]> jfr.im git - yt-dlp.git/blame - yt_dlp/YoutubeDL.py
Fix bugs in 3bec830a597e8c7ab0d9f4e1258dc4a1be0b1de4
[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
32from .extractor.openload import PhantomJSwrapper
33from .minicurses import format_text
34from .postprocessor import _PLUGIN_CLASSES as plugin_postprocessors
35from .postprocessor import (
36 EmbedThumbnailPP,
37 FFmpegFixupDuplicateMoovPP,
38 FFmpegFixupDurationPP,
39 FFmpegFixupM3u8PP,
40 FFmpegFixupM4aPP,
41 FFmpegFixupStretchedPP,
42 FFmpegFixupTimestampPP,
43 FFmpegMergerPP,
44 FFmpegPostProcessor,
ca9def71 45 FFmpegVideoConvertorPP,
f8271158 46 MoveFilesAfterDownloadPP,
47 get_postprocessor,
48)
ca9def71 49from .postprocessor.ffmpeg import resolve_mapping as resolve_recode_mapping
f8271158 50from .update import detect_variant
8c25f81b 51from .utils import (
f8271158 52 DEFAULT_OUTTMPL,
7b2c3f47 53 IDENTITY,
f8271158 54 LINK_TEMPLATES,
55 NO_DEFAULT,
1d485a1a 56 NUMBER_RE,
f8271158 57 OUTTMPL_TYPES,
58 POSTPROCESS_WHEN,
59 STR_FORMAT_RE_TMPL,
60 STR_FORMAT_TYPES,
61 ContentTooShortError,
62 DateRange,
63 DownloadCancelled,
64 DownloadError,
65 EntryNotInPlaylist,
66 ExistingVideoReached,
67 ExtractorError,
68 GeoRestrictedError,
69 HEADRequest,
f8271158 70 ISO3166Utils,
71 LazyList,
72 MaxDownloadsReached,
19a03940 73 Namespace,
f8271158 74 PagedList,
75 PerRequestProxyHandler,
7e88d7d7 76 PlaylistEntries,
f8271158 77 Popen,
78 PostProcessingError,
79 ReExtractInfo,
80 RejectedVideoReached,
81 SameFileError,
82 UnavailableVideoError,
693f0600 83 UserNotLive,
f8271158 84 YoutubeDLCookieProcessor,
85 YoutubeDLHandler,
86 YoutubeDLRedirectHandler,
eedb7ba5
S
87 age_restricted,
88 args_to_str,
cb794ee0 89 bug_reports_message,
ce02ed60 90 date_from_str,
ce02ed60 91 determine_ext,
b5559424 92 determine_protocol,
c0384f22 93 encode_compat_str,
ce02ed60 94 encodeFilename,
a06916d9 95 error_to_compat_str,
47cdc68e 96 escapeHTML,
590bc6f6 97 expand_path,
90137ca4 98 filter_dict,
e29663c6 99 float_or_none,
02dbf93f 100 format_bytes,
e0fd9573 101 format_decimal_suffix,
f8271158 102 format_field,
525ef922 103 formatSeconds,
0bb322b9 104 get_domain,
c9969434 105 int_or_none,
732044af 106 iri_to_uri,
34921b43 107 join_nonempty,
ce02ed60 108 locked_file,
0202b52a 109 make_dir,
dca08720 110 make_HTTPS_handler,
8b7539d2 111 merge_headers,
3158150c 112 network_exceptions,
ec11a9f4 113 number_of_digits,
cd6fc19e 114 orderedSet,
083c9df9 115 parse_filesize,
ce02ed60 116 preferredencoding,
eedb7ba5 117 prepend_extension,
51fb4995 118 register_socks_protocols,
3efb96a6 119 remove_terminal_sequences,
cfb56d1a 120 render_table,
eedb7ba5 121 replace_extension,
ce02ed60 122 sanitize_filename,
1bb5c511 123 sanitize_path,
dcf77cf1 124 sanitize_url,
67dda517 125 sanitized_Request,
e5660ee6 126 std_headers,
1211bb6d 127 str_or_none,
e29663c6 128 strftime_or_none,
ce02ed60 129 subtitles_filename,
819e0531 130 supports_terminal_sequences,
b1f94422 131 system_identifier,
f2ebc5c7 132 timetuple_from_msec,
732044af 133 to_high_limit_path,
324ad820 134 traverse_obj,
6033d980 135 try_get,
29eb5174 136 url_basename,
7d1eb38a 137 variadic,
58b1f00d 138 version_tuple,
53973b4d 139 windows_enable_vt_mode,
ce02ed60
PH
140 write_json_file,
141 write_string,
4f026faf 142)
f8271158 143from .version import RELEASE_GIT_HEAD, __version__
8222d8de 144
e9c0cdd3
YCH
145if compat_os_name == 'nt':
146 import ctypes
147
2459b6e1 148
86e5f3ed 149class YoutubeDL:
8222d8de
JMF
150 """YoutubeDL class.
151
152 YoutubeDL objects are the ones responsible of downloading the
153 actual video file and writing it to disk if the user has requested
154 it, among some other tasks. In most cases there should be one per
155 program. As, given a video URL, the downloader doesn't know how to
156 extract all the needed information, task that InfoExtractors do, it
157 has to pass the URL to one of them.
158
159 For this, YoutubeDL objects have a method that allows
160 InfoExtractors to be registered in a given order. When it is passed
161 a URL, the YoutubeDL object handles it to the first InfoExtractor it
162 finds that reports being able to handle it. The InfoExtractor extracts
163 all the information about the video or videos the URL refers to, and
164 YoutubeDL process the extracted information, possibly using a File
165 Downloader to download the video.
166
167 YoutubeDL objects accept a lot of parameters. In order not to saturate
168 the object constructor with arguments, it receives a dictionary of
169 options instead. These options are available through the params
170 attribute for the InfoExtractors to use. The YoutubeDL also
171 registers itself as the downloader in charge for the InfoExtractors
172 that are added to it, so this is a "mutual registration".
173
174 Available options:
175
176 username: Username for authentication purposes.
177 password: Password for authentication purposes.
180940e0 178 videopassword: Password for accessing a video.
1da50aa3
S
179 ap_mso: Adobe Pass multiple-system operator identifier.
180 ap_username: Multiple-system operator account username.
181 ap_password: Multiple-system operator account password.
8222d8de
JMF
182 usenetrc: Use netrc for authentication instead.
183 verbose: Print additional info to stdout.
184 quiet: Do not print messages to stdout.
ad8915b7 185 no_warnings: Do not print out anything for warnings.
bb66c247 186 forceprint: A dict with keys WHEN mapped to a list of templates to
187 print to stdout. The allowed keys are video or any of the
188 items in utils.POSTPROCESS_WHEN.
ca30f449 189 For compatibility, a single list is also accepted
bb66c247 190 print_to_file: A dict with keys WHEN (same as forceprint) mapped to
191 a list of tuples with (template, filename)
8694c600 192 forcejson: Force printing info_dict as JSON.
63e0be34
PH
193 dump_single_json: Force printing the info_dict of the whole playlist
194 (or video) as a single JSON line.
c25228e5 195 force_write_download_archive: Force writing download archive regardless
196 of 'skip_download' or 'simulate'.
b7b04c78 197 simulate: Do not download the video files. If unset (or None),
198 simulate only if listsubtitles, listformats or list_thumbnails is used
eb8a4433 199 format: Video format code. see "FORMAT SELECTION" for more details.
093a1710 200 You can also pass a function. The function takes 'ctx' as
201 argument and returns the formats to download.
202 See "build_format_selector" for an implementation
63ad4d43 203 allow_unplayable_formats: Allow unplayable formats to be extracted and downloaded.
b7da73eb 204 ignore_no_formats_error: Ignore "No video formats" error. Usefull for
205 extracting metadata even if the video is not actually
206 available for download (experimental)
0930b11f 207 format_sort: A list of fields by which to sort the video formats.
208 See "Sorting Formats" for more details.
c25228e5 209 format_sort_force: Force the given format_sort. see "Sorting Formats"
210 for more details.
08d30158 211 prefer_free_formats: Whether to prefer video formats with free containers
212 over non-free ones of same quality.
c25228e5 213 allow_multiple_video_streams: Allow multiple video streams to be merged
214 into a single file
215 allow_multiple_audio_streams: Allow multiple audio streams to be merged
216 into a single file
0ba692ac 217 check_formats Whether to test if the formats are downloadable.
9f1a1c36 218 Can be True (check all), False (check none),
219 'selected' (check selected formats),
0ba692ac 220 or None (check only if requested by extractor)
4524baf0 221 paths: Dictionary of output paths. The allowed keys are 'home'
222 'temp' and the keys of OUTTMPL_TYPES (in utils.py)
de6000d9 223 outtmpl: Dictionary of templates for output names. Allowed keys
4524baf0 224 are 'default' and the keys of OUTTMPL_TYPES (in utils.py).
34488702 225 For compatibility with youtube-dl, a single string can also be used
a820dc72
RA
226 outtmpl_na_placeholder: Placeholder for unavailable meta fields.
227 restrictfilenames: Do not allow "&" and spaces in file names
228 trim_file_name: Limit length of filename (extension excluded)
4524baf0 229 windowsfilenames: Force the filenames to be windows compatible
b1940459 230 ignoreerrors: Do not stop on download/postprocessing errors.
231 Can be 'only_download' to ignore only download errors.
232 Default is 'only_download' for CLI, but False for API
26e2805c 233 skip_playlist_after_errors: Number of allowed failures until the rest of
234 the playlist is skipped
d22dec74 235 force_generic_extractor: Force downloader to use the generic extractor
0c3d0f51 236 overwrites: Overwrite all video and metadata files if True,
237 overwrite only non-video files if None
238 and don't overwrite any file if False
34488702 239 For compatibility with youtube-dl,
240 "nooverwrites" may also be used instead
c14e88f0 241 playlist_items: Specific indices of playlist to download.
75822ca7 242 playlistrandom: Download playlist items in random order.
7e9a6125 243 lazy_playlist: Process playlist entries as they are received.
8222d8de
JMF
244 matchtitle: Download only matching titles.
245 rejecttitle: Reject downloads for matching titles.
8bf9319e 246 logger: Log messages to a logging.Logger instance.
8222d8de 247 logtostderr: Log messages to stderr instead of stdout.
819e0531 248 consoletitle: Display progress in console window's titlebar.
8222d8de
JMF
249 writedescription: Write the video description to a .description file
250 writeinfojson: Write the video description to a .info.json file
75d43ca0 251 clean_infojson: Remove private fields from the infojson
34488702 252 getcomments: Extract video comments. This will not be written to disk
06167fbb 253 unless writeinfojson is also given
1fb07d10 254 writeannotations: Write the video annotations to a .annotations.xml file
8222d8de 255 writethumbnail: Write the thumbnail image to a file
c25228e5 256 allow_playlist_files: Whether to write playlists' description, infojson etc
257 also to disk when using the 'write*' options
ec82d85a 258 write_all_thumbnails: Write all thumbnail formats to files
732044af 259 writelink: Write an internet shortcut file, depending on the
260 current platform (.url/.webloc/.desktop)
261 writeurllink: Write a Windows internet shortcut file (.url)
262 writewebloclink: Write a macOS internet shortcut file (.webloc)
263 writedesktoplink: Write a Linux internet shortcut file (.desktop)
8222d8de 264 writesubtitles: Write the video subtitles to a file
741dd8ea 265 writeautomaticsub: Write the automatically generated subtitles to a file
8222d8de 266 listsubtitles: Lists all available subtitles for the video
a504ced0 267 subtitlesformat: The format code for subtitles
c32b0aab 268 subtitleslangs: List of languages of the subtitles to download (can be regex).
269 The list may contain "all" to refer to all the available
270 subtitles. The language can be prefixed with a "-" to
271 exclude it from the requested languages. Eg: ['all', '-live_chat']
8222d8de
JMF
272 keepvideo: Keep the video file after post-processing
273 daterange: A DateRange object, download only if the upload_date is in the range.
274 skip_download: Skip the actual download of the video file
c35f9e72 275 cachedir: Location of the cache files in the filesystem.
a0e07d31 276 False to disable filesystem cache.
47192f92 277 noplaylist: Download single video instead of a playlist if in doubt.
8dbe9899
PH
278 age_limit: An integer representing the user's age in years.
279 Unsuitable videos for the given age are skipped.
5fe18bdb
PH
280 min_views: An integer representing the minimum view count the video
281 must have in order to not be skipped.
282 Videos without view count information are always
283 downloaded. None for no limit.
284 max_views: An integer representing the maximum view count.
285 Videos that are more popular than that are not
286 downloaded.
287 Videos without view count information are always
288 downloaded. None for no limit.
289 download_archive: File name of a file where all downloads are recorded.
c1c9a79c
PH
290 Videos already present in the file are not downloaded
291 again.
8a51f564 292 break_on_existing: Stop the download process after attempting to download a
293 file that is in the archive.
294 break_on_reject: Stop the download process when encountering a video that
295 has been filtered out.
b222c271 296 break_per_url: Whether break_on_reject and break_on_existing
297 should act on each input URL as opposed to for the entire queue
d76fa1f3 298 cookiefile: File name or text stream from where cookies should be read and dumped to
f59f5ef8
MB
299 cookiesfrombrowser: A tuple containing the name of the browser, the profile
300 name/pathfrom where cookies are loaded, and the name of the
301 keyring. Eg: ('chrome', ) or ('vivaldi', 'default', 'BASICTEXT')
f81c62a6 302 legacyserverconnect: Explicitly allow HTTPS connection to servers that do not
303 support RFC 5746 secure renegotiation
f59f5ef8 304 nocheckcertificate: Do not verify SSL certificates
bb58c9ed 305 client_certificate: Path to client certificate file in PEM format. May include the private key
306 client_certificate_key: Path to private key file for client certificate
307 client_certificate_password: Password for client certificate private key, if encrypted.
308 If not provided and the key is encrypted, yt-dlp will ask interactively
7e8c0af0 309 prefer_insecure: Use HTTP instead of HTTPS to retrieve information.
c6e07cf1 310 (Only supported by some extractors)
8b7539d2 311 http_headers: A dictionary of custom headers to be used for all requests
a1ee09e8 312 proxy: URL of the proxy server to use
38cce791 313 geo_verification_proxy: URL of the proxy to use for IP address verification
504f20dd 314 on geo-restricted sites.
e344693b 315 socket_timeout: Time to wait for unresponsive hosts, in seconds
0783b09b
PH
316 bidi_workaround: Work around buggy terminals without bidirectional text
317 support, using fridibi
a0ddb8a2 318 debug_printtraffic:Print out sent and received HTTP traffic
04b4d394
PH
319 default_search: Prepend this string if an input url is not valid.
320 'auto' for elaborate guessing
62fec3b2 321 encoding: Use this encoding instead of the system-specified.
134c913c 322 extract_flat: Whether to resolve and process url_results further
323 * False: Always process (default)
324 * True: Never process
325 * 'in_playlist': Do not process inside playlist/multi_video
326 * 'discard': Always process, but don't return the result
327 from inside playlist/multi_video
328 * 'discard_in_playlist': Same as "discard", but only for
329 playlists (not multi_video)
f2ebc5c7 330 wait_for_video: If given, wait for scheduled streams to become available.
331 The value should be a tuple containing the range
332 (min_secs, max_secs) to wait between retries
4f026faf 333 postprocessors: A list of dictionaries, each with an entry
71b640cc 334 * key: The name of the postprocessor. See
7a5c1cfe 335 yt_dlp/postprocessor/__init__.py for a list.
bb66c247 336 * when: When to run the postprocessor. Allowed values are
337 the entries of utils.POSTPROCESS_WHEN
56d868db 338 Assumed to be 'post_process' if not given
71b640cc
PH
339 progress_hooks: A list of functions that get called on download
340 progress, with a dictionary with the entries
5cda4eda 341 * status: One of "downloading", "error", or "finished".
ee69b99a 342 Check this first and ignore unknown values.
3ba7740d 343 * info_dict: The extracted info_dict
71b640cc 344
5cda4eda 345 If status is one of "downloading", or "finished", the
ee69b99a
PH
346 following properties may also be present:
347 * filename: The final filename (always present)
5cda4eda 348 * tmpfilename: The filename we're currently writing to
71b640cc
PH
349 * downloaded_bytes: Bytes on disk
350 * total_bytes: Size of the whole file, None if unknown
5cda4eda
PH
351 * total_bytes_estimate: Guess of the eventual file size,
352 None if unavailable.
353 * elapsed: The number of seconds since download started.
71b640cc
PH
354 * eta: The estimated time in seconds, None if unknown
355 * speed: The download speed in bytes/second, None if
356 unknown
5cda4eda
PH
357 * fragment_index: The counter of the currently
358 downloaded video fragment.
359 * fragment_count: The number of fragments (= individual
360 files that will be merged)
71b640cc
PH
361
362 Progress hooks are guaranteed to be called at least once
363 (with status "finished") if the download is successful.
819e0531 364 postprocessor_hooks: A list of functions that get called on postprocessing
365 progress, with a dictionary with the entries
366 * status: One of "started", "processing", or "finished".
367 Check this first and ignore unknown values.
368 * postprocessor: Name of the postprocessor
369 * info_dict: The extracted info_dict
370
371 Progress hooks are guaranteed to be called at least twice
372 (with status "started" and "finished") if the processing is successful.
45598f15 373 merge_output_format: Extension to use when merging formats.
6b591b29 374 final_ext: Expected final extension; used to detect when the file was
59a7a13e 375 already downloaded and converted
6271f1ca
PH
376 fixup: Automatically correct known faults of the file.
377 One of:
378 - "never": do nothing
379 - "warn": only emit a warning
380 - "detect_or_warn": check whether we can do anything
62cd676c 381 about it, warn otherwise (default)
504f20dd 382 source_address: Client-side IP address to bind to.
1cf376f5 383 sleep_interval_requests: Number of seconds to sleep between requests
384 during extraction
7aa589a5
S
385 sleep_interval: Number of seconds to sleep before each download when
386 used alone or a lower bound of a range for randomized
387 sleep before each download (minimum possible number
388 of seconds to sleep) when used along with
389 max_sleep_interval.
390 max_sleep_interval:Upper bound of a range for randomized sleep before each
391 download (maximum possible number of seconds to sleep).
392 Must only be used along with sleep_interval.
393 Actual sleep time will be a random float from range
394 [sleep_interval; max_sleep_interval].
1cf376f5 395 sleep_interval_subtitles: Number of seconds to sleep before each subtitle download
cfb56d1a
PH
396 listformats: Print an overview of available video formats and exit.
397 list_thumbnails: Print a table of all thumbnails and exit.
0a41f331 398 match_filter: A function that gets called for every video with the signature
399 (info_dict, *, incomplete: bool) -> Optional[str]
400 For backward compatibility with youtube-dl, the signature
401 (info_dict) -> Optional[str] is also allowed.
402 - If it returns a message, the video is ignored.
403 - If it returns None, the video is downloaded.
404 - If it returns utils.NO_DEFAULT, the user is interactively
405 asked whether to download the video.
347de493 406 match_filter_func in utils.py is one example for this.
7e5db8c9 407 no_color: Do not emit color codes in output.
0a840f58 408 geo_bypass: Bypass geographic restriction via faking X-Forwarded-For
504f20dd 409 HTTP header
0a840f58 410 geo_bypass_country:
773f291d
S
411 Two-letter ISO 3166-2 country code that will be used for
412 explicit geographic restriction bypassing via faking
504f20dd 413 X-Forwarded-For HTTP header
5f95927a
S
414 geo_bypass_ip_block:
415 IP range in CIDR notation that will be used similarly to
504f20dd 416 geo_bypass_country
52a8a1e1 417 external_downloader: A dictionary of protocol keys and the executable of the
418 external downloader to use for it. The allowed protocols
419 are default|http|ftp|m3u8|dash|rtsp|rtmp|mms.
420 Set the value to 'native' to use the native downloader
53ed7066 421 compat_opts: Compatibility options. See "Differences in default behavior".
3acf6d38 422 The following options do not work when used through the API:
b5ae35ee 423 filename, abort-on-error, multistreams, no-live-chat, format-sort
dac5df5a 424 no-clean-infojson, no-playlist-metafiles, no-keep-subs, no-attach-info-json.
e4f02757 425 Refer __init__.py for their implementation
819e0531 426 progress_template: Dictionary of templates for progress outputs.
427 Allowed keys are 'download', 'postprocess',
428 'download-title' (console title) and 'postprocess-title'.
429 The template is mapped on a dictionary with keys 'progress' and 'info'
23326151 430 retry_sleep_functions: Dictionary of functions that takes the number of attempts
431 as argument and returns the time to sleep in seconds.
432 Allowed keys are 'http', 'fragment', 'file_access'
0f446365
SW
433 download_ranges: A callback function that gets called for every video with
434 the signature (info_dict, ydl) -> Iterable[Section].
435 Only the returned sections will be downloaded.
436 Each Section is a dict with the following keys:
5ec1b6b7 437 * start_time: Start time of the section in seconds
438 * end_time: End time of the section in seconds
439 * title: Section title (Optional)
440 * index: Section number (Optional)
0f446365 441 force_keyframes_at_cuts: Re-encode the video when downloading ranges to get precise cuts
a7dc6a89 442 noprogress: Do not print the progress bar
fe7e0c98 443
8222d8de 444 The following parameters are not used by YoutubeDL itself, they are used by
7a5c1cfe 445 the downloader (see yt_dlp/downloader/common.py):
51d9739f 446 nopart, updatetime, buffersize, ratelimit, throttledratelimit, min_filesize,
205a0654 447 max_filesize, test, noresizebuffer, retries, file_access_retries, fragment_retries,
a7dc6a89 448 continuedl, xattr_set_filesize, hls_use_mpegts, http_chunk_size,
59a7a13e 449 external_downloader_args, concurrent_fragment_downloads.
76b1bd67
JMF
450
451 The following options are used by the post processors:
c0b7d117
S
452 ffmpeg_location: Location of the ffmpeg/avconv binary; either the path
453 to the binary or its containing directory.
43820c03 454 postprocessor_args: A dictionary of postprocessor/executable keys (in lower case)
34488702 455 and a list of additional command-line arguments for the
456 postprocessor/executable. The dict can also have "PP+EXE" keys
457 which are used when the given exe is used by the given PP.
458 Use 'default' as the name for arguments to passed to all PP
459 For compatibility with youtube-dl, a single list of args
460 can also be used
e409895f 461
462 The following options are used by the extractors:
62bff2c1 463 extractor_retries: Number of times to retry for known errors
464 dynamic_mpd: Whether to process dynamic DASH manifests (default: True)
e409895f 465 hls_split_discontinuity: Split HLS playlists to different formats at
62bff2c1 466 discontinuities such as ad breaks (default: False)
5d3a0e79 467 extractor_args: A dictionary of arguments to be passed to the extractors.
468 See "EXTRACTOR ARGUMENTS" for details.
469 Eg: {'youtube': {'skip': ['dash', 'hls']}}
88f23a18 470 mark_watched: Mark videos watched (even with --simulate). Only for YouTube
1890fc63 471
472 The following options are deprecated and may be removed in the future:
473
7e9a6125 474 playliststart: - Use playlist_items
475 Playlist item to start at.
476 playlistend: - Use playlist_items
477 Playlist item to end at.
478 playlistreverse: - Use playlist_items
479 Download playlist items in reverse order.
1890fc63 480 forceurl: - Use forceprint
481 Force printing final URL.
482 forcetitle: - Use forceprint
483 Force printing title.
484 forceid: - Use forceprint
485 Force printing ID.
486 forcethumbnail: - Use forceprint
487 Force printing thumbnail URL.
488 forcedescription: - Use forceprint
489 Force printing description.
490 forcefilename: - Use forceprint
491 Force printing final filename.
492 forceduration: - Use forceprint
493 Force printing duration.
494 allsubtitles: - Use subtitleslangs = ['all']
495 Downloads all the subtitles of the video
496 (requires writesubtitles or writeautomaticsub)
497 include_ads: - Doesn't work
498 Download ads as well
499 call_home: - Not implemented
500 Boolean, true iff we are allowed to contact the
501 yt-dlp servers for debugging.
502 post_hooks: - Register a custom postprocessor
503 A list of functions that get called as the final step
504 for each video file, after all postprocessors have been
505 called. The filename will be passed as the only argument.
506 hls_prefer_native: - Use external_downloader = {'m3u8': 'native'} or {'m3u8': 'ffmpeg'}.
507 Use the native HLS downloader instead of ffmpeg/avconv
508 if True, otherwise use ffmpeg/avconv if False, otherwise
509 use downloader suggested by extractor if None.
510 prefer_ffmpeg: - avconv support is deprecated
511 If False, use avconv instead of ffmpeg if both are available,
512 otherwise prefer ffmpeg.
513 youtube_include_dash_manifest: - Use extractor_args
5d3a0e79 514 If True (default), DASH manifests and related
62bff2c1 515 data will be downloaded and processed by extractor.
516 You can reduce network I/O by disabling it if you don't
517 care about DASH. (only for youtube)
1890fc63 518 youtube_include_hls_manifest: - Use extractor_args
5d3a0e79 519 If True (default), HLS manifests and related
62bff2c1 520 data will be downloaded and processed by extractor.
