]> jfr.im git - yt-dlp.git/blame - yt_dlp/YoutubeDL.py
[adobepass] Add Suddenlink MSO (#2977)
[yt-dlp.git] / yt_dlp / YoutubeDL.py
CommitLineData
cc52de43 1#!/usr/bin/env python3
dcdb292f 2# coding: utf-8
8222d8de 3
6febd1c1 4from __future__ import absolute_import, unicode_literals
8222d8de 5
26e63931 6import collections
31bd3925 7import contextlib
9d2ecdbc 8import datetime
c1c9a79c 9import errno
31bd3925 10import fileinput
b5ae35ee 11import functools
8222d8de 12import io
b82f815f 13import itertools
8694c600 14import json
62fec3b2 15import locale
083c9df9 16import operator
8222d8de 17import os
dca08720 18import platform
8222d8de
JMF
19import re
20import shutil
dca08720 21import subprocess
8222d8de 22import sys
21cd8fae 23import tempfile
8222d8de 24import time
67134eab 25import tokenize
8222d8de 26import traceback
75822ca7 27import random
524e2e4f 28import unicodedata
8222d8de 29
ec11a9f4 30from enum import Enum
961ea474
S
31from string import ascii_letters
32
8c25f81b 33from .compat import (
82d8a8b6 34 compat_basestring,
003c69a8 35 compat_get_terminal_size,
4f026faf 36 compat_kwargs,
d0d9ade4 37 compat_numeric_types,
e9c0cdd3 38 compat_os_name,
edf65256 39 compat_pycrypto_AES,
7d1eb38a 40 compat_shlex_quote,
ce02ed60 41 compat_str,
67134eab 42 compat_tokenize_tokenize,
ce02ed60
PH
43 compat_urllib_error,
44 compat_urllib_request,
8b172c2e 45 compat_urllib_request_DataHandler,
819e0531 46 windows_enable_vt_mode,
8c25f81b 47)
982ee69a 48from .cookies import load_cookies
8c25f81b 49from .utils import (
eedb7ba5
S
50 age_restricted,
51 args_to_str,
ce02ed60
PH
52 ContentTooShortError,
53 date_from_str,
54 DateRange,
acd69589 55 DEFAULT_OUTTMPL,
ce02ed60 56 determine_ext,
b5559424 57 determine_protocol,
48f79687 58 DownloadCancelled,
ce02ed60 59 DownloadError,
c0384f22 60 encode_compat_str,
ce02ed60 61 encodeFilename,
498f5606 62 EntryNotInPlaylist,
a06916d9 63 error_to_compat_str,
8b0d7497 64 ExistingVideoReached,
590bc6f6 65 expand_path,
ce02ed60 66 ExtractorError,
e29663c6 67 float_or_none,
02dbf93f 68 format_bytes,
76d321f6 69 format_field,
e0fd9573 70 format_decimal_suffix,
525ef922 71 formatSeconds,
773f291d 72 GeoRestrictedError,
0bb322b9 73 get_domain,
b0249bca 74 HEADRequest,
d37707bd 75 InAdvancePagedList,
c9969434 76 int_or_none,
732044af 77 iri_to_uri,
773f291d 78 ISO3166Utils,
34921b43 79 join_nonempty,
56a8fb4f 80 LazyList,
08438d2c 81 LINK_TEMPLATES,
ce02ed60 82 locked_file,
0202b52a 83 make_dir,
dca08720 84 make_HTTPS_handler,
ce02ed60 85 MaxDownloadsReached,
8b7539d2 86 merge_headers,
3158150c 87 network_exceptions,
ec11a9f4 88 number_of_digits,
cd6fc19e 89 orderedSet,
a06916d9 90 OUTTMPL_TYPES,
b7ab0590 91 PagedList,
083c9df9 92 parse_filesize,
91410c9b 93 PerRequestProxyHandler,
dca08720 94 platform_name,
d3c93ec2 95 Popen,
1e43a6f7 96 POSTPROCESS_WHEN,
eedb7ba5 97 PostProcessingError,
ce02ed60 98 preferredencoding,
eedb7ba5 99 prepend_extension,
f2ebc5c7 100 ReExtractInfo,
51fb4995 101 register_socks_protocols,
a06916d9 102 RejectedVideoReached,
3efb96a6 103 remove_terminal_sequences,
cfb56d1a 104 render_table,
eedb7ba5 105 replace_extension,
ce02ed60
PH
106 SameFileError,
107 sanitize_filename,
1bb5c511 108 sanitize_path,
dcf77cf1 109 sanitize_url,
67dda517 110 sanitized_Request,
e5660ee6 111 std_headers,
819e0531 112 STR_FORMAT_RE_TMPL,
113 STR_FORMAT_TYPES,
1211bb6d 114 str_or_none,
e29663c6 115 strftime_or_none,
ce02ed60 116 subtitles_filename,
819e0531 117 supports_terminal_sequences,
f2ebc5c7 118 timetuple_from_msec,
732044af 119 to_high_limit_path,
324ad820 120 traverse_obj,
6033d980 121 try_get,
ce02ed60 122 UnavailableVideoError,
29eb5174 123 url_basename,
7d1eb38a 124 variadic,
58b1f00d 125 version_tuple,
ce02ed60
PH
126 write_json_file,
127 write_string,
6a3f4c3f 128 YoutubeDLCookieProcessor,
dca08720 129 YoutubeDLHandler,
fca6dba8 130 YoutubeDLRedirectHandler,
ce02ed60 131)
a0e07d31 132from .cache import Cache
ec11a9f4 133from .minicurses import format_text
52a8a1e1 134from .extractor import (
135 gen_extractor_classes,
136 get_info_extractor,
137 _LAZY_LOADER,
3ae5e797 138 _PLUGIN_CLASSES as plugin_extractors
52a8a1e1 139)
4c54b89e 140from .extractor.openload import PhantomJSwrapper
52a8a1e1 141from .downloader import (
dbf5416a 142 FFmpegFD,
52a8a1e1 143 get_suitable_downloader,
144 shorten_protocol_name
145)
4c83c967 146from .downloader.rtmp import rtmpdump_version
4f026faf 147from .postprocessor import (
e36d50c5 148 get_postprocessor,
4e3b637d 149 EmbedThumbnailPP,
adbc4ec4 150 FFmpegFixupDuplicateMoovPP,
e36d50c5 151 FFmpegFixupDurationPP,
f17f8651 152 FFmpegFixupM3u8PP,
62cd676c 153 FFmpegFixupM4aPP,
6271f1ca 154 FFmpegFixupStretchedPP,
e36d50c5 155 FFmpegFixupTimestampPP,
4f026faf
PH
156 FFmpegMergerPP,
157 FFmpegPostProcessor,
0202b52a 158 MoveFilesAfterDownloadPP,
3ae5e797 159 _PLUGIN_CLASSES as plugin_postprocessors
4f026faf 160)
4c88ff87 161from .update import detect_variant
36eaf303 162from .version import __version__, RELEASE_GIT_HEAD
8222d8de 163
e9c0cdd3
YCH
164if compat_os_name == 'nt':
165 import ctypes
166
2459b6e1 167
8222d8de
JMF
168class YoutubeDL(object):
169 """YoutubeDL class.
170
171 YoutubeDL objects are the ones responsible of downloading the
172 actual video file and writing it to disk if the user has requested
173 it, among some other tasks. In most cases there should be one per
174 program. As, given a video URL, the downloader doesn't know how to
175 extract all the needed information, task that InfoExtractors do, it
176 has to pass the URL to one of them.
177
178 For this, YoutubeDL objects have a method that allows
179 InfoExtractors to be registered in a given order. When it is passed
180 a URL, the YoutubeDL object handles it to the first InfoExtractor it
181 finds that reports being able to handle it. The InfoExtractor extracts
182 all the information about the video or videos the URL refers to, and
183 YoutubeDL process the extracted information, possibly using a File
184 Downloader to download the video.
185
186 YoutubeDL objects accept a lot of parameters. In order not to saturate
187 the object constructor with arguments, it receives a dictionary of
188 options instead. These options are available through the params
189 attribute for the InfoExtractors to use. The YoutubeDL also
190 registers itself as the downloader in charge for the InfoExtractors
191 that are added to it, so this is a "mutual registration".
192
193 Available options:
194
195 username: Username for authentication purposes.
196 password: Password for authentication purposes.
180940e0 197 videopassword: Password for accessing a video.
1da50aa3
S
198 ap_mso: Adobe Pass multiple-system operator identifier.
199 ap_username: Multiple-system operator account username.
200 ap_password: Multiple-system operator account password.
8222d8de
JMF
201 usenetrc: Use netrc for authentication instead.
202 verbose: Print additional info to stdout.
203 quiet: Do not print messages to stdout.
ad8915b7 204 no_warnings: Do not print out anything for warnings.
bb66c247 205 forceprint: A dict with keys WHEN mapped to a list of templates to
206 print to stdout. The allowed keys are video or any of the
207 items in utils.POSTPROCESS_WHEN.
ca30f449 208 For compatibility, a single list is also accepted
bb66c247 209 print_to_file: A dict with keys WHEN (same as forceprint) mapped to
210 a list of tuples with (template, filename)
53c18592 211 forceurl: Force printing final URL. (Deprecated)
212 forcetitle: Force printing title. (Deprecated)
213 forceid: Force printing ID. (Deprecated)
214 forcethumbnail: Force printing thumbnail URL. (Deprecated)
215 forcedescription: Force printing description. (Deprecated)
216 forcefilename: Force printing final filename. (Deprecated)
217 forceduration: Force printing duration. (Deprecated)
8694c600 218 forcejson: Force printing info_dict as JSON.
63e0be34
PH
219 dump_single_json: Force printing the info_dict of the whole playlist
220 (or video) as a single JSON line.
c25228e5 221 force_write_download_archive: Force writing download archive regardless
222 of 'skip_download' or 'simulate'.
b7b04c78 223 simulate: Do not download the video files. If unset (or None),
224 simulate only if listsubtitles, listformats or list_thumbnails is used
eb8a4433 225 format: Video format code. see "FORMAT SELECTION" for more details.
093a1710 226 You can also pass a function. The function takes 'ctx' as
227 argument and returns the formats to download.
228 See "build_format_selector" for an implementation
63ad4d43 229 allow_unplayable_formats: Allow unplayable formats to be extracted and downloaded.
b7da73eb 230 ignore_no_formats_error: Ignore "No video formats" error. Usefull for
231 extracting metadata even if the video is not actually
232 available for download (experimental)
0930b11f 233 format_sort: A list of fields by which to sort the video formats.
234 See "Sorting Formats" for more details.
c25228e5 235 format_sort_force: Force the given format_sort. see "Sorting Formats"
236 for more details.
237 allow_multiple_video_streams: Allow multiple video streams to be merged
238 into a single file
239 allow_multiple_audio_streams: Allow multiple audio streams to be merged
240 into a single file
0ba692ac 241 check_formats Whether to test if the formats are downloadable.
9f1a1c36 242 Can be True (check all), False (check none),
243 'selected' (check selected formats),
0ba692ac 244 or None (check only if requested by extractor)
4524baf0 245 paths: Dictionary of output paths. The allowed keys are 'home'
246 'temp' and the keys of OUTTMPL_TYPES (in utils.py)
de6000d9 247 outtmpl: Dictionary of templates for output names. Allowed keys
4524baf0 248 are 'default' and the keys of OUTTMPL_TYPES (in utils.py).
34488702 249 For compatibility with youtube-dl, a single string can also be used
a820dc72
RA
250 outtmpl_na_placeholder: Placeholder for unavailable meta fields.
251 restrictfilenames: Do not allow "&" and spaces in file names
252 trim_file_name: Limit length of filename (extension excluded)
4524baf0 253 windowsfilenames: Force the filenames to be windows compatible
b1940459 254 ignoreerrors: Do not stop on download/postprocessing errors.
255 Can be 'only_download' to ignore only download errors.
256 Default is 'only_download' for CLI, but False for API
26e2805c 257 skip_playlist_after_errors: Number of allowed failures until the rest of
258 the playlist is skipped
d22dec74 259 force_generic_extractor: Force downloader to use the generic extractor
0c3d0f51 260 overwrites: Overwrite all video and metadata files if True,
261 overwrite only non-video files if None
262 and don't overwrite any file if False
34488702 263 For compatibility with youtube-dl,
264 "nooverwrites" may also be used instead
8222d8de
JMF
265 playliststart: Playlist item to start at.
266 playlistend: Playlist item to end at.
c14e88f0 267 playlist_items: Specific indices of playlist to download.
ff815fe6 268 playlistreverse: Download playlist items in reverse order.
75822ca7 269 playlistrandom: Download playlist items in random order.
8222d8de
JMF
270 matchtitle: Download only matching titles.
271 rejecttitle: Reject downloads for matching titles.
8bf9319e 272 logger: Log messages to a logging.Logger instance.
8222d8de 273 logtostderr: Log messages to stderr instead of stdout.
819e0531 274 consoletitle: Display progress in console window's titlebar.
8222d8de
JMF
275 writedescription: Write the video description to a .description file
276 writeinfojson: Write the video description to a .info.json file
75d43ca0 277 clean_infojson: Remove private fields from the infojson
34488702 278 getcomments: Extract video comments. This will not be written to disk
06167fbb 279 unless writeinfojson is also given
1fb07d10 280 writeannotations: Write the video annotations to a .annotations.xml file
8222d8de 281 writethumbnail: Write the thumbnail image to a file
c25228e5 282 allow_playlist_files: Whether to write playlists' description, infojson etc
283 also to disk when using the 'write*' options
ec82d85a 284 write_all_thumbnails: Write all thumbnail formats to files
732044af 285 writelink: Write an internet shortcut file, depending on the
286 current platform (.url/.webloc/.desktop)
287 writeurllink: Write a Windows internet shortcut file (.url)
288 writewebloclink: Write a macOS internet shortcut file (.webloc)
289 writedesktoplink: Write a Linux internet shortcut file (.desktop)
8222d8de 290 writesubtitles: Write the video subtitles to a file
741dd8ea 291 writeautomaticsub: Write the automatically generated subtitles to a file
245524e6 292 allsubtitles: Deprecated - Use subtitleslangs = ['all']
c32b0aab 293 Downloads all the subtitles of the video
0b7f3118 294 (requires writesubtitles or writeautomaticsub)
8222d8de 295 listsubtitles: Lists all available subtitles for the video
a504ced0 296 subtitlesformat: The format code for subtitles
c32b0aab 297 subtitleslangs: List of languages of the subtitles to download (can be regex).
298 The list may contain "all" to refer to all the available
299 subtitles. The language can be prefixed with a "-" to
300 exclude it from the requested languages. Eg: ['all', '-live_chat']
8222d8de
JMF
301 keepvideo: Keep the video file after post-processing
302 daterange: A DateRange object, download only if the upload_date is in the range.
303 skip_download: Skip the actual download of the video file
c35f9e72 304 cachedir: Location of the cache files in the filesystem.
a0e07d31 305 False to disable filesystem cache.
47192f92 306 noplaylist: Download single video instead of a playlist if in doubt.
8dbe9899
PH
307 age_limit: An integer representing the user's age in years.
308 Unsuitable videos for the given age are skipped.
5fe18bdb
PH
309 min_views: An integer representing the minimum view count the video
310 must have in order to not be skipped.
311 Videos without view count information are always
312 downloaded. None for no limit.
313 max_views: An integer representing the maximum view count.
314 Videos that are more popular than that are not
315 downloaded.
316 Videos without view count information are always
317 downloaded. None for no limit.
318 download_archive: File name of a file where all downloads are recorded.
c1c9a79c
PH
319 Videos already present in the file are not downloaded
320 again.
8a51f564 321 break_on_existing: Stop the download process after attempting to download a
322 file that is in the archive.
323 break_on_reject: Stop the download process when encountering a video that
324 has been filtered out.
b222c271 325 break_per_url: Whether break_on_reject and break_on_existing
326 should act on each input URL as opposed to for the entire queue
8a51f564 327 cookiefile: File name where cookies should be read from and dumped to
f59f5ef8
MB
328 cookiesfrombrowser: A tuple containing the name of the browser, the profile
329 name/pathfrom where cookies are loaded, and the name of the
330 keyring. Eg: ('chrome', ) or ('vivaldi', 'default', 'BASICTEXT')
f81c62a6 331 legacyserverconnect: Explicitly allow HTTPS connection to servers that do not
332 support RFC 5746 secure renegotiation
f59f5ef8 333 nocheckcertificate: Do not verify SSL certificates
7e8c0af0
PH
334 prefer_insecure: Use HTTP instead of HTTPS to retrieve information.
335 At the moment, this is only supported by YouTube.
8b7539d2 336 http_headers: A dictionary of custom headers to be used for all requests
a1ee09e8 337 proxy: URL of the proxy server to use
38cce791 338 geo_verification_proxy: URL of the proxy to use for IP address verification
504f20dd 339 on geo-restricted sites.
e344693b 340 socket_timeout: Time to wait for unresponsive hosts, in seconds
0783b09b
PH
341 bidi_workaround: Work around buggy terminals without bidirectional text
342 support, using fridibi
a0ddb8a2 343 debug_printtraffic:Print out sent and received HTTP traffic
91f071af 344 include_ads: Download ads as well (deprecated)
04b4d394
PH
345 default_search: Prepend this string if an input url is not valid.
346 'auto' for elaborate guessing
62fec3b2 347 encoding: Use this encoding instead of the system-specified.
e8ee972c 348 extract_flat: Do not resolve URLs, return the immediate result.
057a5206
PH
349 Pass in 'in_playlist' to only show this behavior for
350 playlist items.
f2ebc5c7 351 wait_for_video: If given, wait for scheduled streams to become available.
352 The value should be a tuple containing the range
353 (min_secs, max_secs) to wait between retries
4f026faf 354 postprocessors: A list of dictionaries, each with an entry
71b640cc 355 * key: The name of the postprocessor. See
7a5c1cfe 356 yt_dlp/postprocessor/__init__.py for a list.
bb66c247 357 * when: When to run the postprocessor. Allowed values are
358 the entries of utils.POSTPROCESS_WHEN
56d868db 359 Assumed to be 'post_process' if not given
b5ae35ee 360 post_hooks: Deprecated - Register a custom postprocessor instead
361 A list of functions that get called as the final step
ab8e5e51
AM
362 for each video file, after all postprocessors have been
363 called. The filename will be passed as the only argument.
71b640cc
PH
364 progress_hooks: A list of functions that get called on download
365 progress, with a dictionary with the entries
5cda4eda 366 * status: One of "downloading", "error", or "finished".
ee69b99a 367 Check this first and ignore unknown values.
3ba7740d 368 * info_dict: The extracted info_dict
71b640cc 369
5cda4eda 370 If status is one of "downloading", or "finished", the
ee69b99a
PH
371 following properties may also be present:
372 * filename: The final filename (always present)
5cda4eda 373 * tmpfilename: The filename we're currently writing to
71b640cc
PH
374 * downloaded_bytes: Bytes on disk
375 * total_bytes: Size of the whole file, None if unknown
5cda4eda
PH
376 * total_bytes_estimate: Guess of the eventual file size,
377 None if unavailable.
378 * elapsed: The number of seconds since download started.
71b640cc
PH
379 * eta: The estimated time in seconds, None if unknown
380 * speed: The download speed in bytes/second, None if
381 unknown
5cda4eda
PH
382 * fragment_index: The counter of the currently
383 downloaded video fragment.
384 * fragment_count: The number of fragments (= individual
385 files that will be merged)
71b640cc
PH
386
387 Progress hooks are guaranteed to be called at least once
388 (with status "finished") if the download is successful.
819e0531 389 postprocessor_hooks: A list of functions that get called on postprocessing
390 progress, with a dictionary with the entries
391 * status: One of "started", "processing", or "finished".
392 Check this first and ignore unknown values.
393 * postprocessor: Name of the postprocessor
394 * info_dict: The extracted info_dict
395
396 Progress hooks are guaranteed to be called at least twice
397 (with status "started" and "finished") if the processing is successful.
45598f15 398 merge_output_format: Extension to use when merging formats.
6b591b29 399 final_ext: Expected final extension; used to detect when the file was
59a7a13e 400 already downloaded and converted
6271f1ca
PH
401 fixup: Automatically correct known faults of the file.
402 One of:
403 - "never": do nothing
404 - "warn": only emit a warning
405 - "detect_or_warn": check whether we can do anything
62cd676c 406 about it, warn otherwise (default)
504f20dd 407 source_address: Client-side IP address to bind to.
6ec6cb4e 408 call_home: Boolean, true iff we are allowed to contact the
7a5c1cfe 409 yt-dlp servers for debugging. (BROKEN)
1cf376f5 410 sleep_interval_requests: Number of seconds to sleep between requests
411 during extraction
7aa589a5
S
412 sleep_interval: Number of seconds to sleep before each download when
413 used alone or a lower bound of a range for randomized
414 sleep before each download (minimum possible number
415 of seconds to sleep) when used along with
416 max_sleep_interval.
417 max_sleep_interval:Upper bound of a range for randomized sleep before each
418 download (maximum possible number of seconds to sleep).
419 Must only be used along with sleep_interval.
420 Actual sleep time will be a random float from range
421 [sleep_interval; max_sleep_interval].
1cf376f5 422 sleep_interval_subtitles: Number of seconds to sleep before each subtitle download
cfb56d1a
PH
423 listformats: Print an overview of available video formats and exit.
424 list_thumbnails: Print a table of all thumbnails and exit.
347de493
PH
425 match_filter: A function that gets called with the info_dict of
426 every video.
427 If it returns a message, the video is ignored.
428 If it returns None, the video is downloaded.
429 match_filter_func in utils.py is one example for this.
7e5db8c9 430 no_color: Do not emit color codes in output.
0a840f58 431 geo_bypass: Bypass geographic restriction via faking X-Forwarded-For
504f20dd 432 HTTP header
0a840f58 433 geo_bypass_country:
773f291d
S
434 Two-letter ISO 3166-2 country code that will be used for
435 explicit geographic restriction bypassing via faking
504f20dd 436 X-Forwarded-For HTTP header
5f95927a
S
437 geo_bypass_ip_block:
438 IP range in CIDR notation that will be used similarly to
504f20dd 439 geo_bypass_country
71b640cc 440
85729c51 441 The following options determine which downloader is picked:
52a8a1e1 442 external_downloader: A dictionary of protocol keys and the executable of the
443 external downloader to use for it. The allowed protocols
444 are default|http|ftp|m3u8|dash|rtsp|rtmp|mms.
445 Set the value to 'native' to use the native downloader
446 hls_prefer_native: Deprecated - Use external_downloader = {'m3u8': 'native'}
447 or {'m3u8': 'ffmpeg'} instead.
448 Use the native HLS downloader instead of ffmpeg/avconv
bf09af3a
S
449 if True, otherwise use ffmpeg/avconv if False, otherwise
450 use downloader suggested by extractor if None.
53ed7066 451 compat_opts: Compatibility options. See "Differences in default behavior".
3acf6d38 452 The following options do not work when used through the API:
b5ae35ee 453 filename, abort-on-error, multistreams, no-live-chat, format-sort
dac5df5a 454 no-clean-infojson, no-playlist-metafiles, no-keep-subs, no-attach-info-json.
e4f02757 455 Refer __init__.py for their implementation
819e0531 456 progress_template: Dictionary of templates for progress outputs.
457 Allowed keys are 'download', 'postprocess',
458 'download-title' (console title) and 'postprocess-title'.
459 The template is mapped on a dictionary with keys 'progress' and 'info'
fe7e0c98 460
8222d8de 461 The following parameters are not used by YoutubeDL itself, they are used by
7a5c1cfe 462 the downloader (see yt_dlp/downloader/common.py):
51d9739f 463 nopart, updatetime, buffersize, ratelimit, throttledratelimit, min_filesize,
205a0654
EH
464 max_filesize, test, noresizebuffer, retries, file_access_retries, fragment_retries,
465 continuedl, noprogress, xattr_set_filesize, hls_use_mpegts, http_chunk_size,
59a7a13e 466 external_downloader_args, concurrent_fragment_downloads.
