]> jfr.im git - yt-dlp.git/blame - yt_dlp/YoutubeDL.py
[ffmpeg] Fix features detection
[yt-dlp.git] / yt_dlp / YoutubeDL.py
CommitLineData
cc52de43 1#!/usr/bin/env python3
26e63931 2import collections
31bd3925 3import contextlib
9d2ecdbc 4import datetime
c1c9a79c 5import errno
31bd3925 6import fileinput
b5ae35ee 7import functools
8222d8de 8import io
b82f815f 9import itertools
8694c600 10import json
62fec3b2 11import locale
083c9df9 12import operator
8222d8de 13import os
dca08720 14import platform
f8271158 15import random
8222d8de
JMF
16import re
17import shutil
dca08720 18import subprocess
8222d8de 19import sys
21cd8fae 20import tempfile
8222d8de 21import time
67134eab 22import tokenize
8222d8de 23import traceback
524e2e4f 24import unicodedata
f9934b96 25import urllib.request
961ea474
S
26from string import ascii_letters
27
f8271158 28from .cache import Cache
8c25f81b 29from .compat import (
003c69a8 30 compat_get_terminal_size,
e9c0cdd3 31 compat_os_name,
7d1eb38a 32 compat_shlex_quote,
ce02ed60
PH
33 compat_str,
34 compat_urllib_error,
35 compat_urllib_request,
819e0531 36 windows_enable_vt_mode,
8c25f81b 37)
982ee69a 38from .cookies import load_cookies
f8271158 39from .downloader import FFmpegFD, get_suitable_downloader, shorten_protocol_name
40from .downloader.rtmp import rtmpdump_version
41from .extractor import _LAZY_LOADER
42from .extractor import _PLUGIN_CLASSES as plugin_extractors
43from .extractor import gen_extractor_classes, get_info_extractor
44from .extractor.openload import PhantomJSwrapper
45from .minicurses import format_text
46from .postprocessor import _PLUGIN_CLASSES as plugin_postprocessors
47from .postprocessor import (
48 EmbedThumbnailPP,
49 FFmpegFixupDuplicateMoovPP,
50 FFmpegFixupDurationPP,
51 FFmpegFixupM3u8PP,
52 FFmpegFixupM4aPP,
53 FFmpegFixupStretchedPP,
54 FFmpegFixupTimestampPP,
55 FFmpegMergerPP,
56 FFmpegPostProcessor,
57 MoveFilesAfterDownloadPP,
58 get_postprocessor,
59)
60from .update import detect_variant
8c25f81b 61from .utils import (
f8271158 62 DEFAULT_OUTTMPL,
63 LINK_TEMPLATES,
64 NO_DEFAULT,
65 OUTTMPL_TYPES,
66 POSTPROCESS_WHEN,
67 STR_FORMAT_RE_TMPL,
68 STR_FORMAT_TYPES,
69 ContentTooShortError,
70 DateRange,
71 DownloadCancelled,
72 DownloadError,
73 EntryNotInPlaylist,
74 ExistingVideoReached,
75 ExtractorError,
76 GeoRestrictedError,
77 HEADRequest,
78 InAdvancePagedList,
79 ISO3166Utils,
80 LazyList,
81 MaxDownloadsReached,
19a03940 82 Namespace,
f8271158 83 PagedList,
84 PerRequestProxyHandler,
85 Popen,
86 PostProcessingError,
87 ReExtractInfo,
88 RejectedVideoReached,
89 SameFileError,
90 UnavailableVideoError,
91 YoutubeDLCookieProcessor,
92 YoutubeDLHandler,
93 YoutubeDLRedirectHandler,
eedb7ba5
S
94 age_restricted,
95 args_to_str,
ce02ed60 96 date_from_str,
ce02ed60 97 determine_ext,
b5559424 98 determine_protocol,
c0384f22 99 encode_compat_str,
ce02ed60 100 encodeFilename,
a06916d9 101 error_to_compat_str,
590bc6f6 102 expand_path,
90137ca4 103 filter_dict,
e29663c6 104 float_or_none,
02dbf93f 105 format_bytes,
e0fd9573 106 format_decimal_suffix,
f8271158 107 format_field,
525ef922 108 formatSeconds,
0bb322b9 109 get_domain,
c9969434 110 int_or_none,
732044af 111 iri_to_uri,
34921b43 112 join_nonempty,
ce02ed60 113 locked_file,
0202b52a 114 make_dir,
dca08720 115 make_HTTPS_handler,
8b7539d2 116 merge_headers,
3158150c 117 network_exceptions,
ec11a9f4 118 number_of_digits,
cd6fc19e 119 orderedSet,
083c9df9 120 parse_filesize,
dca08720 121 platform_name,
ce02ed60 122 preferredencoding,
eedb7ba5 123 prepend_extension,
51fb4995 124 register_socks_protocols,
3efb96a6 125 remove_terminal_sequences,
cfb56d1a 126 render_table,
eedb7ba5 127 replace_extension,
ce02ed60 128 sanitize_filename,
1bb5c511 129 sanitize_path,
dcf77cf1 130 sanitize_url,
67dda517 131 sanitized_Request,
e5660ee6 132 std_headers,
1211bb6d 133 str_or_none,
e29663c6 134 strftime_or_none,
ce02ed60 135 subtitles_filename,
819e0531 136 supports_terminal_sequences,
f2ebc5c7 137 timetuple_from_msec,
732044af 138 to_high_limit_path,
324ad820 139 traverse_obj,
6033d980 140 try_get,
29eb5174 141 url_basename,
7d1eb38a 142 variadic,
58b1f00d 143 version_tuple,
ce02ed60
PH
144 write_json_file,
145 write_string,
4f026faf 146)
f8271158 147from .version import RELEASE_GIT_HEAD, __version__
8222d8de 148
e9c0cdd3
YCH
149if compat_os_name == 'nt':
150 import ctypes
151
2459b6e1 152
86e5f3ed 153class YoutubeDL:
8222d8de
JMF
154 """YoutubeDL class.
155
156 YoutubeDL objects are the ones responsible of downloading the
157 actual video file and writing it to disk if the user has requested
158 it, among some other tasks. In most cases there should be one per
159 program. As, given a video URL, the downloader doesn't know how to
160 extract all the needed information, task that InfoExtractors do, it
161 has to pass the URL to one of them.
162
163 For this, YoutubeDL objects have a method that allows
164 InfoExtractors to be registered in a given order. When it is passed
165 a URL, the YoutubeDL object handles it to the first InfoExtractor it
166 finds that reports being able to handle it. The InfoExtractor extracts
167 all the information about the video or videos the URL refers to, and
168 YoutubeDL process the extracted information, possibly using a File
169 Downloader to download the video.
170
171 YoutubeDL objects accept a lot of parameters. In order not to saturate
172 the object constructor with arguments, it receives a dictionary of
173 options instead. These options are available through the params
174 attribute for the InfoExtractors to use. The YoutubeDL also
175 registers itself as the downloader in charge for the InfoExtractors
176 that are added to it, so this is a "mutual registration".
177
178 Available options:
179
180 username: Username for authentication purposes.
181 password: Password for authentication purposes.
180940e0 182 videopassword: Password for accessing a video.
1da50aa3
S
183 ap_mso: Adobe Pass multiple-system operator identifier.
184 ap_username: Multiple-system operator account username.
185 ap_password: Multiple-system operator account password.
8222d8de
JMF
186 usenetrc: Use netrc for authentication instead.
187 verbose: Print additional info to stdout.
188 quiet: Do not print messages to stdout.
ad8915b7 189 no_warnings: Do not print out anything for warnings.
bb66c247 190 forceprint: A dict with keys WHEN mapped to a list of templates to
191 print to stdout. The allowed keys are video or any of the
192 items in utils.POSTPROCESS_WHEN.
ca30f449 193 For compatibility, a single list is also accepted
bb66c247 194 print_to_file: A dict with keys WHEN (same as forceprint) mapped to
195 a list of tuples with (template, filename)
53c18592 196 forceurl: Force printing final URL. (Deprecated)
197 forcetitle: Force printing title. (Deprecated)
198 forceid: Force printing ID. (Deprecated)
199 forcethumbnail: Force printing thumbnail URL. (Deprecated)
200 forcedescription: Force printing description. (Deprecated)
201 forcefilename: Force printing final filename. (Deprecated)
202 forceduration: Force printing duration. (Deprecated)
8694c600 203 forcejson: Force printing info_dict as JSON.
63e0be34
PH
204 dump_single_json: Force printing the info_dict of the whole playlist
205 (or video) as a single JSON line.
c25228e5 206 force_write_download_archive: Force writing download archive regardless
207 of 'skip_download' or 'simulate'.
b7b04c78 208 simulate: Do not download the video files. If unset (or None),
209 simulate only if listsubtitles, listformats or list_thumbnails is used
eb8a4433 210 format: Video format code. see "FORMAT SELECTION" for more details.
093a1710 211 You can also pass a function. The function takes 'ctx' as
212 argument and returns the formats to download.
213 See "build_format_selector" for an implementation
63ad4d43 214 allow_unplayable_formats: Allow unplayable formats to be extracted and downloaded.
b7da73eb 215 ignore_no_formats_error: Ignore "No video formats" error. Usefull for
216 extracting metadata even if the video is not actually
217 available for download (experimental)
0930b11f 218 format_sort: A list of fields by which to sort the video formats.
219 See "Sorting Formats" for more details.
c25228e5 220 format_sort_force: Force the given format_sort. see "Sorting Formats"
221 for more details.
08d30158 222 prefer_free_formats: Whether to prefer video formats with free containers
223 over non-free ones of same quality.
c25228e5 224 allow_multiple_video_streams: Allow multiple video streams to be merged
225 into a single file
226 allow_multiple_audio_streams: Allow multiple audio streams to be merged
227 into a single file
0ba692ac 228 check_formats Whether to test if the formats are downloadable.
9f1a1c36 229 Can be True (check all), False (check none),
230 'selected' (check selected formats),
0ba692ac 231 or None (check only if requested by extractor)
4524baf0 232 paths: Dictionary of output paths. The allowed keys are 'home'
233 'temp' and the keys of OUTTMPL_TYPES (in utils.py)
de6000d9 234 outtmpl: Dictionary of templates for output names. Allowed keys
4524baf0 235 are 'default' and the keys of OUTTMPL_TYPES (in utils.py).
34488702 236 For compatibility with youtube-dl, a single string can also be used
a820dc72
RA
237 outtmpl_na_placeholder: Placeholder for unavailable meta fields.
238 restrictfilenames: Do not allow "&" and spaces in file names
239 trim_file_name: Limit length of filename (extension excluded)
4524baf0 240 windowsfilenames: Force the filenames to be windows compatible
b1940459 241 ignoreerrors: Do not stop on download/postprocessing errors.
242 Can be 'only_download' to ignore only download errors.
243 Default is 'only_download' for CLI, but False for API
26e2805c 244 skip_playlist_after_errors: Number of allowed failures until the rest of
245 the playlist is skipped
d22dec74 246 force_generic_extractor: Force downloader to use the generic extractor
0c3d0f51 247 overwrites: Overwrite all video and metadata files if True,
248 overwrite only non-video files if None
249 and don't overwrite any file if False
34488702 250 For compatibility with youtube-dl,
251 "nooverwrites" may also be used instead
8222d8de
JMF
252 playliststart: Playlist item to start at.
253 playlistend: Playlist item to end at.
c14e88f0 254 playlist_items: Specific indices of playlist to download.
ff815fe6 255 playlistreverse: Download playlist items in reverse order.
75822ca7 256 playlistrandom: Download playlist items in random order.
8222d8de
JMF
257 matchtitle: Download only matching titles.
258 rejecttitle: Reject downloads for matching titles.
8bf9319e 259 logger: Log messages to a logging.Logger instance.
8222d8de 260 logtostderr: Log messages to stderr instead of stdout.
819e0531 261 consoletitle: Display progress in console window's titlebar.
8222d8de
JMF
262 writedescription: Write the video description to a .description file
263 writeinfojson: Write the video description to a .info.json file
75d43ca0 264 clean_infojson: Remove private fields from the infojson
34488702 265 getcomments: Extract video comments. This will not be written to disk
06167fbb 266 unless writeinfojson is also given
1fb07d10 267 writeannotations: Write the video annotations to a .annotations.xml file
8222d8de 268 writethumbnail: Write the thumbnail image to a file
c25228e5 269 allow_playlist_files: Whether to write playlists' description, infojson etc
270 also to disk when using the 'write*' options
ec82d85a 271 write_all_thumbnails: Write all thumbnail formats to files
732044af 272 writelink: Write an internet shortcut file, depending on the
273 current platform (.url/.webloc/.desktop)
274 writeurllink: Write a Windows internet shortcut file (.url)
275 writewebloclink: Write a macOS internet shortcut file (.webloc)
276 writedesktoplink: Write a Linux internet shortcut file (.desktop)
8222d8de 277 writesubtitles: Write the video subtitles to a file
741dd8ea 278 writeautomaticsub: Write the automatically generated subtitles to a file
245524e6 279 allsubtitles: Deprecated - Use subtitleslangs = ['all']
c32b0aab 280 Downloads all the subtitles of the video
0b7f3118 281 (requires writesubtitles or writeautomaticsub)
8222d8de 282 listsubtitles: Lists all available subtitles for the video
a504ced0 283 subtitlesformat: The format code for subtitles
c32b0aab 284 subtitleslangs: List of languages of the subtitles to download (can be regex).
285 The list may contain "all" to refer to all the available
286 subtitles. The language can be prefixed with a "-" to
287 exclude it from the requested languages. Eg: ['all', '-live_chat']
8222d8de
JMF
288 keepvideo: Keep the video file after post-processing
289 daterange: A DateRange object, download only if the upload_date is in the range.
290 skip_download: Skip the actual download of the video file
c35f9e72 291 cachedir: Location of the cache files in the filesystem.
a0e07d31 292 False to disable filesystem cache.
47192f92 293 noplaylist: Download single video instead of a playlist if in doubt.
8dbe9899
PH
294 age_limit: An integer representing the user's age in years.
295 Unsuitable videos for the given age are skipped.
5fe18bdb
PH
296 min_views: An integer representing the minimum view count the video
297 must have in order to not be skipped.
298 Videos without view count information are always
299 downloaded. None for no limit.
300 max_views: An integer representing the maximum view count.
301 Videos that are more popular than that are not
302 downloaded.
303 Videos without view count information are always
304 downloaded. None for no limit.
305 download_archive: File name of a file where all downloads are recorded.
c1c9a79c
PH
306 Videos already present in the file are not downloaded
307 again.
8a51f564 308 break_on_existing: Stop the download process after attempting to download a
309 file that is in the archive.
310 break_on_reject: Stop the download process when encountering a video that
311 has been filtered out.
b222c271 312 break_per_url: Whether break_on_reject and break_on_existing
313 should act on each input URL as opposed to for the entire queue
8a51f564 314 cookiefile: File name where cookies should be read from and dumped to
f59f5ef8
MB
315 cookiesfrombrowser: A tuple containing the name of the browser, the profile
316 name/pathfrom where cookies are loaded, and the name of the
317 keyring. Eg: ('chrome', ) or ('vivaldi', 'default', 'BASICTEXT')
f81c62a6 318 legacyserverconnect: Explicitly allow HTTPS connection to servers that do not
319 support RFC 5746 secure renegotiation
f59f5ef8 320 nocheckcertificate: Do not verify SSL certificates
7e8c0af0
PH
321 prefer_insecure: Use HTTP instead of HTTPS to retrieve information.
322 At the moment, this is only supported by YouTube.
8b7539d2 323 http_headers: A dictionary of custom headers to be used for all requests
a1ee09e8 324 proxy: URL of the proxy server to use
38cce791 325 geo_verification_proxy: URL of the proxy to use for IP address verification
504f20dd 326 on geo-restricted sites.
e344693b 327 socket_timeout: Time to wait for unresponsive hosts, in seconds
0783b09b
PH
328 bidi_workaround: Work around buggy terminals without bidirectional text
329 support, using fridibi
a0ddb8a2 330 debug_printtraffic:Print out sent and received HTTP traffic
91f071af 331 include_ads: Download ads as well (deprecated)
04b4d394
PH
332 default_search: Prepend this string if an input url is not valid.
333 'auto' for elaborate guessing
62fec3b2 334 encoding: Use this encoding instead of the system-specified.
e8ee972c 335 extract_flat: Do not resolve URLs, return the immediate result.
057a5206
PH
336 Pass in 'in_playlist' to only show this behavior for
337 playlist items.
f2ebc5c7 338 wait_for_video: If given, wait for scheduled streams to become available.
339 The value should be a tuple containing the range
340 (min_secs, max_secs) to wait between retries
4f026faf 341 postprocessors: A list of dictionaries, each with an entry
71b640cc 342 * key: The name of the postprocessor. See
7a5c1cfe 343 yt_dlp/postprocessor/__init__.py for a list.
bb66c247 344 * when: When to run the postprocessor. Allowed values are
345 the entries of utils.POSTPROCESS_WHEN
56d868db 346 Assumed to be 'post_process' if not given
b5ae35ee 347 post_hooks: Deprecated - Register a custom postprocessor instead
348 A list of functions that get called as the final step
ab8e5e51
AM
349 for each video file, after all postprocessors have been
350 called. The filename will be passed as the only argument.
71b640cc
PH
351 progress_hooks: A list of functions that get called on download
352 progress, with a dictionary with the entries
5cda4eda 353 * status: One of "downloading", "error", or "finished".
ee69b99a 354 Check this first and ignore unknown values.
3ba7740d 355 * info_dict: The extracted info_dict
71b640cc 356
5cda4eda 357 If status is one of "downloading", or "finished", the
ee69b99a
PH
358 following properties may also be present:
359 * filename: The final filename (always present)
5cda4eda 360 * tmpfilename: The filename we're currently writing to
71b640cc
PH
361 * downloaded_bytes: Bytes on disk
362 * total_bytes: Size of the whole file, None if unknown
5cda4eda
PH
363 * total_bytes_estimate: Guess of the eventual file size,
364 None if unavailable.
365 * elapsed: The number of seconds since download started.
71b640cc
PH
366 * eta: The estimated time in seconds, None if unknown
367 * speed: The download speed in bytes/second, None if
368 unknown
5cda4eda
PH
369 * fragment_index: The counter of the currently
370 downloaded video fragment.
371 * fragment_count: The number of fragments (= individual
372 files that will be merged)
71b640cc
PH
373
374 Progress hooks are guaranteed to be called at least once
375 (with status "finished") if the download is successful.
819e0531 376 postprocessor_hooks: A list of functions that get called on postprocessing
377 progress, with a dictionary with the entries
378 * status: One of "started", "processing", or "finished".
379 Check this first and ignore unknown values.
380 * postprocessor: Name of the postprocessor
381 * info_dict: The extracted info_dict
382
383 Progress hooks are guaranteed to be called at least twice
384 (with status "started" and "finished") if the processing is successful.
45598f15 385 merge_output_format: Extension to use when merging formats.
6b591b29 386 final_ext: Expected final extension; used to detect when the file was
59a7a13e 387 already downloaded and converted
6271f1ca
PH
388 fixup: Automatically correct known faults of the file.
389 One of:
390 - "never": do nothing
391 - "warn": only emit a warning
392 - "detect_or_warn": check whether we can do anything
62cd676c 393 about it, warn otherwise (default)
504f20dd 394 source_address: Client-side IP address to bind to.
6ec6cb4e 395 call_home: Boolean, true iff we are allowed to contact the
7a5c1cfe 396 yt-dlp servers for debugging. (BROKEN)
1cf376f5 397 sleep_interval_requests: Number of seconds to sleep between requests
398 during extraction
7aa589a5
S
399 sleep_interval: Number of seconds to sleep before each download when
400 used alone or a lower bound of a range for randomized
401 sleep before each download (minimum possible number
402 of seconds to sleep) when used along with
403 max_sleep_interval.
404 max_sleep_interval:Upper bound of a range for randomized sleep before each
405 download (maximum possible number of seconds to sleep).
406 Must only be used along with sleep_interval.
407 Actual sleep time will be a random float from range
408 [sleep_interval; max_sleep_interval].
1cf376f5 409 sleep_interval_subtitles: Number of seconds to sleep before each subtitle download
cfb56d1a
PH
410 listformats: Print an overview of available video formats and exit.
411 list_thumbnails: Print a table of all thumbnails and exit.
347de493
PH
412 match_filter: A function that gets called with the info_dict of
413 every video.
414 If it returns a message, the video is ignored.
415 If it returns None, the video is downloaded.
416 match_filter_func in utils.py is one example for this.
7e5db8c9 417 no_color: Do not emit color codes in output.
0a840f58 418 geo_bypass: Bypass geographic restriction via faking X-Forwarded-For
504f20dd 419 HTTP header
0a840f58 420 geo_bypass_country:
773f291d
S
421 Two-letter ISO 3166-2 country code that will be used for
422 explicit geographic restriction bypassing via faking
504f20dd 423 X-Forwarded-For HTTP header
5f95927a
S
424 geo_bypass_ip_block:
425 IP range in CIDR notation that will be used similarly to
504f20dd 426 geo_bypass_country
71b640cc 427
85729c51 428 The following options determine which downloader is picked:
52a8a1e1 429 external_downloader: A dictionary of protocol keys and the executable of the
430 external downloader to use for it. The allowed protocols
431 are default|http|ftp|m3u8|dash|rtsp|rtmp|mms.
432 Set the value to 'native' to use the native downloader
433 hls_prefer_native: Deprecated - Use external_downloader = {'m3u8': 'native'}
434 or {'m3u8': 'ffmpeg'} instead.
435 Use the native HLS downloader instead of ffmpeg/avconv
bf09af3a
S
436 if True, otherwise use ffmpeg/avconv if False, otherwise
437 use downloader suggested by extractor if None.
53ed7066 438 compat_opts: Compatibility options. See "Differences in default behavior".
3acf6d38 439 The following options do not work when used through the API:
b5ae35ee 440 filename, abort-on-error, multistreams, no-live-chat, format-sort
dac5df5a 441 no-clean-infojson, no-playlist-metafiles, no-keep-subs, no-attach-info-json.
e4f02757 442 Refer __init__.py for their implementation
819e0531 443 progress_template: Dictionary of templates for progress outputs.
444 Allowed keys are 'download', 'postprocess',
445 'download-title' (console title) and 'postprocess-title'.
446 The template is mapped on a dictionary with keys 'progress' and 'info'
fe7e0c98 447
8222d8de 448 The following parameters are not used by YoutubeDL itself, they are used by
7a5c1cfe 449 the downloader (see yt_dlp/downloader/common.py):
51d9739f 450 nopart, updatetime, buffersize, ratelimit, throttledratelimit, min_filesize,
205a0654
EH
451 max_filesize, test, noresizebuffer, retries, file_access_retries, fragment_retries,
452 continuedl, noprogress, xattr_set_filesize, hls_use_mpegts, http_chunk_size,
59a7a13e 453 external_downloader_args, concurrent_fragment_downloads.
76b1bd67
JMF
454
455 The following options are used by the post processors:
d4a24f40 456 prefer_ffmpeg: If False, use avconv instead of ffmpeg if both are available,
e4172ac9 457 otherwise prefer ffmpeg. (avconv support is deprecated)
c0b7d117
S
458 ffmpeg_location: Location of the ffmpeg/avconv binary; either the path
459 to the binary or its containing directory.
43820c03 460 postprocessor_args: A dictionary of postprocessor/executable keys (in lower case)
34488702 461 and a list of additional command-line arguments for the
462 postprocessor/executable. The dict can also have "PP+EXE" keys
463 which are used when the given exe is used by the given PP.
464 Use 'default' as the name for arguments to passed to all PP
465 For compatibility with youtube-dl, a single list of args
466 can also be used
e409895f 467
468 The following options are used by the extractors:
62bff2c1 469 extractor_retries: Number of times to retry for known errors
470 dynamic_mpd: Whether to process dynamic DASH manifests (default: True)
e409895f 471 hls_split_discontinuity: Split HLS playlists to different formats at
62bff2c1 472 discontinuities such as ad breaks (default: False)
5d3a0e79 473 extractor_args: A dictionary of arguments to be passed to the extractors.
474 See "EXTRACTOR ARGUMENTS" for details.
475 Eg: {'youtube': {'skip': ['dash', 'hls']}}
88f23a18 476 mark_watched: Mark videos watched (even with --simulate). Only for YouTube
5d3a0e79 477 youtube_include_dash_manifest: Deprecated - Use extractor_args instead.
478 If True (default), DASH manifests and related
62bff2c1 479 data will be downloaded and processed by extractor.