521 You can reduce network I/O by disabling it if you don't
522 care about HLS. (only for youtube)
8222d8de
JMF
523 """
524
86e5f3ed 525 _NUMERIC_FIELDS = {
c9969434 526 'width', 'height', 'tbr', 'abr', 'asr', 'vbr', 'fps', 'filesize', 'filesize_approx',
e6f21b3d 527 'timestamp', 'release_timestamp',
c9969434
S
528 'duration', 'view_count', 'like_count', 'dislike_count', 'repost_count',
529 'average_rating', 'comment_count', 'age_limit',
530 'start_time', 'end_time',
531 'chapter_number', 'season_number', 'episode_number',
532 'track_number', 'disc_number', 'release_year',
86e5f3ed 533 }
c9969434 534
6db9c4d5 535 _format_fields = {
536 # NB: Keep in sync with the docstring of extractor/common.py
a44ca5a4 537 'url', 'manifest_url', 'manifest_stream_number', 'ext', 'format', 'format_id', 'format_note',
6db9c4d5 538 'width', 'height', 'resolution', 'dynamic_range', 'tbr', 'abr', 'acodec', 'asr',
539 'vbr', 'fps', 'vcodec', 'container', 'filesize', 'filesize_approx',
540 'player_url', 'protocol', 'fragment_base_url', 'fragments', 'is_from_start',
541 'preference', 'language', 'language_preference', 'quality', 'source_preference',
542 'http_headers', 'stretched_ratio', 'no_resume', 'has_drm', 'downloader_options',
543 'page_url', 'app', 'play_path', 'tc_url', 'flash_version', 'rtmp_live', 'rtmp_conn', 'rtmp_protocol', 'rtmp_real_time'
544 }
48ee10ee 545 _format_selection_exts = {
546 'audio': {'m4a', 'mp3', 'ogg', 'aac'},
547 'video': {'mp4', 'flv', 'webm', '3gp'},
548 'storyboards': {'mhtml'},
549 }
550
3511266b 551 def __init__(self, params=None, auto_init=True):
883d4b1e 552 """Create a FileDownloader object with the given options.
553 @param auto_init Whether to load the default extractors and print header (if verbose).
49a57e70 554 Set to 'no_verbose_header' to not print the header
883d4b1e 555 """
e9f9a10f
JMF
556 if params is None:
557 params = {}
592b7485 558 self.params = params
8b7491c8 559 self._ies = {}
56c73665 560 self._ies_instances = {}
1e43a6f7 561 self._pps = {k: [] for k in POSTPROCESS_WHEN}
b35496d8 562 self._printed_messages = set()
1cf376f5 563 self._first_webpage_request = True
ab8e5e51 564 self._post_hooks = []
933605d7 565 self._progress_hooks = []
819e0531 566 self._postprocessor_hooks = []
8222d8de
JMF
567 self._download_retcode = 0
568 self._num_downloads = 0
9c906919 569 self._num_videos = 0
592b7485 570 self._playlist_level = 0
571 self._playlist_urls = set()
a0e07d31 572 self.cache = Cache(self)
34308b30 573
819e0531 574 windows_enable_vt_mode()
591bb9d3 575 stdout = sys.stderr if self.params.get('logtostderr') else sys.stdout
576 self._out_files = Namespace(
577 out=stdout,
578 error=sys.stderr,
579 screen=sys.stderr if self.params.get('quiet') else stdout,
580 console=None if compat_os_name == 'nt' else next(
cf4f42cb 581 filter(supports_terminal_sequences, (sys.stderr, sys.stdout)), None)
591bb9d3 582 )
583 self._allow_colors = Namespace(**{
584 type_: not self.params.get('no_color') and supports_terminal_sequences(stream)
64fa820c 585 for type_, stream in self._out_files.items_ if type_ != 'console'
591bb9d3 586 })
819e0531 587
6929b41a 588 # The code is left like this to be reused for future deprecations
589 MIN_SUPPORTED, MIN_RECOMMENDED = (3, 7), (3, 7)
eff42759 590 current_version = sys.version_info[:2]
591 if current_version < MIN_RECOMMENDED:
9d339c41 592 msg = ('Support for Python version %d.%d has been deprecated. '
24093d52 593 'See https://github.com/yt-dlp/yt-dlp/issues/3764 for more details.'
c6e07cf1 594 '\n You will no longer receive updates on this version')
eff42759 595 if current_version < MIN_SUPPORTED:
596 msg = 'Python version %d.%d is no longer supported'
597 self.deprecation_warning(
598 f'{msg}! Please update to Python %d.%d or above' % (*current_version, *MIN_RECOMMENDED))
a61f4b28 599
88acdbc2 600 if self.params.get('allow_unplayable_formats'):
601 self.report_warning(
ec11a9f4 602 f'You have asked for {self._format_err("UNPLAYABLE", self.Styles.EMPHASIS)} formats to be listed/downloaded. '
819e0531 603 'This is a developer option intended for debugging. \n'
604 ' If you experience any issues while using this option, '
ec11a9f4 605 f'{self._format_err("DO NOT", self.Styles.ERROR)} open a bug report')
88acdbc2 606
be5df5ee
S
607 def check_deprecated(param, option, suggestion):
608 if self.params.get(param) is not None:
86e5f3ed 609 self.report_warning(f'{option} is deprecated. Use {suggestion} instead')
be5df5ee
S
610 return True
611 return False
612
613 if check_deprecated('cn_verification_proxy', '--cn-verification-proxy', '--geo-verification-proxy'):
38cce791
YCH
614 if self.params.get('geo_verification_proxy') is None:
615 self.params['geo_verification_proxy'] = self.params['cn_verification_proxy']
616
0d1bb027 617 check_deprecated('autonumber', '--auto-number', '-o "%(autonumber)s-%(title)s.%(ext)s"')
618 check_deprecated('usetitle', '--title', '-o "%(title)s-%(id)s.%(ext)s"')
53ed7066 619 check_deprecated('useid', '--id', '-o "%(id)s.%(ext)s"')
0d1bb027 620
49a57e70 621 for msg in self.params.get('_warnings', []):
0d1bb027 622 self.report_warning(msg)
ee8dd27a 623 for msg in self.params.get('_deprecation_warnings', []):
624 self.deprecation_warning(msg)
0d1bb027 625
8a82af35 626 self.params['compat_opts'] = set(self.params.get('compat_opts', ()))
8a82af35 627 if 'list-formats' in self.params['compat_opts']:
ec11a9f4 628 self.params['listformats_table'] = False
629
b5ae35ee 630 if 'overwrites' not in self.params and self.params.get('nooverwrites') is not None:
b868936c 631 # nooverwrites was unnecessarily changed to overwrites
632 # in 0c3d0f51778b153f65c21906031c2e091fcfb641
633 # This ensures compatibility with both keys
634 self.params['overwrites'] = not self.params['nooverwrites']
b5ae35ee 635 elif self.params.get('overwrites') is None:
636 self.params.pop('overwrites', None)
b868936c 637 else:
638 self.params['nooverwrites'] = not self.params['overwrites']
b9d973be 639
455a15e2 640 self.params.setdefault('forceprint', {})
641 self.params.setdefault('print_to_file', {})
bb66c247 642
643 # Compatibility with older syntax
ca30f449 644 if not isinstance(params['forceprint'], dict):
455a15e2 645 self.params['forceprint'] = {'video': params['forceprint']}
ca30f449 646
455a15e2 647 if self.params.get('bidi_workaround', False):
1c088fa8
PH
648 try:
649 import pty
650 master, slave = pty.openpty()
ac668111 651 width = shutil.get_terminal_size().columns
591bb9d3 652 width_args = [] if width is None else ['-w', str(width)]
653 sp_kwargs = {'stdin': subprocess.PIPE, 'stdout': slave, 'stderr': self._out_files.error}
5d681e96 654 try:
d3c93ec2 655 self._output_process = Popen(['bidiv'] + width_args, **sp_kwargs)
5d681e96 656 except OSError:
d3c93ec2 657 self._output_process = Popen(['fribidi', '-c', 'UTF-8'] + width_args, **sp_kwargs)
5d681e96 658 self._output_channel = os.fdopen(master, 'rb')
1c088fa8 659 except OSError as ose:
66e7ace1 660 if ose.errno == errno.ENOENT:
49a57e70 661 self.report_warning(
662 'Could not find fribidi executable, ignoring --bidi-workaround. '
663 'Make sure that fribidi is an executable file in one of the directories in your $PATH.')
1c088fa8
PH
664 else:
665 raise
0783b09b 666
97ec5bc5 667 if auto_init:
668 if auto_init != 'no_verbose_header':
669 self.print_debug_header()
670 self.add_default_info_extractors()
671
3089bc74
S
672 if (sys.platform != 'win32'
673 and sys.getfilesystemencoding() in ['ascii', 'ANSI_X3.4-1968']
455a15e2 674 and not self.params.get('restrictfilenames', False)):
e9137224 675 # Unicode filesystem API will throw errors (#1474, #13027)
34308b30 676 self.report_warning(
6febd1c1 677 'Assuming --restrict-filenames since file system encoding '
1b725173 678 'cannot encode all characters. '
6febd1c1 679 'Set the LC_ALL environment variable to fix this.')
4a98cdbf 680 self.params['restrictfilenames'] = True
34308b30 681
bf1824b3 682 self._parse_outtmpl()
486dd09e 683
187986a8 684 # Creating format selector here allows us to catch syntax errors before the extraction
685 self.format_selector = (
fa9f30b8 686 self.params.get('format') if self.params.get('format') in (None, '-')
093a1710 687 else self.params['format'] if callable(self.params['format'])
187986a8 688 else self.build_format_selector(self.params['format']))
689
8b7539d2 690 # Set http_headers defaults according to std_headers
691 self.params['http_headers'] = merge_headers(std_headers, self.params.get('http_headers', {}))
692
013b50b7 693 hooks = {
694 'post_hooks': self.add_post_hook,
695 'progress_hooks': self.add_progress_hook,
696 'postprocessor_hooks': self.add_postprocessor_hook,
697 }
698 for opt, fn in hooks.items():
699 for ph in self.params.get(opt, []):
700 fn(ph)
71b640cc 701
5bfc8bee 702 for pp_def_raw in self.params.get('postprocessors', []):
703 pp_def = dict(pp_def_raw)
704 when = pp_def.pop('when', 'post_process')
705 self.add_post_processor(
f9934b96 706 get_postprocessor(pp_def.pop('key'))(self, **pp_def),
5bfc8bee 707 when=when)
708
97ec5bc5 709 self._setup_opener()
51fb4995
YCH
710 register_socks_protocols()
711
ed39cac5 712 def preload_download_archive(fn):
713 """Preload the archive, if any is specified"""
714 if fn is None:
715 return False
49a57e70 716 self.write_debug(f'Loading archive file {fn!r}')
ed39cac5 717 try:
718 with locked_file(fn, 'r', encoding='utf-8') as archive_file:
719 for line in archive_file:
720 self.archive.add(line.strip())
86e5f3ed 721 except OSError as ioe:
ed39cac5 722 if ioe.errno != errno.ENOENT:
723 raise
724 return False
725 return True
726
727 self.archive = set()
728 preload_download_archive(self.params.get('download_archive'))
729
7d4111ed
PH
730 def warn_if_short_id(self, argv):
731 # short YouTube ID starting with dash?
732 idxs = [
733 i for i, a in enumerate(argv)
734 if re.match(r'^-[0-9A-Za-z_-]{10}$', a)]
735 if idxs:
736 correct_argv = (
7a5c1cfe 737 ['yt-dlp']
3089bc74
S
738 + [a for i, a in enumerate(argv) if i not in idxs]
739 + ['--'] + [argv[i] for i in idxs]
7d4111ed
PH
740 )
741 self.report_warning(
742 'Long argument string detected. '
49a57e70 743 'Use -- to separate parameters and URLs, like this:\n%s' %
7d4111ed
PH
744 args_to_str(correct_argv))
745
8222d8de
JMF
746 def add_info_extractor(self, ie):
747 """Add an InfoExtractor object to the end of the list."""
8b7491c8 748 ie_key = ie.ie_key()
749 self._ies[ie_key] = ie
e52d7f85 750 if not isinstance(ie, type):
8b7491c8 751 self._ies_instances[ie_key] = ie
e52d7f85 752 ie.set_downloader(self)
8222d8de 753
8b7491c8 754 def _get_info_extractor_class(self, ie_key):
755 ie = self._ies.get(ie_key)
756 if ie is None:
757 ie = get_info_extractor(ie_key)
758 self.add_info_extractor(ie)
759 return ie
760
56c73665
JMF
761 def get_info_extractor(self, ie_key):
762 """
763 Get an instance of an IE with name ie_key, it will try to get one from
764 the _ies list, if there's no instance it will create a new one and add
765 it to the extractor list.
766 """
767 ie = self._ies_instances.get(ie_key)
768 if ie is None:
769 ie = get_info_extractor(ie_key)()
770 self.add_info_extractor(ie)
771 return ie
772
023fa8c4
JMF
773 def add_default_info_extractors(self):
774 """
775 Add the InfoExtractors returned by gen_extractors to the end of the list
776 """
e52d7f85 777 for ie in gen_extractor_classes():
023fa8c4
JMF
778 self.add_info_extractor(ie)
779
56d868db 780 def add_post_processor(self, pp, when='post_process'):
8222d8de 781 """Add a PostProcessor object to the end of the chain."""
8aa0e7cd 782 assert when in POSTPROCESS_WHEN, f'Invalid when={when}'
5bfa4862 783 self._pps[when].append(pp)
8222d8de
JMF
784 pp.set_downloader(self)
785
ab8e5e51
AM
786 def add_post_hook(self, ph):
787 """Add the post hook"""
788 self._post_hooks.append(ph)
789
933605d7 790 def add_progress_hook(self, ph):
819e0531 791 """Add the download progress hook"""
933605d7 792 self._progress_hooks.append(ph)
8ab470f1 793
819e0531 794 def add_postprocessor_hook(self, ph):
795 """Add the postprocessing progress hook"""
796 self._postprocessor_hooks.append(ph)
5bfc8bee 797 for pps in self._pps.values():
798 for pp in pps:
799 pp.add_progress_hook(ph)
819e0531 800
1c088fa8 801 def _bidi_workaround(self, message):
5d681e96 802 if not hasattr(self, '_output_channel'):
1c088fa8
PH
803 return message
804
5d681e96 805 assert hasattr(self, '_output_process')
14f25df2 806 assert isinstance(message, str)
6febd1c1 807 line_count = message.count('\n') + 1
0f06bcd7 808 self._output_process.stdin.write((message + '\n').encode())
5d681e96 809 self._output_process.stdin.flush()
0f06bcd7 810 res = ''.join(self._output_channel.readline().decode()
9e1a5b84 811 for _ in range(line_count))
6febd1c1 812 return res[:-len('\n')]
1c088fa8 813
b35496d8 814 def _write_string(self, message, out=None, only_once=False):
815 if only_once:
816 if message in self._printed_messages:
817 return
818 self._printed_messages.add(message)
819 write_string(message, out=out, encoding=self.params.get('encoding'))
734f90bb 820
cf4f42cb 821 def to_stdout(self, message, skip_eol=False, quiet=None):
0760b0a7 822 """Print message to stdout"""
cf4f42cb 823 if quiet is not None:
ae6a1b95 824 self.deprecation_warning('"YoutubeDL.to_stdout" no longer accepts the argument quiet. Use "YoutubeDL.to_screen" instead')
8a82af35 825 if skip_eol is not False:
826 self.deprecation_warning('"YoutubeDL.to_stdout" no longer accepts the argument skip_eol. Use "YoutubeDL.to_screen" instead')
0bf9dc1e 827 self._write_string(f'{self._bidi_workaround(message)}\n', self._out_files.out)
cf4f42cb 828
829 def to_screen(self, message, skip_eol=False, quiet=None):
830 """Print message to screen if not in quiet mode"""
8bf9319e 831 if self.params.get('logger'):
43afe285 832 self.params['logger'].debug(message)
cf4f42cb 833 return
834 if (self.params.get('quiet') if quiet is None else quiet) and not self.params.get('verbose'):
835 return
836 self._write_string(
837 '%s%s' % (self._bidi_workaround(message), ('' if skip_eol else '\n')),
591bb9d3 838 self._out_files.screen)
8222d8de 839
b35496d8 840 def to_stderr(self, message, only_once=False):
0760b0a7 841 """Print message to stderr"""
14f25df2 842 assert isinstance(message, str)
8bf9319e 843 if self.params.get('logger'):
43afe285
IB
844 self.params['logger'].error(message)
845 else:
5792c950 846 self._write_string(f'{self._bidi_workaround(message)}\n', self._out_files.error, only_once=only_once)
cf4f42cb 847
848 def _send_console_code(self, code):
591bb9d3 849 if compat_os_name == 'nt' or not self._out_files.console:
cf4f42cb 850 return
591bb9d3 851 self._write_string(code, self._out_files.console)
8222d8de 852
1e5b9a95
PH
853 def to_console_title(self, message):
854 if not self.params.get('consoletitle', False):
855 return
3efb96a6 856 message = remove_terminal_sequences(message)
4bede0d8
C
857 if compat_os_name == 'nt':
858 if ctypes.windll.kernel32.GetConsoleWindow():
859 # c_wchar_p() might not be necessary if `message` is
860 # already of type unicode()
861 ctypes.windll.kernel32.SetConsoleTitleW(ctypes.c_wchar_p(message))
cf4f42cb 862 else:
863 self._send_console_code(f'\033]0;{message}\007')
1e5b9a95 864
bdde425c 865 def save_console_title(self):
cf4f42cb 866 if not self.params.get('consoletitle') or self.params.get('simulate'):
bdde425c 867 return
592b7485 868 self._send_console_code('\033[22;0t') # Save the title on stack
bdde425c
PH
869
870 def restore_console_title(self):
cf4f42cb 871 if not self.params.get('consoletitle') or self.params.get('simulate'):
bdde425c 872 return
592b7485 873 self._send_console_code('\033[23;0t') # Restore the title from stack
bdde425c
PH
874
875 def __enter__(self):
876 self.save_console_title()
877 return self
878
879 def __exit__(self, *args):
880 self.restore_console_title()
f89197d7 881
dca08720 882 if self.params.get('cookiefile') is not None:
1bab3437 883 self.cookiejar.save(ignore_discard=True, ignore_expires=True)
bdde425c 884
fa9f30b8 885 def trouble(self, message=None, tb=None, is_error=True):
8222d8de
JMF
886 """Determine action to take when a download problem appears.
887
888 Depending on if the downloader has been configured to ignore
889 download errors or not, this method may throw an exception or
890 not when errors are found, after printing the message.
891
fa9f30b8 892 @param tb If given, is additional traceback information
893 @param is_error Whether to raise error according to ignorerrors
8222d8de
JMF
894 """
895 if message is not None:
896 self.to_stderr(message)
897 if self.params.get('verbose'):
898 if tb is None:
899 if sys.exc_info()[0]: # if .trouble has been called from an except block
6febd1c1 900 tb = ''
8222d8de 901 if hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
6febd1c1 902 tb += ''.join(traceback.format_exception(*sys.exc_info()[1].exc_info))
c0384f22 903 tb += encode_compat_str(traceback.format_exc())
8222d8de
JMF
904 else:
905 tb_data = traceback.format_list(traceback.extract_stack())
6febd1c1 906 tb = ''.join(tb_data)
c19bc311 907 if tb:
908 self.to_stderr(tb)
fa9f30b8 909 if not is_error:
910 return
b1940459 911 if not self.params.get('ignoreerrors'):
8222d8de
JMF
912 if sys.exc_info()[0] and hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
913 exc_info = sys.exc_info()[1].exc_info
914 else:
915 exc_info = sys.exc_info()
916 raise DownloadError(message, exc_info)
917 self._download_retcode = 1
918
19a03940 919 Styles = Namespace(
920 HEADERS='yellow',
921 EMPHASIS='light blue',
492272fe 922 FILENAME='green',
19a03940 923 ID='green',
924 DELIM='blue',
925 ERROR='red',
926 WARNING='yellow',
927 SUPPRESS='light black',
928 )
ec11a9f4 929
7578d77d 930 def _format_text(self, handle, allow_colors, text, f, fallback=None, *, test_encoding=False):
e5a998f3 931 text = str(text)
ec11a9f4 932 if test_encoding:
933 original_text = text
5c104538 934 # handle.encoding can be None. See https://github.com/yt-dlp/yt-dlp/issues/2711
935 encoding = self.params.get('encoding') or getattr(handle, 'encoding', None) or 'ascii'
ec11a9f4 936 text = text.encode(encoding, 'ignore').decode(encoding)
937 if fallback is not None and text != original_text:
938 text = fallback
7578d77d 939 return format_text(text, f) if allow_colors else text if fallback is None else fallback
ec11a9f4 940
591bb9d3 941 def _format_out(self, *args, **kwargs):
942 return self._format_text(self._out_files.out, self._allow_colors.out, *args, **kwargs)
943
ec11a9f4 944 def _format_screen(self, *args, **kwargs):
591bb9d3 945 return self._format_text(self._out_files.screen, self._allow_colors.screen, *args, **kwargs)
ec11a9f4 946
947 def _format_err(self, *args, **kwargs):
591bb9d3 948 return self._format_text(self._out_files.error, self._allow_colors.error, *args, **kwargs)
819e0531 949
c84aeac6 950 def report_warning(self, message, only_once=False):
8222d8de
JMF
951 '''
952 Print the message to stderr, it will be prefixed with 'WARNING:'
953 If stderr is a tty file the 'WARNING:' will be colored
954 '''
6d07ce01
JMF
955 if self.params.get('logger') is not None:
956 self.params['logger'].warning(message)
8222d8de 957 else:
ad8915b7
PH
958 if self.params.get('no_warnings'):
959 return
ec11a9f4 960 self.to_stderr(f'{self._format_err("WARNING:", self.Styles.WARNING)} {message}', only_once)
8222d8de 961
ee8dd27a 962 def deprecation_warning(self, message):
963 if self.params.get('logger') is not None:
a44ca5a4 964 self.params['logger'].warning(f'DeprecationWarning: {message}')
ee8dd27a 965 else:
966 self.to_stderr(f'{self._format_err("DeprecationWarning:", self.Styles.ERROR)} {message}', True)
967
fa9f30b8 968 def report_error(self, message, *args, **kwargs):
8222d8de
JMF
969 '''
970 Do the same as trouble, but prefixes the message with 'ERROR:', colored
971 in red if stderr is a tty file.
972 '''
fa9f30b8 973 self.trouble(f'{self._format_err("ERROR:", self.Styles.ERROR)} {message}', *args, **kwargs)
8222d8de 974
b35496d8 975 def write_debug(self, message, only_once=False):
0760b0a7 976 '''Log debug message or Print message to stderr'''
977 if not self.params.get('verbose', False):
978 return
8a82af35 979 message = f'[debug] {message}'
0760b0a7 980 if self.params.get('logger'):
981 self.params['logger'].debug(message)
982 else:
b35496d8 983 self.to_stderr(message, only_once)
0760b0a7 984
8222d8de
JMF
985 def report_file_already_downloaded(self, file_name):
986 """Report file has already been fully downloaded."""
987 try:
6febd1c1 988 self.to_screen('[download] %s has already been downloaded' % file_name)
ce02ed60 989 except UnicodeEncodeError:
6febd1c1 990 self.to_screen('[download] The file has already been downloaded')
8222d8de 991
0c3d0f51 992 def report_file_delete(self, file_name):
993 """Report that existing file will be deleted."""
994 try:
c25228e5 995 self.to_screen('Deleting existing file %s' % file_name)
0c3d0f51 996 except UnicodeEncodeError:
c25228e5 997 self.to_screen('Deleting existing file')
0c3d0f51 998
319b6059 999 def raise_no_formats(self, info, forced=False, *, msg=None):
0a5a191a 1000 has_drm = info.get('_has_drm')
319b6059 1001 ignored, expected = self.params.get('ignore_no_formats_error'), bool(msg)
1002 msg = msg or has_drm and 'This video is DRM protected' or 'No video formats found!'
1003 if forced or not ignored:
1151c407 1004 raise ExtractorError(msg, video_id=info['id'], ie=info['extractor'],
319b6059 1005 expected=has_drm or ignored or expected)
88acdbc2 1006 else:
1007 self.report_warning(msg)
1008
de6000d9 1009 def parse_outtmpl(self):
bf1824b3 1010 self.deprecation_warning('"YoutubeDL.parse_outtmpl" is deprecated and may be removed in a future version')
1011 self._parse_outtmpl()
1012 return self.params['outtmpl']
1013
1014 def _parse_outtmpl(self):
7b2c3f47 1015 sanitize = IDENTITY
bf1824b3 1016 if self.params.get('restrictfilenames'): # Remove spaces in the default template
71ce444a 1017 sanitize = lambda x: x.replace(' - ', ' ').replace(' ', '-')
bf1824b3 1018
1019 outtmpl = self.params.setdefault('outtmpl', {})
1020 if not isinstance(outtmpl, dict):
1021 self.params['outtmpl'] = outtmpl = {'default': outtmpl}
1022 outtmpl.update({k: sanitize(v) for k, v in DEFAULT_OUTTMPL.items() if outtmpl.get(k) is None})
de6000d9 1023
21cd8fae 1024 def get_output_path(self, dir_type='', filename=None):
1025 paths = self.params.get('paths', {})
1026 assert isinstance(paths, dict)
1027 path = os.path.join(
1028 expand_path(paths.get('home', '').strip()),
1029 expand_path(paths.get(dir_type, '').strip()) if dir_type else '',
1030 filename or '')
21cd8fae 1031 return sanitize_path(path, force=self.params.get('windowsfilenames'))
1032
76a264ac 1033 @staticmethod
901130bb 1034 def _outtmpl_expandpath(outtmpl):
1035 # expand_path translates '%%' into '%' and '$$' into '$'
1036 # correspondingly that is not what we want since we need to keep
1037 # '%%' intact for template dict substitution step. Working around
1038 # with boundary-alike separator hack.
