
e9f4ccd1 1645 # TODO: Improve MetadataParserPP to allow setting a list
14f25df2 1646 if isinstance(additional_urls, str):
9c2b75b5 1647 additional_urls = [additional_urls]
1648 self.to_screen(
1649 '[info] %s: %d additional URL(s) requested' % (ie_result['id'], len(additional_urls)))
1650 self.write_debug('Additional URLs: "%s"' % '", "'.join(additional_urls))
1651 ie_result['additional_entries'] = [
1652 self.extract_info(
b69fd25c 1653 url, download, extra_info=extra_info,
9c2b75b5 1654 force_generic_extractor=self.params.get('force_generic_extractor'))
1655 for url in additional_urls
1656 ]
1657 return ie_result
8222d8de
JMF
1658 elif result_type == 'url':
1659 # We have to add extra_info to the results because it may be
1660 # contained in a playlist
07cce701 1661 return self.extract_info(
1662 ie_result['url'], download,
1663 ie_key=ie_result.get('ie_key'),
1664 extra_info=extra_info)
7fc3fa05
PH
1665 elif result_type == 'url_transparent':
1666 # Use the information from the embedding page
1667 info = self.extract_info(
1668 ie_result['url'], ie_key=ie_result.get('ie_key'),
1669 extra_info=extra_info, download=False, process=False)
1670
1640eb09
S
1671 # extract_info may return None when ignoreerrors is enabled and
1672 # extraction failed with an error, don't crash and return early
1673 # in this case
1674 if not info:
1675 return info
1676
3975b4d2 1677 exempted_fields = {'_type', 'url', 'ie_key'}
1678 if not ie_result.get('section_end') and ie_result.get('section_start') is None:
1679 # For video clips, the id etc of the clip extractor should be used
1680 exempted_fields |= {'id', 'extractor', 'extractor_key'}
1681
412c617d 1682 new_result = info.copy()
3975b4d2 1683 new_result.update(filter_dict(ie_result, lambda k, v: v is not None and k not in exempted_fields))
7fc3fa05 1684
0563f7ac
S
1685 # Extracted info may not be a video result (i.e.
1686 # info.get('_type', 'video') != video) but rather an url or
1687 # url_transparent. In such cases outer metadata (from ie_result)
1688 # should be propagated to inner one (info). For this to happen
1689 # _type of info should be overridden with url_transparent. This
067aa17e 1690 # fixes issue from https://github.com/ytdl-org/youtube-dl/pull/11163.
0563f7ac
S
1691 if new_result.get('_type') == 'url':
1692 new_result['_type'] = 'url_transparent'
7fc3fa05
PH
1693
1694 return self.process_ie_result(
1695 new_result, download=download, extra_info=extra_info)
40fcba5e 1696 elif result_type in ('playlist', 'multi_video'):
30a074c2 1697 # Protect from infinite recursion due to recursively nested playlists
1698 # (see https://github.com/ytdl-org/youtube-dl/issues/27833)
0bd5a039 1699 webpage_url = ie_result.get('webpage_url') # Playlists maynot have webpage_url
1700 if webpage_url and webpage_url in self._playlist_urls:
7e85e872 1701 self.to_screen(
30a074c2 1702 '[download] Skipping already downloaded playlist: %s'
1703 % ie_result.get('title') or ie_result.get('id'))
1704 return
7e85e872 1705
30a074c2 1706 self._playlist_level += 1
1707 self._playlist_urls.add(webpage_url)
03f83004 1708 self._fill_common_fields(ie_result, False)
bc516a3f 1709 self._sanitize_thumbnails(ie_result)
30a074c2 1710 try:
1711 return self.__process_playlist(ie_result, download)
1712 finally:
1713 self._playlist_level -= 1
1714 if not self._playlist_level:
1715 self._playlist_urls.clear()
8222d8de 1716 elif result_type == 'compat_list':
c9bf4114
PH
1717 self.report_warning(
1718 'Extractor %s returned a compat_list result. '
1719 'It needs to be updated.' % ie_result.get('extractor'))
5f6a1245 1720
8222d8de 1721 def _fixup(r):
b868936c 1722 self.add_extra_info(r, {
1723 'extractor': ie_result['extractor'],
1724 'webpage_url': ie_result['webpage_url'],
1725 'webpage_url_basename': url_basename(ie_result['webpage_url']),
0bb322b9 1726 'webpage_url_domain': get_domain(ie_result['webpage_url']),
b868936c 1727 'extractor_key': ie_result['extractor_key'],
1728 })
8222d8de
JMF
1729 return r
1730 ie_result['entries'] = [
b6c45014 1731 self.process_ie_result(_fixup(r), download, extra_info)
8222d8de
JMF
1732 for r in ie_result['entries']
1733 ]
1734 return ie_result
1735 else:
1736 raise Exception('Invalid result type: %s' % result_type)
1737
e92caff5 1738 def _ensure_dir_exists(self, path):
1739 return make_dir(path, self.report_error)
1740
3b603dbd 1741 @staticmethod
3bec830a 1742 def _playlist_infodict(ie_result, strict=False, **kwargs):
1743 info = {
1744 'playlist_count': ie_result.get('playlist_count'),
3b603dbd 1745 'playlist': ie_result.get('title') or ie_result.get('id'),
1746 'playlist_id': ie_result.get('id'),
1747 'playlist_title': ie_result.get('title'),
1748 'playlist_uploader': ie_result.get('uploader'),
1749 'playlist_uploader_id': ie_result.get('uploader_id'),
3b603dbd 1750 **kwargs,
1751 }
3bec830a 1752 if strict:
1753 return info
0bd5a039 1754 if ie_result.get('webpage_url'):
1755 info.update({
1756 'webpage_url': ie_result['webpage_url'],
1757 'webpage_url_basename': url_basename(ie_result['webpage_url']),
1758 'webpage_url_domain': get_domain(ie_result['webpage_url']),
1759 })
3bec830a 1760 return {
1761 **info,
1762 'playlist_index': 0,
1763 '__last_playlist_index': max(ie_result['requested_entries'] or (0, 0)),
1764 'extractor': ie_result['extractor'],
3bec830a 1765 'extractor_key': ie_result['extractor_key'],
1766 }
3b603dbd 1767
30a074c2 1768 def __process_playlist(self, ie_result, download):
7e88d7d7 1769 """Process each entry in the playlist"""
f5ea4748 1770 assert ie_result['_type'] in ('playlist', 'multi_video')
1771
3bec830a 1772 common_info = self._playlist_infodict(ie_result, strict=True)
3955b207 1773 title = common_info.get('playlist') or '<Untitled>'
3bec830a 1774 if self._match_entry(common_info, incomplete=True) is not None:
1775 return
c6e07cf1 1776 self.to_screen(f'[download] Downloading {ie_result["_type"]}: {title}')
f0d785d3 1777
7e88d7d7 1778 all_entries = PlaylistEntries(self, ie_result)
7e9a6125 1779 entries = orderedSet(all_entries.get_requested_items(), lazy=True)
1780
1781 lazy = self.params.get('lazy_playlist')
1782 if lazy:
1783 resolved_entries, n_entries = [], 'N/A'
1784 ie_result['requested_entries'], ie_result['entries'] = None, None
1785 else:
1786 entries = resolved_entries = list(entries)
1787 n_entries = len(resolved_entries)
1788 ie_result['requested_entries'], ie_result['entries'] = tuple(zip(*resolved_entries)) or ([], [])
1789 if not ie_result.get('playlist_count'):
1790 # Better to do this after potentially exhausting entries
1791 ie_result['playlist_count'] = all_entries.get_full_count()
498f5606 1792
0647d925 1793 extra = self._playlist_infodict(ie_result, n_entries=int_or_none(n_entries))
1794 ie_copy = collections.ChainMap(ie_result, extra)
3bec830a 1795
e08a85d8 1796 _infojson_written = False
0bfc53d0 1797 write_playlist_files = self.params.get('allow_playlist_files', True)
1798 if write_playlist_files and self.params.get('list_thumbnails'):
1799 self.list_thumbnails(ie_result)
1800 if write_playlist_files and not self.params.get('simulate'):
e08a85d8 1801 _infojson_written = self._write_info_json(
1802 'playlist', ie_result, self.prepare_filename(ie_copy, 'pl_infojson'))
1803 if _infojson_written is None:
80c03fa9 1804 return
1805 if self._write_description('playlist', ie_result,
1806 self.prepare_filename(ie_copy, 'pl_description')) is None:
1807 return
681de68e 1808 # TODO: This should be passed to ThumbnailsConvertor if necessary
3bec830a 1809 self._write_thumbnails('playlist', ie_result, self.prepare_filename(ie_copy, 'pl_thumbnail'))
30a074c2 1810
7e9a6125 1811 if lazy:
1812 if self.params.get('playlistreverse') or self.params.get('playlistrandom'):
1813 self.report_warning('playlistreverse and playlistrandom are not supported with lazy_playlist', only_once=True)
1814 elif self.params.get('playlistreverse'):
1815 entries.reverse()
1816 elif self.params.get('playlistrandom'):
30a074c2 1817 random.shuffle(entries)
1818
bc5c2f8a 1819 self.to_screen(f'[{ie_result["extractor"]}] Playlist {title}: Downloading {n_entries} items'
7e88d7d7 1820 f'{format_field(ie_result, "playlist_count", " of %s")}')
30a074c2 1821
134c913c 1822 keep_resolved_entries = self.params.get('extract_flat') != 'discard'
1823 if self.params.get('extract_flat') == 'discard_in_playlist':
1824 keep_resolved_entries = ie_result['_type'] != 'playlist'
1825 if keep_resolved_entries:
1826 self.write_debug('The information of all playlist entries will be held in memory')
1827
26e2805c 1828 failures = 0
1829 max_failures = self.params.get('skip_playlist_after_errors') or float('inf')
7e9a6125 1830 for i, (playlist_index, entry) in enumerate(entries):
1831 if lazy:
1832 resolved_entries.append((playlist_index, entry))
3bec830a 1833 if not entry:
7e88d7d7 1834 continue
1835
7e88d7d7 1836 entry['__x_forwarded_for_ip'] = ie_result.get('__x_forwarded_for_ip')
7e9a6125 1837 if not lazy and 'playlist-index' in self.params.get('compat_opts', []):
1838 playlist_index = ie_result['requested_entries'][i]
1839
0647d925 1840 entry_copy = collections.ChainMap(entry, {
3bec830a 1841 **common_info,
3955b207 1842 'n_entries': int_or_none(n_entries),
71729754 1843 'playlist_index': playlist_index,
7e9a6125 1844 'playlist_autonumber': i + 1,
0647d925 1845 })
3bec830a 1846
0647d925 1847 if self._match_entry(entry_copy, incomplete=True) is not None:
f0ad6f8c 1848 # For compatabilty with youtube-dl. See https://github.com/yt-dlp/yt-dlp/issues/4369
1849 resolved_entries[i] = (playlist_index, NO_DEFAULT)
3bec830a 1850 continue
1851
bc5c2f8a 1852 self.to_screen('[download] Downloading item %s of %s' % (
3bec830a 1853 self._format_screen(i + 1, self.Styles.ID), self._format_screen(n_entries, self.Styles.EMPHASIS)))
1854
a6ca61d4 1855 extra.update({
1856 'playlist_index': playlist_index,
1857 'playlist_autonumber': i + 1,
1858 })
3bec830a 1859 entry_result = self.__process_iterable_entry(entry, download, extra)
26e2805c 1860 if not entry_result:
1861 failures += 1
1862 if failures >= max_failures:
1863 self.report_error(
7e88d7d7 1864 f'Skipping the remaining entries in playlist "{title}" since {failures} items failed extraction')
26e2805c 1865 break
134c913c 1866 if keep_resolved_entries:
1867 resolved_entries[i] = (playlist_index, entry_result)
7e88d7d7 1868
1869 # Update with processed data
f0ad6f8c 1870 ie_result['entries'] = [e for _, e in resolved_entries if e is not NO_DEFAULT]
bc5c2f8a 1871 ie_result['requested_entries'] = [i for i, e in resolved_entries if e is not NO_DEFAULT]
1872 if ie_result['requested_entries'] == try_call(lambda: list(range(1, ie_result['playlist_count'] + 1))):
1873 # Do not set for full playlist
1874 ie_result.pop('requested_entries')
e08a85d8 1875
1876 # Write the updated info to json
cb96c5be 1877 if _infojson_written is True and self._write_info_json(
e08a85d8 1878 'updated playlist', ie_result,
1879 self.prepare_filename(ie_copy, 'pl_infojson'), overwrite=True) is None:
1880 return
ca30f449 1881
ed5835b4 1882 ie_result = self.run_all_pps('playlist', ie_result)
7e88d7d7 1883 self.to_screen(f'[download] Finished downloading playlist: {title}')
30a074c2 1884 return ie_result
1885
7e88d7d7 1886 @_handle_extraction_exceptions
a0566bbf 1887 def __process_iterable_entry(self, entry, download, extra_info):
1888 return self.process_ie_result(
1889 entry, download=download, extra_info=extra_info)
1890
67134eab
JMF
1891 def _build_format_filter(self, filter_spec):
1892 " Returns a function to filter the formats according to the filter_spec "
083c9df9
PH
1893
1894 OPERATORS = {
1895 '<': operator.lt,
1896 '<=': operator.le,
1897 '>': operator.gt,
1898 '>=': operator.ge,
1899 '=': operator.eq,
1900 '!=': operator.ne,
1901 }
67134eab 1902 operator_rex = re.compile(r'''(?x)\s*
187986a8 1903 (?P<key>width|height|tbr|abr|vbr|asr|filesize|filesize_approx|fps)\s*
1904 (?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
1905 (?P<value>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?)\s*
083c9df9 1906 ''' % '|'.join(map(re.escape, OPERATORS.keys())))
187986a8 1907 m = operator_rex.fullmatch(filter_spec)
9ddb6925
S
1908 if m:
1909 try:
1910 comparison_value = int(m.group('value'))
1911 except ValueError:
1912 comparison_value = parse_filesize(m.group('value'))
1913 if comparison_value is None:
1914 comparison_value = parse_filesize(m.group('value') + 'B')
1915 if comparison_value is None:
1916 raise ValueError(
1917 'Invalid value %r in format specification %r' % (
67134eab 1918 m.group('value'), filter_spec))
9ddb6925
S
1919 op = OPERATORS[m.group('op')]
1920
083c9df9 1921 if not m:
9ddb6925
S
1922 STR_OPERATORS = {
1923 '=': operator.eq,
10d33b34
YCH
1924 '^=': lambda attr, value: attr.startswith(value),
1925 '$=': lambda attr, value: attr.endswith(value),
1926 '*=': lambda attr, value: value in attr,
1ce9a3cb 1927 '~=': lambda attr, value: value.search(attr) is not None
9ddb6925 1928 }
187986a8 1929 str_operator_rex = re.compile(r'''(?x)\s*
1930 (?P<key>[a-zA-Z0-9._-]+)\s*
1ce9a3cb
LF
1931 (?P<negation>!\s*)?(?P<op>%s)\s*(?P<none_inclusive>\?\s*)?
1932 (?P<quote>["'])?