480 You can reduce network I/O by disabling it if you don't
481 care about DASH. (only for youtube)
5d3a0e79 482 youtube_include_hls_manifest: Deprecated - Use extractor_args instead.
483 If True (default), HLS manifests and related
62bff2c1 484 data will be downloaded and processed by extractor.
485 You can reduce network I/O by disabling it if you don't
486 care about HLS. (only for youtube)
8222d8de
JMF
487 """
488
86e5f3ed 489 _NUMERIC_FIELDS = {
c9969434 490 'width', 'height', 'tbr', 'abr', 'asr', 'vbr', 'fps', 'filesize', 'filesize_approx',
e6f21b3d 491 'timestamp', 'release_timestamp',
c9969434
S
492 'duration', 'view_count', 'like_count', 'dislike_count', 'repost_count',
493 'average_rating', 'comment_count', 'age_limit',
494 'start_time', 'end_time',
495 'chapter_number', 'season_number', 'episode_number',
496 'track_number', 'disc_number', 'release_year',
86e5f3ed 497 }
c9969434 498
6db9c4d5 499 _format_fields = {
500 # NB: Keep in sync with the docstring of extractor/common.py
a44ca5a4 501 'url', 'manifest_url', 'manifest_stream_number', 'ext', 'format', 'format_id', 'format_note',
6db9c4d5 502 'width', 'height', 'resolution', 'dynamic_range', 'tbr', 'abr', 'acodec', 'asr',
503 'vbr', 'fps', 'vcodec', 'container', 'filesize', 'filesize_approx',
504 'player_url', 'protocol', 'fragment_base_url', 'fragments', 'is_from_start',
505 'preference', 'language', 'language_preference', 'quality', 'source_preference',
506 'http_headers', 'stretched_ratio', 'no_resume', 'has_drm', 'downloader_options',
507 'page_url', 'app', 'play_path', 'tc_url', 'flash_version', 'rtmp_live', 'rtmp_conn', 'rtmp_protocol', 'rtmp_real_time'
508 }
48ee10ee 509 _format_selection_exts = {
510 'audio': {'m4a', 'mp3', 'ogg', 'aac'},
511 'video': {'mp4', 'flv', 'webm', '3gp'},
512 'storyboards': {'mhtml'},
513 }
514
3511266b 515 def __init__(self, params=None, auto_init=True):
883d4b1e 516 """Create a FileDownloader object with the given options.
517 @param auto_init Whether to load the default extractors and print header (if verbose).
49a57e70 518 Set to 'no_verbose_header' to not print the header
883d4b1e 519 """
e9f9a10f
JMF
520 if params is None:
521 params = {}
592b7485 522 self.params = params
8b7491c8 523 self._ies = {}
56c73665 524 self._ies_instances = {}
1e43a6f7 525 self._pps = {k: [] for k in POSTPROCESS_WHEN}
b35496d8 526 self._printed_messages = set()
1cf376f5 527 self._first_webpage_request = True
ab8e5e51 528 self._post_hooks = []
933605d7 529 self._progress_hooks = []
819e0531 530 self._postprocessor_hooks = []
8222d8de
JMF
531 self._download_retcode = 0
532 self._num_downloads = 0
9c906919 533 self._num_videos = 0
592b7485 534 self._playlist_level = 0
535 self._playlist_urls = set()
a0e07d31 536 self.cache = Cache(self)
34308b30 537
819e0531 538 windows_enable_vt_mode()
cf4f42cb 539 self._out_files = {
540 'error': sys.stderr,
541 'print': sys.stderr if self.params.get('logtostderr') else sys.stdout,
542 'console': None if compat_os_name == 'nt' else next(
543 filter(supports_terminal_sequences, (sys.stderr, sys.stdout)), None)
544 }
545 self._out_files['screen'] = sys.stderr if self.params.get('quiet') else self._out_files['print']
ec11a9f4 546 self._allow_colors = {
cf4f42cb 547 type_: not self.params.get('no_color') and supports_terminal_sequences(self._out_files[type_])
548 for type_ in ('screen', 'error')
ec11a9f4 549 }
819e0531 550
a61f4b28 551 if sys.version_info < (3, 6):
552 self.report_warning(
0181adef 553 'Python version %d.%d is not supported! Please update to Python 3.6 or above' % sys.version_info[:2])
a61f4b28 554
88acdbc2 555 if self.params.get('allow_unplayable_formats'):
556 self.report_warning(
ec11a9f4 557 f'You have asked for {self._format_err("UNPLAYABLE", self.Styles.EMPHASIS)} formats to be listed/downloaded. '
819e0531 558 'This is a developer option intended for debugging. \n'
559 ' If you experience any issues while using this option, '
ec11a9f4 560 f'{self._format_err("DO NOT", self.Styles.ERROR)} open a bug report')
88acdbc2 561
be5df5ee
S
562 def check_deprecated(param, option, suggestion):
563 if self.params.get(param) is not None:
86e5f3ed 564 self.report_warning(f'{option} is deprecated. Use {suggestion} instead')
be5df5ee
S
565 return True
566 return False
567
568 if check_deprecated('cn_verification_proxy', '--cn-verification-proxy', '--geo-verification-proxy'):
38cce791
YCH
569 if self.params.get('geo_verification_proxy') is None:
570 self.params['geo_verification_proxy'] = self.params['cn_verification_proxy']
571
0d1bb027 572 check_deprecated('autonumber', '--auto-number', '-o "%(autonumber)s-%(title)s.%(ext)s"')
573 check_deprecated('usetitle', '--title', '-o "%(title)s-%(id)s.%(ext)s"')
53ed7066 574 check_deprecated('useid', '--id', '-o "%(id)s.%(ext)s"')
0d1bb027 575
49a57e70 576 for msg in self.params.get('_warnings', []):
0d1bb027 577 self.report_warning(msg)
ee8dd27a 578 for msg in self.params.get('_deprecation_warnings', []):
579 self.deprecation_warning(msg)
0d1bb027 580
ec11a9f4 581 if 'list-formats' in self.params.get('compat_opts', []):
582 self.params['listformats_table'] = False
583
b5ae35ee 584 if 'overwrites' not in self.params and self.params.get('nooverwrites') is not None:
b868936c 585 # nooverwrites was unnecessarily changed to overwrites
586 # in 0c3d0f51778b153f65c21906031c2e091fcfb641
587 # This ensures compatibility with both keys
588 self.params['overwrites'] = not self.params['nooverwrites']
b5ae35ee 589 elif self.params.get('overwrites') is None:
590 self.params.pop('overwrites', None)
b868936c 591 else:
592 self.params['nooverwrites'] = not self.params['overwrites']
b9d973be 593
455a15e2 594 self.params.setdefault('forceprint', {})
595 self.params.setdefault('print_to_file', {})
bb66c247 596
597 # Compatibility with older syntax
ca30f449 598 if not isinstance(params['forceprint'], dict):
455a15e2 599 self.params['forceprint'] = {'video': params['forceprint']}
ca30f449 600
455a15e2 601 if self.params.get('bidi_workaround', False):
1c088fa8
PH
602 try:
603 import pty
604 master, slave = pty.openpty()
003c69a8 605 width = compat_get_terminal_size().columns
1c088fa8
PH
606 if width is None:
607 width_args = []
608 else:
609 width_args = ['-w', str(width)]
5d681e96 610 sp_kwargs = dict(
1c088fa8
PH
611 stdin=subprocess.PIPE,
612 stdout=slave,
cf4f42cb 613 stderr=self._out_files['error'])
5d681e96 614 try:
d3c93ec2 615 self._output_process = Popen(['bidiv'] + width_args, **sp_kwargs)
5d681e96 616 except OSError:
d3c93ec2 617 self._output_process = Popen(['fribidi', '-c', 'UTF-8'] + width_args, **sp_kwargs)
5d681e96 618 self._output_channel = os.fdopen(master, 'rb')
1c088fa8 619 except OSError as ose:
66e7ace1 620 if ose.errno == errno.ENOENT:
49a57e70 621 self.report_warning(
622 'Could not find fribidi executable, ignoring --bidi-workaround. '
623 'Make sure that fribidi is an executable file in one of the directories in your $PATH.')
1c088fa8
PH
624 else:
625 raise
0783b09b 626
97ec5bc5 627 if auto_init:
628 if auto_init != 'no_verbose_header':
629 self.print_debug_header()
630 self.add_default_info_extractors()
631
3089bc74
S
632 if (sys.platform != 'win32'
633 and sys.getfilesystemencoding() in ['ascii', 'ANSI_X3.4-1968']
455a15e2 634 and not self.params.get('restrictfilenames', False)):
e9137224 635 # Unicode filesystem API will throw errors (#1474, #13027)
34308b30 636 self.report_warning(
6febd1c1 637 'Assuming --restrict-filenames since file system encoding '
1b725173 638 'cannot encode all characters. '
6febd1c1 639 'Set the LC_ALL environment variable to fix this.')
4a98cdbf 640 self.params['restrictfilenames'] = True
34308b30 641
de6000d9 642 self.outtmpl_dict = self.parse_outtmpl()
486dd09e 643
187986a8 644 # Creating format selector here allows us to catch syntax errors before the extraction
645 self.format_selector = (
fa9f30b8 646 self.params.get('format') if self.params.get('format') in (None, '-')
093a1710 647 else self.params['format'] if callable(self.params['format'])
187986a8 648 else self.build_format_selector(self.params['format']))
649
8b7539d2 650 # Set http_headers defaults according to std_headers
651 self.params['http_headers'] = merge_headers(std_headers, self.params.get('http_headers', {}))
652
013b50b7 653 hooks = {
654 'post_hooks': self.add_post_hook,
655 'progress_hooks': self.add_progress_hook,
656 'postprocessor_hooks': self.add_postprocessor_hook,
657 }
658 for opt, fn in hooks.items():
659 for ph in self.params.get(opt, []):
660 fn(ph)
71b640cc 661
5bfc8bee 662 for pp_def_raw in self.params.get('postprocessors', []):
663 pp_def = dict(pp_def_raw)
664 when = pp_def.pop('when', 'post_process')
665 self.add_post_processor(
f9934b96 666 get_postprocessor(pp_def.pop('key'))(self, **pp_def),
5bfc8bee 667 when=when)
668
97ec5bc5 669 self._setup_opener()
51fb4995
YCH
670 register_socks_protocols()
671
ed39cac5 672 def preload_download_archive(fn):
673 """Preload the archive, if any is specified"""
674 if fn is None:
675 return False
49a57e70 676 self.write_debug(f'Loading archive file {fn!r}')
ed39cac5 677 try:
678 with locked_file(fn, 'r', encoding='utf-8') as archive_file:
679 for line in archive_file:
680 self.archive.add(line.strip())
86e5f3ed 681 except OSError as ioe:
ed39cac5 682 if ioe.errno != errno.ENOENT:
683 raise
684 return False
685 return True
686
687 self.archive = set()
688 preload_download_archive(self.params.get('download_archive'))
689
7d4111ed
PH
690 def warn_if_short_id(self, argv):
691 # short YouTube ID starting with dash?
692 idxs = [
693 i for i, a in enumerate(argv)
694 if re.match(r'^-[0-9A-Za-z_-]{10}$', a)]
695 if idxs:
696 correct_argv = (
7a5c1cfe 697 ['yt-dlp']
3089bc74
S
698 + [a for i, a in enumerate(argv) if i not in idxs]
699 + ['--'] + [argv[i] for i in idxs]
7d4111ed
PH
700 )
701 self.report_warning(
702 'Long argument string detected. '
49a57e70 703 'Use -- to separate parameters and URLs, like this:\n%s' %
7d4111ed
PH
704 args_to_str(correct_argv))
705
8222d8de
JMF
706 def add_info_extractor(self, ie):
707 """Add an InfoExtractor object to the end of the list."""
8b7491c8 708 ie_key = ie.ie_key()
709 self._ies[ie_key] = ie
e52d7f85 710 if not isinstance(ie, type):
8b7491c8 711 self._ies_instances[ie_key] = ie
e52d7f85 712 ie.set_downloader(self)
8222d8de 713
8b7491c8 714 def _get_info_extractor_class(self, ie_key):
715 ie = self._ies.get(ie_key)
716 if ie is None:
717 ie = get_info_extractor(ie_key)
718 self.add_info_extractor(ie)
719 return ie
720
56c73665
JMF
721 def get_info_extractor(self, ie_key):
722 """
723 Get an instance of an IE with name ie_key, it will try to get one from
724 the _ies list, if there's no instance it will create a new one and add
725 it to the extractor list.
726 """
727 ie = self._ies_instances.get(ie_key)
728 if ie is None:
729 ie = get_info_extractor(ie_key)()
730 self.add_info_extractor(ie)
731 return ie
732
023fa8c4
JMF
733 def add_default_info_extractors(self):
734 """
735 Add the InfoExtractors returned by gen_extractors to the end of the list
736 """
e52d7f85 737 for ie in gen_extractor_classes():
023fa8c4
JMF
738 self.add_info_extractor(ie)
739
56d868db 740 def add_post_processor(self, pp, when='post_process'):
8222d8de 741 """Add a PostProcessor object to the end of the chain."""
5bfa4862 742 self._pps[when].append(pp)
8222d8de
JMF
743 pp.set_downloader(self)
744
ab8e5e51
AM
745 def add_post_hook(self, ph):
746 """Add the post hook"""
747 self._post_hooks.append(ph)
748
933605d7 749 def add_progress_hook(self, ph):
819e0531 750 """Add the download progress hook"""
933605d7 751 self._progress_hooks.append(ph)
8ab470f1 752
819e0531 753 def add_postprocessor_hook(self, ph):
754 """Add the postprocessing progress hook"""
755 self._postprocessor_hooks.append(ph)
5bfc8bee 756 for pps in self._pps.values():
757 for pp in pps:
758 pp.add_progress_hook(ph)
819e0531 759
1c088fa8 760 def _bidi_workaround(self, message):
5d681e96 761 if not hasattr(self, '_output_channel'):
1c088fa8
PH
762 return message
763
5d681e96 764 assert hasattr(self, '_output_process')
11b85ce6 765 assert isinstance(message, compat_str)
6febd1c1
PH
766 line_count = message.count('\n') + 1
767 self._output_process.stdin.write((message + '\n').encode('utf-8'))
5d681e96 768 self._output_process.stdin.flush()
6febd1c1 769 res = ''.join(self._output_channel.readline().decode('utf-8')
9e1a5b84 770 for _ in range(line_count))
6febd1c1 771 return res[:-len('\n')]
1c088fa8 772
b35496d8 773 def _write_string(self, message, out=None, only_once=False):
774 if only_once:
775 if message in self._printed_messages:
776 return
777 self._printed_messages.add(message)
778 write_string(message, out=out, encoding=self.params.get('encoding'))
734f90bb 779
cf4f42cb 780 def to_stdout(self, message, skip_eol=False, quiet=None):
0760b0a7 781 """Print message to stdout"""
cf4f42cb 782 if quiet is not None:
ae6a1b95 783 self.deprecation_warning('"YoutubeDL.to_stdout" no longer accepts the argument quiet. Use "YoutubeDL.to_screen" instead')
cf4f42cb 784 self._write_string(
785 '%s%s' % (self._bidi_workaround(message), ('' if skip_eol else '\n')),
786 self._out_files['print'])
787
788 def to_screen(self, message, skip_eol=False, quiet=None):
789 """Print message to screen if not in quiet mode"""
8bf9319e 790 if self.params.get('logger'):
43afe285 791 self.params['logger'].debug(message)
cf4f42cb 792 return
793 if (self.params.get('quiet') if quiet is None else quiet) and not self.params.get('verbose'):
794 return
795 self._write_string(
796 '%s%s' % (self._bidi_workaround(message), ('' if skip_eol else '\n')),
797 self._out_files['screen'])
8222d8de 798
b35496d8 799 def to_stderr(self, message, only_once=False):
0760b0a7 800 """Print message to stderr"""
11b85ce6 801 assert isinstance(message, compat_str)
8bf9319e 802 if self.params.get('logger'):
43afe285
IB
803 self.params['logger'].error(message)
804 else:
cf4f42cb 805 self._write_string('%s\n' % self._bidi_workaround(message), self._out_files['error'], only_once=only_once)
806
807 def _send_console_code(self, code):
808 if compat_os_name == 'nt' or not self._out_files['console']:
809 return
810 self._write_string(code, self._out_files['console'])
8222d8de 811
1e5b9a95
PH
812 def to_console_title(self, message):
813 if not self.params.get('consoletitle', False):
814 return
3efb96a6 815 message = remove_terminal_sequences(message)
4bede0d8
C
816 if compat_os_name == 'nt':
817 if ctypes.windll.kernel32.GetConsoleWindow():
818 # c_wchar_p() might not be necessary if `message` is
819 # already of type unicode()
820 ctypes.windll.kernel32.SetConsoleTitleW(ctypes.c_wchar_p(message))
cf4f42cb 821 else:
822 self._send_console_code(f'\033]0;{message}\007')
1e5b9a95 823
bdde425c 824 def save_console_title(self):
cf4f42cb 825 if not self.params.get('consoletitle') or self.params.get('simulate'):
bdde425c 826 return
592b7485 827 self._send_console_code('\033[22;0t') # Save the title on stack
bdde425c
PH
828
829 def restore_console_title(self):
cf4f42cb 830 if not self.params.get('consoletitle') or self.params.get('simulate'):
bdde425c 831 return
592b7485 832 self._send_console_code('\033[23;0t') # Restore the title from stack
bdde425c
PH
833
834 def __enter__(self):
835 self.save_console_title()
836 return self
837
838 def __exit__(self, *args):
839 self.restore_console_title()
f89197d7 840
dca08720 841 if self.params.get('cookiefile') is not None:
1bab3437 842 self.cookiejar.save(ignore_discard=True, ignore_expires=True)
bdde425c 843
fa9f30b8 844 def trouble(self, message=None, tb=None, is_error=True):
8222d8de
JMF
845 """Determine action to take when a download problem appears.
846
847 Depending on if the downloader has been configured to ignore
848 download errors or not, this method may throw an exception or
849 not when errors are found, after printing the message.
850
fa9f30b8 851 @param tb If given, is additional traceback information
852 @param is_error Whether to raise error according to ignorerrors
8222d8de
JMF
853 """
854 if message is not None:
855 self.to_stderr(message)
856 if self.params.get('verbose'):
857 if tb is None:
858 if sys.exc_info()[0]: # if .trouble has been called from an except block
6febd1c1 859 tb = ''
8222d8de 860 if hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
6febd1c1 861 tb += ''.join(traceback.format_exception(*sys.exc_info()[1].exc_info))
c0384f22 862 tb += encode_compat_str(traceback.format_exc())
8222d8de
JMF
863 else:
864 tb_data = traceback.format_list(traceback.extract_stack())
6febd1c1 865 tb = ''.join(tb_data)
c19bc311 866 if tb:
867 self.to_stderr(tb)
fa9f30b8 868 if not is_error:
869 return
b1940459 870 if not self.params.get('ignoreerrors'):
8222d8de
JMF
871 if sys.exc_info()[0] and hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
872 exc_info = sys.exc_info()[1].exc_info
873 else:
874 exc_info = sys.exc_info()
875 raise DownloadError(message, exc_info)
876 self._download_retcode = 1
877
19a03940 878 Styles = Namespace(
879 HEADERS='yellow',
880 EMPHASIS='light blue',
881 ID='green',
882 DELIM='blue',
883 ERROR='red',
884 WARNING='yellow',
885 SUPPRESS='light black',
886 )
ec11a9f4 887
7578d77d 888 def _format_text(self, handle, allow_colors, text, f, fallback=None, *, test_encoding=False):
e5a998f3 889 text = str(text)
ec11a9f4 890 if test_encoding:
891 original_text = text
5c104538 892 # handle.encoding can be None. See https://github.com/yt-dlp/yt-dlp/issues/2711
893 encoding = self.params.get('encoding') or getattr(handle, 'encoding', None) or 'ascii'
ec11a9f4 894 text = text.encode(encoding, 'ignore').decode(encoding)
895 if fallback is not None and text != original_text:
896 text = fallback
7578d77d 897 return format_text(text, f) if allow_colors else text if fallback is None else fallback
ec11a9f4 898
899 def _format_screen(self, *args, **kwargs):
7578d77d 900 return self._format_text(
cf4f42cb 901 self._out_files['screen'], self._allow_colors['screen'], *args, **kwargs)
ec11a9f4 902
903 def _format_err(self, *args, **kwargs):
7578d77d 904 return self._format_text(
cf4f42cb 905 self._out_files['error'], self._allow_colors['error'], *args, **kwargs)
819e0531 906
c84aeac6 907 def report_warning(self, message, only_once=False):
8222d8de
JMF
908 '''
909 Print the message to stderr, it will be prefixed with 'WARNING:'
910 If stderr is a tty file the 'WARNING:' will be colored
911 '''
6d07ce01
JMF
912 if self.params.get('logger') is not None:
913 self.params['logger'].warning(message)
8222d8de 914 else:
ad8915b7
PH
915 if self.params.get('no_warnings'):
916 return
ec11a9f4 917 self.to_stderr(f'{self._format_err("WARNING:", self.Styles.WARNING)} {message}', only_once)
8222d8de 918
ee8dd27a 919 def deprecation_warning(self, message):
920 if self.params.get('logger') is not None:
a44ca5a4 921 self.params['logger'].warning(f'DeprecationWarning: {message}')
ee8dd27a 922 else:
923 self.to_stderr(f'{self._format_err("DeprecationWarning:", self.Styles.ERROR)} {message}', True)
924
fa9f30b8 925 def report_error(self, message, *args, **kwargs):
8222d8de
JMF
926 '''
927 Do the same as trouble, but prefixes the message with 'ERROR:', colored
928 in red if stderr is a tty file.
929 '''
fa9f30b8 930 self.trouble(f'{self._format_err("ERROR:", self.Styles.ERROR)} {message}', *args, **kwargs)
8222d8de 931
b35496d8 932 def write_debug(self, message, only_once=False):
0760b0a7 933 '''Log debug message or Print message to stderr'''
934 if not self.params.get('verbose', False):
935 return
936 message = '[debug] %s' % message
937 if self.params.get('logger'):
938 self.params['logger'].debug(message)
939 else:
b35496d8 940 self.to_stderr(message, only_once)
0760b0a7 941
8222d8de
JMF
942 def report_file_already_downloaded(self, file_name):
943 """Report file has already been fully downloaded."""
944 try:
6febd1c1 945 self.to_screen('[download] %s has already been downloaded' % file_name)
ce02ed60 946 except UnicodeEncodeError:
6febd1c1 947 self.to_screen('[download] The file has already been downloaded')
8222d8de 948
0c3d0f51 949 def report_file_delete(self, file_name):
950 """Report that existing file will be deleted."""
951 try:
c25228e5 952 self.to_screen('Deleting existing file %s' % file_name)
0c3d0f51 953 except UnicodeEncodeError:
c25228e5 954 self.to_screen('Deleting existing file')
0c3d0f51 955
319b6059 956 def raise_no_formats(self, info, forced=False, *, msg=None):
1151c407 957 has_drm = info.get('__has_drm')
319b6059 958 ignored, expected = self.params.get('ignore_no_formats_error'), bool(msg)
959 msg = msg or has_drm and 'This video is DRM protected' or 'No video formats found!'
960 if forced or not ignored:
1151c407 961 raise ExtractorError(msg, video_id=info['id'], ie=info['extractor'],
319b6059 962 expected=has_drm or ignored or expected)
88acdbc2 963 else:
964 self.report_warning(msg)
965
de6000d9 966 def parse_outtmpl(self):
967 outtmpl_dict = self.params.get('outtmpl', {})
968 if not isinstance(outtmpl_dict, dict):
969 outtmpl_dict = {'default': outtmpl_dict}
71ce444a 970 # Remove spaces in the default template
971 if self.params.get('restrictfilenames'):
972 sanitize = lambda x: x.replace(' - ', ' ').replace(' ', '-')
973 else:
974 sanitize = lambda x: x
de6000d9 975 outtmpl_dict.update({
71ce444a 976 k: sanitize(v) for k, v in DEFAULT_OUTTMPL.items()
80c03fa9 977 if outtmpl_dict.get(k) is None})
86e5f3ed 978 for _, val in outtmpl_dict.items():
de6000d9 979 if isinstance(val, bytes):
86e5f3ed 980 self.report_warning('Parameter outtmpl is bytes, but should be a unicode string')
de6000d9 981 return outtmpl_dict
982
21cd8fae 983 def get_output_path(self, dir_type='', filename=None):
984 paths = self.params.get('paths', {})
985 assert isinstance(paths, dict)
986 path = os.path.join(
987 expand_path(paths.get('home', '').strip()),
988 expand_path(paths.get(dir_type, '').strip()) if dir_type else '',
989 filename or '')
21cd8fae 990 return sanitize_path(path, force=self.params.get('windowsfilenames'))
991
76a264ac 992 @staticmethod
901130bb 993 def _outtmpl_expandpath(outtmpl):
994 # expand_path translates '%%' into '%' and '$$' into '$'
995 # correspondingly that is not what we want since we need to keep
996 # '%%' intact for template dict substitution step. Working around
997 # with boundary-alike separator hack.