1039 sep = ''.join([random.choice(ascii_letters) for _ in range(32)])
86e5f3ed 1040 outtmpl = outtmpl.replace('%%', f'%{sep}%').replace('$$', f'${sep}$')
901130bb 1041
1042 # outtmpl should be expand_path'ed before template dict substitution
1043 # because meta fields may contain env variables we don't want to
1044 # be expanded. For example, for outtmpl "%(title)s.%(ext)s" and
1045 # title "Hello $PATH", we don't want `$PATH` to be expanded.
1046 return expand_path(outtmpl).replace(sep, '')
1047
1048 @staticmethod
1049 def escape_outtmpl(outtmpl):
1050 ''' Escape any remaining strings like %s, %abc% etc. '''
1051 return re.sub(
1052 STR_FORMAT_RE_TMPL.format('', '(?![%(\0])'),
1053 lambda mobj: ('' if mobj.group('has_key') else '%') + mobj.group(0),
1054 outtmpl)
1055
1056 @classmethod
1057 def validate_outtmpl(cls, outtmpl):
76a264ac 1058 ''' @return None or Exception object '''
7d1eb38a 1059 outtmpl = re.sub(
47cdc68e 1060 STR_FORMAT_RE_TMPL.format('[^)]*', '[ljhqBUDS]'),
7d1eb38a 1061 lambda mobj: f'{mobj.group(0)[:-1]}s',
1062 cls._outtmpl_expandpath(outtmpl))
76a264ac 1063 try:
7d1eb38a 1064 cls.escape_outtmpl(outtmpl) % collections.defaultdict(int)
76a264ac 1065 return None
1066 except ValueError as err:
1067 return err
1068
03b4de72 1069 @staticmethod
1070 def _copy_infodict(info_dict):
1071 info_dict = dict(info_dict)
09b49e1f 1072 info_dict.pop('__postprocessors', None)
415f8d51 1073 info_dict.pop('__pending_error', None)
03b4de72 1074 return info_dict
1075
e0fd9573 1076 def prepare_outtmpl(self, outtmpl, info_dict, sanitize=False):
1077 """ Make the outtmpl and info_dict suitable for substitution: ydl.escape_outtmpl(outtmpl) % info_dict
1078 @param sanitize Whether to sanitize the output as a filename.
1079 For backward compatibility, a function can also be passed
1080 """
1081
6e84b215 1082 info_dict.setdefault('epoch', int(time.time())) # keep epoch consistent once set
143db31d 1083
03b4de72 1084 info_dict = self._copy_infodict(info_dict)
752cda38 1085 info_dict['duration_string'] = ( # %(duration>%H-%M-%S)s is wrong if duration > 24hrs
53c18592 1086 formatSeconds(info_dict['duration'], '-' if sanitize else ':')
143db31d 1087 if info_dict.get('duration', None) is not None
1088 else None)
1d485a1a 1089 info_dict['autonumber'] = int(self.params.get('autonumber_start', 1) - 1 + self._num_downloads)
9c906919 1090 info_dict['video_autonumber'] = self._num_videos
752cda38 1091 if info_dict.get('resolution') is None:
1092 info_dict['resolution'] = self.format_resolution(info_dict, default=None)
143db31d 1093
e6f21b3d 1094 # For fields playlist_index, playlist_autonumber and autonumber convert all occurrences
143db31d 1095 # of %(field)s to %(field)0Nd for backward compatibility
1096 field_size_compat_map = {
0a5a191a 1097 'playlist_index': number_of_digits(info_dict.get('__last_playlist_index') or 0),
ec11a9f4 1098 'playlist_autonumber': number_of_digits(info_dict.get('n_entries') or 0),
752cda38 1099 'autonumber': self.params.get('autonumber_size') or 5,
143db31d 1100 }
752cda38 1101
385a27fa 1102 TMPL_DICT = {}
47cdc68e 1103 EXTERNAL_FORMAT_RE = re.compile(STR_FORMAT_RE_TMPL.format('[^)]*', f'[{STR_FORMAT_TYPES}ljhqBUDS]'))
385a27fa 1104 MATH_FUNCTIONS = {
1105 '+': float.__add__,
1106 '-': float.__sub__,
1107 }
e625be0d 1108 # Field is of the form key1.key2...
1109 # where keys (except first) can be string, int or slice
2b8a2973 1110 FIELD_RE = r'\w*(?:\.(?:\w+|{num}|{num}?(?::{num}?){{1,2}}))*'.format(num=r'(?:-?\d+)')
1d485a1a 1111 MATH_FIELD_RE = rf'(?:{FIELD_RE}|-?{NUMBER_RE})'
385a27fa 1112 MATH_OPERATORS_RE = r'(?:%s)' % '|'.join(map(re.escape, MATH_FUNCTIONS.keys()))
1d485a1a 1113 INTERNAL_FORMAT_RE = re.compile(rf'''(?x)
e625be0d 1114 (?P<negate>-)?
1d485a1a 1115 (?P<fields>{FIELD_RE})
1116 (?P<maths>(?:{MATH_OPERATORS_RE}{MATH_FIELD_RE})*)
e625be0d 1117 (?:>(?P<strf_format>.+?))?
34baa9fd 1118 (?P<remaining>
1119 (?P<alternate>(?<!\\),[^|&)]+)?
1120 (?:&(?P<replacement>.*?))?
1121 (?:\|(?P<default>.*?))?
1d485a1a 1122 )$''')
752cda38 1123
2b8a2973 1124 def _traverse_infodict(k):
1125 k = k.split('.')
1126 if k[0] == '':
1127 k.pop(0)
1128 return traverse_obj(info_dict, k, is_user_input=True, traverse_string=True)
76a264ac 1129
752cda38 1130 def get_value(mdict):
1131 # Object traversal
2b8a2973 1132 value = _traverse_infodict(mdict['fields'])
752cda38 1133 # Negative
1134 if mdict['negate']:
1135 value = float_or_none(value)
1136 if value is not None:
1137 value *= -1
1138 # Do maths
385a27fa 1139 offset_key = mdict['maths']
1140 if offset_key:
752cda38 1141 value = float_or_none(value)
1142 operator = None
385a27fa 1143 while offset_key:
1144 item = re.match(
1145 MATH_FIELD_RE if operator else MATH_OPERATORS_RE,
1146 offset_key).group(0)
1147 offset_key = offset_key[len(item):]
1148 if operator is None:
752cda38 1149 operator = MATH_FUNCTIONS[item]
385a27fa 1150 continue
1151 item, multiplier = (item[1:], -1) if item[0] == '-' else (item, 1)
1152 offset = float_or_none(item)
1153 if offset is None:
2b8a2973 1154 offset = float_or_none(_traverse_infodict(item))
385a27fa 1155 try:
1156 value = operator(value, multiplier * offset)
1157 except (TypeError, ZeroDivisionError):
1158 return None
1159 operator = None
752cda38 1160 # Datetime formatting
1161 if mdict['strf_format']:
7c37ff97 1162 value = strftime_or_none(value, mdict['strf_format'].replace('\\,', ','))
752cda38 1163
1164 return value
1165
b868936c 1166 na = self.params.get('outtmpl_na_placeholder', 'NA')
1167
e0fd9573 1168 def filename_sanitizer(key, value, restricted=self.params.get('restrictfilenames')):
5c3895ff 1169 return sanitize_filename(str(value), restricted=restricted, is_id=(
1170 bool(re.search(r'(^|[_.])id(\.|$)', key))
8a82af35 1171 if 'filename-sanitization' in self.params['compat_opts']
5c3895ff 1172 else NO_DEFAULT))
e0fd9573 1173
1174 sanitizer = sanitize if callable(sanitize) else filename_sanitizer
1175 sanitize = bool(sanitize)
1176
6e84b215 1177 def _dumpjson_default(obj):
1178 if isinstance(obj, (set, LazyList)):
1179 return list(obj)
adbc4ec4 1180 return repr(obj)
6e84b215 1181
752cda38 1182 def create_key(outer_mobj):
1183 if not outer_mobj.group('has_key'):
b836dc94 1184 return outer_mobj.group(0)
752cda38 1185 key = outer_mobj.group('key')
752cda38 1186 mobj = re.match(INTERNAL_FORMAT_RE, key)
e0fd9573 1187 initial_field = mobj.group('fields') if mobj else ''
e978789f 1188 value, replacement, default = None, None, na
7c37ff97 1189 while mobj:
e625be0d 1190 mobj = mobj.groupdict()
7c37ff97 1191 default = mobj['default'] if mobj['default'] is not None else default
752cda38 1192 value = get_value(mobj)
e978789f 1193 replacement = mobj['replacement']
7c37ff97 1194 if value is None and mobj['alternate']:
34baa9fd 1195 mobj = re.match(INTERNAL_FORMAT_RE, mobj['remaining'][1:])
7c37ff97 1196 else:
1197 break
752cda38 1198
b868936c 1199 fmt = outer_mobj.group('format')
752cda38 1200 if fmt == 's' and value is not None and key in field_size_compat_map.keys():
86e5f3ed 1201 fmt = f'0{field_size_compat_map[key]:d}d'
752cda38 1202
e978789f 1203 value = default if value is None else value if replacement is None else replacement
752cda38 1204
4476d2c7 1205 flags = outer_mobj.group('conversion') or ''
7d1eb38a 1206 str_fmt = f'{fmt[:-1]}s'
524e2e4f 1207 if fmt[-1] == 'l': # list
4476d2c7 1208 delim = '\n' if '#' in flags else ', '
9e907ebd 1209 value, fmt = delim.join(map(str, variadic(value, allowed_types=(str, bytes)))), str_fmt
524e2e4f 1210 elif fmt[-1] == 'j': # json
4476d2c7 1211 value, fmt = json.dumps(value, default=_dumpjson_default, indent=4 if '#' in flags else None), str_fmt
47cdc68e 1212 elif fmt[-1] == 'h': # html
1213 value, fmt = escapeHTML(value), str_fmt
524e2e4f 1214 elif fmt[-1] == 'q': # quoted
4476d2c7 1215 value = map(str, variadic(value) if '#' in flags else [value])
1216 value, fmt = ' '.join(map(compat_shlex_quote, value)), str_fmt
524e2e4f 1217 elif fmt[-1] == 'B': # bytes
0f06bcd7 1218 value = f'%{str_fmt}'.encode() % str(value).encode()
f5aa5cfb 1219 value, fmt = value.decode('utf-8', 'ignore'), 's'
524e2e4f 1220 elif fmt[-1] == 'U': # unicode normalized
524e2e4f 1221 value, fmt = unicodedata.normalize(
1222 # "+" = compatibility equivalence, "#" = NFD
4476d2c7 1223 'NF%s%s' % ('K' if '+' in flags else '', 'D' if '#' in flags else 'C'),
524e2e4f 1224 value), str_fmt
e0fd9573 1225 elif fmt[-1] == 'D': # decimal suffix
abbeeebc 1226 num_fmt, fmt = fmt[:-1].replace('#', ''), 's'
1227 value = format_decimal_suffix(value, f'%{num_fmt}f%s' if num_fmt else '%d%s',
1228 factor=1024 if '#' in flags else 1000)
37893bb0 1229 elif fmt[-1] == 'S': # filename sanitization
e0fd9573 1230 value, fmt = filename_sanitizer(initial_field, value, restricted='#' in flags), str_fmt
7d1eb38a 1231 elif fmt[-1] == 'c':
524e2e4f 1232 if value:
1233 value = str(value)[0]
76a264ac 1234 else:
524e2e4f 1235 fmt = str_fmt
76a264ac 1236 elif fmt[-1] not in 'rs': # numeric
a439a3a4 1237 value = float_or_none(value)
752cda38 1238 if value is None:
1239 value, fmt = default, 's'
901130bb 1240
752cda38 1241 if sanitize:
1242 if fmt[-1] == 'r':
1243 # If value is an object, sanitize might convert it to a string
1244 # So we convert it to repr first
7d1eb38a 1245 value, fmt = repr(value), str_fmt
639f1cea 1246 if fmt[-1] in 'csr':
e0fd9573 1247 value = sanitizer(initial_field, value)
901130bb 1248
b868936c 1249 key = '%s\0%s' % (key.replace('%', '%\0'), outer_mobj.group('format'))
385a27fa 1250 TMPL_DICT[key] = value
b868936c 1251 return '{prefix}%({key}){fmt}'.format(key=key, fmt=fmt, prefix=outer_mobj.group('prefix'))
752cda38 1252
385a27fa 1253 return EXTERNAL_FORMAT_RE.sub(create_key, outtmpl), TMPL_DICT
143db31d 1254
819e0531 1255 def evaluate_outtmpl(self, outtmpl, info_dict, *args, **kwargs):
1256 outtmpl, info_dict = self.prepare_outtmpl(outtmpl, info_dict, *args, **kwargs)
1257 return self.escape_outtmpl(outtmpl) % info_dict
1258
5127e92a 1259 def _prepare_filename(self, info_dict, *, outtmpl=None, tmpl_type=None):
1260 assert None in (outtmpl, tmpl_type), 'outtmpl and tmpl_type are mutually exclusive'
1261 if outtmpl is None:
bf1824b3 1262 outtmpl = self.params['outtmpl'].get(tmpl_type or 'default', self.params['outtmpl']['default'])
8222d8de 1263 try:
5127e92a 1264 outtmpl = self._outtmpl_expandpath(outtmpl)
e0fd9573 1265 filename = self.evaluate_outtmpl(outtmpl, info_dict, True)
6a0546e3 1266 if not filename:
1267 return None
15da37c7 1268
5127e92a 1269 if tmpl_type in ('', 'temp'):
6a0546e3 1270 final_ext, ext = self.params.get('final_ext'), info_dict.get('ext')
1271 if final_ext and ext and final_ext != ext and filename.endswith(f'.{final_ext}'):
1272 filename = replace_extension(filename, ext, final_ext)
5127e92a 1273 elif tmpl_type:
6a0546e3 1274 force_ext = OUTTMPL_TYPES[tmpl_type]
1275 if force_ext:
1276 filename = replace_extension(filename, force_ext, info_dict.get('ext'))
de6000d9 1277
bdc3fd2f
U
1278 # https://github.com/blackjack4494/youtube-dlc/issues/85
1279 trim_file_name = self.params.get('trim_file_name', False)
1280 if trim_file_name:
5c22c63d 1281 no_ext, *ext = filename.rsplit('.', 2)
1282 filename = join_nonempty(no_ext[:trim_file_name], *ext, delim='.')
bdc3fd2f 1283
0202b52a 1284 return filename
8222d8de 1285 except ValueError as err:
6febd1c1 1286 self.report_error('Error in output template: ' + str(err) + ' (encoding: ' + repr(preferredencoding()) + ')')
8222d8de
JMF
1287 return None
1288
5127e92a 1289 def prepare_filename(self, info_dict, dir_type='', *, outtmpl=None, warn=False):
1290 """Generate the output filename"""
1291 if outtmpl:
1292 assert not dir_type, 'outtmpl and dir_type are mutually exclusive'
1293 dir_type = None
1294 filename = self._prepare_filename(info_dict, tmpl_type=dir_type, outtmpl=outtmpl)
80c03fa9 1295 if not filename and dir_type not in ('', 'temp'):
1296 return ''
de6000d9 1297
c84aeac6 1298 if warn:
21cd8fae 1299 if not self.params.get('paths'):
de6000d9 1300 pass
1301 elif filename == '-':
c84aeac6 1302 self.report_warning('--paths is ignored when an outputting to stdout', only_once=True)
de6000d9 1303 elif os.path.isabs(filename):
c84aeac6 1304 self.report_warning('--paths is ignored since an absolute path is given in output template', only_once=True)
de6000d9 1305 if filename == '-' or not filename:
1306 return filename
1307
21cd8fae 1308 return self.get_output_path(dir_type, filename)
0202b52a 1309
120fe513 1310 def _match_entry(self, info_dict, incomplete=False, silent=False):
ecdec191 1311 """ Returns None if the file should be downloaded """
8222d8de 1312
3bec830a 1313 video_title = info_dict.get('title', info_dict.get('id', 'entry'))
c77495e3 1314
8b0d7497 1315 def check_filter():
8b0d7497 1316 if 'title' in info_dict:
1317 # This can happen when we're just evaluating the playlist
1318 title = info_dict['title']
1319 matchtitle = self.params.get('matchtitle', False)
1320 if matchtitle:
1321 if not re.search(matchtitle, title, re.IGNORECASE):
1322 return '"' + title + '" title did not match pattern "' + matchtitle + '"'
1323 rejecttitle = self.params.get('rejecttitle', False)
1324 if rejecttitle:
1325 if re.search(rejecttitle, title, re.IGNORECASE):
1326 return '"' + title + '" title matched reject pattern "' + rejecttitle + '"'
1327 date = info_dict.get('upload_date')
1328 if date is not None:
1329 dateRange = self.params.get('daterange', DateRange())
1330 if date not in dateRange:
86e5f3ed 1331 return f'{date_from_str(date).isoformat()} upload date is not in range {dateRange}'
8b0d7497 1332 view_count = info_dict.get('view_count')
1333 if view_count is not None:
1334 min_views = self.params.get('min_views')
1335 if min_views is not None and view_count < min_views:
1336 return 'Skipping %s, because it has not reached minimum view count (%d/%d)' % (video_title, view_count, min_views)
1337 max_views = self.params.get('max_views')
1338 if max_views is not None and view_count > max_views:
1339 return 'Skipping %s, because it has exceeded the maximum view count (%d/%d)' % (video_title, view_count, max_views)
1340 if age_restricted(info_dict.get('age_limit'), self.params.get('age_limit')):
1341 return 'Skipping "%s" because it is age restricted' % video_title
8b0d7497 1342
8f18aca8 1343 match_filter = self.params.get('match_filter')
1344 if match_filter is not None:
1345 try:
1346 ret = match_filter(info_dict, incomplete=incomplete)
1347 except TypeError:
1348 # For backward compatibility
1349 ret = None if incomplete else match_filter(info_dict)
492272fe 1350 if ret is NO_DEFAULT:
1351 while True:
1352 filename = self._format_screen(self.prepare_filename(info_dict), self.Styles.FILENAME)
1353 reply = input(self._format_screen(
1354 f'Download "{filename}"? (Y/n): ', self.Styles.EMPHASIS)).lower().strip()
1355 if reply in {'y', ''}:
1356 return None
1357 elif reply == 'n':
1358 return f'Skipping {video_title}'
492272fe 1359 elif ret is not None:
8f18aca8 1360 return ret
8b0d7497 1361 return None
1362
c77495e3 1363 if self.in_download_archive(info_dict):
1364 reason = '%s has already been recorded in the archive' % video_title
1365 break_opt, break_err = 'break_on_existing', ExistingVideoReached
1366 else:
1367 reason = check_filter()
1368 break_opt, break_err = 'break_on_reject', RejectedVideoReached
8b0d7497 1369 if reason is not None:
120fe513 1370 if not silent:
1371 self.to_screen('[download] ' + reason)
c77495e3 1372 if self.params.get(break_opt, False):
1373 raise break_err()
8b0d7497 1374 return reason
fe7e0c98 1375
b6c45014
JMF
1376 @staticmethod
1377 def add_extra_info(info_dict, extra_info):
1378 '''Set the keys from extra_info in info dict if they are missing'''
1379 for key, value in extra_info.items():
1380 info_dict.setdefault(key, value)
1381
409e1828 1382 def extract_info(self, url, download=True, ie_key=None, extra_info=None,
61aa5ba3 1383 process=True, force_generic_extractor=False):
41d1cca3 1384 """
1385 Return a list with a dictionary for each video extracted.
1386
1387 Arguments:
1388 url -- URL to extract
1389
1390 Keyword arguments:
1391 download -- whether to download videos during extraction
1392 ie_key -- extractor key hint
1393 extra_info -- dictionary containing the extra values to add to each result
1394 process -- whether to resolve all unresolved references (URLs, playlist items),
1395 must be True for download to work.
1396 force_generic_extractor -- force using the generic extractor
1397 """
fe7e0c98 1398
409e1828 1399 if extra_info is None:
1400 extra_info = {}
1401
61aa5ba3 1402 if not ie_key and force_generic_extractor:
d22dec74
S
1403 ie_key = 'Generic'
1404
8222d8de 1405 if ie_key:
8b7491c8 1406 ies = {ie_key: self._get_info_extractor_class(ie_key)}
8222d8de
JMF
1407 else:
1408 ies = self._ies
1409
8b7491c8 1410 for ie_key, ie in ies.items():
8222d8de
JMF
1411 if not ie.suitable(url):
1412 continue
1413
1414 if not ie.working():
6febd1c1
PH
1415 self.report_warning('The program functionality for this site has been marked as broken, '
1416 'and will probably not work.')
8222d8de 1417
1151c407 1418 temp_id = ie.get_temp_id(url)
a0566bbf 1419 if temp_id is not None and self.in_download_archive({'id': temp_id, 'ie_key': ie_key}):
5e5be0c0 1420 self.to_screen(f'[{ie_key}] {temp_id}: has already been recorded in the archive')
1421 if self.params.get('break_on_existing', False):
1422 raise ExistingVideoReached()
a0566bbf 1423 break
8b7491c8 1424 return self.__extract_info(url, self.get_info_extractor(ie_key), download, extra_info, process)
a0566bbf 1425 else:
1426 self.report_error('no suitable InfoExtractor for URL %s' % url)
1427
7e88d7d7 1428 def _handle_extraction_exceptions(func):
b5ae35ee 1429 @functools.wraps(func)
a0566bbf 1430 def wrapper(self, *args, **kwargs):
6da22e7d 1431 while True:
1432 try:
1433 return func(self, *args, **kwargs)
1434 except (DownloadCancelled, LazyList.IndexError, PagedList.IndexError):
8222d8de 1435 raise
6da22e7d 1436 except ReExtractInfo as e:
1437 if e.expected:
1438 self.to_screen(f'{e}; Re-extracting data')
1439 else:
1440 self.to_stderr('\r')
1441 self.report_warning(f'{e}; Re-extracting data')
1442 continue
1443 except GeoRestrictedError as e:
1444 msg = e.msg
1445 if e.countries:
1446 msg += '\nThis video is available in %s.' % ', '.join(
1447 map(ISO3166Utils.short2full, e.countries))
1448 msg += '\nYou might want to use a VPN or a proxy server (with --proxy) to workaround.'