76b1bd67
JMF
467
468 The following options are used by the post processors:
d4a24f40 469 prefer_ffmpeg: If False, use avconv instead of ffmpeg if both are available,
e4172ac9 470 otherwise prefer ffmpeg. (avconv support is deprecated)
c0b7d117
S
471 ffmpeg_location: Location of the ffmpeg/avconv binary; either the path
472 to the binary or its containing directory.
43820c03 473 postprocessor_args: A dictionary of postprocessor/executable keys (in lower case)
34488702 474 and a list of additional command-line arguments for the
475 postprocessor/executable. The dict can also have "PP+EXE" keys
476 which are used when the given exe is used by the given PP.
477 Use 'default' as the name for arguments to passed to all PP
478 For compatibility with youtube-dl, a single list of args
479 can also be used
e409895f 480
481 The following options are used by the extractors:
62bff2c1 482 extractor_retries: Number of times to retry for known errors
483 dynamic_mpd: Whether to process dynamic DASH manifests (default: True)
e409895f 484 hls_split_discontinuity: Split HLS playlists to different formats at
62bff2c1 485 discontinuities such as ad breaks (default: False)
5d3a0e79 486 extractor_args: A dictionary of arguments to be passed to the extractors.
487 See "EXTRACTOR ARGUMENTS" for details.
488 Eg: {'youtube': {'skip': ['dash', 'hls']}}
88f23a18 489 mark_watched: Mark videos watched (even with --simulate). Only for YouTube
5d3a0e79 490 youtube_include_dash_manifest: Deprecated - Use extractor_args instead.
491 If True (default), DASH manifests and related
62bff2c1 492 data will be downloaded and processed by extractor.
493 You can reduce network I/O by disabling it if you don't
494 care about DASH. (only for youtube)
5d3a0e79 495 youtube_include_hls_manifest: Deprecated - Use extractor_args instead.
496 If True (default), HLS manifests and related
62bff2c1 497 data will be downloaded and processed by extractor.
498 You can reduce network I/O by disabling it if you don't
499 care about HLS. (only for youtube)
8222d8de
JMF
500 """
501
c9969434
S
502 _NUMERIC_FIELDS = set((
503 'width', 'height', 'tbr', 'abr', 'asr', 'vbr', 'fps', 'filesize', 'filesize_approx',
e6f21b3d 504 'timestamp', 'release_timestamp',
c9969434
S
505 'duration', 'view_count', 'like_count', 'dislike_count', 'repost_count',
506 'average_rating', 'comment_count', 'age_limit',
507 'start_time', 'end_time',
508 'chapter_number', 'season_number', 'episode_number',
509 'track_number', 'disc_number', 'release_year',
c9969434
S
510 ))
511
48ee10ee 512 _format_selection_exts = {
513 'audio': {'m4a', 'mp3', 'ogg', 'aac'},
514 'video': {'mp4', 'flv', 'webm', '3gp'},
515 'storyboards': {'mhtml'},
516 }
517
8222d8de 518 params = None
8b7491c8 519 _ies = {}
1e43a6f7 520 _pps = {k: [] for k in POSTPROCESS_WHEN}
b35496d8 521 _printed_messages = set()
1cf376f5 522 _first_webpage_request = True
8222d8de
JMF
523 _download_retcode = None
524 _num_downloads = None
30a074c2 525 _playlist_level = 0
526 _playlist_urls = set()
8222d8de
JMF
527 _screen_file = None
528
3511266b 529 def __init__(self, params=None, auto_init=True):
883d4b1e 530 """Create a FileDownloader object with the given options.
531 @param auto_init Whether to load the default extractors and print header (if verbose).
49a57e70 532 Set to 'no_verbose_header' to not print the header
883d4b1e 533 """
e9f9a10f
JMF
534 if params is None:
535 params = {}
8b7491c8 536 self._ies = {}
56c73665 537 self._ies_instances = {}
1e43a6f7 538 self._pps = {k: [] for k in POSTPROCESS_WHEN}
b35496d8 539 self._printed_messages = set()
1cf376f5 540 self._first_webpage_request = True
ab8e5e51 541 self._post_hooks = []
933605d7 542 self._progress_hooks = []
819e0531 543 self._postprocessor_hooks = []
8222d8de
JMF
544 self._download_retcode = 0
545 self._num_downloads = 0
9c906919 546 self._num_videos = 0
8222d8de 547 self._screen_file = [sys.stdout, sys.stderr][params.get('logtostderr', False)]
0783b09b 548 self._err_file = sys.stderr
819e0531 549 self.params = params
a0e07d31 550 self.cache = Cache(self)
34308b30 551
819e0531 552 windows_enable_vt_mode()
ec11a9f4 553 self._allow_colors = {
554 'screen': not self.params.get('no_color') and supports_terminal_sequences(self._screen_file),
555 'err': not self.params.get('no_color') and supports_terminal_sequences(self._err_file),
556 }
819e0531 557
a61f4b28 558 if sys.version_info < (3, 6):
559 self.report_warning(
0181adef 560 'Python version %d.%d is not supported! Please update to Python 3.6 or above' % sys.version_info[:2])
a61f4b28 561
88acdbc2 562 if self.params.get('allow_unplayable_formats'):
563 self.report_warning(
ec11a9f4 564 f'You have asked for {self._format_err("UNPLAYABLE", self.Styles.EMPHASIS)} formats to be listed/downloaded. '
819e0531 565 'This is a developer option intended for debugging. \n'
566 ' If you experience any issues while using this option, '
ec11a9f4 567 f'{self._format_err("DO NOT", self.Styles.ERROR)} open a bug report')
88acdbc2 568
be5df5ee
S
569 def check_deprecated(param, option, suggestion):
570 if self.params.get(param) is not None:
53ed7066 571 self.report_warning('%s is deprecated. Use %s instead' % (option, suggestion))
be5df5ee
S
572 return True
573 return False
574
575 if check_deprecated('cn_verification_proxy', '--cn-verification-proxy', '--geo-verification-proxy'):
38cce791
YCH
576 if self.params.get('geo_verification_proxy') is None:
577 self.params['geo_verification_proxy'] = self.params['cn_verification_proxy']
578
0d1bb027 579 check_deprecated('autonumber', '--auto-number', '-o "%(autonumber)s-%(title)s.%(ext)s"')
580 check_deprecated('usetitle', '--title', '-o "%(title)s-%(id)s.%(ext)s"')
53ed7066 581 check_deprecated('useid', '--id', '-o "%(id)s.%(ext)s"')
0d1bb027 582
49a57e70 583 for msg in self.params.get('_warnings', []):
0d1bb027 584 self.report_warning(msg)
ee8dd27a 585 for msg in self.params.get('_deprecation_warnings', []):
586 self.deprecation_warning(msg)
0d1bb027 587
ec11a9f4 588 if 'list-formats' in self.params.get('compat_opts', []):
589 self.params['listformats_table'] = False
590
b5ae35ee 591 if 'overwrites' not in self.params and self.params.get('nooverwrites') is not None:
b868936c 592 # nooverwrites was unnecessarily changed to overwrites
593 # in 0c3d0f51778b153f65c21906031c2e091fcfb641
594 # This ensures compatibility with both keys
595 self.params['overwrites'] = not self.params['nooverwrites']
b5ae35ee 596 elif self.params.get('overwrites') is None:
597 self.params.pop('overwrites', None)
b868936c 598 else:
599 self.params['nooverwrites'] = not self.params['overwrites']
b9d973be 600
455a15e2 601 self.params.setdefault('forceprint', {})
602 self.params.setdefault('print_to_file', {})
bb66c247 603
604 # Compatibility with older syntax
ca30f449 605 if not isinstance(params['forceprint'], dict):
455a15e2 606 self.params['forceprint'] = {'video': params['forceprint']}
ca30f449 607
455a15e2 608 if self.params.get('bidi_workaround', False):
1c088fa8
PH
609 try:
610 import pty
611 master, slave = pty.openpty()
003c69a8 612 width = compat_get_terminal_size().columns
1c088fa8
PH
613 if width is None:
614 width_args = []
615 else:
616 width_args = ['-w', str(width)]
5d681e96 617 sp_kwargs = dict(
1c088fa8
PH
618 stdin=subprocess.PIPE,
619 stdout=slave,
620 stderr=self._err_file)
5d681e96 621 try:
d3c93ec2 622 self._output_process = Popen(['bidiv'] + width_args, **sp_kwargs)
5d681e96 623 except OSError:
d3c93ec2 624 self._output_process = Popen(['fribidi', '-c', 'UTF-8'] + width_args, **sp_kwargs)
5d681e96 625 self._output_channel = os.fdopen(master, 'rb')
1c088fa8 626 except OSError as ose:
66e7ace1 627 if ose.errno == errno.ENOENT:
49a57e70 628 self.report_warning(
629 'Could not find fribidi executable, ignoring --bidi-workaround. '
630 'Make sure that fribidi is an executable file in one of the directories in your $PATH.')
1c088fa8
PH
631 else:
632 raise
0783b09b 633
3089bc74
S
634 if (sys.platform != 'win32'
635 and sys.getfilesystemencoding() in ['ascii', 'ANSI_X3.4-1968']
455a15e2 636 and not self.params.get('restrictfilenames', False)):
e9137224 637 # Unicode filesystem API will throw errors (#1474, #13027)
34308b30 638 self.report_warning(
6febd1c1 639 'Assuming --restrict-filenames since file system encoding '
1b725173 640 'cannot encode all characters. '
6febd1c1 641 'Set the LC_ALL environment variable to fix this.')
4a98cdbf 642 self.params['restrictfilenames'] = True
34308b30 643
de6000d9 644 self.outtmpl_dict = self.parse_outtmpl()
486dd09e 645
187986a8 646 # Creating format selector here allows us to catch syntax errors before the extraction
647 self.format_selector = (
fa9f30b8 648 self.params.get('format') if self.params.get('format') in (None, '-')
093a1710 649 else self.params['format'] if callable(self.params['format'])
187986a8 650 else self.build_format_selector(self.params['format']))
651
8b7539d2 652 # Set http_headers defaults according to std_headers
653 self.params['http_headers'] = merge_headers(std_headers, self.params.get('http_headers', {}))
654
dca08720
PH
655 self._setup_opener()
656
3511266b 657 if auto_init:
883d4b1e 658 if auto_init != 'no_verbose_header':
659 self.print_debug_header()
3511266b
PH
660 self.add_default_info_extractors()
661
013b50b7 662 hooks = {
663 'post_hooks': self.add_post_hook,
664 'progress_hooks': self.add_progress_hook,
665 'postprocessor_hooks': self.add_postprocessor_hook,
666 }
667 for opt, fn in hooks.items():
668 for ph in self.params.get(opt, []):
669 fn(ph)
71b640cc 670
5bfc8bee 671 for pp_def_raw in self.params.get('postprocessors', []):
672 pp_def = dict(pp_def_raw)
673 when = pp_def.pop('when', 'post_process')
674 self.add_post_processor(
675 get_postprocessor(pp_def.pop('key'))(self, **compat_kwargs(pp_def)),
676 when=when)
677
51fb4995
YCH
678 register_socks_protocols()
679
ed39cac5 680 def preload_download_archive(fn):
681 """Preload the archive, if any is specified"""
682 if fn is None:
683 return False
49a57e70 684 self.write_debug(f'Loading archive file {fn!r}')
ed39cac5 685 try:
686 with locked_file(fn, 'r', encoding='utf-8') as archive_file:
687 for line in archive_file:
688 self.archive.add(line.strip())
689 except IOError as ioe:
690 if ioe.errno != errno.ENOENT:
691 raise
692 return False
693 return True
694
695 self.archive = set()
696 preload_download_archive(self.params.get('download_archive'))
697
7d4111ed
PH
698 def warn_if_short_id(self, argv):
699 # short YouTube ID starting with dash?
700 idxs = [
701 i for i, a in enumerate(argv)
702 if re.match(r'^-[0-9A-Za-z_-]{10}$', a)]
703 if idxs:
704 correct_argv = (
7a5c1cfe 705 ['yt-dlp']
3089bc74
S
706 + [a for i, a in enumerate(argv) if i not in idxs]
707 + ['--'] + [argv[i] for i in idxs]
7d4111ed
PH
708 )
709 self.report_warning(
710 'Long argument string detected. '
49a57e70 711 'Use -- to separate parameters and URLs, like this:\n%s' %
7d4111ed
PH
712 args_to_str(correct_argv))
713
8222d8de
JMF
714 def add_info_extractor(self, ie):
715 """Add an InfoExtractor object to the end of the list."""
8b7491c8 716 ie_key = ie.ie_key()
717 self._ies[ie_key] = ie
e52d7f85 718 if not isinstance(ie, type):
8b7491c8 719 self._ies_instances[ie_key] = ie
e52d7f85 720 ie.set_downloader(self)
8222d8de 721
8b7491c8 722 def _get_info_extractor_class(self, ie_key):
723 ie = self._ies.get(ie_key)
724 if ie is None:
725 ie = get_info_extractor(ie_key)
726 self.add_info_extractor(ie)
727 return ie
728
56c73665
JMF
729 def get_info_extractor(self, ie_key):
730 """
731 Get an instance of an IE with name ie_key, it will try to get one from
732 the _ies list, if there's no instance it will create a new one and add
733 it to the extractor list.
734 """
735 ie = self._ies_instances.get(ie_key)
736 if ie is None:
737 ie = get_info_extractor(ie_key)()
738 self.add_info_extractor(ie)
739 return ie
740
023fa8c4
JMF
741 def add_default_info_extractors(self):
742 """
743 Add the InfoExtractors returned by gen_extractors to the end of the list
744 """
e52d7f85 745 for ie in gen_extractor_classes():
023fa8c4
JMF
746 self.add_info_extractor(ie)
747
56d868db 748 def add_post_processor(self, pp, when='post_process'):
8222d8de 749 """Add a PostProcessor object to the end of the chain."""
5bfa4862 750 self._pps[when].append(pp)
8222d8de
JMF
751 pp.set_downloader(self)
752
ab8e5e51
AM
753 def add_post_hook(self, ph):
754 """Add the post hook"""
755 self._post_hooks.append(ph)
756
933605d7 757 def add_progress_hook(self, ph):
819e0531 758 """Add the download progress hook"""
933605d7 759 self._progress_hooks.append(ph)
8ab470f1 760
819e0531 761 def add_postprocessor_hook(self, ph):
762 """Add the postprocessing progress hook"""
763 self._postprocessor_hooks.append(ph)
5bfc8bee 764 for pps in self._pps.values():
765 for pp in pps:
766 pp.add_progress_hook(ph)
819e0531 767
1c088fa8 768 def _bidi_workaround(self, message):
5d681e96 769 if not hasattr(self, '_output_channel'):
1c088fa8
PH
770 return message
771
5d681e96 772 assert hasattr(self, '_output_process')
11b85ce6 773 assert isinstance(message, compat_str)
6febd1c1
PH
774 line_count = message.count('\n') + 1
775 self._output_process.stdin.write((message + '\n').encode('utf-8'))
5d681e96 776 self._output_process.stdin.flush()
6febd1c1 777 res = ''.join(self._output_channel.readline().decode('utf-8')
9e1a5b84 778 for _ in range(line_count))
6febd1c1 779 return res[:-len('\n')]
1c088fa8 780
b35496d8 781 def _write_string(self, message, out=None, only_once=False):
782 if only_once:
783 if message in self._printed_messages:
784 return
785 self._printed_messages.add(message)
786 write_string(message, out=out, encoding=self.params.get('encoding'))
734f90bb 787
848887eb 788 def to_stdout(self, message, skip_eol=False, quiet=False):
0760b0a7 789 """Print message to stdout"""
8bf9319e 790 if self.params.get('logger'):
43afe285 791 self.params['logger'].debug(message)
835a1478 792 elif not quiet or self.params.get('verbose'):
793 self._write_string(
794 '%s%s' % (self._bidi_workaround(message), ('' if skip_eol else '\n')),
795 self._err_file if quiet else self._screen_file)
8222d8de 796
b35496d8 797 def to_stderr(self, message, only_once=False):
0760b0a7 798 """Print message to stderr"""
11b85ce6 799 assert isinstance(message, compat_str)
8bf9319e 800 if self.params.get('logger'):
43afe285
IB
801 self.params['logger'].error(message)
802 else:
b35496d8 803 self._write_string('%s\n' % self._bidi_workaround(message), self._err_file, only_once=only_once)
8222d8de 804
1e5b9a95
PH
805 def to_console_title(self, message):
806 if not self.params.get('consoletitle', False):
807 return
3efb96a6 808 message = remove_terminal_sequences(message)
4bede0d8
C
809 if compat_os_name == 'nt':
810 if ctypes.windll.kernel32.GetConsoleWindow():
811 # c_wchar_p() might not be necessary if `message` is
812 # already of type unicode()
813 ctypes.windll.kernel32.SetConsoleTitleW(ctypes.c_wchar_p(message))
1e5b9a95 814 elif 'TERM' in os.environ:
b46696bd 815 self._write_string('\033]0;%s\007' % message, self._screen_file)
1e5b9a95 816
bdde425c
PH
817 def save_console_title(self):
818 if not self.params.get('consoletitle', False):
819 return
b7b04c78 820 if self.params.get('simulate'):
94c3442e 821 return
4bede0d8 822 if compat_os_name != 'nt' and 'TERM' in os.environ:
efd6c574 823 # Save the title on stack
734f90bb 824 self._write_string('\033[22;0t', self._screen_file)
bdde425c
PH
825
826 def restore_console_title(self):
827 if not self.params.get('consoletitle', False):
828 return
b7b04c78 829 if self.params.get('simulate'):
94c3442e 830 return
4bede0d8 831 if compat_os_name != 'nt' and 'TERM' in os.environ:
efd6c574 832 # Restore the title from stack
734f90bb 833 self._write_string('\033[23;0t', self._screen_file)
bdde425c
PH
834
835 def __enter__(self):
836 self.save_console_title()
837 return self
838
839 def __exit__(self, *args):
840 self.restore_console_title()
f89197d7 841
dca08720 842 if self.params.get('cookiefile') is not None:
1bab3437 843 self.cookiejar.save(ignore_discard=True, ignore_expires=True)
bdde425c 844
fa9f30b8 845 def trouble(self, message=None, tb=None, is_error=True):
8222d8de
JMF
846 """Determine action to take when a download problem appears.
847
848 Depending on if the downloader has been configured to ignore
849 download errors or not, this method may throw an exception or
850 not when errors are found, after printing the message.
851
fa9f30b8 852 @param tb If given, is additional traceback information
853 @param is_error Whether to raise error according to ignorerrors
8222d8de
JMF
854 """
855 if message is not None:
856 self.to_stderr(message)
857 if self.params.get('verbose'):
858 if tb is None:
859 if sys.exc_info()[0]: # if .trouble has been called from an except block
6febd1c1 860 tb = ''
8222d8de 861 if hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
6febd1c1 862 tb += ''.join(traceback.format_exception(*sys.exc_info()[1].exc_info))
c0384f22 863 tb += encode_compat_str(traceback.format_exc())
8222d8de
JMF
864 else:
865 tb_data = traceback.format_list(traceback.extract_stack())
6febd1c1 866 tb = ''.join(tb_data)
c19bc311 867 if tb:
868 self.to_stderr(tb)
fa9f30b8 869 if not is_error:
870 return
b1940459 871 if not self.params.get('ignoreerrors'):
8222d8de
JMF
872 if sys.exc_info()[0] and hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
873 exc_info = sys.exc_info()[1].exc_info
874 else:
875 exc_info = sys.exc_info()
876 raise DownloadError(message, exc_info)
877 self._download_retcode = 1
878
0760b0a7 879 def to_screen(self, message, skip_eol=False):
880 """Print message to stdout if not in quiet mode"""
881 self.to_stdout(
882 message, skip_eol, quiet=self.params.get('quiet', False))
883
ec11a9f4 884 class Styles(Enum):
885 HEADERS = 'yellow'
f304da8a 886 EMPHASIS = 'light blue'
ec11a9f4 887 ID = 'green'
888 DELIM = 'blue'
889 ERROR = 'red'
890 WARNING = 'yellow'
ff51ed58 891 SUPPRESS = 'light black'
ec11a9f4 892
7578d77d 893 def _format_text(self, handle, allow_colors, text, f, fallback=None, *, test_encoding=False):
ec11a9f4 894 if test_encoding:
895 original_text = text
5c104538 896 # handle.encoding can be None. See https://github.com/yt-dlp/yt-dlp/issues/2711
897 encoding = self.params.get('encoding') or getattr(handle, 'encoding', None) or 'ascii'
ec11a9f4 898 text = text.encode(encoding, 'ignore').decode(encoding)
899 if fallback is not None and text != original_text:
900 text = fallback
901 if isinstance(f, self.Styles):
f304da8a 902 f = f.value
7578d77d 903 return format_text(text, f) if allow_colors else text if fallback is None else fallback
ec11a9f4 904
905 def _format_screen(self, *args, **kwargs):
7578d77d 906 return self._format_text(
907 self._screen_file, self._allow_colors['screen'], *args, **kwargs)
ec11a9f4 908
909 def _format_err(self, *args, **kwargs):
7578d77d 910 return self._format_text(
911 self._err_file, self._allow_colors['err'], *args, **kwargs)
819e0531 912
c84aeac6 913 def report_warning(self, message, only_once=False):
8222d8de
JMF
914 '''
915 Print the message to stderr, it will be prefixed with 'WARNING:'
916 If stderr is a tty file the 'WARNING:' will be colored
917 '''
6d07ce01
JMF
918 if self.params.get('logger') is not None:
919 self.params['logger'].warning(message)
8222d8de 920 else:
ad8915b7
PH
921 if self.params.get('no_warnings'):
922 return
ec11a9f4 923 self.to_stderr(f'{self._format_err("WARNING:", self.Styles.WARNING)} {message}', only_once)
8222d8de 924
ee8dd27a 925 def deprecation_warning(self, message):
926 if self.params.get('logger') is not None:
927 self.params['logger'].warning('DeprecationWarning: {message}')
928 else:
929 self.to_stderr(f'{self._format_err("DeprecationWarning:", self.Styles.ERROR)} {message}', True)
930
fa9f30b8 931 def report_error(self, message, *args, **kwargs):
8222d8de
JMF
932 '''
933 Do the same as trouble, but prefixes the message with 'ERROR:', colored
934 in red if stderr is a tty file.
935 '''
fa9f30b8 936 self.trouble(f'{self._format_err("ERROR:", self.Styles.ERROR)} {message}', *args, **kwargs)
8222d8de 937
b35496d8 938 def write_debug(self, message, only_once=False):
0760b0a7 939 '''Log debug message or Print message to stderr'''
940 if not self.params.get('verbose', False):
941 return
942 message = '[debug] %s' % message
943 if self.params.get('logger'):
944 self.params['logger'].debug(message)
945 else:
b35496d8 946 self.to_stderr(message, only_once)
0760b0a7 947
8222d8de
JMF
948 def report_file_already_downloaded(self, file_name):
949 """Report file has already been fully downloaded."""
950 try:
6febd1c1 951 self.to_screen('[download] %s has already been downloaded' % file_name)
ce02ed60 952 except UnicodeEncodeError:
6febd1c1 953 self.to_screen('[download] The file has already been downloaded')
8222d8de 954
0c3d0f51 955 def report_file_delete(self, file_name):
956 """Report that existing file will be deleted."""
957 try:
c25228e5 958 self.to_screen('Deleting existing file %s' % file_name)
0c3d0f51 959 except UnicodeEncodeError:
c25228e5 960 self.to_screen('Deleting existing file')
0c3d0f51 961
319b6059 962 def raise_no_formats(self, info, forced=False, *, msg=None):
1151c407 963 has_drm = info.get('__has_drm')
319b6059 964 ignored, expected = self.params.get('ignore_no_formats_error'), bool(msg)
965 msg = msg or has_drm and 'This video is DRM protected' or 'No video formats found!'