1933 (?P<value>(?(quote)(?:(?!(?P=quote))[^\\]|\\.)+|[\w.-]+))
1934 (?(quote)(?P=quote))\s*
9ddb6925 1935 ''' % '|'.join(map(re.escape, STR_OPERATORS.keys())))
187986a8 1936 m = str_operator_rex.fullmatch(filter_spec)
9ddb6925 1937 if m:
1ce9a3cb
LF
1938 if m.group('op') == '~=':
1939 comparison_value = re.compile(m.group('value'))
1940 else:
1941 comparison_value = re.sub(r'''\\([\\"'])''', r'\1', m.group('value'))
2cc779f4
S
1942 str_op = STR_OPERATORS[m.group('op')]
1943 if m.group('negation'):
e118a879 1944 op = lambda attr, value: not str_op(attr, value)
2cc779f4
S
1945 else:
1946 op = str_op
083c9df9 1947
9ddb6925 1948 if not m:
187986a8 1949 raise SyntaxError('Invalid filter specification %r' % filter_spec)
083c9df9
PH
1950
1951 def _filter(f):
1952 actual_value = f.get(m.group('key'))
1953 if actual_value is None:
1954 return m.group('none_inclusive')
1955 return op(actual_value, comparison_value)
67134eab
JMF
1956 return _filter
1957
9f1a1c36 1958 def _check_formats(self, formats):
1959 for f in formats:
1960 self.to_screen('[info] Testing format %s' % f['format_id'])
75689fe5 1961 path = self.get_output_path('temp')
1962 if not self._ensure_dir_exists(f'{path}/'):
1963 continue
1964 temp_file = tempfile.NamedTemporaryFile(suffix='.tmp', delete=False, dir=path or None)
9f1a1c36 1965 temp_file.close()
1966 try:
1967 success, _ = self.dl(temp_file.name, f, test=True)
8a82af35 1968 except (DownloadError, OSError, ValueError) + network_exceptions:
9f1a1c36 1969 success = False
1970 finally:
1971 if os.path.exists(temp_file.name):
1972 try:
1973 os.remove(temp_file.name)
1974 except OSError:
1975 self.report_warning('Unable to delete temporary file "%s"' % temp_file.name)
1976 if success:
1977 yield f
1978 else:
1979 self.to_screen('[info] Unable to download format %s. Skipping...' % f['format_id'])
1980
0017d9ad 1981 def _default_format_spec(self, info_dict, download=True):
0017d9ad 1982
af0f7428
S
1983 def can_merge():
1984 merger = FFmpegMergerPP(self)
1985 return merger.available and merger.can_merge()
1986
91ebc640 1987 prefer_best = (
b7b04c78 1988 not self.params.get('simulate')
91ebc640 1989 and download
1990 and (
1991 not can_merge()
21633673 1992 or info_dict.get('is_live') and not self.params.get('live_from_start')
bf1824b3 1993 or self.params['outtmpl']['default'] == '-'))
53ed7066 1994 compat = (
1995 prefer_best
1996 or self.params.get('allow_multiple_audio_streams', False)
8a82af35 1997 or 'format-spec' in self.params['compat_opts'])
91ebc640 1998
1999 return (
53ed7066 2000 'best/bestvideo+bestaudio' if prefer_best
2001 else 'bestvideo*+bestaudio/best' if not compat
91ebc640 2002 else 'bestvideo+bestaudio/best')
0017d9ad 2003
67134eab
JMF
2004 def build_format_selector(self, format_spec):
2005 def syntax_error(note, start):
2006 message = (
2007 'Invalid format specification: '
86e5f3ed 2008 '{}\n\t{}\n\t{}^'.format(note, format_spec, ' ' * start[1]))
67134eab
JMF
2009 return SyntaxError(message)
2010
2011 PICKFIRST = 'PICKFIRST'
2012 MERGE = 'MERGE'
2013 SINGLE = 'SINGLE'
0130afb7 2014 GROUP = 'GROUP'
67134eab
JMF
2015 FormatSelector = collections.namedtuple('FormatSelector', ['type', 'selector', 'filters'])
2016
91ebc640 2017 allow_multiple_streams = {'audio': self.params.get('allow_multiple_audio_streams', False),
2018 'video': self.params.get('allow_multiple_video_streams', False)}
909d24dd 2019
9f1a1c36 2020 check_formats = self.params.get('check_formats') == 'selected'
e8e73840 2021
67134eab
JMF
2022 def _parse_filter(tokens):
2023 filter_parts = []
2024 for type, string, start, _, _ in tokens:
2025 if type == tokenize.OP and string == ']':
2026 return ''.join(filter_parts)
2027 else:
2028 filter_parts.append(string)
2029
232541df 2030 def _remove_unused_ops(tokens):
62b58c09
L
2031 # Remove operators that we don't use and join them with the surrounding strings.
2032 # E.g. 'mp4' '-' 'baseline' '-' '16x9' is converted to 'mp4-baseline-16x9'
232541df
JMF
2033 ALLOWED_OPS = ('/', '+', ',', '(', ')')
2034 last_string, last_start, last_end, last_line = None, None, None, None
2035 for type, string, start, end, line in tokens:
2036 if type == tokenize.OP and string == '[':
2037 if last_string:
2038 yield tokenize.NAME, last_string, last_start, last_end, last_line
2039 last_string = None
2040 yield type, string, start, end, line
2041 # everything inside brackets will be handled by _parse_filter
2042 for type, string, start, end, line in tokens:
2043 yield type, string, start, end, line
2044 if type == tokenize.OP and string == ']':
2045 break
2046 elif type == tokenize.OP and string in ALLOWED_OPS:
2047 if last_string:
2048 yield tokenize.NAME, last_string, last_start, last_end, last_line
2049 last_string = None
2050 yield type, string, start, end, line
2051 elif type in [tokenize.NAME, tokenize.NUMBER, tokenize.OP]:
2052 if not last_string:
2053 last_string = string
2054 last_start = start
2055 last_end = end
2056 else:
2057 last_string += string
2058 if last_string:
2059 yield tokenize.NAME, last_string, last_start, last_end, last_line
2060
cf2ac6df 2061 def _parse_format_selection(tokens, inside_merge=False, inside_choice=False, inside_group=False):
67134eab
JMF
2062 selectors = []
2063 current_selector = None
2064 for type, string, start, _, _ in tokens:
2065 # ENCODING is only defined in python 3.x
2066 if type == getattr(tokenize, 'ENCODING', None):
2067 continue
2068 elif type in [tokenize.NAME, tokenize.NUMBER]:
2069 current_selector = FormatSelector(SINGLE, string, [])
2070 elif type == tokenize.OP:
cf2ac6df
JMF
2071 if string == ')':
2072 if not inside_group:
2073 # ')' will be handled by the parentheses group
2074 tokens.restore_last_token()
67134eab 2075 break
cf2ac6df 2076 elif inside_merge and string in ['/', ',']:
0130afb7
JMF
2077 tokens.restore_last_token()
2078 break
cf2ac6df
JMF
2079 elif inside_choice and string == ',':
2080 tokens.restore_last_token()
2081 break
2082 elif string == ',':
0a31a350
JMF
2083 if not current_selector:
2084 raise syntax_error('"," must follow a format selector', start)
67134eab
JMF
2085 selectors.append(current_selector)
2086 current_selector = None
2087 elif string == '/':
d96d604e
JMF
2088 if not current_selector:
2089 raise syntax_error('"/" must follow a format selector', start)
67134eab 2090 first_choice = current_selector
cf2ac6df 2091 second_choice = _parse_format_selection(tokens, inside_choice=True)
f5f4a27a 2092 current_selector = FormatSelector(PICKFIRST, (first_choice, second_choice), [])
67134eab
JMF
2093 elif string == '[':
2094 if not current_selector:
2095 current_selector = FormatSelector(SINGLE, 'best', [])
2096 format_filter = _parse_filter(tokens)
2097 current_selector.filters.append(format_filter)
0130afb7
JMF
2098 elif string == '(':
2099 if current_selector:
2100 raise syntax_error('Unexpected "("', start)
cf2ac6df
JMF
2101 group = _parse_format_selection(tokens, inside_group=True)
2102 current_selector = FormatSelector(GROUP, group, [])
67134eab 2103 elif string == '+':
d03cfdce 2104 if not current_selector:
2105 raise syntax_error('Unexpected "+"', start)
2106 selector_1 = current_selector
2107 selector_2 = _parse_format_selection(tokens, inside_merge=True)
2108 if not selector_2:
2109 raise syntax_error('Expected a selector', start)
2110 current_selector = FormatSelector(MERGE, (selector_1, selector_2), [])
67134eab 2111 else:
86e5f3ed 2112 raise syntax_error(f'Operator not recognized: "{string}"', start)
67134eab
JMF
2113 elif type == tokenize.ENDMARKER:
2114 break
2115 if current_selector:
2116 selectors.append(current_selector)
2117 return selectors
2118
f8d4ad9a 2119 def _merge(formats_pair):
2120 format_1, format_2 = formats_pair
2121
2122 formats_info = []
2123 formats_info.extend(format_1.get('requested_formats', (format_1,)))
2124 formats_info.extend(format_2.get('requested_formats', (format_2,)))
2125
2126 if not allow_multiple_streams['video'] or not allow_multiple_streams['audio']:
551f9388 2127 get_no_more = {'video': False, 'audio': False}
f8d4ad9a 2128 for (i, fmt_info) in enumerate(formats_info):
551f9388 2129 if fmt_info.get('acodec') == fmt_info.get('vcodec') == 'none':
2130 formats_info.pop(i)
2131 continue
2132 for aud_vid in ['audio', 'video']:
f8d4ad9a 2133 if not allow_multiple_streams[aud_vid] and fmt_info.get(aud_vid[0] + 'codec') != 'none':
2134 if get_no_more[aud_vid]:
2135 formats_info.pop(i)
f5510afe 2136 break
f8d4ad9a 2137 get_no_more[aud_vid] = True
2138
2139 if len(formats_info) == 1:
2140 return formats_info[0]
2141
2142 video_fmts = [fmt_info for fmt_info in formats_info if fmt_info.get('vcodec') != 'none']
2143 audio_fmts = [fmt_info for fmt_info in formats_info if fmt_info.get('acodec') != 'none']
2144
2145 the_only_video = video_fmts[0] if len(video_fmts) == 1 else None
2146 the_only_audio = audio_fmts[0] if len(audio_fmts) == 1 else None
2147
fc61aff4
LL
2148 output_ext = get_compatible_ext(
2149 vcodecs=[f.get('vcodec') for f in video_fmts],
2150 acodecs=[f.get('acodec') for f in audio_fmts],
2151 vexts=[f['ext'] for f in video_fmts],
2152 aexts=[f['ext'] for f in audio_fmts],
2153 preferences=(try_call(lambda: self.params['merge_output_format'].split('/'))
2154 or self.params.get('prefer_free_formats') and ('webm', 'mkv')))
f8d4ad9a 2155
975a0d0d 2156 filtered = lambda *keys: filter(None, (traverse_obj(fmt, *keys) for fmt in formats_info))
2157
f8d4ad9a 2158 new_dict = {
2159 'requested_formats': formats_info,
975a0d0d 2160 'format': '+'.join(filtered('format')),
2161 'format_id': '+'.join(filtered('format_id')),
f8d4ad9a 2162 'ext': output_ext,
975a0d0d 2163 'protocol': '+'.join(map(determine_protocol, formats_info)),
093a1710 2164 'language': '+'.join(orderedSet(filtered('language'))) or None,
2165 'format_note': '+'.join(orderedSet(filtered('format_note'))) or None,
2166 'filesize_approx': sum(filtered('filesize', 'filesize_approx')) or None,
975a0d0d 2167 'tbr': sum(filtered('tbr', 'vbr', 'abr')),
f8d4ad9a 2168 }
2169
2170 if the_only_video:
2171 new_dict.update({
2172 'width': the_only_video.get('width'),
2173 'height': the_only_video.get('height'),
2174 'resolution': the_only_video.get('resolution') or self.format_resolution(the_only_video),
2175 'fps': the_only_video.get('fps'),
49a57e70 2176 'dynamic_range': the_only_video.get('dynamic_range'),
f8d4ad9a 2177 'vcodec': the_only_video.get('vcodec'),
2178 'vbr': the_only_video.get('vbr'),
2179 'stretched_ratio': the_only_video.get('stretched_ratio'),
2180 })
2181
2182 if the_only_audio:
2183 new_dict.update({
2184 'acodec': the_only_audio.get('acodec'),
2185 'abr': the_only_audio.get('abr'),
975a0d0d 2186 'asr': the_only_audio.get('asr'),
b8ed0f15 2187 'audio_channels': the_only_audio.get('audio_channels')
f8d4ad9a 2188 })
2189
2190 return new_dict
2191
e8e73840 2192 def _check_formats(formats):
981052c9 2193 if not check_formats:
2194 yield from formats
b5ac45b1 2195 return
9f1a1c36 2196 yield from self._check_formats(formats)
e8e73840 2197
67134eab 2198 def _build_selector_function(selector):
909d24dd 2199 if isinstance(selector, list): # ,
67134eab
JMF
2200 fs = [_build_selector_function(s) for s in selector]
2201
317f7ab6 2202 def selector_function(ctx):
67134eab 2203 for f in fs:
981052c9 2204 yield from f(ctx)
67134eab 2205 return selector_function
909d24dd 2206
2207 elif selector.type == GROUP: # ()
0130afb7 2208 selector_function = _build_selector_function(selector.selector)
909d24dd 2209
2210 elif selector.type == PICKFIRST: # /
67134eab
JMF
2211 fs = [_build_selector_function(s) for s in selector.selector]
2212
317f7ab6 2213 def selector_function(ctx):
67134eab 2214 for f in fs:
317f7ab6 2215 picked_formats = list(f(ctx))
67134eab
JMF
2216 if picked_formats:
2217 return picked_formats
2218 return []
67134eab 2219
981052c9 2220 elif selector.type == MERGE: # +
2221 selector_1, selector_2 = map(_build_selector_function, selector.selector)
2222
2223 def selector_function(ctx):
adbc4ec4 2224 for pair in itertools.product(selector_1(ctx), selector_2(ctx)):
981052c9 2225 yield _merge(pair)
2226
909d24dd 2227 elif selector.type == SINGLE: # atom
598d185d 2228 format_spec = selector.selector or 'best'
909d24dd 2229
f8d4ad9a 2230 # TODO: Add allvideo, allaudio etc by generalizing the code with best/worst selector
909d24dd 2231 if format_spec == 'all':
2232 def selector_function(ctx):
9222c381 2233 yield from _check_formats(ctx['formats'][::-1])
f8d4ad9a 2234 elif format_spec == 'mergeall':
2235 def selector_function(ctx):
316f2650 2236 formats = list(_check_formats(
2237 f for f in ctx['formats'] if f.get('vcodec') != 'none' or f.get('acodec') != 'none'))
e01d6aa4 2238 if not formats:
2239 return
921b76ca 2240 merged_format = formats[-1]
2241 for f in formats[-2::-1]:
f8d4ad9a 2242 merged_format = _merge((merged_format, f))
2243 yield merged_format
909d24dd 2244
2245 else:
85e801a9 2246 format_fallback, seperate_fallback, format_reverse, format_idx = False, None, True, 1
eff63539 2247 mobj = re.match(
2248 r'(?P<bw>best|worst|b|w)(?P<type>video|audio|v|a)?(?P<mod>\*)?(?:\.(?P<n>[1-9]\d*))?$',
2249 format_spec)
2250 if mobj is not None:
2251 format_idx = int_or_none(mobj.group('n'), default=1)
e8e73840 2252 format_reverse = mobj.group('bw')[0] == 'b'
eff63539 2253 format_type = (mobj.group('type') or [None])[0]
2254 not_format_type = {'v': 'a', 'a': 'v'}.get(format_type)
2255 format_modified = mobj.group('mod') is not None
909d24dd 2256
2257 format_fallback = not format_type and not format_modified # for b, w
8326b00a 2258 _filter_f = (
eff63539 2259 (lambda f: f.get('%scodec' % format_type) != 'none')
2260 if format_type and format_modified # bv*, ba*, wv*, wa*
2261 else (lambda f: f.get('%scodec' % not_format_type) == 'none')