1449 self.report_error(msg)
1450 except ExtractorError as e: # An error we somewhat expected
1451 self.report_error(str(e), e.format_traceback())
1452 except Exception as e:
1453 if self.params.get('ignoreerrors'):
1454 self.report_error(str(e), tb=encode_compat_str(traceback.format_exc()))
1455 else:
1456 raise
1457 break
a0566bbf 1458 return wrapper
1459
693f0600 1460 def _wait_for_video(self, ie_result={}):
f2ebc5c7 1461 if (not self.params.get('wait_for_video')
1462 or ie_result.get('_type', 'video') != 'video'
1463 or ie_result.get('formats') or ie_result.get('url')):
1464 return
1465
1466 format_dur = lambda dur: '%02d:%02d:%02d' % timetuple_from_msec(dur * 1000)[:-1]
1467 last_msg = ''
1468
1469 def progress(msg):
1470 nonlocal last_msg
a7dc6a89 1471 full_msg = f'{msg}\n'
1472 if not self.params.get('noprogress'):
1473 full_msg = msg + ' ' * (len(last_msg) - len(msg)) + '\r'
1474 elif last_msg:
1475 return
1476 self.to_screen(full_msg, skip_eol=True)
f2ebc5c7 1477 last_msg = msg
1478
1479 min_wait, max_wait = self.params.get('wait_for_video')
1480 diff = try_get(ie_result, lambda x: x['release_timestamp'] - time.time())
1481 if diff is None and ie_result.get('live_status') == 'is_upcoming':
16c620bc 1482 diff = round(random.uniform(min_wait, max_wait) if (max_wait and min_wait) else (max_wait or min_wait), 0)
f2ebc5c7 1483 self.report_warning('Release time of video is not known')
693f0600 1484 elif ie_result and (diff or 0) <= 0:
f2ebc5c7 1485 self.report_warning('Video should already be available according to extracted info')
38d79fd1 1486 diff = min(max(diff or 0, min_wait or 0), max_wait or float('inf'))
f2ebc5c7 1487 self.to_screen(f'[wait] Waiting for {format_dur(diff)} - Press Ctrl+C to try now')
1488
1489 wait_till = time.time() + diff
1490 try:
1491 while True:
1492 diff = wait_till - time.time()
1493 if diff <= 0:
1494 progress('')
1495 raise ReExtractInfo('[wait] Wait period ended', expected=True)
1496 progress(f'[wait] Remaining time until next attempt: {self._format_screen(format_dur(diff), self.Styles.EMPHASIS)}')
1497 time.sleep(1)
1498 except KeyboardInterrupt:
1499 progress('')
1500 raise ReExtractInfo('[wait] Interrupted by user', expected=True)
1501 except BaseException as e:
1502 if not isinstance(e, ReExtractInfo):
1503 self.to_screen('')
1504 raise
1505
7e88d7d7 1506 @_handle_extraction_exceptions
58f197b7 1507 def __extract_info(self, url, ie, download, extra_info, process):
693f0600 1508 try:
1509 ie_result = ie.extract(url)
1510 except UserNotLive as e:
1511 if process:
1512 if self.params.get('wait_for_video'):
1513 self.report_warning(e)
1514 self._wait_for_video()
1515 raise
a0566bbf 1516 if ie_result is None: # Finished already (backwards compatibility; listformats and friends should be moved here)
cb794ee0 1517 self.report_warning(f'Extractor {ie.IE_NAME} returned nothing{bug_reports_message()}')
a0566bbf 1518 return
1519 if isinstance(ie_result, list):
1520 # Backwards compatibility: old IE result format
1521 ie_result = {
1522 '_type': 'compat_list',
1523 'entries': ie_result,
1524 }
e37d0efb 1525 if extra_info.get('original_url'):
1526 ie_result.setdefault('original_url', extra_info['original_url'])
a0566bbf 1527 self.add_default_extra_info(ie_result, ie, url)
1528 if process:
f2ebc5c7 1529 self._wait_for_video(ie_result)
a0566bbf 1530 return self.process_ie_result(ie_result, download, extra_info)
8222d8de 1531 else:
a0566bbf 1532 return ie_result
fe7e0c98 1533
ea38e55f 1534 def add_default_extra_info(self, ie_result, ie, url):
6033d980 1535 if url is not None:
1536 self.add_extra_info(ie_result, {
1537 'webpage_url': url,
1538 'original_url': url,
57ebfca3 1539 })
1540 webpage_url = ie_result.get('webpage_url')
1541 if webpage_url:
1542 self.add_extra_info(ie_result, {
1543 'webpage_url_basename': url_basename(webpage_url),
1544 'webpage_url_domain': get_domain(webpage_url),
6033d980 1545 })
1546 if ie is not None:
1547 self.add_extra_info(ie_result, {
1548 'extractor': ie.IE_NAME,
1549 'extractor_key': ie.ie_key(),
1550 })
ea38e55f 1551
58adec46 1552 def process_ie_result(self, ie_result, download=True, extra_info=None):
8222d8de
JMF
1553 """
1554 Take the result of the ie(may be modified) and resolve all unresolved
1555 references (URLs, playlist items).
1556
1557 It will also download the videos if 'download'.
1558 Returns the resolved ie_result.
1559 """
58adec46 1560 if extra_info is None:
1561 extra_info = {}
e8ee972c
PH
1562 result_type = ie_result.get('_type', 'video')
1563
057a5206 1564 if result_type in ('url', 'url_transparent'):
134c6ea8 1565 ie_result['url'] = sanitize_url(ie_result['url'])
e37d0efb 1566 if ie_result.get('original_url'):
1567 extra_info.setdefault('original_url', ie_result['original_url'])
1568
057a5206 1569 extract_flat = self.params.get('extract_flat', False)
3089bc74
S
1570 if ((extract_flat == 'in_playlist' and 'playlist' in extra_info)
1571 or extract_flat is True):
ecb54191 1572 info_copy = ie_result.copy()
6033d980 1573 ie = try_get(ie_result.get('ie_key'), self.get_info_extractor)
360167b9 1574 if ie and not ie_result.get('id'):
4614bc22 1575 info_copy['id'] = ie.get_temp_id(ie_result['url'])
6033d980 1576 self.add_default_extra_info(info_copy, ie, ie_result['url'])
4614bc22 1577 self.add_extra_info(info_copy, extra_info)
b5475f11 1578 info_copy, _ = self.pre_process(info_copy)
ecb54191 1579 self.__forced_printings(info_copy, self.prepare_filename(info_copy), incomplete=True)
415f8d51 1580 self._raise_pending_errors(info_copy)
4614bc22 1581 if self.params.get('force_write_download_archive', False):
1582 self.record_download_archive(info_copy)
e8ee972c
PH
1583 return ie_result
1584
8222d8de 1585 if result_type == 'video':
b6c45014 1586 self.add_extra_info(ie_result, extra_info)
9c2b75b5 1587 ie_result = self.process_video_result(ie_result, download=download)
415f8d51 1588 self._raise_pending_errors(ie_result)
28b0eb0f 1589 additional_urls = (ie_result or {}).get('additional_urls')
9c2b75b5 1590 if additional_urls:
e9f4ccd1 1591 # TODO: Improve MetadataParserPP to allow setting a list
14f25df2 1592 if isinstance(additional_urls, str):
9c2b75b5 1593 additional_urls = [additional_urls]
1594 self.to_screen(
1595 '[info] %s: %d additional URL(s) requested' % (ie_result['id'], len(additional_urls)))
1596 self.write_debug('Additional URLs: "%s"' % '", "'.join(additional_urls))
1597 ie_result['additional_entries'] = [
1598 self.extract_info(
b69fd25c 1599 url, download, extra_info=extra_info,
9c2b75b5 1600 force_generic_extractor=self.params.get('force_generic_extractor'))
1601 for url in additional_urls
1602 ]
1603 return ie_result
8222d8de
JMF
1604 elif result_type == 'url':
1605 # We have to add extra_info to the results because it may be
1606 # contained in a playlist
07cce701 1607 return self.extract_info(
1608 ie_result['url'], download,
1609 ie_key=ie_result.get('ie_key'),
1610 extra_info=extra_info)
7fc3fa05
PH
1611 elif result_type == 'url_transparent':
1612 # Use the information from the embedding page
1613 info = self.extract_info(
1614 ie_result['url'], ie_key=ie_result.get('ie_key'),
1615 extra_info=extra_info, download=False, process=False)
1616
1640eb09
S
1617 # extract_info may return None when ignoreerrors is enabled and
1618 # extraction failed with an error, don't crash and return early
1619 # in this case
1620 if not info:
1621 return info
1622
3975b4d2 1623 exempted_fields = {'_type', 'url', 'ie_key'}
1624 if not ie_result.get('section_end') and ie_result.get('section_start') is None:
1625 # For video clips, the id etc of the clip extractor should be used
1626 exempted_fields |= {'id', 'extractor', 'extractor_key'}
1627
412c617d 1628 new_result = info.copy()
3975b4d2 1629 new_result.update(filter_dict(ie_result, lambda k, v: v is not None and k not in exempted_fields))
7fc3fa05 1630
0563f7ac
S
1631 # Extracted info may not be a video result (i.e.
1632 # info.get('_type', 'video') != video) but rather an url or
1633 # url_transparent. In such cases outer metadata (from ie_result)
1634 # should be propagated to inner one (info). For this to happen
1635 # _type of info should be overridden with url_transparent. This
067aa17e 1636 # fixes issue from https://github.com/ytdl-org/youtube-dl/pull/11163.
0563f7ac
S
1637 if new_result.get('_type') == 'url':
1638 new_result['_type'] = 'url_transparent'
7fc3fa05
PH
1639
1640 return self.process_ie_result(
1641 new_result, download=download, extra_info=extra_info)
40fcba5e 1642 elif result_type in ('playlist', 'multi_video'):
30a074c2 1643 # Protect from infinite recursion due to recursively nested playlists
1644 # (see https://github.com/ytdl-org/youtube-dl/issues/27833)
1645 webpage_url = ie_result['webpage_url']
1646 if webpage_url in self._playlist_urls:
7e85e872 1647 self.to_screen(
30a074c2 1648 '[download] Skipping already downloaded playlist: %s'
1649 % ie_result.get('title') or ie_result.get('id'))
1650 return
7e85e872 1651
30a074c2 1652 self._playlist_level += 1
1653 self._playlist_urls.add(webpage_url)
03f83004 1654 self._fill_common_fields(ie_result, False)
bc516a3f 1655 self._sanitize_thumbnails(ie_result)
30a074c2 1656 try:
1657 return self.__process_playlist(ie_result, download)
1658 finally:
1659 self._playlist_level -= 1
1660 if not self._playlist_level:
1661 self._playlist_urls.clear()
8222d8de 1662 elif result_type == 'compat_list':
c9bf4114
PH
1663 self.report_warning(
1664 'Extractor %s returned a compat_list result. '
1665 'It needs to be updated.' % ie_result.get('extractor'))
5f6a1245 1666
8222d8de 1667 def _fixup(r):
b868936c 1668 self.add_extra_info(r, {
1669 'extractor': ie_result['extractor'],
1670 'webpage_url': ie_result['webpage_url'],
1671 'webpage_url_basename': url_basename(ie_result['webpage_url']),
0bb322b9 1672 'webpage_url_domain': get_domain(ie_result['webpage_url']),
b868936c 1673 'extractor_key': ie_result['extractor_key'],
1674 })
8222d8de
JMF
1675 return r
1676 ie_result['entries'] = [
b6c45014 1677 self.process_ie_result(_fixup(r), download, extra_info)
8222d8de
JMF
1678 for r in ie_result['entries']
1679 ]
1680 return ie_result
1681 else:
1682 raise Exception('Invalid result type: %s' % result_type)
1683
e92caff5 1684 def _ensure_dir_exists(self, path):
1685 return make_dir(path, self.report_error)
1686
3b603dbd 1687 @staticmethod
3bec830a 1688 def _playlist_infodict(ie_result, strict=False, **kwargs):
1689 info = {
1690 'playlist_count': ie_result.get('playlist_count'),
3b603dbd 1691 'playlist': ie_result.get('title') or ie_result.get('id'),
1692 'playlist_id': ie_result.get('id'),
1693 'playlist_title': ie_result.get('title'),
1694 'playlist_uploader': ie_result.get('uploader'),
1695 'playlist_uploader_id': ie_result.get('uploader_id'),
3b603dbd 1696 **kwargs,
1697 }
3bec830a 1698 if strict:
1699 return info
1700 return {
1701 **info,
1702 'playlist_index': 0,
1703 '__last_playlist_index': max(ie_result['requested_entries'] or (0, 0)),
1704 'extractor': ie_result['extractor'],
1705 'webpage_url': ie_result['webpage_url'],
1706 'webpage_url_basename': url_basename(ie_result['webpage_url']),
1707 'webpage_url_domain': get_domain(ie_result['webpage_url']),
1708 'extractor_key': ie_result['extractor_key'],
1709 }
3b603dbd 1710
30a074c2 1711 def __process_playlist(self, ie_result, download):
7e88d7d7 1712 """Process each entry in the playlist"""
f5ea4748 1713 assert ie_result['_type'] in ('playlist', 'multi_video')
1714
3bec830a 1715 common_info = self._playlist_infodict(ie_result, strict=True)
3955b207 1716 title = common_info.get('playlist') or '<Untitled>'
3bec830a 1717 if self._match_entry(common_info, incomplete=True) is not None:
1718 return
c6e07cf1 1719 self.to_screen(f'[download] Downloading {ie_result["_type"]}: {title}')
f0d785d3 1720
7e88d7d7 1721 all_entries = PlaylistEntries(self, ie_result)
7e9a6125 1722 entries = orderedSet(all_entries.get_requested_items(), lazy=True)
1723
1724 lazy = self.params.get('lazy_playlist')
1725 if lazy:
1726 resolved_entries, n_entries = [], 'N/A'
1727 ie_result['requested_entries'], ie_result['entries'] = None, None
1728 else:
1729 entries = resolved_entries = list(entries)
1730 n_entries = len(resolved_entries)
1731 ie_result['requested_entries'], ie_result['entries'] = tuple(zip(*resolved_entries)) or ([], [])
1732 if not ie_result.get('playlist_count'):
1733 # Better to do this after potentially exhausting entries
1734 ie_result['playlist_count'] = all_entries.get_full_count()
498f5606 1735
3955b207 1736 ie_copy = collections.ChainMap(
1737 ie_result, self._playlist_infodict(ie_result, n_entries=int_or_none(n_entries)))
3bec830a 1738
e08a85d8 1739 _infojson_written = False
0bfc53d0 1740 write_playlist_files = self.params.get('allow_playlist_files', True)
1741 if write_playlist_files and self.params.get('list_thumbnails'):
1742 self.list_thumbnails(ie_result)
1743 if write_playlist_files and not self.params.get('simulate'):
e08a85d8 1744 _infojson_written = self._write_info_json(
1745 'playlist', ie_result, self.prepare_filename(ie_copy, 'pl_infojson'))
1746 if _infojson_written is None:
80c03fa9 1747 return
1748 if self._write_description('playlist', ie_result,
1749 self.prepare_filename(ie_copy, 'pl_description')) is None:
1750 return
681de68e 1751 # TODO: This should be passed to ThumbnailsConvertor if necessary
3bec830a 1752 self._write_thumbnails('playlist', ie_result, self.prepare_filename(ie_copy, 'pl_thumbnail'))
30a074c2 1753
7e9a6125 1754 if lazy:
1755 if self.params.get('playlistreverse') or self.params.get('playlistrandom'):
1756 self.report_warning('playlistreverse and playlistrandom are not supported with lazy_playlist', only_once=True)
1757 elif self.params.get('playlistreverse'):
1758 entries.reverse()
1759 elif self.params.get('playlistrandom'):
30a074c2 1760 random.shuffle(entries)
1761
7e88d7d7 1762 self.to_screen(f'[{ie_result["extractor"]}] Playlist {title}: Downloading {n_entries} videos'
1763 f'{format_field(ie_result, "playlist_count", " of %s")}')
30a074c2 1764
134c913c 1765 keep_resolved_entries = self.params.get('extract_flat') != 'discard'
1766 if self.params.get('extract_flat') == 'discard_in_playlist':
1767 keep_resolved_entries = ie_result['_type'] != 'playlist'
1768 if keep_resolved_entries:
1769 self.write_debug('The information of all playlist entries will be held in memory')
1770
26e2805c 1771 failures = 0
1772 max_failures = self.params.get('skip_playlist_after_errors') or float('inf')
7e9a6125 1773 for i, (playlist_index, entry) in enumerate(entries):
1774 if lazy:
1775 resolved_entries.append((playlist_index, entry))
3bec830a 1776 if not entry:
7e88d7d7 1777 continue
1778
7e88d7d7 1779 entry['__x_forwarded_for_ip'] = ie_result.get('__x_forwarded_for_ip')
7e9a6125 1780 if not lazy and 'playlist-index' in self.params.get('compat_opts', []):
1781 playlist_index = ie_result['requested_entries'][i]
1782
3bec830a 1783 extra = {
1784 **common_info,
3955b207 1785 'n_entries': int_or_none(n_entries),
71729754 1786 'playlist_index': playlist_index,
7e9a6125 1787 'playlist_autonumber': i + 1,
3bec830a 1788 }
1789
1790 if self._match_entry(collections.ChainMap(entry, extra), incomplete=True) is not None:
1791 continue
1792
1793 self.to_screen('[download] Downloading video %s of %s' % (
1794 self._format_screen(i + 1, self.Styles.ID), self._format_screen(n_entries, self.Styles.EMPHASIS)))
1795
1796 entry_result = self.__process_iterable_entry(entry, download, extra)
26e2805c 1797 if not entry_result:
1798 failures += 1
1799 if failures >= max_failures:
1800 self.report_error(
7e88d7d7 1801 f'Skipping the remaining entries in playlist "{title}" since {failures} items failed extraction')
26e2805c 1802 break
134c913c 1803 if keep_resolved_entries:
1804 resolved_entries[i] = (playlist_index, entry_result)
7e88d7d7 1805
1806 # Update with processed data
7e9a6125 1807 ie_result['requested_entries'], ie_result['entries'] = tuple(zip(*resolved_entries)) or ([], [])
e08a85d8 1808
1809 # Write the updated info to json
cb96c5be 1810 if _infojson_written is True and self._write_info_json(
e08a85d8 1811 'updated playlist', ie_result,
1812 self.prepare_filename(ie_copy, 'pl_infojson'), overwrite=True) is None:
1813 return
ca30f449 1814
ed5835b4 1815 ie_result = self.run_all_pps('playlist', ie_result)
7e88d7d7 1816 self.to_screen(f'[download] Finished downloading playlist: {title}')
30a074c2 1817 return ie_result
1818
7e88d7d7 1819 @_handle_extraction_exceptions
a0566bbf 1820 def __process_iterable_entry(self, entry, download, extra_info):
1821 return self.process_ie_result(
1822 entry, download=download, extra_info=extra_info)
1823
67134eab
JMF
1824 def _build_format_filter(self, filter_spec):
1825 " Returns a function to filter the formats according to the filter_spec "
083c9df9
PH
1826
1827 OPERATORS = {
1828 '<': operator.lt,
1829 '<=': operator.le,
1830 '>': operator.gt,
1831 '>=': operator.ge,
1832 '=': operator.eq,
1833 '!=': operator.ne,
1834 }
67134eab 1835 operator_rex = re.compile(r'''(?x)\s*
187986a8 1836 (?P<key>width|height|tbr|abr|vbr|asr|filesize|filesize_approx|fps)\s*
1837 (?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
1838 (?P<value>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?)\s*
083c9df9 1839 ''' % '|'.join(map(re.escape, OPERATORS.keys())))
187986a8 1840 m = operator_rex.fullmatch(filter_spec)
9ddb6925
S
1841 if m:
1842 try:
1843 comparison_value = int(m.group('value'))
1844 except ValueError:
1845 comparison_value = parse_filesize(m.group('value'))
1846 if comparison_value is None:
1847 comparison_value = parse_filesize(m.group('value') + 'B')
1848 if comparison_value is None:
1849 raise ValueError(
1850 'Invalid value %r in format specification %r' % (
67134eab 1851 m.group('value'), filter_spec))
9ddb6925
S
1852 op = OPERATORS[m.group('op')]
1853
083c9df9 1854 if not m:
9ddb6925
S
1855 STR_OPERATORS = {
1856 '=': operator.eq,
10d33b34
YCH
1857 '^=': lambda attr, value: attr.startswith(value),
1858 '$=': lambda attr, value: attr.endswith(value),
1859 '*=': lambda attr, value: value in attr,
1ce9a3cb 1860 '~=': lambda attr, value: value.search(attr) is not None
9ddb6925 1861 }
187986a8 1862 str_operator_rex = re.compile(r'''(?x)\s*
1863 (?P<key>[a-zA-Z0-9._-]+)\s*
1ce9a3cb
LF
1864 (?P<negation>!\s*)?(?P<op>%s)\s*(?P<none_inclusive>\?\s*)?
1865 (?P<quote>["'])?