966 if forced or not ignored:
1151c407 967 raise ExtractorError(msg, video_id=info['id'], ie=info['extractor'],
319b6059 968 expected=has_drm or ignored or expected)
88acdbc2 969 else:
970 self.report_warning(msg)
971
de6000d9 972 def parse_outtmpl(self):
973 outtmpl_dict = self.params.get('outtmpl', {})
974 if not isinstance(outtmpl_dict, dict):
975 outtmpl_dict = {'default': outtmpl_dict}
71ce444a 976 # Remove spaces in the default template
977 if self.params.get('restrictfilenames'):
978 sanitize = lambda x: x.replace(' - ', ' ').replace(' ', '-')
979 else:
980 sanitize = lambda x: x
de6000d9 981 outtmpl_dict.update({
71ce444a 982 k: sanitize(v) for k, v in DEFAULT_OUTTMPL.items()
80c03fa9 983 if outtmpl_dict.get(k) is None})
de6000d9 984 for key, val in outtmpl_dict.items():
985 if isinstance(val, bytes):
986 self.report_warning(
987 'Parameter outtmpl is bytes, but should be a unicode string. '
988 'Put from __future__ import unicode_literals at the top of your code file or consider switching to Python 3.x.')
989 return outtmpl_dict
990
21cd8fae 991 def get_output_path(self, dir_type='', filename=None):
992 paths = self.params.get('paths', {})
993 assert isinstance(paths, dict)
994 path = os.path.join(
995 expand_path(paths.get('home', '').strip()),
996 expand_path(paths.get(dir_type, '').strip()) if dir_type else '',
997 filename or '')
998
999 # Temporary fix for #4787
1000 # 'Treat' all problem characters by passing filename through preferredencoding
1001 # to workaround encoding issues with subprocess on python2 @ Windows
1002 if sys.version_info < (3, 0) and sys.platform == 'win32':
1003 path = encodeFilename(path, True).decode(preferredencoding())
1004 return sanitize_path(path, force=self.params.get('windowsfilenames'))
1005
76a264ac 1006 @staticmethod
901130bb 1007 def _outtmpl_expandpath(outtmpl):
1008 # expand_path translates '%%' into '%' and '$$' into '$'
1009 # correspondingly that is not what we want since we need to keep
1010 # '%%' intact for template dict substitution step. Working around
1011 # with boundary-alike separator hack.
1012 sep = ''.join([random.choice(ascii_letters) for _ in range(32)])
1013 outtmpl = outtmpl.replace('%%', '%{0}%'.format(sep)).replace('$$', '${0}$'.format(sep))
1014
1015 # outtmpl should be expand_path'ed before template dict substitution
1016 # because meta fields may contain env variables we don't want to
1017 # be expanded. For example, for outtmpl "%(title)s.%(ext)s" and
1018 # title "Hello $PATH", we don't want `$PATH` to be expanded.
1019 return expand_path(outtmpl).replace(sep, '')
1020
1021 @staticmethod
1022 def escape_outtmpl(outtmpl):
1023 ''' Escape any remaining strings like %s, %abc% etc. '''
1024 return re.sub(
1025 STR_FORMAT_RE_TMPL.format('', '(?![%(\0])'),
1026 lambda mobj: ('' if mobj.group('has_key') else '%') + mobj.group(0),
1027 outtmpl)
1028
1029 @classmethod
1030 def validate_outtmpl(cls, outtmpl):
76a264ac 1031 ''' @return None or Exception object '''
7d1eb38a 1032 outtmpl = re.sub(
37893bb0 1033 STR_FORMAT_RE_TMPL.format('[^)]*', '[ljqBUDS]'),
7d1eb38a 1034 lambda mobj: f'{mobj.group(0)[:-1]}s',
1035 cls._outtmpl_expandpath(outtmpl))
76a264ac 1036 try:
7d1eb38a 1037 cls.escape_outtmpl(outtmpl) % collections.defaultdict(int)
76a264ac 1038 return None
1039 except ValueError as err:
1040 return err
1041
03b4de72 1042 @staticmethod
1043 def _copy_infodict(info_dict):
1044 info_dict = dict(info_dict)
09b49e1f 1045 info_dict.pop('__postprocessors', None)
03b4de72 1046 return info_dict
1047
e0fd9573 1048 def prepare_outtmpl(self, outtmpl, info_dict, sanitize=False):
1049 """ Make the outtmpl and info_dict suitable for substitution: ydl.escape_outtmpl(outtmpl) % info_dict
1050 @param sanitize Whether to sanitize the output as a filename.
1051 For backward compatibility, a function can also be passed
1052 """
1053
6e84b215 1054 info_dict.setdefault('epoch', int(time.time())) # keep epoch consistent once set
143db31d 1055
03b4de72 1056 info_dict = self._copy_infodict(info_dict)
752cda38 1057 info_dict['duration_string'] = ( # %(duration>%H-%M-%S)s is wrong if duration > 24hrs
53c18592 1058 formatSeconds(info_dict['duration'], '-' if sanitize else ':')
143db31d 1059 if info_dict.get('duration', None) is not None
1060 else None)
752cda38 1061 info_dict['autonumber'] = self.params.get('autonumber_start', 1) - 1 + self._num_downloads
9c906919 1062 info_dict['video_autonumber'] = self._num_videos
752cda38 1063 if info_dict.get('resolution') is None:
1064 info_dict['resolution'] = self.format_resolution(info_dict, default=None)
143db31d 1065
e6f21b3d 1066 # For fields playlist_index, playlist_autonumber and autonumber convert all occurrences
143db31d 1067 # of %(field)s to %(field)0Nd for backward compatibility
1068 field_size_compat_map = {
ec11a9f4 1069 'playlist_index': number_of_digits(info_dict.get('_last_playlist_index') or 0),
1070 'playlist_autonumber': number_of_digits(info_dict.get('n_entries') or 0),
752cda38 1071 'autonumber': self.params.get('autonumber_size') or 5,
143db31d 1072 }
752cda38 1073
385a27fa 1074 TMPL_DICT = {}
37893bb0 1075 EXTERNAL_FORMAT_RE = re.compile(STR_FORMAT_RE_TMPL.format('[^)]*', f'[{STR_FORMAT_TYPES}ljqBUDS]'))
385a27fa 1076 MATH_FUNCTIONS = {
1077 '+': float.__add__,
1078 '-': float.__sub__,
1079 }
e625be0d 1080 # Field is of the form key1.key2...
1081 # where keys (except first) can be string, int or slice
2b8a2973 1082 FIELD_RE = r'\w*(?:\.(?:\w+|{num}|{num}?(?::{num}?){{1,2}}))*'.format(num=r'(?:-?\d+)')
e0fd9573 1083 MATH_FIELD_RE = r'''(?:{field}|{num})'''.format(field=FIELD_RE, num=r'-?\d+(?:.\d+)?')
385a27fa 1084 MATH_OPERATORS_RE = r'(?:%s)' % '|'.join(map(re.escape, MATH_FUNCTIONS.keys()))
e625be0d 1085 INTERNAL_FORMAT_RE = re.compile(r'''(?x)
1086 (?P<negate>-)?
385a27fa 1087 (?P<fields>{field})
1088 (?P<maths>(?:{math_op}{math_field})*)
e625be0d 1089 (?:>(?P<strf_format>.+?))?
e978789f
P
1090 (?P<alternate>(?<!\\),[^|&)]+)?
1091 (?:&(?P<replacement>.*?))?
e625be0d 1092 (?:\|(?P<default>.*?))?
385a27fa 1093 $'''.format(field=FIELD_RE, math_op=MATH_OPERATORS_RE, math_field=MATH_FIELD_RE))
752cda38 1094
2b8a2973 1095 def _traverse_infodict(k):
1096 k = k.split('.')
1097 if k[0] == '':
1098 k.pop(0)
1099 return traverse_obj(info_dict, k, is_user_input=True, traverse_string=True)
76a264ac 1100
752cda38 1101 def get_value(mdict):
1102 # Object traversal
2b8a2973 1103 value = _traverse_infodict(mdict['fields'])
752cda38 1104 # Negative
1105 if mdict['negate']:
1106 value = float_or_none(value)
1107 if value is not None:
1108 value *= -1
1109 # Do maths
385a27fa 1110 offset_key = mdict['maths']
1111 if offset_key:
752cda38 1112 value = float_or_none(value)
1113 operator = None
385a27fa 1114 while offset_key:
1115 item = re.match(
1116 MATH_FIELD_RE if operator else MATH_OPERATORS_RE,
1117 offset_key).group(0)
1118 offset_key = offset_key[len(item):]
1119 if operator is None:
752cda38 1120 operator = MATH_FUNCTIONS[item]
385a27fa 1121 continue
1122 item, multiplier = (item[1:], -1) if item[0] == '-' else (item, 1)
1123 offset = float_or_none(item)
1124 if offset is None:
2b8a2973 1125 offset = float_or_none(_traverse_infodict(item))
385a27fa 1126 try:
1127 value = operator(value, multiplier * offset)
1128 except (TypeError, ZeroDivisionError):
1129 return None
1130 operator = None
752cda38 1131 # Datetime formatting
1132 if mdict['strf_format']:
7c37ff97 1133 value = strftime_or_none(value, mdict['strf_format'].replace('\\,', ','))
752cda38 1134
1135 return value
1136
b868936c 1137 na = self.params.get('outtmpl_na_placeholder', 'NA')
1138
e0fd9573 1139 def filename_sanitizer(key, value, restricted=self.params.get('restrictfilenames')):
1140 return sanitize_filename(str(value), restricted=restricted,
1141 is_id=re.search(r'(^|[_.])id(\.|$)', key))
1142
1143 sanitizer = sanitize if callable(sanitize) else filename_sanitizer
1144 sanitize = bool(sanitize)
1145
6e84b215 1146 def _dumpjson_default(obj):
1147 if isinstance(obj, (set, LazyList)):
1148 return list(obj)
adbc4ec4 1149 return repr(obj)
6e84b215 1150
752cda38 1151 def create_key(outer_mobj):
1152 if not outer_mobj.group('has_key'):
b836dc94 1153 return outer_mobj.group(0)
752cda38 1154 key = outer_mobj.group('key')
752cda38 1155 mobj = re.match(INTERNAL_FORMAT_RE, key)
e0fd9573 1156 initial_field = mobj.group('fields') if mobj else ''
e978789f 1157 value, replacement, default = None, None, na
7c37ff97 1158 while mobj:
e625be0d 1159 mobj = mobj.groupdict()
7c37ff97 1160 default = mobj['default'] if mobj['default'] is not None else default
752cda38 1161 value = get_value(mobj)
e978789f 1162 replacement = mobj['replacement']
7c37ff97 1163 if value is None and mobj['alternate']:
1164 mobj = re.match(INTERNAL_FORMAT_RE, mobj['alternate'][1:])
1165 else:
1166 break
752cda38 1167
b868936c 1168 fmt = outer_mobj.group('format')
752cda38 1169 if fmt == 's' and value is not None and key in field_size_compat_map.keys():
1170 fmt = '0{:d}d'.format(field_size_compat_map[key])
1171
e978789f 1172 value = default if value is None else value if replacement is None else replacement
752cda38 1173
4476d2c7 1174 flags = outer_mobj.group('conversion') or ''
7d1eb38a 1175 str_fmt = f'{fmt[:-1]}s'
524e2e4f 1176 if fmt[-1] == 'l': # list
4476d2c7 1177 delim = '\n' if '#' in flags else ', '
9e907ebd 1178 value, fmt = delim.join(map(str, variadic(value, allowed_types=(str, bytes)))), str_fmt
524e2e4f 1179 elif fmt[-1] == 'j': # json
4476d2c7 1180 value, fmt = json.dumps(value, default=_dumpjson_default, indent=4 if '#' in flags else None), str_fmt
524e2e4f 1181 elif fmt[-1] == 'q': # quoted
4476d2c7 1182 value = map(str, variadic(value) if '#' in flags else [value])
1183 value, fmt = ' '.join(map(compat_shlex_quote, value)), str_fmt
524e2e4f 1184 elif fmt[-1] == 'B': # bytes
f5aa5cfb 1185 value = f'%{str_fmt}'.encode('utf-8') % str(value).encode('utf-8')
1186 value, fmt = value.decode('utf-8', 'ignore'), 's'
524e2e4f 1187 elif fmt[-1] == 'U': # unicode normalized
524e2e4f 1188 value, fmt = unicodedata.normalize(
1189 # "+" = compatibility equivalence, "#" = NFD
4476d2c7 1190 'NF%s%s' % ('K' if '+' in flags else '', 'D' if '#' in flags else 'C'),
524e2e4f 1191 value), str_fmt
e0fd9573 1192 elif fmt[-1] == 'D': # decimal suffix
abbeeebc 1193 num_fmt, fmt = fmt[:-1].replace('#', ''), 's'
1194 value = format_decimal_suffix(value, f'%{num_fmt}f%s' if num_fmt else '%d%s',
1195 factor=1024 if '#' in flags else 1000)
37893bb0 1196 elif fmt[-1] == 'S': # filename sanitization
e0fd9573 1197 value, fmt = filename_sanitizer(initial_field, value, restricted='#' in flags), str_fmt
7d1eb38a 1198 elif fmt[-1] == 'c':
524e2e4f 1199 if value:
1200 value = str(value)[0]
76a264ac 1201 else:
524e2e4f 1202 fmt = str_fmt
76a264ac 1203 elif fmt[-1] not in 'rs': # numeric
a439a3a4 1204 value = float_or_none(value)
752cda38 1205 if value is None:
1206 value, fmt = default, 's'
901130bb 1207
752cda38 1208 if sanitize:
1209 if fmt[-1] == 'r':
1210 # If value is an object, sanitize might convert it to a string
1211 # So we convert it to repr first
7d1eb38a 1212 value, fmt = repr(value), str_fmt
639f1cea 1213 if fmt[-1] in 'csr':
e0fd9573 1214 value = sanitizer(initial_field, value)
901130bb 1215
b868936c 1216 key = '%s\0%s' % (key.replace('%', '%\0'), outer_mobj.group('format'))
385a27fa 1217 TMPL_DICT[key] = value
b868936c 1218 return '{prefix}%({key}){fmt}'.format(key=key, fmt=fmt, prefix=outer_mobj.group('prefix'))
752cda38 1219
385a27fa 1220 return EXTERNAL_FORMAT_RE.sub(create_key, outtmpl), TMPL_DICT
143db31d 1221
819e0531 1222 def evaluate_outtmpl(self, outtmpl, info_dict, *args, **kwargs):
1223 outtmpl, info_dict = self.prepare_outtmpl(outtmpl, info_dict, *args, **kwargs)
1224 return self.escape_outtmpl(outtmpl) % info_dict
1225
de6000d9 1226 def _prepare_filename(self, info_dict, tmpl_type='default'):
8222d8de 1227 try:
b836dc94 1228 outtmpl = self._outtmpl_expandpath(self.outtmpl_dict.get(tmpl_type, self.outtmpl_dict['default']))
e0fd9573 1229 filename = self.evaluate_outtmpl(outtmpl, info_dict, True)
6a0546e3 1230 if not filename:
1231 return None
15da37c7 1232
6a0546e3 1233 if tmpl_type in ('default', 'temp'):
1234 final_ext, ext = self.params.get('final_ext'), info_dict.get('ext')
1235 if final_ext and ext and final_ext != ext and filename.endswith(f'.{final_ext}'):
1236 filename = replace_extension(filename, ext, final_ext)
1237 else:
1238 force_ext = OUTTMPL_TYPES[tmpl_type]
1239 if force_ext:
1240 filename = replace_extension(filename, force_ext, info_dict.get('ext'))
de6000d9 1241
bdc3fd2f
U
1242 # https://github.com/blackjack4494/youtube-dlc/issues/85
1243 trim_file_name = self.params.get('trim_file_name', False)
1244 if trim_file_name:
5c22c63d 1245 no_ext, *ext = filename.rsplit('.', 2)
1246 filename = join_nonempty(no_ext[:trim_file_name], *ext, delim='.')
bdc3fd2f 1247
0202b52a 1248 return filename
8222d8de 1249 except ValueError as err:
6febd1c1 1250 self.report_error('Error in output template: ' + str(err) + ' (encoding: ' + repr(preferredencoding()) + ')')
8222d8de
JMF
1251 return None
1252
de6000d9 1253 def prepare_filename(self, info_dict, dir_type='', warn=False):
1254 """Generate the output filename."""
21cd8fae 1255
de6000d9 1256 filename = self._prepare_filename(info_dict, dir_type or 'default')
80c03fa9 1257 if not filename and dir_type not in ('', 'temp'):
1258 return ''
de6000d9 1259
c84aeac6 1260 if warn:
21cd8fae 1261 if not self.params.get('paths'):
de6000d9 1262 pass
1263 elif filename == '-':
c84aeac6 1264 self.report_warning('--paths is ignored when an outputting to stdout', only_once=True)
de6000d9 1265 elif os.path.isabs(filename):
c84aeac6 1266 self.report_warning('--paths is ignored since an absolute path is given in output template', only_once=True)
de6000d9 1267 if filename == '-' or not filename:
1268 return filename
1269
21cd8fae 1270 return self.get_output_path(dir_type, filename)
0202b52a 1271
120fe513 1272 def _match_entry(self, info_dict, incomplete=False, silent=False):
ecdec191 1273 """ Returns None if the file should be downloaded """
8222d8de 1274
c77495e3 1275 video_title = info_dict.get('title', info_dict.get('id', 'video'))
1276
8b0d7497 1277 def check_filter():
8b0d7497 1278 if 'title' in info_dict:
1279 # This can happen when we're just evaluating the playlist
1280 title = info_dict['title']
1281 matchtitle = self.params.get('matchtitle', False)
1282 if matchtitle:
1283 if not re.search(matchtitle, title, re.IGNORECASE):
1284 return '"' + title + '" title did not match pattern "' + matchtitle + '"'
1285 rejecttitle = self.params.get('rejecttitle', False)
1286 if rejecttitle:
1287 if re.search(rejecttitle, title, re.IGNORECASE):
1288 return '"' + title + '" title matched reject pattern "' + rejecttitle + '"'
1289 date = info_dict.get('upload_date')
1290 if date is not None:
1291 dateRange = self.params.get('daterange', DateRange())
1292 if date not in dateRange:
1293 return '%s upload date is not in range %s' % (date_from_str(date).isoformat(), dateRange)
1294 view_count = info_dict.get('view_count')
1295 if view_count is not None:
1296 min_views = self.params.get('min_views')
1297 if min_views is not None and view_count < min_views:
1298 return 'Skipping %s, because it has not reached minimum view count (%d/%d)' % (video_title, view_count, min_views)
1299 max_views = self.params.get('max_views')
1300 if max_views is not None and view_count > max_views:
1301 return 'Skipping %s, because it has exceeded the maximum view count (%d/%d)' % (video_title, view_count, max_views)
1302 if age_restricted(info_dict.get('age_limit'), self.params.get('age_limit')):
1303 return 'Skipping "%s" because it is age restricted' % video_title
8b0d7497 1304
8f18aca8 1305 match_filter = self.params.get('match_filter')
1306 if match_filter is not None:
1307 try:
1308 ret = match_filter(info_dict, incomplete=incomplete)
1309 except TypeError:
1310 # For backward compatibility
1311 ret = None if incomplete else match_filter(info_dict)
1312 if ret is not None:
1313 return ret
8b0d7497 1314 return None
1315
c77495e3 1316 if self.in_download_archive(info_dict):
1317 reason = '%s has already been recorded in the archive' % video_title
1318 break_opt, break_err = 'break_on_existing', ExistingVideoReached
1319 else:
1320 reason = check_filter()
1321 break_opt, break_err = 'break_on_reject', RejectedVideoReached
8b0d7497 1322 if reason is not None:
120fe513 1323 if not silent:
1324 self.to_screen('[download] ' + reason)
c77495e3 1325 if self.params.get(break_opt, False):
1326 raise break_err()
8b0d7497 1327 return reason
fe7e0c98 1328
b6c45014
JMF
1329 @staticmethod
1330 def add_extra_info(info_dict, extra_info):
1331 '''Set the keys from extra_info in info dict if they are missing'''
1332 for key, value in extra_info.items():
1333 info_dict.setdefault(key, value)
1334
409e1828 1335 def extract_info(self, url, download=True, ie_key=None, extra_info=None,
61aa5ba3 1336 process=True, force_generic_extractor=False):
41d1cca3 1337 """
1338 Return a list with a dictionary for each video extracted.
1339
1340 Arguments:
1341 url -- URL to extract
1342
1343 Keyword arguments:
1344 download -- whether to download videos during extraction
1345 ie_key -- extractor key hint
1346 extra_info -- dictionary containing the extra values to add to each result
1347 process -- whether to resolve all unresolved references (URLs, playlist items),
1348 must be True for download to work.
1349 force_generic_extractor -- force using the generic extractor
1350 """
fe7e0c98 1351
409e1828 1352 if extra_info is None:
1353 extra_info = {}
1354
61aa5ba3 1355 if not ie_key and force_generic_extractor:
d22dec74
S
1356 ie_key = 'Generic'
1357
8222d8de 1358 if ie_key:
8b7491c8 1359 ies = {ie_key: self._get_info_extractor_class(ie_key)}
8222d8de
JMF
1360 else:
1361 ies = self._ies
1362
8b7491c8 1363 for ie_key, ie in ies.items():
8222d8de
JMF
1364 if not ie.suitable(url):
1365 continue
1366
1367 if not ie.working():
6febd1c1
PH
1368 self.report_warning('The program functionality for this site has been marked as broken, '
1369 'and will probably not work.')
8222d8de 1370
1151c407 1371 temp_id = ie.get_temp_id(url)
a0566bbf 1372 if temp_id is not None and self.in_download_archive({'id': temp_id, 'ie_key': ie_key}):
5e5be0c0 1373 self.to_screen(f'[{ie_key}] {temp_id}: has already been recorded in the archive')
1374 if self.params.get('break_on_existing', False):
1375 raise ExistingVideoReached()
a0566bbf 1376 break
8b7491c8 1377 return self.__extract_info(url, self.get_info_extractor(ie_key), download, extra_info, process)
a0566bbf 1378 else:
1379 self.report_error('no suitable InfoExtractor for URL %s' % url)
1380
8e5fecc8 1381 def __handle_extraction_exceptions(func):
b5ae35ee 1382 @functools.wraps(func)
a0566bbf 1383 def wrapper(self, *args, **kwargs):
6da22e7d 1384 while True:
1385 try:
1386 return func(self, *args, **kwargs)
1387 except (DownloadCancelled, LazyList.IndexError, PagedList.IndexError):
8222d8de 1388 raise
6da22e7d 1389 except ReExtractInfo as e:
1390 if e.expected:
1391 self.to_screen(f'{e}; Re-extracting data')
1392 else:
1393 self.to_stderr('\r')
1394 self.report_warning(f'{e}; Re-extracting data')
1395 continue
1396 except GeoRestrictedError as e:
1397 msg = e.msg
1398 if e.countries:
1399 msg += '\nThis video is available in %s.' % ', '.join(
1400 map(ISO3166Utils.short2full, e.countries))
1401 msg += '\nYou might want to use a VPN or a proxy server (with --proxy) to workaround.'