998 sep = ''.join([random.choice(ascii_letters) for _ in range(32)])
86e5f3ed 999 outtmpl = outtmpl.replace('%%', f'%{sep}%').replace('$$', f'${sep}$')
901130bb 1000
1001 # outtmpl should be expand_path'ed before template dict substitution
1002 # because meta fields may contain env variables we don't want to
1003 # be expanded. For example, for outtmpl "%(title)s.%(ext)s" and
1004 # title "Hello $PATH", we don't want `$PATH` to be expanded.
1005 return expand_path(outtmpl).replace(sep, '')
1006
1007 @staticmethod
1008 def escape_outtmpl(outtmpl):
1009 ''' Escape any remaining strings like %s, %abc% etc. '''
1010 return re.sub(
1011 STR_FORMAT_RE_TMPL.format('', '(?![%(\0])'),
1012 lambda mobj: ('' if mobj.group('has_key') else '%') + mobj.group(0),
1013 outtmpl)
1014
1015 @classmethod
1016 def validate_outtmpl(cls, outtmpl):
76a264ac 1017 ''' @return None or Exception object '''
7d1eb38a 1018 outtmpl = re.sub(
37893bb0 1019 STR_FORMAT_RE_TMPL.format('[^)]*', '[ljqBUDS]'),
7d1eb38a 1020 lambda mobj: f'{mobj.group(0)[:-1]}s',
1021 cls._outtmpl_expandpath(outtmpl))
76a264ac 1022 try:
7d1eb38a 1023 cls.escape_outtmpl(outtmpl) % collections.defaultdict(int)
76a264ac 1024 return None
1025 except ValueError as err:
1026 return err
1027
03b4de72 1028 @staticmethod
1029 def _copy_infodict(info_dict):
1030 info_dict = dict(info_dict)
09b49e1f 1031 info_dict.pop('__postprocessors', None)
03b4de72 1032 return info_dict
1033
e0fd9573 1034 def prepare_outtmpl(self, outtmpl, info_dict, sanitize=False):
1035 """ Make the outtmpl and info_dict suitable for substitution: ydl.escape_outtmpl(outtmpl) % info_dict
1036 @param sanitize Whether to sanitize the output as a filename.
1037 For backward compatibility, a function can also be passed
1038 """
1039
6e84b215 1040 info_dict.setdefault('epoch', int(time.time())) # keep epoch consistent once set
143db31d 1041
03b4de72 1042 info_dict = self._copy_infodict(info_dict)
752cda38 1043 info_dict['duration_string'] = ( # %(duration>%H-%M-%S)s is wrong if duration > 24hrs
53c18592 1044 formatSeconds(info_dict['duration'], '-' if sanitize else ':')
143db31d 1045 if info_dict.get('duration', None) is not None
1046 else None)
752cda38 1047 info_dict['autonumber'] = self.params.get('autonumber_start', 1) - 1 + self._num_downloads
9c906919 1048 info_dict['video_autonumber'] = self._num_videos
752cda38 1049 if info_dict.get('resolution') is None:
1050 info_dict['resolution'] = self.format_resolution(info_dict, default=None)
143db31d 1051
e6f21b3d 1052 # For fields playlist_index, playlist_autonumber and autonumber convert all occurrences
143db31d 1053 # of %(field)s to %(field)0Nd for backward compatibility
1054 field_size_compat_map = {
ec11a9f4 1055 'playlist_index': number_of_digits(info_dict.get('_last_playlist_index') or 0),
1056 'playlist_autonumber': number_of_digits(info_dict.get('n_entries') or 0),
752cda38 1057 'autonumber': self.params.get('autonumber_size') or 5,
143db31d 1058 }
752cda38 1059
385a27fa 1060 TMPL_DICT = {}
37893bb0 1061 EXTERNAL_FORMAT_RE = re.compile(STR_FORMAT_RE_TMPL.format('[^)]*', f'[{STR_FORMAT_TYPES}ljqBUDS]'))
385a27fa 1062 MATH_FUNCTIONS = {
1063 '+': float.__add__,
1064 '-': float.__sub__,
1065 }
e625be0d 1066 # Field is of the form key1.key2...
1067 # where keys (except first) can be string, int or slice
2b8a2973 1068 FIELD_RE = r'\w*(?:\.(?:\w+|{num}|{num}?(?::{num}?){{1,2}}))*'.format(num=r'(?:-?\d+)')
e0fd9573 1069 MATH_FIELD_RE = r'''(?:{field}|{num})'''.format(field=FIELD_RE, num=r'-?\d+(?:.\d+)?')
385a27fa 1070 MATH_OPERATORS_RE = r'(?:%s)' % '|'.join(map(re.escape, MATH_FUNCTIONS.keys()))
e625be0d 1071 INTERNAL_FORMAT_RE = re.compile(r'''(?x)
1072 (?P<negate>-)?
385a27fa 1073 (?P<fields>{field})
1074 (?P<maths>(?:{math_op}{math_field})*)
e625be0d 1075 (?:>(?P<strf_format>.+?))?
34baa9fd 1076 (?P<remaining>
1077 (?P<alternate>(?<!\\),[^|&)]+)?
1078 (?:&(?P<replacement>.*?))?
1079 (?:\|(?P<default>.*?))?
1080 )$'''.format(field=FIELD_RE, math_op=MATH_OPERATORS_RE, math_field=MATH_FIELD_RE))
752cda38 1081
2b8a2973 1082 def _traverse_infodict(k):
1083 k = k.split('.')
1084 if k[0] == '':
1085 k.pop(0)
1086 return traverse_obj(info_dict, k, is_user_input=True, traverse_string=True)
76a264ac 1087
752cda38 1088 def get_value(mdict):
1089 # Object traversal
2b8a2973 1090 value = _traverse_infodict(mdict['fields'])
752cda38 1091 # Negative
1092 if mdict['negate']:
1093 value = float_or_none(value)
1094 if value is not None:
1095 value *= -1
1096 # Do maths
385a27fa 1097 offset_key = mdict['maths']
1098 if offset_key:
752cda38 1099 value = float_or_none(value)
1100 operator = None
385a27fa 1101 while offset_key:
1102 item = re.match(
1103 MATH_FIELD_RE if operator else MATH_OPERATORS_RE,
1104 offset_key).group(0)
1105 offset_key = offset_key[len(item):]
1106 if operator is None:
752cda38 1107 operator = MATH_FUNCTIONS[item]
385a27fa 1108 continue
1109 item, multiplier = (item[1:], -1) if item[0] == '-' else (item, 1)
1110 offset = float_or_none(item)
1111 if offset is None:
2b8a2973 1112 offset = float_or_none(_traverse_infodict(item))
385a27fa 1113 try:
1114 value = operator(value, multiplier * offset)
1115 except (TypeError, ZeroDivisionError):
1116 return None
1117 operator = None
752cda38 1118 # Datetime formatting
1119 if mdict['strf_format']:
7c37ff97 1120 value = strftime_or_none(value, mdict['strf_format'].replace('\\,', ','))
752cda38 1121
1122 return value
1123
b868936c 1124 na = self.params.get('outtmpl_na_placeholder', 'NA')
1125
e0fd9573 1126 def filename_sanitizer(key, value, restricted=self.params.get('restrictfilenames')):
5c3895ff 1127 return sanitize_filename(str(value), restricted=restricted, is_id=(
1128 bool(re.search(r'(^|[_.])id(\.|$)', key))
1129 if 'filename-sanitization' in self.params.get('compat_opts', [])
1130 else NO_DEFAULT))
e0fd9573 1131
1132 sanitizer = sanitize if callable(sanitize) else filename_sanitizer
1133 sanitize = bool(sanitize)
1134
6e84b215 1135 def _dumpjson_default(obj):
1136 if isinstance(obj, (set, LazyList)):
1137 return list(obj)
adbc4ec4 1138 return repr(obj)
6e84b215 1139
752cda38 1140 def create_key(outer_mobj):
1141 if not outer_mobj.group('has_key'):
b836dc94 1142 return outer_mobj.group(0)
752cda38 1143 key = outer_mobj.group('key')
752cda38 1144 mobj = re.match(INTERNAL_FORMAT_RE, key)
e0fd9573 1145 initial_field = mobj.group('fields') if mobj else ''
e978789f 1146 value, replacement, default = None, None, na
7c37ff97 1147 while mobj:
e625be0d 1148 mobj = mobj.groupdict()
7c37ff97 1149 default = mobj['default'] if mobj['default'] is not None else default
752cda38 1150 value = get_value(mobj)
e978789f 1151 replacement = mobj['replacement']
7c37ff97 1152 if value is None and mobj['alternate']:
34baa9fd 1153 mobj = re.match(INTERNAL_FORMAT_RE, mobj['remaining'][1:])
7c37ff97 1154 else:
1155 break
752cda38 1156
b868936c 1157 fmt = outer_mobj.group('format')
752cda38 1158 if fmt == 's' and value is not None and key in field_size_compat_map.keys():
86e5f3ed 1159 fmt = f'0{field_size_compat_map[key]:d}d'
752cda38 1160
e978789f 1161 value = default if value is None else value if replacement is None else replacement
752cda38 1162
4476d2c7 1163 flags = outer_mobj.group('conversion') or ''
7d1eb38a 1164 str_fmt = f'{fmt[:-1]}s'
524e2e4f 1165 if fmt[-1] == 'l': # list
4476d2c7 1166 delim = '\n' if '#' in flags else ', '
9e907ebd 1167 value, fmt = delim.join(map(str, variadic(value, allowed_types=(str, bytes)))), str_fmt
524e2e4f 1168 elif fmt[-1] == 'j': # json
4476d2c7 1169 value, fmt = json.dumps(value, default=_dumpjson_default, indent=4 if '#' in flags else None), str_fmt
524e2e4f 1170 elif fmt[-1] == 'q': # quoted
4476d2c7 1171 value = map(str, variadic(value) if '#' in flags else [value])
1172 value, fmt = ' '.join(map(compat_shlex_quote, value)), str_fmt
524e2e4f 1173 elif fmt[-1] == 'B': # bytes
86e5f3ed 1174 value = f'%{str_fmt}'.encode() % str(value).encode('utf-8')
f5aa5cfb 1175 value, fmt = value.decode('utf-8', 'ignore'), 's'
524e2e4f 1176 elif fmt[-1] == 'U': # unicode normalized
524e2e4f 1177 value, fmt = unicodedata.normalize(
1178 # "+" = compatibility equivalence, "#" = NFD
4476d2c7 1179 'NF%s%s' % ('K' if '+' in flags else '', 'D' if '#' in flags else 'C'),
524e2e4f 1180 value), str_fmt
e0fd9573 1181 elif fmt[-1] == 'D': # decimal suffix
abbeeebc 1182 num_fmt, fmt = fmt[:-1].replace('#', ''), 's'
1183 value = format_decimal_suffix(value, f'%{num_fmt}f%s' if num_fmt else '%d%s',
1184 factor=1024 if '#' in flags else 1000)
37893bb0 1185 elif fmt[-1] == 'S': # filename sanitization
e0fd9573 1186 value, fmt = filename_sanitizer(initial_field, value, restricted='#' in flags), str_fmt
7d1eb38a 1187 elif fmt[-1] == 'c':
524e2e4f 1188 if value:
1189 value = str(value)[0]
76a264ac 1190 else:
524e2e4f 1191 fmt = str_fmt
76a264ac 1192 elif fmt[-1] not in 'rs': # numeric
a439a3a4 1193 value = float_or_none(value)
752cda38 1194 if value is None:
1195 value, fmt = default, 's'
901130bb 1196
752cda38 1197 if sanitize:
1198 if fmt[-1] == 'r':
1199 # If value is an object, sanitize might convert it to a string
1200 # So we convert it to repr first
7d1eb38a 1201 value, fmt = repr(value), str_fmt
639f1cea 1202 if fmt[-1] in 'csr':
e0fd9573 1203 value = sanitizer(initial_field, value)
901130bb 1204
b868936c 1205 key = '%s\0%s' % (key.replace('%', '%\0'), outer_mobj.group('format'))
385a27fa 1206 TMPL_DICT[key] = value
b868936c 1207 return '{prefix}%({key}){fmt}'.format(key=key, fmt=fmt, prefix=outer_mobj.group('prefix'))
752cda38 1208
385a27fa 1209 return EXTERNAL_FORMAT_RE.sub(create_key, outtmpl), TMPL_DICT
143db31d 1210
819e0531 1211 def evaluate_outtmpl(self, outtmpl, info_dict, *args, **kwargs):
1212 outtmpl, info_dict = self.prepare_outtmpl(outtmpl, info_dict, *args, **kwargs)
1213 return self.escape_outtmpl(outtmpl) % info_dict
1214
5127e92a 1215 def _prepare_filename(self, info_dict, *, outtmpl=None, tmpl_type=None):
1216 assert None in (outtmpl, tmpl_type), 'outtmpl and tmpl_type are mutually exclusive'
1217 if outtmpl is None:
1218 outtmpl = self.outtmpl_dict.get(tmpl_type or 'default', self.outtmpl_dict['default'])
8222d8de 1219 try:
5127e92a 1220 outtmpl = self._outtmpl_expandpath(outtmpl)
e0fd9573 1221 filename = self.evaluate_outtmpl(outtmpl, info_dict, True)
6a0546e3 1222 if not filename:
1223 return None
15da37c7 1224
5127e92a 1225 if tmpl_type in ('', 'temp'):
6a0546e3 1226 final_ext, ext = self.params.get('final_ext'), info_dict.get('ext')
1227 if final_ext and ext and final_ext != ext and filename.endswith(f'.{final_ext}'):
1228 filename = replace_extension(filename, ext, final_ext)
5127e92a 1229 elif tmpl_type:
6a0546e3 1230 force_ext = OUTTMPL_TYPES[tmpl_type]
1231 if force_ext:
1232 filename = replace_extension(filename, force_ext, info_dict.get('ext'))
de6000d9 1233
bdc3fd2f
U
1234 # https://github.com/blackjack4494/youtube-dlc/issues/85
1235 trim_file_name = self.params.get('trim_file_name', False)
1236 if trim_file_name:
5c22c63d 1237 no_ext, *ext = filename.rsplit('.', 2)
1238 filename = join_nonempty(no_ext[:trim_file_name], *ext, delim='.')
bdc3fd2f 1239
0202b52a 1240 return filename
8222d8de 1241 except ValueError as err:
6febd1c1 1242 self.report_error('Error in output template: ' + str(err) + ' (encoding: ' + repr(preferredencoding()) + ')')
8222d8de
JMF
1243 return None
1244
5127e92a 1245 def prepare_filename(self, info_dict, dir_type='', *, outtmpl=None, warn=False):
1246 """Generate the output filename"""
1247 if outtmpl:
1248 assert not dir_type, 'outtmpl and dir_type are mutually exclusive'
1249 dir_type = None
1250 filename = self._prepare_filename(info_dict, tmpl_type=dir_type, outtmpl=outtmpl)
80c03fa9 1251 if not filename and dir_type not in ('', 'temp'):
1252 return ''
de6000d9 1253
c84aeac6 1254 if warn:
21cd8fae 1255 if not self.params.get('paths'):
de6000d9 1256 pass
1257 elif filename == '-':
c84aeac6 1258 self.report_warning('--paths is ignored when an outputting to stdout', only_once=True)
de6000d9 1259 elif os.path.isabs(filename):
c84aeac6 1260 self.report_warning('--paths is ignored since an absolute path is given in output template', only_once=True)
de6000d9 1261 if filename == '-' or not filename:
1262 return filename
1263
21cd8fae 1264 return self.get_output_path(dir_type, filename)
0202b52a 1265
120fe513 1266 def _match_entry(self, info_dict, incomplete=False, silent=False):
ecdec191 1267 """ Returns None if the file should be downloaded """
8222d8de 1268
c77495e3 1269 video_title = info_dict.get('title', info_dict.get('id', 'video'))
1270
8b0d7497 1271 def check_filter():
8b0d7497 1272 if 'title' in info_dict:
1273 # This can happen when we're just evaluating the playlist
1274 title = info_dict['title']
1275 matchtitle = self.params.get('matchtitle', False)
1276 if matchtitle:
1277 if not re.search(matchtitle, title, re.IGNORECASE):
1278 return '"' + title + '" title did not match pattern "' + matchtitle + '"'
1279 rejecttitle = self.params.get('rejecttitle', False)
1280 if rejecttitle:
1281 if re.search(rejecttitle, title, re.IGNORECASE):
1282 return '"' + title + '" title matched reject pattern "' + rejecttitle + '"'
1283 date = info_dict.get('upload_date')
1284 if date is not None:
1285 dateRange = self.params.get('daterange', DateRange())
1286 if date not in dateRange:
86e5f3ed 1287 return f'{date_from_str(date).isoformat()} upload date is not in range {dateRange}'
8b0d7497 1288 view_count = info_dict.get('view_count')
1289 if view_count is not None:
1290 min_views = self.params.get('min_views')
1291 if min_views is not None and view_count < min_views:
1292 return 'Skipping %s, because it has not reached minimum view count (%d/%d)' % (video_title, view_count, min_views)
1293 max_views = self.params.get('max_views')
1294 if max_views is not None and view_count > max_views:
1295 return 'Skipping %s, because it has exceeded the maximum view count (%d/%d)' % (video_title, view_count, max_views)
1296 if age_restricted(info_dict.get('age_limit'), self.params.get('age_limit')):
1297 return 'Skipping "%s" because it is age restricted' % video_title
8b0d7497 1298
8f18aca8 1299 match_filter = self.params.get('match_filter')
1300 if match_filter is not None:
1301 try:
1302 ret = match_filter(info_dict, incomplete=incomplete)
1303 except TypeError:
1304 # For backward compatibility
1305 ret = None if incomplete else match_filter(info_dict)
1306 if ret is not None:
1307 return ret
8b0d7497 1308 return None
1309
c77495e3 1310 if self.in_download_archive(info_dict):
1311 reason = '%s has already been recorded in the archive' % video_title
1312 break_opt, break_err = 'break_on_existing', ExistingVideoReached
1313 else:
1314 reason = check_filter()
1315 break_opt, break_err = 'break_on_reject', RejectedVideoReached
8b0d7497 1316 if reason is not None:
120fe513 1317 if not silent:
1318 self.to_screen('[download] ' + reason)
c77495e3 1319 if self.params.get(break_opt, False):
1320 raise break_err()
8b0d7497 1321 return reason
fe7e0c98 1322
b6c45014
JMF
1323 @staticmethod
1324 def add_extra_info(info_dict, extra_info):
1325 '''Set the keys from extra_info in info dict if they are missing'''
1326 for key, value in extra_info.items():
1327 info_dict.setdefault(key, value)
1328
409e1828 1329 def extract_info(self, url, download=True, ie_key=None, extra_info=None,
61aa5ba3 1330 process=True, force_generic_extractor=False):
41d1cca3 1331 """
1332 Return a list with a dictionary for each video extracted.
1333
1334 Arguments:
1335 url -- URL to extract
1336
1337 Keyword arguments:
1338 download -- whether to download videos during extraction
1339 ie_key -- extractor key hint
1340 extra_info -- dictionary containing the extra values to add to each result
1341 process -- whether to resolve all unresolved references (URLs, playlist items),
1342 must be True for download to work.
1343 force_generic_extractor -- force using the generic extractor
1344 """
fe7e0c98 1345
409e1828 1346 if extra_info is None:
1347 extra_info = {}
1348
61aa5ba3 1349 if not ie_key and force_generic_extractor:
d22dec74
S
1350 ie_key = 'Generic'
1351
8222d8de 1352 if ie_key:
8b7491c8 1353 ies = {ie_key: self._get_info_extractor_class(ie_key)}
8222d8de
JMF
1354 else:
1355 ies = self._ies
1356
8b7491c8 1357 for ie_key, ie in ies.items():
8222d8de
JMF
1358 if not ie.suitable(url):
1359 continue
1360
1361 if not ie.working():
6febd1c1
PH
1362 self.report_warning('The program functionality for this site has been marked as broken, '
1363 'and will probably not work.')
8222d8de 1364
1151c407 1365 temp_id = ie.get_temp_id(url)
a0566bbf 1366 if temp_id is not None and self.in_download_archive({'id': temp_id, 'ie_key': ie_key}):
5e5be0c0 1367 self.to_screen(f'[{ie_key}] {temp_id}: has already been recorded in the archive')
1368 if self.params.get('break_on_existing', False):
1369 raise ExistingVideoReached()
a0566bbf 1370 break
8b7491c8 1371 return self.__extract_info(url, self.get_info_extractor(ie_key), download, extra_info, process)
a0566bbf 1372 else:
1373 self.report_error('no suitable InfoExtractor for URL %s' % url)
1374
8e5fecc8 1375 def __handle_extraction_exceptions(func):
b5ae35ee 1376 @functools.wraps(func)
a0566bbf 1377 def wrapper(self, *args, **kwargs):
6da22e7d 1378 while True:
1379 try:
1380 return func(self, *args, **kwargs)
1381 except (DownloadCancelled, LazyList.IndexError, PagedList.IndexError):
8222d8de 1382 raise
6da22e7d 1383 except ReExtractInfo as e:
1384 if e.expected:
1385 self.to_screen(f'{e}; Re-extracting data')
1386 else:
1387 self.to_stderr('\r')
1388 self.report_warning(f'{e}; Re-extracting data')
1389 continue
1390 except GeoRestrictedError as e:
1391 msg = e.msg
1392 if e.countries:
1393 msg += '\nThis video is available in %s.' % ', '.join(
1394 map(ISO3166Utils.short2full, e.countries))
1395 msg += '\nYou might want to use a VPN or a proxy server (with --proxy) to workaround.'
1396 self.report_error(msg)
1397 except ExtractorError as e: # An error we somewhat expected
1398 self.report_error(str(e), e.format_traceback())
1399 except Exception as e:
1400 if self.params.get('ignoreerrors'):
1401 self.report_error(str(e), tb=encode_compat_str(traceback.format_exc()))
1402 else:
1403 raise
1404 break
a0566bbf 1405 return wrapper
1406
f2ebc5c7 1407 def _wait_for_video(self, ie_result):
1408 if (not self.params.get('wait_for_video')
1409 or ie_result.get('_type', 'video') != 'video'
1410 or ie_result.get('formats') or ie_result.get('url')):
1411 return
1412
1413 format_dur = lambda dur: '%02d:%02d:%02d' % timetuple_from_msec(dur * 1000)[:-1]
1414 last_msg = ''
1415
1416 def progress(msg):
1417 nonlocal last_msg
1418 self.to_screen(msg + ' ' * (len(last_msg) - len(msg)) + '\r', skip_eol=True)
1419 last_msg = msg
1420
1421 min_wait, max_wait = self.params.get('wait_for_video')
1422 diff = try_get(ie_result, lambda x: x['release_timestamp'] - time.time())
1423 if diff is None and ie_result.get('live_status') == 'is_upcoming':
16c620bc 1424 diff = round(random.uniform(min_wait, max_wait) if (max_wait and min_wait) else (max_wait or min_wait), 0)
f2ebc5c7 1425 self.report_warning('Release time of video is not known')
1426 elif (diff or 0) <= 0:
1427 self.report_warning('Video should already be available according to extracted info')
38d79fd1 1428 diff = min(max(diff or 0, min_wait or 0), max_wait or float('inf'))
f2ebc5c7 1429 self.to_screen(f'[wait] Waiting for {format_dur(diff)} - Press Ctrl+C to try now')
1430
1431 wait_till = time.time() + diff
1432 try:
1433 while True:
1434 diff = wait_till - time.time()
1435 if diff <= 0:
1436 progress('')
1437 raise ReExtractInfo('[wait] Wait period ended', expected=True)
1438 progress(f'[wait] Remaining time until next attempt: {self._format_screen(format_dur(diff), self.Styles.EMPHASIS)}')
1439 time.sleep(1)
1440 except KeyboardInterrupt:
1441 progress('')
1442 raise ReExtractInfo('[wait] Interrupted by user', expected=True)
1443 except BaseException as e:
1444 if not isinstance(e, ReExtractInfo):
1445 self.to_screen('')
1446 raise
1447
a0566bbf 1448 @__handle_extraction_exceptions
58f197b7 1449 def __extract_info(self, url, ie, download, extra_info, process):
a0566bbf 1450 ie_result = ie.extract(url)
1451 if ie_result is None: # Finished already (backwards compatibility; listformats and friends should be moved here)
1452 return
1453 if isinstance(ie_result, list):
1454 # Backwards compatibility: old IE result format
1455 ie_result = {
1456 '_type': 'compat_list',
1457 'entries': ie_result,
1458 }
e37d0efb 1459 if extra_info.get('original_url'):
1460 ie_result.setdefault('original_url', extra_info['original_url'])
a0566bbf 1461 self.add_default_extra_info(ie_result, ie, url)
1462 if process:
f2ebc5c7 1463 self._wait_for_video(ie_result)
a0566bbf 1464 return self.process_ie_result(ie_result, download, extra_info)
8222d8de 1465 else:
a0566bbf 1466 return ie_result
fe7e0c98 1467
ea38e55f 1468 def add_default_extra_info(self, ie_result, ie, url):
6033d980 1469 if url is not None:
1470 self.add_extra_info(ie_result, {
1471 'webpage_url': url,
1472 'original_url': url,
57ebfca3 1473 })
1474 webpage_url = ie_result.get('webpage_url')
1475 if webpage_url:
1476 self.add_extra_info(ie_result, {
1477 'webpage_url_basename': url_basename(webpage_url),
1478 'webpage_url_domain': get_domain(webpage_url),
6033d980 1479 })
1480 if ie is not None:
1481 self.add_extra_info(ie_result, {
1482 'extractor': ie.IE_NAME,
1483 'extractor_key': ie.ie_key(),
1484 })
ea38e55f 1485
58adec46 1486 def process_ie_result(self, ie_result, download=True, extra_info=None):
8222d8de
JMF
1487 """
1488 Take the result of the ie(may be modified) and resolve all unresolved
1489 references (URLs, playlist items).