2262 if format_type # bv, ba, wv, wa
2263 else (lambda f: f.get('vcodec') != 'none' and f.get('acodec') != 'none')
2264 if not format_modified # b, w
8326b00a 2265 else lambda f: True) # b*, w*
2266 filter_f = lambda f: _filter_f(f) and (
2267 f.get('vcodec') != 'none' or f.get('acodec') != 'none')
67134eab 2268 else:
48ee10ee 2269 if format_spec in self._format_selection_exts['audio']:
b11c04a8 2270 filter_f = lambda f: f.get('ext') == format_spec and f.get('acodec') != 'none'
48ee10ee 2271 elif format_spec in self._format_selection_exts['video']:
b11c04a8 2272 filter_f = lambda f: f.get('ext') == format_spec and f.get('acodec') != 'none' and f.get('vcodec') != 'none'
85e801a9 2273 seperate_fallback = lambda f: f.get('ext') == format_spec and f.get('vcodec') != 'none'
48ee10ee 2274 elif format_spec in self._format_selection_exts['storyboards']:
b11c04a8 2275 filter_f = lambda f: f.get('ext') == format_spec and f.get('acodec') == 'none' and f.get('vcodec') == 'none'
2276 else:
b5ae35ee 2277 filter_f = lambda f: f.get('format_id') == format_spec # id
909d24dd 2278
2279 def selector_function(ctx):
2280 formats = list(ctx['formats'])
909d24dd 2281 matches = list(filter(filter_f, formats)) if filter_f is not None else formats
85e801a9 2282 if not matches:
2283 if format_fallback and ctx['incomplete_formats']:
2284 # for extractors with incomplete formats (audio only (soundcloud)
2285 # or video only (imgur)) best/worst will fallback to
2286 # best/worst {video,audio}-only format
2287 matches = formats
2288 elif seperate_fallback and not ctx['has_merged_format']:
2289 # for compatibility with youtube-dl when there is no pre-merged format
2290 matches = list(filter(seperate_fallback, formats))
981052c9 2291 matches = LazyList(_check_formats(matches[::-1 if format_reverse else 1]))
2292 try:
e8e73840 2293 yield matches[format_idx - 1]
4abea8ca 2294 except LazyList.IndexError:
981052c9 2295 return
083c9df9 2296
67134eab 2297 filters = [self._build_format_filter(f) for f in selector.filters]
083c9df9 2298
317f7ab6 2299 def final_selector(ctx):
adbc4ec4 2300 ctx_copy = dict(ctx)
67134eab 2301 for _filter in filters:
317f7ab6
S
2302 ctx_copy['formats'] = list(filter(_filter, ctx_copy['formats']))
2303 return selector_function(ctx_copy)
67134eab 2304 return final_selector
083c9df9 2305
0f06bcd7 2306 stream = io.BytesIO(format_spec.encode())
0130afb7 2307 try:
f9934b96 2308 tokens = list(_remove_unused_ops(tokenize.tokenize(stream.readline)))
0130afb7
JMF
2309 except tokenize.TokenError:
2310 raise syntax_error('Missing closing/opening brackets or parenthesis', (0, len(format_spec)))
2311
86e5f3ed 2312 class TokenIterator:
0130afb7
JMF
2313 def __init__(self, tokens):
2314 self.tokens = tokens
2315 self.counter = 0
2316
2317 def __iter__(self):
2318 return self
2319
2320 def __next__(self):
2321 if self.counter >= len(self.tokens):
2322 raise StopIteration()
2323 value = self.tokens[self.counter]
2324 self.counter += 1
2325 return value
2326
2327 next = __next__
2328
2329 def restore_last_token(self):
2330 self.counter -= 1
2331
2332 parsed_selector = _parse_format_selection(iter(TokenIterator(tokens)))
67134eab 2333 return _build_selector_function(parsed_selector)
a9c58ad9 2334
e5660ee6 2335 def _calc_headers(self, info_dict):
8b7539d2 2336 res = merge_headers(self.params['http_headers'], info_dict.get('http_headers') or {})
e5660ee6 2337
c487cf00 2338 cookies = self._calc_cookies(info_dict['url'])
e5660ee6
JMF
2339 if cookies:
2340 res['Cookie'] = cookies
2341
0016b84e
S
2342 if 'X-Forwarded-For' not in res:
2343 x_forwarded_for_ip = info_dict.get('__x_forwarded_for_ip')
2344 if x_forwarded_for_ip:
2345 res['X-Forwarded-For'] = x_forwarded_for_ip
2346
e5660ee6
JMF
2347 return res
2348
c487cf00 2349 def _calc_cookies(self, url):
2350 pr = sanitized_Request(url)
e5660ee6 2351 self.cookiejar.add_cookie_header(pr)
662435f7 2352 return pr.get_header('Cookie')
e5660ee6 2353
9f1a1c36 2354 def _sort_thumbnails(self, thumbnails):
2355 thumbnails.sort(key=lambda t: (
2356 t.get('preference') if t.get('preference') is not None else -1,
2357 t.get('width') if t.get('width') is not None else -1,
2358 t.get('height') if t.get('height') is not None else -1,
2359 t.get('id') if t.get('id') is not None else '',
2360 t.get('url')))
2361
b0249bca 2362 def _sanitize_thumbnails(self, info_dict):
bc516a3f 2363 thumbnails = info_dict.get('thumbnails')
2364 if thumbnails is None:
2365 thumbnail = info_dict.get('thumbnail')
2366 if thumbnail:
2367 info_dict['thumbnails'] = thumbnails = [{'url': thumbnail}]
9f1a1c36 2368 if not thumbnails:
2369 return
2370
2371 def check_thumbnails(thumbnails):
2372 for t in thumbnails:
2373 self.to_screen(f'[info] Testing thumbnail {t["id"]}')
2374 try:
2375 self.urlopen(HEADRequest(t['url']))
2376 except network_exceptions as err:
2377 self.to_screen(f'[info] Unable to connect to thumbnail {t["id"]} URL {t["url"]!r} - {err}. Skipping...')
2378 continue
2379 yield t
2380
2381 self._sort_thumbnails(thumbnails)
2382 for i, t in enumerate(thumbnails):
2383 if t.get('id') is None:
2384 t['id'] = '%d' % i
2385 if t.get('width') and t.get('height'):
2386 t['resolution'] = '%dx%d' % (t['width'], t['height'])
2387 t['url'] = sanitize_url(t['url'])
2388
2389 if self.params.get('check_formats') is True:
282f5709 2390 info_dict['thumbnails'] = LazyList(check_thumbnails(thumbnails[::-1]), reverse=True)
9f1a1c36 2391 else:
2392 info_dict['thumbnails'] = thumbnails
bc516a3f 2393
94dc8604 2394 def _fill_common_fields(self, info_dict, final=True):
03f83004 2395 # TODO: move sanitization here
94dc8604 2396 if final:
d4736fdb 2397 title = info_dict.get('title', NO_DEFAULT)
2398 if title is NO_DEFAULT:
03f83004
LNO
2399 raise ExtractorError('Missing "title" field in extractor result',
2400 video_id=info_dict['id'], ie=info_dict['extractor'])
d4736fdb 2401 info_dict['fulltitle'] = title
2402 if not title:
2403 if title == '':
2404 self.write_debug('Extractor gave empty title. Creating a generic title')
2405 else:
2406 self.report_warning('Extractor failed to obtain "title". Creating a generic title instead')
1d485a1a 2407 info_dict['title'] = f'{info_dict["extractor"].replace(":", "-")} video #{info_dict["id"]}'
03f83004
LNO
2408
2409 if info_dict.get('duration') is not None:
2410 info_dict['duration_string'] = formatSeconds(info_dict['duration'])
2411
2412 for ts_key, date_key in (
2413 ('timestamp', 'upload_date'),
2414 ('release_timestamp', 'release_date'),
2415 ('modified_timestamp', 'modified_date'),
2416 ):
2417 if info_dict.get(date_key) is None and info_dict.get(ts_key) is not None:
2418 # Working around out-of-range timestamp values (e.g. negative ones on Windows,
2419 # see http://bugs.python.org/issue1646728)
19a03940 2420 with contextlib.suppress(ValueError, OverflowError, OSError):
03f83004
LNO
2421 upload_date = datetime.datetime.utcfromtimestamp(info_dict[ts_key])
2422 info_dict[date_key] = upload_date.strftime('%Y%m%d')
03f83004
LNO
2423
2424 live_keys = ('is_live', 'was_live')
2425 live_status = info_dict.get('live_status')
2426 if live_status is None:
2427 for key in live_keys:
2428 if info_dict.get(key) is False:
2429 continue
2430 if info_dict.get(key):
2431 live_status = key
2432 break
2433 if all(info_dict.get(key) is False for key in live_keys):
2434 live_status = 'not_live'
2435 if live_status:
2436 info_dict['live_status'] = live_status
2437 for key in live_keys:
2438 if info_dict.get(key) is None:
2439 info_dict[key] = (live_status == key)
a057779d 2440 if live_status == 'post_live':
2441 info_dict['was_live'] = True
03f83004
LNO
2442
2443 # Auto generate title fields corresponding to the *_number fields when missing
2444 # in order to always have clean titles. This is very common for TV series.
2445 for field in ('chapter', 'season', 'episode'):
94dc8604 2446 if final and info_dict.get('%s_number' % field) is not None and not info_dict.get(field):
03f83004
LNO
2447 info_dict[field] = '%s %d' % (field.capitalize(), info_dict['%s_number' % field])
2448
415f8d51 2449 def _raise_pending_errors(self, info):
2450 err = info.pop('__pending_error', None)
2451 if err:
2452 self.report_error(err, tb=False)
2453
dd82ffea
JMF
2454 def process_video_result(self, info_dict, download=True):
2455 assert info_dict.get('_type', 'video') == 'video'
9c906919 2456 self._num_videos += 1
dd82ffea 2457
bec1fad2 2458 if 'id' not in info_dict:
fc08bdd6 2459 raise ExtractorError('Missing "id" field in extractor result', ie=info_dict['extractor'])
2460 elif not info_dict.get('id'):
2461 raise ExtractorError('Extractor failed to obtain "id"', ie=info_dict['extractor'])
455a15e2 2462
c9969434
S
2463 def report_force_conversion(field, field_not, conversion):
2464 self.report_warning(
2465 '"%s" field is not %s - forcing %s conversion, there is an error in extractor'
2466 % (field, field_not, conversion))
2467
2468 def sanitize_string_field(info, string_field):
2469 field = info.get(string_field)
14f25df2 2470 if field is None or isinstance(field, str):
c9969434
S
2471 return
2472 report_force_conversion(string_field, 'a string', 'string')
14f25df2 2473 info[string_field] = str(field)
c9969434
S
2474
2475 def sanitize_numeric_fields(info):
2476 for numeric_field in self._NUMERIC_FIELDS:
2477 field = info.get(numeric_field)
f9934b96 2478 if field is None or isinstance(field, (int, float)):
c9969434
S
2479 continue
2480 report_force_conversion(numeric_field, 'numeric', 'int')
2481 info[numeric_field] = int_or_none(field)
2482
2483 sanitize_string_field(info_dict, 'id')
2484 sanitize_numeric_fields(info_dict)
3975b4d2 2485 if info_dict.get('section_end') and info_dict.get('section_start') is not None:
2486 info_dict['duration'] = round(info_dict['section_end'] - info_dict['section_start'], 3)
4c3f8c3f 2487 if (info_dict.get('duration') or 0) <= 0 and info_dict.pop('duration', None):
50e93e03 2488 self.report_warning('"duration" field is negative, there is an error in extractor')
be6217b2 2489
9eef7c4e 2490 chapters = info_dict.get('chapters') or []
a3976e07 2491 if chapters and chapters[0].get('start_time'):
2492 chapters.insert(0, {'start_time': 0})
2493
9eef7c4e 2494 dummy_chapter = {'end_time': 0, 'start_time': info_dict.get('duration')}
a3976e07 2495 for idx, (prev, current, next_) in enumerate(zip(
2496 (dummy_chapter, *chapters), chapters, (*chapters[1:], dummy_chapter)), 1):
9eef7c4e 2497 if current.get('start_time') is None:
2498 current['start_time'] = prev.get('end_time')
2499 if not current.get('end_time'):
2500 current['end_time'] = next_.get('start_time')
a3976e07 2501 if not current.get('title'):
2502 current['title'] = f'<Untitled Chapter {idx}>'
9eef7c4e 2503
dd82ffea
JMF
2504 if 'playlist' not in info_dict:
2505 # It isn't part of a playlist
2506 info_dict['playlist'] = None
2507 info_dict['playlist_index'] = None
2508
bc516a3f 2509 self._sanitize_thumbnails(info_dict)
d5519808 2510
536a55da 2511 thumbnail = info_dict.get('thumbnail')
bc516a3f 2512 thumbnails = info_dict.get('thumbnails')
536a55da
S
2513 if thumbnail:
2514 info_dict['thumbnail'] = sanitize_url(thumbnail)
2515 elif thumbnails:
d5519808
PH
2516 info_dict['thumbnail'] = thumbnails[-1]['url']
2517
ae30b840 2518 if info_dict.get('display_id') is None and 'id' in info_dict:
0afef30b
PH
2519 info_dict['display_id'] = info_dict['id']
2520
03f83004 2521 self._fill_common_fields(info_dict)
33d2fc2f 2522
05108a49
S
2523 for cc_kind in ('subtitles', 'automatic_captions'):
2524 cc = info_dict.get(cc_kind)
2525 if cc:
2526 for _, subtitle in cc.items():
2527 for subtitle_format in subtitle:
2528 if subtitle_format.get('url'):
2529 subtitle_format['url'] = sanitize_url(subtitle_format['url'])
2530 if subtitle_format.get('ext') is None:
2531 subtitle_format['ext'] = determine_ext(subtitle_format['url']).lower()
2532
2533 automatic_captions = info_dict.get('automatic_captions')
4bba3716 2534 subtitles = info_dict.get('subtitles')
4bba3716 2535
360e1ca5 2536 info_dict['requested_subtitles'] = self.process_subtitles(
05108a49 2537 info_dict['id'], subtitles, automatic_captions)
a504ced0 2538
aebb4f4b 2539 formats = self._get_formats(info_dict)
dd82ffea 2540
0a5a191a 2541 # or None ensures --clean-infojson removes it
2542 info_dict['_has_drm'] = any(f.get('has_drm') for f in formats) or None
88acdbc2 2543 if not self.params.get('allow_unplayable_formats'):
2544 formats = [f for f in formats if not f.get('has_drm')]
17ffed18 2545
2546 if formats and all(f.get('acodec') == f.get('vcodec') == 'none' for f in formats):
2547 self.report_warning(
2548 f'{"This video is DRM protected and " if info_dict["_has_drm"] else ""}'
2549 'only images are available for download. Use --list-formats to see them'.capitalize())
88acdbc2 2550
319b6059 2551 get_from_start = not info_dict.get('is_live') or bool(self.params.get('live_from_start'))
2552 if not get_from_start:
2553 info_dict['title'] += ' ' + datetime.datetime.now().strftime('%Y-%m-%d %H:%M')
2554 if info_dict.get('is_live') and formats:
adbc4ec4 2555 formats = [f for f in formats if bool(f.get('is_from_start')) == get_from_start]
319b6059 2556 if get_from_start and not formats:
a44ca5a4 2557 self.raise_no_formats(info_dict, msg=(
2558 '--live-from-start is passed, but there are no formats that can be downloaded from the start. '
2559 'If you want to download from the current time, use --no-live-from-start'))
adbc4ec4 2560
73af5cc8
S
2561 def is_wellformed(f):
2562 url = f.get('url')
a5ac0c47 2563 if not url:
73af5cc8
S
2564 self.report_warning(
2565 '"url" field is missing or empty - skipping format, '
2566 'there is an error in extractor')
a5ac0c47
S
2567 return False
2568 if isinstance(url, bytes):