1402 self.report_error(msg)
1403 except ExtractorError as e: # An error we somewhat expected
1404 self.report_error(str(e), e.format_traceback())
1405 except Exception as e:
1406 if self.params.get('ignoreerrors'):
1407 self.report_error(str(e), tb=encode_compat_str(traceback.format_exc()))
1408 else:
1409 raise
1410 break
a0566bbf 1411 return wrapper
1412
f2ebc5c7 1413 def _wait_for_video(self, ie_result):
1414 if (not self.params.get('wait_for_video')
1415 or ie_result.get('_type', 'video') != 'video'
1416 or ie_result.get('formats') or ie_result.get('url')):
1417 return
1418
1419 format_dur = lambda dur: '%02d:%02d:%02d' % timetuple_from_msec(dur * 1000)[:-1]
1420 last_msg = ''
1421
1422 def progress(msg):
1423 nonlocal last_msg
1424 self.to_screen(msg + ' ' * (len(last_msg) - len(msg)) + '\r', skip_eol=True)
1425 last_msg = msg
1426
1427 min_wait, max_wait = self.params.get('wait_for_video')
1428 diff = try_get(ie_result, lambda x: x['release_timestamp'] - time.time())
1429 if diff is None and ie_result.get('live_status') == 'is_upcoming':
38d79fd1 1430 diff = random.randrange(min_wait, max_wait) if (max_wait and min_wait) else (max_wait or min_wait)
f2ebc5c7 1431 self.report_warning('Release time of video is not known')
1432 elif (diff or 0) <= 0:
1433 self.report_warning('Video should already be available according to extracted info')
38d79fd1 1434 diff = min(max(diff or 0, min_wait or 0), max_wait or float('inf'))
f2ebc5c7 1435 self.to_screen(f'[wait] Waiting for {format_dur(diff)} - Press Ctrl+C to try now')
1436
1437 wait_till = time.time() + diff
1438 try:
1439 while True:
1440 diff = wait_till - time.time()
1441 if diff <= 0:
1442 progress('')
1443 raise ReExtractInfo('[wait] Wait period ended', expected=True)
1444 progress(f'[wait] Remaining time until next attempt: {self._format_screen(format_dur(diff), self.Styles.EMPHASIS)}')
1445 time.sleep(1)
1446 except KeyboardInterrupt:
1447 progress('')
1448 raise ReExtractInfo('[wait] Interrupted by user', expected=True)
1449 except BaseException as e:
1450 if not isinstance(e, ReExtractInfo):
1451 self.to_screen('')
1452 raise
1453
a0566bbf 1454 @__handle_extraction_exceptions
58f197b7 1455 def __extract_info(self, url, ie, download, extra_info, process):
a0566bbf 1456 ie_result = ie.extract(url)
1457 if ie_result is None: # Finished already (backwards compatibility; listformats and friends should be moved here)
1458 return
1459 if isinstance(ie_result, list):
1460 # Backwards compatibility: old IE result format
1461 ie_result = {
1462 '_type': 'compat_list',
1463 'entries': ie_result,
1464 }
e37d0efb 1465 if extra_info.get('original_url'):
1466 ie_result.setdefault('original_url', extra_info['original_url'])
a0566bbf 1467 self.add_default_extra_info(ie_result, ie, url)
1468 if process:
f2ebc5c7 1469 self._wait_for_video(ie_result)
a0566bbf 1470 return self.process_ie_result(ie_result, download, extra_info)
8222d8de 1471 else:
a0566bbf 1472 return ie_result
fe7e0c98 1473
ea38e55f 1474 def add_default_extra_info(self, ie_result, ie, url):
6033d980 1475 if url is not None:
1476 self.add_extra_info(ie_result, {
1477 'webpage_url': url,
1478 'original_url': url,
57ebfca3 1479 })
1480 webpage_url = ie_result.get('webpage_url')
1481 if webpage_url:
1482 self.add_extra_info(ie_result, {
1483 'webpage_url_basename': url_basename(webpage_url),
1484 'webpage_url_domain': get_domain(webpage_url),
6033d980 1485 })
1486 if ie is not None:
1487 self.add_extra_info(ie_result, {
1488 'extractor': ie.IE_NAME,
1489 'extractor_key': ie.ie_key(),
1490 })
ea38e55f 1491
58adec46 1492 def process_ie_result(self, ie_result, download=True, extra_info=None):
8222d8de
JMF
1493 """
1494 Take the result of the ie(may be modified) and resolve all unresolved
1495 references (URLs, playlist items).
1496
1497 It will also download the videos if 'download'.
1498 Returns the resolved ie_result.
1499 """
58adec46 1500 if extra_info is None:
1501 extra_info = {}
e8ee972c
PH
1502 result_type = ie_result.get('_type', 'video')
1503
057a5206 1504 if result_type in ('url', 'url_transparent'):
134c6ea8 1505 ie_result['url'] = sanitize_url(ie_result['url'])
e37d0efb 1506 if ie_result.get('original_url'):
1507 extra_info.setdefault('original_url', ie_result['original_url'])
1508
057a5206 1509 extract_flat = self.params.get('extract_flat', False)
3089bc74
S
1510 if ((extract_flat == 'in_playlist' and 'playlist' in extra_info)
1511 or extract_flat is True):
ecb54191 1512 info_copy = ie_result.copy()
6033d980 1513 ie = try_get(ie_result.get('ie_key'), self.get_info_extractor)
360167b9 1514 if ie and not ie_result.get('id'):
4614bc22 1515 info_copy['id'] = ie.get_temp_id(ie_result['url'])
6033d980 1516 self.add_default_extra_info(info_copy, ie, ie_result['url'])
4614bc22 1517 self.add_extra_info(info_copy, extra_info)
b5475f11 1518 info_copy, _ = self.pre_process(info_copy)
ecb54191 1519 self.__forced_printings(info_copy, self.prepare_filename(info_copy), incomplete=True)
4614bc22 1520 if self.params.get('force_write_download_archive', False):
1521 self.record_download_archive(info_copy)
e8ee972c
PH
1522 return ie_result
1523
8222d8de 1524 if result_type == 'video':
b6c45014 1525 self.add_extra_info(ie_result, extra_info)
9c2b75b5 1526 ie_result = self.process_video_result(ie_result, download=download)
28b0eb0f 1527 additional_urls = (ie_result or {}).get('additional_urls')
9c2b75b5 1528 if additional_urls:
e9f4ccd1 1529 # TODO: Improve MetadataParserPP to allow setting a list
9c2b75b5 1530 if isinstance(additional_urls, compat_str):
1531 additional_urls = [additional_urls]
1532 self.to_screen(
1533 '[info] %s: %d additional URL(s) requested' % (ie_result['id'], len(additional_urls)))
1534 self.write_debug('Additional URLs: "%s"' % '", "'.join(additional_urls))
1535 ie_result['additional_entries'] = [
1536 self.extract_info(
b69fd25c 1537 url, download, extra_info=extra_info,
9c2b75b5 1538 force_generic_extractor=self.params.get('force_generic_extractor'))
1539 for url in additional_urls
1540 ]
1541 return ie_result
8222d8de
JMF
1542 elif result_type == 'url':
1543 # We have to add extra_info to the results because it may be
1544 # contained in a playlist
07cce701 1545 return self.extract_info(
1546 ie_result['url'], download,
1547 ie_key=ie_result.get('ie_key'),
1548 extra_info=extra_info)
7fc3fa05
PH
1549 elif result_type == 'url_transparent':
1550 # Use the information from the embedding page
1551 info = self.extract_info(
1552 ie_result['url'], ie_key=ie_result.get('ie_key'),
1553 extra_info=extra_info, download=False, process=False)
1554
1640eb09
S
1555 # extract_info may return None when ignoreerrors is enabled and
1556 # extraction failed with an error, don't crash and return early
1557 # in this case
1558 if not info:
1559 return info
1560
412c617d
PH
1561 force_properties = dict(
1562 (k, v) for k, v in ie_result.items() if v is not None)
0396806f 1563 for f in ('_type', 'url', 'id', 'extractor', 'extractor_key', 'ie_key'):
412c617d
PH
1564 if f in force_properties:
1565 del force_properties[f]
1566 new_result = info.copy()
1567 new_result.update(force_properties)
7fc3fa05 1568
0563f7ac
S
1569 # Extracted info may not be a video result (i.e.
1570 # info.get('_type', 'video') != video) but rather an url or
1571 # url_transparent. In such cases outer metadata (from ie_result)
1572 # should be propagated to inner one (info). For this to happen
1573 # _type of info should be overridden with url_transparent. This
067aa17e 1574 # fixes issue from https://github.com/ytdl-org/youtube-dl/pull/11163.
0563f7ac
S
1575 if new_result.get('_type') == 'url':
1576 new_result['_type'] = 'url_transparent'
7fc3fa05
PH
1577
1578 return self.process_ie_result(
1579 new_result, download=download, extra_info=extra_info)
40fcba5e 1580 elif result_type in ('playlist', 'multi_video'):
30a074c2 1581 # Protect from infinite recursion due to recursively nested playlists
1582 # (see https://github.com/ytdl-org/youtube-dl/issues/27833)
1583 webpage_url = ie_result['webpage_url']
1584 if webpage_url in self._playlist_urls:
7e85e872 1585 self.to_screen(
30a074c2 1586 '[download] Skipping already downloaded playlist: %s'
1587 % ie_result.get('title') or ie_result.get('id'))
1588 return
7e85e872 1589
30a074c2 1590 self._playlist_level += 1
1591 self._playlist_urls.add(webpage_url)
03f83004 1592 self._fill_common_fields(ie_result, False)
bc516a3f 1593 self._sanitize_thumbnails(ie_result)
30a074c2 1594 try:
1595 return self.__process_playlist(ie_result, download)
1596 finally:
1597 self._playlist_level -= 1
1598 if not self._playlist_level:
1599 self._playlist_urls.clear()
8222d8de 1600 elif result_type == 'compat_list':
c9bf4114
PH
1601 self.report_warning(
1602 'Extractor %s returned a compat_list result. '
1603 'It needs to be updated.' % ie_result.get('extractor'))
5f6a1245 1604
8222d8de 1605 def _fixup(r):
b868936c 1606 self.add_extra_info(r, {
1607 'extractor': ie_result['extractor'],
1608 'webpage_url': ie_result['webpage_url'],
1609 'webpage_url_basename': url_basename(ie_result['webpage_url']),
0bb322b9 1610 'webpage_url_domain': get_domain(ie_result['webpage_url']),
b868936c 1611 'extractor_key': ie_result['extractor_key'],
1612 })
8222d8de
JMF
1613 return r
1614 ie_result['entries'] = [
b6c45014 1615 self.process_ie_result(_fixup(r), download, extra_info)
8222d8de
JMF
1616 for r in ie_result['entries']
1617 ]
1618 return ie_result
1619 else:
1620 raise Exception('Invalid result type: %s' % result_type)
1621
e92caff5 1622 def _ensure_dir_exists(self, path):
1623 return make_dir(path, self.report_error)
1624
3b603dbd 1625 @staticmethod
1626 def _playlist_infodict(ie_result, **kwargs):
1627 return {
1628 **ie_result,
1629 'playlist': ie_result.get('title') or ie_result.get('id'),
1630 'playlist_id': ie_result.get('id'),
1631 'playlist_title': ie_result.get('title'),
1632 'playlist_uploader': ie_result.get('uploader'),
1633 'playlist_uploader_id': ie_result.get('uploader_id'),
1634 'playlist_index': 0,
1635 **kwargs,
1636 }
1637
30a074c2 1638 def __process_playlist(self, ie_result, download):
1639 # We process each entry in the playlist
1640 playlist = ie_result.get('title') or ie_result.get('id')
1641 self.to_screen('[download] Downloading playlist: %s' % playlist)
1642
498f5606 1643 if 'entries' not in ie_result:
aa9369a2 1644 raise EntryNotInPlaylist('There are no entries')
7c7f7161 1645
1646 MissingEntry = object()
498f5606 1647 incomplete_entries = bool(ie_result.get('requested_entries'))
1648 if incomplete_entries:
bf5f605e 1649 def fill_missing_entries(entries, indices):
7c7f7161 1650 ret = [MissingEntry] * max(indices)
bf5f605e 1651 for i, entry in zip(indices, entries):
498f5606 1652 ret[i - 1] = entry
1653 return ret
1654 ie_result['entries'] = fill_missing_entries(ie_result['entries'], ie_result['requested_entries'])
02fd60d3 1655
30a074c2 1656 playlist_results = []
1657
56a8fb4f 1658 playliststart = self.params.get('playliststart', 1)
30a074c2 1659 playlistend = self.params.get('playlistend')
1660 # For backwards compatibility, interpret -1 as whole list
1661 if playlistend == -1:
1662 playlistend = None
1663
1664 playlistitems_str = self.params.get('playlist_items')
1665 playlistitems = None
1666 if playlistitems_str is not None:
1667 def iter_playlistitems(format):
1668 for string_segment in format.split(','):
1669 if '-' in string_segment:
1670 start, end = string_segment.split('-')
1671 for item in range(int(start), int(end) + 1):
1672 yield int(item)
1673 else:
1674 yield int(string_segment)
1675 playlistitems = orderedSet(iter_playlistitems(playlistitems_str))
1676
1677 ie_entries = ie_result['entries']
8e5fecc8 1678 if isinstance(ie_entries, list):
ed8d87f9 1679 playlist_count = len(ie_entries)
f0d785d3 1680 msg = f'Collected {playlist_count} videos; downloading %d of them'
1681 ie_result['playlist_count'] = ie_result.get('playlist_count') or playlist_count
1682
8e5fecc8 1683 def get_entry(i):
1684 return ie_entries[i - 1]
1685 else:
f0d785d3 1686 msg = 'Downloading %d videos'
c586f9e8 1687 if not isinstance(ie_entries, (PagedList, LazyList)):
8e5fecc8 1688 ie_entries = LazyList(ie_entries)
d37707bd 1689 elif isinstance(ie_entries, InAdvancePagedList):
1690 if ie_entries._pagesize == 1:
1691 playlist_count = ie_entries._pagecount
8e5fecc8 1692
1693 def get_entry(i):
1694 return YoutubeDL.__handle_extraction_exceptions(
1695 lambda self, i: ie_entries[i - 1]
1696 )(self, i)
50fed816 1697
f0d785d3 1698 entries, broken = [], False
ff1c7fc9 1699 items = playlistitems if playlistitems is not None else itertools.count(playliststart)
1700 for i in items:
1701 if i == 0:
1702 continue
56a8fb4f 1703 if playlistitems is None and playlistend is not None and playlistend < i:
1704 break
1705 entry = None
1706 try:
50fed816 1707 entry = get_entry(i)
7c7f7161 1708 if entry is MissingEntry:
498f5606 1709 raise EntryNotInPlaylist()
56a8fb4f 1710 except (IndexError, EntryNotInPlaylist):
1711 if incomplete_entries:
aa9369a2 1712 raise EntryNotInPlaylist(f'Entry {i} cannot be found')
56a8fb4f 1713 elif not playlistitems:
1714 break
1715 entries.append(entry)
120fe513 1716 try:
1717 if entry is not None:
1718 self._match_entry(entry, incomplete=True, silent=True)
1719 except (ExistingVideoReached, RejectedVideoReached):
f0d785d3 1720 broken = True
120fe513 1721 break
56a8fb4f 1722 ie_result['entries'] = entries
30a074c2 1723
56a8fb4f 1724 # Save playlist_index before re-ordering
1725 entries = [
9e598870 1726 ((playlistitems[i - 1] if playlistitems else i + playliststart - 1), entry)
56a8fb4f 1727 for i, entry in enumerate(entries, 1)
1728 if entry is not None]
1729 n_entries = len(entries)
498f5606 1730
f0d785d3 1731 if not (ie_result.get('playlist_count') or broken or playlistitems or playlistend):
1732 ie_result['playlist_count'] = n_entries
1733
e08a85d8 1734 if not playlistitems and (playliststart != 1 or playlistend):
56a8fb4f 1735 playlistitems = list(range(playliststart, playliststart + n_entries))
498f5606 1736 ie_result['requested_entries'] = playlistitems
1737
e08a85d8 1738 _infojson_written = False
0bfc53d0 1739 write_playlist_files = self.params.get('allow_playlist_files', True)
1740 if write_playlist_files and self.params.get('list_thumbnails'):
1741 self.list_thumbnails(ie_result)
1742 if write_playlist_files and not self.params.get('simulate'):
3b603dbd 1743 ie_copy = self._playlist_infodict(ie_result, n_entries=n_entries)
e08a85d8 1744 _infojson_written = self._write_info_json(
1745 'playlist', ie_result, self.prepare_filename(ie_copy, 'pl_infojson'))
1746 if _infojson_written is None:
80c03fa9 1747 return
1748 if self._write_description('playlist', ie_result,
1749 self.prepare_filename(ie_copy, 'pl_description')) is None:
1750 return
681de68e 1751 # TODO: This should be passed to ThumbnailsConvertor if necessary
80c03fa9 1752 self._write_thumbnails('playlist', ie_copy, self.prepare_filename(ie_copy, 'pl_thumbnail'))
30a074c2 1753
1754 if self.params.get('playlistreverse', False):
1755 entries = entries[::-1]
30a074c2 1756 if self.params.get('playlistrandom', False):
1757 random.shuffle(entries)
1758
1759 x_forwarded_for = ie_result.get('__x_forwarded_for_ip')
1760
56a8fb4f 1761 self.to_screen('[%s] playlist %s: %s' % (ie_result['extractor'], playlist, msg % n_entries))
26e2805c 1762 failures = 0
1763 max_failures = self.params.get('skip_playlist_after_errors') or float('inf')
71729754 1764 for i, entry_tuple in enumerate(entries, 1):
1765 playlist_index, entry = entry_tuple
81139999 1766 if 'playlist-index' in self.params.get('compat_opts', []):
1767 playlist_index = playlistitems[i - 1] if playlistitems else i + playliststart - 1
30a074c2 1768 self.to_screen('[download] Downloading video %s of %s' % (i, n_entries))
1769 # This __x_forwarded_for_ip thing is a bit ugly but requires
1770 # minimal changes
1771 if x_forwarded_for:
1772 entry['__x_forwarded_for_ip'] = x_forwarded_for
1773 extra = {
1774 'n_entries': n_entries,
f59ae581 1775 '_last_playlist_index': max(playlistitems) if playlistitems else (playlistend or n_entries),
f0d785d3 1776 'playlist_count': ie_result.get('playlist_count'),
71729754 1777 'playlist_index': playlist_index,
1778 'playlist_autonumber': i,
30a074c2 1779 'playlist': playlist,
1780 'playlist_id': ie_result.get('id'),
1781 'playlist_title': ie_result.get('title'),
1782 'playlist_uploader': ie_result.get('uploader'),
1783 'playlist_uploader_id': ie_result.get('uploader_id'),
30a074c2 1784 'extractor': ie_result['extractor'],
1785 'webpage_url': ie_result['webpage_url'],
1786 'webpage_url_basename': url_basename(ie_result['webpage_url']),
0bb322b9 1787 'webpage_url_domain': get_domain(ie_result['webpage_url']),
30a074c2 1788 'extractor_key': ie_result['extractor_key'],
1789 }
1790
1791 if self._match_entry(entry, incomplete=True) is not None:
1792 continue
1793
1794 entry_result = self.__process_iterable_entry(entry, download, extra)
26e2805c 1795 if not entry_result:
1796 failures += 1
1797 if failures >= max_failures:
1798 self.report_error(
1799 'Skipping the remaining entries in playlist "%s" since %d items failed extraction' % (playlist, failures))
1800 break
30a074c2 1801 playlist_results.append(entry_result)
1802 ie_result['entries'] = playlist_results
e08a85d8 1803
1804 # Write the updated info to json
1805 if _infojson_written and self._write_info_json(
1806 'updated playlist', ie_result,
1807 self.prepare_filename(ie_copy, 'pl_infojson'), overwrite=True) is None:
1808 return
ca30f449 1809
ed5835b4 1810 ie_result = self.run_all_pps('playlist', ie_result)
1811 self.to_screen(f'[download] Finished downloading playlist: {playlist}')
30a074c2 1812 return ie_result
1813
a0566bbf 1814 @__handle_extraction_exceptions
1815 def __process_iterable_entry(self, entry, download, extra_info):
1816 return self.process_ie_result(
1817 entry, download=download, extra_info=extra_info)
1818
67134eab
JMF
1819 def _build_format_filter(self, filter_spec):
1820 " Returns a function to filter the formats according to the filter_spec "
083c9df9
PH
1821
1822 OPERATORS = {
1823 '<': operator.lt,
1824 '<=': operator.le,
1825 '>': operator.gt,
1826 '>=': operator.ge,
1827 '=': operator.eq,
1828 '!=': operator.ne,
1829 }
67134eab 1830 operator_rex = re.compile(r'''(?x)\s*
187986a8 1831 (?P<key>width|height|tbr|abr|vbr|asr|filesize|filesize_approx|fps)\s*
1832 (?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
1833 (?P<value>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?)\s*
083c9df9 1834 ''' % '|'.join(map(re.escape, OPERATORS.keys())))
187986a8 1835 m = operator_rex.fullmatch(filter_spec)
9ddb6925
S
1836 if m:
1837 try:
1838 comparison_value = int(m.group('value'))
1839 except ValueError:
1840 comparison_value = parse_filesize(m.group('value'))
1841 if comparison_value is None:
1842 comparison_value = parse_filesize(m.group('value') + 'B')
1843 if comparison_value is None:
1844 raise ValueError(
1845 'Invalid value %r in format specification %r' % (
67134eab 1846 m.group('value'), filter_spec))
9ddb6925
S
1847 op = OPERATORS[m.group('op')]
1848
083c9df9 1849 if not m:
9ddb6925
S
1850 STR_OPERATORS = {
1851 '=': operator.eq,
10d33b34
YCH
1852 '^=': lambda attr, value: attr.startswith(value),
1853 '$=': lambda attr, value: attr.endswith(value),
1854 '*=': lambda attr, value: value in attr,
1ce9a3cb 1855 '~=': lambda attr, value: value.search(attr) is not None
9ddb6925 1856 }
187986a8 1857 str_operator_rex = re.compile(r'''(?x)\s*
1858 (?P<key>[a-zA-Z0-9._-]+)\s*
1ce9a3cb
LF
1859 (?P<negation>!\s*)?(?P<op>%s)\s*(?P<none_inclusive>\?\s*)?
1860 (?P<quote>["'])?