1866 (?P<value>(?(quote)(?:(?!(?P=quote))[^\\]|\\.)+|[\w.-]+))
1867 (?(quote)(?P=quote))\s*
9ddb6925 1868 ''' % '|'.join(map(re.escape, STR_OPERATORS.keys())))
187986a8 1869 m = str_operator_rex.fullmatch(filter_spec)
9ddb6925 1870 if m:
1ce9a3cb
LF
1871 if m.group('op') == '~=':
1872 comparison_value = re.compile(m.group('value'))
1873 else:
1874 comparison_value = re.sub(r'''\\([\\"'])''', r'\1', m.group('value'))
2cc779f4
S
1875 str_op = STR_OPERATORS[m.group('op')]
1876 if m.group('negation'):
e118a879 1877 op = lambda attr, value: not str_op(attr, value)
2cc779f4
S
1878 else:
1879 op = str_op
083c9df9 1880
9ddb6925 1881 if not m:
187986a8 1882 raise SyntaxError('Invalid filter specification %r' % filter_spec)
083c9df9
PH
1883
1884 def _filter(f):
1885 actual_value = f.get(m.group('key'))
1886 if actual_value is None:
1887 return m.group('none_inclusive')
1888 return op(actual_value, comparison_value)
67134eab
JMF
1889 return _filter
1890
9f1a1c36 1891 def _check_formats(self, formats):
1892 for f in formats:
1893 self.to_screen('[info] Testing format %s' % f['format_id'])
75689fe5 1894 path = self.get_output_path('temp')
1895 if not self._ensure_dir_exists(f'{path}/'):
1896 continue
1897 temp_file = tempfile.NamedTemporaryFile(suffix='.tmp', delete=False, dir=path or None)
9f1a1c36 1898 temp_file.close()
1899 try:
1900 success, _ = self.dl(temp_file.name, f, test=True)
8a82af35 1901 except (DownloadError, OSError, ValueError) + network_exceptions:
9f1a1c36 1902 success = False
1903 finally:
1904 if os.path.exists(temp_file.name):
1905 try:
1906 os.remove(temp_file.name)
1907 except OSError:
1908 self.report_warning('Unable to delete temporary file "%s"' % temp_file.name)
1909 if success:
1910 yield f
1911 else:
1912 self.to_screen('[info] Unable to download format %s. Skipping...' % f['format_id'])
1913
0017d9ad 1914 def _default_format_spec(self, info_dict, download=True):
0017d9ad 1915
af0f7428
S
1916 def can_merge():
1917 merger = FFmpegMergerPP(self)
1918 return merger.available and merger.can_merge()
1919
91ebc640 1920 prefer_best = (
b7b04c78 1921 not self.params.get('simulate')
91ebc640 1922 and download
1923 and (
1924 not can_merge()
21633673 1925 or info_dict.get('is_live') and not self.params.get('live_from_start')
bf1824b3 1926 or self.params['outtmpl']['default'] == '-'))
53ed7066 1927 compat = (
1928 prefer_best
1929 or self.params.get('allow_multiple_audio_streams', False)
8a82af35 1930 or 'format-spec' in self.params['compat_opts'])
91ebc640 1931
1932 return (
53ed7066 1933 'best/bestvideo+bestaudio' if prefer_best
1934 else 'bestvideo*+bestaudio/best' if not compat
91ebc640 1935 else 'bestvideo+bestaudio/best')
0017d9ad 1936
67134eab
JMF
1937 def build_format_selector(self, format_spec):
1938 def syntax_error(note, start):
1939 message = (
1940 'Invalid format specification: '
86e5f3ed 1941 '{}\n\t{}\n\t{}^'.format(note, format_spec, ' ' * start[1]))
67134eab
JMF
1942 return SyntaxError(message)
1943
1944 PICKFIRST = 'PICKFIRST'
1945 MERGE = 'MERGE'
1946 SINGLE = 'SINGLE'
0130afb7 1947 GROUP = 'GROUP'
67134eab
JMF
1948 FormatSelector = collections.namedtuple('FormatSelector', ['type', 'selector', 'filters'])
1949
91ebc640 1950 allow_multiple_streams = {'audio': self.params.get('allow_multiple_audio_streams', False),
1951 'video': self.params.get('allow_multiple_video_streams', False)}
909d24dd 1952
9f1a1c36 1953 check_formats = self.params.get('check_formats') == 'selected'
e8e73840 1954
67134eab
JMF
1955 def _parse_filter(tokens):
1956 filter_parts = []
1957 for type, string, start, _, _ in tokens:
1958 if type == tokenize.OP and string == ']':
1959 return ''.join(filter_parts)
1960 else:
1961 filter_parts.append(string)
1962
232541df 1963 def _remove_unused_ops(tokens):
17cc1534 1964 # Remove operators that we don't use and join them with the surrounding strings
232541df
JMF
1965 # for example: 'mp4' '-' 'baseline' '-' '16x9' is converted to 'mp4-baseline-16x9'
1966 ALLOWED_OPS = ('/', '+', ',', '(', ')')
1967 last_string, last_start, last_end, last_line = None, None, None, None
1968 for type, string, start, end, line in tokens:
1969 if type == tokenize.OP and string == '[':
1970 if last_string:
1971 yield tokenize.NAME, last_string, last_start, last_end, last_line
1972 last_string = None
1973 yield type, string, start, end, line
1974 # everything inside brackets will be handled by _parse_filter
1975 for type, string, start, end, line in tokens:
1976 yield type, string, start, end, line
1977 if type == tokenize.OP and string == ']':
1978 break
1979 elif type == tokenize.OP and string in ALLOWED_OPS:
1980 if last_string:
1981 yield tokenize.NAME, last_string, last_start, last_end, last_line
1982 last_string = None
1983 yield type, string, start, end, line
1984 elif type in [tokenize.NAME, tokenize.NUMBER, tokenize.OP]:
1985 if not last_string:
1986 last_string = string
1987 last_start = start
1988 last_end = end
1989 else:
1990 last_string += string
1991 if last_string:
1992 yield tokenize.NAME, last_string, last_start, last_end, last_line
1993
cf2ac6df 1994 def _parse_format_selection(tokens, inside_merge=False, inside_choice=False, inside_group=False):
67134eab
JMF
1995 selectors = []
1996 current_selector = None
1997 for type, string, start, _, _ in tokens:
1998 # ENCODING is only defined in python 3.x
1999 if type == getattr(tokenize, 'ENCODING', None):
2000 continue
2001 elif type in [tokenize.NAME, tokenize.NUMBER]:
2002 current_selector = FormatSelector(SINGLE, string, [])
2003 elif type == tokenize.OP:
cf2ac6df
JMF
2004 if string == ')':
2005 if not inside_group:
2006 # ')' will be handled by the parentheses group
2007 tokens.restore_last_token()
67134eab 2008 break
cf2ac6df 2009 elif inside_merge and string in ['/', ',']:
0130afb7
JMF
2010 tokens.restore_last_token()
2011 break
cf2ac6df
JMF
2012 elif inside_choice and string == ',':
2013 tokens.restore_last_token()
2014 break
2015 elif string == ',':
0a31a350
JMF
2016 if not current_selector:
2017 raise syntax_error('"," must follow a format selector', start)
67134eab
JMF
2018 selectors.append(current_selector)
2019 current_selector = None
2020 elif string == '/':
d96d604e
JMF
2021 if not current_selector:
2022 raise syntax_error('"/" must follow a format selector', start)
67134eab 2023 first_choice = current_selector
cf2ac6df 2024 second_choice = _parse_format_selection(tokens, inside_choice=True)
f5f4a27a 2025 current_selector = FormatSelector(PICKFIRST, (first_choice, second_choice), [])
67134eab
JMF
2026 elif string == '[':
2027 if not current_selector:
2028 current_selector = FormatSelector(SINGLE, 'best', [])
2029 format_filter = _parse_filter(tokens)
2030 current_selector.filters.append(format_filter)
0130afb7
JMF
2031 elif string == '(':
2032 if current_selector:
2033 raise syntax_error('Unexpected "("', start)
cf2ac6df
JMF
2034 group = _parse_format_selection(tokens, inside_group=True)
2035 current_selector = FormatSelector(GROUP, group, [])
67134eab 2036 elif string == '+':
d03cfdce 2037 if not current_selector:
2038 raise syntax_error('Unexpected "+"', start)
2039 selector_1 = current_selector
2040 selector_2 = _parse_format_selection(tokens, inside_merge=True)
2041 if not selector_2:
2042 raise syntax_error('Expected a selector', start)
2043 current_selector = FormatSelector(MERGE, (selector_1, selector_2), [])
67134eab 2044 else:
86e5f3ed 2045 raise syntax_error(f'Operator not recognized: "{string}"', start)
67134eab
JMF
2046 elif type == tokenize.ENDMARKER:
2047 break
2048 if current_selector:
2049 selectors.append(current_selector)
2050 return selectors
2051
f8d4ad9a 2052 def _merge(formats_pair):
2053 format_1, format_2 = formats_pair
2054
2055 formats_info = []
2056 formats_info.extend(format_1.get('requested_formats', (format_1,)))
2057 formats_info.extend(format_2.get('requested_formats', (format_2,)))
2058
2059 if not allow_multiple_streams['video'] or not allow_multiple_streams['audio']:
551f9388 2060 get_no_more = {'video': False, 'audio': False}
f8d4ad9a 2061 for (i, fmt_info) in enumerate(formats_info):
551f9388 2062 if fmt_info.get('acodec') == fmt_info.get('vcodec') == 'none':
2063 formats_info.pop(i)
2064 continue
2065 for aud_vid in ['audio', 'video']:
f8d4ad9a 2066 if not allow_multiple_streams[aud_vid] and fmt_info.get(aud_vid[0] + 'codec') != 'none':
2067 if get_no_more[aud_vid]:
2068 formats_info.pop(i)
f5510afe 2069 break
f8d4ad9a 2070 get_no_more[aud_vid] = True
2071
2072 if len(formats_info) == 1:
2073 return formats_info[0]
2074
2075 video_fmts = [fmt_info for fmt_info in formats_info if fmt_info.get('vcodec') != 'none']
2076 audio_fmts = [fmt_info for fmt_info in formats_info if fmt_info.get('acodec') != 'none']
2077
2078 the_only_video = video_fmts[0] if len(video_fmts) == 1 else None
2079 the_only_audio = audio_fmts[0] if len(audio_fmts) == 1 else None
2080
2081 output_ext = self.params.get('merge_output_format')
2082 if not output_ext:
2083 if the_only_video:
2084 output_ext = the_only_video['ext']
2085 elif the_only_audio and not video_fmts:
2086 output_ext = the_only_audio['ext']
2087 else:
2088 output_ext = 'mkv'
2089
975a0d0d 2090 filtered = lambda *keys: filter(None, (traverse_obj(fmt, *keys) for fmt in formats_info))
2091
f8d4ad9a 2092 new_dict = {
2093 'requested_formats': formats_info,
975a0d0d 2094 'format': '+'.join(filtered('format')),
2095 'format_id': '+'.join(filtered('format_id')),
f8d4ad9a 2096 'ext': output_ext,
975a0d0d 2097 'protocol': '+'.join(map(determine_protocol, formats_info)),
093a1710 2098 'language': '+'.join(orderedSet(filtered('language'))) or None,
2099 'format_note': '+'.join(orderedSet(filtered('format_note'))) or None,
2100 'filesize_approx': sum(filtered('filesize', 'filesize_approx')) or None,
975a0d0d 2101 'tbr': sum(filtered('tbr', 'vbr', 'abr')),
f8d4ad9a 2102 }
2103
2104 if the_only_video:
2105 new_dict.update({
2106 'width': the_only_video.get('width'),
2107 'height': the_only_video.get('height'),
2108 'resolution': the_only_video.get('resolution') or self.format_resolution(the_only_video),
2109 'fps': the_only_video.get('fps'),
49a57e70 2110 'dynamic_range': the_only_video.get('dynamic_range'),
f8d4ad9a 2111 'vcodec': the_only_video.get('vcodec'),
2112 'vbr': the_only_video.get('vbr'),
2113 'stretched_ratio': the_only_video.get('stretched_ratio'),
2114 })
2115
2116 if the_only_audio:
2117 new_dict.update({
2118 'acodec': the_only_audio.get('acodec'),
2119 'abr': the_only_audio.get('abr'),
975a0d0d 2120 'asr': the_only_audio.get('asr'),
f8d4ad9a 2121 })
2122
2123 return new_dict
2124
e8e73840 2125 def _check_formats(formats):
981052c9 2126 if not check_formats:
2127 yield from formats
b5ac45b1 2128 return
9f1a1c36 2129 yield from self._check_formats(formats)
e8e73840 2130
67134eab 2131 def _build_selector_function(selector):
909d24dd 2132 if isinstance(selector, list): # ,
67134eab
JMF
2133 fs = [_build_selector_function(s) for s in selector]
2134
317f7ab6 2135 def selector_function(ctx):
67134eab 2136 for f in fs:
981052c9 2137 yield from f(ctx)
67134eab 2138 return selector_function
909d24dd 2139
2140 elif selector.type == GROUP: # ()
0130afb7 2141 selector_function = _build_selector_function(selector.selector)
909d24dd 2142
2143 elif selector.type == PICKFIRST: # /
67134eab
JMF
2144 fs = [_build_selector_function(s) for s in selector.selector]
2145
317f7ab6 2146 def selector_function(ctx):
67134eab 2147 for f in fs:
317f7ab6 2148 picked_formats = list(f(ctx))
67134eab
JMF
2149 if picked_formats:
2150 return picked_formats
2151 return []
67134eab 2152
981052c9 2153 elif selector.type == MERGE: # +
2154 selector_1, selector_2 = map(_build_selector_function, selector.selector)
2155
2156 def selector_function(ctx):
adbc4ec4 2157 for pair in itertools.product(selector_1(ctx), selector_2(ctx)):
981052c9 2158 yield _merge(pair)
2159
909d24dd 2160 elif selector.type == SINGLE: # atom
598d185d 2161 format_spec = selector.selector or 'best'
909d24dd 2162
f8d4ad9a 2163 # TODO: Add allvideo, allaudio etc by generalizing the code with best/worst selector
909d24dd 2164 if format_spec == 'all':
2165 def selector_function(ctx):
9222c381 2166 yield from _check_formats(ctx['formats'][::-1])
f8d4ad9a 2167 elif format_spec == 'mergeall':
2168 def selector_function(ctx):
316f2650 2169 formats = list(_check_formats(
2170 f for f in ctx['formats'] if f.get('vcodec') != 'none' or f.get('acodec') != 'none'))
e01d6aa4 2171 if not formats:
2172 return
921b76ca 2173 merged_format = formats[-1]
2174 for f in formats[-2::-1]:
f8d4ad9a 2175 merged_format = _merge((merged_format, f))
2176 yield merged_format
909d24dd 2177
2178 else:
85e801a9 2179 format_fallback, seperate_fallback, format_reverse, format_idx = False, None, True, 1
eff63539 2180 mobj = re.match(
2181 r'(?P<bw>best|worst|b|w)(?P<type>video|audio|v|a)?(?P<mod>\*)?(?:\.(?P<n>[1-9]\d*))?$',
2182 format_spec)
2183 if mobj is not None:
2184 format_idx = int_or_none(mobj.group('n'), default=1)
e8e73840 2185 format_reverse = mobj.group('bw')[0] == 'b'
eff63539 2186 format_type = (mobj.group('type') or [None])[0]
2187 not_format_type = {'v': 'a', 'a': 'v'}.get(format_type)
2188 format_modified = mobj.group('mod') is not None
909d24dd 2189
2190 format_fallback = not format_type and not format_modified # for b, w
8326b00a 2191 _filter_f = (
eff63539 2192 (lambda f: f.get('%scodec' % format_type) != 'none')
2193 if format_type and format_modified # bv*, ba*, wv*, wa*
2194 else (lambda f: f.get('%scodec' % not_format_type) == 'none')
2195 if format_type # bv, ba, wv, wa
2196 else (lambda f: f.get('vcodec') != 'none' and f.get('acodec') != 'none')
2197 if not format_modified # b, w
8326b00a 2198 else lambda f: True) # b*, w*
2199 filter_f = lambda f: _filter_f(f) and (
2200 f.get('vcodec') != 'none' or f.get('acodec') != 'none')
67134eab 2201 else:
48ee10ee 2202 if format_spec in self._format_selection_exts['audio']:
b11c04a8 2203 filter_f = lambda f: f.get('ext') == format_spec and f.get('acodec') != 'none'
48ee10ee 2204 elif format_spec in self._format_selection_exts['video']:
b11c04a8 2205 filter_f = lambda f: f.get('ext') == format_spec and f.get('acodec') != 'none' and f.get('vcodec') != 'none'
85e801a9 2206 seperate_fallback = lambda f: f.get('ext') == format_spec and f.get('vcodec') != 'none'
48ee10ee 2207 elif format_spec in self._format_selection_exts['storyboards']:
b11c04a8 2208 filter_f = lambda f: f.get('ext') == format_spec and f.get('acodec') == 'none' and f.get('vcodec') == 'none'
2209 else:
b5ae35ee 2210 filter_f = lambda f: f.get('format_id') == format_spec # id
909d24dd 2211
2212 def selector_function(ctx):
2213 formats = list(ctx['formats'])
909d24dd 2214 matches = list(filter(filter_f, formats)) if filter_f is not None else formats
85e801a9 2215 if not matches:
2216 if format_fallback and ctx['incomplete_formats']:
2217 # for extractors with incomplete formats (audio only (soundcloud)
2218 # or video only (imgur)) best/worst will fallback to
2219 # best/worst {video,audio}-only format
2220 matches = formats
2221 elif seperate_fallback and not ctx['has_merged_format']:
2222 # for compatibility with youtube-dl when there is no pre-merged format
2223 matches = list(filter(seperate_fallback, formats))
981052c9 2224 matches = LazyList(_check_formats(matches[::-1 if format_reverse else 1]))
2225 try:
e8e73840 2226 yield matches[format_idx - 1]
4abea8ca 2227 except LazyList.IndexError:
981052c9 2228 return
083c9df9 2229
67134eab 2230 filters = [self._build_format_filter(f) for f in selector.filters]
083c9df9 2231
317f7ab6 2232 def final_selector(ctx):
adbc4ec4 2233 ctx_copy = dict(ctx)
67134eab 2234 for _filter in filters:
317f7ab6
S
2235 ctx_copy['formats'] = list(filter(_filter, ctx_copy['formats']))
2236 return selector_function(ctx_copy)
67134eab 2237 return final_selector
083c9df9 2238
0f06bcd7 2239 stream = io.BytesIO(format_spec.encode())
0130afb7 2240 try:
f9934b96 2241 tokens = list(_remove_unused_ops(tokenize.tokenize(stream.readline)))
0130afb7
JMF
2242 except tokenize.TokenError:
2243 raise syntax_error('Missing closing/opening brackets or parenthesis', (0, len(format_spec)))
2244
86e5f3ed 2245 class TokenIterator:
0130afb7
JMF
2246 def __init__(self, tokens):
2247 self.tokens = tokens
2248 self.counter = 0
2249
2250 def __iter__(self):
2251 return self
2252
2253 def __next__(self):
2254 if self.counter >= len(self.tokens):
2255 raise StopIteration()
2256 value = self.tokens[self.counter]
2257 self.counter += 1
2258 return value
2259
2260 next = __next__
2261
2262 def restore_last_token(self):
2263 self.counter -= 1
2264
2265 parsed_selector = _parse_format_selection(iter(TokenIterator(tokens)))
67134eab 2266 return _build_selector_function(parsed_selector)
a9c58ad9 2267
e5660ee6 2268 def _calc_headers(self, info_dict):
8b7539d2 2269 res = merge_headers(self.params['http_headers'], info_dict.get('http_headers') or {})
e5660ee6 2270
c487cf00 2271 cookies = self._calc_cookies(info_dict['url'])
e5660ee6
JMF
2272 if cookies:
2273 res['Cookie'] = cookies
2274
0016b84e
S
2275 if 'X-Forwarded-For' not in res:
2276 x_forwarded_for_ip = info_dict.get('__x_forwarded_for_ip')
2277 if x_forwarded_for_ip:
2278 res['X-Forwarded-For'] = x_forwarded_for_ip
2279
e5660ee6
JMF
2280 return res
2281
c487cf00 2282 def _calc_cookies(self, url):
2283 pr = sanitized_Request(url)
e5660ee6 2284 self.cookiejar.add_cookie_header(pr)
662435f7 2285 return pr.get_header('Cookie')
e5660ee6 2286
9f1a1c36 2287 def _sort_thumbnails(self, thumbnails):
2288 thumbnails.sort(key=lambda t: (
2289 t.get('preference') if t.get('preference') is not None else -1,
2290 t.get('width') if t.get('width') is not None else -1,
2291 t.get('height') if t.get('height') is not None else -1,
2292 t.get('id') if t.get('id') is not None else '',
2293 t.get('url')))
2294
b0249bca 2295 def _sanitize_thumbnails(self, info_dict):
bc516a3f 2296 thumbnails = info_dict.get('thumbnails')
2297 if thumbnails is None:
2298 thumbnail = info_dict.get('thumbnail')
2299 if thumbnail:
2300 info_dict['thumbnails'] = thumbnails = [{'url': thumbnail}]
9f1a1c36 2301 if not thumbnails:
2302 return
2303
2304 def check_thumbnails(thumbnails):
2305 for t in thumbnails:
2306 self.to_screen(f'[info] Testing thumbnail {t["id"]}')
2307 try:
2308 self.urlopen(HEADRequest(t['url']))
2309 except network_exceptions as err:
2310 self.to_screen(f'[info] Unable to connect to thumbnail {t["id"]} URL {t["url"]!r} - {err}. Skipping...')
2311 continue
2312 yield t
2313
2314 self._sort_thumbnails(thumbnails)
2315 for i, t in enumerate(thumbnails):
2316 if t.get('id') is None:
2317 t['id'] = '%d' % i
2318 if t.get('width') and t.get('height'):
2319 t['resolution'] = '%dx%d' % (t['width'], t['height'])
2320 t['url'] = sanitize_url(t['url'])
2321
2322 if self.params.get('check_formats') is True:
282f5709 2323 info_dict['thumbnails'] = LazyList(check_thumbnails(thumbnails[::-1]), reverse=True)
9f1a1c36 2324 else:
2325 info_dict['thumbnails'] = thumbnails
bc516a3f 2326
03f83004
LNO
2327 def _fill_common_fields(self, info_dict, is_video=True):
2328 # TODO: move sanitization here
2329 if is_video:
2330 # playlists are allowed to lack "title"
d4736fdb 2331 title = info_dict.get('title', NO_DEFAULT)
2332 if title is NO_DEFAULT:
03f83004
LNO
2333 raise ExtractorError('Missing "title" field in extractor result',
2334 video_id=info_dict['id'], ie=info_dict['extractor'])
d4736fdb 2335 info_dict['fulltitle'] = title
2336 if not title:
2337 if title == '':
2338 self.write_debug('Extractor gave empty title. Creating a generic title')
2339 else:
2340 self.report_warning('Extractor failed to obtain "title". Creating a generic title instead')
1d485a1a 2341 info_dict['title'] = f'{info_dict["extractor"].replace(":", "-")} video #{info_dict["id"]}'
03f83004
LNO
2342
2343 if info_dict.get('duration') is not None:
2344 info_dict['duration_string'] = formatSeconds(info_dict['duration'])
2345
2346 for ts_key, date_key in (
2347 ('timestamp', 'upload_date'),
2348 ('release_timestamp', 'release_date'),
2349 ('modified_timestamp', 'modified_date'),
2350 ):
2351 if info_dict.get(date_key) is None and info_dict.get(ts_key) is not None:
2352 # Working around out-of-range timestamp values (e.g. negative ones on Windows,
2353 # see http://bugs.python.org/issue1646728)
19a03940 2354 with contextlib.suppress(ValueError, OverflowError, OSError):
03f83004
LNO
2355 upload_date = datetime.datetime.utcfromtimestamp(info_dict[ts_key])
2356 info_dict[date_key] = upload_date.strftime('%Y%m%d')
03f83004
LNO
2357
2358 live_keys = ('is_live', 'was_live')
2359 live_status = info_dict.get('live_status')
2360 if live_status is None:
2361 for key in live_keys:
2362 if info_dict.get(key) is False:
2363 continue
2364 if info_dict.get(key):
2365 live_status = key
2366 break
2367 if all(info_dict.get(key) is False for key in live_keys):
2368 live_status = 'not_live'
2369 if live_status:
2370 info_dict['live_status'] = live_status
2371 for key in live_keys:
2372 if info_dict.get(key) is None:
2373 info_dict[key] = (live_status == key)
2374
2375 # Auto generate title fields corresponding to the *_number fields when missing
2376 # in order to always have clean titles. This is very common for TV series.