2569 sanitize_string_field(f, 'url')
2570 return True
73af5cc8
S
2571
2572 # Filter out malformed formats for better extraction robustness
1ac7f461 2573 formats = list(filter(is_wellformed, formats or []))
2574
2575 if not formats:
2576 self.raise_no_formats(info_dict)
73af5cc8 2577
181c7053
S
2578 formats_dict = {}
2579
dd82ffea 2580 # We check that all the formats have the format and format_id fields
db95dc13 2581 for i, format in enumerate(formats):
c9969434
S
2582 sanitize_string_field(format, 'format_id')
2583 sanitize_numeric_fields(format)
dcf77cf1 2584 format['url'] = sanitize_url(format['url'])
e74e3b63 2585 if not format.get('format_id'):
14f25df2 2586 format['format_id'] = str(i)
e2effb08
S
2587 else:
2588 # Sanitize format_id from characters used in format selector expression
ec85ded8 2589 format['format_id'] = re.sub(r'[\s,/+\[\]()]', '_', format['format_id'])
181c7053
S
2590 format_id = format['format_id']
2591 if format_id not in formats_dict:
2592 formats_dict[format_id] = []
2593 formats_dict[format_id].append(format)
2594
2595 # Make sure all formats have unique format_id
03b4de72 2596 common_exts = set(itertools.chain(*self._format_selection_exts.values()))
181c7053 2597 for format_id, ambiguous_formats in formats_dict.items():
48ee10ee 2598 ambigious_id = len(ambiguous_formats) > 1
2599 for i, format in enumerate(ambiguous_formats):
2600 if ambigious_id:
181c7053 2601 format['format_id'] = '%s-%d' % (format_id, i)
48ee10ee 2602 if format.get('ext') is None:
2603 format['ext'] = determine_ext(format['url']).lower()
2604 # Ensure there is no conflict between id and ext in format selection
2605 # See https://github.com/yt-dlp/yt-dlp/issues/1282
2606 if format['format_id'] != format['ext'] and format['format_id'] in common_exts:
2607 format['format_id'] = 'f%s' % format['format_id']
181c7053
S
2608
2609 for i, format in enumerate(formats):
8c51aa65 2610 if format.get('format') is None:
6febd1c1 2611 format['format'] = '{id} - {res}{note}'.format(
8c51aa65
JMF
2612 id=format['format_id'],
2613 res=self.format_resolution(format),
b868936c 2614 note=format_field(format, 'format_note', ' (%s)'),
8c51aa65 2615 )
6f0be937 2616 if format.get('protocol') is None:
b5559424 2617 format['protocol'] = determine_protocol(format)
239df021 2618 if format.get('resolution') is None:
2619 format['resolution'] = self.format_resolution(format, default=None)
176f1866 2620 if format.get('dynamic_range') is None and format.get('vcodec') != 'none':
2621 format['dynamic_range'] = 'SDR'
f2fe69c7 2622 if (info_dict.get('duration') and format.get('tbr')
2623 and not format.get('filesize') and not format.get('filesize_approx')):
56ba69e4 2624 format['filesize_approx'] = int(info_dict['duration'] * format['tbr'] * (1024 / 8))
f2fe69c7 2625
e5660ee6
JMF
2626 # Add HTTP headers, so that external programs can use them from the
2627 # json output
2628 full_format_info = info_dict.copy()
2629 full_format_info.update(format)
2630 format['http_headers'] = self._calc_headers(full_format_info)
0016b84e
S
2631 # Remove private housekeeping stuff
2632 if '__x_forwarded_for_ip' in info_dict:
2633 del info_dict['__x_forwarded_for_ip']
dd82ffea 2634
9f1a1c36 2635 if self.params.get('check_formats') is True:
282f5709 2636 formats = LazyList(self._check_formats(formats[::-1]), reverse=True)
9f1a1c36 2637
88acdbc2 2638 if not formats or formats[0] is not info_dict:
b3d9ef88
JMF
2639 # only set the 'formats' fields if the original info_dict list them
2640 # otherwise we end up with a circular reference, the first (and unique)
f89197d7 2641 # element in the 'formats' field in info_dict is info_dict itself,
dfb1b146 2642 # which can't be exported to json
b3d9ef88 2643 info_dict['formats'] = formats
4ec82a72 2644
2645 info_dict, _ = self.pre_process(info_dict)
2646
6db9c4d5 2647 if self._match_entry(info_dict, incomplete=self._format_fields) is not None:
09b49e1f 2648 return info_dict
2649
2650 self.post_extract(info_dict)
2651 info_dict, _ = self.pre_process(info_dict, 'after_filter')
2652
093a1710 2653 # The pre-processors may have modified the formats
aebb4f4b 2654 formats = self._get_formats(info_dict)
093a1710 2655
e4221b70 2656 list_only = self.params.get('simulate') == 'list_only'
fa9f30b8 2657 interactive_format_selection = not list_only and self.format_selector == '-'
b7b04c78 2658 if self.params.get('list_thumbnails'):
2659 self.list_thumbnails(info_dict)
b7b04c78 2660 if self.params.get('listsubtitles'):
2661 if 'automatic_captions' in info_dict:
2662 self.list_subtitles(
2663 info_dict['id'], automatic_captions, 'automatic captions')
2664 self.list_subtitles(info_dict['id'], subtitles, 'subtitles')
fa9f30b8 2665 if self.params.get('listformats') or interactive_format_selection:
b69fd25c 2666 self.list_formats(info_dict)
169dbde9 2667 if list_only:
b7b04c78 2668 # Without this printing, -F --print-json will not work
169dbde9 2669 self.__forced_printings(info_dict, self.prepare_filename(info_dict), incomplete=True)
c487cf00 2670 return info_dict
bfaae0a7 2671
187986a8 2672 format_selector = self.format_selector
2673 if format_selector is None:
0017d9ad 2674 req_format = self._default_format_spec(info_dict, download=download)
0760b0a7 2675 self.write_debug('Default format spec: %s' % req_format)
187986a8 2676 format_selector = self.build_format_selector(req_format)
317f7ab6 2677
fa9f30b8 2678 while True:
2679 if interactive_format_selection:
2680 req_format = input(
2681 self._format_screen('\nEnter format selector: ', self.Styles.EMPHASIS))
2682 try:
2683 format_selector = self.build_format_selector(req_format)
2684 except SyntaxError as err:
2685 self.report_error(err, tb=False, is_error=False)
2686 continue
2687
85e801a9 2688 formats_to_download = list(format_selector({
fa9f30b8 2689 'formats': formats,
85e801a9 2690 'has_merged_format': any('none' not in (f.get('acodec'), f.get('vcodec')) for f in formats),
2691 'incomplete_formats': (
2692 # All formats are video-only or
2693 all(f.get('vcodec') != 'none' and f.get('acodec') == 'none' for f in formats)
2694 # all formats are audio-only
2695 or all(f.get('vcodec') == 'none' and f.get('acodec') != 'none' for f in formats)),
2696 }))
fa9f30b8 2697 if interactive_format_selection and not formats_to_download:
2698 self.report_error('Requested format is not available', tb=False, is_error=False)
2699 continue
2700 break
317f7ab6 2701
dd82ffea 2702 if not formats_to_download:
b7da73eb 2703 if not self.params.get('ignore_no_formats_error'):
c0b6e5c7 2704 raise ExtractorError(
2705 'Requested format is not available. Use --list-formats for a list of available formats',
2706 expected=True, video_id=info_dict['id'], ie=info_dict['extractor'])
b62fa6d7 2707 self.report_warning('Requested format is not available')
2708 # Process what we can, even without any available formats.
2709 formats_to_download = [{}]
a13e6848 2710
0500ee3d 2711 requested_ranges = tuple(self.params.get('download_ranges', lambda *_: [{}])(info_dict, self))
5ec1b6b7 2712 best_format, downloaded_formats = formats_to_download[-1], []
b62fa6d7 2713 if download:
0500ee3d 2714 if best_format and requested_ranges:
5ec1b6b7 2715 def to_screen(*msg):
2716 self.to_screen(f'[info] {info_dict["id"]}: {" ".join(", ".join(variadic(m)) for m in msg)}')
2717
2718 to_screen(f'Downloading {len(formats_to_download)} format(s):',
2719 (f['format_id'] for f in formats_to_download))
0500ee3d 2720 if requested_ranges != ({}, ):
5ec1b6b7 2721 to_screen(f'Downloading {len(requested_ranges)} time ranges:',
fc2ba496 2722 (f'{c["start_time"]:.1f}-{c["end_time"]:.1f}' for c in requested_ranges))
a13e6848 2723 max_downloads_reached = False
5ec1b6b7 2724
0500ee3d 2725 for fmt, chapter in itertools.product(formats_to_download, requested_ranges):
5ec1b6b7 2726 new_info = self._copy_infodict(info_dict)
b7da73eb 2727 new_info.update(fmt)
3975b4d2 2728 offset, duration = info_dict.get('section_start') or 0, info_dict.get('duration') or float('inf')
fc2ba496 2729 end_time = offset + min(chapter.get('end_time', duration), duration)
3975b4d2 2730 if chapter or offset:
5ec1b6b7 2731 new_info.update({
3975b4d2 2732 'section_start': offset + chapter.get('start_time', 0),
2576d53a 2733 # duration may not be accurate. So allow deviations <1sec
2734 'section_end': end_time if end_time <= offset + duration + 1 else None,
5ec1b6b7 2735 'section_title': chapter.get('title'),
2736 'section_number': chapter.get('index'),
2737 })
2738 downloaded_formats.append(new_info)
a13e6848 2739 try:
2740 self.process_info(new_info)
2741 except MaxDownloadsReached:
2742 max_downloads_reached = True
415f8d51 2743 self._raise_pending_errors(new_info)
f46e2f9d 2744 # Remove copied info
2745 for key, val in tuple(new_info.items()):
2746 if info_dict.get(key) == val:
2747 new_info.pop(key)
a13e6848 2748 if max_downloads_reached:
2749 break
ebed8b37 2750
5ec1b6b7 2751 write_archive = {f.get('__write_download_archive', False) for f in downloaded_formats}
a13e6848 2752 assert write_archive.issubset({True, False, 'ignore'})
2753 if True in write_archive and False not in write_archive:
2754 self.record_download_archive(info_dict)
be72c624 2755
5ec1b6b7 2756 info_dict['requested_downloads'] = downloaded_formats
ed5835b4 2757 info_dict = self.run_all_pps('after_video', info_dict)
a13e6848 2758 if max_downloads_reached:
2759 raise MaxDownloadsReached()
ebed8b37 2760
49a57e70 2761 # We update the info dict with the selected best quality format (backwards compatibility)
be72c624 2762 info_dict.update(best_format)
dd82ffea
JMF
2763 return info_dict
2764
98c70d6f 2765 def process_subtitles(self, video_id, normal_subtitles, automatic_captions):
a504ced0 2766 """Select the requested subtitles and their format"""
d8a58ddc 2767 available_subs, normal_sub_langs = {}, []
98c70d6f
JMF
2768 if normal_subtitles and self.params.get('writesubtitles'):
2769 available_subs.update(normal_subtitles)
d8a58ddc 2770 normal_sub_langs = tuple(normal_subtitles.keys())
98c70d6f
JMF
2771 if automatic_captions and self.params.get('writeautomaticsub'):
2772 for lang, cap_info in automatic_captions.items():
360e1ca5
JMF
2773 if lang not in available_subs:
2774 available_subs[lang] = cap_info
2775
d2c8aadf 2776 if not available_subs or (
2777 not self.params.get('writesubtitles')
2778 and not self.params.get('writeautomaticsub')):
4d171848 2779 return None
a504ced0 2780
d8a58ddc 2781 all_sub_langs = tuple(available_subs.keys())
a504ced0 2782 if self.params.get('allsubtitles', False):
c32b0aab 2783 requested_langs = all_sub_langs
2784 elif self.params.get('subtitleslangs', False):
5314b521 2785 try:
2786 requested_langs = orderedSet_from_options(
2787 self.params.get('subtitleslangs'), {'all': all_sub_langs}, use_regex=True)
2788 except re.error as e:
2789 raise ValueError(f'Wrong regex for subtitlelangs: {e.pattern}')
d8a58ddc 2790 elif normal_sub_langs:
2791 requested_langs = ['en'] if 'en' in normal_sub_langs else normal_sub_langs[:1]
a504ced0 2792 else:
d8a58ddc 2793 requested_langs = ['en'] if 'en' in all_sub_langs else all_sub_langs[:1]
ad3dc496 2794 if requested_langs:
d2c8aadf 2795 self.to_screen(f'[info] {video_id}: Downloading subtitles: {", ".join(requested_langs)}')
a504ced0
JMF
2796
2797 formats_query = self.params.get('subtitlesformat', 'best')
2798 formats_preference = formats_query.split('/') if formats_query else []
2799 subs = {}
2800 for lang in requested_langs:
2801 formats = available_subs.get(lang)
2802 if formats is None:
86e5f3ed 2803 self.report_warning(f'{lang} subtitles not available for {video_id}')
a504ced0 2804 continue
a504ced0
JMF
2805 for ext in formats_preference:
2806 if ext == 'best':
2807 f = formats[-1]
2808 break
2809 matches = list(filter(lambda f: f['ext'] == ext, formats))
2810 if matches:
2811 f = matches[-1]
2812 break
2813 else:
2814 f = formats[-1]
2815 self.report_warning(
2816 'No subtitle format found matching "%s" for language %s, '
2817 'using %s' % (formats_query, lang, f['ext']))
2818 subs[lang] = f
2819 return subs
2820
bb66c247 2821 def _forceprint(self, key, info_dict):
2822 if info_dict is None:
2823 return
2824 info_copy = info_dict.copy()
2825 info_copy['formats_table'] = self.render_formats_table(info_dict)
2826 info_copy['thumbnails_table'] = self.render_thumbnails_table(info_dict)
2827 info_copy['subtitles_table'] = self.render_subtitles_table(info_dict.get('id'), info_dict.get('subtitles'))
2828 info_copy['automatic_captions_table'] = self.render_subtitles_table(info_dict.get('id'), info_dict.get('automatic_captions'))
2829
2830 def format_tmpl(tmpl):
48c8424b 2831 mobj = re.fullmatch(r'([\w.:,]|-\d|(?P<dict>{([\w.:,]|-\d)+}))+=?', tmpl)
07a1250e 2832 if not mobj:
2833 return tmpl
48c8424b 2834
2835 fmt = '%({})s'
2836 if tmpl.startswith('{'):
2837 tmpl = f'.{tmpl}'
2838 if tmpl.endswith('='):
2839 tmpl, fmt = tmpl[:-1], '{0} = %({0})#j'
2840 return '\n'.join(map(fmt.format, [tmpl] if mobj.group('dict') else tmpl.split(',')))
8130779d 2841
bb66c247 2842 for tmpl in self.params['forceprint'].get(key, []):
2843 self.to_stdout(self.evaluate_outtmpl(format_tmpl(tmpl), info_copy))
2844
2845 for tmpl, file_tmpl in self.params['print_to_file'].get(key, []):
5127e92a 2846 filename = self.prepare_filename(info_dict, outtmpl=file_tmpl)
bb66c247 2847 tmpl = format_tmpl(tmpl)
2848 self.to_screen(f'[info] Writing {tmpl!r} to: {filename}')
8d93e69d 2849 if self._ensure_dir_exists(filename):
86e5f3ed 2850 with open(filename, 'a', encoding='utf-8') as f:
8d93e69d 2851 f.write(self.evaluate_outtmpl(tmpl, info_copy) + '\n')
ca30f449 2852
d06daf23 2853 def __forced_printings(self, info_dict, filename, incomplete):
53c18592 2854 def print_mandatory(field, actual_field=None):