1861 (?P<value>(?(quote)(?:(?!(?P=quote))[^\\]|\\.)+|[\w.-]+))
1862 (?(quote)(?P=quote))\s*
9ddb6925 1863 ''' % '|'.join(map(re.escape, STR_OPERATORS.keys())))
187986a8 1864 m = str_operator_rex.fullmatch(filter_spec)
9ddb6925 1865 if m:
1ce9a3cb
LF
1866 if m.group('op') == '~=':
1867 comparison_value = re.compile(m.group('value'))
1868 else:
1869 comparison_value = re.sub(r'''\\([\\"'])''', r'\1', m.group('value'))
2cc779f4
S
1870 str_op = STR_OPERATORS[m.group('op')]
1871 if m.group('negation'):
e118a879 1872 op = lambda attr, value: not str_op(attr, value)
2cc779f4
S
1873 else:
1874 op = str_op
083c9df9 1875
9ddb6925 1876 if not m:
187986a8 1877 raise SyntaxError('Invalid filter specification %r' % filter_spec)
083c9df9
PH
1878
1879 def _filter(f):
1880 actual_value = f.get(m.group('key'))
1881 if actual_value is None:
1882 return m.group('none_inclusive')
1883 return op(actual_value, comparison_value)
67134eab
JMF
1884 return _filter
1885
9f1a1c36 1886 def _check_formats(self, formats):
1887 for f in formats:
1888 self.to_screen('[info] Testing format %s' % f['format_id'])
75689fe5 1889 path = self.get_output_path('temp')
1890 if not self._ensure_dir_exists(f'{path}/'):
1891 continue
1892 temp_file = tempfile.NamedTemporaryFile(suffix='.tmp', delete=False, dir=path or None)
9f1a1c36 1893 temp_file.close()
1894 try:
1895 success, _ = self.dl(temp_file.name, f, test=True)
1896 except (DownloadError, IOError, OSError, ValueError) + network_exceptions:
1897 success = False
1898 finally:
1899 if os.path.exists(temp_file.name):
1900 try:
1901 os.remove(temp_file.name)
1902 except OSError:
1903 self.report_warning('Unable to delete temporary file "%s"' % temp_file.name)
1904 if success:
1905 yield f
1906 else:
1907 self.to_screen('[info] Unable to download format %s. Skipping...' % f['format_id'])
1908
0017d9ad 1909 def _default_format_spec(self, info_dict, download=True):
0017d9ad 1910
af0f7428
S
1911 def can_merge():
1912 merger = FFmpegMergerPP(self)
1913 return merger.available and merger.can_merge()
1914
91ebc640 1915 prefer_best = (
b7b04c78 1916 not self.params.get('simulate')
91ebc640 1917 and download
1918 and (
1919 not can_merge()
19807826 1920 or info_dict.get('is_live', False)
de6000d9 1921 or self.outtmpl_dict['default'] == '-'))
53ed7066 1922 compat = (
1923 prefer_best
1924 or self.params.get('allow_multiple_audio_streams', False)
1925 or 'format-spec' in self.params.get('compat_opts', []))
91ebc640 1926
1927 return (
53ed7066 1928 'best/bestvideo+bestaudio' if prefer_best
1929 else 'bestvideo*+bestaudio/best' if not compat
91ebc640 1930 else 'bestvideo+bestaudio/best')
0017d9ad 1931
67134eab
JMF
1932 def build_format_selector(self, format_spec):
1933 def syntax_error(note, start):
1934 message = (
1935 'Invalid format specification: '
1936 '{0}\n\t{1}\n\t{2}^'.format(note, format_spec, ' ' * start[1]))
1937 return SyntaxError(message)
1938
1939 PICKFIRST = 'PICKFIRST'
1940 MERGE = 'MERGE'
1941 SINGLE = 'SINGLE'
0130afb7 1942 GROUP = 'GROUP'
67134eab
JMF
1943 FormatSelector = collections.namedtuple('FormatSelector', ['type', 'selector', 'filters'])
1944
91ebc640 1945 allow_multiple_streams = {'audio': self.params.get('allow_multiple_audio_streams', False),
1946 'video': self.params.get('allow_multiple_video_streams', False)}
909d24dd 1947
9f1a1c36 1948 check_formats = self.params.get('check_formats') == 'selected'
e8e73840 1949
67134eab
JMF
1950 def _parse_filter(tokens):
1951 filter_parts = []
1952 for type, string, start, _, _ in tokens:
1953 if type == tokenize.OP and string == ']':
1954 return ''.join(filter_parts)
1955 else:
1956 filter_parts.append(string)
1957
232541df 1958 def _remove_unused_ops(tokens):
17cc1534 1959 # Remove operators that we don't use and join them with the surrounding strings
232541df
JMF
1960 # for example: 'mp4' '-' 'baseline' '-' '16x9' is converted to 'mp4-baseline-16x9'
1961 ALLOWED_OPS = ('/', '+', ',', '(', ')')
1962 last_string, last_start, last_end, last_line = None, None, None, None
1963 for type, string, start, end, line in tokens:
1964 if type == tokenize.OP and string == '[':
1965 if last_string:
1966 yield tokenize.NAME, last_string, last_start, last_end, last_line
1967 last_string = None
1968 yield type, string, start, end, line
1969 # everything inside brackets will be handled by _parse_filter
1970 for type, string, start, end, line in tokens:
1971 yield type, string, start, end, line
1972 if type == tokenize.OP and string == ']':
1973 break
1974 elif type == tokenize.OP and string in ALLOWED_OPS:
1975 if last_string:
1976 yield tokenize.NAME, last_string, last_start, last_end, last_line
1977 last_string = None
1978 yield type, string, start, end, line
1979 elif type in [tokenize.NAME, tokenize.NUMBER, tokenize.OP]:
1980 if not last_string:
1981 last_string = string
1982 last_start = start
1983 last_end = end
1984 else:
1985 last_string += string
1986 if last_string:
1987 yield tokenize.NAME, last_string, last_start, last_end, last_line
1988
cf2ac6df 1989 def _parse_format_selection(tokens, inside_merge=False, inside_choice=False, inside_group=False):
67134eab
JMF
1990 selectors = []
1991 current_selector = None
1992 for type, string, start, _, _ in tokens:
1993 # ENCODING is only defined in python 3.x
1994 if type == getattr(tokenize, 'ENCODING', None):
1995 continue
1996 elif type in [tokenize.NAME, tokenize.NUMBER]:
1997 current_selector = FormatSelector(SINGLE, string, [])
1998 elif type == tokenize.OP:
cf2ac6df
JMF
1999 if string == ')':
2000 if not inside_group:
2001 # ')' will be handled by the parentheses group
2002 tokens.restore_last_token()
67134eab 2003 break
cf2ac6df 2004 elif inside_merge and string in ['/', ',']:
0130afb7
JMF
2005 tokens.restore_last_token()
2006 break
cf2ac6df
JMF
2007 elif inside_choice and string == ',':
2008 tokens.restore_last_token()
2009 break
2010 elif string == ',':
0a31a350
JMF
2011 if not current_selector:
2012 raise syntax_error('"," must follow a format selector', start)
67134eab
JMF
2013 selectors.append(current_selector)
2014 current_selector = None
2015 elif string == '/':
d96d604e
JMF
2016 if not current_selector:
2017 raise syntax_error('"/" must follow a format selector', start)
67134eab 2018 first_choice = current_selector
cf2ac6df 2019 second_choice = _parse_format_selection(tokens, inside_choice=True)
f5f4a27a 2020 current_selector = FormatSelector(PICKFIRST, (first_choice, second_choice), [])
67134eab
JMF
2021 elif string == '[':
2022 if not current_selector:
2023 current_selector = FormatSelector(SINGLE, 'best', [])
2024 format_filter = _parse_filter(tokens)
2025 current_selector.filters.append(format_filter)
0130afb7
JMF
2026 elif string == '(':
2027 if current_selector:
2028 raise syntax_error('Unexpected "("', start)
cf2ac6df
JMF
2029 group = _parse_format_selection(tokens, inside_group=True)
2030 current_selector = FormatSelector(GROUP, group, [])
67134eab 2031 elif string == '+':
d03cfdce 2032 if not current_selector:
2033 raise syntax_error('Unexpected "+"', start)
2034 selector_1 = current_selector
2035 selector_2 = _parse_format_selection(tokens, inside_merge=True)
2036 if not selector_2:
2037 raise syntax_error('Expected a selector', start)
2038 current_selector = FormatSelector(MERGE, (selector_1, selector_2), [])
67134eab
JMF
2039 else:
2040 raise syntax_error('Operator not recognized: "{0}"'.format(string), start)
2041 elif type == tokenize.ENDMARKER:
2042 break
2043 if current_selector:
2044 selectors.append(current_selector)
2045 return selectors
2046
f8d4ad9a 2047 def _merge(formats_pair):
2048 format_1, format_2 = formats_pair
2049
2050 formats_info = []
2051 formats_info.extend(format_1.get('requested_formats', (format_1,)))
2052 formats_info.extend(format_2.get('requested_formats', (format_2,)))
2053
2054 if not allow_multiple_streams['video'] or not allow_multiple_streams['audio']:
551f9388 2055 get_no_more = {'video': False, 'audio': False}
f8d4ad9a 2056 for (i, fmt_info) in enumerate(formats_info):
551f9388 2057 if fmt_info.get('acodec') == fmt_info.get('vcodec') == 'none':
2058 formats_info.pop(i)
2059 continue
2060 for aud_vid in ['audio', 'video']:
f8d4ad9a 2061 if not allow_multiple_streams[aud_vid] and fmt_info.get(aud_vid[0] + 'codec') != 'none':
2062 if get_no_more[aud_vid]:
2063 formats_info.pop(i)
f5510afe 2064 break
f8d4ad9a 2065 get_no_more[aud_vid] = True
2066
2067 if len(formats_info) == 1:
2068 return formats_info[0]
2069
2070 video_fmts = [fmt_info for fmt_info in formats_info if fmt_info.get('vcodec') != 'none']
2071 audio_fmts = [fmt_info for fmt_info in formats_info if fmt_info.get('acodec') != 'none']
2072
2073 the_only_video = video_fmts[0] if len(video_fmts) == 1 else None
2074 the_only_audio = audio_fmts[0] if len(audio_fmts) == 1 else None
2075
2076 output_ext = self.params.get('merge_output_format')
2077 if not output_ext:
2078 if the_only_video:
2079 output_ext = the_only_video['ext']
2080 elif the_only_audio and not video_fmts:
2081 output_ext = the_only_audio['ext']
2082 else:
2083 output_ext = 'mkv'
2084
975a0d0d 2085 filtered = lambda *keys: filter(None, (traverse_obj(fmt, *keys) for fmt in formats_info))
2086
f8d4ad9a 2087 new_dict = {
2088 'requested_formats': formats_info,
975a0d0d 2089 'format': '+'.join(filtered('format')),
2090 'format_id': '+'.join(filtered('format_id')),
f8d4ad9a 2091 'ext': output_ext,
975a0d0d 2092 'protocol': '+'.join(map(determine_protocol, formats_info)),
093a1710 2093 'language': '+'.join(orderedSet(filtered('language'))) or None,
2094 'format_note': '+'.join(orderedSet(filtered('format_note'))) or None,
2095 'filesize_approx': sum(filtered('filesize', 'filesize_approx')) or None,
975a0d0d 2096 'tbr': sum(filtered('tbr', 'vbr', 'abr')),
f8d4ad9a 2097 }
2098
2099 if the_only_video:
2100 new_dict.update({
2101 'width': the_only_video.get('width'),
2102 'height': the_only_video.get('height'),
2103 'resolution': the_only_video.get('resolution') or self.format_resolution(the_only_video),
2104 'fps': the_only_video.get('fps'),
49a57e70 2105 'dynamic_range': the_only_video.get('dynamic_range'),
f8d4ad9a 2106 'vcodec': the_only_video.get('vcodec'),
2107 'vbr': the_only_video.get('vbr'),
2108 'stretched_ratio': the_only_video.get('stretched_ratio'),
2109 })
2110
2111 if the_only_audio:
2112 new_dict.update({
2113 'acodec': the_only_audio.get('acodec'),
2114 'abr': the_only_audio.get('abr'),
975a0d0d 2115 'asr': the_only_audio.get('asr'),
f8d4ad9a 2116 })
2117
2118 return new_dict
2119
e8e73840 2120 def _check_formats(formats):
981052c9 2121 if not check_formats:
2122 yield from formats
b5ac45b1 2123 return
9f1a1c36 2124 yield from self._check_formats(formats)
e8e73840 2125
67134eab 2126 def _build_selector_function(selector):
909d24dd 2127 if isinstance(selector, list): # ,
67134eab
JMF
2128 fs = [_build_selector_function(s) for s in selector]
2129
317f7ab6 2130 def selector_function(ctx):
67134eab 2131 for f in fs:
981052c9 2132 yield from f(ctx)
67134eab 2133 return selector_function
909d24dd 2134
2135 elif selector.type == GROUP: # ()
0130afb7 2136 selector_function = _build_selector_function(selector.selector)
909d24dd 2137
2138 elif selector.type == PICKFIRST: # /
67134eab
JMF
2139 fs = [_build_selector_function(s) for s in selector.selector]
2140
317f7ab6 2141 def selector_function(ctx):
67134eab 2142 for f in fs:
317f7ab6 2143 picked_formats = list(f(ctx))
67134eab
JMF
2144 if picked_formats:
2145 return picked_formats
2146 return []
67134eab 2147
981052c9 2148 elif selector.type == MERGE: # +
2149 selector_1, selector_2 = map(_build_selector_function, selector.selector)
2150
2151 def selector_function(ctx):
adbc4ec4 2152 for pair in itertools.product(selector_1(ctx), selector_2(ctx)):
981052c9 2153 yield _merge(pair)
2154
909d24dd 2155 elif selector.type == SINGLE: # atom
598d185d 2156 format_spec = selector.selector or 'best'
909d24dd 2157
f8d4ad9a 2158 # TODO: Add allvideo, allaudio etc by generalizing the code with best/worst selector
909d24dd 2159 if format_spec == 'all':
2160 def selector_function(ctx):
9222c381 2161 yield from _check_formats(ctx['formats'][::-1])
f8d4ad9a 2162 elif format_spec == 'mergeall':
2163 def selector_function(ctx):
dd2a987d 2164 formats = list(_check_formats(ctx['formats']))
e01d6aa4 2165 if not formats:
2166 return
921b76ca 2167 merged_format = formats[-1]
2168 for f in formats[-2::-1]:
f8d4ad9a 2169 merged_format = _merge((merged_format, f))
2170 yield merged_format
909d24dd 2171
2172 else:
e8e73840 2173 format_fallback, format_reverse, format_idx = False, True, 1
eff63539 2174 mobj = re.match(
2175 r'(?P<bw>best|worst|b|w)(?P<type>video|audio|v|a)?(?P<mod>\*)?(?:\.(?P<n>[1-9]\d*))?$',
2176 format_spec)
2177 if mobj is not None:
2178 format_idx = int_or_none(mobj.group('n'), default=1)
e8e73840 2179 format_reverse = mobj.group('bw')[0] == 'b'
eff63539 2180 format_type = (mobj.group('type') or [None])[0]
2181 not_format_type = {'v': 'a', 'a': 'v'}.get(format_type)
2182 format_modified = mobj.group('mod') is not None
909d24dd 2183
2184 format_fallback = not format_type and not format_modified # for b, w
8326b00a 2185 _filter_f = (
eff63539 2186 (lambda f: f.get('%scodec' % format_type) != 'none')
2187 if format_type and format_modified # bv*, ba*, wv*, wa*
2188 else (lambda f: f.get('%scodec' % not_format_type) == 'none')
2189 if format_type # bv, ba, wv, wa
2190 else (lambda f: f.get('vcodec') != 'none' and f.get('acodec') != 'none')
2191 if not format_modified # b, w
8326b00a 2192 else lambda f: True) # b*, w*
2193 filter_f = lambda f: _filter_f(f) and (
2194 f.get('vcodec') != 'none' or f.get('acodec') != 'none')
67134eab 2195 else:
48ee10ee 2196 if format_spec in self._format_selection_exts['audio']:
b11c04a8 2197 filter_f = lambda f: f.get('ext') == format_spec and f.get('acodec') != 'none'
48ee10ee 2198 elif format_spec in self._format_selection_exts['video']:
b11c04a8 2199 filter_f = lambda f: f.get('ext') == format_spec and f.get('acodec') != 'none' and f.get('vcodec') != 'none'
48ee10ee 2200 elif format_spec in self._format_selection_exts['storyboards']:
b11c04a8 2201 filter_f = lambda f: f.get('ext') == format_spec and f.get('acodec') == 'none' and f.get('vcodec') == 'none'
2202 else:
b5ae35ee 2203 filter_f = lambda f: f.get('format_id') == format_spec # id
909d24dd 2204
2205 def selector_function(ctx):
2206 formats = list(ctx['formats'])
909d24dd 2207 matches = list(filter(filter_f, formats)) if filter_f is not None else formats
e8e73840 2208 if format_fallback and ctx['incomplete_formats'] and not matches:
909d24dd 2209 # for extractors with incomplete formats (audio only (soundcloud)
2210 # or video only (imgur)) best/worst will fallback to
2211 # best/worst {video,audio}-only format
e8e73840 2212 matches = formats
981052c9 2213 matches = LazyList(_check_formats(matches[::-1 if format_reverse else 1]))
2214 try:
e8e73840 2215 yield matches[format_idx - 1]
981052c9 2216 except IndexError:
2217 return
083c9df9 2218
67134eab 2219 filters = [self._build_format_filter(f) for f in selector.filters]
083c9df9 2220
317f7ab6 2221 def final_selector(ctx):
adbc4ec4 2222 ctx_copy = dict(ctx)
67134eab 2223 for _filter in filters:
317f7ab6
S
2224 ctx_copy['formats'] = list(filter(_filter, ctx_copy['formats']))
2225 return selector_function(ctx_copy)
67134eab 2226 return final_selector
083c9df9 2227
67134eab 2228 stream = io.BytesIO(format_spec.encode('utf-8'))
0130afb7 2229 try:
232541df 2230 tokens = list(_remove_unused_ops(compat_tokenize_tokenize(stream.readline)))
0130afb7
JMF
2231 except tokenize.TokenError:
2232 raise syntax_error('Missing closing/opening brackets or parenthesis', (0, len(format_spec)))
2233
2234 class TokenIterator(object):
2235 def __init__(self, tokens):
2236 self.tokens = tokens
2237 self.counter = 0
2238
2239 def __iter__(self):
2240 return self
2241
2242 def __next__(self):
2243 if self.counter >= len(self.tokens):
2244 raise StopIteration()
2245 value = self.tokens[self.counter]
2246 self.counter += 1
2247 return value
2248
2249 next = __next__
2250
2251 def restore_last_token(self):
2252 self.counter -= 1
2253
2254 parsed_selector = _parse_format_selection(iter(TokenIterator(tokens)))
67134eab 2255 return _build_selector_function(parsed_selector)
a9c58ad9 2256
e5660ee6 2257 def _calc_headers(self, info_dict):
8b7539d2 2258 res = merge_headers(self.params['http_headers'], info_dict.get('http_headers') or {})
e5660ee6
JMF
2259
2260 cookies = self._calc_cookies(info_dict)
2261 if cookies:
2262 res['Cookie'] = cookies
2263
0016b84e
S
2264 if 'X-Forwarded-For' not in res:
2265 x_forwarded_for_ip = info_dict.get('__x_forwarded_for_ip')
2266 if x_forwarded_for_ip:
2267 res['X-Forwarded-For'] = x_forwarded_for_ip
2268
e5660ee6
JMF
2269 return res
2270
2271 def _calc_cookies(self, info_dict):
5c2266df 2272 pr = sanitized_Request(info_dict['url'])
e5660ee6 2273 self.cookiejar.add_cookie_header(pr)
662435f7 2274 return pr.get_header('Cookie')
e5660ee6 2275
9f1a1c36 2276 def _sort_thumbnails(self, thumbnails):
2277 thumbnails.sort(key=lambda t: (
2278 t.get('preference') if t.get('preference') is not None else -1,
2279 t.get('width') if t.get('width') is not None else -1,
2280 t.get('height') if t.get('height') is not None else -1,
2281 t.get('id') if t.get('id') is not None else '',
2282 t.get('url')))
2283
b0249bca 2284 def _sanitize_thumbnails(self, info_dict):
bc516a3f 2285 thumbnails = info_dict.get('thumbnails')
2286 if thumbnails is None:
2287 thumbnail = info_dict.get('thumbnail')
2288 if thumbnail:
2289 info_dict['thumbnails'] = thumbnails = [{'url': thumbnail}]
9f1a1c36 2290 if not thumbnails:
2291 return
2292
2293 def check_thumbnails(thumbnails):
2294 for t in thumbnails:
2295 self.to_screen(f'[info] Testing thumbnail {t["id"]}')
2296 try:
2297 self.urlopen(HEADRequest(t['url']))
2298 except network_exceptions as err:
2299 self.to_screen(f'[info] Unable to connect to thumbnail {t["id"]} URL {t["url"]!r} - {err}. Skipping...')
2300 continue
2301 yield t
2302
2303 self._sort_thumbnails(thumbnails)
2304 for i, t in enumerate(thumbnails):
2305 if t.get('id') is None:
2306 t['id'] = '%d' % i
2307 if t.get('width') and t.get('height'):
2308 t['resolution'] = '%dx%d' % (t['width'], t['height'])
2309 t['url'] = sanitize_url(t['url'])
2310
2311 if self.params.get('check_formats') is True:
282f5709 2312 info_dict['thumbnails'] = LazyList(check_thumbnails(thumbnails[::-1]), reverse=True)
9f1a1c36 2313 else:
2314 info_dict['thumbnails'] = thumbnails
bc516a3f 2315
03f83004
LNO
2316 def _fill_common_fields(self, info_dict, is_video=True):
2317 # TODO: move sanitization here
2318 if is_video:
2319 # playlists are allowed to lack "title"
2320 info_dict['fulltitle'] = info_dict.get('title')
2321 if 'title' not in info_dict:
2322 raise ExtractorError('Missing "title" field in extractor result',
2323 video_id=info_dict['id'], ie=info_dict['extractor'])
2324 elif not info_dict.get('title'):
2325 self.report_warning('Extractor failed to obtain "title". Creating a generic title instead')
2326 info_dict['title'] = f'{info_dict["extractor"]} video #{info_dict["id"]}'
2327
2328 if info_dict.get('duration') is not None:
2329 info_dict['duration_string'] = formatSeconds(info_dict['duration'])
2330
2331 for ts_key, date_key in (
2332 ('timestamp', 'upload_date'),
2333 ('release_timestamp', 'release_date'),
2334 ('modified_timestamp', 'modified_date'),
2335 ):
2336 if info_dict.get(date_key) is None and info_dict.get(ts_key) is not None:
2337 # Working around out-of-range timestamp values (e.g. negative ones on Windows,
2338 # see http://bugs.python.org/issue1646728)
2339 try:
2340 upload_date = datetime.datetime.utcfromtimestamp(info_dict[ts_key])
2341 info_dict[date_key] = upload_date.strftime('%Y%m%d')
2342 except (ValueError, OverflowError, OSError):
2343 pass
2344
2345 live_keys = ('is_live', 'was_live')
2346 live_status = info_dict.get('live_status')
2347 if live_status is None:
2348 for key in live_keys:
2349 if info_dict.get(key) is False:
2350 continue
2351 if info_dict.get(key):
2352 live_status = key
2353 break
2354 if all(info_dict.get(key) is False for key in live_keys):
2355 live_status = 'not_live'
2356 if live_status:
2357 info_dict['live_status'] = live_status
2358 for key in live_keys:
2359 if info_dict.get(key) is None:
2360 info_dict[key] = (live_status == key)
2361
2362 # Auto generate title fields corresponding to the *_number fields when missing
2363 # in order to always have clean titles. This is very common for TV series.