1490
1491 It will also download the videos if 'download'.
1492 Returns the resolved ie_result.
1493 """
58adec46 1494 if extra_info is None:
1495 extra_info = {}
e8ee972c
PH
1496 result_type = ie_result.get('_type', 'video')
1497
057a5206 1498 if result_type in ('url', 'url_transparent'):
134c6ea8 1499 ie_result['url'] = sanitize_url(ie_result['url'])
e37d0efb 1500 if ie_result.get('original_url'):
1501 extra_info.setdefault('original_url', ie_result['original_url'])
1502
057a5206 1503 extract_flat = self.params.get('extract_flat', False)
3089bc74
S
1504 if ((extract_flat == 'in_playlist' and 'playlist' in extra_info)
1505 or extract_flat is True):
ecb54191 1506 info_copy = ie_result.copy()
6033d980 1507 ie = try_get(ie_result.get('ie_key'), self.get_info_extractor)
360167b9 1508 if ie and not ie_result.get('id'):
4614bc22 1509 info_copy['id'] = ie.get_temp_id(ie_result['url'])
6033d980 1510 self.add_default_extra_info(info_copy, ie, ie_result['url'])
4614bc22 1511 self.add_extra_info(info_copy, extra_info)
b5475f11 1512 info_copy, _ = self.pre_process(info_copy)
ecb54191 1513 self.__forced_printings(info_copy, self.prepare_filename(info_copy), incomplete=True)
4614bc22 1514 if self.params.get('force_write_download_archive', False):
1515 self.record_download_archive(info_copy)
e8ee972c
PH
1516 return ie_result
1517
8222d8de 1518 if result_type == 'video':
b6c45014 1519 self.add_extra_info(ie_result, extra_info)
9c2b75b5 1520 ie_result = self.process_video_result(ie_result, download=download)
28b0eb0f 1521 additional_urls = (ie_result or {}).get('additional_urls')
9c2b75b5 1522 if additional_urls:
e9f4ccd1 1523 # TODO: Improve MetadataParserPP to allow setting a list
9c2b75b5 1524 if isinstance(additional_urls, compat_str):
1525 additional_urls = [additional_urls]
1526 self.to_screen(
1527 '[info] %s: %d additional URL(s) requested' % (ie_result['id'], len(additional_urls)))
1528 self.write_debug('Additional URLs: "%s"' % '", "'.join(additional_urls))
1529 ie_result['additional_entries'] = [
1530 self.extract_info(
b69fd25c 1531 url, download, extra_info=extra_info,
9c2b75b5 1532 force_generic_extractor=self.params.get('force_generic_extractor'))
1533 for url in additional_urls
1534 ]
1535 return ie_result
8222d8de
JMF
1536 elif result_type == 'url':
1537 # We have to add extra_info to the results because it may be
1538 # contained in a playlist
07cce701 1539 return self.extract_info(
1540 ie_result['url'], download,
1541 ie_key=ie_result.get('ie_key'),
1542 extra_info=extra_info)
7fc3fa05
PH
1543 elif result_type == 'url_transparent':
1544 # Use the information from the embedding page
1545 info = self.extract_info(
1546 ie_result['url'], ie_key=ie_result.get('ie_key'),
1547 extra_info=extra_info, download=False, process=False)
1548
1640eb09
S
1549 # extract_info may return None when ignoreerrors is enabled and
1550 # extraction failed with an error, don't crash and return early
1551 # in this case
1552 if not info:
1553 return info
1554
412c617d 1555 new_result = info.copy()
90137ca4 1556 new_result.update(filter_dict(ie_result, lambda k, v: (
1557 v is not None and k not in {'_type', 'url', 'id', 'extractor', 'extractor_key', 'ie_key'})))
7fc3fa05 1558
0563f7ac
S
1559 # Extracted info may not be a video result (i.e.
1560 # info.get('_type', 'video') != video) but rather an url or
1561 # url_transparent. In such cases outer metadata (from ie_result)
1562 # should be propagated to inner one (info). For this to happen
1563 # _type of info should be overridden with url_transparent. This
067aa17e 1564 # fixes issue from https://github.com/ytdl-org/youtube-dl/pull/11163.
0563f7ac
S
1565 if new_result.get('_type') == 'url':
1566 new_result['_type'] = 'url_transparent'
7fc3fa05
PH
1567
1568 return self.process_ie_result(
1569 new_result, download=download, extra_info=extra_info)
40fcba5e 1570 elif result_type in ('playlist', 'multi_video'):
30a074c2 1571 # Protect from infinite recursion due to recursively nested playlists
1572 # (see https://github.com/ytdl-org/youtube-dl/issues/27833)
1573 webpage_url = ie_result['webpage_url']
1574 if webpage_url in self._playlist_urls:
7e85e872 1575 self.to_screen(
30a074c2 1576 '[download] Skipping already downloaded playlist: %s'
1577 % ie_result.get('title') or ie_result.get('id'))
1578 return
7e85e872 1579
30a074c2 1580 self._playlist_level += 1
1581 self._playlist_urls.add(webpage_url)
03f83004 1582 self._fill_common_fields(ie_result, False)
bc516a3f 1583 self._sanitize_thumbnails(ie_result)
30a074c2 1584 try:
1585 return self.__process_playlist(ie_result, download)
1586 finally:
1587 self._playlist_level -= 1
1588 if not self._playlist_level:
1589 self._playlist_urls.clear()
8222d8de 1590 elif result_type == 'compat_list':
c9bf4114
PH
1591 self.report_warning(
1592 'Extractor %s returned a compat_list result. '
1593 'It needs to be updated.' % ie_result.get('extractor'))
5f6a1245 1594
8222d8de 1595 def _fixup(r):
b868936c 1596 self.add_extra_info(r, {
1597 'extractor': ie_result['extractor'],
1598 'webpage_url': ie_result['webpage_url'],
1599 'webpage_url_basename': url_basename(ie_result['webpage_url']),
0bb322b9 1600 'webpage_url_domain': get_domain(ie_result['webpage_url']),
b868936c 1601 'extractor_key': ie_result['extractor_key'],
1602 })
8222d8de
JMF
1603 return r
1604 ie_result['entries'] = [
b6c45014 1605 self.process_ie_result(_fixup(r), download, extra_info)
8222d8de
JMF
1606 for r in ie_result['entries']
1607 ]
1608 return ie_result
1609 else:
1610 raise Exception('Invalid result type: %s' % result_type)
1611
e92caff5 1612 def _ensure_dir_exists(self, path):
1613 return make_dir(path, self.report_error)
1614
3b603dbd 1615 @staticmethod
1616 def _playlist_infodict(ie_result, **kwargs):
1617 return {
1618 **ie_result,
1619 'playlist': ie_result.get('title') or ie_result.get('id'),
1620 'playlist_id': ie_result.get('id'),
1621 'playlist_title': ie_result.get('title'),
1622 'playlist_uploader': ie_result.get('uploader'),
1623 'playlist_uploader_id': ie_result.get('uploader_id'),
1624 'playlist_index': 0,
1625 **kwargs,
1626 }
1627
30a074c2 1628 def __process_playlist(self, ie_result, download):
1629 # We process each entry in the playlist
1630 playlist = ie_result.get('title') or ie_result.get('id')
1631 self.to_screen('[download] Downloading playlist: %s' % playlist)
1632
498f5606 1633 if 'entries' not in ie_result:
aa9369a2 1634 raise EntryNotInPlaylist('There are no entries')
7c7f7161 1635
1636 MissingEntry = object()
498f5606 1637 incomplete_entries = bool(ie_result.get('requested_entries'))
1638 if incomplete_entries:
bf5f605e 1639 def fill_missing_entries(entries, indices):
7c7f7161 1640 ret = [MissingEntry] * max(indices)
bf5f605e 1641 for i, entry in zip(indices, entries):
498f5606 1642 ret[i - 1] = entry
1643 return ret
1644 ie_result['entries'] = fill_missing_entries(ie_result['entries'], ie_result['requested_entries'])
02fd60d3 1645
30a074c2 1646 playlist_results = []
1647
56a8fb4f 1648 playliststart = self.params.get('playliststart', 1)
30a074c2 1649 playlistend = self.params.get('playlistend')
1650 # For backwards compatibility, interpret -1 as whole list
1651 if playlistend == -1:
1652 playlistend = None
1653
1654 playlistitems_str = self.params.get('playlist_items')
1655 playlistitems = None
1656 if playlistitems_str is not None:
1657 def iter_playlistitems(format):
1658 for string_segment in format.split(','):
1659 if '-' in string_segment:
1660 start, end = string_segment.split('-')
1661 for item in range(int(start), int(end) + 1):
1662 yield int(item)
1663 else:
1664 yield int(string_segment)
1665 playlistitems = orderedSet(iter_playlistitems(playlistitems_str))
1666
1667 ie_entries = ie_result['entries']
8e5fecc8 1668 if isinstance(ie_entries, list):
ed8d87f9 1669 playlist_count = len(ie_entries)
f0d785d3 1670 msg = f'Collected {playlist_count} videos; downloading %d of them'
1671 ie_result['playlist_count'] = ie_result.get('playlist_count') or playlist_count
1672
8e5fecc8 1673 def get_entry(i):
1674 return ie_entries[i - 1]
1675 else:
f0d785d3 1676 msg = 'Downloading %d videos'
c586f9e8 1677 if not isinstance(ie_entries, (PagedList, LazyList)):
8e5fecc8 1678 ie_entries = LazyList(ie_entries)
d37707bd 1679 elif isinstance(ie_entries, InAdvancePagedList):
1680 if ie_entries._pagesize == 1:
1681 playlist_count = ie_entries._pagecount
8e5fecc8 1682
1683 def get_entry(i):
1684 return YoutubeDL.__handle_extraction_exceptions(
1685 lambda self, i: ie_entries[i - 1]
1686 )(self, i)
50fed816 1687
f0d785d3 1688 entries, broken = [], False
ff1c7fc9 1689 items = playlistitems if playlistitems is not None else itertools.count(playliststart)
1690 for i in items:
1691 if i == 0:
1692 continue
56a8fb4f 1693 if playlistitems is None and playlistend is not None and playlistend < i:
1694 break
1695 entry = None
1696 try:
50fed816 1697 entry = get_entry(i)
7c7f7161 1698 if entry is MissingEntry:
498f5606 1699 raise EntryNotInPlaylist()
56a8fb4f 1700 except (IndexError, EntryNotInPlaylist):
1701 if incomplete_entries:
aa9369a2 1702 raise EntryNotInPlaylist(f'Entry {i} cannot be found')
56a8fb4f 1703 elif not playlistitems:
1704 break
1705 entries.append(entry)
120fe513 1706 try:
1707 if entry is not None:
e5a998f3 1708 # TODO: Add auto-generated fields
120fe513 1709 self._match_entry(entry, incomplete=True, silent=True)
1710 except (ExistingVideoReached, RejectedVideoReached):
f0d785d3 1711 broken = True
120fe513 1712 break
56a8fb4f 1713 ie_result['entries'] = entries
30a074c2 1714
56a8fb4f 1715 # Save playlist_index before re-ordering
1716 entries = [
9e598870 1717 ((playlistitems[i - 1] if playlistitems else i + playliststart - 1), entry)
56a8fb4f 1718 for i, entry in enumerate(entries, 1)
1719 if entry is not None]
1720 n_entries = len(entries)
498f5606 1721
f0d785d3 1722 if not (ie_result.get('playlist_count') or broken or playlistitems or playlistend):
1723 ie_result['playlist_count'] = n_entries
1724
e08a85d8 1725 if not playlistitems and (playliststart != 1 or playlistend):
56a8fb4f 1726 playlistitems = list(range(playliststart, playliststart + n_entries))
498f5606 1727 ie_result['requested_entries'] = playlistitems
1728
e08a85d8 1729 _infojson_written = False
0bfc53d0 1730 write_playlist_files = self.params.get('allow_playlist_files', True)
1731 if write_playlist_files and self.params.get('list_thumbnails'):
1732 self.list_thumbnails(ie_result)
1733 if write_playlist_files and not self.params.get('simulate'):
3b603dbd 1734 ie_copy = self._playlist_infodict(ie_result, n_entries=n_entries)
e08a85d8 1735 _infojson_written = self._write_info_json(
1736 'playlist', ie_result, self.prepare_filename(ie_copy, 'pl_infojson'))
1737 if _infojson_written is None:
80c03fa9 1738 return
1739 if self._write_description('playlist', ie_result,
1740 self.prepare_filename(ie_copy, 'pl_description')) is None:
1741 return
681de68e 1742 # TODO: This should be passed to ThumbnailsConvertor if necessary
80c03fa9 1743 self._write_thumbnails('playlist', ie_copy, self.prepare_filename(ie_copy, 'pl_thumbnail'))
30a074c2 1744
1745 if self.params.get('playlistreverse', False):
1746 entries = entries[::-1]
30a074c2 1747 if self.params.get('playlistrandom', False):
1748 random.shuffle(entries)
1749
1750 x_forwarded_for = ie_result.get('__x_forwarded_for_ip')
1751
86e5f3ed 1752 self.to_screen(f'[{ie_result["extractor"]}] playlist {playlist}: {msg % n_entries}')
26e2805c 1753 failures = 0
1754 max_failures = self.params.get('skip_playlist_after_errors') or float('inf')
71729754 1755 for i, entry_tuple in enumerate(entries, 1):
1756 playlist_index, entry = entry_tuple
81139999 1757 if 'playlist-index' in self.params.get('compat_opts', []):
1758 playlist_index = playlistitems[i - 1] if playlistitems else i + playliststart - 1
19a03940 1759 self.to_screen('[download] Downloading video %s of %s' % (
1760 self._format_screen(i, self.Styles.ID), self._format_screen(n_entries, self.Styles.EMPHASIS)))
30a074c2 1761 # This __x_forwarded_for_ip thing is a bit ugly but requires
1762 # minimal changes
1763 if x_forwarded_for:
1764 entry['__x_forwarded_for_ip'] = x_forwarded_for
1765 extra = {
1766 'n_entries': n_entries,
f59ae581 1767 '_last_playlist_index': max(playlistitems) if playlistitems else (playlistend or n_entries),
f0d785d3 1768 'playlist_count': ie_result.get('playlist_count'),
71729754 1769 'playlist_index': playlist_index,
1770 'playlist_autonumber': i,
30a074c2 1771 'playlist': playlist,
1772 'playlist_id': ie_result.get('id'),
1773 'playlist_title': ie_result.get('title'),
1774 'playlist_uploader': ie_result.get('uploader'),
1775 'playlist_uploader_id': ie_result.get('uploader_id'),
30a074c2 1776 'extractor': ie_result['extractor'],
1777 'webpage_url': ie_result['webpage_url'],
1778 'webpage_url_basename': url_basename(ie_result['webpage_url']),
0bb322b9 1779 'webpage_url_domain': get_domain(ie_result['webpage_url']),
30a074c2 1780 'extractor_key': ie_result['extractor_key'],
1781 }
1782
1783 if self._match_entry(entry, incomplete=True) is not None:
1784 continue
1785
1786 entry_result = self.__process_iterable_entry(entry, download, extra)
26e2805c 1787 if not entry_result:
1788 failures += 1
1789 if failures >= max_failures:
1790 self.report_error(
1791 'Skipping the remaining entries in playlist "%s" since %d items failed extraction' % (playlist, failures))
1792 break
30a074c2 1793 playlist_results.append(entry_result)
1794 ie_result['entries'] = playlist_results
e08a85d8 1795
1796 # Write the updated info to json
cb96c5be 1797 if _infojson_written is True and self._write_info_json(
e08a85d8 1798 'updated playlist', ie_result,
1799 self.prepare_filename(ie_copy, 'pl_infojson'), overwrite=True) is None:
1800 return
ca30f449 1801
ed5835b4 1802 ie_result = self.run_all_pps('playlist', ie_result)
1803 self.to_screen(f'[download] Finished downloading playlist: {playlist}')
30a074c2 1804 return ie_result
1805
a0566bbf 1806 @__handle_extraction_exceptions
1807 def __process_iterable_entry(self, entry, download, extra_info):
1808 return self.process_ie_result(
1809 entry, download=download, extra_info=extra_info)
1810
67134eab
JMF
1811 def _build_format_filter(self, filter_spec):
1812 " Returns a function to filter the formats according to the filter_spec "
083c9df9
PH
1813
1814 OPERATORS = {
1815 '<': operator.lt,
1816 '<=': operator.le,
1817 '>': operator.gt,
1818 '>=': operator.ge,
1819 '=': operator.eq,
1820 '!=': operator.ne,
1821 }
67134eab 1822 operator_rex = re.compile(r'''(?x)\s*
187986a8 1823 (?P<key>width|height|tbr|abr|vbr|asr|filesize|filesize_approx|fps)\s*
1824 (?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
1825 (?P<value>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?)\s*
083c9df9 1826 ''' % '|'.join(map(re.escape, OPERATORS.keys())))
187986a8 1827 m = operator_rex.fullmatch(filter_spec)
9ddb6925
S
1828 if m:
1829 try:
1830 comparison_value = int(m.group('value'))
1831 except ValueError:
1832 comparison_value = parse_filesize(m.group('value'))
1833 if comparison_value is None:
1834 comparison_value = parse_filesize(m.group('value') + 'B')
1835 if comparison_value is None:
1836 raise ValueError(
1837 'Invalid value %r in format specification %r' % (
67134eab 1838 m.group('value'), filter_spec))
9ddb6925
S
1839 op = OPERATORS[m.group('op')]
1840
083c9df9 1841 if not m:
9ddb6925
S
1842 STR_OPERATORS = {
1843 '=': operator.eq,
10d33b34
YCH
1844 '^=': lambda attr, value: attr.startswith(value),
1845 '$=': lambda attr, value: attr.endswith(value),
1846 '*=': lambda attr, value: value in attr,
1ce9a3cb 1847 '~=': lambda attr, value: value.search(attr) is not None
9ddb6925 1848 }
187986a8 1849 str_operator_rex = re.compile(r'''(?x)\s*
1850 (?P<key>[a-zA-Z0-9._-]+)\s*
1ce9a3cb
LF
1851 (?P<negation>!\s*)?(?P<op>%s)\s*(?P<none_inclusive>\?\s*)?
1852 (?P<quote>["'])?