2377 for field in ('chapter', 'season', 'episode'):
2378 if info_dict.get('%s_number' % field) is not None and not info_dict.get(field):
2379 info_dict[field] = '%s %d' % (field.capitalize(), info_dict['%s_number' % field])
2380
415f8d51 2381 def _raise_pending_errors(self, info):
2382 err = info.pop('__pending_error', None)
2383 if err:
2384 self.report_error(err, tb=False)
2385
dd82ffea
JMF
2386 def process_video_result(self, info_dict, download=True):
2387 assert info_dict.get('_type', 'video') == 'video'
9c906919 2388 self._num_videos += 1
dd82ffea 2389
bec1fad2 2390 if 'id' not in info_dict:
fc08bdd6 2391 raise ExtractorError('Missing "id" field in extractor result', ie=info_dict['extractor'])
2392 elif not info_dict.get('id'):
2393 raise ExtractorError('Extractor failed to obtain "id"', ie=info_dict['extractor'])
455a15e2 2394
c9969434
S
2395 def report_force_conversion(field, field_not, conversion):
2396 self.report_warning(
2397 '"%s" field is not %s - forcing %s conversion, there is an error in extractor'
2398 % (field, field_not, conversion))
2399
2400 def sanitize_string_field(info, string_field):
2401 field = info.get(string_field)
14f25df2 2402 if field is None or isinstance(field, str):
c9969434
S
2403 return
2404 report_force_conversion(string_field, 'a string', 'string')
14f25df2 2405 info[string_field] = str(field)
c9969434
S
2406
2407 def sanitize_numeric_fields(info):
2408 for numeric_field in self._NUMERIC_FIELDS:
2409 field = info.get(numeric_field)
f9934b96 2410 if field is None or isinstance(field, (int, float)):
c9969434
S
2411 continue
2412 report_force_conversion(numeric_field, 'numeric', 'int')
2413 info[numeric_field] = int_or_none(field)
2414
2415 sanitize_string_field(info_dict, 'id')
2416 sanitize_numeric_fields(info_dict)
3975b4d2 2417 if info_dict.get('section_end') and info_dict.get('section_start') is not None:
2418 info_dict['duration'] = round(info_dict['section_end'] - info_dict['section_start'], 3)
4c3f8c3f 2419 if (info_dict.get('duration') or 0) <= 0 and info_dict.pop('duration', None):
50e93e03 2420 self.report_warning('"duration" field is negative, there is an error in extractor')
be6217b2 2421
9eef7c4e 2422 chapters = info_dict.get('chapters') or []
a3976e07 2423 if chapters and chapters[0].get('start_time'):
2424 chapters.insert(0, {'start_time': 0})
2425
9eef7c4e 2426 dummy_chapter = {'end_time': 0, 'start_time': info_dict.get('duration')}
a3976e07 2427 for idx, (prev, current, next_) in enumerate(zip(
2428 (dummy_chapter, *chapters), chapters, (*chapters[1:], dummy_chapter)), 1):
9eef7c4e 2429 if current.get('start_time') is None:
2430 current['start_time'] = prev.get('end_time')
2431 if not current.get('end_time'):
2432 current['end_time'] = next_.get('start_time')
a3976e07 2433 if not current.get('title'):
2434 current['title'] = f'<Untitled Chapter {idx}>'
9eef7c4e 2435
dd82ffea
JMF
2436 if 'playlist' not in info_dict:
2437 # It isn't part of a playlist
2438 info_dict['playlist'] = None
2439 info_dict['playlist_index'] = None
2440
bc516a3f 2441 self._sanitize_thumbnails(info_dict)
d5519808 2442
536a55da 2443 thumbnail = info_dict.get('thumbnail')
bc516a3f 2444 thumbnails = info_dict.get('thumbnails')
536a55da
S
2445 if thumbnail:
2446 info_dict['thumbnail'] = sanitize_url(thumbnail)
2447 elif thumbnails:
d5519808
PH
2448 info_dict['thumbnail'] = thumbnails[-1]['url']
2449
ae30b840 2450 if info_dict.get('display_id') is None and 'id' in info_dict:
0afef30b
PH
2451 info_dict['display_id'] = info_dict['id']
2452
03f83004 2453 self._fill_common_fields(info_dict)
33d2fc2f 2454
05108a49
S
2455 for cc_kind in ('subtitles', 'automatic_captions'):
2456 cc = info_dict.get(cc_kind)
2457 if cc:
2458 for _, subtitle in cc.items():
2459 for subtitle_format in subtitle:
2460 if subtitle_format.get('url'):
2461 subtitle_format['url'] = sanitize_url(subtitle_format['url'])
2462 if subtitle_format.get('ext') is None:
2463 subtitle_format['ext'] = determine_ext(subtitle_format['url']).lower()
2464
2465 automatic_captions = info_dict.get('automatic_captions')
4bba3716 2466 subtitles = info_dict.get('subtitles')
4bba3716 2467
360e1ca5 2468 info_dict['requested_subtitles'] = self.process_subtitles(
05108a49 2469 info_dict['id'], subtitles, automatic_captions)
a504ced0 2470
dd82ffea
JMF
2471 if info_dict.get('formats') is None:
2472 # There's only one format available
2473 formats = [info_dict]
2474 else:
2475 formats = info_dict['formats']
2476
0a5a191a 2477 # or None ensures --clean-infojson removes it
2478 info_dict['_has_drm'] = any(f.get('has_drm') for f in formats) or None
88acdbc2 2479 if not self.params.get('allow_unplayable_formats'):
2480 formats = [f for f in formats if not f.get('has_drm')]
0a5a191a 2481 if info_dict['_has_drm'] and all(
c0b6e5c7 2482 f.get('acodec') == f.get('vcodec') == 'none' for f in formats):
2483 self.report_warning(
2484 'This video is DRM protected and only images are available for download. '
2485 'Use --list-formats to see them')
88acdbc2 2486
319b6059 2487 get_from_start = not info_dict.get('is_live') or bool(self.params.get('live_from_start'))
2488 if not get_from_start:
2489 info_dict['title'] += ' ' + datetime.datetime.now().strftime('%Y-%m-%d %H:%M')
2490 if info_dict.get('is_live') and formats:
adbc4ec4 2491 formats = [f for f in formats if bool(f.get('is_from_start')) == get_from_start]
319b6059 2492 if get_from_start and not formats:
a44ca5a4 2493 self.raise_no_formats(info_dict, msg=(
2494 '--live-from-start is passed, but there are no formats that can be downloaded from the start. '
2495 'If you want to download from the current time, use --no-live-from-start'))
adbc4ec4 2496
db95dc13 2497 if not formats:
1151c407 2498 self.raise_no_formats(info_dict)
db95dc13 2499
73af5cc8
S
2500 def is_wellformed(f):
2501 url = f.get('url')
a5ac0c47 2502 if not url:
73af5cc8
S
2503 self.report_warning(
2504 '"url" field is missing or empty - skipping format, '
2505 'there is an error in extractor')
a5ac0c47
S
2506 return False
2507 if isinstance(url, bytes):
2508 sanitize_string_field(f, 'url')
2509 return True
73af5cc8
S
2510
2511 # Filter out malformed formats for better extraction robustness
2512 formats = list(filter(is_wellformed, formats))
2513
181c7053
S
2514 formats_dict = {}
2515
dd82ffea 2516 # We check that all the formats have the format and format_id fields
db95dc13 2517 for i, format in enumerate(formats):
c9969434
S
2518 sanitize_string_field(format, 'format_id')
2519 sanitize_numeric_fields(format)
dcf77cf1 2520 format['url'] = sanitize_url(format['url'])
e74e3b63 2521 if not format.get('format_id'):
14f25df2 2522 format['format_id'] = str(i)
e2effb08
S
2523 else:
2524 # Sanitize format_id from characters used in format selector expression
ec85ded8 2525 format['format_id'] = re.sub(r'[\s,/+\[\]()]', '_', format['format_id'])
181c7053
S
2526 format_id = format['format_id']
2527 if format_id not in formats_dict:
2528 formats_dict[format_id] = []
2529 formats_dict[format_id].append(format)
2530
2531 # Make sure all formats have unique format_id
03b4de72 2532 common_exts = set(itertools.chain(*self._format_selection_exts.values()))
181c7053 2533 for format_id, ambiguous_formats in formats_dict.items():
48ee10ee 2534 ambigious_id = len(ambiguous_formats) > 1
2535 for i, format in enumerate(ambiguous_formats):
2536 if ambigious_id:
181c7053 2537 format['format_id'] = '%s-%d' % (format_id, i)
48ee10ee 2538 if format.get('ext') is None:
2539 format['ext'] = determine_ext(format['url']).lower()
2540 # Ensure there is no conflict between id and ext in format selection
2541 # See https://github.com/yt-dlp/yt-dlp/issues/1282
2542 if format['format_id'] != format['ext'] and format['format_id'] in common_exts:
2543 format['format_id'] = 'f%s' % format['format_id']
181c7053
S
2544
2545 for i, format in enumerate(formats):
8c51aa65 2546 if format.get('format') is None:
6febd1c1 2547 format['format'] = '{id} - {res}{note}'.format(
8c51aa65
JMF
2548 id=format['format_id'],
2549 res=self.format_resolution(format),
b868936c 2550 note=format_field(format, 'format_note', ' (%s)'),
8c51aa65 2551 )
6f0be937 2552 if format.get('protocol') is None:
b5559424 2553 format['protocol'] = determine_protocol(format)
239df021 2554 if format.get('resolution') is None:
2555 format['resolution'] = self.format_resolution(format, default=None)
176f1866 2556 if format.get('dynamic_range') is None and format.get('vcodec') != 'none':
2557 format['dynamic_range'] = 'SDR'
f2fe69c7 2558 if (info_dict.get('duration') and format.get('tbr')
2559 and not format.get('filesize') and not format.get('filesize_approx')):
56ba69e4 2560 format['filesize_approx'] = int(info_dict['duration'] * format['tbr'] * (1024 / 8))
f2fe69c7 2561
e5660ee6
JMF
2562 # Add HTTP headers, so that external programs can use them from the
2563 # json output
2564 full_format_info = info_dict.copy()
2565 full_format_info.update(format)
2566 format['http_headers'] = self._calc_headers(full_format_info)
0016b84e
S
2567 # Remove private housekeeping stuff
2568 if '__x_forwarded_for_ip' in info_dict:
2569 del info_dict['__x_forwarded_for_ip']
dd82ffea 2570
9f1a1c36 2571 if self.params.get('check_formats') is True:
282f5709 2572 formats = LazyList(self._check_formats(formats[::-1]), reverse=True)
9f1a1c36 2573
88acdbc2 2574 if not formats or formats[0] is not info_dict:
b3d9ef88
JMF
2575 # only set the 'formats' fields if the original info_dict list them
2576 # otherwise we end up with a circular reference, the first (and unique)
f89197d7 2577 # element in the 'formats' field in info_dict is info_dict itself,
dfb1b146 2578 # which can't be exported to json
b3d9ef88 2579 info_dict['formats'] = formats
4ec82a72 2580
2581 info_dict, _ = self.pre_process(info_dict)
2582
6db9c4d5 2583 if self._match_entry(info_dict, incomplete=self._format_fields) is not None:
09b49e1f 2584 return info_dict
2585
2586 self.post_extract(info_dict)
2587 info_dict, _ = self.pre_process(info_dict, 'after_filter')
2588
093a1710 2589 # The pre-processors may have modified the formats
2590 formats = info_dict.get('formats', [info_dict])
2591
fa9f30b8 2592 list_only = self.params.get('simulate') is None and (
2593 self.params.get('list_thumbnails') or self.params.get('listformats') or self.params.get('listsubtitles'))
2594 interactive_format_selection = not list_only and self.format_selector == '-'
b7b04c78 2595 if self.params.get('list_thumbnails'):
2596 self.list_thumbnails(info_dict)
b7b04c78 2597 if self.params.get('listsubtitles'):
2598 if 'automatic_captions' in info_dict:
2599 self.list_subtitles(
2600 info_dict['id'], automatic_captions, 'automatic captions')
2601 self.list_subtitles(info_dict['id'], subtitles, 'subtitles')
fa9f30b8 2602 if self.params.get('listformats') or interactive_format_selection:
b69fd25c 2603 self.list_formats(info_dict)
169dbde9 2604 if list_only:
b7b04c78 2605 # Without this printing, -F --print-json will not work
169dbde9 2606 self.__forced_printings(info_dict, self.prepare_filename(info_dict), incomplete=True)
c487cf00 2607 return info_dict
bfaae0a7 2608
187986a8 2609 format_selector = self.format_selector
2610 if format_selector is None:
0017d9ad 2611 req_format = self._default_format_spec(info_dict, download=download)
0760b0a7 2612 self.write_debug('Default format spec: %s' % req_format)
187986a8 2613 format_selector = self.build_format_selector(req_format)
317f7ab6 2614
fa9f30b8 2615 while True:
2616 if interactive_format_selection:
2617 req_format = input(
2618 self._format_screen('\nEnter format selector: ', self.Styles.EMPHASIS))
2619 try:
2620 format_selector = self.build_format_selector(req_format)
2621 except SyntaxError as err:
2622 self.report_error(err, tb=False, is_error=False)
2623 continue
2624
85e801a9 2625 formats_to_download = list(format_selector({
fa9f30b8 2626 'formats': formats,
85e801a9 2627 'has_merged_format': any('none' not in (f.get('acodec'), f.get('vcodec')) for f in formats),
2628 'incomplete_formats': (
2629 # All formats are video-only or
2630 all(f.get('vcodec') != 'none' and f.get('acodec') == 'none' for f in formats)
2631 # all formats are audio-only
2632 or all(f.get('vcodec') == 'none' and f.get('acodec') != 'none' for f in formats)),
2633 }))
fa9f30b8 2634 if interactive_format_selection and not formats_to_download:
2635 self.report_error('Requested format is not available', tb=False, is_error=False)
2636 continue
2637 break
317f7ab6 2638
dd82ffea 2639 if not formats_to_download:
b7da73eb 2640 if not self.params.get('ignore_no_formats_error'):
c0b6e5c7 2641 raise ExtractorError(
2642 'Requested format is not available. Use --list-formats for a list of available formats',
2643 expected=True, video_id=info_dict['id'], ie=info_dict['extractor'])
b62fa6d7 2644 self.report_warning('Requested format is not available')
2645 # Process what we can, even without any available formats.
2646 formats_to_download = [{}]
a13e6848 2647
5ec1b6b7 2648 requested_ranges = self.params.get('download_ranges')
2649 if requested_ranges:
2650 requested_ranges = tuple(requested_ranges(info_dict, self))
2651
2652 best_format, downloaded_formats = formats_to_download[-1], []
b62fa6d7 2653 if download:
2654 if best_format:
5ec1b6b7 2655 def to_screen(*msg):
2656 self.to_screen(f'[info] {info_dict["id"]}: {" ".join(", ".join(variadic(m)) for m in msg)}')
2657
2658 to_screen(f'Downloading {len(formats_to_download)} format(s):',
2659 (f['format_id'] for f in formats_to_download))
2660 if requested_ranges:
2661 to_screen(f'Downloading {len(requested_ranges)} time ranges:',
2662 (f'{int(c["start_time"])}-{int(c["end_time"])}' for c in requested_ranges))
a13e6848 2663 max_downloads_reached = False
5ec1b6b7 2664
2665 for fmt, chapter in itertools.product(formats_to_download, requested_ranges or [{}]):
2666 new_info = self._copy_infodict(info_dict)
b7da73eb 2667 new_info.update(fmt)
3975b4d2 2668 offset, duration = info_dict.get('section_start') or 0, info_dict.get('duration') or float('inf')
2669 if chapter or offset:
5ec1b6b7 2670 new_info.update({
3975b4d2 2671 'section_start': offset + chapter.get('start_time', 0),
bc401608 2672 'section_end': offset + min(chapter.get('end_time', duration), duration),
5ec1b6b7 2673 'section_title': chapter.get('title'),
2674 'section_number': chapter.get('index'),
2675 })
2676 downloaded_formats.append(new_info)
a13e6848 2677 try:
2678 self.process_info(new_info)
2679 except MaxDownloadsReached:
2680 max_downloads_reached = True
415f8d51 2681 self._raise_pending_errors(new_info)
f46e2f9d 2682 # Remove copied info
2683 for key, val in tuple(new_info.items()):
2684 if info_dict.get(key) == val:
2685 new_info.pop(key)
a13e6848 2686 if max_downloads_reached:
2687 break
ebed8b37 2688
5ec1b6b7 2689 write_archive = {f.get('__write_download_archive', False) for f in downloaded_formats}
a13e6848 2690 assert write_archive.issubset({True, False, 'ignore'})
2691 if True in write_archive and False not in write_archive:
2692 self.record_download_archive(info_dict)
be72c624 2693
5ec1b6b7 2694 info_dict['requested_downloads'] = downloaded_formats
ed5835b4 2695 info_dict = self.run_all_pps('after_video', info_dict)
a13e6848 2696 if max_downloads_reached:
2697 raise MaxDownloadsReached()
ebed8b37 2698
49a57e70 2699 # We update the info dict with the selected best quality format (backwards compatibility)
be72c624 2700 info_dict.update(best_format)
dd82ffea
JMF
2701 return info_dict
2702
98c70d6f 2703 def process_subtitles(self, video_id, normal_subtitles, automatic_captions):
a504ced0 2704 """Select the requested subtitles and their format"""
d8a58ddc 2705 available_subs, normal_sub_langs = {}, []
98c70d6f
JMF
2706 if normal_subtitles and self.params.get('writesubtitles'):
2707 available_subs.update(normal_subtitles)
d8a58ddc 2708 normal_sub_langs = tuple(normal_subtitles.keys())
98c70d6f
JMF
2709 if automatic_captions and self.params.get('writeautomaticsub'):
2710 for lang, cap_info in automatic_captions.items():
360e1ca5
JMF
2711 if lang not in available_subs:
2712 available_subs[lang] = cap_info
2713
4d171848
JMF
2714 if (not self.params.get('writesubtitles') and not
2715 self.params.get('writeautomaticsub') or not
2716 available_subs):
2717 return None
a504ced0 2718
d8a58ddc 2719 all_sub_langs = tuple(available_subs.keys())
a504ced0 2720 if self.params.get('allsubtitles', False):
c32b0aab 2721 requested_langs = all_sub_langs
2722 elif self.params.get('subtitleslangs', False):
77c4a9ef 2723 # A list is used so that the order of languages will be the same as
2724 # given in subtitleslangs. See https://github.com/yt-dlp/yt-dlp/issues/1041
2725 requested_langs = []
2726 for lang_re in self.params.get('subtitleslangs'):
77c4a9ef 2727 discard = lang_re[0] == '-'
c32b0aab 2728 if discard:
77c4a9ef 2729 lang_re = lang_re[1:]
3aa91540 2730 if lang_re == 'all':
2731 if discard:
2732 requested_langs = []
2733 else:
2734 requested_langs.extend(all_sub_langs)
2735 continue
77c4a9ef 2736 current_langs = filter(re.compile(lang_re + '$').match, all_sub_langs)
c32b0aab 2737 if discard:
2738 for lang in current_langs:
77c4a9ef 2739 while lang in requested_langs:
2740 requested_langs.remove(lang)
c32b0aab 2741 else:
77c4a9ef 2742 requested_langs.extend(current_langs)
2743 requested_langs = orderedSet(requested_langs)
d8a58ddc 2744 elif normal_sub_langs:
2745 requested_langs = ['en'] if 'en' in normal_sub_langs else normal_sub_langs[:1]
a504ced0 2746 else:
d8a58ddc 2747 requested_langs = ['en'] if 'en' in all_sub_langs else all_sub_langs[:1]
ad3dc496 2748 if requested_langs:
2749 self.write_debug('Downloading subtitles: %s' % ', '.join(requested_langs))
a504ced0
JMF
2750
2751 formats_query = self.params.get('subtitlesformat', 'best')
2752 formats_preference = formats_query.split('/') if formats_query else []
2753 subs = {}
2754 for lang in requested_langs:
2755 formats = available_subs.get(lang)
2756 if formats is None:
86e5f3ed 2757 self.report_warning(f'{lang} subtitles not available for {video_id}')
a504ced0 2758 continue
a504ced0
JMF
2759 for ext in formats_preference:
2760 if ext == 'best':
2761 f = formats[-1]
2762 break
2763 matches = list(filter(lambda f: f['ext'] == ext, formats))
2764 if matches:
2765 f = matches[-1]
2766 break
2767 else:
2768 f = formats[-1]
2769 self.report_warning(
2770 'No subtitle format found matching "%s" for language %s, '
2771 'using %s' % (formats_query, lang, f['ext']))
2772 subs[lang] = f
2773 return subs
2774
bb66c247 2775 def _forceprint(self, key, info_dict):
2776 if info_dict is None:
2777 return
2778 info_copy = info_dict.copy()
2779 info_copy['formats_table'] = self.render_formats_table(info_dict)
2780 info_copy['thumbnails_table'] = self.render_thumbnails_table(info_dict)
2781 info_copy['subtitles_table'] = self.render_subtitles_table(info_dict.get('id'), info_dict.get('subtitles'))
2782 info_copy['automatic_captions_table'] = self.render_subtitles_table(info_dict.get('id'), info_dict.get('automatic_captions'))
2783
2784 def format_tmpl(tmpl):
2785 mobj = re.match(r'\w+(=?)$', tmpl)
2786 if mobj and mobj.group(1):
2787 return f'{tmpl[:-1]} = %({tmpl[:-1]})r'
2788 elif mobj:
2789 return f'%({tmpl})s'
2790 return tmpl
8130779d 2791
bb66c247 2792 for tmpl in self.params['forceprint'].get(key, []):
2793 self.to_stdout(self.evaluate_outtmpl(format_tmpl(tmpl), info_copy))
2794
2795 for tmpl, file_tmpl in self.params['print_to_file'].get(key, []):
5127e92a 2796 filename = self.prepare_filename(info_dict, outtmpl=file_tmpl)
bb66c247 2797 tmpl = format_tmpl(tmpl)
2798 self.to_screen(f'[info] Writing {tmpl!r} to: {filename}')
8d93e69d 2799 if self._ensure_dir_exists(filename):
86e5f3ed 2800 with open(filename, 'a', encoding='utf-8') as f:
8d93e69d 2801 f.write(self.evaluate_outtmpl(tmpl, info_copy) + '\n')
ca30f449 2802
d06daf23 2803 def __forced_printings(self, info_dict, filename, incomplete):
53c18592 2804 def print_mandatory(field, actual_field=None):
2805 if actual_field is None:
2806 actual_field = field
d06daf23 2807 if (self.params.get('force%s' % field, False)
53c18592 2808 and (not incomplete or info_dict.get(actual_field) is not None)):
2809 self.to_stdout(info_dict[actual_field])
d06daf23
S
2810
2811 def print_optional(field):
2812 if (self.params.get('force%s' % field, False)
2813 and info_dict.get(field) is not None):
2814 self.to_stdout(info_dict[field])
2815
53c18592 2816 info_dict = info_dict.copy()
2817 if filename is not None:
2818 info_dict['filename'] = filename
2819 if info_dict.get('requested_formats') is not None:
2820 # For RTMP URLs, also include the playpath
2821 info_dict['urls'] = '\n'.join(f['url'] + f.get('play_path', '') for f in info_dict['requested_formats'])
10331a26 2822 elif info_dict.get('url'):
53c18592 2823 info_dict['urls'] = info_dict['url'] + info_dict.get('play_path', '')
2824
bb66c247 2825 if (self.params.get('forcejson')
2826 or self.params['forceprint'].get('video')
2827 or self.params['print_to_file'].get('video')):
2b8a2973 2828 self.post_extract(info_dict)
bb66c247 2829 self._forceprint('video', info_dict)
53c18592 2830
d06daf23
S
2831 print_mandatory('title')
2832 print_mandatory('id')
53c18592 2833 print_mandatory('url', 'urls')
d06daf23
S
2834 print_optional('thumbnail')
2835 print_optional('description')
53c18592 2836 print_optional('filename')
b868936c 2837 if self.params.get('forceduration') and info_dict.get('duration') is not None:
d06daf23
S
2838 self.to_stdout(formatSeconds(info_dict['duration']))
2839 print_mandatory('format')
53c18592 2840
2b8a2973 2841 if self.params.get('forcejson'):
6e84b215 2842 self.to_stdout(json.dumps(self.sanitize_info(info_dict)))
d06daf23 2843
e8e73840 2844 def dl(self, name, info, subtitle=False, test=False):
88acdbc2 2845 if not info.get('url'):
1151c407 2846 self.raise_no_formats(info, True)
e8e73840 2847
2848 if test:
2849 verbose = self.params.get('verbose')
2850 params = {
2851 'test': True,
a169858f 2852 'quiet': self.params.get('quiet') or not verbose,
e8e73840 2853 'verbose': verbose,
2854 'noprogress': not verbose,
2855 'nopart': True,
2856 'skip_unavailable_fragments': False,
2857 'keep_fragments': False,
2858 'overwrites': True,
2859 '_no_ytdl_file': True,
2860 }
2861 else:
2862 params = self.params
96fccc10 2863 fd = get_suitable_downloader(info, params, to_stdout=(name == '-'))(self, params)
e8e73840 2864 if not test:
2865 for ph in self._progress_hooks:
2866 fd.add_progress_hook(ph)
42676437
M
2867 urls = '", "'.join(
2868 (f['url'].split(',')[0] + ',<data>' if f['url'].startswith('data:') else f['url'])
2869 for f in info.get('requested_formats', []) or [info])
3a408f9d 2870 self.write_debug(f'Invoking {fd.FD_NAME} downloader on "{urls}"')
03b4de72 2871
adbc4ec4
THD
2872 # Note: Ideally info should be a deep-copied so that hooks cannot modify it.
2873 # But it may contain objects that are not deep-copyable
2874 new_info = self._copy_infodict(info)
e8e73840 2875 if new_info.get('http_headers') is None:
2876 new_info['http_headers'] = self._calc_headers(new_info)
2877 return fd.download(name, new_info, subtitle)
2878
e04938ab 2879 def existing_file(self, filepaths, *, default_overwrite=True):
2880 existing_files = list(filter(os.path.exists, orderedSet(filepaths)))
2881 if existing_files and not self.params.get('overwrites', default_overwrite):
2882 return existing_files[0]
2883
2884 for file in existing_files:
2885 self.report_file_delete(file)
2886 os.remove(file)
2887 return None
2888
8222d8de 2889 def process_info(self, info_dict):
09b49e1f 2890 """Process a single resolved IE result. (Modifies it in-place)"""
8222d8de
JMF
2891
2892 assert info_dict.get('_type', 'video') == 'video'
f46e2f9d 2893 original_infodict = info_dict
fd288278 2894
4513a41a 2895 if 'format' not in info_dict and 'ext' in info_dict:
8222d8de
JMF
2896 info_dict['format'] = info_dict['ext']
2897
09b49e1f 2898 # This is mostly just for backward compatibility of process_info
2899 # As a side-effect, this allows for format-specific filters
c77495e3 2900 if self._match_entry(info_dict) is not None:
9e907ebd 2901 info_dict['__write_download_archive'] = 'ignore'
8222d8de
JMF
2902 return
2903
09b49e1f 2904 # Does nothing under normal operation - for backward compatibility of process_info
277d6ff5 2905 self.post_extract(info_dict)
0c14d66a 2906 self._num_downloads += 1
8222d8de 2907
dcf64d43 2908 # info_dict['_filename'] needs to be set for backward compatibility
de6000d9 2909 info_dict['_filename'] = full_filename = self.prepare_filename(info_dict, warn=True)
2910 temp_filename = self.prepare_filename(info_dict, 'temp')
0202b52a 2911 files_to_move = {}
8222d8de
JMF
2912
2913 # Forced printings
4513a41a 2914 self.__forced_printings(info_dict, full_filename, incomplete=('format' not in info_dict))
8222d8de 2915
ca6d59d2 2916 def check_max_downloads():
2917 if self._num_downloads >= float(self.params.get('max_downloads') or 'inf'):
2918 raise MaxDownloadsReached()
2919
b7b04c78 2920 if self.params.get('simulate'):
9e907ebd 2921 info_dict['__write_download_archive'] = self.params.get('force_write_download_archive')
ca6d59d2 2922 check_max_downloads()
8222d8de
JMF
2923 return
2924
de6000d9 2925 if full_filename is None:
8222d8de 2926 return
e92caff5 2927 if not self._ensure_dir_exists(encodeFilename(full_filename)):
0202b52a 2928 return
e92caff5 2929 if not self._ensure_dir_exists(encodeFilename(temp_filename)):
8222d8de
JMF
2930 return
2931
80c03fa9 2932 if self._write_description('video', info_dict,
2933 self.prepare_filename(info_dict, 'description')) is None:
2934 return
2935
2936 sub_files = self._write_subtitles(info_dict, temp_filename)
2937 if sub_files is None:
2938 return
2939 files_to_move.update(dict(sub_files))
2940
2941 thumb_files = self._write_thumbnails(
2942 'video', info_dict, temp_filename, self.prepare_filename(info_dict, 'thumbnail'))
2943 if thumb_files is None:
2944 return
2945 files_to_move.update(dict(thumb_files))
8222d8de 2946
80c03fa9 2947 infofn = self.prepare_filename(info_dict, 'infojson')
2948 _infojson_written = self._write_info_json('video', info_dict, infofn)
2949 if _infojson_written:
dac5df5a 2950 info_dict['infojson_filename'] = infofn
e75bb0d6 2951 # For backward compatibility, even though it was a private field
80c03fa9 2952 info_dict['__infojson_filename'] = infofn
2953 elif _infojson_written is None:
2954 return
2955
2956 # Note: Annotations are deprecated
2957 annofn = None
1fb07d10 2958 if self.params.get('writeannotations', False):
de6000d9 2959 annofn = self.prepare_filename(info_dict, 'annotation')
80c03fa9 2960 if annofn:
e92caff5 2961 if not self._ensure_dir_exists(encodeFilename(annofn)):
0202b52a 2962 return
0c3d0f51 2963 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(annofn)):
6febd1c1 2964 self.to_screen('[info] Video annotations are already present')
ffddb112
RA
2965 elif not info_dict.get('annotations'):
2966 self.report_warning('There are no annotations to write.')
7b6fefc9
PH
2967 else:
2968 try:
6febd1c1 2969 self.to_screen('[info] Writing video annotations to: ' + annofn)
86e5f3ed 2970 with open(encodeFilename(annofn), 'w', encoding='utf-8') as annofile:
7b6fefc9
PH
2971 annofile.write(info_dict['annotations'])
2972 except (KeyError, TypeError):
6febd1c1 2973 self.report_warning('There are no annotations to write.')