2364 for field in ('chapter', 'season', 'episode'):
2365 if info_dict.get('%s_number' % field) is not None and not info_dict.get(field):
2366 info_dict[field] = '%s %d' % (field.capitalize(), info_dict['%s_number' % field])
2367
dd82ffea
JMF
2368 def process_video_result(self, info_dict, download=True):
2369 assert info_dict.get('_type', 'video') == 'video'
9c906919 2370 self._num_videos += 1
dd82ffea 2371
bec1fad2 2372 if 'id' not in info_dict:
fc08bdd6 2373 raise ExtractorError('Missing "id" field in extractor result', ie=info_dict['extractor'])
2374 elif not info_dict.get('id'):
2375 raise ExtractorError('Extractor failed to obtain "id"', ie=info_dict['extractor'])
455a15e2 2376
c9969434
S
2377 def report_force_conversion(field, field_not, conversion):
2378 self.report_warning(
2379 '"%s" field is not %s - forcing %s conversion, there is an error in extractor'
2380 % (field, field_not, conversion))
2381
2382 def sanitize_string_field(info, string_field):
2383 field = info.get(string_field)
2384 if field is None or isinstance(field, compat_str):
2385 return
2386 report_force_conversion(string_field, 'a string', 'string')
2387 info[string_field] = compat_str(field)
2388
2389 def sanitize_numeric_fields(info):
2390 for numeric_field in self._NUMERIC_FIELDS:
2391 field = info.get(numeric_field)
2392 if field is None or isinstance(field, compat_numeric_types):
2393 continue
2394 report_force_conversion(numeric_field, 'numeric', 'int')
2395 info[numeric_field] = int_or_none(field)
2396
2397 sanitize_string_field(info_dict, 'id')
2398 sanitize_numeric_fields(info_dict)
4c3f8c3f 2399 if (info_dict.get('duration') or 0) <= 0 and info_dict.pop('duration', None):
50e93e03 2400 self.report_warning('"duration" field is negative, there is an error in extractor')
be6217b2 2401
dd82ffea
JMF
2402 if 'playlist' not in info_dict:
2403 # It isn't part of a playlist
2404 info_dict['playlist'] = None
2405 info_dict['playlist_index'] = None
2406
bc516a3f 2407 self._sanitize_thumbnails(info_dict)
d5519808 2408
536a55da 2409 thumbnail = info_dict.get('thumbnail')
bc516a3f 2410 thumbnails = info_dict.get('thumbnails')
536a55da
S
2411 if thumbnail:
2412 info_dict['thumbnail'] = sanitize_url(thumbnail)
2413 elif thumbnails:
d5519808
PH
2414 info_dict['thumbnail'] = thumbnails[-1]['url']
2415
ae30b840 2416 if info_dict.get('display_id') is None and 'id' in info_dict:
0afef30b
PH
2417 info_dict['display_id'] = info_dict['id']
2418
03f83004 2419 self._fill_common_fields(info_dict)
33d2fc2f 2420
05108a49
S
2421 for cc_kind in ('subtitles', 'automatic_captions'):
2422 cc = info_dict.get(cc_kind)
2423 if cc:
2424 for _, subtitle in cc.items():
2425 for subtitle_format in subtitle:
2426 if subtitle_format.get('url'):
2427 subtitle_format['url'] = sanitize_url(subtitle_format['url'])
2428 if subtitle_format.get('ext') is None:
2429 subtitle_format['ext'] = determine_ext(subtitle_format['url']).lower()
2430
2431 automatic_captions = info_dict.get('automatic_captions')
4bba3716 2432 subtitles = info_dict.get('subtitles')
4bba3716 2433
360e1ca5 2434 info_dict['requested_subtitles'] = self.process_subtitles(
05108a49 2435 info_dict['id'], subtitles, automatic_captions)
a504ced0 2436
dd82ffea
JMF
2437 if info_dict.get('formats') is None:
2438 # There's only one format available
2439 formats = [info_dict]
2440 else:
2441 formats = info_dict['formats']
2442
e0493e90 2443 info_dict['__has_drm'] = any(f.get('has_drm') for f in formats)
88acdbc2 2444 if not self.params.get('allow_unplayable_formats'):
2445 formats = [f for f in formats if not f.get('has_drm')]
88acdbc2 2446
319b6059 2447 get_from_start = not info_dict.get('is_live') or bool(self.params.get('live_from_start'))
2448 if not get_from_start:
2449 info_dict['title'] += ' ' + datetime.datetime.now().strftime('%Y-%m-%d %H:%M')
2450 if info_dict.get('is_live') and formats:
adbc4ec4 2451 formats = [f for f in formats if bool(f.get('is_from_start')) == get_from_start]
319b6059 2452 if get_from_start and not formats:
2453 self.raise_no_formats(info_dict, msg='--live-from-start is passed, but there are no formats that can be downloaded from the start. '
2454 'If you want to download from the current time, pass --no-live-from-start')
adbc4ec4 2455
db95dc13 2456 if not formats:
1151c407 2457 self.raise_no_formats(info_dict)
db95dc13 2458
73af5cc8
S
2459 def is_wellformed(f):
2460 url = f.get('url')
a5ac0c47 2461 if not url:
73af5cc8
S
2462 self.report_warning(
2463 '"url" field is missing or empty - skipping format, '
2464 'there is an error in extractor')
a5ac0c47
S
2465 return False
2466 if isinstance(url, bytes):
2467 sanitize_string_field(f, 'url')
2468 return True
73af5cc8
S
2469
2470 # Filter out malformed formats for better extraction robustness
2471 formats = list(filter(is_wellformed, formats))
2472
181c7053
S
2473 formats_dict = {}
2474
dd82ffea 2475 # We check that all the formats have the format and format_id fields
db95dc13 2476 for i, format in enumerate(formats):
c9969434
S
2477 sanitize_string_field(format, 'format_id')
2478 sanitize_numeric_fields(format)
dcf77cf1 2479 format['url'] = sanitize_url(format['url'])
e74e3b63 2480 if not format.get('format_id'):
8016c922 2481 format['format_id'] = compat_str(i)
e2effb08
S
2482 else:
2483 # Sanitize format_id from characters used in format selector expression
ec85ded8 2484 format['format_id'] = re.sub(r'[\s,/+\[\]()]', '_', format['format_id'])
181c7053
S
2485 format_id = format['format_id']
2486 if format_id not in formats_dict:
2487 formats_dict[format_id] = []
2488 formats_dict[format_id].append(format)
2489
2490 # Make sure all formats have unique format_id
03b4de72 2491 common_exts = set(itertools.chain(*self._format_selection_exts.values()))
181c7053 2492 for format_id, ambiguous_formats in formats_dict.items():
48ee10ee 2493 ambigious_id = len(ambiguous_formats) > 1
2494 for i, format in enumerate(ambiguous_formats):
2495 if ambigious_id:
181c7053 2496 format['format_id'] = '%s-%d' % (format_id, i)
48ee10ee 2497 if format.get('ext') is None:
2498 format['ext'] = determine_ext(format['url']).lower()
2499 # Ensure there is no conflict between id and ext in format selection
2500 # See https://github.com/yt-dlp/yt-dlp/issues/1282
2501 if format['format_id'] != format['ext'] and format['format_id'] in common_exts:
2502 format['format_id'] = 'f%s' % format['format_id']
181c7053
S
2503
2504 for i, format in enumerate(formats):
8c51aa65 2505 if format.get('format') is None:
6febd1c1 2506 format['format'] = '{id} - {res}{note}'.format(
8c51aa65
JMF
2507 id=format['format_id'],
2508 res=self.format_resolution(format),
b868936c 2509 note=format_field(format, 'format_note', ' (%s)'),
8c51aa65 2510 )
6f0be937 2511 if format.get('protocol') is None:
b5559424 2512 format['protocol'] = determine_protocol(format)
239df021 2513 if format.get('resolution') is None:
2514 format['resolution'] = self.format_resolution(format, default=None)
176f1866 2515 if format.get('dynamic_range') is None and format.get('vcodec') != 'none':
2516 format['dynamic_range'] = 'SDR'
f2fe69c7 2517 if (info_dict.get('duration') and format.get('tbr')
2518 and not format.get('filesize') and not format.get('filesize_approx')):
2519 format['filesize_approx'] = info_dict['duration'] * format['tbr'] * (1024 / 8)
2520
e5660ee6
JMF
2521 # Add HTTP headers, so that external programs can use them from the
2522 # json output
2523 full_format_info = info_dict.copy()
2524 full_format_info.update(format)
2525 format['http_headers'] = self._calc_headers(full_format_info)
0016b84e
S
2526 # Remove private housekeeping stuff
2527 if '__x_forwarded_for_ip' in info_dict:
2528 del info_dict['__x_forwarded_for_ip']
dd82ffea 2529
9f1a1c36 2530 if self.params.get('check_formats') is True:
282f5709 2531 formats = LazyList(self._check_formats(formats[::-1]), reverse=True)
9f1a1c36 2532
88acdbc2 2533 if not formats or formats[0] is not info_dict:
b3d9ef88
JMF
2534 # only set the 'formats' fields if the original info_dict list them
2535 # otherwise we end up with a circular reference, the first (and unique)
f89197d7 2536 # element in the 'formats' field in info_dict is info_dict itself,
dfb1b146 2537 # which can't be exported to json
b3d9ef88 2538 info_dict['formats'] = formats
4ec82a72 2539
2540 info_dict, _ = self.pre_process(info_dict)
2541
09b49e1f 2542 if self._match_entry(info_dict) is not None:
2543 return info_dict
2544
2545 self.post_extract(info_dict)
2546 info_dict, _ = self.pre_process(info_dict, 'after_filter')
2547
093a1710 2548 # The pre-processors may have modified the formats
2549 formats = info_dict.get('formats', [info_dict])
2550
fa9f30b8 2551 list_only = self.params.get('simulate') is None and (
2552 self.params.get('list_thumbnails') or self.params.get('listformats') or self.params.get('listsubtitles'))
2553 interactive_format_selection = not list_only and self.format_selector == '-'
b7b04c78 2554 if self.params.get('list_thumbnails'):
2555 self.list_thumbnails(info_dict)
b7b04c78 2556 if self.params.get('listsubtitles'):
2557 if 'automatic_captions' in info_dict:
2558 self.list_subtitles(
2559 info_dict['id'], automatic_captions, 'automatic captions')
2560 self.list_subtitles(info_dict['id'], subtitles, 'subtitles')
fa9f30b8 2561 if self.params.get('listformats') or interactive_format_selection:
b69fd25c 2562 self.list_formats(info_dict)
169dbde9 2563 if list_only:
b7b04c78 2564 # Without this printing, -F --print-json will not work
169dbde9 2565 self.__forced_printings(info_dict, self.prepare_filename(info_dict), incomplete=True)
bfaae0a7 2566 return
2567
187986a8 2568 format_selector = self.format_selector
2569 if format_selector is None:
0017d9ad 2570 req_format = self._default_format_spec(info_dict, download=download)
0760b0a7 2571 self.write_debug('Default format spec: %s' % req_format)
187986a8 2572 format_selector = self.build_format_selector(req_format)
317f7ab6 2573
fa9f30b8 2574 while True:
2575 if interactive_format_selection:
2576 req_format = input(
2577 self._format_screen('\nEnter format selector: ', self.Styles.EMPHASIS))
2578 try:
2579 format_selector = self.build_format_selector(req_format)
2580 except SyntaxError as err:
2581 self.report_error(err, tb=False, is_error=False)
2582 continue
2583
2584 # While in format selection we may need to have an access to the original
2585 # format set in order to calculate some metrics or do some processing.
2586 # For now we need to be able to guess whether original formats provided
2587 # by extractor are incomplete or not (i.e. whether extractor provides only
2588 # video-only or audio-only formats) for proper formats selection for
2589 # extractors with such incomplete formats (see
2590 # https://github.com/ytdl-org/youtube-dl/pull/5556).
2591 # Since formats may be filtered during format selection and may not match
2592 # the original formats the results may be incorrect. Thus original formats
2593 # or pre-calculated metrics should be passed to format selection routines
2594 # as well.
2595 # We will pass a context object containing all necessary additional data
2596 # instead of just formats.
2597 # This fixes incorrect format selection issue (see
2598 # https://github.com/ytdl-org/youtube-dl/issues/10083).
2599 incomplete_formats = (
2600 # All formats are video-only or
2601 all(f.get('vcodec') != 'none' and f.get('acodec') == 'none' for f in formats)
2602 # all formats are audio-only
2603 or all(f.get('vcodec') == 'none' and f.get('acodec') != 'none' for f in formats))
2604
2605 ctx = {
2606 'formats': formats,
2607 'incomplete_formats': incomplete_formats,
2608 }
2609
2610 formats_to_download = list(format_selector(ctx))
2611 if interactive_format_selection and not formats_to_download:
2612 self.report_error('Requested format is not available', tb=False, is_error=False)
2613 continue
2614 break
317f7ab6 2615
dd82ffea 2616 if not formats_to_download:
b7da73eb 2617 if not self.params.get('ignore_no_formats_error'):
1151c407 2618 raise ExtractorError('Requested format is not available', expected=True,
2619 video_id=info_dict['id'], ie=info_dict['extractor'])
b62fa6d7 2620 self.report_warning('Requested format is not available')
2621 # Process what we can, even without any available formats.
2622 formats_to_download = [{}]
a13e6848 2623
b62fa6d7 2624 best_format = formats_to_download[-1]
2625 if download:
2626 if best_format:
2627 self.to_screen(
2628 f'[info] {info_dict["id"]}: Downloading {len(formats_to_download)} format(s): '
2629 + ', '.join([f['format_id'] for f in formats_to_download]))
a13e6848 2630 max_downloads_reached = False
f46e2f9d 2631 for i, fmt in enumerate(formats_to_download):
09b49e1f 2632 formats_to_download[i] = new_info = self._copy_infodict(info_dict)
b7da73eb 2633 new_info.update(fmt)
a13e6848 2634 try:
2635 self.process_info(new_info)
2636 except MaxDownloadsReached:
2637 max_downloads_reached = True
f46e2f9d 2638 # Remove copied info
2639 for key, val in tuple(new_info.items()):
2640 if info_dict.get(key) == val:
2641 new_info.pop(key)
a13e6848 2642 if max_downloads_reached:
2643 break
ebed8b37 2644
9e907ebd 2645 write_archive = set(f.get('__write_download_archive', False) for f in formats_to_download)
a13e6848 2646 assert write_archive.issubset({True, False, 'ignore'})
2647 if True in write_archive and False not in write_archive:
2648 self.record_download_archive(info_dict)
be72c624 2649
2650 info_dict['requested_downloads'] = formats_to_download
ed5835b4 2651 info_dict = self.run_all_pps('after_video', info_dict)
a13e6848 2652 if max_downloads_reached:
2653 raise MaxDownloadsReached()
ebed8b37 2654
49a57e70 2655 # We update the info dict with the selected best quality format (backwards compatibility)
be72c624 2656 info_dict.update(best_format)
dd82ffea
JMF
2657 return info_dict
2658
98c70d6f 2659 def process_subtitles(self, video_id, normal_subtitles, automatic_captions):
a504ced0 2660 """Select the requested subtitles and their format"""
98c70d6f
JMF
2661 available_subs = {}
2662 if normal_subtitles and self.params.get('writesubtitles'):
2663 available_subs.update(normal_subtitles)
2664 if automatic_captions and self.params.get('writeautomaticsub'):
2665 for lang, cap_info in automatic_captions.items():
360e1ca5
JMF
2666 if lang not in available_subs:
2667 available_subs[lang] = cap_info
2668
4d171848
JMF
2669 if (not self.params.get('writesubtitles') and not
2670 self.params.get('writeautomaticsub') or not
2671 available_subs):
2672 return None
a504ced0 2673
c32b0aab 2674 all_sub_langs = available_subs.keys()
a504ced0 2675 if self.params.get('allsubtitles', False):
c32b0aab 2676 requested_langs = all_sub_langs
2677 elif self.params.get('subtitleslangs', False):
77c4a9ef 2678 # A list is used so that the order of languages will be the same as
2679 # given in subtitleslangs. See https://github.com/yt-dlp/yt-dlp/issues/1041
2680 requested_langs = []
2681 for lang_re in self.params.get('subtitleslangs'):
77c4a9ef 2682 discard = lang_re[0] == '-'
c32b0aab 2683 if discard:
77c4a9ef 2684 lang_re = lang_re[1:]
3aa91540 2685 if lang_re == 'all':
2686 if discard:
2687 requested_langs = []
2688 else:
2689 requested_langs.extend(all_sub_langs)
2690 continue
77c4a9ef 2691 current_langs = filter(re.compile(lang_re + '$').match, all_sub_langs)
c32b0aab 2692 if discard:
2693 for lang in current_langs:
77c4a9ef 2694 while lang in requested_langs:
2695 requested_langs.remove(lang)
c32b0aab 2696 else:
77c4a9ef 2697 requested_langs.extend(current_langs)
2698 requested_langs = orderedSet(requested_langs)
c32b0aab 2699 elif 'en' in available_subs:
2700 requested_langs = ['en']
a504ced0 2701 else:
c32b0aab 2702 requested_langs = [list(all_sub_langs)[0]]
ad3dc496 2703 if requested_langs:
2704 self.write_debug('Downloading subtitles: %s' % ', '.join(requested_langs))
a504ced0
JMF
2705
2706 formats_query = self.params.get('subtitlesformat', 'best')
2707 formats_preference = formats_query.split('/') if formats_query else []
2708 subs = {}
2709 for lang in requested_langs:
2710 formats = available_subs.get(lang)
2711 if formats is None:
2712 self.report_warning('%s subtitles not available for %s' % (lang, video_id))
2713 continue
a504ced0
JMF
2714 for ext in formats_preference:
2715 if ext == 'best':
2716 f = formats[-1]
2717 break
2718 matches = list(filter(lambda f: f['ext'] == ext, formats))
2719 if matches:
2720 f = matches[-1]
2721 break
2722 else:
2723 f = formats[-1]
2724 self.report_warning(
2725 'No subtitle format found matching "%s" for language %s, '
2726 'using %s' % (formats_query, lang, f['ext']))
2727 subs[lang] = f
2728 return subs
2729
bb66c247 2730 def _forceprint(self, key, info_dict):
2731 if info_dict is None:
2732 return
2733 info_copy = info_dict.copy()
2734 info_copy['formats_table'] = self.render_formats_table(info_dict)
2735 info_copy['thumbnails_table'] = self.render_thumbnails_table(info_dict)
2736 info_copy['subtitles_table'] = self.render_subtitles_table(info_dict.get('id'), info_dict.get('subtitles'))
2737 info_copy['automatic_captions_table'] = self.render_subtitles_table(info_dict.get('id'), info_dict.get('automatic_captions'))
2738
2739 def format_tmpl(tmpl):
2740 mobj = re.match(r'\w+(=?)$', tmpl)
2741 if mobj and mobj.group(1):
2742 return f'{tmpl[:-1]} = %({tmpl[:-1]})r'
2743 elif mobj:
2744 return f'%({tmpl})s'
2745 return tmpl
8130779d 2746
bb66c247 2747 for tmpl in self.params['forceprint'].get(key, []):
2748 self.to_stdout(self.evaluate_outtmpl(format_tmpl(tmpl), info_copy))
2749
2750 for tmpl, file_tmpl in self.params['print_to_file'].get(key, []):
2751 filename = self.evaluate_outtmpl(file_tmpl, info_dict)
2752 tmpl = format_tmpl(tmpl)
2753 self.to_screen(f'[info] Writing {tmpl!r} to: {filename}')
8d93e69d 2754 if self._ensure_dir_exists(filename):
2755 with io.open(filename, 'a', encoding='utf-8') as f:
2756 f.write(self.evaluate_outtmpl(tmpl, info_copy) + '\n')
ca30f449 2757
d06daf23 2758 def __forced_printings(self, info_dict, filename, incomplete):
53c18592 2759 def print_mandatory(field, actual_field=None):
2760 if actual_field is None:
2761 actual_field = field
d06daf23 2762 if (self.params.get('force%s' % field, False)
53c18592 2763 and (not incomplete or info_dict.get(actual_field) is not None)):
2764 self.to_stdout(info_dict[actual_field])
d06daf23
S
2765
2766 def print_optional(field):
2767 if (self.params.get('force%s' % field, False)
2768 and info_dict.get(field) is not None):
2769 self.to_stdout(info_dict[field])
2770
53c18592 2771 info_dict = info_dict.copy()
2772 if filename is not None:
2773 info_dict['filename'] = filename
2774 if info_dict.get('requested_formats') is not None:
2775 # For RTMP URLs, also include the playpath
2776 info_dict['urls'] = '\n'.join(f['url'] + f.get('play_path', '') for f in info_dict['requested_formats'])
2777 elif 'url' in info_dict:
2778 info_dict['urls'] = info_dict['url'] + info_dict.get('play_path', '')
2779
bb66c247 2780 if (self.params.get('forcejson')
2781 or self.params['forceprint'].get('video')
2782 or self.params['print_to_file'].get('video')):
2b8a2973 2783 self.post_extract(info_dict)
bb66c247 2784 self._forceprint('video', info_dict)
53c18592 2785
d06daf23
S
2786 print_mandatory('title')
2787 print_mandatory('id')
53c18592 2788 print_mandatory('url', 'urls')
d06daf23
S
2789 print_optional('thumbnail')
2790 print_optional('description')
53c18592 2791 print_optional('filename')
b868936c 2792 if self.params.get('forceduration') and info_dict.get('duration') is not None:
d06daf23
S
2793 self.to_stdout(formatSeconds(info_dict['duration']))
2794 print_mandatory('format')
53c18592 2795
2b8a2973 2796 if self.params.get('forcejson'):
6e84b215 2797 self.to_stdout(json.dumps(self.sanitize_info(info_dict)))
d06daf23 2798
e8e73840 2799 def dl(self, name, info, subtitle=False, test=False):
88acdbc2 2800 if not info.get('url'):
1151c407 2801 self.raise_no_formats(info, True)
e8e73840 2802
2803 if test:
2804 verbose = self.params.get('verbose')
2805 params = {
2806 'test': True,
a169858f 2807 'quiet': self.params.get('quiet') or not verbose,
e8e73840 2808 'verbose': verbose,
2809 'noprogress': not verbose,
2810 'nopart': True,
2811 'skip_unavailable_fragments': False,
2812 'keep_fragments': False,
2813 'overwrites': True,
2814 '_no_ytdl_file': True,
2815 }
2816 else:
2817 params = self.params
96fccc10 2818 fd = get_suitable_downloader(info, params, to_stdout=(name == '-'))(self, params)
e8e73840 2819 if not test:
2820 for ph in self._progress_hooks:
2821 fd.add_progress_hook(ph)
42676437
M
2822 urls = '", "'.join(
2823 (f['url'].split(',')[0] + ',<data>' if f['url'].startswith('data:') else f['url'])
2824 for f in info.get('requested_formats', []) or [info])
18e674b4 2825 self.write_debug('Invoking downloader on "%s"' % urls)
03b4de72 2826
adbc4ec4
THD
2827 # Note: Ideally info should be a deep-copied so that hooks cannot modify it.
2828 # But it may contain objects that are not deep-copyable
2829 new_info = self._copy_infodict(info)
e8e73840 2830 if new_info.get('http_headers') is None:
2831 new_info['http_headers'] = self._calc_headers(new_info)
2832 return fd.download(name, new_info, subtitle)
2833
e04938ab 2834 def existing_file(self, filepaths, *, default_overwrite=True):
2835 existing_files = list(filter(os.path.exists, orderedSet(filepaths)))
2836 if existing_files and not self.params.get('overwrites', default_overwrite):
2837 return existing_files[0]
2838
2839 for file in existing_files:
2840 self.report_file_delete(file)
2841 os.remove(file)
2842 return None
2843
8222d8de 2844 def process_info(self, info_dict):
09b49e1f 2845 """Process a single resolved IE result. (Modifies it in-place)"""
8222d8de
JMF
2846
2847 assert info_dict.get('_type', 'video') == 'video'
f46e2f9d 2848 original_infodict = info_dict
fd288278 2849
4513a41a 2850 if 'format' not in info_dict and 'ext' in info_dict:
8222d8de
JMF
2851 info_dict['format'] = info_dict['ext']
2852
09b49e1f 2853 # This is mostly just for backward compatibility of process_info
2854 # As a side-effect, this allows for format-specific filters
c77495e3 2855 if self._match_entry(info_dict) is not None:
9e907ebd 2856 info_dict['__write_download_archive'] = 'ignore'
8222d8de
JMF
2857 return
2858
09b49e1f 2859 # Does nothing under normal operation - for backward compatibility of process_info
277d6ff5 2860 self.post_extract(info_dict)
8222d8de 2861
dcf64d43 2862 # info_dict['_filename'] needs to be set for backward compatibility
de6000d9 2863 info_dict['_filename'] = full_filename = self.prepare_filename(info_dict, warn=True)
2864 temp_filename = self.prepare_filename(info_dict, 'temp')
0202b52a 2865 files_to_move = {}
8222d8de 2866
09b49e1f 2867 self._num_downloads += 1
2868
8222d8de 2869 # Forced printings
4513a41a 2870 self.__forced_printings(info_dict, full_filename, incomplete=('format' not in info_dict))
8222d8de 2871
b7b04c78 2872 if self.params.get('simulate'):
9e907ebd 2873 info_dict['__write_download_archive'] = self.params.get('force_write_download_archive')
8222d8de
JMF
2874 return
2875
de6000d9 2876 if full_filename is None:
8222d8de 2877 return
e92caff5 2878 if not self._ensure_dir_exists(encodeFilename(full_filename)):
0202b52a 2879 return
e92caff5 2880 if not self._ensure_dir_exists(encodeFilename(temp_filename)):
8222d8de
JMF
2881 return
2882
80c03fa9 2883 if self._write_description('video', info_dict,
2884 self.prepare_filename(info_dict, 'description')) is None:
2885 return
2886
2887 sub_files = self._write_subtitles(info_dict, temp_filename)
2888 if sub_files is None:
2889 return
2890 files_to_move.update(dict(sub_files))
2891
2892 thumb_files = self._write_thumbnails(
2893 'video', info_dict, temp_filename, self.prepare_filename(info_dict, 'thumbnail'))
2894 if thumb_files is None:
2895 return
2896 files_to_move.update(dict(thumb_files))
8222d8de 2897
80c03fa9 2898 infofn = self.prepare_filename(info_dict, 'infojson')
2899 _infojson_written = self._write_info_json('video', info_dict, infofn)
2900 if _infojson_written:
dac5df5a 2901 info_dict['infojson_filename'] = infofn
e75bb0d6 2902 # For backward compatibility, even though it was a private field
80c03fa9 2903 info_dict['__infojson_filename'] = infofn
2904 elif _infojson_written is None:
2905 return
2906
2907 # Note: Annotations are deprecated
2908 annofn = None
1fb07d10 2909 if self.params.get('writeannotations', False):
de6000d9 2910 annofn = self.prepare_filename(info_dict, 'annotation')
80c03fa9 2911 if annofn:
e92caff5 2912 if not self._ensure_dir_exists(encodeFilename(annofn)):
0202b52a 2913 return
0c3d0f51 2914 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(annofn)):
6febd1c1 2915 self.to_screen('[info] Video annotations are already present')
ffddb112
RA
2916 elif not info_dict.get('annotations'):
2917 self.report_warning('There are no annotations to write.')