1853 (?P<value>(?(quote)(?:(?!(?P=quote))[^\\]|\\.)+|[\w.-]+))
1854 (?(quote)(?P=quote))\s*
9ddb6925 1855 ''' % '|'.join(map(re.escape, STR_OPERATORS.keys())))
187986a8 1856 m = str_operator_rex.fullmatch(filter_spec)
9ddb6925 1857 if m:
1ce9a3cb
LF
1858 if m.group('op') == '~=':
1859 comparison_value = re.compile(m.group('value'))
1860 else:
1861 comparison_value = re.sub(r'''\\([\\"'])''', r'\1', m.group('value'))
2cc779f4
S
1862 str_op = STR_OPERATORS[m.group('op')]
1863 if m.group('negation'):
e118a879 1864 op = lambda attr, value: not str_op(attr, value)
2cc779f4
S
1865 else:
1866 op = str_op
083c9df9 1867
9ddb6925 1868 if not m:
187986a8 1869 raise SyntaxError('Invalid filter specification %r' % filter_spec)
083c9df9
PH
1870
1871 def _filter(f):
1872 actual_value = f.get(m.group('key'))
1873 if actual_value is None:
1874 return m.group('none_inclusive')
1875 return op(actual_value, comparison_value)
67134eab
JMF
1876 return _filter
1877
9f1a1c36 1878 def _check_formats(self, formats):
1879 for f in formats:
1880 self.to_screen('[info] Testing format %s' % f['format_id'])
75689fe5 1881 path = self.get_output_path('temp')
1882 if not self._ensure_dir_exists(f'{path}/'):
1883 continue
1884 temp_file = tempfile.NamedTemporaryFile(suffix='.tmp', delete=False, dir=path or None)
9f1a1c36 1885 temp_file.close()
1886 try:
1887 success, _ = self.dl(temp_file.name, f, test=True)
1888 except (DownloadError, IOError, OSError, ValueError) + network_exceptions:
1889 success = False
1890 finally:
1891 if os.path.exists(temp_file.name):
1892 try:
1893 os.remove(temp_file.name)
1894 except OSError:
1895 self.report_warning('Unable to delete temporary file "%s"' % temp_file.name)
1896 if success:
1897 yield f
1898 else:
1899 self.to_screen('[info] Unable to download format %s. Skipping...' % f['format_id'])
1900
0017d9ad 1901 def _default_format_spec(self, info_dict, download=True):
0017d9ad 1902
af0f7428
S
1903 def can_merge():
1904 merger = FFmpegMergerPP(self)
1905 return merger.available and merger.can_merge()
1906
91ebc640 1907 prefer_best = (
b7b04c78 1908 not self.params.get('simulate')
91ebc640 1909 and download
1910 and (
1911 not can_merge()
19807826 1912 or info_dict.get('is_live', False)
de6000d9 1913 or self.outtmpl_dict['default'] == '-'))
53ed7066 1914 compat = (
1915 prefer_best
1916 or self.params.get('allow_multiple_audio_streams', False)
1917 or 'format-spec' in self.params.get('compat_opts', []))
91ebc640 1918
1919 return (
53ed7066 1920 'best/bestvideo+bestaudio' if prefer_best
1921 else 'bestvideo*+bestaudio/best' if not compat
91ebc640 1922 else 'bestvideo+bestaudio/best')
0017d9ad 1923
67134eab
JMF
1924 def build_format_selector(self, format_spec):
1925 def syntax_error(note, start):
1926 message = (
1927 'Invalid format specification: '
86e5f3ed 1928 '{}\n\t{}\n\t{}^'.format(note, format_spec, ' ' * start[1]))
67134eab
JMF
1929 return SyntaxError(message)
1930
1931 PICKFIRST = 'PICKFIRST'
1932 MERGE = 'MERGE'
1933 SINGLE = 'SINGLE'
0130afb7 1934 GROUP = 'GROUP'
67134eab
JMF
1935 FormatSelector = collections.namedtuple('FormatSelector', ['type', 'selector', 'filters'])
1936
91ebc640 1937 allow_multiple_streams = {'audio': self.params.get('allow_multiple_audio_streams', False),
1938 'video': self.params.get('allow_multiple_video_streams', False)}
909d24dd 1939
9f1a1c36 1940 check_formats = self.params.get('check_formats') == 'selected'
e8e73840 1941
67134eab
JMF
1942 def _parse_filter(tokens):
1943 filter_parts = []
1944 for type, string, start, _, _ in tokens:
1945 if type == tokenize.OP and string == ']':
1946 return ''.join(filter_parts)
1947 else:
1948 filter_parts.append(string)
1949
232541df 1950 def _remove_unused_ops(tokens):
17cc1534 1951 # Remove operators that we don't use and join them with the surrounding strings
232541df
JMF
1952 # for example: 'mp4' '-' 'baseline' '-' '16x9' is converted to 'mp4-baseline-16x9'
1953 ALLOWED_OPS = ('/', '+', ',', '(', ')')
1954 last_string, last_start, last_end, last_line = None, None, None, None
1955 for type, string, start, end, line in tokens:
1956 if type == tokenize.OP and string == '[':
1957 if last_string:
1958 yield tokenize.NAME, last_string, last_start, last_end, last_line
1959 last_string = None
1960 yield type, string, start, end, line
1961 # everything inside brackets will be handled by _parse_filter
1962 for type, string, start, end, line in tokens:
1963 yield type, string, start, end, line
1964 if type == tokenize.OP and string == ']':
1965 break
1966 elif type == tokenize.OP and string in ALLOWED_OPS:
1967 if last_string:
1968 yield tokenize.NAME, last_string, last_start, last_end, last_line
1969 last_string = None
1970 yield type, string, start, end, line
1971 elif type in [tokenize.NAME, tokenize.NUMBER, tokenize.OP]:
1972 if not last_string:
1973 last_string = string
1974 last_start = start
1975 last_end = end
1976 else:
1977 last_string += string
1978 if last_string:
1979 yield tokenize.NAME, last_string, last_start, last_end, last_line
1980
cf2ac6df 1981 def _parse_format_selection(tokens, inside_merge=False, inside_choice=False, inside_group=False):
67134eab
JMF
1982 selectors = []
1983 current_selector = None
1984 for type, string, start, _, _ in tokens:
1985 # ENCODING is only defined in python 3.x
1986 if type == getattr(tokenize, 'ENCODING', None):
1987 continue
1988 elif type in [tokenize.NAME, tokenize.NUMBER]:
1989 current_selector = FormatSelector(SINGLE, string, [])
1990 elif type == tokenize.OP:
cf2ac6df
JMF
1991 if string == ')':
1992 if not inside_group:
1993 # ')' will be handled by the parentheses group
1994 tokens.restore_last_token()
67134eab 1995 break
cf2ac6df 1996 elif inside_merge and string in ['/', ',']:
0130afb7
JMF
1997 tokens.restore_last_token()
1998 break
cf2ac6df
JMF
1999 elif inside_choice and string == ',':
2000 tokens.restore_last_token()
2001 break
2002 elif string == ',':
0a31a350
JMF
2003 if not current_selector:
2004 raise syntax_error('"," must follow a format selector', start)
67134eab
JMF
2005 selectors.append(current_selector)
2006 current_selector = None
2007 elif string == '/':
d96d604e
JMF
2008 if not current_selector:
2009 raise syntax_error('"/" must follow a format selector', start)
67134eab 2010 first_choice = current_selector
cf2ac6df 2011 second_choice = _parse_format_selection(tokens, inside_choice=True)
f5f4a27a 2012 current_selector = FormatSelector(PICKFIRST, (first_choice, second_choice), [])
67134eab
JMF
2013 elif string == '[':
2014 if not current_selector:
2015 current_selector = FormatSelector(SINGLE, 'best', [])
2016 format_filter = _parse_filter(tokens)
2017 current_selector.filters.append(format_filter)
0130afb7
JMF
2018 elif string == '(':
2019 if current_selector:
2020 raise syntax_error('Unexpected "("', start)
cf2ac6df
JMF
2021 group = _parse_format_selection(tokens, inside_group=True)
2022 current_selector = FormatSelector(GROUP, group, [])
67134eab 2023 elif string == '+':
d03cfdce 2024 if not current_selector:
2025 raise syntax_error('Unexpected "+"', start)
2026 selector_1 = current_selector
2027 selector_2 = _parse_format_selection(tokens, inside_merge=True)
2028 if not selector_2:
2029 raise syntax_error('Expected a selector', start)
2030 current_selector = FormatSelector(MERGE, (selector_1, selector_2), [])
67134eab 2031 else:
86e5f3ed 2032 raise syntax_error(f'Operator not recognized: "{string}"', start)
67134eab
JMF
2033 elif type == tokenize.ENDMARKER:
2034 break
2035 if current_selector:
2036 selectors.append(current_selector)
2037 return selectors
2038
f8d4ad9a 2039 def _merge(formats_pair):
2040 format_1, format_2 = formats_pair
2041
2042 formats_info = []
2043 formats_info.extend(format_1.get('requested_formats', (format_1,)))
2044 formats_info.extend(format_2.get('requested_formats', (format_2,)))
2045
2046 if not allow_multiple_streams['video'] or not allow_multiple_streams['audio']:
551f9388 2047 get_no_more = {'video': False, 'audio': False}
f8d4ad9a 2048 for (i, fmt_info) in enumerate(formats_info):
551f9388 2049 if fmt_info.get('acodec') == fmt_info.get('vcodec') == 'none':
2050 formats_info.pop(i)
2051 continue
2052 for aud_vid in ['audio', 'video']:
f8d4ad9a 2053 if not allow_multiple_streams[aud_vid] and fmt_info.get(aud_vid[0] + 'codec') != 'none':
2054 if get_no_more[aud_vid]:
2055 formats_info.pop(i)
f5510afe 2056 break
f8d4ad9a 2057 get_no_more[aud_vid] = True
2058
2059 if len(formats_info) == 1:
2060 return formats_info[0]
2061
2062 video_fmts = [fmt_info for fmt_info in formats_info if fmt_info.get('vcodec') != 'none']
2063 audio_fmts = [fmt_info for fmt_info in formats_info if fmt_info.get('acodec') != 'none']
2064
2065 the_only_video = video_fmts[0] if len(video_fmts) == 1 else None
2066 the_only_audio = audio_fmts[0] if len(audio_fmts) == 1 else None
2067
2068 output_ext = self.params.get('merge_output_format')
2069 if not output_ext:
2070 if the_only_video:
2071 output_ext = the_only_video['ext']
2072 elif the_only_audio and not video_fmts:
2073 output_ext = the_only_audio['ext']
2074 else:
2075 output_ext = 'mkv'
2076
975a0d0d 2077 filtered = lambda *keys: filter(None, (traverse_obj(fmt, *keys) for fmt in formats_info))
2078
f8d4ad9a 2079 new_dict = {
2080 'requested_formats': formats_info,
975a0d0d 2081 'format': '+'.join(filtered('format')),
2082 'format_id': '+'.join(filtered('format_id')),
f8d4ad9a 2083 'ext': output_ext,
975a0d0d 2084 'protocol': '+'.join(map(determine_protocol, formats_info)),
093a1710 2085 'language': '+'.join(orderedSet(filtered('language'))) or None,
2086 'format_note': '+'.join(orderedSet(filtered('format_note'))) or None,
2087 'filesize_approx': sum(filtered('filesize', 'filesize_approx')) or None,
975a0d0d 2088 'tbr': sum(filtered('tbr', 'vbr', 'abr')),
f8d4ad9a 2089 }
2090
2091 if the_only_video:
2092 new_dict.update({
2093 'width': the_only_video.get('width'),
2094 'height': the_only_video.get('height'),
2095 'resolution': the_only_video.get('resolution') or self.format_resolution(the_only_video),
2096 'fps': the_only_video.get('fps'),
49a57e70 2097 'dynamic_range': the_only_video.get('dynamic_range'),
f8d4ad9a 2098 'vcodec': the_only_video.get('vcodec'),
2099 'vbr': the_only_video.get('vbr'),
2100 'stretched_ratio': the_only_video.get('stretched_ratio'),
2101 })
2102
2103 if the_only_audio:
2104 new_dict.update({
2105 'acodec': the_only_audio.get('acodec'),
2106 'abr': the_only_audio.get('abr'),
975a0d0d 2107 'asr': the_only_audio.get('asr'),
f8d4ad9a 2108 })
2109
2110 return new_dict
2111
e8e73840 2112 def _check_formats(formats):
981052c9 2113 if not check_formats:
2114 yield from formats
b5ac45b1 2115 return
9f1a1c36 2116 yield from self._check_formats(formats)
e8e73840 2117
67134eab 2118 def _build_selector_function(selector):
909d24dd 2119 if isinstance(selector, list): # ,
67134eab
JMF
2120 fs = [_build_selector_function(s) for s in selector]
2121
317f7ab6 2122 def selector_function(ctx):
67134eab 2123 for f in fs:
981052c9 2124 yield from f(ctx)
67134eab 2125 return selector_function
909d24dd 2126
2127 elif selector.type == GROUP: # ()
0130afb7 2128 selector_function = _build_selector_function(selector.selector)
909d24dd 2129
2130 elif selector.type == PICKFIRST: # /
67134eab
JMF
2131 fs = [_build_selector_function(s) for s in selector.selector]
2132
317f7ab6 2133 def selector_function(ctx):
67134eab 2134 for f in fs:
317f7ab6 2135 picked_formats = list(f(ctx))
67134eab
JMF
2136 if picked_formats:
2137 return picked_formats
2138 return []
67134eab 2139
981052c9 2140 elif selector.type == MERGE: # +
2141 selector_1, selector_2 = map(_build_selector_function, selector.selector)
2142
2143 def selector_function(ctx):
adbc4ec4 2144 for pair in itertools.product(selector_1(ctx), selector_2(ctx)):
981052c9 2145 yield _merge(pair)
2146
909d24dd 2147 elif selector.type == SINGLE: # atom
598d185d 2148 format_spec = selector.selector or 'best'
909d24dd 2149
f8d4ad9a 2150 # TODO: Add allvideo, allaudio etc by generalizing the code with best/worst selector
909d24dd 2151 if format_spec == 'all':
2152 def selector_function(ctx):
9222c381 2153 yield from _check_formats(ctx['formats'][::-1])
f8d4ad9a 2154 elif format_spec == 'mergeall':
2155 def selector_function(ctx):
316f2650 2156 formats = list(_check_formats(
2157 f for f in ctx['formats'] if f.get('vcodec') != 'none' or f.get('acodec') != 'none'))
e01d6aa4 2158 if not formats:
2159 return
921b76ca 2160 merged_format = formats[-1]
2161 for f in formats[-2::-1]:
f8d4ad9a 2162 merged_format = _merge((merged_format, f))
2163 yield merged_format
909d24dd 2164
2165 else:
85e801a9 2166 format_fallback, seperate_fallback, format_reverse, format_idx = False, None, True, 1
eff63539 2167 mobj = re.match(
2168 r'(?P<bw>best|worst|b|w)(?P<type>video|audio|v|a)?(?P<mod>\*)?(?:\.(?P<n>[1-9]\d*))?$',
2169 format_spec)
2170 if mobj is not None:
2171 format_idx = int_or_none(mobj.group('n'), default=1)
e8e73840 2172 format_reverse = mobj.group('bw')[0] == 'b'
eff63539 2173 format_type = (mobj.group('type') or [None])[0]
2174 not_format_type = {'v': 'a', 'a': 'v'}.get(format_type)
2175 format_modified = mobj.group('mod') is not None
909d24dd 2176
2177 format_fallback = not format_type and not format_modified # for b, w
8326b00a 2178 _filter_f = (
eff63539 2179 (lambda f: f.get('%scodec' % format_type) != 'none')
2180 if format_type and format_modified # bv*, ba*, wv*, wa*
2181 else (lambda f: f.get('%scodec' % not_format_type) == 'none')
2182 if format_type # bv, ba, wv, wa
2183 else (lambda f: f.get('vcodec') != 'none' and f.get('acodec') != 'none')
2184 if not format_modified # b, w
8326b00a 2185 else lambda f: True) # b*, w*
2186 filter_f = lambda f: _filter_f(f) and (
2187 f.get('vcodec') != 'none' or f.get('acodec') != 'none')
67134eab 2188 else:
48ee10ee 2189 if format_spec in self._format_selection_exts['audio']:
b11c04a8 2190 filter_f = lambda f: f.get('ext') == format_spec and f.get('acodec') != 'none'
48ee10ee 2191 elif format_spec in self._format_selection_exts['video']:
b11c04a8 2192 filter_f = lambda f: f.get('ext') == format_spec and f.get('acodec') != 'none' and f.get('vcodec') != 'none'
85e801a9 2193 seperate_fallback = lambda f: f.get('ext') == format_spec and f.get('vcodec') != 'none'
48ee10ee 2194 elif format_spec in self._format_selection_exts['storyboards']:
b11c04a8 2195 filter_f = lambda f: f.get('ext') == format_spec and f.get('acodec') == 'none' and f.get('vcodec') == 'none'
2196 else:
b5ae35ee 2197 filter_f = lambda f: f.get('format_id') == format_spec # id
909d24dd 2198
2199 def selector_function(ctx):
2200 formats = list(ctx['formats'])
909d24dd 2201 matches = list(filter(filter_f, formats)) if filter_f is not None else formats
85e801a9 2202 if not matches:
2203 if format_fallback and ctx['incomplete_formats']:
2204 # for extractors with incomplete formats (audio only (soundcloud)
2205 # or video only (imgur)) best/worst will fallback to
2206 # best/worst {video,audio}-only format
2207 matches = formats
2208 elif seperate_fallback and not ctx['has_merged_format']:
2209 # for compatibility with youtube-dl when there is no pre-merged format
2210 matches = list(filter(seperate_fallback, formats))
981052c9 2211 matches = LazyList(_check_formats(matches[::-1 if format_reverse else 1]))
2212 try:
e8e73840 2213 yield matches[format_idx - 1]
4abea8ca 2214 except LazyList.IndexError:
981052c9 2215 return
083c9df9 2216
67134eab 2217 filters = [self._build_format_filter(f) for f in selector.filters]
083c9df9 2218
317f7ab6 2219 def final_selector(ctx):
adbc4ec4 2220 ctx_copy = dict(ctx)
67134eab 2221 for _filter in filters:
317f7ab6
S
2222 ctx_copy['formats'] = list(filter(_filter, ctx_copy['formats']))
2223 return selector_function(ctx_copy)
67134eab 2224 return final_selector
083c9df9 2225
67134eab 2226 stream = io.BytesIO(format_spec.encode('utf-8'))
0130afb7 2227 try:
f9934b96 2228 tokens = list(_remove_unused_ops(tokenize.tokenize(stream.readline)))
0130afb7
JMF
2229 except tokenize.TokenError:
2230 raise syntax_error('Missing closing/opening brackets or parenthesis', (0, len(format_spec)))
2231
86e5f3ed 2232 class TokenIterator:
0130afb7
JMF
2233 def __init__(self, tokens):
2234 self.tokens = tokens
2235 self.counter = 0
2236
2237 def __iter__(self):
2238 return self
2239
2240 def __next__(self):
2241 if self.counter >= len(self.tokens):
2242 raise StopIteration()
2243 value = self.tokens[self.counter]
2244 self.counter += 1
2245 return value
2246
2247 next = __next__
2248
2249 def restore_last_token(self):
2250 self.counter -= 1
2251
2252 parsed_selector = _parse_format_selection(iter(TokenIterator(tokens)))
67134eab 2253 return _build_selector_function(parsed_selector)
a9c58ad9 2254
e5660ee6 2255 def _calc_headers(self, info_dict):
8b7539d2 2256 res = merge_headers(self.params['http_headers'], info_dict.get('http_headers') or {})
e5660ee6
JMF
2257
2258 cookies = self._calc_cookies(info_dict)
2259 if cookies:
2260 res['Cookie'] = cookies
2261
0016b84e
S
2262 if 'X-Forwarded-For' not in res:
2263 x_forwarded_for_ip = info_dict.get('__x_forwarded_for_ip')
2264 if x_forwarded_for_ip:
2265 res['X-Forwarded-For'] = x_forwarded_for_ip
2266
e5660ee6
JMF
2267 return res
2268
2269 def _calc_cookies(self, info_dict):
5c2266df 2270 pr = sanitized_Request(info_dict['url'])
e5660ee6 2271 self.cookiejar.add_cookie_header(pr)
662435f7 2272 return pr.get_header('Cookie')
e5660ee6 2273
9f1a1c36 2274 def _sort_thumbnails(self, thumbnails):
2275 thumbnails.sort(key=lambda t: (
2276 t.get('preference') if t.get('preference') is not None else -1,
2277 t.get('width') if t.get('width') is not None else -1,
2278 t.get('height') if t.get('height') is not None else -1,
2279 t.get('id') if t.get('id') is not None else '',
2280 t.get('url')))
2281
b0249bca 2282 def _sanitize_thumbnails(self, info_dict):
bc516a3f 2283 thumbnails = info_dict.get('thumbnails')
2284 if thumbnails is None:
2285 thumbnail = info_dict.get('thumbnail')
2286 if thumbnail:
2287 info_dict['thumbnails'] = thumbnails = [{'url': thumbnail}]
9f1a1c36 2288 if not thumbnails:
2289 return
2290
2291 def check_thumbnails(thumbnails):
2292 for t in thumbnails:
2293 self.to_screen(f'[info] Testing thumbnail {t["id"]}')
2294 try:
2295 self.urlopen(HEADRequest(t['url']))
2296 except network_exceptions as err:
2297 self.to_screen(f'[info] Unable to connect to thumbnail {t["id"]} URL {t["url"]!r} - {err}. Skipping...')
2298 continue
2299 yield t
2300
2301 self._sort_thumbnails(thumbnails)
2302 for i, t in enumerate(thumbnails):
2303 if t.get('id') is None:
2304 t['id'] = '%d' % i
2305 if t.get('width') and t.get('height'):
2306 t['resolution'] = '%dx%d' % (t['width'], t['height'])
2307 t['url'] = sanitize_url(t['url'])
2308
2309 if self.params.get('check_formats') is True:
282f5709 2310 info_dict['thumbnails'] = LazyList(check_thumbnails(thumbnails[::-1]), reverse=True)
9f1a1c36 2311 else:
2312 info_dict['thumbnails'] = thumbnails
bc516a3f 2313
03f83004
LNO
2314 def _fill_common_fields(self, info_dict, is_video=True):
2315 # TODO: move sanitization here
2316 if is_video:
2317 # playlists are allowed to lack "title"
2318 info_dict['fulltitle'] = info_dict.get('title')
2319 if 'title' not in info_dict:
2320 raise ExtractorError('Missing "title" field in extractor result',
2321 video_id=info_dict['id'], ie=info_dict['extractor'])
2322 elif not info_dict.get('title'):
2323 self.report_warning('Extractor failed to obtain "title". Creating a generic title instead')
2324 info_dict['title'] = f'{info_dict["extractor"]} video #{info_dict["id"]}'
2325
2326 if info_dict.get('duration') is not None:
2327 info_dict['duration_string'] = formatSeconds(info_dict['duration'])
2328
2329 for ts_key, date_key in (
2330 ('timestamp', 'upload_date'),
2331 ('release_timestamp', 'release_date'),
2332 ('modified_timestamp', 'modified_date'),
2333 ):
2334 if info_dict.get(date_key) is None and info_dict.get(ts_key) is not None:
2335 # Working around out-of-range timestamp values (e.g. negative ones on Windows,
2336 # see http://bugs.python.org/issue1646728)
19a03940 2337 with contextlib.suppress(ValueError, OverflowError, OSError):
03f83004
LNO
2338 upload_date = datetime.datetime.utcfromtimestamp(info_dict[ts_key])
2339 info_dict[date_key] = upload_date.strftime('%Y%m%d')
03f83004
LNO
2340
2341 live_keys = ('is_live', 'was_live')
2342 live_status = info_dict.get('live_status')
2343 if live_status is None:
2344 for key in live_keys:
2345 if info_dict.get(key) is False:
2346 continue
2347 if info_dict.get(key):
2348 live_status = key
2349 break
2350 if all(info_dict.get(key) is False for key in live_keys):
2351 live_status = 'not_live'
2352 if live_status:
2353 info_dict['live_status'] = live_status
2354 for key in live_keys:
2355 if info_dict.get(key) is None:
2356 info_dict[key] = (live_status == key)
2357
2358 # Auto generate title fields corresponding to the *_number fields when missing
2359 # in order to always have clean titles. This is very common for TV series.