2855 if actual_field is None:
2856 actual_field = field
d06daf23 2857 if (self.params.get('force%s' % field, False)
53c18592 2858 and (not incomplete or info_dict.get(actual_field) is not None)):
2859 self.to_stdout(info_dict[actual_field])
d06daf23
S
2860
2861 def print_optional(field):
2862 if (self.params.get('force%s' % field, False)
2863 and info_dict.get(field) is not None):
2864 self.to_stdout(info_dict[field])
2865
53c18592 2866 info_dict = info_dict.copy()
2867 if filename is not None:
2868 info_dict['filename'] = filename
2869 if info_dict.get('requested_formats') is not None:
2870 # For RTMP URLs, also include the playpath
2871 info_dict['urls'] = '\n'.join(f['url'] + f.get('play_path', '') for f in info_dict['requested_formats'])
10331a26 2872 elif info_dict.get('url'):
53c18592 2873 info_dict['urls'] = info_dict['url'] + info_dict.get('play_path', '')
2874
bb66c247 2875 if (self.params.get('forcejson')
2876 or self.params['forceprint'].get('video')
2877 or self.params['print_to_file'].get('video')):
2b8a2973 2878 self.post_extract(info_dict)
bb66c247 2879 self._forceprint('video', info_dict)
53c18592 2880
d06daf23
S
2881 print_mandatory('title')
2882 print_mandatory('id')
53c18592 2883 print_mandatory('url', 'urls')
d06daf23
S
2884 print_optional('thumbnail')
2885 print_optional('description')
53c18592 2886 print_optional('filename')
b868936c 2887 if self.params.get('forceduration') and info_dict.get('duration') is not None:
d06daf23
S
2888 self.to_stdout(formatSeconds(info_dict['duration']))
2889 print_mandatory('format')
53c18592 2890
2b8a2973 2891 if self.params.get('forcejson'):
6e84b215 2892 self.to_stdout(json.dumps(self.sanitize_info(info_dict)))
d06daf23 2893
e8e73840 2894 def dl(self, name, info, subtitle=False, test=False):
88acdbc2 2895 if not info.get('url'):
1151c407 2896 self.raise_no_formats(info, True)
e8e73840 2897
2898 if test:
2899 verbose = self.params.get('verbose')
2900 params = {
2901 'test': True,
a169858f 2902 'quiet': self.params.get('quiet') or not verbose,
e8e73840 2903 'verbose': verbose,
2904 'noprogress': not verbose,
2905 'nopart': True,
2906 'skip_unavailable_fragments': False,
2907 'keep_fragments': False,
2908 'overwrites': True,
2909 '_no_ytdl_file': True,
2910 }
2911 else:
2912 params = self.params
96fccc10 2913 fd = get_suitable_downloader(info, params, to_stdout=(name == '-'))(self, params)
e8e73840 2914 if not test:
2915 for ph in self._progress_hooks:
2916 fd.add_progress_hook(ph)
42676437
M
2917 urls = '", "'.join(
2918 (f['url'].split(',')[0] + ',<data>' if f['url'].startswith('data:') else f['url'])
2919 for f in info.get('requested_formats', []) or [info])
3a408f9d 2920 self.write_debug(f'Invoking {fd.FD_NAME} downloader on "{urls}"')
03b4de72 2921
adbc4ec4
THD
2922 # Note: Ideally info should be a deep-copied so that hooks cannot modify it.
2923 # But it may contain objects that are not deep-copyable
2924 new_info = self._copy_infodict(info)
e8e73840 2925 if new_info.get('http_headers') is None:
2926 new_info['http_headers'] = self._calc_headers(new_info)
2927 return fd.download(name, new_info, subtitle)
2928
e04938ab 2929 def existing_file(self, filepaths, *, default_overwrite=True):
2930 existing_files = list(filter(os.path.exists, orderedSet(filepaths)))
2931 if existing_files and not self.params.get('overwrites', default_overwrite):
2932 return existing_files[0]
2933
2934 for file in existing_files:
2935 self.report_file_delete(file)
2936 os.remove(file)
2937 return None
2938
8222d8de 2939 def process_info(self, info_dict):
09b49e1f 2940 """Process a single resolved IE result. (Modifies it in-place)"""
8222d8de
JMF
2941
2942 assert info_dict.get('_type', 'video') == 'video'
f46e2f9d 2943 original_infodict = info_dict
fd288278 2944
4513a41a 2945 if 'format' not in info_dict and 'ext' in info_dict:
8222d8de
JMF
2946 info_dict['format'] = info_dict['ext']
2947
09b49e1f 2948 # This is mostly just for backward compatibility of process_info
2949 # As a side-effect, this allows for format-specific filters
c77495e3 2950 if self._match_entry(info_dict) is not None:
9e907ebd 2951 info_dict['__write_download_archive'] = 'ignore'
8222d8de
JMF
2952 return
2953
09b49e1f 2954 # Does nothing under normal operation - for backward compatibility of process_info
277d6ff5 2955 self.post_extract(info_dict)
0c14d66a 2956 self._num_downloads += 1
8222d8de 2957
dcf64d43 2958 # info_dict['_filename'] needs to be set for backward compatibility
de6000d9 2959 info_dict['_filename'] = full_filename = self.prepare_filename(info_dict, warn=True)
2960 temp_filename = self.prepare_filename(info_dict, 'temp')
0202b52a 2961 files_to_move = {}
8222d8de
JMF
2962
2963 # Forced printings
4513a41a 2964 self.__forced_printings(info_dict, full_filename, incomplete=('format' not in info_dict))
8222d8de 2965
ca6d59d2 2966 def check_max_downloads():
2967 if self._num_downloads >= float(self.params.get('max_downloads') or 'inf'):
2968 raise MaxDownloadsReached()
2969
b7b04c78 2970 if self.params.get('simulate'):
9e907ebd 2971 info_dict['__write_download_archive'] = self.params.get('force_write_download_archive')
ca6d59d2 2972 check_max_downloads()
8222d8de
JMF
2973 return
2974
de6000d9 2975 if full_filename is None:
8222d8de 2976 return
e92caff5 2977 if not self._ensure_dir_exists(encodeFilename(full_filename)):
0202b52a 2978 return
e92caff5 2979 if not self._ensure_dir_exists(encodeFilename(temp_filename)):
8222d8de
JMF
2980 return
2981
80c03fa9 2982 if self._write_description('video', info_dict,
2983 self.prepare_filename(info_dict, 'description')) is None:
2984 return
2985
2986 sub_files = self._write_subtitles(info_dict, temp_filename)
2987 if sub_files is None:
2988 return
2989 files_to_move.update(dict(sub_files))
2990
2991 thumb_files = self._write_thumbnails(
2992 'video', info_dict, temp_filename, self.prepare_filename(info_dict, 'thumbnail'))
2993 if thumb_files is None:
2994 return
2995 files_to_move.update(dict(thumb_files))
8222d8de 2996
80c03fa9 2997 infofn = self.prepare_filename(info_dict, 'infojson')
2998 _infojson_written = self._write_info_json('video', info_dict, infofn)
2999 if _infojson_written:
dac5df5a 3000 info_dict['infojson_filename'] = infofn
e75bb0d6 3001 # For backward compatibility, even though it was a private field
80c03fa9 3002 info_dict['__infojson_filename'] = infofn
3003 elif _infojson_written is None:
3004 return
3005
3006 # Note: Annotations are deprecated
3007 annofn = None
1fb07d10 3008 if self.params.get('writeannotations', False):
de6000d9 3009 annofn = self.prepare_filename(info_dict, 'annotation')
80c03fa9 3010 if annofn:
e92caff5 3011 if not self._ensure_dir_exists(encodeFilename(annofn)):
0202b52a 3012 return
0c3d0f51 3013 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(annofn)):
6febd1c1 3014 self.to_screen('[info] Video annotations are already present')
ffddb112
RA
3015 elif not info_dict.get('annotations'):
3016 self.report_warning('There are no annotations to write.')
7b6fefc9
PH
3017 else:
3018 try:
6febd1c1 3019 self.to_screen('[info] Writing video annotations to: ' + annofn)
86e5f3ed 3020 with open(encodeFilename(annofn), 'w', encoding='utf-8') as annofile:
7b6fefc9
PH
3021 annofile.write(info_dict['annotations'])
3022 except (KeyError, TypeError):
6febd1c1 3023 self.report_warning('There are no annotations to write.')
86e5f3ed 3024 except OSError:
6febd1c1 3025 self.report_error('Cannot write annotations file: ' + annofn)
7b6fefc9 3026 return
1fb07d10 3027
732044af 3028 # Write internet shortcut files
08438d2c 3029 def _write_link_file(link_type):
60f3e995 3030 url = try_get(info_dict['webpage_url'], iri_to_uri)
3031 if not url:
3032 self.report_warning(
3033 f'Cannot write internet shortcut file because the actual URL of "{info_dict["webpage_url"]}" is unknown')
3034 return True
08438d2c 3035 linkfn = replace_extension(self.prepare_filename(info_dict, 'link'), link_type, info_dict.get('ext'))
0e6b018a
Z
3036 if not self._ensure_dir_exists(encodeFilename(linkfn)):
3037 return False
10e3742e 3038 if self.params.get('overwrites', True) and os.path.exists(encodeFilename(linkfn)):
08438d2c 3039 self.to_screen(f'[info] Internet shortcut (.{link_type}) is already present')
3040 return True
3041 try:
3042 self.to_screen(f'[info] Writing internet shortcut (.{link_type}) to: {linkfn}')
86e5f3ed 3043 with open(encodeFilename(to_high_limit_path(linkfn)), 'w', encoding='utf-8',
3044 newline='\r\n' if link_type == 'url' else '\n') as linkfile:
60f3e995 3045 template_vars = {'url': url}
08438d2c 3046 if link_type == 'desktop':
3047 template_vars['filename'] = linkfn[:-(len(link_type) + 1)]
3048 linkfile.write(LINK_TEMPLATES[link_type] % template_vars)
86e5f3ed 3049 except OSError:
08438d2c 3050 self.report_error(f'Cannot write internet shortcut {linkfn}')
3051 return False
732044af 3052 return True
3053
08438d2c 3054 write_links = {
3055 'url': self.params.get('writeurllink'),
3056 'webloc': self.params.get('writewebloclink'),
3057 'desktop': self.params.get('writedesktoplink'),
3058 }
3059 if self.params.get('writelink'):
3060 link_type = ('webloc' if sys.platform == 'darwin'
3061 else 'desktop' if sys.platform.startswith('linux')
3062 else 'url')
3063 write_links[link_type] = True
3064
3065 if any(should_write and not _write_link_file(link_type)
3066 for link_type, should_write in write_links.items()):
3067 return
732044af 3068
f46e2f9d 3069 def replace_info_dict(new_info):
3070 nonlocal info_dict
3071 if new_info == info_dict:
3072 return
3073 info_dict.clear()
3074 info_dict.update(new_info)
3075
415f8d51 3076 new_info, files_to_move = self.pre_process(info_dict, 'before_dl', files_to_move)
3077 replace_info_dict(new_info)
56d868db 3078
a13e6848 3079 if self.params.get('skip_download'):
56d868db 3080 info_dict['filepath'] = temp_filename
3081 info_dict['__finaldir'] = os.path.dirname(os.path.abspath(encodeFilename(full_filename)))
3082 info_dict['__files_to_move'] = files_to_move
f46e2f9d 3083 replace_info_dict(self.run_pp(MoveFilesAfterDownloadPP(self, False), info_dict))
9e907ebd 3084 info_dict['__write_download_archive'] = self.params.get('force_write_download_archive')
56d868db 3085 else:
3086 # Download
b868936c 3087 info_dict.setdefault('__postprocessors', [])
4340deca 3088 try:
0202b52a 3089
e04938ab 3090 def existing_video_file(*filepaths):
6b591b29 3091 ext = info_dict.get('ext')
e04938ab 3092 converted = lambda file: replace_extension(file, self.params.get('final_ext') or ext, ext)
3093 file = self.existing_file(itertools.chain(*zip(map(converted, filepaths), filepaths)),
3094 default_overwrite=False)
3095 if file:
3096 info_dict['ext'] = os.path.splitext(file)[1][1:]
3097 return file
0202b52a 3098
7b2c3f47 3099 fd, success = None, True
fccf90e7 3100 if info_dict.get('protocol') or info_dict.get('url'):
56ba69e4 3101 fd = get_suitable_downloader(info_dict, self.params, to_stdout=temp_filename == '-')
3102 if fd is not FFmpegFD and (
3103 info_dict.get('section_start') or info_dict.get('section_end')):
7b2c3f47 3104 msg = ('This format cannot be partially downloaded' if FFmpegFD.available()
56ba69e4 3105 else 'You have requested downloading the video partially, but ffmpeg is not installed')
3106 self.report_error(f'{msg}. Aborting')
5ec1b6b7 3107 return
5ec1b6b7 3108
4340deca 3109 if info_dict.get('requested_formats') is not None:
81cd954a 3110 requested_formats = info_dict['requested_formats']
0202b52a 3111 old_ext = info_dict['ext']
4e3b637d 3112 if self.params.get('merge_output_format') is None:
4e3b637d 3113 if (info_dict['ext'] == 'webm'
3114 and info_dict.get('thumbnails')
3115 # check with type instead of pp_key, __name__, or isinstance
3116 # since we dont want any custom PPs to trigger this
c487cf00 3117 and any(type(pp) == EmbedThumbnailPP for pp in self._pps['post_process'])): # noqa: E721
4e3b637d 3118 info_dict['ext'] = 'mkv'
3119 self.report_warning(
3120 'webm doesn\'t support embedding a thumbnail, mkv will be used')
124bc071 3121 new_ext = info_dict['ext']
0202b52a 3122
124bc071 3123 def correct_ext(filename, ext=new_ext):
96fccc10 3124 if filename == '-':
3125 return filename
0202b52a 3126 filename_real_ext = os.path.splitext(filename)[1][1:]
3127 filename_wo_ext = (
3128 os.path.splitext(filename)[0]
124bc071 3129 if filename_real_ext in (old_ext, new_ext)
0202b52a 3130 else filename)
86e5f3ed 3131 return f'{filename_wo_ext}.{ext}'
0202b52a 3132
38c6902b 3133 # Ensure filename always has a correct extension for successful merge
0202b52a 3134 full_filename = correct_ext(full_filename)
3135 temp_filename = correct_ext(temp_filename)
e04938ab 3136 dl_filename = existing_video_file(full_filename, temp_filename)
1ea24129 3137 info_dict['__real_download'] = False
18e674b4 3138
7b2c3f47 3139 merger = FFmpegMergerPP(self)
adbc4ec4 3140 downloaded = []
dbf5416a 3141 if dl_filename is not None:
6c7274ec 3142 self.report_file_already_downloaded(dl_filename)
adbc4ec4
THD
3143 elif fd:
3144 for f in requested_formats if fd != FFmpegFD else []:
3145 f['filepath'] = fname = prepend_extension(
3146 correct_ext(temp_filename, info_dict['ext']),
3147 'f%s' % f['format_id'], info_dict['ext'])
3148 downloaded.append(fname)
dbf5416a 3149 info_dict['url'] = '\n'.join(f['url'] for f in requested_formats)
3150 success, real_download = self.dl(temp_filename, info_dict)
3151 info_dict['__real_download'] = real_download
18e674b4 3152 else:
18e674b4 3153 if self.params.get('allow_unplayable_formats'):
3154 self.report_warning(
3155 'You have requested merging of multiple formats '
3156 'while also allowing unplayable formats to be downloaded. '
3157 'The formats won\'t be merged to prevent data corruption.')