7b6fefc9
PH
2918 else:
2919 try:
6febd1c1 2920 self.to_screen('[info] Writing video annotations to: ' + annofn)
7b6fefc9
PH
2921 with io.open(encodeFilename(annofn), 'w', encoding='utf-8') as annofile:
2922 annofile.write(info_dict['annotations'])
2923 except (KeyError, TypeError):
6febd1c1 2924 self.report_warning('There are no annotations to write.')
7b6fefc9 2925 except (OSError, IOError):
6febd1c1 2926 self.report_error('Cannot write annotations file: ' + annofn)
7b6fefc9 2927 return
1fb07d10 2928
732044af 2929 # Write internet shortcut files
08438d2c 2930 def _write_link_file(link_type):
60f3e995 2931 url = try_get(info_dict['webpage_url'], iri_to_uri)
2932 if not url:
2933 self.report_warning(
2934 f'Cannot write internet shortcut file because the actual URL of "{info_dict["webpage_url"]}" is unknown')
2935 return True
08438d2c 2936 linkfn = replace_extension(self.prepare_filename(info_dict, 'link'), link_type, info_dict.get('ext'))
0e6b018a
Z
2937 if not self._ensure_dir_exists(encodeFilename(linkfn)):
2938 return False
10e3742e 2939 if self.params.get('overwrites', True) and os.path.exists(encodeFilename(linkfn)):
08438d2c 2940 self.to_screen(f'[info] Internet shortcut (.{link_type}) is already present')
2941 return True
2942 try:
2943 self.to_screen(f'[info] Writing internet shortcut (.{link_type}) to: {linkfn}')
2944 with io.open(encodeFilename(to_high_limit_path(linkfn)), 'w', encoding='utf-8',
2945 newline='\r\n' if link_type == 'url' else '\n') as linkfile:
60f3e995 2946 template_vars = {'url': url}
08438d2c 2947 if link_type == 'desktop':
2948 template_vars['filename'] = linkfn[:-(len(link_type) + 1)]
2949 linkfile.write(LINK_TEMPLATES[link_type] % template_vars)
2950 except (OSError, IOError):
2951 self.report_error(f'Cannot write internet shortcut {linkfn}')
2952 return False
732044af 2953 return True
2954
08438d2c 2955 write_links = {
2956 'url': self.params.get('writeurllink'),
2957 'webloc': self.params.get('writewebloclink'),
2958 'desktop': self.params.get('writedesktoplink'),
2959 }
2960 if self.params.get('writelink'):
2961 link_type = ('webloc' if sys.platform == 'darwin'
2962 else 'desktop' if sys.platform.startswith('linux')
2963 else 'url')
2964 write_links[link_type] = True
2965
2966 if any(should_write and not _write_link_file(link_type)
2967 for link_type, should_write in write_links.items()):
2968 return
732044af 2969
f46e2f9d 2970 def replace_info_dict(new_info):
2971 nonlocal info_dict
2972 if new_info == info_dict:
2973 return
2974 info_dict.clear()
2975 info_dict.update(new_info)
2976
56d868db 2977 try:
f46e2f9d 2978 new_info, files_to_move = self.pre_process(info_dict, 'before_dl', files_to_move)
2979 replace_info_dict(new_info)
56d868db 2980 except PostProcessingError as err:
2981 self.report_error('Preprocessing: %s' % str(err))
2982 return
2983
a13e6848 2984 if self.params.get('skip_download'):
56d868db 2985 info_dict['filepath'] = temp_filename
2986 info_dict['__finaldir'] = os.path.dirname(os.path.abspath(encodeFilename(full_filename)))
2987 info_dict['__files_to_move'] = files_to_move
f46e2f9d 2988 replace_info_dict(self.run_pp(MoveFilesAfterDownloadPP(self, False), info_dict))
9e907ebd 2989 info_dict['__write_download_archive'] = self.params.get('force_write_download_archive')
56d868db 2990 else:
2991 # Download
b868936c 2992 info_dict.setdefault('__postprocessors', [])
4340deca 2993 try:
0202b52a 2994
e04938ab 2995 def existing_video_file(*filepaths):
6b591b29 2996 ext = info_dict.get('ext')
e04938ab 2997 converted = lambda file: replace_extension(file, self.params.get('final_ext') or ext, ext)
2998 file = self.existing_file(itertools.chain(*zip(map(converted, filepaths), filepaths)),
2999 default_overwrite=False)
3000 if file:
3001 info_dict['ext'] = os.path.splitext(file)[1][1:]
3002 return file
0202b52a 3003
3004 success = True
4340deca 3005 if info_dict.get('requested_formats') is not None:
81cd954a
S
3006
3007 def compatible_formats(formats):
d03cfdce 3008 # TODO: some formats actually allow this (mkv, webm, ogg, mp4), but not all of them.
3009 video_formats = [format for format in formats if format.get('vcodec') != 'none']
3010 audio_formats = [format for format in formats if format.get('acodec') != 'none']
3011 if len(video_formats) > 2 or len(audio_formats) > 2:
3012 return False
3013
81cd954a 3014 # Check extension
d03cfdce 3015 exts = set(format.get('ext') for format in formats)
3016 COMPATIBLE_EXTS = (
3017 set(('mp3', 'mp4', 'm4a', 'm4p', 'm4b', 'm4r', 'm4v', 'ismv', 'isma')),
3018 set(('webm',)),
3019 )
3020 for ext_sets in COMPATIBLE_EXTS:
3021 if ext_sets.issuperset(exts):
3022 return True
81cd954a
S
3023 # TODO: Check acodec/vcodec
3024 return False
3025
3026 requested_formats = info_dict['requested_formats']
0202b52a 3027 old_ext = info_dict['ext']
4e3b637d 3028 if self.params.get('merge_output_format') is None:
3029 if not compatible_formats(requested_formats):
3030 info_dict['ext'] = 'mkv'
3031 self.report_warning(
3032 'Requested formats are incompatible for merge and will be merged into mkv')
3033 if (info_dict['ext'] == 'webm'
3034 and info_dict.get('thumbnails')
3035 # check with type instead of pp_key, __name__, or isinstance
3036 # since we dont want any custom PPs to trigger this
3037 and any(type(pp) == EmbedThumbnailPP for pp in self._pps['post_process'])):
3038 info_dict['ext'] = 'mkv'
3039 self.report_warning(
3040 'webm doesn\'t support embedding a thumbnail, mkv will be used')
124bc071 3041 new_ext = info_dict['ext']
0202b52a 3042
124bc071 3043 def correct_ext(filename, ext=new_ext):
96fccc10 3044 if filename == '-':
3045 return filename
0202b52a 3046 filename_real_ext = os.path.splitext(filename)[1][1:]
3047 filename_wo_ext = (
3048 os.path.splitext(filename)[0]
124bc071 3049 if filename_real_ext in (old_ext, new_ext)
0202b52a 3050 else filename)
124bc071 3051 return '%s.%s' % (filename_wo_ext, ext)
0202b52a 3052
38c6902b 3053 # Ensure filename always has a correct extension for successful merge
0202b52a 3054 full_filename = correct_ext(full_filename)
3055 temp_filename = correct_ext(temp_filename)
e04938ab 3056 dl_filename = existing_video_file(full_filename, temp_filename)
1ea24129 3057 info_dict['__real_download'] = False
18e674b4 3058
adbc4ec4
THD
3059 downloaded = []
3060 merger = FFmpegMergerPP(self)
3061
3062 fd = get_suitable_downloader(info_dict, self.params, to_stdout=temp_filename == '-')
dbf5416a 3063 if dl_filename is not None:
6c7274ec 3064 self.report_file_already_downloaded(dl_filename)
adbc4ec4
THD
3065 elif fd:
3066 for f in requested_formats if fd != FFmpegFD else []:
3067 f['filepath'] = fname = prepend_extension(
3068 correct_ext(temp_filename, info_dict['ext']),
3069 'f%s' % f['format_id'], info_dict['ext'])
3070 downloaded.append(fname)
dbf5416a 3071 info_dict['url'] = '\n'.join(f['url'] for f in requested_formats)
3072 success, real_download = self.dl(temp_filename, info_dict)
3073 info_dict['__real_download'] = real_download
18e674b4 3074 else:
18e674b4 3075 if self.params.get('allow_unplayable_formats'):
3076 self.report_warning(
3077 'You have requested merging of multiple formats '
3078 'while also allowing unplayable formats to be downloaded. '
3079 'The formats won\'t be merged to prevent data corruption.')
3080 elif not merger.available:
e8969bda 3081 msg = 'You have requested merging of multiple formats but ffmpeg is not installed'
3082 if not self.params.get('ignoreerrors'):
3083 self.report_error(f'{msg}. Aborting due to --abort-on-error')
3084 return
3085 self.report_warning(f'{msg}. The formats won\'t be merged')
18e674b4 3086
96fccc10 3087 if temp_filename == '-':
adbc4ec4 3088 reason = ('using a downloader other than ffmpeg' if FFmpegFD.can_merge_formats(info_dict, self.params)
96fccc10 3089 else 'but the formats are incompatible for simultaneous download' if merger.available
3090 else 'but ffmpeg is not installed')
3091 self.report_warning(
3092 f'You have requested downloading multiple formats to stdout {reason}. '
3093 'The formats will be streamed one after the other')
3094 fname = temp_filename
dbf5416a 3095 for f in requested_formats:
3096 new_info = dict(info_dict)
3097 del new_info['requested_formats']
3098 new_info.update(f)
96fccc10 3099 if temp_filename != '-':
124bc071 3100 fname = prepend_extension(
3101 correct_ext(temp_filename, new_info['ext']),
3102 'f%s' % f['format_id'], new_info['ext'])
96fccc10 3103 if not self._ensure_dir_exists(fname):
3104 return
a21e0ab1 3105 f['filepath'] = fname
96fccc10 3106 downloaded.append(fname)
dbf5416a 3107 partial_success, real_download = self.dl(fname, new_info)
3108 info_dict['__real_download'] = info_dict['__real_download'] or real_download
3109 success = success and partial_success
adbc4ec4
THD
3110
3111 if downloaded and merger.available and not self.params.get('allow_unplayable_formats'):
3112 info_dict['__postprocessors'].append(merger)
3113 info_dict['__files_to_merge'] = downloaded
3114 # Even if there were no downloads, it is being merged only now
3115 info_dict['__real_download'] = True
3116 else:
3117 for file in downloaded:
3118 files_to_move[file] = None
4340deca
P
3119 else:
3120 # Just a single file
e04938ab 3121 dl_filename = existing_video_file(full_filename, temp_filename)
6c7274ec 3122 if dl_filename is None or dl_filename == temp_filename:
3123 # dl_filename == temp_filename could mean that the file was partially downloaded with --no-part.
3124 # So we should try to resume the download
e8e73840 3125 success, real_download = self.dl(temp_filename, info_dict)
0202b52a 3126 info_dict['__real_download'] = real_download
6c7274ec 3127 else:
3128 self.report_file_already_downloaded(dl_filename)
0202b52a 3129
0202b52a 3130 dl_filename = dl_filename or temp_filename
c571435f 3131 info_dict['__finaldir'] = os.path.dirname(os.path.abspath(encodeFilename(full_filename)))
0202b52a 3132
3158150c 3133 except network_exceptions as err:
7960b056 3134 self.report_error('unable to download video data: %s' % error_to_compat_str(err))
4340deca
P
3135 return
3136 except (OSError, IOError) as err:
3137 raise UnavailableVideoError(err)
3138 except (ContentTooShortError, ) as err:
3139 self.report_error('content too short (expected %s bytes and served %s)' % (err.expected, err.downloaded))
3140 return
8222d8de 3141
de6000d9 3142 if success and full_filename != '-':
f17f8651 3143
fd7cfb64 3144 def fixup():
3145 do_fixup = True
3146 fixup_policy = self.params.get('fixup')
3147 vid = info_dict['id']
3148
3149 if fixup_policy in ('ignore', 'never'):
3150 return
3151 elif fixup_policy == 'warn':
3152 do_fixup = False
f89b3e2d 3153 elif fixup_policy != 'force':
3154 assert fixup_policy in ('detect_or_warn', None)
3155 if not info_dict.get('__real_download'):
3156 do_fixup = False
fd7cfb64 3157
3158 def ffmpeg_fixup(cndn, msg, cls):
3159 if not cndn:
3160 return
3161 if not do_fixup:
3162 self.report_warning(f'{vid}: {msg}')
3163 return
3164 pp = cls(self)
3165 if pp.available:
3166 info_dict['__postprocessors'].append(pp)
3167 else:
3168 self.report_warning(f'{vid}: {msg}. Install ffmpeg to fix this automatically')
3169
3170 stretched_ratio = info_dict.get('stretched_ratio')
3171 ffmpeg_fixup(
3172 stretched_ratio not in (1, None),
3173 f'Non-uniform pixel ratio {stretched_ratio}',
3174 FFmpegFixupStretchedPP)
3175
3176 ffmpeg_fixup(
3177 (info_dict.get('requested_formats') is None
3178 and info_dict.get('container') == 'm4a_dash'
3179 and info_dict.get('ext') == 'm4a'),
3180 'writing DASH m4a. Only some players support this container',
3181 FFmpegFixupM4aPP)
3182
993191c0 3183 downloader = get_suitable_downloader(info_dict, self.params) if 'protocol' in info_dict else None
3184 downloader = downloader.__name__ if downloader else None
adbc4ec4
THD
3185
3186 if info_dict.get('requested_formats') is None: # Not necessary if doing merger
3187 ffmpeg_fixup(downloader == 'HlsFD',
3188 'Possible MPEG-TS in MP4 container or malformed AAC timestamps',
3189 FFmpegFixupM3u8PP)
3190 ffmpeg_fixup(info_dict.get('is_live') and downloader == 'DashSegmentsFD',
3191 'Possible duplicate MOOV atoms', FFmpegFixupDuplicateMoovPP)
3192
e04b003e 3193 ffmpeg_fixup(downloader == 'WebSocketFragmentFD', 'Malformed timestamps detected', FFmpegFixupTimestampPP)
3194 ffmpeg_fixup(downloader == 'WebSocketFragmentFD', 'Malformed duration detected', FFmpegFixupDurationPP)
fd7cfb64 3195
3196 fixup()
8222d8de 3197 try:
f46e2f9d 3198 replace_info_dict(self.post_process(dl_filename, info_dict, files_to_move))
af819c21 3199 except PostProcessingError as err:
3200 self.report_error('Postprocessing: %s' % str(err))
8222d8de 3201 return
ab8e5e51
AM
3202 try:
3203 for ph in self._post_hooks:
23c1a667 3204 ph(info_dict['filepath'])
ab8e5e51
AM
3205 except Exception as err:
3206 self.report_error('post hooks: %s' % str(err))
3207 return
9e907ebd 3208 info_dict['__write_download_archive'] = True
2d30509f 3209
a13e6848 3210 if self.params.get('force_write_download_archive'):
9e907ebd 3211 info_dict['__write_download_archive'] = True
a13e6848 3212
3213 # Make sure the info_dict was modified in-place
f46e2f9d 3214 assert info_dict is original_infodict
a13e6848 3215
c3e6ffba 3216 max_downloads = self.params.get('max_downloads')
3217 if max_downloads is not None and self._num_downloads >= int(max_downloads):
3218 raise MaxDownloadsReached()
8222d8de 3219
aa9369a2 3220 def __download_wrapper(self, func):
3221 @functools.wraps(func)
3222 def wrapper(*args, **kwargs):
3223 try:
3224 res = func(*args, **kwargs)
3225 except UnavailableVideoError as e:
3226 self.report_error(e)
b222c271 3227 except MaxDownloadsReached as e:
aa9369a2 3228 self.to_screen(f'[info] {e}')
3229 raise
b222c271 3230 except DownloadCancelled as e:
3231 self.to_screen(f'[info] {e}')
3232 if not self.params.get('break_per_url'):
3233 raise
aa9369a2 3234 else:
3235 if self.params.get('dump_single_json', False):
3236 self.post_extract(res)
3237 self.to_stdout(json.dumps(self.sanitize_info(res)))
3238 return wrapper
3239
8222d8de
JMF
3240 def download(self, url_list):
3241 """Download a given list of URLs."""
aa9369a2 3242 url_list = variadic(url_list) # Passing a single URL is a common mistake
de6000d9 3243 outtmpl = self.outtmpl_dict['default']
3089bc74
S
3244 if (len(url_list) > 1
3245 and outtmpl != '-'
3246 and '%' not in outtmpl
3247 and self.params.get('max_downloads') != 1):
acd69589 3248 raise SameFileError(outtmpl)
8222d8de
JMF
3249
3250 for url in url_list:
aa9369a2 3251 self.__download_wrapper(self.extract_info)(
3252 url, force_generic_extractor=self.params.get('force_generic_extractor', False))
8222d8de
JMF
3253
3254 return self._download_retcode
3255
1dcc4c0c 3256 def download_with_info_file(self, info_filename):
31bd3925
JMF
3257 with contextlib.closing(fileinput.FileInput(
3258 [info_filename], mode='r',
3259 openhook=fileinput.hook_encoded('utf-8'))) as f:
3260 # FileInput doesn't have a read method, we can't call json.load
8012d892 3261 info = self.sanitize_info(json.loads('\n'.join(f)), self.params.get('clean_infojson', True))
d4943898 3262 try:
aa9369a2 3263 self.__download_wrapper(self.process_ie_result)(info, download=True)
f2ebc5c7 3264 except (DownloadError, EntryNotInPlaylist, ReExtractInfo) as e:
bf5f605e 3265 if not isinstance(e, EntryNotInPlaylist):
3266 self.to_stderr('\r')
d4943898
JMF
3267 webpage_url = info.get('webpage_url')
3268 if webpage_url is not None:
aa9369a2 3269 self.report_warning(f'The info failed to download: {e}; trying with URL {webpage_url}')
d4943898
JMF
3270 return self.download([webpage_url])
3271 else:
3272 raise
3273 return self._download_retcode
1dcc4c0c 3274
cb202fd2 3275 @staticmethod
8012d892 3276 def sanitize_info(info_dict, remove_private_keys=False):
3277 ''' Sanitize the infodict for converting to json '''
3ad56b42 3278 if info_dict is None:
3279 return info_dict
6e84b215 3280 info_dict.setdefault('epoch', int(time.time()))
6a5a30f9 3281 info_dict.setdefault('_type', 'video')
09b49e1f 3282
8012d892 3283 if remove_private_keys:
09b49e1f 3284 reject = lambda k, v: v is None or (k.startswith('_') and k != '_type') or k in {
f46e2f9d 3285 'requested_downloads', 'requested_formats', 'requested_subtitles', 'requested_entries',
3286 'entries', 'filepath', 'infojson_filename', 'original_url', 'playlist_autonumber',
6e84b215 3287 }
ae8f99e6 3288 else:
09b49e1f 3289 reject = lambda k, v: False
adbc4ec4
THD
3290
3291 def filter_fn(obj):
3292 if isinstance(obj, dict):
3293 return {k: filter_fn(v) for k, v in obj.items() if not reject(k, v)}
3294 elif isinstance(obj, (list, tuple, set, LazyList)):
3295 return list(map(filter_fn, obj))
3296 elif obj is None or isinstance(obj, (str, int, float, bool)):
3297 return obj
3298 else:
3299 return repr(obj)
3300
5226731e 3301 return filter_fn(info_dict)
cb202fd2 3302
8012d892 3303 @staticmethod
3304 def filter_requested_info(info_dict, actually_filter=True):
3305 ''' Alias of sanitize_info for backward compatibility '''
3306 return YoutubeDL.sanitize_info(info_dict, actually_filter)
3307
ed5835b4 3308 @staticmethod
3309 def post_extract(info_dict):
3310 def actual_post_extract(info_dict):
3311 if info_dict.get('_type') in ('playlist', 'multi_video'):
3312 for video_dict in info_dict.get('entries', {}):
3313 actual_post_extract(video_dict or {})
3314 return
3315
09b49e1f 3316 post_extractor = info_dict.pop('__post_extractor', None) or (lambda: {})
3317 info_dict.update(post_extractor())
ed5835b4 3318
3319 actual_post_extract(info_dict or {})
3320
dcf64d43 3321 def run_pp(self, pp, infodict):
5bfa4862 3322 files_to_delete = []
dcf64d43 3323 if '__files_to_move' not in infodict:
3324 infodict['__files_to_move'] = {}
b1940459 3325 try:
3326 files_to_delete, infodict = pp.run(infodict)
3327 except PostProcessingError as e:
3328 # Must be True and not 'only_download'
3329 if self.params.get('ignoreerrors') is True:
3330 self.report_error(e)
3331 return infodict
3332 raise
3333
5bfa4862 3334 if not files_to_delete:
dcf64d43 3335 return infodict
5bfa4862 3336 if self.params.get('keepvideo', False):
3337 for f in files_to_delete:
dcf64d43 3338 infodict['__files_to_move'].setdefault(f, '')
5bfa4862 3339 else:
3340 for old_filename in set(files_to_delete):
3341 self.to_screen('Deleting original file %s (pass -k to keep)' % old_filename)
3342 try:
3343 os.remove(encodeFilename(old_filename))
3344 except (IOError, OSError):
3345 self.report_warning('Unable to remove downloaded original file')
dcf64d43 3346 if old_filename in infodict['__files_to_move']:
3347 del infodict['__files_to_move'][old_filename]
3348 return infodict
5bfa4862 3349
ed5835b4 3350 def run_all_pps(self, key, info, *, additional_pps=None):
bb66c247 3351 self._forceprint(key, info)
ed5835b4 3352 for pp in (additional_pps or []) + self._pps[key]:
dc5f409c 3353 info = self.run_pp(pp, info)
ed5835b4 3354 return info
277d6ff5 3355
56d868db 3356 def pre_process(self, ie_info, key='pre_process', files_to_move=None):
5bfa4862 3357 info = dict(ie_info)
56d868db 3358 info['__files_to_move'] = files_to_move or {}
ed5835b4 3359 info = self.run_all_pps(key, info)
56d868db 3360 return info, info.pop('__files_to_move', None)
5bfa4862 3361
f46e2f9d 3362 def post_process(self, filename, info, files_to_move=None):
8222d8de 3363 """Run all the postprocessors on the given file."""