2360 for field in ('chapter', 'season', 'episode'):
2361 if info_dict.get('%s_number' % field) is not None and not info_dict.get(field):
2362 info_dict[field] = '%s %d' % (field.capitalize(), info_dict['%s_number' % field])
2363
dd82ffea
JMF
2364 def process_video_result(self, info_dict, download=True):
2365 assert info_dict.get('_type', 'video') == 'video'
9c906919 2366 self._num_videos += 1
dd82ffea 2367
bec1fad2 2368 if 'id' not in info_dict:
fc08bdd6 2369 raise ExtractorError('Missing "id" field in extractor result', ie=info_dict['extractor'])
2370 elif not info_dict.get('id'):
2371 raise ExtractorError('Extractor failed to obtain "id"', ie=info_dict['extractor'])
455a15e2 2372
c9969434
S
2373 def report_force_conversion(field, field_not, conversion):
2374 self.report_warning(
2375 '"%s" field is not %s - forcing %s conversion, there is an error in extractor'
2376 % (field, field_not, conversion))
2377
2378 def sanitize_string_field(info, string_field):
2379 field = info.get(string_field)
2380 if field is None or isinstance(field, compat_str):
2381 return
2382 report_force_conversion(string_field, 'a string', 'string')
2383 info[string_field] = compat_str(field)
2384
2385 def sanitize_numeric_fields(info):
2386 for numeric_field in self._NUMERIC_FIELDS:
2387 field = info.get(numeric_field)
f9934b96 2388 if field is None or isinstance(field, (int, float)):
c9969434
S
2389 continue
2390 report_force_conversion(numeric_field, 'numeric', 'int')
2391 info[numeric_field] = int_or_none(field)
2392
2393 sanitize_string_field(info_dict, 'id')
2394 sanitize_numeric_fields(info_dict)
4c3f8c3f 2395 if (info_dict.get('duration') or 0) <= 0 and info_dict.pop('duration', None):
50e93e03 2396 self.report_warning('"duration" field is negative, there is an error in extractor')
be6217b2 2397
dd82ffea
JMF
2398 if 'playlist' not in info_dict:
2399 # It isn't part of a playlist
2400 info_dict['playlist'] = None
2401 info_dict['playlist_index'] = None
2402
bc516a3f 2403 self._sanitize_thumbnails(info_dict)
d5519808 2404
536a55da 2405 thumbnail = info_dict.get('thumbnail')
bc516a3f 2406 thumbnails = info_dict.get('thumbnails')
536a55da
S
2407 if thumbnail:
2408 info_dict['thumbnail'] = sanitize_url(thumbnail)
2409 elif thumbnails:
d5519808
PH
2410 info_dict['thumbnail'] = thumbnails[-1]['url']
2411
ae30b840 2412 if info_dict.get('display_id') is None and 'id' in info_dict:
0afef30b
PH
2413 info_dict['display_id'] = info_dict['id']
2414
03f83004 2415 self._fill_common_fields(info_dict)
33d2fc2f 2416
05108a49
S
2417 for cc_kind in ('subtitles', 'automatic_captions'):
2418 cc = info_dict.get(cc_kind)
2419 if cc:
2420 for _, subtitle in cc.items():
2421 for subtitle_format in subtitle:
2422 if subtitle_format.get('url'):
2423 subtitle_format['url'] = sanitize_url(subtitle_format['url'])
2424 if subtitle_format.get('ext') is None:
2425 subtitle_format['ext'] = determine_ext(subtitle_format['url']).lower()
2426
2427 automatic_captions = info_dict.get('automatic_captions')
4bba3716 2428 subtitles = info_dict.get('subtitles')
4bba3716 2429
360e1ca5 2430 info_dict['requested_subtitles'] = self.process_subtitles(
05108a49 2431 info_dict['id'], subtitles, automatic_captions)
a504ced0 2432
dd82ffea
JMF
2433 if info_dict.get('formats') is None:
2434 # There's only one format available
2435 formats = [info_dict]
2436 else:
2437 formats = info_dict['formats']
2438
e0493e90 2439 info_dict['__has_drm'] = any(f.get('has_drm') for f in formats)
88acdbc2 2440 if not self.params.get('allow_unplayable_formats'):
2441 formats = [f for f in formats if not f.get('has_drm')]
c0b6e5c7 2442 if info_dict['__has_drm'] and all(
2443 f.get('acodec') == f.get('vcodec') == 'none' for f in formats):
2444 self.report_warning(
2445 'This video is DRM protected and only images are available for download. '
2446 'Use --list-formats to see them')
88acdbc2 2447
319b6059 2448 get_from_start = not info_dict.get('is_live') or bool(self.params.get('live_from_start'))
2449 if not get_from_start:
2450 info_dict['title'] += ' ' + datetime.datetime.now().strftime('%Y-%m-%d %H:%M')
2451 if info_dict.get('is_live') and formats:
adbc4ec4 2452 formats = [f for f in formats if bool(f.get('is_from_start')) == get_from_start]
319b6059 2453 if get_from_start and not formats:
a44ca5a4 2454 self.raise_no_formats(info_dict, msg=(
2455 '--live-from-start is passed, but there are no formats that can be downloaded from the start. '
2456 'If you want to download from the current time, use --no-live-from-start'))
adbc4ec4 2457
db95dc13 2458 if not formats:
1151c407 2459 self.raise_no_formats(info_dict)
db95dc13 2460
73af5cc8
S
2461 def is_wellformed(f):
2462 url = f.get('url')
a5ac0c47 2463 if not url:
73af5cc8
S
2464 self.report_warning(
2465 '"url" field is missing or empty - skipping format, '
2466 'there is an error in extractor')
a5ac0c47
S
2467 return False
2468 if isinstance(url, bytes):
2469 sanitize_string_field(f, 'url')
2470 return True
73af5cc8
S
2471
2472 # Filter out malformed formats for better extraction robustness
2473 formats = list(filter(is_wellformed, formats))
2474
181c7053
S
2475 formats_dict = {}
2476
dd82ffea 2477 # We check that all the formats have the format and format_id fields
db95dc13 2478 for i, format in enumerate(formats):
c9969434
S
2479 sanitize_string_field(format, 'format_id')
2480 sanitize_numeric_fields(format)
dcf77cf1 2481 format['url'] = sanitize_url(format['url'])
e74e3b63 2482 if not format.get('format_id'):
8016c922 2483 format['format_id'] = compat_str(i)
e2effb08
S
2484 else:
2485 # Sanitize format_id from characters used in format selector expression
ec85ded8 2486 format['format_id'] = re.sub(r'[\s,/+\[\]()]', '_', format['format_id'])
181c7053
S
2487 format_id = format['format_id']
2488 if format_id not in formats_dict:
2489 formats_dict[format_id] = []
2490 formats_dict[format_id].append(format)
2491
2492 # Make sure all formats have unique format_id
03b4de72 2493 common_exts = set(itertools.chain(*self._format_selection_exts.values()))
181c7053 2494 for format_id, ambiguous_formats in formats_dict.items():
48ee10ee 2495 ambigious_id = len(ambiguous_formats) > 1
2496 for i, format in enumerate(ambiguous_formats):
2497 if ambigious_id:
181c7053 2498 format['format_id'] = '%s-%d' % (format_id, i)
48ee10ee 2499 if format.get('ext') is None:
2500 format['ext'] = determine_ext(format['url']).lower()
2501 # Ensure there is no conflict between id and ext in format selection
2502 # See https://github.com/yt-dlp/yt-dlp/issues/1282
2503 if format['format_id'] != format['ext'] and format['format_id'] in common_exts:
2504 format['format_id'] = 'f%s' % format['format_id']
181c7053
S
2505
2506 for i, format in enumerate(formats):
8c51aa65 2507 if format.get('format') is None:
6febd1c1 2508 format['format'] = '{id} - {res}{note}'.format(
8c51aa65
JMF
2509 id=format['format_id'],
2510 res=self.format_resolution(format),
b868936c 2511 note=format_field(format, 'format_note', ' (%s)'),
8c51aa65 2512 )
6f0be937 2513 if format.get('protocol') is None:
b5559424 2514 format['protocol'] = determine_protocol(format)
239df021 2515 if format.get('resolution') is None:
2516 format['resolution'] = self.format_resolution(format, default=None)
176f1866 2517 if format.get('dynamic_range') is None and format.get('vcodec') != 'none':
2518 format['dynamic_range'] = 'SDR'
f2fe69c7 2519 if (info_dict.get('duration') and format.get('tbr')
2520 and not format.get('filesize') and not format.get('filesize_approx')):
2521 format['filesize_approx'] = info_dict['duration'] * format['tbr'] * (1024 / 8)
2522
e5660ee6
JMF
2523 # Add HTTP headers, so that external programs can use them from the
2524 # json output
2525 full_format_info = info_dict.copy()
2526 full_format_info.update(format)
2527 format['http_headers'] = self._calc_headers(full_format_info)
0016b84e
S
2528 # Remove private housekeeping stuff
2529 if '__x_forwarded_for_ip' in info_dict:
2530 del info_dict['__x_forwarded_for_ip']
dd82ffea 2531
9f1a1c36 2532 if self.params.get('check_formats') is True:
282f5709 2533 formats = LazyList(self._check_formats(formats[::-1]), reverse=True)
9f1a1c36 2534
88acdbc2 2535 if not formats or formats[0] is not info_dict:
b3d9ef88
JMF
2536 # only set the 'formats' fields if the original info_dict list them
2537 # otherwise we end up with a circular reference, the first (and unique)
f89197d7 2538 # element in the 'formats' field in info_dict is info_dict itself,
dfb1b146 2539 # which can't be exported to json
b3d9ef88 2540 info_dict['formats'] = formats
4ec82a72 2541
2542 info_dict, _ = self.pre_process(info_dict)
2543
6db9c4d5 2544 if self._match_entry(info_dict, incomplete=self._format_fields) is not None:
09b49e1f 2545 return info_dict
2546
2547 self.post_extract(info_dict)
2548 info_dict, _ = self.pre_process(info_dict, 'after_filter')
2549
093a1710 2550 # The pre-processors may have modified the formats
2551 formats = info_dict.get('formats', [info_dict])
2552
fa9f30b8 2553 list_only = self.params.get('simulate') is None and (
2554 self.params.get('list_thumbnails') or self.params.get('listformats') or self.params.get('listsubtitles'))
2555 interactive_format_selection = not list_only and self.format_selector == '-'
b7b04c78 2556 if self.params.get('list_thumbnails'):
2557 self.list_thumbnails(info_dict)
b7b04c78 2558 if self.params.get('listsubtitles'):
2559 if 'automatic_captions' in info_dict:
2560 self.list_subtitles(
2561 info_dict['id'], automatic_captions, 'automatic captions')
2562 self.list_subtitles(info_dict['id'], subtitles, 'subtitles')
fa9f30b8 2563 if self.params.get('listformats') or interactive_format_selection:
b69fd25c 2564 self.list_formats(info_dict)
169dbde9 2565 if list_only:
b7b04c78 2566 # Without this printing, -F --print-json will not work
169dbde9 2567 self.__forced_printings(info_dict, self.prepare_filename(info_dict), incomplete=True)
bfaae0a7 2568 return
2569
187986a8 2570 format_selector = self.format_selector
2571 if format_selector is None:
0017d9ad 2572 req_format = self._default_format_spec(info_dict, download=download)
0760b0a7 2573 self.write_debug('Default format spec: %s' % req_format)
187986a8 2574 format_selector = self.build_format_selector(req_format)
317f7ab6 2575
fa9f30b8 2576 while True:
2577 if interactive_format_selection:
2578 req_format = input(
2579 self._format_screen('\nEnter format selector: ', self.Styles.EMPHASIS))
2580 try:
2581 format_selector = self.build_format_selector(req_format)
2582 except SyntaxError as err:
2583 self.report_error(err, tb=False, is_error=False)
2584 continue
2585
85e801a9 2586 formats_to_download = list(format_selector({
fa9f30b8 2587 'formats': formats,
85e801a9 2588 'has_merged_format': any('none' not in (f.get('acodec'), f.get('vcodec')) for f in formats),
2589 'incomplete_formats': (
2590 # All formats are video-only or
2591 all(f.get('vcodec') != 'none' and f.get('acodec') == 'none' for f in formats)
2592 # all formats are audio-only
2593 or all(f.get('vcodec') == 'none' and f.get('acodec') != 'none' for f in formats)),
2594 }))
fa9f30b8 2595 if interactive_format_selection and not formats_to_download:
2596 self.report_error('Requested format is not available', tb=False, is_error=False)
2597 continue
2598 break
317f7ab6 2599
dd82ffea 2600 if not formats_to_download:
b7da73eb 2601 if not self.params.get('ignore_no_formats_error'):
c0b6e5c7 2602 raise ExtractorError(
2603 'Requested format is not available. Use --list-formats for a list of available formats',
2604 expected=True, video_id=info_dict['id'], ie=info_dict['extractor'])
b62fa6d7 2605 self.report_warning('Requested format is not available')
2606 # Process what we can, even without any available formats.
2607 formats_to_download = [{}]
a13e6848 2608
b62fa6d7 2609 best_format = formats_to_download[-1]
2610 if download:
2611 if best_format:
2612 self.to_screen(
2613 f'[info] {info_dict["id"]}: Downloading {len(formats_to_download)} format(s): '
2614 + ', '.join([f['format_id'] for f in formats_to_download]))
a13e6848 2615 max_downloads_reached = False
f46e2f9d 2616 for i, fmt in enumerate(formats_to_download):
09b49e1f 2617 formats_to_download[i] = new_info = self._copy_infodict(info_dict)
b7da73eb 2618 new_info.update(fmt)
a13e6848 2619 try:
2620 self.process_info(new_info)
2621 except MaxDownloadsReached:
2622 max_downloads_reached = True
f46e2f9d 2623 # Remove copied info
2624 for key, val in tuple(new_info.items()):
2625 if info_dict.get(key) == val:
2626 new_info.pop(key)
a13e6848 2627 if max_downloads_reached:
2628 break
ebed8b37 2629
86e5f3ed 2630 write_archive = {f.get('__write_download_archive', False) for f in formats_to_download}
a13e6848 2631 assert write_archive.issubset({True, False, 'ignore'})
2632 if True in write_archive and False not in write_archive:
2633 self.record_download_archive(info_dict)
be72c624 2634
2635 info_dict['requested_downloads'] = formats_to_download
ed5835b4 2636 info_dict = self.run_all_pps('after_video', info_dict)
a13e6848 2637 if max_downloads_reached:
2638 raise MaxDownloadsReached()
ebed8b37 2639
49a57e70 2640 # We update the info dict with the selected best quality format (backwards compatibility)
be72c624 2641 info_dict.update(best_format)
dd82ffea
JMF
2642 return info_dict
2643
98c70d6f 2644 def process_subtitles(self, video_id, normal_subtitles, automatic_captions):
a504ced0 2645 """Select the requested subtitles and their format"""
d8a58ddc 2646 available_subs, normal_sub_langs = {}, []
98c70d6f
JMF
2647 if normal_subtitles and self.params.get('writesubtitles'):
2648 available_subs.update(normal_subtitles)
d8a58ddc 2649 normal_sub_langs = tuple(normal_subtitles.keys())
98c70d6f
JMF
2650 if automatic_captions and self.params.get('writeautomaticsub'):
2651 for lang, cap_info in automatic_captions.items():
360e1ca5
JMF
2652 if lang not in available_subs:
2653 available_subs[lang] = cap_info
2654
4d171848
JMF
2655 if (not self.params.get('writesubtitles') and not
2656 self.params.get('writeautomaticsub') or not
2657 available_subs):
2658 return None
a504ced0 2659
d8a58ddc 2660 all_sub_langs = tuple(available_subs.keys())
a504ced0 2661 if self.params.get('allsubtitles', False):
c32b0aab 2662 requested_langs = all_sub_langs
2663 elif self.params.get('subtitleslangs', False):
77c4a9ef 2664 # A list is used so that the order of languages will be the same as
2665 # given in subtitleslangs. See https://github.com/yt-dlp/yt-dlp/issues/1041
2666 requested_langs = []
2667 for lang_re in self.params.get('subtitleslangs'):
77c4a9ef 2668 discard = lang_re[0] == '-'
c32b0aab 2669 if discard:
77c4a9ef 2670 lang_re = lang_re[1:]
3aa91540 2671 if lang_re == 'all':
2672 if discard:
2673 requested_langs = []
2674 else:
2675 requested_langs.extend(all_sub_langs)
2676 continue
77c4a9ef 2677 current_langs = filter(re.compile(lang_re + '$').match, all_sub_langs)
c32b0aab 2678 if discard:
2679 for lang in current_langs:
77c4a9ef 2680 while lang in requested_langs:
2681 requested_langs.remove(lang)
c32b0aab 2682 else:
77c4a9ef 2683 requested_langs.extend(current_langs)
2684 requested_langs = orderedSet(requested_langs)
d8a58ddc 2685 elif normal_sub_langs:
2686 requested_langs = ['en'] if 'en' in normal_sub_langs else normal_sub_langs[:1]
a504ced0 2687 else:
d8a58ddc 2688 requested_langs = ['en'] if 'en' in all_sub_langs else all_sub_langs[:1]
ad3dc496 2689 if requested_langs:
2690 self.write_debug('Downloading subtitles: %s' % ', '.join(requested_langs))
a504ced0
JMF
2691
2692 formats_query = self.params.get('subtitlesformat', 'best')
2693 formats_preference = formats_query.split('/') if formats_query else []
2694 subs = {}
2695 for lang in requested_langs:
2696 formats = available_subs.get(lang)
2697 if formats is None:
86e5f3ed 2698 self.report_warning(f'{lang} subtitles not available for {video_id}')
a504ced0 2699 continue
a504ced0
JMF
2700 for ext in formats_preference:
2701 if ext == 'best':
2702 f = formats[-1]
2703 break
2704 matches = list(filter(lambda f: f['ext'] == ext, formats))
2705 if matches:
2706 f = matches[-1]
2707 break
2708 else:
2709 f = formats[-1]
2710 self.report_warning(
2711 'No subtitle format found matching "%s" for language %s, '
2712 'using %s' % (formats_query, lang, f['ext']))
2713 subs[lang] = f
2714 return subs
2715
bb66c247 2716 def _forceprint(self, key, info_dict):
2717 if info_dict is None:
2718 return
2719 info_copy = info_dict.copy()
2720 info_copy['formats_table'] = self.render_formats_table(info_dict)
2721 info_copy['thumbnails_table'] = self.render_thumbnails_table(info_dict)
2722 info_copy['subtitles_table'] = self.render_subtitles_table(info_dict.get('id'), info_dict.get('subtitles'))
2723 info_copy['automatic_captions_table'] = self.render_subtitles_table(info_dict.get('id'), info_dict.get('automatic_captions'))
2724
2725 def format_tmpl(tmpl):
2726 mobj = re.match(r'\w+(=?)$', tmpl)
2727 if mobj and mobj.group(1):
2728 return f'{tmpl[:-1]} = %({tmpl[:-1]})r'
2729 elif mobj:
2730 return f'%({tmpl})s'
2731 return tmpl
8130779d 2732
bb66c247 2733 for tmpl in self.params['forceprint'].get(key, []):
2734 self.to_stdout(self.evaluate_outtmpl(format_tmpl(tmpl), info_copy))
2735
2736 for tmpl, file_tmpl in self.params['print_to_file'].get(key, []):
5127e92a 2737 filename = self.prepare_filename(info_dict, outtmpl=file_tmpl)
bb66c247 2738 tmpl = format_tmpl(tmpl)
2739 self.to_screen(f'[info] Writing {tmpl!r} to: {filename}')
8d93e69d 2740 if self._ensure_dir_exists(filename):
86e5f3ed 2741 with open(filename, 'a', encoding='utf-8') as f:
8d93e69d 2742 f.write(self.evaluate_outtmpl(tmpl, info_copy) + '\n')
ca30f449 2743
d06daf23 2744 def __forced_printings(self, info_dict, filename, incomplete):
53c18592 2745 def print_mandatory(field, actual_field=None):
2746 if actual_field is None:
2747 actual_field = field
d06daf23 2748 if (self.params.get('force%s' % field, False)
53c18592 2749 and (not incomplete or info_dict.get(actual_field) is not None)):
2750 self.to_stdout(info_dict[actual_field])
d06daf23
S
2751
2752 def print_optional(field):
2753 if (self.params.get('force%s' % field, False)
2754 and info_dict.get(field) is not None):
2755 self.to_stdout(info_dict[field])
2756
53c18592 2757 info_dict = info_dict.copy()
2758 if filename is not None:
2759 info_dict['filename'] = filename
2760 if info_dict.get('requested_formats') is not None:
2761 # For RTMP URLs, also include the playpath
2762 info_dict['urls'] = '\n'.join(f['url'] + f.get('play_path', '') for f in info_dict['requested_formats'])
10331a26 2763 elif info_dict.get('url'):
53c18592 2764 info_dict['urls'] = info_dict['url'] + info_dict.get('play_path', '')
2765
bb66c247 2766 if (self.params.get('forcejson')
2767 or self.params['forceprint'].get('video')
2768 or self.params['print_to_file'].get('video')):
2b8a2973 2769 self.post_extract(info_dict)
bb66c247 2770 self._forceprint('video', info_dict)
53c18592 2771
d06daf23
S
2772 print_mandatory('title')
2773 print_mandatory('id')
53c18592 2774 print_mandatory('url', 'urls')
d06daf23
S
2775 print_optional('thumbnail')
2776 print_optional('description')
53c18592 2777 print_optional('filename')
b868936c 2778 if self.params.get('forceduration') and info_dict.get('duration') is not None:
d06daf23
S
2779 self.to_stdout(formatSeconds(info_dict['duration']))
2780 print_mandatory('format')
53c18592 2781
2b8a2973 2782 if self.params.get('forcejson'):
6e84b215 2783 self.to_stdout(json.dumps(self.sanitize_info(info_dict)))
d06daf23 2784
e8e73840 2785 def dl(self, name, info, subtitle=False, test=False):
88acdbc2 2786 if not info.get('url'):
1151c407 2787 self.raise_no_formats(info, True)
e8e73840 2788
2789 if test:
2790 verbose = self.params.get('verbose')
2791 params = {
2792 'test': True,
a169858f 2793 'quiet': self.params.get('quiet') or not verbose,
e8e73840 2794 'verbose': verbose,
2795 'noprogress': not verbose,
2796 'nopart': True,
2797 'skip_unavailable_fragments': False,
2798 'keep_fragments': False,
2799 'overwrites': True,
2800 '_no_ytdl_file': True,
2801 }
2802 else:
2803 params = self.params
96fccc10 2804 fd = get_suitable_downloader(info, params, to_stdout=(name == '-'))(self, params)
e8e73840 2805 if not test:
2806 for ph in self._progress_hooks:
2807 fd.add_progress_hook(ph)
42676437
M
2808 urls = '", "'.join(
2809 (f['url'].split(',')[0] + ',<data>' if f['url'].startswith('data:') else f['url'])
2810 for f in info.get('requested_formats', []) or [info])
18e674b4 2811 self.write_debug('Invoking downloader on "%s"' % urls)
03b4de72 2812
adbc4ec4
THD
2813 # Note: Ideally info should be a deep-copied so that hooks cannot modify it.
2814 # But it may contain objects that are not deep-copyable
2815 new_info = self._copy_infodict(info)
e8e73840 2816 if new_info.get('http_headers') is None:
2817 new_info['http_headers'] = self._calc_headers(new_info)
2818 return fd.download(name, new_info, subtitle)
2819
e04938ab 2820 def existing_file(self, filepaths, *, default_overwrite=True):
2821 existing_files = list(filter(os.path.exists, orderedSet(filepaths)))
2822 if existing_files and not self.params.get('overwrites', default_overwrite):
2823 return existing_files[0]
2824
2825 for file in existing_files:
2826 self.report_file_delete(file)
2827 os.remove(file)
2828 return None
2829
8222d8de 2830 def process_info(self, info_dict):
09b49e1f 2831 """Process a single resolved IE result. (Modifies it in-place)"""
8222d8de
JMF
2832
2833 assert info_dict.get('_type', 'video') == 'video'
f46e2f9d 2834 original_infodict = info_dict
fd288278 2835
4513a41a 2836 if 'format' not in info_dict and 'ext' in info_dict:
8222d8de
JMF
2837 info_dict['format'] = info_dict['ext']
2838
09b49e1f 2839 # This is mostly just for backward compatibility of process_info
2840 # As a side-effect, this allows for format-specific filters
c77495e3 2841 if self._match_entry(info_dict) is not None:
9e907ebd 2842 info_dict['__write_download_archive'] = 'ignore'
8222d8de
JMF
2843 return
2844
09b49e1f 2845 # Does nothing under normal operation - for backward compatibility of process_info
277d6ff5 2846 self.post_extract(info_dict)
0c14d66a 2847 self._num_downloads += 1
8222d8de 2848
dcf64d43 2849 # info_dict['_filename'] needs to be set for backward compatibility
de6000d9 2850 info_dict['_filename'] = full_filename = self.prepare_filename(info_dict, warn=True)
2851 temp_filename = self.prepare_filename(info_dict, 'temp')
0202b52a 2852 files_to_move = {}
8222d8de
JMF
2853
2854 # Forced printings
4513a41a 2855 self.__forced_printings(info_dict, full_filename, incomplete=('format' not in info_dict))
8222d8de 2856
b7b04c78 2857 if self.params.get('simulate'):
9e907ebd 2858 info_dict['__write_download_archive'] = self.params.get('force_write_download_archive')
8222d8de
JMF
2859 return
2860
de6000d9 2861 if full_filename is None:
8222d8de 2862 return
e92caff5 2863 if not self._ensure_dir_exists(encodeFilename(full_filename)):
0202b52a 2864 return
e92caff5 2865 if not self._ensure_dir_exists(encodeFilename(temp_filename)):
8222d8de
JMF
2866 return
2867
80c03fa9 2868 if self._write_description('video', info_dict,
2869 self.prepare_filename(info_dict, 'description')) is None:
2870 return
2871
2872 sub_files = self._write_subtitles(info_dict, temp_filename)
2873 if sub_files is None:
2874 return
2875 files_to_move.update(dict(sub_files))
2876
2877 thumb_files = self._write_thumbnails(
2878 'video', info_dict, temp_filename, self.prepare_filename(info_dict, 'thumbnail'))
2879 if thumb_files is None:
2880 return
2881 files_to_move.update(dict(thumb_files))
8222d8de 2882
80c03fa9 2883 infofn = self.prepare_filename(info_dict, 'infojson')
2884 _infojson_written = self._write_info_json('video', info_dict, infofn)
2885 if _infojson_written:
dac5df5a 2886 info_dict['infojson_filename'] = infofn
e75bb0d6 2887 # For backward compatibility, even though it was a private field
80c03fa9 2888 info_dict['__infojson_filename'] = infofn
2889 elif _infojson_written is None:
2890 return
2891
2892 # Note: Annotations are deprecated
2893 annofn = None
1fb07d10 2894 if self.params.get('writeannotations', False):
de6000d9 2895 annofn = self.prepare_filename(info_dict, 'annotation')
80c03fa9 2896 if annofn:
e92caff5 2897 if not self._ensure_dir_exists(encodeFilename(annofn)):
0202b52a 2898 return
0c3d0f51 2899 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(annofn)):
6febd1c1 2900 self.to_screen('[info] Video annotations are already present')
ffddb112
RA
2901 elif not info_dict.get('annotations'):
2902 self.report_warning('There are no annotations to write.')
7b6fefc9
PH
2903 else:
2904 try:
6febd1c1 2905 self.to_screen('[info] Writing video annotations to: ' + annofn)
86e5f3ed 2906 with open(encodeFilename(annofn), 'w', encoding='utf-8') as annofile:
7b6fefc9
PH
2907 annofile.write(info_dict['annotations'])
2908 except (KeyError, TypeError):
6febd1c1 2909 self.report_warning('There are no annotations to write.')