3158 elif not merger.available:
e8969bda 3159 msg = 'You have requested merging of multiple formats but ffmpeg is not installed'
3160 if not self.params.get('ignoreerrors'):
3161 self.report_error(f'{msg}. Aborting due to --abort-on-error')
3162 return
3163 self.report_warning(f'{msg}. The formats won\'t be merged')
18e674b4 3164
96fccc10 3165 if temp_filename == '-':
adbc4ec4 3166 reason = ('using a downloader other than ffmpeg' if FFmpegFD.can_merge_formats(info_dict, self.params)
96fccc10 3167 else 'but the formats are incompatible for simultaneous download' if merger.available
3168 else 'but ffmpeg is not installed')
3169 self.report_warning(
3170 f'You have requested downloading multiple formats to stdout {reason}. '
3171 'The formats will be streamed one after the other')
3172 fname = temp_filename
dbf5416a 3173 for f in requested_formats:
3174 new_info = dict(info_dict)
3175 del new_info['requested_formats']
3176 new_info.update(f)
96fccc10 3177 if temp_filename != '-':
124bc071 3178 fname = prepend_extension(
3179 correct_ext(temp_filename, new_info['ext']),
3180 'f%s' % f['format_id'], new_info['ext'])
96fccc10 3181 if not self._ensure_dir_exists(fname):
3182 return
a21e0ab1 3183 f['filepath'] = fname
96fccc10 3184 downloaded.append(fname)
dbf5416a 3185 partial_success, real_download = self.dl(fname, new_info)
3186 info_dict['__real_download'] = info_dict['__real_download'] or real_download
3187 success = success and partial_success
adbc4ec4
THD
3188
3189 if downloaded and merger.available and not self.params.get('allow_unplayable_formats'):
3190 info_dict['__postprocessors'].append(merger)
3191 info_dict['__files_to_merge'] = downloaded
3192 # Even if there were no downloads, it is being merged only now
3193 info_dict['__real_download'] = True
3194 else:
3195 for file in downloaded:
3196 files_to_move[file] = None
4340deca
P
3197 else:
3198 # Just a single file
e04938ab 3199 dl_filename = existing_video_file(full_filename, temp_filename)
6c7274ec 3200 if dl_filename is None or dl_filename == temp_filename:
3201 # dl_filename == temp_filename could mean that the file was partially downloaded with --no-part.
3202 # So we should try to resume the download
e8e73840 3203 success, real_download = self.dl(temp_filename, info_dict)
0202b52a 3204 info_dict['__real_download'] = real_download
6c7274ec 3205 else:
3206 self.report_file_already_downloaded(dl_filename)
0202b52a 3207
0202b52a 3208 dl_filename = dl_filename or temp_filename
c571435f 3209 info_dict['__finaldir'] = os.path.dirname(os.path.abspath(encodeFilename(full_filename)))
0202b52a 3210
3158150c 3211 except network_exceptions as err:
7960b056 3212 self.report_error('unable to download video data: %s' % error_to_compat_str(err))
4340deca 3213 return
86e5f3ed 3214 except OSError as err:
4340deca
P
3215 raise UnavailableVideoError(err)
3216 except (ContentTooShortError, ) as err:
86e5f3ed 3217 self.report_error(f'content too short (expected {err.expected} bytes and served {err.downloaded})')
4340deca 3218 return
8222d8de 3219
415f8d51 3220 self._raise_pending_errors(info_dict)
de6000d9 3221 if success and full_filename != '-':
f17f8651 3222
fd7cfb64 3223 def fixup():
3224 do_fixup = True
3225 fixup_policy = self.params.get('fixup')
3226 vid = info_dict['id']
3227
3228 if fixup_policy in ('ignore', 'never'):
3229 return
3230 elif fixup_policy == 'warn':
3fe75fdc 3231 do_fixup = 'warn'
f89b3e2d 3232 elif fixup_policy != 'force':
3233 assert fixup_policy in ('detect_or_warn', None)
3234 if not info_dict.get('__real_download'):
3235 do_fixup = False
fd7cfb64 3236
3237 def ffmpeg_fixup(cndn, msg, cls):
3fe75fdc 3238 if not (do_fixup and cndn):
fd7cfb64 3239 return
3fe75fdc 3240 elif do_fixup == 'warn':
fd7cfb64 3241 self.report_warning(f'{vid}: {msg}')
3242 return
3243 pp = cls(self)
3244 if pp.available:
3245 info_dict['__postprocessors'].append(pp)
3246 else:
3247 self.report_warning(f'{vid}: {msg}. Install ffmpeg to fix this automatically')
3248
3249 stretched_ratio = info_dict.get('stretched_ratio')
ca9def71
LNO
3250 ffmpeg_fixup(stretched_ratio not in (1, None),
3251 f'Non-uniform pixel ratio {stretched_ratio}',
3252 FFmpegFixupStretchedPP)
fd7cfb64 3253
993191c0 3254 downloader = get_suitable_downloader(info_dict, self.params) if 'protocol' in info_dict else None
24146491 3255 downloader = downloader.FD_NAME if downloader else None
adbc4ec4 3256
ca9def71
LNO
3257 ext = info_dict.get('ext')
3258 postprocessed_by_ffmpeg = info_dict.get('requested_formats') or any((
3259 isinstance(pp, FFmpegVideoConvertorPP)
3260 and resolve_recode_mapping(ext, pp.mapping)[0] not in (ext, None)
3261 ) for pp in self._pps['post_process'])
3262
3263 if not postprocessed_by_ffmpeg:
3264 ffmpeg_fixup(ext == 'm4a' and info_dict.get('container') == 'm4a_dash',
f2df4071 3265 'writing DASH m4a. Only some players support this container',
3266 FFmpegFixupM4aPP)
24146491 3267 ffmpeg_fixup(downloader == 'hlsnative' and not self.params.get('hls_use_mpegts')
494f5230 3268 or info_dict.get('is_live') and self.params.get('hls_use_mpegts') is None,
adbc4ec4
THD
3269 'Possible MPEG-TS in MP4 container or malformed AAC timestamps',
3270 FFmpegFixupM3u8PP)
3271 ffmpeg_fixup(info_dict.get('is_live') and downloader == 'DashSegmentsFD',
3272 'Possible duplicate MOOV atoms', FFmpegFixupDuplicateMoovPP)
3273
24146491 3274 ffmpeg_fixup(downloader == 'web_socket_fragment', 'Malformed timestamps detected', FFmpegFixupTimestampPP)
3275 ffmpeg_fixup(downloader == 'web_socket_fragment', 'Malformed duration detected', FFmpegFixupDurationPP)
fd7cfb64 3276
3277 fixup()
8222d8de 3278 try:
f46e2f9d 3279 replace_info_dict(self.post_process(dl_filename, info_dict, files_to_move))
af819c21 3280 except PostProcessingError as err:
3281 self.report_error('Postprocessing: %s' % str(err))
8222d8de 3282 return
ab8e5e51
AM
3283 try:
3284 for ph in self._post_hooks:
23c1a667 3285 ph(info_dict['filepath'])
ab8e5e51
AM
3286 except Exception as err:
3287 self.report_error('post hooks: %s' % str(err))
3288 return
9e907ebd 3289 info_dict['__write_download_archive'] = True
2d30509f 3290
c487cf00 3291 assert info_dict is original_infodict # Make sure the info_dict was modified in-place
a13e6848 3292 if self.params.get('force_write_download_archive'):
9e907ebd 3293 info_dict['__write_download_archive'] = True
ca6d59d2 3294 check_max_downloads()
8222d8de 3295
aa9369a2 3296 def __download_wrapper(self, func):
3297 @functools.wraps(func)
3298 def wrapper(*args, **kwargs):
3299 try:
3300 res = func(*args, **kwargs)
3301 except UnavailableVideoError as e:
3302 self.report_error(e)
b222c271 3303 except DownloadCancelled as e:
3304 self.to_screen(f'[info] {e}')
3305 if not self.params.get('break_per_url'):
3306 raise
fd404bec 3307 self._num_downloads = 0
aa9369a2 3308 else:
3309 if self.params.get('dump_single_json', False):
3310 self.post_extract(res)
3311 self.to_stdout(json.dumps(self.sanitize_info(res)))
3312 return wrapper
3313
8222d8de
JMF
3314 def download(self, url_list):
3315 """Download a given list of URLs."""
aa9369a2 3316 url_list = variadic(url_list) # Passing a single URL is a common mistake
bf1824b3 3317 outtmpl = self.params['outtmpl']['default']
3089bc74
S
3318 if (len(url_list) > 1
3319 and outtmpl != '-'
3320 and '%' not in outtmpl
3321 and self.params.get('max_downloads') != 1):
acd69589 3322 raise SameFileError(outtmpl)
8222d8de
JMF
3323
3324 for url in url_list:
aa9369a2 3325 self.__download_wrapper(self.extract_info)(
3326 url, force_generic_extractor=self.params.get('force_generic_extractor', False))
8222d8de
JMF
3327
3328 return self._download_retcode
3329
1dcc4c0c 3330 def download_with_info_file(self, info_filename):
31bd3925
JMF
3331 with contextlib.closing(fileinput.FileInput(
3332 [info_filename], mode='r',
3333 openhook=fileinput.hook_encoded('utf-8'))) as f:
3334 # FileInput doesn't have a read method, we can't call json.load
8012d892 3335 info = self.sanitize_info(json.loads('\n'.join(f)), self.params.get('clean_infojson', True))
d4943898 3336 try:
aa9369a2 3337 self.__download_wrapper(self.process_ie_result)(info, download=True)
f2ebc5c7 3338 except (DownloadError, EntryNotInPlaylist, ReExtractInfo) as e:
bf5f605e 3339 if not isinstance(e, EntryNotInPlaylist):
3340 self.to_stderr('\r')
d4943898
JMF
3341 webpage_url = info.get('webpage_url')
3342 if webpage_url is not None:
aa9369a2 3343 self.report_warning(f'The info failed to download: {e}; trying with URL {webpage_url}')
d4943898
JMF
3344 return self.download([webpage_url])
3345 else:
3346 raise
3347 return self._download_retcode
1dcc4c0c 3348
cb202fd2 3349 @staticmethod
8012d892 3350 def sanitize_info(info_dict, remove_private_keys=False):
3351 ''' Sanitize the infodict for converting to json '''
3ad56b42 3352 if info_dict is None:
3353 return info_dict
6e84b215 3354 info_dict.setdefault('epoch', int(time.time()))
6a5a30f9 3355 info_dict.setdefault('_type', 'video')
b5e7a2e6 3356 info_dict.setdefault('_version', {
3357 'version': __version__,
3358 'current_git_head': current_git_head(),
3359 'release_git_head': RELEASE_GIT_HEAD,
3360 'repository': REPOSITORY,
3361 })
09b49e1f 3362
8012d892 3363 if remove_private_keys:
0a5a191a 3364 reject = lambda k, v: v is None or k.startswith('__') or k in {
f46e2f9d 3365 'requested_downloads', 'requested_formats', 'requested_subtitles', 'requested_entries',
0a5a191a 3366 'entries', 'filepath', '_filename', 'infojson_filename', 'original_url', 'playlist_autonumber',
6e84b215 3367 }
ae8f99e6 3368 else:
09b49e1f 3369 reject = lambda k, v: False
adbc4ec4
THD
3370
3371 def filter_fn(obj):
3372 if isinstance(obj, dict):
3373 return {k: filter_fn(v) for k, v in obj.items() if not reject(k, v)}
3374 elif isinstance(obj, (list, tuple, set, LazyList)):
3375 return list(map(filter_fn, obj))
3376 elif obj is None or isinstance(obj, (str, int, float, bool)):
3377 return obj
3378 else:
3379 return repr(obj)
3380
5226731e 3381 return filter_fn(info_dict)
cb202fd2 3382
8012d892 3383 @staticmethod
3384 def filter_requested_info(info_dict, actually_filter=True):
3385 ''' Alias of sanitize_info for backward compatibility '''
3386 return YoutubeDL.sanitize_info(info_dict, actually_filter)
3387
43d7f5a5 3388 def _delete_downloaded_files(self, *files_to_delete, info={}, msg=None):
3389 for filename in set(filter(None, files_to_delete)):
3390 if msg:
3391 self.to_screen(msg % filename)
3392 try:
3393 os.remove(filename)
3394 except OSError:
3395 self.report_warning(f'Unable to delete file {filename}')
3396 if filename in info.get('__files_to_move', []): # NB: Delete even if None
3397 del info['__files_to_move'][filename]
3398
ed5835b4 3399 @staticmethod
3400 def post_extract(info_dict):
3401 def actual_post_extract(info_dict):
3402 if info_dict.get('_type') in ('playlist', 'multi_video'):
3403 for video_dict in info_dict.get('entries', {}):
3404 actual_post_extract(video_dict or {})
3405 return
3406
09b49e1f 3407 post_extractor = info_dict.pop('__post_extractor', None) or (lambda: {})
3408 info_dict.update(post_extractor())
ed5835b4 3409
3410 actual_post_extract(info_dict or {})
3411
dcf64d43 3412 def run_pp(self, pp, infodict):
5bfa4862 3413 files_to_delete = []
dcf64d43 3414 if '__files_to_move' not in infodict:
3415 infodict['__files_to_move'] = {}
b1940459 3416 try:
3417 files_to_delete, infodict = pp.run(infodict)
3418 except PostProcessingError as e:
3419 # Must be True and not 'only_download'
3420 if self.params.get('ignoreerrors') is True:
3421 self.report_error(e)
3422 return infodict
3423 raise
3424
5bfa4862 3425 if not files_to_delete:
dcf64d43 3426 return infodict
5bfa4862 3427 if self.params.get('keepvideo', False):
3428 for f in files_to_delete:
dcf64d43 3429 infodict['__files_to_move'].setdefault(f, '')
5bfa4862 3430 else:
43d7f5a5 3431 self._delete_downloaded_files(
3432 *files_to_delete, info=infodict, msg='Deleting original file %s (pass -k to keep)')
dcf64d43 3433 return infodict
5bfa4862 3434
ed5835b4 3435 def run_all_pps(self, key, info, *, additional_pps=None):
bb66c247 3436 self._forceprint(key, info)
ed5835b4 3437 for pp in (additional_pps or []) + self._pps[key]:
dc5f409c 3438 info = self.run_pp(pp, info)
ed5835b4 3439 return info
277d6ff5 3440
56d868db 3441 def pre_process(self, ie_info, key='pre_process', files_to_move=None):
5bfa4862 3442 info = dict(ie_info)
56d868db 3443 info['__files_to_move'] = files_to_move or {}
415f8d51 3444 try:
3445 info = self.run_all_pps(key, info)
3446 except PostProcessingError as err:
3447 msg = f'Preprocessing: {err}'
3448 info.setdefault('__pending_error', msg)
3449 self.report_error(msg, is_error=False)
56d868db 3450 return info, info.pop('__files_to_move', None)
5bfa4862 3451
f46e2f9d 3452 def post_process(self, filename, info, files_to_move=None):
8222d8de 3453 """Run all the postprocessors on the given file."""