86e5f3ed 2974 except OSError:
6febd1c1 2975 self.report_error('Cannot write annotations file: ' + annofn)
7b6fefc9 2976 return
1fb07d10 2977
732044af 2978 # Write internet shortcut files
08438d2c 2979 def _write_link_file(link_type):
60f3e995 2980 url = try_get(info_dict['webpage_url'], iri_to_uri)
2981 if not url:
2982 self.report_warning(
2983 f'Cannot write internet shortcut file because the actual URL of "{info_dict["webpage_url"]}" is unknown')
2984 return True
08438d2c 2985 linkfn = replace_extension(self.prepare_filename(info_dict, 'link'), link_type, info_dict.get('ext'))
0e6b018a
Z
2986 if not self._ensure_dir_exists(encodeFilename(linkfn)):
2987 return False
10e3742e 2988 if self.params.get('overwrites', True) and os.path.exists(encodeFilename(linkfn)):
08438d2c 2989 self.to_screen(f'[info] Internet shortcut (.{link_type}) is already present')
2990 return True
2991 try:
2992 self.to_screen(f'[info] Writing internet shortcut (.{link_type}) to: {linkfn}')
86e5f3ed 2993 with open(encodeFilename(to_high_limit_path(linkfn)), 'w', encoding='utf-8',
2994 newline='\r\n' if link_type == 'url' else '\n') as linkfile:
60f3e995 2995 template_vars = {'url': url}
08438d2c 2996 if link_type == 'desktop':
2997 template_vars['filename'] = linkfn[:-(len(link_type) + 1)]
2998 linkfile.write(LINK_TEMPLATES[link_type] % template_vars)
86e5f3ed 2999 except OSError:
08438d2c 3000 self.report_error(f'Cannot write internet shortcut {linkfn}')
3001 return False
732044af 3002 return True
3003
08438d2c 3004 write_links = {
3005 'url': self.params.get('writeurllink'),
3006 'webloc': self.params.get('writewebloclink'),
3007 'desktop': self.params.get('writedesktoplink'),
3008 }
3009 if self.params.get('writelink'):
3010 link_type = ('webloc' if sys.platform == 'darwin'
3011 else 'desktop' if sys.platform.startswith('linux')
3012 else 'url')
3013 write_links[link_type] = True
3014
3015 if any(should_write and not _write_link_file(link_type)
3016 for link_type, should_write in write_links.items()):
3017 return
732044af 3018
f46e2f9d 3019 def replace_info_dict(new_info):
3020 nonlocal info_dict
3021 if new_info == info_dict:
3022 return
3023 info_dict.clear()
3024 info_dict.update(new_info)
3025
415f8d51 3026 new_info, files_to_move = self.pre_process(info_dict, 'before_dl', files_to_move)
3027 replace_info_dict(new_info)
56d868db 3028
a13e6848 3029 if self.params.get('skip_download'):
56d868db 3030 info_dict['filepath'] = temp_filename
3031 info_dict['__finaldir'] = os.path.dirname(os.path.abspath(encodeFilename(full_filename)))
3032 info_dict['__files_to_move'] = files_to_move
f46e2f9d 3033 replace_info_dict(self.run_pp(MoveFilesAfterDownloadPP(self, False), info_dict))
9e907ebd 3034 info_dict['__write_download_archive'] = self.params.get('force_write_download_archive')
56d868db 3035 else:
3036 # Download
b868936c 3037 info_dict.setdefault('__postprocessors', [])
4340deca 3038 try:
0202b52a 3039
e04938ab 3040 def existing_video_file(*filepaths):
6b591b29 3041 ext = info_dict.get('ext')
e04938ab 3042 converted = lambda file: replace_extension(file, self.params.get('final_ext') or ext, ext)
3043 file = self.existing_file(itertools.chain(*zip(map(converted, filepaths), filepaths)),
3044 default_overwrite=False)
3045 if file:
3046 info_dict['ext'] = os.path.splitext(file)[1][1:]
3047 return file
0202b52a 3048
7b2c3f47 3049 fd, success = None, True
fccf90e7 3050 if info_dict.get('protocol') or info_dict.get('url'):
56ba69e4 3051 fd = get_suitable_downloader(info_dict, self.params, to_stdout=temp_filename == '-')
3052 if fd is not FFmpegFD and (
3053 info_dict.get('section_start') or info_dict.get('section_end')):
7b2c3f47 3054 msg = ('This format cannot be partially downloaded' if FFmpegFD.available()
56ba69e4 3055 else 'You have requested downloading the video partially, but ffmpeg is not installed')
3056 self.report_error(f'{msg}. Aborting')
5ec1b6b7 3057 return
5ec1b6b7 3058
4340deca 3059 if info_dict.get('requested_formats') is not None:
81cd954a
S
3060
3061 def compatible_formats(formats):
d03cfdce 3062 # TODO: some formats actually allow this (mkv, webm, ogg, mp4), but not all of them.
3063 video_formats = [format for format in formats if format.get('vcodec') != 'none']
3064 audio_formats = [format for format in formats if format.get('acodec') != 'none']
3065 if len(video_formats) > 2 or len(audio_formats) > 2:
3066 return False
3067
81cd954a 3068 # Check extension
86e5f3ed 3069 exts = {format.get('ext') for format in formats}
d03cfdce 3070 COMPATIBLE_EXTS = (
86e5f3ed 3071 {'mp3', 'mp4', 'm4a', 'm4p', 'm4b', 'm4r', 'm4v', 'ismv', 'isma'},
3072 {'webm'},
d03cfdce 3073 )
3074 for ext_sets in COMPATIBLE_EXTS:
3075 if ext_sets.issuperset(exts):
3076 return True
81cd954a
S
3077 # TODO: Check acodec/vcodec
3078 return False
3079
3080 requested_formats = info_dict['requested_formats']
0202b52a 3081 old_ext = info_dict['ext']
4e3b637d 3082 if self.params.get('merge_output_format') is None:
3083 if not compatible_formats(requested_formats):
3084 info_dict['ext'] = 'mkv'
3085 self.report_warning(
3086 'Requested formats are incompatible for merge and will be merged into mkv')
3087 if (info_dict['ext'] == 'webm'
3088 and info_dict.get('thumbnails')
3089 # check with type instead of pp_key, __name__, or isinstance
3090 # since we dont want any custom PPs to trigger this
c487cf00 3091 and any(type(pp) == EmbedThumbnailPP for pp in self._pps['post_process'])): # noqa: E721
4e3b637d 3092 info_dict['ext'] = 'mkv'
3093 self.report_warning(
3094 'webm doesn\'t support embedding a thumbnail, mkv will be used')
124bc071 3095 new_ext = info_dict['ext']
0202b52a 3096
124bc071 3097 def correct_ext(filename, ext=new_ext):
96fccc10 3098 if filename == '-':
3099 return filename
0202b52a 3100 filename_real_ext = os.path.splitext(filename)[1][1:]
3101 filename_wo_ext = (
3102 os.path.splitext(filename)[0]
124bc071 3103 if filename_real_ext in (old_ext, new_ext)
0202b52a 3104 else filename)
86e5f3ed 3105 return f'{filename_wo_ext}.{ext}'
0202b52a 3106
38c6902b 3107 # Ensure filename always has a correct extension for successful merge
0202b52a 3108 full_filename = correct_ext(full_filename)
3109 temp_filename = correct_ext(temp_filename)
e04938ab 3110 dl_filename = existing_video_file(full_filename, temp_filename)
1ea24129 3111 info_dict['__real_download'] = False
18e674b4 3112
7b2c3f47 3113 merger = FFmpegMergerPP(self)
adbc4ec4 3114 downloaded = []
dbf5416a 3115 if dl_filename is not None:
6c7274ec 3116 self.report_file_already_downloaded(dl_filename)
adbc4ec4
THD
3117 elif fd:
3118 for f in requested_formats if fd != FFmpegFD else []:
3119 f['filepath'] = fname = prepend_extension(
3120 correct_ext(temp_filename, info_dict['ext']),
3121 'f%s' % f['format_id'], info_dict['ext'])
3122 downloaded.append(fname)
dbf5416a 3123 info_dict['url'] = '\n'.join(f['url'] for f in requested_formats)
3124 success, real_download = self.dl(temp_filename, info_dict)
3125 info_dict['__real_download'] = real_download
18e674b4 3126 else:
18e674b4 3127 if self.params.get('allow_unplayable_formats'):
3128 self.report_warning(
3129 'You have requested merging of multiple formats '
3130 'while also allowing unplayable formats to be downloaded. '
3131 'The formats won\'t be merged to prevent data corruption.')
3132 elif not merger.available:
e8969bda 3133 msg = 'You have requested merging of multiple formats but ffmpeg is not installed'
3134 if not self.params.get('ignoreerrors'):
3135 self.report_error(f'{msg}. Aborting due to --abort-on-error')
3136 return
3137 self.report_warning(f'{msg}. The formats won\'t be merged')
18e674b4 3138
96fccc10 3139 if temp_filename == '-':
adbc4ec4 3140 reason = ('using a downloader other than ffmpeg' if FFmpegFD.can_merge_formats(info_dict, self.params)
96fccc10 3141 else 'but the formats are incompatible for simultaneous download' if merger.available
3142 else 'but ffmpeg is not installed')
3143 self.report_warning(
3144 f'You have requested downloading multiple formats to stdout {reason}. '
3145 'The formats will be streamed one after the other')
3146 fname = temp_filename
dbf5416a 3147 for f in requested_formats:
3148 new_info = dict(info_dict)
3149 del new_info['requested_formats']
3150 new_info.update(f)
96fccc10 3151 if temp_filename != '-':
124bc071 3152 fname = prepend_extension(
3153 correct_ext(temp_filename, new_info['ext']),
3154 'f%s' % f['format_id'], new_info['ext'])
96fccc10 3155 if not self._ensure_dir_exists(fname):
3156 return
a21e0ab1 3157 f['filepath'] = fname
96fccc10 3158 downloaded.append(fname)
dbf5416a 3159 partial_success, real_download = self.dl(fname, new_info)
3160 info_dict['__real_download'] = info_dict['__real_download'] or real_download
3161 success = success and partial_success
adbc4ec4
THD
3162
3163 if downloaded and merger.available and not self.params.get('allow_unplayable_formats'):
3164 info_dict['__postprocessors'].append(merger)
3165 info_dict['__files_to_merge'] = downloaded
3166 # Even if there were no downloads, it is being merged only now
3167 info_dict['__real_download'] = True
3168 else:
3169 for file in downloaded:
3170 files_to_move[file] = None
4340deca
P
3171 else:
3172 # Just a single file
e04938ab 3173 dl_filename = existing_video_file(full_filename, temp_filename)
6c7274ec 3174 if dl_filename is None or dl_filename == temp_filename:
3175 # dl_filename == temp_filename could mean that the file was partially downloaded with --no-part.
3176 # So we should try to resume the download
e8e73840 3177 success, real_download = self.dl(temp_filename, info_dict)
0202b52a 3178 info_dict['__real_download'] = real_download
6c7274ec 3179 else:
3180 self.report_file_already_downloaded(dl_filename)
0202b52a 3181
0202b52a 3182 dl_filename = dl_filename or temp_filename
c571435f 3183 info_dict['__finaldir'] = os.path.dirname(os.path.abspath(encodeFilename(full_filename)))
0202b52a 3184
3158150c 3185 except network_exceptions as err:
7960b056 3186 self.report_error('unable to download video data: %s' % error_to_compat_str(err))
4340deca 3187 return
86e5f3ed 3188 except OSError as err:
4340deca
P
3189 raise UnavailableVideoError(err)
3190 except (ContentTooShortError, ) as err:
86e5f3ed 3191 self.report_error(f'content too short (expected {err.expected} bytes and served {err.downloaded})')
4340deca 3192 return
8222d8de 3193
415f8d51 3194 self._raise_pending_errors(info_dict)
de6000d9 3195 if success and full_filename != '-':
f17f8651 3196
fd7cfb64 3197 def fixup():
3198 do_fixup = True
3199 fixup_policy = self.params.get('fixup')
3200 vid = info_dict['id']
3201
3202 if fixup_policy in ('ignore', 'never'):
3203 return
3204 elif fixup_policy == 'warn':
3fe75fdc 3205 do_fixup = 'warn'
f89b3e2d 3206 elif fixup_policy != 'force':
3207 assert fixup_policy in ('detect_or_warn', None)
3208 if not info_dict.get('__real_download'):
3209 do_fixup = False
fd7cfb64 3210
3211 def ffmpeg_fixup(cndn, msg, cls):
3fe75fdc 3212 if not (do_fixup and cndn):
fd7cfb64 3213 return
3fe75fdc 3214 elif do_fixup == 'warn':
fd7cfb64 3215 self.report_warning(f'{vid}: {msg}')
3216 return
3217 pp = cls(self)
3218 if pp.available:
3219 info_dict['__postprocessors'].append(pp)
3220 else:
3221 self.report_warning(f'{vid}: {msg}. Install ffmpeg to fix this automatically')
3222
3223 stretched_ratio = info_dict.get('stretched_ratio')
ca9def71
LNO
3224 ffmpeg_fixup(stretched_ratio not in (1, None),
3225 f'Non-uniform pixel ratio {stretched_ratio}',
3226 FFmpegFixupStretchedPP)
fd7cfb64 3227
993191c0 3228 downloader = get_suitable_downloader(info_dict, self.params) if 'protocol' in info_dict else None
24146491 3229 downloader = downloader.FD_NAME if downloader else None
adbc4ec4 3230
ca9def71
LNO
3231 ext = info_dict.get('ext')
3232 postprocessed_by_ffmpeg = info_dict.get('requested_formats') or any((
3233 isinstance(pp, FFmpegVideoConvertorPP)
3234 and resolve_recode_mapping(ext, pp.mapping)[0] not in (ext, None)
3235 ) for pp in self._pps['post_process'])
3236
3237 if not postprocessed_by_ffmpeg:
3238 ffmpeg_fixup(ext == 'm4a' and info_dict.get('container') == 'm4a_dash',
f2df4071 3239 'writing DASH m4a. Only some players support this container',
3240 FFmpegFixupM4aPP)
24146491 3241 ffmpeg_fixup(downloader == 'hlsnative' and not self.params.get('hls_use_mpegts')
494f5230 3242 or info_dict.get('is_live') and self.params.get('hls_use_mpegts') is None,
adbc4ec4
THD
3243 'Possible MPEG-TS in MP4 container or malformed AAC timestamps',
3244 FFmpegFixupM3u8PP)
3245 ffmpeg_fixup(info_dict.get('is_live') and downloader == 'DashSegmentsFD',
3246 'Possible duplicate MOOV atoms', FFmpegFixupDuplicateMoovPP)
3247
24146491 3248 ffmpeg_fixup(downloader == 'web_socket_fragment', 'Malformed timestamps detected', FFmpegFixupTimestampPP)
3249 ffmpeg_fixup(downloader == 'web_socket_fragment', 'Malformed duration detected', FFmpegFixupDurationPP)
fd7cfb64 3250
3251 fixup()
8222d8de 3252 try:
f46e2f9d 3253 replace_info_dict(self.post_process(dl_filename, info_dict, files_to_move))
af819c21 3254 except PostProcessingError as err:
3255 self.report_error('Postprocessing: %s' % str(err))
8222d8de 3256 return
ab8e5e51
AM
3257 try:
3258 for ph in self._post_hooks:
23c1a667 3259 ph(info_dict['filepath'])
ab8e5e51
AM
3260 except Exception as err:
3261 self.report_error('post hooks: %s' % str(err))
3262 return
9e907ebd 3263 info_dict['__write_download_archive'] = True
2d30509f 3264
c487cf00 3265 assert info_dict is original_infodict # Make sure the info_dict was modified in-place
a13e6848 3266 if self.params.get('force_write_download_archive'):
9e907ebd 3267 info_dict['__write_download_archive'] = True
ca6d59d2 3268 check_max_downloads()
8222d8de 3269
aa9369a2 3270 def __download_wrapper(self, func):
3271 @functools.wraps(func)
3272 def wrapper(*args, **kwargs):
3273 try:
3274 res = func(*args, **kwargs)
3275 except UnavailableVideoError as e:
3276 self.report_error(e)
b222c271 3277 except DownloadCancelled as e:
3278 self.to_screen(f'[info] {e}')
3279 if not self.params.get('break_per_url'):
3280 raise
aa9369a2 3281 else:
3282 if self.params.get('dump_single_json', False):
3283 self.post_extract(res)
3284 self.to_stdout(json.dumps(self.sanitize_info(res)))
3285 return wrapper
3286
8222d8de
JMF
3287 def download(self, url_list):
3288 """Download a given list of URLs."""
aa9369a2 3289 url_list = variadic(url_list) # Passing a single URL is a common mistake
bf1824b3 3290 outtmpl = self.params['outtmpl']['default']
3089bc74
S
3291 if (len(url_list) > 1
3292 and outtmpl != '-'
3293 and '%' not in outtmpl
3294 and self.params.get('max_downloads') != 1):
acd69589 3295 raise SameFileError(outtmpl)
8222d8de
JMF
3296
3297 for url in url_list:
aa9369a2 3298 self.__download_wrapper(self.extract_info)(
3299 url, force_generic_extractor=self.params.get('force_generic_extractor', False))
8222d8de
JMF
3300
3301 return self._download_retcode
3302
1dcc4c0c 3303 def download_with_info_file(self, info_filename):
31bd3925
JMF
3304 with contextlib.closing(fileinput.FileInput(
3305 [info_filename], mode='r',
3306 openhook=fileinput.hook_encoded('utf-8'))) as f:
3307 # FileInput doesn't have a read method, we can't call json.load
8012d892 3308 info = self.sanitize_info(json.loads('\n'.join(f)), self.params.get('clean_infojson', True))
d4943898 3309 try:
aa9369a2 3310 self.__download_wrapper(self.process_ie_result)(info, download=True)
f2ebc5c7 3311 except (DownloadError, EntryNotInPlaylist, ReExtractInfo) as e:
bf5f605e 3312 if not isinstance(e, EntryNotInPlaylist):
3313 self.to_stderr('\r')
d4943898
JMF
3314 webpage_url = info.get('webpage_url')
3315 if webpage_url is not None:
aa9369a2 3316 self.report_warning(f'The info failed to download: {e}; trying with URL {webpage_url}')
d4943898
JMF
3317 return self.download([webpage_url])
3318 else:
3319 raise
3320 return self._download_retcode
1dcc4c0c 3321
cb202fd2 3322 @staticmethod
8012d892 3323 def sanitize_info(info_dict, remove_private_keys=False):
3324 ''' Sanitize the infodict for converting to json '''
3ad56b42 3325 if info_dict is None:
3326 return info_dict
6e84b215 3327 info_dict.setdefault('epoch', int(time.time()))
6a5a30f9 3328 info_dict.setdefault('_type', 'video')
09b49e1f 3329
8012d892 3330 if remove_private_keys:
0a5a191a 3331 reject = lambda k, v: v is None or k.startswith('__') or k in {
f46e2f9d 3332 'requested_downloads', 'requested_formats', 'requested_subtitles', 'requested_entries',
0a5a191a 3333 'entries', 'filepath', '_filename', 'infojson_filename', 'original_url', 'playlist_autonumber',
6e84b215 3334 }
ae8f99e6 3335 else:
09b49e1f 3336 reject = lambda k, v: False
adbc4ec4
THD
3337
3338 def filter_fn(obj):
3339 if isinstance(obj, dict):
3340 return {k: filter_fn(v) for k, v in obj.items() if not reject(k, v)}
3341 elif isinstance(obj, (list, tuple, set, LazyList)):
3342 return list(map(filter_fn, obj))
3343 elif obj is None or isinstance(obj, (str, int, float, bool)):
3344 return obj
3345 else:
3346 return repr(obj)
3347
5226731e 3348 return filter_fn(info_dict)
cb202fd2 3349
8012d892 3350 @staticmethod
3351 def filter_requested_info(info_dict, actually_filter=True):
3352 ''' Alias of sanitize_info for backward compatibility '''
3353 return YoutubeDL.sanitize_info(info_dict, actually_filter)
3354
43d7f5a5 3355 def _delete_downloaded_files(self, *files_to_delete, info={}, msg=None):
3356 for filename in set(filter(None, files_to_delete)):
3357 if msg:
3358 self.to_screen(msg % filename)
3359 try:
3360 os.remove(filename)
3361 except OSError:
3362 self.report_warning(f'Unable to delete file {filename}')
3363 if filename in info.get('__files_to_move', []): # NB: Delete even if None
3364 del info['__files_to_move'][filename]
3365
ed5835b4 3366 @staticmethod
3367 def post_extract(info_dict):
3368 def actual_post_extract(info_dict):
3369 if info_dict.get('_type') in ('playlist', 'multi_video'):
3370 for video_dict in info_dict.get('entries', {}):
3371 actual_post_extract(video_dict or {})
3372 return
3373
09b49e1f 3374 post_extractor = info_dict.pop('__post_extractor', None) or (lambda: {})
3375 info_dict.update(post_extractor())
ed5835b4 3376
3377 actual_post_extract(info_dict or {})
3378
dcf64d43 3379 def run_pp(self, pp, infodict):
5bfa4862 3380 files_to_delete = []
dcf64d43 3381 if '__files_to_move' not in infodict:
3382 infodict['__files_to_move'] = {}
b1940459 3383 try:
3384 files_to_delete, infodict = pp.run(infodict)
3385 except PostProcessingError as e:
3386 # Must be True and not 'only_download'
3387 if self.params.get('ignoreerrors') is True:
3388 self.report_error(e)
3389 return infodict
3390 raise
3391
5bfa4862 3392 if not files_to_delete:
dcf64d43 3393 return infodict
5bfa4862 3394 if self.params.get('keepvideo', False):
3395 for f in files_to_delete:
dcf64d43 3396 infodict['__files_to_move'].setdefault(f, '')
5bfa4862 3397 else:
43d7f5a5 3398 self._delete_downloaded_files(
3399 *files_to_delete, info=infodict, msg='Deleting original file %s (pass -k to keep)')
dcf64d43 3400 return infodict
5bfa4862 3401
ed5835b4 3402 def run_all_pps(self, key, info, *, additional_pps=None):
bb66c247 3403 self._forceprint(key, info)
ed5835b4 3404 for pp in (additional_pps or []) + self._pps[key]:
dc5f409c 3405 info = self.run_pp(pp, info)
ed5835b4 3406 return info
277d6ff5 3407
56d868db 3408 def pre_process(self, ie_info, key='pre_process', files_to_move=None):
5bfa4862 3409 info = dict(ie_info)
56d868db 3410 info['__files_to_move'] = files_to_move or {}
415f8d51 3411 try:
3412 info = self.run_all_pps(key, info)
3413 except PostProcessingError as err:
3414 msg = f'Preprocessing: {err}'
3415 info.setdefault('__pending_error', msg)
3416 self.report_error(msg, is_error=False)
56d868db 3417 return info, info.pop('__files_to_move', None)
5bfa4862 3418
f46e2f9d 3419 def post_process(self, filename, info, files_to_move=None):
8222d8de 3420 """Run all the postprocessors on the given file."""