86e5f3ed 2910 except OSError:
6febd1c1 2911 self.report_error('Cannot write annotations file: ' + annofn)
7b6fefc9 2912 return
1fb07d10 2913
732044af 2914 # Write internet shortcut files
08438d2c 2915 def _write_link_file(link_type):
60f3e995 2916 url = try_get(info_dict['webpage_url'], iri_to_uri)
2917 if not url:
2918 self.report_warning(
2919 f'Cannot write internet shortcut file because the actual URL of "{info_dict["webpage_url"]}" is unknown')
2920 return True
08438d2c 2921 linkfn = replace_extension(self.prepare_filename(info_dict, 'link'), link_type, info_dict.get('ext'))
0e6b018a
Z
2922 if not self._ensure_dir_exists(encodeFilename(linkfn)):
2923 return False
10e3742e 2924 if self.params.get('overwrites', True) and os.path.exists(encodeFilename(linkfn)):
08438d2c 2925 self.to_screen(f'[info] Internet shortcut (.{link_type}) is already present')
2926 return True
2927 try:
2928 self.to_screen(f'[info] Writing internet shortcut (.{link_type}) to: {linkfn}')
86e5f3ed 2929 with open(encodeFilename(to_high_limit_path(linkfn)), 'w', encoding='utf-8',
2930 newline='\r\n' if link_type == 'url' else '\n') as linkfile:
60f3e995 2931 template_vars = {'url': url}
08438d2c 2932 if link_type == 'desktop':
2933 template_vars['filename'] = linkfn[:-(len(link_type) + 1)]
2934 linkfile.write(LINK_TEMPLATES[link_type] % template_vars)
86e5f3ed 2935 except OSError:
08438d2c 2936 self.report_error(f'Cannot write internet shortcut {linkfn}')
2937 return False
732044af 2938 return True
2939
08438d2c 2940 write_links = {
2941 'url': self.params.get('writeurllink'),
2942 'webloc': self.params.get('writewebloclink'),
2943 'desktop': self.params.get('writedesktoplink'),
2944 }
2945 if self.params.get('writelink'):
2946 link_type = ('webloc' if sys.platform == 'darwin'
2947 else 'desktop' if sys.platform.startswith('linux')
2948 else 'url')
2949 write_links[link_type] = True
2950
2951 if any(should_write and not _write_link_file(link_type)
2952 for link_type, should_write in write_links.items()):
2953 return
732044af 2954
f46e2f9d 2955 def replace_info_dict(new_info):
2956 nonlocal info_dict
2957 if new_info == info_dict:
2958 return
2959 info_dict.clear()
2960 info_dict.update(new_info)
2961
56d868db 2962 try:
f46e2f9d 2963 new_info, files_to_move = self.pre_process(info_dict, 'before_dl', files_to_move)
2964 replace_info_dict(new_info)
56d868db 2965 except PostProcessingError as err:
2966 self.report_error('Preprocessing: %s' % str(err))
2967 return
2968
a13e6848 2969 if self.params.get('skip_download'):
56d868db 2970 info_dict['filepath'] = temp_filename
2971 info_dict['__finaldir'] = os.path.dirname(os.path.abspath(encodeFilename(full_filename)))
2972 info_dict['__files_to_move'] = files_to_move
f46e2f9d 2973 replace_info_dict(self.run_pp(MoveFilesAfterDownloadPP(self, False), info_dict))
9e907ebd 2974 info_dict['__write_download_archive'] = self.params.get('force_write_download_archive')
56d868db 2975 else:
2976 # Download
b868936c 2977 info_dict.setdefault('__postprocessors', [])
4340deca 2978 try:
0202b52a 2979
e04938ab 2980 def existing_video_file(*filepaths):
6b591b29 2981 ext = info_dict.get('ext')
e04938ab 2982 converted = lambda file: replace_extension(file, self.params.get('final_ext') or ext, ext)
2983 file = self.existing_file(itertools.chain(*zip(map(converted, filepaths), filepaths)),
2984 default_overwrite=False)
2985 if file:
2986 info_dict['ext'] = os.path.splitext(file)[1][1:]
2987 return file
0202b52a 2988
2989 success = True
4340deca 2990 if info_dict.get('requested_formats') is not None:
81cd954a
S
2991
2992 def compatible_formats(formats):
d03cfdce 2993 # TODO: some formats actually allow this (mkv, webm, ogg, mp4), but not all of them.
2994 video_formats = [format for format in formats if format.get('vcodec') != 'none']
2995 audio_formats = [format for format in formats if format.get('acodec') != 'none']
2996 if len(video_formats) > 2 or len(audio_formats) > 2:
2997 return False
2998
81cd954a 2999 # Check extension
86e5f3ed 3000 exts = {format.get('ext') for format in formats}
d03cfdce 3001 COMPATIBLE_EXTS = (
86e5f3ed 3002 {'mp3', 'mp4', 'm4a', 'm4p', 'm4b', 'm4r', 'm4v', 'ismv', 'isma'},
3003 {'webm'},
d03cfdce 3004 )
3005 for ext_sets in COMPATIBLE_EXTS:
3006 if ext_sets.issuperset(exts):
3007 return True
81cd954a
S
3008 # TODO: Check acodec/vcodec
3009 return False
3010
3011 requested_formats = info_dict['requested_formats']
0202b52a 3012 old_ext = info_dict['ext']
4e3b637d 3013 if self.params.get('merge_output_format') is None:
3014 if not compatible_formats(requested_formats):
3015 info_dict['ext'] = 'mkv'
3016 self.report_warning(
3017 'Requested formats are incompatible for merge and will be merged into mkv')
3018 if (info_dict['ext'] == 'webm'
3019 and info_dict.get('thumbnails')
3020 # check with type instead of pp_key, __name__, or isinstance
3021 # since we dont want any custom PPs to trigger this
3022 and any(type(pp) == EmbedThumbnailPP for pp in self._pps['post_process'])):
3023 info_dict['ext'] = 'mkv'
3024 self.report_warning(
3025 'webm doesn\'t support embedding a thumbnail, mkv will be used')
124bc071 3026 new_ext = info_dict['ext']
0202b52a 3027
124bc071 3028 def correct_ext(filename, ext=new_ext):
96fccc10 3029 if filename == '-':
3030 return filename
0202b52a 3031 filename_real_ext = os.path.splitext(filename)[1][1:]
3032 filename_wo_ext = (
3033 os.path.splitext(filename)[0]
124bc071 3034 if filename_real_ext in (old_ext, new_ext)
0202b52a 3035 else filename)
86e5f3ed 3036 return f'{filename_wo_ext}.{ext}'
0202b52a 3037
38c6902b 3038 # Ensure filename always has a correct extension for successful merge
0202b52a 3039 full_filename = correct_ext(full_filename)
3040 temp_filename = correct_ext(temp_filename)
e04938ab 3041 dl_filename = existing_video_file(full_filename, temp_filename)
1ea24129 3042 info_dict['__real_download'] = False
18e674b4 3043
adbc4ec4
THD
3044 downloaded = []
3045 merger = FFmpegMergerPP(self)
3046
3047 fd = get_suitable_downloader(info_dict, self.params, to_stdout=temp_filename == '-')
dbf5416a 3048 if dl_filename is not None:
6c7274ec 3049 self.report_file_already_downloaded(dl_filename)
adbc4ec4
THD
3050 elif fd:
3051 for f in requested_formats if fd != FFmpegFD else []:
3052 f['filepath'] = fname = prepend_extension(
3053 correct_ext(temp_filename, info_dict['ext']),
3054 'f%s' % f['format_id'], info_dict['ext'])
3055 downloaded.append(fname)
dbf5416a 3056 info_dict['url'] = '\n'.join(f['url'] for f in requested_formats)
3057 success, real_download = self.dl(temp_filename, info_dict)
3058 info_dict['__real_download'] = real_download
18e674b4 3059 else:
18e674b4 3060 if self.params.get('allow_unplayable_formats'):
3061 self.report_warning(
3062 'You have requested merging of multiple formats '
3063 'while also allowing unplayable formats to be downloaded. '
3064 'The formats won\'t be merged to prevent data corruption.')
3065 elif not merger.available:
e8969bda 3066 msg = 'You have requested merging of multiple formats but ffmpeg is not installed'
3067 if not self.params.get('ignoreerrors'):
3068 self.report_error(f'{msg}. Aborting due to --abort-on-error')
3069 return
3070 self.report_warning(f'{msg}. The formats won\'t be merged')
18e674b4 3071
96fccc10 3072 if temp_filename == '-':
adbc4ec4 3073 reason = ('using a downloader other than ffmpeg' if FFmpegFD.can_merge_formats(info_dict, self.params)
96fccc10 3074 else 'but the formats are incompatible for simultaneous download' if merger.available
3075 else 'but ffmpeg is not installed')
3076 self.report_warning(
3077 f'You have requested downloading multiple formats to stdout {reason}. '
3078 'The formats will be streamed one after the other')
3079 fname = temp_filename
dbf5416a 3080 for f in requested_formats:
3081 new_info = dict(info_dict)
3082 del new_info['requested_formats']
3083 new_info.update(f)
96fccc10 3084 if temp_filename != '-':
124bc071 3085 fname = prepend_extension(
3086 correct_ext(temp_filename, new_info['ext']),
3087 'f%s' % f['format_id'], new_info['ext'])
96fccc10 3088 if not self._ensure_dir_exists(fname):
3089 return
a21e0ab1 3090 f['filepath'] = fname
96fccc10 3091 downloaded.append(fname)
dbf5416a 3092 partial_success, real_download = self.dl(fname, new_info)
3093 info_dict['__real_download'] = info_dict['__real_download'] or real_download
3094 success = success and partial_success
adbc4ec4
THD
3095
3096 if downloaded and merger.available and not self.params.get('allow_unplayable_formats'):
3097 info_dict['__postprocessors'].append(merger)
3098 info_dict['__files_to_merge'] = downloaded
3099 # Even if there were no downloads, it is being merged only now
3100 info_dict['__real_download'] = True
3101 else:
3102 for file in downloaded:
3103 files_to_move[file] = None
4340deca
P
3104 else:
3105 # Just a single file
e04938ab 3106 dl_filename = existing_video_file(full_filename, temp_filename)
6c7274ec 3107 if dl_filename is None or dl_filename == temp_filename:
3108 # dl_filename == temp_filename could mean that the file was partially downloaded with --no-part.
3109 # So we should try to resume the download
e8e73840 3110 success, real_download = self.dl(temp_filename, info_dict)
0202b52a 3111 info_dict['__real_download'] = real_download
6c7274ec 3112 else:
3113 self.report_file_already_downloaded(dl_filename)
0202b52a 3114
0202b52a 3115 dl_filename = dl_filename or temp_filename
c571435f 3116 info_dict['__finaldir'] = os.path.dirname(os.path.abspath(encodeFilename(full_filename)))
0202b52a 3117
3158150c 3118 except network_exceptions as err:
7960b056 3119 self.report_error('unable to download video data: %s' % error_to_compat_str(err))
4340deca 3120 return
86e5f3ed 3121 except OSError as err:
4340deca
P
3122 raise UnavailableVideoError(err)
3123 except (ContentTooShortError, ) as err:
86e5f3ed 3124 self.report_error(f'content too short (expected {err.expected} bytes and served {err.downloaded})')
4340deca 3125 return
8222d8de 3126
de6000d9 3127 if success and full_filename != '-':
f17f8651 3128
fd7cfb64 3129 def fixup():
3130 do_fixup = True
3131 fixup_policy = self.params.get('fixup')
3132 vid = info_dict['id']
3133
3134 if fixup_policy in ('ignore', 'never'):
3135 return
3136 elif fixup_policy == 'warn':
3137 do_fixup = False
f89b3e2d 3138 elif fixup_policy != 'force':
3139 assert fixup_policy in ('detect_or_warn', None)
3140 if not info_dict.get('__real_download'):
3141 do_fixup = False
fd7cfb64 3142
3143 def ffmpeg_fixup(cndn, msg, cls):
3144 if not cndn:
3145 return
3146 if not do_fixup:
3147 self.report_warning(f'{vid}: {msg}')
3148 return
3149 pp = cls(self)
3150 if pp.available:
3151 info_dict['__postprocessors'].append(pp)
3152 else:
3153 self.report_warning(f'{vid}: {msg}. Install ffmpeg to fix this automatically')
3154
3155 stretched_ratio = info_dict.get('stretched_ratio')
3156 ffmpeg_fixup(
3157 stretched_ratio not in (1, None),
3158 f'Non-uniform pixel ratio {stretched_ratio}',
3159 FFmpegFixupStretchedPP)
3160
3161 ffmpeg_fixup(
3162 (info_dict.get('requested_formats') is None
3163 and info_dict.get('container') == 'm4a_dash'
3164 and info_dict.get('ext') == 'm4a'),
3165 'writing DASH m4a. Only some players support this container',
3166 FFmpegFixupM4aPP)
3167
993191c0 3168 downloader = get_suitable_downloader(info_dict, self.params) if 'protocol' in info_dict else None
3169 downloader = downloader.__name__ if downloader else None
adbc4ec4
THD
3170
3171 if info_dict.get('requested_formats') is None: # Not necessary if doing merger
3172 ffmpeg_fixup(downloader == 'HlsFD',
3173 'Possible MPEG-TS in MP4 container or malformed AAC timestamps',
3174 FFmpegFixupM3u8PP)
3175 ffmpeg_fixup(info_dict.get('is_live') and downloader == 'DashSegmentsFD',
3176 'Possible duplicate MOOV atoms', FFmpegFixupDuplicateMoovPP)
3177
e04b003e 3178 ffmpeg_fixup(downloader == 'WebSocketFragmentFD', 'Malformed timestamps detected', FFmpegFixupTimestampPP)
3179 ffmpeg_fixup(downloader == 'WebSocketFragmentFD', 'Malformed duration detected', FFmpegFixupDurationPP)
fd7cfb64 3180
3181 fixup()
8222d8de 3182 try:
f46e2f9d 3183 replace_info_dict(self.post_process(dl_filename, info_dict, files_to_move))
af819c21 3184 except PostProcessingError as err:
3185 self.report_error('Postprocessing: %s' % str(err))
8222d8de 3186 return
ab8e5e51
AM
3187 try:
3188 for ph in self._post_hooks:
23c1a667 3189 ph(info_dict['filepath'])
ab8e5e51
AM
3190 except Exception as err:
3191 self.report_error('post hooks: %s' % str(err))
3192 return
9e907ebd 3193 info_dict['__write_download_archive'] = True
2d30509f 3194
a13e6848 3195 if self.params.get('force_write_download_archive'):
9e907ebd 3196 info_dict['__write_download_archive'] = True
a13e6848 3197
3198 # Make sure the info_dict was modified in-place
f46e2f9d 3199 assert info_dict is original_infodict
a13e6848 3200
c3e6ffba 3201 max_downloads = self.params.get('max_downloads')
3202 if max_downloads is not None and self._num_downloads >= int(max_downloads):
3203 raise MaxDownloadsReached()
8222d8de 3204
aa9369a2 3205 def __download_wrapper(self, func):
3206 @functools.wraps(func)
3207 def wrapper(*args, **kwargs):
3208 try:
3209 res = func(*args, **kwargs)
3210 except UnavailableVideoError as e:
3211 self.report_error(e)
b222c271 3212 except MaxDownloadsReached as e:
aa9369a2 3213 self.to_screen(f'[info] {e}')
3214 raise
b222c271 3215 except DownloadCancelled as e:
3216 self.to_screen(f'[info] {e}')
3217 if not self.params.get('break_per_url'):
3218 raise
aa9369a2 3219 else:
3220 if self.params.get('dump_single_json', False):
3221 self.post_extract(res)
3222 self.to_stdout(json.dumps(self.sanitize_info(res)))
3223 return wrapper
3224
8222d8de
JMF
3225 def download(self, url_list):
3226 """Download a given list of URLs."""
aa9369a2 3227 url_list = variadic(url_list) # Passing a single URL is a common mistake
de6000d9 3228 outtmpl = self.outtmpl_dict['default']
3089bc74
S
3229 if (len(url_list) > 1
3230 and outtmpl != '-'
3231 and '%' not in outtmpl
3232 and self.params.get('max_downloads') != 1):
acd69589 3233 raise SameFileError(outtmpl)
8222d8de
JMF
3234
3235 for url in url_list:
aa9369a2 3236 self.__download_wrapper(self.extract_info)(
3237 url, force_generic_extractor=self.params.get('force_generic_extractor', False))
8222d8de
JMF
3238
3239 return self._download_retcode
3240
1dcc4c0c 3241 def download_with_info_file(self, info_filename):
31bd3925
JMF
3242 with contextlib.closing(fileinput.FileInput(
3243 [info_filename], mode='r',
3244 openhook=fileinput.hook_encoded('utf-8'))) as f:
3245 # FileInput doesn't have a read method, we can't call json.load
8012d892 3246 info = self.sanitize_info(json.loads('\n'.join(f)), self.params.get('clean_infojson', True))
d4943898 3247 try:
aa9369a2 3248 self.__download_wrapper(self.process_ie_result)(info, download=True)
f2ebc5c7 3249 except (DownloadError, EntryNotInPlaylist, ReExtractInfo) as e:
bf5f605e 3250 if not isinstance(e, EntryNotInPlaylist):
3251 self.to_stderr('\r')
d4943898
JMF
3252 webpage_url = info.get('webpage_url')
3253 if webpage_url is not None:
aa9369a2 3254 self.report_warning(f'The info failed to download: {e}; trying with URL {webpage_url}')
d4943898
JMF
3255 return self.download([webpage_url])
3256 else:
3257 raise
3258 return self._download_retcode
1dcc4c0c 3259
cb202fd2 3260 @staticmethod
8012d892 3261 def sanitize_info(info_dict, remove_private_keys=False):
3262 ''' Sanitize the infodict for converting to json '''
3ad56b42 3263 if info_dict is None:
3264 return info_dict
6e84b215 3265 info_dict.setdefault('epoch', int(time.time()))
6a5a30f9 3266 info_dict.setdefault('_type', 'video')
09b49e1f 3267
8012d892 3268 if remove_private_keys:
09b49e1f 3269 reject = lambda k, v: v is None or (k.startswith('_') and k != '_type') or k in {
f46e2f9d 3270 'requested_downloads', 'requested_formats', 'requested_subtitles', 'requested_entries',
3271 'entries', 'filepath', 'infojson_filename', 'original_url', 'playlist_autonumber',
6e84b215 3272 }
ae8f99e6 3273 else:
09b49e1f 3274 reject = lambda k, v: False
adbc4ec4
THD
3275
3276 def filter_fn(obj):
3277 if isinstance(obj, dict):
3278 return {k: filter_fn(v) for k, v in obj.items() if not reject(k, v)}
3279 elif isinstance(obj, (list, tuple, set, LazyList)):
3280 return list(map(filter_fn, obj))
3281 elif obj is None or isinstance(obj, (str, int, float, bool)):
3282 return obj
3283 else:
3284 return repr(obj)
3285
5226731e 3286 return filter_fn(info_dict)
cb202fd2 3287
8012d892 3288 @staticmethod
3289 def filter_requested_info(info_dict, actually_filter=True):
3290 ''' Alias of sanitize_info for backward compatibility '''
3291 return YoutubeDL.sanitize_info(info_dict, actually_filter)
3292
ed5835b4 3293 @staticmethod
3294 def post_extract(info_dict):
3295 def actual_post_extract(info_dict):
3296 if info_dict.get('_type') in ('playlist', 'multi_video'):
3297 for video_dict in info_dict.get('entries', {}):
3298 actual_post_extract(video_dict or {})
3299 return
3300
09b49e1f 3301 post_extractor = info_dict.pop('__post_extractor', None) or (lambda: {})
3302 info_dict.update(post_extractor())
ed5835b4 3303
3304 actual_post_extract(info_dict or {})
3305
dcf64d43 3306 def run_pp(self, pp, infodict):
5bfa4862 3307 files_to_delete = []
dcf64d43 3308 if '__files_to_move' not in infodict:
3309 infodict['__files_to_move'] = {}
b1940459 3310 try:
3311 files_to_delete, infodict = pp.run(infodict)
3312 except PostProcessingError as e:
3313 # Must be True and not 'only_download'
3314 if self.params.get('ignoreerrors') is True:
3315 self.report_error(e)
3316 return infodict
3317 raise
3318
5bfa4862 3319 if not files_to_delete:
dcf64d43 3320 return infodict
5bfa4862 3321 if self.params.get('keepvideo', False):
3322 for f in files_to_delete:
dcf64d43 3323 infodict['__files_to_move'].setdefault(f, '')
5bfa4862 3324 else:
3325 for old_filename in set(files_to_delete):
3326 self.to_screen('Deleting original file %s (pass -k to keep)' % old_filename)
3327 try:
3328 os.remove(encodeFilename(old_filename))
86e5f3ed 3329 except OSError:
5bfa4862 3330 self.report_warning('Unable to remove downloaded original file')
dcf64d43 3331 if old_filename in infodict['__files_to_move']:
3332 del infodict['__files_to_move'][old_filename]
3333 return infodict
5bfa4862 3334
ed5835b4 3335 def run_all_pps(self, key, info, *, additional_pps=None):
bb66c247 3336 self._forceprint(key, info)
ed5835b4 3337 for pp in (additional_pps or []) + self._pps[key]:
dc5f409c 3338 info = self.run_pp(pp, info)
ed5835b4 3339 return info
277d6ff5 3340
56d868db 3341 def pre_process(self, ie_info, key='pre_process', files_to_move=None):
5bfa4862 3342 info = dict(ie_info)
56d868db 3343 info['__files_to_move'] = files_to_move or {}
ed5835b4 3344 info = self.run_all_pps(key, info)
56d868db 3345 return info, info.pop('__files_to_move', None)
5bfa4862 3346
f46e2f9d 3347 def post_process(self, filename, info, files_to_move=None):
8222d8de 3348 """Run all the postprocessors on the given file."""