8222d8de 3454 info['filepath'] = filename
dcf64d43 3455 info['__files_to_move'] = files_to_move or {}
ed5835b4 3456 info = self.run_all_pps('post_process', info, additional_pps=info.get('__postprocessors'))
dcf64d43 3457 info = self.run_pp(MoveFilesAfterDownloadPP(self), info)
3458 del info['__files_to_move']
ed5835b4 3459 return self.run_all_pps('after_move', info)
c1c9a79c 3460
5db07df6 3461 def _make_archive_id(self, info_dict):
e9fef7ee
S
3462 video_id = info_dict.get('id')
3463 if not video_id:
3464 return
5db07df6
PH
3465 # Future-proof against any change in case
3466 # and backwards compatibility with prior versions
e9fef7ee 3467 extractor = info_dict.get('extractor_key') or info_dict.get('ie_key') # key in a playlist
7012b23c 3468 if extractor is None:
1211bb6d
S
3469 url = str_or_none(info_dict.get('url'))
3470 if not url:
3471 return
e9fef7ee 3472 # Try to find matching extractor for the URL and take its ie_key
8b7491c8 3473 for ie_key, ie in self._ies.items():
1211bb6d 3474 if ie.suitable(url):
8b7491c8 3475 extractor = ie_key
e9fef7ee
S
3476 break
3477 else:
3478 return
0647d925 3479 return make_archive_id(extractor, video_id)
5db07df6
PH
3480
3481 def in_download_archive(self, info_dict):
ae103564 3482 if not self.archive:
5db07df6
PH
3483 return False
3484
1e8fe57e 3485 vid_ids = [self._make_archive_id(info_dict)]
c200096c 3486 vid_ids.extend(info_dict.get('_old_archive_ids') or [])
1e8fe57e 3487 return any(id_ in self.archive for id_ in vid_ids)
c1c9a79c
PH
3488
3489 def record_download_archive(self, info_dict):
3490 fn = self.params.get('download_archive')
3491 if fn is None:
3492 return
5db07df6
PH
3493 vid_id = self._make_archive_id(info_dict)
3494 assert vid_id
ae103564 3495
a13e6848 3496 self.write_debug(f'Adding to archive: {vid_id}')
9c935fbc 3497 if is_path_like(fn):
ae103564 3498 with locked_file(fn, 'a', encoding='utf-8') as archive_file:
3499 archive_file.write(vid_id + '\n')
a45e8619 3500 self.archive.add(vid_id)
dd82ffea 3501
8c51aa65 3502 @staticmethod
8abeeb94 3503 def format_resolution(format, default='unknown'):
9359f3d4 3504 if format.get('vcodec') == 'none' and format.get('acodec') != 'none':
fb04e403 3505 return 'audio only'
f49d89ee
PH
3506 if format.get('resolution') is not None:
3507 return format['resolution']
35615307 3508 if format.get('width') and format.get('height'):
ff51ed58 3509 return '%dx%d' % (format['width'], format['height'])
35615307 3510 elif format.get('height'):
ff51ed58 3511 return '%sp' % format['height']
35615307 3512 elif format.get('width'):
ff51ed58 3513 return '%dx?' % format['width']
3514 return default
8c51aa65 3515
8130779d 3516 def _list_format_headers(self, *headers):
3517 if self.params.get('listformats_table', True) is not False:
591bb9d3 3518 return [self._format_out(header, self.Styles.HEADERS) for header in headers]
8130779d 3519 return headers
3520
c57f7757
PH
3521 def _format_note(self, fdict):
3522 res = ''
3523 if fdict.get('ext') in ['f4f', 'f4m']:
f304da8a 3524 res += '(unsupported)'
32f90364
PH
3525 if fdict.get('language'):
3526 if res:
3527 res += ' '
f304da8a 3528 res += '[%s]' % fdict['language']
c57f7757 3529 if fdict.get('format_note') is not None:
f304da8a 3530 if res:
3531 res += ' '
3532 res += fdict['format_note']
c57f7757 3533 if fdict.get('tbr') is not None:
f304da8a 3534 if res:
3535 res += ', '
3536 res += '%4dk' % fdict['tbr']
c57f7757
PH
3537 if fdict.get('container') is not None:
3538 if res:
3539 res += ', '
3540 res += '%s container' % fdict['container']
3089bc74
S
3541 if (fdict.get('vcodec') is not None
3542 and fdict.get('vcodec') != 'none'):
c57f7757
PH
3543 if res:
3544 res += ', '
3545 res += fdict['vcodec']
91c7271a 3546 if fdict.get('vbr') is not None:
c57f7757
PH
3547 res += '@'
3548 elif fdict.get('vbr') is not None and fdict.get('abr') is not None:
3549 res += 'video@'
3550 if fdict.get('vbr') is not None:
3551 res += '%4dk' % fdict['vbr']
fbb21cf5 3552 if fdict.get('fps') is not None:
5d583bdf
S
3553 if res:
3554 res += ', '
3555 res += '%sfps' % fdict['fps']
c57f7757
PH
3556 if fdict.get('acodec') is not None:
3557 if res:
3558 res += ', '
3559 if fdict['acodec'] == 'none':
3560 res += 'video only'
3561 else:
3562 res += '%-5s' % fdict['acodec']
3563 elif fdict.get('abr') is not None:
3564 if res:
3565 res += ', '
3566 res += 'audio'
3567 if fdict.get('abr') is not None:
3568 res += '@%3dk' % fdict['abr']
3569 if fdict.get('asr') is not None:
3570 res += ' (%5dHz)' % fdict['asr']
3571 if fdict.get('filesize') is not None:
3572 if res:
3573 res += ', '
3574 res += format_bytes(fdict['filesize'])
9732d77e
PH
3575 elif fdict.get('filesize_approx') is not None:
3576 if res:
3577 res += ', '
3578 res += '~' + format_bytes(fdict['filesize_approx'])
c57f7757 3579 return res
91c7271a 3580
aebb4f4b 3581 def _get_formats(self, info_dict):
3582 if info_dict.get('formats') is None:
3583 if info_dict.get('url') and info_dict.get('_type', 'video') == 'video':
3584 return [info_dict]
3585 return []
3586 return info_dict['formats']
b69fd25c 3587
aebb4f4b 3588 def render_formats_table(self, info_dict):
3589 formats = self._get_formats(info_dict)
3590 if not formats:
3591 return
8130779d 3592 if not self.params.get('listformats_table', True) is not False:
76d321f6 3593 table = [
3594 [
3595 format_field(f, 'format_id'),
3596 format_field(f, 'ext'),
3597 self.format_resolution(f),
8130779d 3598 self._format_note(f)
d5d1df8a 3599 ] for f in formats if (f.get('preference') or 0) >= -1000]
8130779d 3600 return render_table(['format code', 'extension', 'resolution', 'note'], table, extra_gap=1)
3601
d816f61f 3602 def simplified_codec(f, field):
3603 assert field in ('acodec', 'vcodec')
3604 codec = f.get(field, 'unknown')
f5ea4748 3605 if not codec:
3606 return 'unknown'
3607 elif codec != 'none':
d816f61f 3608 return '.'.join(codec.split('.')[:4])
3609
3610 if field == 'vcodec' and f.get('acodec') == 'none':
3611 return 'images'
3612 elif field == 'acodec' and f.get('vcodec') == 'none':
3613 return ''
3614 return self._format_out('audio only' if field == 'vcodec' else 'video only',
3615 self.Styles.SUPPRESS)
3616
591bb9d3 3617 delim = self._format_out('\u2502', self.Styles.DELIM, '|', test_encoding=True)
8130779d 3618 table = [
3619 [
591bb9d3 3620 self._format_out(format_field(f, 'format_id'), self.Styles.ID),
8130779d 3621 format_field(f, 'ext'),
3622 format_field(f, func=self.format_resolution, ignore=('audio only', 'images')),
563e0bf8 3623 format_field(f, 'fps', '\t%d', func=round),
8130779d 3624 format_field(f, 'dynamic_range', '%s', ignore=(None, 'SDR')).replace('HDR', ''),
b8ed0f15 3625 format_field(f, 'audio_channels', '\t%s'),
8130779d 3626 delim,
3627 format_field(f, 'filesize', ' \t%s', func=format_bytes) + format_field(f, 'filesize_approx', '~\t%s', func=format_bytes),
563e0bf8 3628 format_field(f, 'tbr', '\t%dk', func=round),
8130779d 3629 shorten_protocol_name(f.get('protocol', '')),
3630 delim,
d816f61f 3631 simplified_codec(f, 'vcodec'),
563e0bf8 3632 format_field(f, 'vbr', '\t%dk', func=round),
d816f61f 3633 simplified_codec(f, 'acodec'),
563e0bf8 3634 format_field(f, 'abr', '\t%dk', func=round),
ae61d108 3635 format_field(f, 'asr', '\t%s', func=format_decimal_suffix),
8130779d 3636 join_nonempty(
591bb9d3 3637 self._format_out('UNSUPPORTED', 'light red') if f.get('ext') in ('f4f', 'f4m') else None,
8130779d 3638 format_field(f, 'language', '[%s]'),
3639 join_nonempty(format_field(f, 'format_note'),
3640 format_field(f, 'container', ignore=(None, f.get('ext'))),
3641 delim=', '),
3642 delim=' '),
3643 ] for f in formats if f.get('preference') is None or f['preference'] >= -1000]
3644 header_line = self._list_format_headers(
b8ed0f15 3645 'ID', 'EXT', 'RESOLUTION', '\tFPS', 'HDR', 'CH', delim, '\tFILESIZE', '\tTBR', 'PROTO',
8130779d 3646 delim, 'VCODEC', '\tVBR', 'ACODEC', '\tABR', '\tASR', 'MORE INFO')
3647
3648 return render_table(
3649 header_line, table, hide_empty=True,
591bb9d3 3650 delim=self._format_out('\u2500', self.Styles.DELIM, '-', test_encoding=True))
8130779d 3651
3652 def render_thumbnails_table(self, info_dict):
88f23a18 3653 thumbnails = list(info_dict.get('thumbnails') or [])
cfb56d1a 3654 if not thumbnails:
8130779d 3655 return None
3656 return render_table(
ec11a9f4 3657 self._list_format_headers('ID', 'Width', 'Height', 'URL'),
177662e0 3658 [[t.get('id'), t.get('width') or 'unknown', t.get('height') or 'unknown', t['url']] for t in thumbnails])
2412044c 3659
8130779d 3660 def render_subtitles_table(self, video_id, subtitles):
2412044c 3661 def _row(lang, formats):
49c258e1 3662 exts, names = zip(*((f['ext'], f.get('name') or 'unknown') for f in reversed(formats)))
2412044c 3663 if len(set(names)) == 1:
7aee40c1 3664 names = [] if names[0] == 'unknown' else names[:1]
2412044c 3665 return [lang, ', '.join(names), ', '.join(exts)]
3666
8130779d 3667 if not subtitles:
3668 return None
3669 return render_table(
ec11a9f4 3670 self._list_format_headers('Language', 'Name', 'Formats'),
2412044c 3671 [_row(lang, formats) for lang, formats in subtitles.items()],
8130779d 3672 hide_empty=True)
3673
3674 def __list_table(self, video_id, name, func, *args):
3675 table = func(*args)
3676 if not table:
3677 self.to_screen(f'{video_id} has no {name}')
3678 return
3679 self.to_screen(f'[info] Available {name} for {video_id}:')
3680 self.to_stdout(table)
3681
3682 def list_formats(self, info_dict):
3683 self.__list_table(info_dict['id'], 'formats', self.render_formats_table, info_dict)
3684
3685 def list_thumbnails(self, info_dict):
3686 self.__list_table(info_dict['id'], 'thumbnails', self.render_thumbnails_table, info_dict)
3687
3688 def list_subtitles(self, video_id, subtitles, name='subtitles'):
3689 self.__list_table(video_id, name, self.render_subtitles_table, video_id, subtitles)
a504ced0 3690
dca08720
PH
3691 def urlopen(self, req):
3692 """ Start an HTTP download """
f9934b96 3693 if isinstance(req, str):
67dda517 3694 req = sanitized_Request(req)
19a41fc6 3695 return self._opener.open(req, timeout=self._socket_timeout)
dca08720
PH
3696
3697 def print_debug_header(self):
3698 if not self.params.get('verbose'):
3699 return
49a57e70 3700
a057779d 3701 from . import _IN_CLI # Must be delayed import
3702
560738f3 3703 # These imports can be slow. So import them only as needed
3704 from .extractor.extractors import _LAZY_LOADER
3705 from .extractor.extractors import _PLUGIN_CLASSES as plugin_extractors
3706
49a57e70 3707 def get_encoding(stream):
2a938746 3708 ret = str(getattr(stream, 'encoding', 'missing (%s)' % type(stream).__name__))
49a57e70 3709 if not supports_terminal_sequences(stream):
53973b4d 3710 from .utils import WINDOWS_VT_MODE # Must be imported locally
e3c7d495 3711 ret += ' (No VT)' if WINDOWS_VT_MODE is False else ' (No ANSI)'
49a57e70 3712 return ret
3713
591bb9d3 3714 encoding_str = 'Encodings: locale %s, fs %s, pref %s, %s' % (
49a57e70 3715 locale.getpreferredencoding(),
3716 sys.getfilesystemencoding(),
591bb9d3 3717 self.get_encoding(),
3718 ', '.join(
64fa820c 3719 f'{key} {get_encoding(stream)}' for key, stream in self._out_files.items_
591bb9d3 3720 if stream is not None and key != 'console')
3721 )
883d4b1e 3722
3723 logger = self.params.get('logger')
3724 if logger:
3725 write_debug = lambda msg: logger.debug(f'[debug] {msg}')
3726 write_debug(encoding_str)
3727 else:
96565c7e 3728 write_string(f'[debug] {encoding_str}\n', encoding=None)
49a57e70 3729 write_debug = lambda msg: self._write_string(f'[debug] {msg}\n')
734f90bb 3730
4c88ff87 3731 source = detect_variant()
70b23409 3732 if VARIANT not in (None, 'pip'):
3733 source += '*'
36eaf303 3734 write_debug(join_nonempty(
b5e7a2e6 3735 f'{"yt-dlp" if REPOSITORY == "yt-dlp/yt-dlp" else REPOSITORY} version',
3736 __version__,
36eaf303 3737 f'[{RELEASE_GIT_HEAD}]' if RELEASE_GIT_HEAD else '',
3738 '' if source == 'unknown' else f'({source})',
a057779d 3739 '' if _IN_CLI else 'API',
36eaf303 3740 delim=' '))
497074f0 3741
3742 if not _IN_CLI:
3743 write_debug(f'params: {self.params}')
3744
6e21fdd2 3745 if not _LAZY_LOADER:
3746 if os.environ.get('YTDLP_NO_LAZY_EXTRACTORS'):
49a57e70 3747 write_debug('Lazy loading extractors is forcibly disabled')
6e21fdd2 3748 else:
49a57e70 3749 write_debug('Lazy loading extractors is disabled')
3ae5e797 3750 if plugin_extractors or plugin_postprocessors:
49a57e70 3751 write_debug('Plugins: %s' % [