8222d8de 3421 info['filepath'] = filename
dcf64d43 3422 info['__files_to_move'] = files_to_move or {}
ed5835b4 3423 info = self.run_all_pps('post_process', info, additional_pps=info.get('__postprocessors'))
dcf64d43 3424 info = self.run_pp(MoveFilesAfterDownloadPP(self), info)
3425 del info['__files_to_move']
ed5835b4 3426 return self.run_all_pps('after_move', info)
c1c9a79c 3427
5db07df6 3428 def _make_archive_id(self, info_dict):
e9fef7ee
S
3429 video_id = info_dict.get('id')
3430 if not video_id:
3431 return
5db07df6
PH
3432 # Future-proof against any change in case
3433 # and backwards compatibility with prior versions
e9fef7ee 3434 extractor = info_dict.get('extractor_key') or info_dict.get('ie_key') # key in a playlist
7012b23c 3435 if extractor is None:
1211bb6d
S
3436 url = str_or_none(info_dict.get('url'))
3437 if not url:
3438 return
e9fef7ee 3439 # Try to find matching extractor for the URL and take its ie_key
8b7491c8 3440 for ie_key, ie in self._ies.items():
1211bb6d 3441 if ie.suitable(url):
8b7491c8 3442 extractor = ie_key
e9fef7ee
S
3443 break
3444 else:
3445 return
86e5f3ed 3446 return f'{extractor.lower()} {video_id}'
5db07df6
PH
3447
3448 def in_download_archive(self, info_dict):
3449 fn = self.params.get('download_archive')
3450 if fn is None:
3451 return False
3452
3453 vid_id = self._make_archive_id(info_dict)
e9fef7ee 3454 if not vid_id:
7012b23c 3455 return False # Incomplete video information
5db07df6 3456
a45e8619 3457 return vid_id in self.archive
c1c9a79c
PH
3458
3459 def record_download_archive(self, info_dict):
3460 fn = self.params.get('download_archive')
3461 if fn is None:
3462 return
5db07df6
PH
3463 vid_id = self._make_archive_id(info_dict)
3464 assert vid_id
a13e6848 3465 self.write_debug(f'Adding to archive: {vid_id}')
c1c9a79c 3466 with locked_file(fn, 'a', encoding='utf-8') as archive_file:
6febd1c1 3467 archive_file.write(vid_id + '\n')
a45e8619 3468 self.archive.add(vid_id)
dd82ffea 3469
8c51aa65 3470 @staticmethod
8abeeb94 3471 def format_resolution(format, default='unknown'):
9359f3d4 3472 if format.get('vcodec') == 'none' and format.get('acodec') != 'none':
fb04e403 3473 return 'audio only'
f49d89ee
PH
3474 if format.get('resolution') is not None:
3475 return format['resolution']
35615307 3476 if format.get('width') and format.get('height'):
ff51ed58 3477 return '%dx%d' % (format['width'], format['height'])
35615307 3478 elif format.get('height'):
ff51ed58 3479 return '%sp' % format['height']
35615307 3480 elif format.get('width'):
ff51ed58 3481 return '%dx?' % format['width']
3482 return default
8c51aa65 3483
8130779d 3484 def _list_format_headers(self, *headers):
3485 if self.params.get('listformats_table', True) is not False:
591bb9d3 3486 return [self._format_out(header, self.Styles.HEADERS) for header in headers]
8130779d 3487 return headers
3488
c57f7757
PH
3489 def _format_note(self, fdict):
3490 res = ''
3491 if fdict.get('ext') in ['f4f', 'f4m']:
f304da8a 3492 res += '(unsupported)'
32f90364
PH
3493 if fdict.get('language'):
3494 if res:
3495 res += ' '
f304da8a 3496 res += '[%s]' % fdict['language']
c57f7757 3497 if fdict.get('format_note') is not None:
f304da8a 3498 if res:
3499 res += ' '
3500 res += fdict['format_note']
c57f7757 3501 if fdict.get('tbr') is not None:
f304da8a 3502 if res:
3503 res += ', '
3504 res += '%4dk' % fdict['tbr']
c57f7757
PH
3505 if fdict.get('container') is not None:
3506 if res:
3507 res += ', '
3508 res += '%s container' % fdict['container']
3089bc74
S
3509 if (fdict.get('vcodec') is not None
3510 and fdict.get('vcodec') != 'none'):
c57f7757
PH
3511 if res:
3512 res += ', '
3513 res += fdict['vcodec']
91c7271a 3514 if fdict.get('vbr') is not None:
c57f7757
PH
3515 res += '@'
3516 elif fdict.get('vbr') is not None and fdict.get('abr') is not None:
3517 res += 'video@'
3518 if fdict.get('vbr') is not None:
3519 res += '%4dk' % fdict['vbr']
fbb21cf5 3520 if fdict.get('fps') is not None:
5d583bdf
S
3521 if res:
3522 res += ', '
3523 res += '%sfps' % fdict['fps']
c57f7757
PH
3524 if fdict.get('acodec') is not None:
3525 if res:
3526 res += ', '
3527 if fdict['acodec'] == 'none':
3528 res += 'video only'
3529 else:
3530 res += '%-5s' % fdict['acodec']
3531 elif fdict.get('abr') is not None:
3532 if res:
3533 res += ', '
3534 res += 'audio'
3535 if fdict.get('abr') is not None:
3536 res += '@%3dk' % fdict['abr']
3537 if fdict.get('asr') is not None:
3538 res += ' (%5dHz)' % fdict['asr']
3539 if fdict.get('filesize') is not None:
3540 if res:
3541 res += ', '
3542 res += format_bytes(fdict['filesize'])
9732d77e
PH
3543 elif fdict.get('filesize_approx') is not None:
3544 if res:
3545 res += ', '
3546 res += '~' + format_bytes(fdict['filesize_approx'])
c57f7757 3547 return res
91c7271a 3548
8130779d 3549 def render_formats_table(self, info_dict):
b69fd25c 3550 if not info_dict.get('formats') and not info_dict.get('url'):
8130779d 3551 return None
b69fd25c 3552
94badb25 3553 formats = info_dict.get('formats', [info_dict])
8130779d 3554 if not self.params.get('listformats_table', True) is not False:
76d321f6 3555 table = [
3556 [
3557 format_field(f, 'format_id'),
3558 format_field(f, 'ext'),
3559 self.format_resolution(f),
8130779d 3560 self._format_note(f)
3561 ] for f in formats if f.get('preference') is None or f['preference'] >= -1000]
3562 return render_table(['format code', 'extension', 'resolution', 'note'], table, extra_gap=1)
3563
d816f61f 3564 def simplified_codec(f, field):
3565 assert field in ('acodec', 'vcodec')
3566 codec = f.get(field, 'unknown')
f5ea4748 3567 if not codec:
3568 return 'unknown'
3569 elif codec != 'none':
d816f61f 3570 return '.'.join(codec.split('.')[:4])
3571
3572 if field == 'vcodec' and f.get('acodec') == 'none':
3573 return 'images'
3574 elif field == 'acodec' and f.get('vcodec') == 'none':
3575 return ''
3576 return self._format_out('audio only' if field == 'vcodec' else 'video only',
3577 self.Styles.SUPPRESS)
3578
591bb9d3 3579 delim = self._format_out('\u2502', self.Styles.DELIM, '|', test_encoding=True)
8130779d 3580 table = [
3581 [
591bb9d3 3582 self._format_out(format_field(f, 'format_id'), self.Styles.ID),
8130779d 3583 format_field(f, 'ext'),
3584 format_field(f, func=self.format_resolution, ignore=('audio only', 'images')),
563e0bf8 3585 format_field(f, 'fps', '\t%d', func=round),
8130779d 3586 format_field(f, 'dynamic_range', '%s', ignore=(None, 'SDR')).replace('HDR', ''),
3587 delim,
3588 format_field(f, 'filesize', ' \t%s', func=format_bytes) + format_field(f, 'filesize_approx', '~\t%s', func=format_bytes),
563e0bf8 3589 format_field(f, 'tbr', '\t%dk', func=round),
8130779d 3590 shorten_protocol_name(f.get('protocol', '')),
3591 delim,
d816f61f 3592 simplified_codec(f, 'vcodec'),
563e0bf8 3593 format_field(f, 'vbr', '\t%dk', func=round),
d816f61f 3594 simplified_codec(f, 'acodec'),
563e0bf8 3595 format_field(f, 'abr', '\t%dk', func=round),
ae61d108 3596 format_field(f, 'asr', '\t%s', func=format_decimal_suffix),
8130779d 3597 join_nonempty(
591bb9d3 3598 self._format_out('UNSUPPORTED', 'light red') if f.get('ext') in ('f4f', 'f4m') else None,
8130779d 3599 format_field(f, 'language', '[%s]'),
3600 join_nonempty(format_field(f, 'format_note'),
3601 format_field(f, 'container', ignore=(None, f.get('ext'))),
3602 delim=', '),
3603 delim=' '),
3604 ] for f in formats if f.get('preference') is None or f['preference'] >= -1000]
3605 header_line = self._list_format_headers(
3606 'ID', 'EXT', 'RESOLUTION', '\tFPS', 'HDR', delim, '\tFILESIZE', '\tTBR', 'PROTO',
3607 delim, 'VCODEC', '\tVBR', 'ACODEC', '\tABR', '\tASR', 'MORE INFO')
3608
3609 return render_table(
3610 header_line, table, hide_empty=True,
591bb9d3 3611 delim=self._format_out('\u2500', self.Styles.DELIM, '-', test_encoding=True))
8130779d 3612
3613 def render_thumbnails_table(self, info_dict):
88f23a18 3614 thumbnails = list(info_dict.get('thumbnails') or [])
cfb56d1a 3615 if not thumbnails:
8130779d 3616 return None
3617 return render_table(
ec11a9f4 3618 self._list_format_headers('ID', 'Width', 'Height', 'URL'),
6970b600 3619 [[t.get('id'), t.get('width', 'unknown'), t.get('height', 'unknown'), t['url']] for t in thumbnails])
2412044c 3620
8130779d 3621 def render_subtitles_table(self, video_id, subtitles):
2412044c 3622 def _row(lang, formats):
49c258e1 3623 exts, names = zip(*((f['ext'], f.get('name') or 'unknown') for f in reversed(formats)))
2412044c 3624 if len(set(names)) == 1:
7aee40c1 3625 names = [] if names[0] == 'unknown' else names[:1]
2412044c 3626 return [lang, ', '.join(names), ', '.join(exts)]
3627
8130779d 3628 if not subtitles:
3629 return None
3630 return render_table(
ec11a9f4 3631 self._list_format_headers('Language', 'Name', 'Formats'),
2412044c 3632 [_row(lang, formats) for lang, formats in subtitles.items()],
8130779d 3633 hide_empty=True)
3634
3635 def __list_table(self, video_id, name, func, *args):
3636 table = func(*args)
3637 if not table:
3638 self.to_screen(f'{video_id} has no {name}')
3639 return
3640 self.to_screen(f'[info] Available {name} for {video_id}:')
3641 self.to_stdout(table)
3642
3643 def list_formats(self, info_dict):
3644 self.__list_table(info_dict['id'], 'formats', self.render_formats_table, info_dict)
3645
3646 def list_thumbnails(self, info_dict):
3647 self.__list_table(info_dict['id'], 'thumbnails', self.render_thumbnails_table, info_dict)
3648
3649 def list_subtitles(self, video_id, subtitles, name='subtitles'):
3650 self.__list_table(video_id, name, self.render_subtitles_table, video_id, subtitles)
a504ced0 3651
dca08720
PH
3652 def urlopen(self, req):
3653 """ Start an HTTP download """
f9934b96 3654 if isinstance(req, str):
67dda517 3655 req = sanitized_Request(req)
19a41fc6 3656 return self._opener.open(req, timeout=self._socket_timeout)
dca08720
PH
3657
3658 def print_debug_header(self):
3659 if not self.params.get('verbose'):
3660 return
49a57e70 3661
560738f3 3662 # These imports can be slow. So import them only as needed
3663 from .extractor.extractors import _LAZY_LOADER
3664 from .extractor.extractors import _PLUGIN_CLASSES as plugin_extractors
3665
49a57e70 3666 def get_encoding(stream):
2a938746 3667 ret = str(getattr(stream, 'encoding', 'missing (%s)' % type(stream).__name__))
49a57e70 3668 if not supports_terminal_sequences(stream):
53973b4d 3669 from .utils import WINDOWS_VT_MODE # Must be imported locally
e3c7d495 3670 ret += ' (No VT)' if WINDOWS_VT_MODE is False else ' (No ANSI)'
49a57e70 3671 return ret
3672
591bb9d3 3673 encoding_str = 'Encodings: locale %s, fs %s, pref %s, %s' % (
49a57e70 3674 locale.getpreferredencoding(),
3675 sys.getfilesystemencoding(),
591bb9d3 3676 self.get_encoding(),
3677 ', '.join(
64fa820c 3678 f'{key} {get_encoding(stream)}' for key, stream in self._out_files.items_
591bb9d3 3679 if stream is not None and key != 'console')
3680 )
883d4b1e 3681
3682 logger = self.params.get('logger')
3683 if logger:
3684 write_debug = lambda msg: logger.debug(f'[debug] {msg}')
3685 write_debug(encoding_str)
3686 else:
96565c7e 3687 write_string(f'[debug] {encoding_str}\n', encoding=None)
49a57e70 3688 write_debug = lambda msg: self._write_string(f'[debug] {msg}\n')
734f90bb 3689
4c88ff87 3690 source = detect_variant()
36eaf303 3691 write_debug(join_nonempty(
3692 'yt-dlp version', __version__,
3693 f'[{RELEASE_GIT_HEAD}]' if RELEASE_GIT_HEAD else '',
3694 '' if source == 'unknown' else f'({source})',
3695 delim=' '))
6e21fdd2 3696 if not _LAZY_LOADER:
3697 if os.environ.get('YTDLP_NO_LAZY_EXTRACTORS'):
49a57e70 3698 write_debug('Lazy loading extractors is forcibly disabled')
6e21fdd2 3699 else:
49a57e70 3700 write_debug('Lazy loading extractors is disabled')
3ae5e797 3701 if plugin_extractors or plugin_postprocessors:
49a57e70 3702 write_debug('Plugins: %s' % [
3ae5e797 3703 '%s%s' % (klass.__name__, '' if klass.__name__ == name else f' as {name}')
3704 for name, klass in itertools.chain(plugin_extractors.items(), plugin_postprocessors.items())])
8a82af35 3705 if self.params['compat_opts']:
3706 write_debug('Compatibility options: %s' % ', '.join(self.params['compat_opts']))
36eaf303 3707
3708 if source == 'source':
dca08720 3709 try:
f0c9fb96 3710 stdout, _, _ = Popen.run(
36eaf303 3711 ['git', 'rev-parse', '--short', 'HEAD'],
f0c9fb96 3712 text=True, cwd=os.path.dirname(os.path.abspath(__file__)),
3713 stdout=subprocess.PIPE, stderr=subprocess.PIPE)
3714 if re.fullmatch('[0-9a-f]+', stdout.strip()):
3715 write_debug(f'Git HEAD: {stdout.strip()}')
70a1165b 3716 except Exception:
19a03940 3717 with contextlib.suppress(Exception):
36eaf303 3718 sys.exc_clear()
b300cda4 3719
b1f94422 3720 write_debug(system_identifier())
d28b5171 3721
8913ef74 3722 exe_versions, ffmpeg_features = FFmpegPostProcessor.get_versions_and_features(self)
3723 ffmpeg_features = {key for key, val in ffmpeg_features.items() if val}
3724 if ffmpeg_features:
19a03940 3725 exe_versions['ffmpeg'] += ' (%s)' % ','.join(sorted(ffmpeg_features))
8913ef74 3726
4c83c967 3727 exe_versions['rtmpdump'] = rtmpdump_version()
feee8d32 3728 exe_versions['phantomjs'] = PhantomJSwrapper._version()
d28b5171 3729 exe_str = ', '.join(
2831b468 3730 f'{exe} {v}' for exe, v in sorted(exe_versions.items()) if v
3731 ) or 'none'
49a57e70 3732 write_debug('exe versions: %s' % exe_str)
dca08720 3733
1d485a1a 3734 from .compat.compat_utils import get_package_info
9b8ee23b 3735 from .dependencies import available_dependencies
3736
3737 write_debug('Optional libraries: %s' % (', '.join(sorted({
1d485a1a 3738 join_nonempty(*get_package_info(m)) for m in available_dependencies.values()
9b8ee23b 3739 })) or 'none'))
2831b468 3740
97ec5bc5 3741 self._setup_opener()
dca08720
PH
3742 proxy_map = {}
3743 for handler in self._opener.handlers:
3744 if hasattr(handler, 'proxies'):
3745 proxy_map.update(handler.proxies)
49a57e70 3746 write_debug(f'Proxy map: {proxy_map}')
dca08720 3747
49a57e70 3748 # Not implemented
3749 if False and self.params.get('call_home'):
0f06bcd7 3750 ipaddr = self.urlopen('https://yt-dl.org/ip').read().decode()
49a57e70 3751 write_debug('Public IP address: %s' % ipaddr)
58b1f00d 3752 latest_version = self.urlopen(
0f06bcd7 3753 'https://yt-dl.org/latest/version').read().decode()
58b1f00d
PH
3754 if version_tuple(latest_version) > version_tuple(__version__):
3755 self.report_warning(
3756 'You are using an outdated version (newest version: %s)! '
3757 'See https://yt-dl.org/update if you need help updating.' %
3758 latest_version)
3759
e344693b 3760 def _setup_opener(self):
97ec5bc5 3761 if hasattr(self, '_opener'):
3762 return
6ad14cab 3763 timeout_val = self.params.get('socket_timeout')
17bddf3e 3764 self._socket_timeout = 20 if timeout_val is None else float(timeout_val)
6ad14cab 3765
982ee69a 3766 opts_cookiesfrombrowser = self.params.get('cookiesfrombrowser')
dca08720
PH
3767 opts_cookiefile = self.params.get('cookiefile')
3768 opts_proxy = self.params.get('proxy')
3769
982ee69a 3770 self.cookiejar = load_cookies(opts_cookiefile, opts_cookiesfrombrowser, self)
dca08720 3771
6a3f4c3f 3772 cookie_processor = YoutubeDLCookieProcessor(self.cookiejar)
dca08720
PH
3773 if opts_proxy is not None:
3774 if opts_proxy == '':
3775 proxies = {}
3776 else:
3777 proxies = {'http': opts_proxy, 'https': opts_proxy}
3778 else:
ac668111 3779 proxies = urllib.request.getproxies()
067aa17e 3780 # Set HTTPS proxy to HTTP one if given (https://github.com/ytdl-org/youtube-dl/issues/805)
dca08720
PH
3781 if 'http' in proxies and 'https' not in proxies:
3782 proxies['https'] = proxies['http']
91410c9b 3783 proxy_handler = PerRequestProxyHandler(proxies)
a0ddb8a2
PH
3784
3785 debuglevel = 1 if self.params.get('debug_printtraffic') else 0
be4a824d
PH
3786 https_handler = make_HTTPS_handler(self.params, debuglevel=debuglevel)
3787 ydlh = YoutubeDLHandler(self.params, debuglevel=debuglevel)
fca6dba8 3788 redirect_handler = YoutubeDLRedirectHandler()
f9934b96 3789 data_handler = urllib.request.DataHandler()
6240b0a2
JMF
3790
3791 # When passing our own FileHandler instance, build_opener won't add the
3792 # default FileHandler and allows us to disable the file protocol, which
3793 # can be used for malicious purposes (see
067aa17e 3794 # https://github.com/ytdl-org/youtube-dl/issues/8227)
ac668111 3795 file_handler = urllib.request.FileHandler()
6240b0a2
JMF
3796
3797 def file_open(*args, **kwargs):
ac668111 3798 raise urllib.error.URLError('file:// scheme is explicitly disabled in yt-dlp for security reasons')
6240b0a2
JMF
3799 file_handler.file_open = file_open
3800
ac668111 3801 opener = urllib.request.build_opener(
fca6dba8 3802 proxy_handler, https_handler, cookie_processor, ydlh, redirect_handler, data_handler, file_handler)
2461f79d 3803
dca08720
PH
3804 # Delete the default user-agent header, which would otherwise apply in
3805 # cases where our custom HTTP handler doesn't come into play
067aa17e 3806 # (See https://github.com/ytdl-org/youtube-dl/issues/1309 for details)
dca08720
PH
3807 opener.addheaders = []
3808 self._opener = opener
62fec3b2
PH
3809
3810 def encode(self, s):
3811 if isinstance(s, bytes):
3812 return s # Already encoded
3813
3814 try:
3815 return s.encode(self.get_encoding())
3816 except UnicodeEncodeError as err:
3817 err.reason = err.reason + '. Check your system encoding configuration or use the --encoding option.'
3818 raise
3819
3820 def get_encoding(self):
3821 encoding = self.params.get('encoding')
3822 if encoding is None:
3823 encoding = preferredencoding()
3824 return encoding
ec82d85a 3825
e08a85d8 3826 def _write_info_json(self, label, ie_result, infofn, overwrite=None):
cb96c5be 3827 ''' Write infojson and returns True = written, 'exists' = Already exists, False = skip, None = error '''
e08a85d8 3828 if overwrite is None:
3829 overwrite = self.params.get('overwrites', True)
80c03fa9 3830 if not self.params.get('writeinfojson'):
3831 return False
3832 elif not infofn:
3833 self.write_debug(f'Skipping writing {label} infojson')
3834 return False
3835 elif not self._ensure_dir_exists(infofn):
3836 return None
e08a85d8 3837 elif not overwrite and os.path.exists(infofn):
80c03fa9 3838 self.to_screen(f'[info] {label.title()} metadata is already present')
cb96c5be 3839 return 'exists'
3840
3841 self.to_screen(f'[info] Writing {label} metadata as JSON to: {infofn}')
3842 try:
3843 write_json_file(self.sanitize_info(ie_result, self.params.get('clean_infojson', True)), infofn)
3844 return True
86e5f3ed 3845 except OSError:
cb96c5be 3846 self.report_error(f'Cannot write {label} metadata to JSON file {infofn}')
3847 return None
80c03fa9 3848
3849 def _write_description(self, label, ie_result, descfn):
3850 ''' Write description and returns True = written, False = skip, None = error '''
3851 if not self.params.get('writedescription'):
3852 return False
3853 elif not descfn:
3854 self.write_debug(f'Skipping writing {label} description')
3855 return False
3856 elif not self._ensure_dir_exists(descfn):
3857 return None
3858 elif not self.params.get('overwrites', True) and os.path.exists(descfn):
3859 self.to_screen(f'[info] {label.title()} description is already present')
3860 elif ie_result.get('description') is None:
3861 self.report_warning(f'There\'s no {label} description to write')
3862 return False
3863 else:
3864 try:
3865 self.to_screen(f'[info] Writing {label} description to: {descfn}')
86e5f3ed 3866 with open(encodeFilename(descfn), 'w', encoding='utf-8') as descfile:
80c03fa9 3867 descfile.write(ie_result['description'])
86e5f3ed 3868 except OSError:
80c03fa9 3869 self.report_error(f'Cannot write {label} description file {descfn}')
3870 return None
3871 return True
3872
3873 def _write_subtitles(self, info_dict, filename):
3874 ''' Write subtitles to file and return list of (sub_filename, final_sub_filename); or None if error'''
3875 ret = []
3876 subtitles = info_dict.get('requested_subtitles')
3877 if not subtitles or not (self.params.get('writesubtitles') or self.params.get('writeautomaticsub')):
3878 # subtitles download errors are already managed as troubles in relevant IE
3879 # that way it will silently go on when used with unsupporting IE
3880 return ret
3881
3882 sub_filename_base = self.prepare_filename(info_dict, 'subtitle')
3883 if not sub_filename_base:
3884 self.to_screen('[info] Skipping writing video subtitles')
3885 return ret
3886 for sub_lang, sub_info in subtitles.items():
3887 sub_format = sub_info['ext']
3888 sub_filename = subtitles_filename(filename, sub_lang, sub_format, info_dict.get('ext'))
3889 sub_filename_final = subtitles_filename(sub_filename_base, sub_lang, sub_format, info_dict.get('ext'))
e04938ab 3890 existing_sub = self.existing_file((sub_filename_final, sub_filename))
3891 if existing_sub:
80c03fa9 3892 self.to_screen(f'[info] Video subtitle {sub_lang}.{sub_format} is already present')
e04938ab 3893 sub_info['filepath'] = existing_sub
3894 ret.append((existing_sub, sub_filename_final))
80c03fa9 3895 continue
3896
3897 self.to_screen(f'[info] Writing video subtitles to: {sub_filename}')
3898 if sub_info.get('data') is not None:
3899 try:
3900 # Use newline='' to prevent conversion of newline characters
3901 # See https://github.com/ytdl-org/youtube-dl/issues/10268
86e5f3ed 3902 with open(sub_filename, 'w', encoding='utf-8', newline='') as subfile:
80c03fa9 3903 subfile.write(sub_info['data'])
3904 sub_info['filepath'] = sub_filename
3905 ret.append((sub_filename, sub_filename_final))
3906 continue
86e5f3ed 3907 except OSError:
80c03fa9 3908 self.report_error(f'Cannot write video subtitles file {sub_filename}')
3909 return None
3910
3911 try:
3912 sub_copy = sub_info.copy()
3913 sub_copy.setdefault('http_headers', info_dict.get('http_headers'))
3914 self.dl(sub_filename, sub_copy, subtitle=True)
3915 sub_info['filepath'] = sub_filename
3916 ret.append((sub_filename, sub_filename_final))
6020e05d 3917 except (DownloadError, ExtractorError, IOError, OSError, ValueError) + network_exceptions as err:
c70c418d 3918 msg = f'Unable to download video subtitles for {sub_lang!r}: {err}'
6020e05d 3919 if self.params.get('ignoreerrors') is not True: # False or 'only_download'
c70c418d 3920 if not self.params.get('ignoreerrors'):
3921 self.report_error(msg)
3922 raise DownloadError(msg)
3923 self.report_warning(msg)
519804a9 3924 return ret
80c03fa9 3925
3926 def _write_thumbnails(self, label, info_dict, filename, thumb_filename_base=None):
3927 ''' Write thumbnails to file and return list of (thumb_filename, final_thumb_filename) '''
6c4fd172 3928 write_all = self.params.get('write_all_thumbnails', False)
80c03fa9 3929 thumbnails, ret = [], []
6c4fd172 3930 if write_all or self.params.get('writethumbnail', False):
0202b52a 3931 thumbnails = info_dict.get('thumbnails') or []
6c4fd172 3932 multiple = write_all and len(thumbnails) > 1
ec82d85a 3933
80c03fa9 3934 if thumb_filename_base is None:
3935 thumb_filename_base = filename
3936 if thumbnails and not thumb_filename_base:
3937 self.write_debug(f'Skipping writing {label} thumbnail')
3938 return ret
3939
dd0228ce 3940 for idx, t in list(enumerate(thumbnails))[::-1]:
80c03fa9 3941 thumb_ext = (f'{t["id"]}.' if multiple else '') + determine_ext(t['url'], 'jpg')
aa9369a2 3942 thumb_display_id = f'{label} thumbnail {t["id"]}'
80c03fa9 3943 thumb_filename = replace_extension(filename, thumb_ext, info_dict.get('ext'))
3944 thumb_filename_final = replace_extension(thumb_filename_base, thumb_ext, info_dict.get('ext'))
ec82d85a 3945
e04938ab 3946 existing_thumb = self.existing_file((thumb_filename_final, thumb_filename))
3947 if existing_thumb:
aa9369a2 3948 self.to_screen('[info] %s is already present' % (
3949 thumb_display_id if multiple else f'{label} thumbnail').capitalize())
e04938ab 3950 t['filepath'] = existing_thumb
3951 ret.append((existing_thumb, thumb_filename_final))
ec82d85a 3952 else:
80c03fa9 3953 self.to_screen(f'[info] Downloading {thumb_display_id} ...')
ec82d85a 3954 try:
297e9952 3955 uf = self.urlopen(sanitized_Request(t['url'], headers=t.get('http_headers', {})))
80c03fa9 3956 self.to_screen(f'[info] Writing {thumb_display_id} to: {thumb_filename}')
d3d89c32 3957 with open(encodeFilename(thumb_filename), 'wb') as thumbf:
ec82d85a 3958 shutil.copyfileobj(uf, thumbf)
80c03fa9 3959 ret.append((thumb_filename, thumb_filename_final))
885cc0b7 3960 t['filepath'] = thumb_filename
3158150c 3961 except network_exceptions as err:
dd0228ce 3962 thumbnails.pop(idx)
80c03fa9 3963 self.report_warning(f'Unable to download {thumb_display_id}: {err}')
6c4fd172 3964 if ret and not write_all:
3965 break
0202b52a 3966 return ret