8222d8de 3349 info['filepath'] = filename
dcf64d43 3350 info['__files_to_move'] = files_to_move or {}
ed5835b4 3351 info = self.run_all_pps('post_process', info, additional_pps=info.get('__postprocessors'))
dcf64d43 3352 info = self.run_pp(MoveFilesAfterDownloadPP(self), info)
3353 del info['__files_to_move']
ed5835b4 3354 return self.run_all_pps('after_move', info)
c1c9a79c 3355
5db07df6 3356 def _make_archive_id(self, info_dict):
e9fef7ee
S
3357 video_id = info_dict.get('id')
3358 if not video_id:
3359 return
5db07df6
PH
3360 # Future-proof against any change in case
3361 # and backwards compatibility with prior versions
e9fef7ee 3362 extractor = info_dict.get('extractor_key') or info_dict.get('ie_key') # key in a playlist
7012b23c 3363 if extractor is None:
1211bb6d
S
3364 url = str_or_none(info_dict.get('url'))
3365 if not url:
3366 return
e9fef7ee 3367 # Try to find matching extractor for the URL and take its ie_key
8b7491c8 3368 for ie_key, ie in self._ies.items():
1211bb6d 3369 if ie.suitable(url):
8b7491c8 3370 extractor = ie_key
e9fef7ee
S
3371 break
3372 else:
3373 return
86e5f3ed 3374 return f'{extractor.lower()} {video_id}'
5db07df6
PH
3375
3376 def in_download_archive(self, info_dict):
3377 fn = self.params.get('download_archive')
3378 if fn is None:
3379 return False
3380
3381 vid_id = self._make_archive_id(info_dict)
e9fef7ee 3382 if not vid_id:
7012b23c 3383 return False # Incomplete video information
5db07df6 3384
a45e8619 3385 return vid_id in self.archive
c1c9a79c
PH
3386
3387 def record_download_archive(self, info_dict):
3388 fn = self.params.get('download_archive')
3389 if fn is None:
3390 return
5db07df6
PH
3391 vid_id = self._make_archive_id(info_dict)
3392 assert vid_id
a13e6848 3393 self.write_debug(f'Adding to archive: {vid_id}')
c1c9a79c 3394 with locked_file(fn, 'a', encoding='utf-8') as archive_file:
6febd1c1 3395 archive_file.write(vid_id + '\n')
a45e8619 3396 self.archive.add(vid_id)
dd82ffea 3397
8c51aa65 3398 @staticmethod
8abeeb94 3399 def format_resolution(format, default='unknown'):
9359f3d4 3400 if format.get('vcodec') == 'none' and format.get('acodec') != 'none':
fb04e403 3401 return 'audio only'
f49d89ee
PH
3402 if format.get('resolution') is not None:
3403 return format['resolution']
35615307 3404 if format.get('width') and format.get('height'):
ff51ed58 3405 return '%dx%d' % (format['width'], format['height'])
35615307 3406 elif format.get('height'):
ff51ed58 3407 return '%sp' % format['height']
35615307 3408 elif format.get('width'):
ff51ed58 3409 return '%dx?' % format['width']
3410 return default
8c51aa65 3411
8130779d 3412 def _list_format_headers(self, *headers):
3413 if self.params.get('listformats_table', True) is not False:
3414 return [self._format_screen(header, self.Styles.HEADERS) for header in headers]
3415 return headers
3416
c57f7757
PH
3417 def _format_note(self, fdict):
3418 res = ''
3419 if fdict.get('ext') in ['f4f', 'f4m']:
f304da8a 3420 res += '(unsupported)'
32f90364
PH
3421 if fdict.get('language'):
3422 if res:
3423 res += ' '
f304da8a 3424 res += '[%s]' % fdict['language']
c57f7757 3425 if fdict.get('format_note') is not None:
f304da8a 3426 if res:
3427 res += ' '
3428 res += fdict['format_note']
c57f7757 3429 if fdict.get('tbr') is not None:
f304da8a 3430 if res:
3431 res += ', '
3432 res += '%4dk' % fdict['tbr']
c57f7757
PH
3433 if fdict.get('container') is not None:
3434 if res:
3435 res += ', '
3436 res += '%s container' % fdict['container']
3089bc74
S
3437 if (fdict.get('vcodec') is not None
3438 and fdict.get('vcodec') != 'none'):
c57f7757
PH
3439 if res:
3440 res += ', '
3441 res += fdict['vcodec']
91c7271a 3442 if fdict.get('vbr') is not None:
c57f7757
PH
3443 res += '@'
3444 elif fdict.get('vbr') is not None and fdict.get('abr') is not None:
3445 res += 'video@'
3446 if fdict.get('vbr') is not None:
3447 res += '%4dk' % fdict['vbr']
fbb21cf5 3448 if fdict.get('fps') is not None:
5d583bdf
S
3449 if res:
3450 res += ', '
3451 res += '%sfps' % fdict['fps']
c57f7757
PH
3452 if fdict.get('acodec') is not None:
3453 if res:
3454 res += ', '
3455 if fdict['acodec'] == 'none':
3456 res += 'video only'
3457 else:
3458 res += '%-5s' % fdict['acodec']
3459 elif fdict.get('abr') is not None:
3460 if res:
3461 res += ', '
3462 res += 'audio'
3463 if fdict.get('abr') is not None:
3464 res += '@%3dk' % fdict['abr']
3465 if fdict.get('asr') is not None:
3466 res += ' (%5dHz)' % fdict['asr']
3467 if fdict.get('filesize') is not None:
3468 if res:
3469 res += ', '
3470 res += format_bytes(fdict['filesize'])
9732d77e
PH
3471 elif fdict.get('filesize_approx') is not None:
3472 if res:
3473 res += ', '
3474 res += '~' + format_bytes(fdict['filesize_approx'])
c57f7757 3475 return res
91c7271a 3476
8130779d 3477 def render_formats_table(self, info_dict):
b69fd25c 3478 if not info_dict.get('formats') and not info_dict.get('url'):
8130779d 3479 return None
b69fd25c 3480
94badb25 3481 formats = info_dict.get('formats', [info_dict])
8130779d 3482 if not self.params.get('listformats_table', True) is not False:
76d321f6 3483 table = [
3484 [
3485 format_field(f, 'format_id'),
3486 format_field(f, 'ext'),
3487 self.format_resolution(f),
8130779d 3488 self._format_note(f)
3489 ] for f in formats if f.get('preference') is None or f['preference'] >= -1000]
3490 return render_table(['format code', 'extension', 'resolution', 'note'], table, extra_gap=1)
3491
3492 delim = self._format_screen('\u2502', self.Styles.DELIM, '|', test_encoding=True)
3493 table = [
3494 [
3495 self._format_screen(format_field(f, 'format_id'), self.Styles.ID),
3496 format_field(f, 'ext'),
3497 format_field(f, func=self.format_resolution, ignore=('audio only', 'images')),
3498 format_field(f, 'fps', '\t%d'),
3499 format_field(f, 'dynamic_range', '%s', ignore=(None, 'SDR')).replace('HDR', ''),
3500 delim,
3501 format_field(f, 'filesize', ' \t%s', func=format_bytes) + format_field(f, 'filesize_approx', '~\t%s', func=format_bytes),
3502 format_field(f, 'tbr', '\t%dk'),
3503 shorten_protocol_name(f.get('protocol', '')),
3504 delim,
3505 format_field(f, 'vcodec', default='unknown').replace(
3506 'none', 'images' if f.get('acodec') == 'none'
3507 else self._format_screen('audio only', self.Styles.SUPPRESS)),
3508 format_field(f, 'vbr', '\t%dk'),
3509 format_field(f, 'acodec', default='unknown').replace(
3510 'none', '' if f.get('vcodec') == 'none'
3511 else self._format_screen('video only', self.Styles.SUPPRESS)),
3512 format_field(f, 'abr', '\t%dk'),
3513 format_field(f, 'asr', '\t%dHz'),
3514 join_nonempty(
3515 self._format_screen('UNSUPPORTED', 'light red') if f.get('ext') in ('f4f', 'f4m') else None,
3516 format_field(f, 'language', '[%s]'),
3517 join_nonempty(format_field(f, 'format_note'),
3518 format_field(f, 'container', ignore=(None, f.get('ext'))),
3519 delim=', '),
3520 delim=' '),
3521 ] for f in formats if f.get('preference') is None or f['preference'] >= -1000]
3522 header_line = self._list_format_headers(
3523 'ID', 'EXT', 'RESOLUTION', '\tFPS', 'HDR', delim, '\tFILESIZE', '\tTBR', 'PROTO',
3524 delim, 'VCODEC', '\tVBR', 'ACODEC', '\tABR', '\tASR', 'MORE INFO')
3525
3526 return render_table(
3527 header_line, table, hide_empty=True,
3528 delim=self._format_screen('\u2500', self.Styles.DELIM, '-', test_encoding=True))
3529
3530 def render_thumbnails_table(self, info_dict):
88f23a18 3531 thumbnails = list(info_dict.get('thumbnails') or [])
cfb56d1a 3532 if not thumbnails:
8130779d 3533 return None
3534 return render_table(
ec11a9f4 3535 self._list_format_headers('ID', 'Width', 'Height', 'URL'),
6970b600 3536 [[t.get('id'), t.get('width', 'unknown'), t.get('height', 'unknown'), t['url']] for t in thumbnails])
2412044c 3537
8130779d 3538 def render_subtitles_table(self, video_id, subtitles):
2412044c 3539 def _row(lang, formats):
49c258e1 3540 exts, names = zip(*((f['ext'], f.get('name') or 'unknown') for f in reversed(formats)))
2412044c 3541 if len(set(names)) == 1:
7aee40c1 3542 names = [] if names[0] == 'unknown' else names[:1]
2412044c 3543 return [lang, ', '.join(names), ', '.join(exts)]
3544
8130779d 3545 if not subtitles:
3546 return None
3547 return render_table(
ec11a9f4 3548 self._list_format_headers('Language', 'Name', 'Formats'),
2412044c 3549 [_row(lang, formats) for lang, formats in subtitles.items()],
8130779d 3550 hide_empty=True)
3551
3552 def __list_table(self, video_id, name, func, *args):
3553 table = func(*args)
3554 if not table:
3555 self.to_screen(f'{video_id} has no {name}')
3556 return
3557 self.to_screen(f'[info] Available {name} for {video_id}:')
3558 self.to_stdout(table)
3559
3560 def list_formats(self, info_dict):
3561 self.__list_table(info_dict['id'], 'formats', self.render_formats_table, info_dict)
3562
3563 def list_thumbnails(self, info_dict):
3564 self.__list_table(info_dict['id'], 'thumbnails', self.render_thumbnails_table, info_dict)
3565
3566 def list_subtitles(self, video_id, subtitles, name='subtitles'):
3567 self.__list_table(video_id, name, self.render_subtitles_table, video_id, subtitles)
a504ced0 3568
dca08720
PH
3569 def urlopen(self, req):
3570 """ Start an HTTP download """
f9934b96 3571 if isinstance(req, str):
67dda517 3572 req = sanitized_Request(req)
19a41fc6 3573 return self._opener.open(req, timeout=self._socket_timeout)
dca08720
PH
3574
3575 def print_debug_header(self):
3576 if not self.params.get('verbose'):
3577 return
49a57e70 3578
3579 def get_encoding(stream):
2a938746 3580 ret = str(getattr(stream, 'encoding', 'missing (%s)' % type(stream).__name__))
49a57e70 3581 if not supports_terminal_sequences(stream):
e3c7d495 3582 from .compat import WINDOWS_VT_MODE
3583 ret += ' (No VT)' if WINDOWS_VT_MODE is False else ' (No ANSI)'
49a57e70 3584 return ret
3585
3586 encoding_str = 'Encodings: locale %s, fs %s, out %s, err %s, pref %s' % (
3587 locale.getpreferredencoding(),
3588 sys.getfilesystemencoding(),
cf4f42cb 3589 get_encoding(self._out_files['screen']), get_encoding(self._out_files['error']),
49a57e70 3590 self.get_encoding())
883d4b1e 3591
3592 logger = self.params.get('logger')
3593 if logger:
3594 write_debug = lambda msg: logger.debug(f'[debug] {msg}')
3595 write_debug(encoding_str)
3596 else:
96565c7e 3597 write_string(f'[debug] {encoding_str}\n', encoding=None)
49a57e70 3598 write_debug = lambda msg: self._write_string(f'[debug] {msg}\n')
734f90bb 3599
4c88ff87 3600 source = detect_variant()
36eaf303 3601 write_debug(join_nonempty(
3602 'yt-dlp version', __version__,
3603 f'[{RELEASE_GIT_HEAD}]' if RELEASE_GIT_HEAD else '',
3604 '' if source == 'unknown' else f'({source})',
3605 delim=' '))
6e21fdd2 3606 if not _LAZY_LOADER:
3607 if os.environ.get('YTDLP_NO_LAZY_EXTRACTORS'):
49a57e70 3608 write_debug('Lazy loading extractors is forcibly disabled')
6e21fdd2 3609 else:
49a57e70 3610 write_debug('Lazy loading extractors is disabled')
3ae5e797 3611 if plugin_extractors or plugin_postprocessors:
49a57e70 3612 write_debug('Plugins: %s' % [
3ae5e797 3613 '%s%s' % (klass.__name__, '' if klass.__name__ == name else f' as {name}')
3614 for name, klass in itertools.chain(plugin_extractors.items(), plugin_postprocessors.items())])
53ed7066 3615 if self.params.get('compat_opts'):
49a57e70 3616 write_debug('Compatibility options: %s' % ', '.join(self.params.get('compat_opts')))
36eaf303 3617
3618 if source == 'source':
dca08720 3619 try:
36eaf303 3620 sp = Popen(
3621 ['git', 'rev-parse', '--short', 'HEAD'],
3622 stdout=subprocess.PIPE, stderr=subprocess.PIPE,
3623 cwd=os.path.dirname(os.path.abspath(__file__)))
3624 out, err = sp.communicate_or_kill()
3625 out = out.decode().strip()
3626 if re.match('[0-9a-f]+', out):
3627 write_debug('Git HEAD: %s' % out)
70a1165b 3628 except Exception:
19a03940 3629 with contextlib.suppress(Exception):
36eaf303 3630 sys.exc_clear()
b300cda4
S
3631
3632 def python_implementation():
3633 impl_name = platform.python_implementation()
3634 if impl_name == 'PyPy' and hasattr(sys, 'pypy_version_info'):
3635 return impl_name + ' version %d.%d.%d' % sys.pypy_version_info[:3]
3636 return impl_name
3637
49a57e70 3638 write_debug('Python version %s (%s %s) - %s' % (
e5813e53 3639 platform.python_version(),
3640 python_implementation(),
3641 platform.architecture()[0],
b300cda4 3642 platform_name()))
d28b5171 3643
8913ef74 3644 exe_versions, ffmpeg_features = FFmpegPostProcessor.get_versions_and_features(self)
3645 ffmpeg_features = {key for key, val in ffmpeg_features.items() if val}
3646 if ffmpeg_features:
19a03940 3647 exe_versions['ffmpeg'] += ' (%s)' % ','.join(sorted(ffmpeg_features))
8913ef74 3648
4c83c967 3649 exe_versions['rtmpdump'] = rtmpdump_version()
feee8d32 3650 exe_versions['phantomjs'] = PhantomJSwrapper._version()
d28b5171 3651 exe_str = ', '.join(
2831b468 3652 f'{exe} {v}' for exe, v in sorted(exe_versions.items()) if v
3653 ) or 'none'
49a57e70 3654 write_debug('exe versions: %s' % exe_str)
dca08720 3655
9b8ee23b 3656 from .dependencies import available_dependencies
3657
3658 write_debug('Optional libraries: %s' % (', '.join(sorted({
3659 module.__name__.split('.')[0] for module in available_dependencies.values()
3660 })) or 'none'))
2831b468 3661
97ec5bc5 3662 self._setup_opener()
dca08720
PH
3663 proxy_map = {}
3664 for handler in self._opener.handlers:
3665 if hasattr(handler, 'proxies'):
3666 proxy_map.update(handler.proxies)
49a57e70 3667 write_debug(f'Proxy map: {proxy_map}')
dca08720 3668
49a57e70 3669 # Not implemented
3670 if False and self.params.get('call_home'):
58b1f00d 3671 ipaddr = self.urlopen('https://yt-dl.org/ip').read().decode('utf-8')
49a57e70 3672 write_debug('Public IP address: %s' % ipaddr)
58b1f00d
PH
3673 latest_version = self.urlopen(
3674 'https://yt-dl.org/latest/version').read().decode('utf-8')
3675 if version_tuple(latest_version) > version_tuple(__version__):
3676 self.report_warning(
3677 'You are using an outdated version (newest version: %s)! '
3678 'See https://yt-dl.org/update if you need help updating.' %
3679 latest_version)
3680
e344693b 3681 def _setup_opener(self):
97ec5bc5 3682 if hasattr(self, '_opener'):
3683 return
6ad14cab 3684 timeout_val = self.params.get('socket_timeout')
17bddf3e 3685 self._socket_timeout = 20 if timeout_val is None else float(timeout_val)
6ad14cab 3686
982ee69a 3687 opts_cookiesfrombrowser = self.params.get('cookiesfrombrowser')
dca08720
PH
3688 opts_cookiefile = self.params.get('cookiefile')
3689 opts_proxy = self.params.get('proxy')
3690
982ee69a 3691 self.cookiejar = load_cookies(opts_cookiefile, opts_cookiesfrombrowser, self)
dca08720 3692
6a3f4c3f 3693 cookie_processor = YoutubeDLCookieProcessor(self.cookiejar)
dca08720
PH
3694 if opts_proxy is not None:
3695 if opts_proxy == '':
3696 proxies = {}
3697 else:
3698 proxies = {'http': opts_proxy, 'https': opts_proxy}
3699 else:
3700 proxies = compat_urllib_request.getproxies()
067aa17e 3701 # Set HTTPS proxy to HTTP one if given (https://github.com/ytdl-org/youtube-dl/issues/805)
dca08720
PH
3702 if 'http' in proxies and 'https' not in proxies:
3703 proxies['https'] = proxies['http']
91410c9b 3704 proxy_handler = PerRequestProxyHandler(proxies)
a0ddb8a2
PH
3705
3706 debuglevel = 1 if self.params.get('debug_printtraffic') else 0
be4a824d
PH
3707 https_handler = make_HTTPS_handler(self.params, debuglevel=debuglevel)
3708 ydlh = YoutubeDLHandler(self.params, debuglevel=debuglevel)
fca6dba8 3709 redirect_handler = YoutubeDLRedirectHandler()
f9934b96 3710 data_handler = urllib.request.DataHandler()
6240b0a2
JMF
3711
3712 # When passing our own FileHandler instance, build_opener won't add the
3713 # default FileHandler and allows us to disable the file protocol, which
3714 # can be used for malicious purposes (see
067aa17e 3715 # https://github.com/ytdl-org/youtube-dl/issues/8227)
6240b0a2
JMF
3716 file_handler = compat_urllib_request.FileHandler()
3717
3718 def file_open(*args, **kwargs):
7a5c1cfe 3719 raise compat_urllib_error.URLError('file:// scheme is explicitly disabled in yt-dlp for security reasons')
6240b0a2
JMF
3720 file_handler.file_open = file_open
3721
3722 opener = compat_urllib_request.build_opener(
fca6dba8 3723 proxy_handler, https_handler, cookie_processor, ydlh, redirect_handler, data_handler, file_handler)
2461f79d 3724
dca08720
PH
3725 # Delete the default user-agent header, which would otherwise apply in
3726 # cases where our custom HTTP handler doesn't come into play
067aa17e 3727 # (See https://github.com/ytdl-org/youtube-dl/issues/1309 for details)
dca08720
PH
3728 opener.addheaders = []
3729 self._opener = opener
62fec3b2
PH
3730
3731 def encode(self, s):
3732 if isinstance(s, bytes):
3733 return s # Already encoded
3734
3735 try:
3736 return s.encode(self.get_encoding())
3737 except UnicodeEncodeError as err:
3738 err.reason = err.reason + '. Check your system encoding configuration or use the --encoding option.'
3739 raise
3740
3741 def get_encoding(self):
3742 encoding = self.params.get('encoding')
3743 if encoding is None:
3744 encoding = preferredencoding()
3745 return encoding
ec82d85a 3746
e08a85d8 3747 def _write_info_json(self, label, ie_result, infofn, overwrite=None):
cb96c5be 3748 ''' Write infojson and returns True = written, 'exists' = Already exists, False = skip, None = error '''
e08a85d8 3749 if overwrite is None:
3750 overwrite = self.params.get('overwrites', True)
80c03fa9 3751 if not self.params.get('writeinfojson'):
3752 return False
3753 elif not infofn:
3754 self.write_debug(f'Skipping writing {label} infojson')
3755 return False
3756 elif not self._ensure_dir_exists(infofn):
3757 return None
e08a85d8 3758 elif not overwrite and os.path.exists(infofn):
80c03fa9 3759 self.to_screen(f'[info] {label.title()} metadata is already present')
cb96c5be 3760 return 'exists'
3761
3762 self.to_screen(f'[info] Writing {label} metadata as JSON to: {infofn}')
3763 try:
3764 write_json_file(self.sanitize_info(ie_result, self.params.get('clean_infojson', True)), infofn)
3765 return True
86e5f3ed 3766 except OSError:
cb96c5be 3767 self.report_error(f'Cannot write {label} metadata to JSON file {infofn}')
3768 return None
80c03fa9 3769
3770 def _write_description(self, label, ie_result, descfn):
3771 ''' Write description and returns True = written, False = skip, None = error '''
3772 if not self.params.get('writedescription'):
3773 return False
3774 elif not descfn:
3775 self.write_debug(f'Skipping writing {label} description')
3776 return False
3777 elif not self._ensure_dir_exists(descfn):
3778 return None
3779 elif not self.params.get('overwrites', True) and os.path.exists(descfn):
3780 self.to_screen(f'[info] {label.title()} description is already present')
3781 elif ie_result.get('description') is None:
3782 self.report_warning(f'There\'s no {label} description to write')
3783 return False
3784 else:
3785 try:
3786 self.to_screen(f'[info] Writing {label} description to: {descfn}')
86e5f3ed 3787 with open(encodeFilename(descfn), 'w', encoding='utf-8') as descfile:
80c03fa9 3788 descfile.write(ie_result['description'])
86e5f3ed 3789 except OSError:
80c03fa9 3790 self.report_error(f'Cannot write {label} description file {descfn}')
3791 return None
3792 return True
3793
3794 def _write_subtitles(self, info_dict, filename):
3795 ''' Write subtitles to file and return list of (sub_filename, final_sub_filename); or None if error'''
3796 ret = []
3797 subtitles = info_dict.get('requested_subtitles')
3798 if not subtitles or not (self.params.get('writesubtitles') or self.params.get('writeautomaticsub')):
3799 # subtitles download errors are already managed as troubles in relevant IE
3800 # that way it will silently go on when used with unsupporting IE
3801 return ret
3802
3803 sub_filename_base = self.prepare_filename(info_dict, 'subtitle')
3804 if not sub_filename_base:
3805 self.to_screen('[info] Skipping writing video subtitles')
3806 return ret
3807 for sub_lang, sub_info in subtitles.items():
3808 sub_format = sub_info['ext']
3809 sub_filename = subtitles_filename(filename, sub_lang, sub_format, info_dict.get('ext'))
3810 sub_filename_final = subtitles_filename(sub_filename_base, sub_lang, sub_format, info_dict.get('ext'))
e04938ab 3811 existing_sub = self.existing_file((sub_filename_final, sub_filename))
3812 if existing_sub:
80c03fa9 3813 self.to_screen(f'[info] Video subtitle {sub_lang}.{sub_format} is already present')
e04938ab 3814 sub_info['filepath'] = existing_sub
3815 ret.append((existing_sub, sub_filename_final))
80c03fa9 3816 continue
3817
3818 self.to_screen(f'[info] Writing video subtitles to: {sub_filename}')
3819 if sub_info.get('data') is not None:
3820 try:
3821 # Use newline='' to prevent conversion of newline characters
3822 # See https://github.com/ytdl-org/youtube-dl/issues/10268
86e5f3ed 3823 with open(sub_filename, 'w', encoding='utf-8', newline='') as subfile:
80c03fa9 3824 subfile.write(sub_info['data'])
3825 sub_info['filepath'] = sub_filename
3826 ret.append((sub_filename, sub_filename_final))
3827 continue
86e5f3ed 3828 except OSError:
80c03fa9 3829 self.report_error(f'Cannot write video subtitles file {sub_filename}')
3830 return None
3831
3832 try:
3833 sub_copy = sub_info.copy()
3834 sub_copy.setdefault('http_headers', info_dict.get('http_headers'))
3835 self.dl(sub_filename, sub_copy, subtitle=True)
3836 sub_info['filepath'] = sub_filename
3837 ret.append((sub_filename, sub_filename_final))
6020e05d 3838 except (DownloadError, ExtractorError, IOError, OSError, ValueError) + network_exceptions as err:
c70c418d 3839 msg = f'Unable to download video subtitles for {sub_lang!r}: {err}'
6020e05d 3840 if self.params.get('ignoreerrors') is not True: # False or 'only_download'
c70c418d 3841 if not self.params.get('ignoreerrors'):
3842 self.report_error(msg)
3843 raise DownloadError(msg)
3844 self.report_warning(msg)
519804a9 3845 return ret
80c03fa9 3846
3847 def _write_thumbnails(self, label, info_dict, filename, thumb_filename_base=None):
3848 ''' Write thumbnails to file and return list of (thumb_filename, final_thumb_filename) '''
6c4fd172 3849 write_all = self.params.get('write_all_thumbnails', False)
80c03fa9 3850 thumbnails, ret = [], []
6c4fd172 3851 if write_all or self.params.get('writethumbnail', False):
0202b52a 3852 thumbnails = info_dict.get('thumbnails') or []
6c4fd172 3853 multiple = write_all and len(thumbnails) > 1
ec82d85a 3854
80c03fa9 3855 if thumb_filename_base is None:
3856 thumb_filename_base = filename
3857 if thumbnails and not thumb_filename_base:
3858 self.write_debug(f'Skipping writing {label} thumbnail')
3859 return ret
3860
dd0228ce 3861 for idx, t in list(enumerate(thumbnails))[::-1]:
80c03fa9 3862 thumb_ext = (f'{t["id"]}.' if multiple else '') + determine_ext(t['url'], 'jpg')
aa9369a2 3863 thumb_display_id = f'{label} thumbnail {t["id"]}'
80c03fa9 3864 thumb_filename = replace_extension(filename, thumb_ext, info_dict.get('ext'))
3865 thumb_filename_final = replace_extension(thumb_filename_base, thumb_ext, info_dict.get('ext'))
ec82d85a 3866
e04938ab 3867 existing_thumb = self.existing_file((thumb_filename_final, thumb_filename))
3868 if existing_thumb:
aa9369a2 3869 self.to_screen('[info] %s is already present' % (
3870 thumb_display_id if multiple else f'{label} thumbnail').capitalize())
e04938ab 3871 t['filepath'] = existing_thumb
3872 ret.append((existing_thumb, thumb_filename_final))
ec82d85a 3873 else:
80c03fa9 3874 self.to_screen(f'[info] Downloading {thumb_display_id} ...')
ec82d85a 3875 try:
297e9952 3876 uf = self.urlopen(sanitized_Request(t['url'], headers=t.get('http_headers', {})))
80c03fa9 3877 self.to_screen(f'[info] Writing {thumb_display_id} to: {thumb_filename}')
d3d89c32 3878 with open(encodeFilename(thumb_filename), 'wb') as thumbf:
ec82d85a 3879 shutil.copyfileobj(uf, thumbf)
80c03fa9 3880 ret.append((thumb_filename, thumb_filename_final))
885cc0b7 3881 t['filepath'] = thumb_filename
3158150c 3882 except network_exceptions as err:
dd0228ce 3883 thumbnails.pop(idx)
80c03fa9 3884 self.report_warning(f'Unable to download {thumb_display_id}: {err}')
6c4fd172 3885 if ret and not write_all:
3886 break
0202b52a 3887 return ret