3ae5e797 3752 '%s%s' % (klass.__name__, '' if klass.__name__ == name else f' as {name}')
3753 for name, klass in itertools.chain(plugin_extractors.items(), plugin_postprocessors.items())])
8a82af35 3754 if self.params['compat_opts']:
3755 write_debug('Compatibility options: %s' % ', '.join(self.params['compat_opts']))
36eaf303 3756
b5e7a2e6 3757 if current_git_head():
3758 write_debug(f'Git HEAD: {current_git_head()}')
b1f94422 3759 write_debug(system_identifier())
d28b5171 3760
8913ef74 3761 exe_versions, ffmpeg_features = FFmpegPostProcessor.get_versions_and_features(self)
3762 ffmpeg_features = {key for key, val in ffmpeg_features.items() if val}
3763 if ffmpeg_features:
19a03940 3764 exe_versions['ffmpeg'] += ' (%s)' % ','.join(sorted(ffmpeg_features))
8913ef74 3765
4c83c967 3766 exe_versions['rtmpdump'] = rtmpdump_version()
feee8d32 3767 exe_versions['phantomjs'] = PhantomJSwrapper._version()
d28b5171 3768 exe_str = ', '.join(
2831b468 3769 f'{exe} {v}' for exe, v in sorted(exe_versions.items()) if v
3770 ) or 'none'
49a57e70 3771 write_debug('exe versions: %s' % exe_str)
dca08720 3772
1d485a1a 3773 from .compat.compat_utils import get_package_info
9b8ee23b 3774 from .dependencies import available_dependencies
3775
3776 write_debug('Optional libraries: %s' % (', '.join(sorted({
1d485a1a 3777 join_nonempty(*get_package_info(m)) for m in available_dependencies.values()
9b8ee23b 3778 })) or 'none'))
2831b468 3779
97ec5bc5 3780 self._setup_opener()
dca08720
PH
3781 proxy_map = {}
3782 for handler in self._opener.handlers:
3783 if hasattr(handler, 'proxies'):
3784 proxy_map.update(handler.proxies)
49a57e70 3785 write_debug(f'Proxy map: {proxy_map}')
dca08720 3786
49a57e70 3787 # Not implemented
3788 if False and self.params.get('call_home'):
0f06bcd7 3789 ipaddr = self.urlopen('https://yt-dl.org/ip').read().decode()
49a57e70 3790 write_debug('Public IP address: %s' % ipaddr)
58b1f00d 3791 latest_version = self.urlopen(
0f06bcd7 3792 'https://yt-dl.org/latest/version').read().decode()
58b1f00d
PH
3793 if version_tuple(latest_version) > version_tuple(__version__):
3794 self.report_warning(
3795 'You are using an outdated version (newest version: %s)! '
3796 'See https://yt-dl.org/update if you need help updating.' %
3797 latest_version)
3798
e344693b 3799 def _setup_opener(self):
97ec5bc5 3800 if hasattr(self, '_opener'):
3801 return
6ad14cab 3802 timeout_val = self.params.get('socket_timeout')
17bddf3e 3803 self._socket_timeout = 20 if timeout_val is None else float(timeout_val)
6ad14cab 3804
982ee69a 3805 opts_cookiesfrombrowser = self.params.get('cookiesfrombrowser')
dca08720
PH
3806 opts_cookiefile = self.params.get('cookiefile')
3807 opts_proxy = self.params.get('proxy')
3808
982ee69a 3809 self.cookiejar = load_cookies(opts_cookiefile, opts_cookiesfrombrowser, self)
dca08720 3810
6a3f4c3f 3811 cookie_processor = YoutubeDLCookieProcessor(self.cookiejar)
dca08720
PH
3812 if opts_proxy is not None:
3813 if opts_proxy == '':
3814 proxies = {}
3815 else:
3816 proxies = {'http': opts_proxy, 'https': opts_proxy}
3817 else:
ac668111 3818 proxies = urllib.request.getproxies()
067aa17e 3819 # Set HTTPS proxy to HTTP one if given (https://github.com/ytdl-org/youtube-dl/issues/805)
dca08720
PH
3820 if 'http' in proxies and 'https' not in proxies:
3821 proxies['https'] = proxies['http']
91410c9b 3822 proxy_handler = PerRequestProxyHandler(proxies)
a0ddb8a2
PH
3823
3824 debuglevel = 1 if self.params.get('debug_printtraffic') else 0
be4a824d
PH
3825 https_handler = make_HTTPS_handler(self.params, debuglevel=debuglevel)
3826 ydlh = YoutubeDLHandler(self.params, debuglevel=debuglevel)
fca6dba8 3827 redirect_handler = YoutubeDLRedirectHandler()
f9934b96 3828 data_handler = urllib.request.DataHandler()
6240b0a2
JMF
3829
3830 # When passing our own FileHandler instance, build_opener won't add the
3831 # default FileHandler and allows us to disable the file protocol, which
3832 # can be used for malicious purposes (see
067aa17e 3833 # https://github.com/ytdl-org/youtube-dl/issues/8227)
ac668111 3834 file_handler = urllib.request.FileHandler()
6240b0a2
JMF
3835
3836 def file_open(*args, **kwargs):
ac668111 3837 raise urllib.error.URLError('file:// scheme is explicitly disabled in yt-dlp for security reasons')
6240b0a2
JMF
3838 file_handler.file_open = file_open
3839
ac668111 3840 opener = urllib.request.build_opener(
fca6dba8 3841 proxy_handler, https_handler, cookie_processor, ydlh, redirect_handler, data_handler, file_handler)
2461f79d 3842
dca08720
PH
3843 # Delete the default user-agent header, which would otherwise apply in
3844 # cases where our custom HTTP handler doesn't come into play
067aa17e 3845 # (See https://github.com/ytdl-org/youtube-dl/issues/1309 for details)
dca08720
PH
3846 opener.addheaders = []
3847 self._opener = opener
62fec3b2
PH
3848
3849 def encode(self, s):
3850 if isinstance(s, bytes):
3851 return s # Already encoded
3852
3853 try:
3854 return s.encode(self.get_encoding())
3855 except UnicodeEncodeError as err:
3856 err.reason = err.reason + '. Check your system encoding configuration or use the --encoding option.'
3857 raise
3858
3859 def get_encoding(self):
3860 encoding = self.params.get('encoding')
3861 if encoding is None:
3862 encoding = preferredencoding()
3863 return encoding
ec82d85a 3864
e08a85d8 3865 def _write_info_json(self, label, ie_result, infofn, overwrite=None):
cb96c5be 3866 ''' Write infojson and returns True = written, 'exists' = Already exists, False = skip, None = error '''
e08a85d8 3867 if overwrite is None:
3868 overwrite = self.params.get('overwrites', True)
80c03fa9 3869 if not self.params.get('writeinfojson'):
3870 return False
3871 elif not infofn:
3872 self.write_debug(f'Skipping writing {label} infojson')
3873 return False
3874 elif not self._ensure_dir_exists(infofn):
3875 return None
e08a85d8 3876 elif not overwrite and os.path.exists(infofn):
80c03fa9 3877 self.to_screen(f'[info] {label.title()} metadata is already present')
cb96c5be 3878 return 'exists'
3879
3880 self.to_screen(f'[info] Writing {label} metadata as JSON to: {infofn}')
3881 try:
3882 write_json_file(self.sanitize_info(ie_result, self.params.get('clean_infojson', True)), infofn)
3883 return True
86e5f3ed 3884 except OSError:
cb96c5be 3885 self.report_error(f'Cannot write {label} metadata to JSON file {infofn}')
3886 return None
80c03fa9 3887
3888 def _write_description(self, label, ie_result, descfn):
3889 ''' Write description and returns True = written, False = skip, None = error '''
3890 if not self.params.get('writedescription'):
3891 return False
3892 elif not descfn:
3893 self.write_debug(f'Skipping writing {label} description')
3894 return False
3895 elif not self._ensure_dir_exists(descfn):
3896 return None
3897 elif not self.params.get('overwrites', True) and os.path.exists(descfn):
3898 self.to_screen(f'[info] {label.title()} description is already present')
3899 elif ie_result.get('description') is None:
3900 self.report_warning(f'There\'s no {label} description to write')
3901 return False
3902 else:
3903 try:
3904 self.to_screen(f'[info] Writing {label} description to: {descfn}')
86e5f3ed 3905 with open(encodeFilename(descfn), 'w', encoding='utf-8') as descfile:
80c03fa9 3906 descfile.write(ie_result['description'])
86e5f3ed 3907 except OSError:
80c03fa9 3908 self.report_error(f'Cannot write {label} description file {descfn}')
3909 return None
3910 return True
3911
3912 def _write_subtitles(self, info_dict, filename):
3913 ''' Write subtitles to file and return list of (sub_filename, final_sub_filename); or None if error'''
3914 ret = []
3915 subtitles = info_dict.get('requested_subtitles')
3916 if not subtitles or not (self.params.get('writesubtitles') or self.params.get('writeautomaticsub')):
3917 # subtitles download errors are already managed as troubles in relevant IE
3918 # that way it will silently go on when used with unsupporting IE
3919 return ret
3920
3921 sub_filename_base = self.prepare_filename(info_dict, 'subtitle')
3922 if not sub_filename_base:
3923 self.to_screen('[info] Skipping writing video subtitles')
3924 return ret
3925 for sub_lang, sub_info in subtitles.items():
3926 sub_format = sub_info['ext']
3927 sub_filename = subtitles_filename(filename, sub_lang, sub_format, info_dict.get('ext'))
3928 sub_filename_final = subtitles_filename(sub_filename_base, sub_lang, sub_format, info_dict.get('ext'))
e04938ab 3929 existing_sub = self.existing_file((sub_filename_final, sub_filename))
3930 if existing_sub:
80c03fa9 3931 self.to_screen(f'[info] Video subtitle {sub_lang}.{sub_format} is already present')
e04938ab 3932 sub_info['filepath'] = existing_sub
3933 ret.append((existing_sub, sub_filename_final))
80c03fa9 3934 continue
3935
3936 self.to_screen(f'[info] Writing video subtitles to: {sub_filename}')
3937 if sub_info.get('data') is not None:
3938 try:
3939 # Use newline='' to prevent conversion of newline characters
3940 # See https://github.com/ytdl-org/youtube-dl/issues/10268
86e5f3ed 3941 with open(sub_filename, 'w', encoding='utf-8', newline='') as subfile:
80c03fa9 3942 subfile.write(sub_info['data'])
3943 sub_info['filepath'] = sub_filename
3944 ret.append((sub_filename, sub_filename_final))
3945 continue
86e5f3ed 3946 except OSError:
80c03fa9 3947 self.report_error(f'Cannot write video subtitles file {sub_filename}')
3948 return None
3949
3950 try:
3951 sub_copy = sub_info.copy()
3952 sub_copy.setdefault('http_headers', info_dict.get('http_headers'))
3953 self.dl(sub_filename, sub_copy, subtitle=True)
3954 sub_info['filepath'] = sub_filename
3955 ret.append((sub_filename, sub_filename_final))
6020e05d 3956 except (DownloadError, ExtractorError, IOError, OSError, ValueError) + network_exceptions as err:
c70c418d 3957 msg = f'Unable to download video subtitles for {sub_lang!r}: {err}'
6020e05d 3958 if self.params.get('ignoreerrors') is not True: # False or 'only_download'
c70c418d 3959 if not self.params.get('ignoreerrors'):
3960 self.report_error(msg)
3961 raise DownloadError(msg)
3962 self.report_warning(msg)
519804a9 3963 return ret
80c03fa9 3964
3965 def _write_thumbnails(self, label, info_dict, filename, thumb_filename_base=None):
3966 ''' Write thumbnails to file and return list of (thumb_filename, final_thumb_filename) '''
6c4fd172 3967 write_all = self.params.get('write_all_thumbnails', False)
80c03fa9 3968 thumbnails, ret = [], []
6c4fd172 3969 if write_all or self.params.get('writethumbnail', False):
0202b52a 3970 thumbnails = info_dict.get('thumbnails') or []
6c4fd172 3971 multiple = write_all and len(thumbnails) > 1
ec82d85a 3972
80c03fa9 3973 if thumb_filename_base is None:
3974 thumb_filename_base = filename
3975 if thumbnails and not thumb_filename_base:
3976 self.write_debug(f'Skipping writing {label} thumbnail')
3977 return ret
3978
dd0228ce 3979 for idx, t in list(enumerate(thumbnails))[::-1]:
80c03fa9 3980 thumb_ext = (f'{t["id"]}.' if multiple else '') + determine_ext(t['url'], 'jpg')
aa9369a2 3981 thumb_display_id = f'{label} thumbnail {t["id"]}'
80c03fa9 3982 thumb_filename = replace_extension(filename, thumb_ext, info_dict.get('ext'))
3983 thumb_filename_final = replace_extension(thumb_filename_base, thumb_ext, info_dict.get('ext'))
ec82d85a 3984
e04938ab 3985 existing_thumb = self.existing_file((thumb_filename_final, thumb_filename))
3986 if existing_thumb:
aa9369a2 3987 self.to_screen('[info] %s is already present' % (
3988 thumb_display_id if multiple else f'{label} thumbnail').capitalize())
e04938ab 3989 t['filepath'] = existing_thumb
3990 ret.append((existing_thumb, thumb_filename_final))
ec82d85a 3991 else:
80c03fa9 3992 self.to_screen(f'[info] Downloading {thumb_display_id} ...')
ec82d85a 3993 try:
297e9952 3994 uf = self.urlopen(sanitized_Request(t['url'], headers=t.get('http_headers', {})))
80c03fa9 3995 self.to_screen(f'[info] Writing {thumb_display_id} to: {thumb_filename}')
d3d89c32 3996 with open(encodeFilename(thumb_filename), 'wb') as thumbf:
ec82d85a 3997 shutil.copyfileobj(uf, thumbf)
80c03fa9 3998 ret.append((thumb_filename, thumb_filename_final))
885cc0b7 3999 t['filepath'] = thumb_filename
3158150c 4000 except network_exceptions as err:
dd0228ce 4001 thumbnails.pop(idx)
80c03fa9 4002 self.report_warning(f'Unable to download {thumb_display_id}: {err}')
6c4fd172 4003 if ret and not write_all:
4004 break
0202b52a 4005 return ret