8222d8de 3364 info['filepath'] = filename
dcf64d43 3365 info['__files_to_move'] = files_to_move or {}
ed5835b4 3366 info = self.run_all_pps('post_process', info, additional_pps=info.get('__postprocessors'))
dcf64d43 3367 info = self.run_pp(MoveFilesAfterDownloadPP(self), info)
3368 del info['__files_to_move']
ed5835b4 3369 return self.run_all_pps('after_move', info)
c1c9a79c 3370
5db07df6 3371 def _make_archive_id(self, info_dict):
e9fef7ee
S
3372 video_id = info_dict.get('id')
3373 if not video_id:
3374 return
5db07df6
PH
3375 # Future-proof against any change in case
3376 # and backwards compatibility with prior versions
e9fef7ee 3377 extractor = info_dict.get('extractor_key') or info_dict.get('ie_key') # key in a playlist
7012b23c 3378 if extractor is None:
1211bb6d
S
3379 url = str_or_none(info_dict.get('url'))
3380 if not url:
3381 return
e9fef7ee 3382 # Try to find matching extractor for the URL and take its ie_key
8b7491c8 3383 for ie_key, ie in self._ies.items():
1211bb6d 3384 if ie.suitable(url):
8b7491c8 3385 extractor = ie_key
e9fef7ee
S
3386 break
3387 else:
3388 return
d0757229 3389 return '%s %s' % (extractor.lower(), video_id)
5db07df6
PH
3390
3391 def in_download_archive(self, info_dict):
3392 fn = self.params.get('download_archive')
3393 if fn is None:
3394 return False
3395
3396 vid_id = self._make_archive_id(info_dict)
e9fef7ee 3397 if not vid_id:
7012b23c 3398 return False # Incomplete video information
5db07df6 3399
a45e8619 3400 return vid_id in self.archive
c1c9a79c
PH
3401
3402 def record_download_archive(self, info_dict):
3403 fn = self.params.get('download_archive')
3404 if fn is None:
3405 return
5db07df6
PH
3406 vid_id = self._make_archive_id(info_dict)
3407 assert vid_id
a13e6848 3408 self.write_debug(f'Adding to archive: {vid_id}')
c1c9a79c 3409 with locked_file(fn, 'a', encoding='utf-8') as archive_file:
6febd1c1 3410 archive_file.write(vid_id + '\n')
a45e8619 3411 self.archive.add(vid_id)
dd82ffea 3412
8c51aa65 3413 @staticmethod
8abeeb94 3414 def format_resolution(format, default='unknown'):
9359f3d4 3415 if format.get('vcodec') == 'none' and format.get('acodec') != 'none':
fb04e403 3416 return 'audio only'
f49d89ee
PH
3417 if format.get('resolution') is not None:
3418 return format['resolution']
35615307 3419 if format.get('width') and format.get('height'):
ff51ed58 3420 return '%dx%d' % (format['width'], format['height'])
35615307 3421 elif format.get('height'):
ff51ed58 3422 return '%sp' % format['height']
35615307 3423 elif format.get('width'):
ff51ed58 3424 return '%dx?' % format['width']
3425 return default
8c51aa65 3426
8130779d 3427 def _list_format_headers(self, *headers):
3428 if self.params.get('listformats_table', True) is not False:
3429 return [self._format_screen(header, self.Styles.HEADERS) for header in headers]
3430 return headers
3431
c57f7757
PH
3432 def _format_note(self, fdict):
3433 res = ''
3434 if fdict.get('ext') in ['f4f', 'f4m']:
f304da8a 3435 res += '(unsupported)'
32f90364
PH
3436 if fdict.get('language'):
3437 if res:
3438 res += ' '
f304da8a 3439 res += '[%s]' % fdict['language']
c57f7757 3440 if fdict.get('format_note') is not None:
f304da8a 3441 if res:
3442 res += ' '
3443 res += fdict['format_note']
c57f7757 3444 if fdict.get('tbr') is not None:
f304da8a 3445 if res:
3446 res += ', '
3447 res += '%4dk' % fdict['tbr']
c57f7757
PH
3448 if fdict.get('container') is not None:
3449 if res:
3450 res += ', '
3451 res += '%s container' % fdict['container']
3089bc74
S
3452 if (fdict.get('vcodec') is not None
3453 and fdict.get('vcodec') != 'none'):
c57f7757
PH
3454 if res:
3455 res += ', '
3456 res += fdict['vcodec']
91c7271a 3457 if fdict.get('vbr') is not None:
c57f7757
PH
3458 res += '@'
3459 elif fdict.get('vbr') is not None and fdict.get('abr') is not None:
3460 res += 'video@'
3461 if fdict.get('vbr') is not None:
3462 res += '%4dk' % fdict['vbr']
fbb21cf5 3463 if fdict.get('fps') is not None:
5d583bdf
S
3464 if res:
3465 res += ', '
3466 res += '%sfps' % fdict['fps']
c57f7757
PH
3467 if fdict.get('acodec') is not None:
3468 if res:
3469 res += ', '
3470 if fdict['acodec'] == 'none':
3471 res += 'video only'
3472 else:
3473 res += '%-5s' % fdict['acodec']
3474 elif fdict.get('abr') is not None:
3475 if res:
3476 res += ', '
3477 res += 'audio'
3478 if fdict.get('abr') is not None:
3479 res += '@%3dk' % fdict['abr']
3480 if fdict.get('asr') is not None:
3481 res += ' (%5dHz)' % fdict['asr']
3482 if fdict.get('filesize') is not None:
3483 if res:
3484 res += ', '
3485 res += format_bytes(fdict['filesize'])
9732d77e
PH
3486 elif fdict.get('filesize_approx') is not None:
3487 if res:
3488 res += ', '
3489 res += '~' + format_bytes(fdict['filesize_approx'])
c57f7757 3490 return res
91c7271a 3491
8130779d 3492 def render_formats_table(self, info_dict):
b69fd25c 3493 if not info_dict.get('formats') and not info_dict.get('url'):
8130779d 3494 return None
b69fd25c 3495
94badb25 3496 formats = info_dict.get('formats', [info_dict])
8130779d 3497 if not self.params.get('listformats_table', True) is not False:
76d321f6 3498 table = [
3499 [
3500 format_field(f, 'format_id'),
3501 format_field(f, 'ext'),
3502 self.format_resolution(f),
8130779d 3503 self._format_note(f)
3504 ] for f in formats if f.get('preference') is None or f['preference'] >= -1000]
3505 return render_table(['format code', 'extension', 'resolution', 'note'], table, extra_gap=1)
3506
3507 delim = self._format_screen('\u2502', self.Styles.DELIM, '|', test_encoding=True)
3508 table = [
3509 [
3510 self._format_screen(format_field(f, 'format_id'), self.Styles.ID),
3511 format_field(f, 'ext'),
3512 format_field(f, func=self.format_resolution, ignore=('audio only', 'images')),
3513 format_field(f, 'fps', '\t%d'),
3514 format_field(f, 'dynamic_range', '%s', ignore=(None, 'SDR')).replace('HDR', ''),
3515 delim,
3516 format_field(f, 'filesize', ' \t%s', func=format_bytes) + format_field(f, 'filesize_approx', '~\t%s', func=format_bytes),
3517 format_field(f, 'tbr', '\t%dk'),
3518 shorten_protocol_name(f.get('protocol', '')),
3519 delim,
3520 format_field(f, 'vcodec', default='unknown').replace(
3521 'none', 'images' if f.get('acodec') == 'none'
3522 else self._format_screen('audio only', self.Styles.SUPPRESS)),
3523 format_field(f, 'vbr', '\t%dk'),
3524 format_field(f, 'acodec', default='unknown').replace(
3525 'none', '' if f.get('vcodec') == 'none'
3526 else self._format_screen('video only', self.Styles.SUPPRESS)),
3527 format_field(f, 'abr', '\t%dk'),
3528 format_field(f, 'asr', '\t%dHz'),
3529 join_nonempty(
3530 self._format_screen('UNSUPPORTED', 'light red') if f.get('ext') in ('f4f', 'f4m') else None,
3531 format_field(f, 'language', '[%s]'),
3532 join_nonempty(format_field(f, 'format_note'),
3533 format_field(f, 'container', ignore=(None, f.get('ext'))),
3534 delim=', '),
3535 delim=' '),
3536 ] for f in formats if f.get('preference') is None or f['preference'] >= -1000]
3537 header_line = self._list_format_headers(
3538 'ID', 'EXT', 'RESOLUTION', '\tFPS', 'HDR', delim, '\tFILESIZE', '\tTBR', 'PROTO',
3539 delim, 'VCODEC', '\tVBR', 'ACODEC', '\tABR', '\tASR', 'MORE INFO')
3540
3541 return render_table(
3542 header_line, table, hide_empty=True,
3543 delim=self._format_screen('\u2500', self.Styles.DELIM, '-', test_encoding=True))
3544
3545 def render_thumbnails_table(self, info_dict):
88f23a18 3546 thumbnails = list(info_dict.get('thumbnails') or [])
cfb56d1a 3547 if not thumbnails:
8130779d 3548 return None
3549 return render_table(
ec11a9f4 3550 self._list_format_headers('ID', 'Width', 'Height', 'URL'),
6970b600 3551 [[t.get('id'), t.get('width', 'unknown'), t.get('height', 'unknown'), t['url']] for t in thumbnails])
2412044c 3552
8130779d 3553 def render_subtitles_table(self, video_id, subtitles):
2412044c 3554 def _row(lang, formats):
49c258e1 3555 exts, names = zip(*((f['ext'], f.get('name') or 'unknown') for f in reversed(formats)))
2412044c 3556 if len(set(names)) == 1:
7aee40c1 3557 names = [] if names[0] == 'unknown' else names[:1]
2412044c 3558 return [lang, ', '.join(names), ', '.join(exts)]
3559
8130779d 3560 if not subtitles:
3561 return None
3562 return render_table(
ec11a9f4 3563 self._list_format_headers('Language', 'Name', 'Formats'),
2412044c 3564 [_row(lang, formats) for lang, formats in subtitles.items()],
8130779d 3565 hide_empty=True)
3566
3567 def __list_table(self, video_id, name, func, *args):
3568 table = func(*args)
3569 if not table:
3570 self.to_screen(f'{video_id} has no {name}')
3571 return
3572 self.to_screen(f'[info] Available {name} for {video_id}:')
3573 self.to_stdout(table)
3574
3575 def list_formats(self, info_dict):
3576 self.__list_table(info_dict['id'], 'formats', self.render_formats_table, info_dict)
3577
3578 def list_thumbnails(self, info_dict):
3579 self.__list_table(info_dict['id'], 'thumbnails', self.render_thumbnails_table, info_dict)
3580
3581 def list_subtitles(self, video_id, subtitles, name='subtitles'):
3582 self.__list_table(video_id, name, self.render_subtitles_table, video_id, subtitles)
a504ced0 3583
dca08720
PH
3584 def urlopen(self, req):
3585 """ Start an HTTP download """
82d8a8b6 3586 if isinstance(req, compat_basestring):
67dda517 3587 req = sanitized_Request(req)
19a41fc6 3588 return self._opener.open(req, timeout=self._socket_timeout)
dca08720
PH
3589
3590 def print_debug_header(self):
3591 if not self.params.get('verbose'):
3592 return
49a57e70 3593
3594 def get_encoding(stream):
2a938746 3595 ret = str(getattr(stream, 'encoding', 'missing (%s)' % type(stream).__name__))
49a57e70 3596 if not supports_terminal_sequences(stream):
e3c7d495 3597 from .compat import WINDOWS_VT_MODE
3598 ret += ' (No VT)' if WINDOWS_VT_MODE is False else ' (No ANSI)'
49a57e70 3599 return ret
3600
3601 encoding_str = 'Encodings: locale %s, fs %s, out %s, err %s, pref %s' % (
3602 locale.getpreferredencoding(),
3603 sys.getfilesystemencoding(),
3604 get_encoding(self._screen_file), get_encoding(self._err_file),
3605 self.get_encoding())
883d4b1e 3606
3607 logger = self.params.get('logger')
3608 if logger:
3609 write_debug = lambda msg: logger.debug(f'[debug] {msg}')
3610 write_debug(encoding_str)
3611 else:
96565c7e 3612 write_string(f'[debug] {encoding_str}\n', encoding=None)
49a57e70 3613 write_debug = lambda msg: self._write_string(f'[debug] {msg}\n')
734f90bb 3614
4c88ff87 3615 source = detect_variant()
36eaf303 3616 write_debug(join_nonempty(
3617 'yt-dlp version', __version__,
3618 f'[{RELEASE_GIT_HEAD}]' if RELEASE_GIT_HEAD else '',
3619 '' if source == 'unknown' else f'({source})',
3620 delim=' '))
6e21fdd2 3621 if not _LAZY_LOADER:
3622 if os.environ.get('YTDLP_NO_LAZY_EXTRACTORS'):
49a57e70 3623 write_debug('Lazy loading extractors is forcibly disabled')
6e21fdd2 3624 else:
49a57e70 3625 write_debug('Lazy loading extractors is disabled')
3ae5e797 3626 if plugin_extractors or plugin_postprocessors:
49a57e70 3627 write_debug('Plugins: %s' % [
3ae5e797 3628 '%s%s' % (klass.__name__, '' if klass.__name__ == name else f' as {name}')
3629 for name, klass in itertools.chain(plugin_extractors.items(), plugin_postprocessors.items())])
53ed7066 3630 if self.params.get('compat_opts'):
49a57e70 3631 write_debug('Compatibility options: %s' % ', '.join(self.params.get('compat_opts')))
36eaf303 3632
3633 if source == 'source':
dca08720 3634 try:
36eaf303 3635 sp = Popen(
3636 ['git', 'rev-parse', '--short', 'HEAD'],
3637 stdout=subprocess.PIPE, stderr=subprocess.PIPE,
3638 cwd=os.path.dirname(os.path.abspath(__file__)))
3639 out, err = sp.communicate_or_kill()
3640 out = out.decode().strip()
3641 if re.match('[0-9a-f]+', out):
3642 write_debug('Git HEAD: %s' % out)
70a1165b 3643 except Exception:
36eaf303 3644 try:
3645 sys.exc_clear()
3646 except Exception:
3647 pass
b300cda4
S
3648
3649 def python_implementation():
3650 impl_name = platform.python_implementation()
3651 if impl_name == 'PyPy' and hasattr(sys, 'pypy_version_info'):
3652 return impl_name + ' version %d.%d.%d' % sys.pypy_version_info[:3]
3653 return impl_name
3654
49a57e70 3655 write_debug('Python version %s (%s %s) - %s' % (
e5813e53 3656 platform.python_version(),
3657 python_implementation(),
3658 platform.architecture()[0],
b300cda4 3659 platform_name()))
d28b5171 3660
8913ef74 3661 exe_versions, ffmpeg_features = FFmpegPostProcessor.get_versions_and_features(self)
3662 ffmpeg_features = {key for key, val in ffmpeg_features.items() if val}
3663 if ffmpeg_features:
a4211baf 3664 exe_versions['ffmpeg'] += ' (%s)' % ','.join(ffmpeg_features)
8913ef74 3665
4c83c967 3666 exe_versions['rtmpdump'] = rtmpdump_version()
feee8d32 3667 exe_versions['phantomjs'] = PhantomJSwrapper._version()
d28b5171 3668 exe_str = ', '.join(
2831b468 3669 f'{exe} {v}' for exe, v in sorted(exe_versions.items()) if v
3670 ) or 'none'
49a57e70 3671 write_debug('exe versions: %s' % exe_str)
dca08720 3672
2831b468 3673 from .downloader.websocket import has_websockets
3674 from .postprocessor.embedthumbnail import has_mutagen
f59f5ef8 3675 from .cookies import SQLITE_AVAILABLE, SECRETSTORAGE_AVAILABLE
2831b468 3676
c586f9e8 3677 lib_str = join_nonempty(
edf65256 3678 compat_pycrypto_AES and compat_pycrypto_AES.__name__.split('.')[0],
f59f5ef8 3679 SECRETSTORAGE_AVAILABLE and 'secretstorage',
2831b468 3680 has_mutagen and 'mutagen',
3681 SQLITE_AVAILABLE and 'sqlite',
c586f9e8 3682 has_websockets and 'websockets',
3683 delim=', ') or 'none'
49a57e70 3684 write_debug('Optional libraries: %s' % lib_str)
2831b468 3685
dca08720
PH
3686 proxy_map = {}
3687 for handler in self._opener.handlers:
3688 if hasattr(handler, 'proxies'):
3689 proxy_map.update(handler.proxies)
49a57e70 3690 write_debug(f'Proxy map: {proxy_map}')
dca08720 3691
49a57e70 3692 # Not implemented
3693 if False and self.params.get('call_home'):
58b1f00d 3694 ipaddr = self.urlopen('https://yt-dl.org/ip').read().decode('utf-8')
49a57e70 3695 write_debug('Public IP address: %s' % ipaddr)
58b1f00d
PH
3696 latest_version = self.urlopen(
3697 'https://yt-dl.org/latest/version').read().decode('utf-8')
3698 if version_tuple(latest_version) > version_tuple(__version__):
3699 self.report_warning(
3700 'You are using an outdated version (newest version: %s)! '
3701 'See https://yt-dl.org/update if you need help updating.' %
3702 latest_version)
3703
e344693b 3704 def _setup_opener(self):
6ad14cab 3705 timeout_val = self.params.get('socket_timeout')
17bddf3e 3706 self._socket_timeout = 20 if timeout_val is None else float(timeout_val)
6ad14cab 3707
982ee69a 3708 opts_cookiesfrombrowser = self.params.get('cookiesfrombrowser')
dca08720
PH
3709 opts_cookiefile = self.params.get('cookiefile')
3710 opts_proxy = self.params.get('proxy')
3711
982ee69a 3712 self.cookiejar = load_cookies(opts_cookiefile, opts_cookiesfrombrowser, self)
dca08720 3713
6a3f4c3f 3714 cookie_processor = YoutubeDLCookieProcessor(self.cookiejar)
dca08720
PH
3715 if opts_proxy is not None:
3716 if opts_proxy == '':
3717 proxies = {}
3718 else:
3719 proxies = {'http': opts_proxy, 'https': opts_proxy}
3720 else:
3721 proxies = compat_urllib_request.getproxies()
067aa17e 3722 # Set HTTPS proxy to HTTP one if given (https://github.com/ytdl-org/youtube-dl/issues/805)
dca08720
PH
3723 if 'http' in proxies and 'https' not in proxies:
3724 proxies['https'] = proxies['http']
91410c9b 3725 proxy_handler = PerRequestProxyHandler(proxies)
a0ddb8a2
PH
3726
3727 debuglevel = 1 if self.params.get('debug_printtraffic') else 0
be4a824d
PH
3728 https_handler = make_HTTPS_handler(self.params, debuglevel=debuglevel)
3729 ydlh = YoutubeDLHandler(self.params, debuglevel=debuglevel)
fca6dba8 3730 redirect_handler = YoutubeDLRedirectHandler()
8b172c2e 3731 data_handler = compat_urllib_request_DataHandler()
6240b0a2
JMF
3732
3733 # When passing our own FileHandler instance, build_opener won't add the
3734 # default FileHandler and allows us to disable the file protocol, which
3735 # can be used for malicious purposes (see
067aa17e 3736 # https://github.com/ytdl-org/youtube-dl/issues/8227)
6240b0a2
JMF
3737 file_handler = compat_urllib_request.FileHandler()
3738
3739 def file_open(*args, **kwargs):
7a5c1cfe 3740 raise compat_urllib_error.URLError('file:// scheme is explicitly disabled in yt-dlp for security reasons')
6240b0a2
JMF
3741 file_handler.file_open = file_open
3742
3743 opener = compat_urllib_request.build_opener(
fca6dba8 3744 proxy_handler, https_handler, cookie_processor, ydlh, redirect_handler, data_handler, file_handler)
2461f79d 3745
dca08720
PH
3746 # Delete the default user-agent header, which would otherwise apply in
3747 # cases where our custom HTTP handler doesn't come into play
067aa17e 3748 # (See https://github.com/ytdl-org/youtube-dl/issues/1309 for details)
dca08720
PH
3749 opener.addheaders = []
3750 self._opener = opener
62fec3b2
PH
3751
3752 def encode(self, s):
3753 if isinstance(s, bytes):
3754 return s # Already encoded
3755
3756 try:
3757 return s.encode(self.get_encoding())
3758 except UnicodeEncodeError as err:
3759 err.reason = err.reason + '. Check your system encoding configuration or use the --encoding option.'
3760 raise
3761
3762 def get_encoding(self):
3763 encoding = self.params.get('encoding')
3764 if encoding is None:
3765 encoding = preferredencoding()
3766 return encoding
ec82d85a 3767
e08a85d8 3768 def _write_info_json(self, label, ie_result, infofn, overwrite=None):
80c03fa9 3769 ''' Write infojson and returns True = written, False = skip, None = error '''
e08a85d8 3770 if overwrite is None:
3771 overwrite = self.params.get('overwrites', True)
80c03fa9 3772 if not self.params.get('writeinfojson'):
3773 return False
3774 elif not infofn:
3775 self.write_debug(f'Skipping writing {label} infojson')
3776 return False
3777 elif not self._ensure_dir_exists(infofn):
3778 return None
e08a85d8 3779 elif not overwrite and os.path.exists(infofn):
80c03fa9 3780 self.to_screen(f'[info] {label.title()} metadata is already present')
3781 else:
3782 self.to_screen(f'[info] Writing {label} metadata as JSON to: {infofn}')
3783 try:
3784 write_json_file(self.sanitize_info(ie_result, self.params.get('clean_infojson', True)), infofn)
3785 except (OSError, IOError):
3786 self.report_error(f'Cannot write {label} metadata to JSON file {infofn}')
3787 return None
3788 return True
3789
3790 def _write_description(self, label, ie_result, descfn):
3791 ''' Write description and returns True = written, False = skip, None = error '''
3792 if not self.params.get('writedescription'):
3793 return False
3794 elif not descfn:
3795 self.write_debug(f'Skipping writing {label} description')
3796 return False
3797 elif not self._ensure_dir_exists(descfn):
3798 return None
3799 elif not self.params.get('overwrites', True) and os.path.exists(descfn):
3800 self.to_screen(f'[info] {label.title()} description is already present')
3801 elif ie_result.get('description') is None:
3802 self.report_warning(f'There\'s no {label} description to write')
3803 return False
3804 else:
3805 try:
3806 self.to_screen(f'[info] Writing {label} description to: {descfn}')
3807 with io.open(encodeFilename(descfn), 'w', encoding='utf-8') as descfile:
3808 descfile.write(ie_result['description'])
3809 except (OSError, IOError):
3810 self.report_error(f'Cannot write {label} description file {descfn}')
3811 return None
3812 return True
3813
3814 def _write_subtitles(self, info_dict, filename):
3815 ''' Write subtitles to file and return list of (sub_filename, final_sub_filename); or None if error'''
3816 ret = []
3817 subtitles = info_dict.get('requested_subtitles')
3818 if not subtitles or not (self.params.get('writesubtitles') or self.params.get('writeautomaticsub')):
3819 # subtitles download errors are already managed as troubles in relevant IE
3820 # that way it will silently go on when used with unsupporting IE
3821 return ret
3822
3823 sub_filename_base = self.prepare_filename(info_dict, 'subtitle')
3824 if not sub_filename_base:
3825 self.to_screen('[info] Skipping writing video subtitles')
3826 return ret
3827 for sub_lang, sub_info in subtitles.items():
3828 sub_format = sub_info['ext']
3829 sub_filename = subtitles_filename(filename, sub_lang, sub_format, info_dict.get('ext'))
3830 sub_filename_final = subtitles_filename(sub_filename_base, sub_lang, sub_format, info_dict.get('ext'))
e04938ab 3831 existing_sub = self.existing_file((sub_filename_final, sub_filename))
3832 if existing_sub:
80c03fa9 3833 self.to_screen(f'[info] Video subtitle {sub_lang}.{sub_format} is already present')
e04938ab 3834 sub_info['filepath'] = existing_sub
3835 ret.append((existing_sub, sub_filename_final))
80c03fa9 3836 continue
3837
3838 self.to_screen(f'[info] Writing video subtitles to: {sub_filename}')
3839 if sub_info.get('data') is not None:
3840 try:
3841 # Use newline='' to prevent conversion of newline characters
3842 # See https://github.com/ytdl-org/youtube-dl/issues/10268
3843 with io.open(sub_filename, 'w', encoding='utf-8', newline='') as subfile:
3844 subfile.write(sub_info['data'])
3845 sub_info['filepath'] = sub_filename
3846 ret.append((sub_filename, sub_filename_final))
3847 continue
3848 except (OSError, IOError):
3849 self.report_error(f'Cannot write video subtitles file {sub_filename}')
3850 return None
3851
3852 try:
3853 sub_copy = sub_info.copy()
3854 sub_copy.setdefault('http_headers', info_dict.get('http_headers'))
3855 self.dl(sub_filename, sub_copy, subtitle=True)
3856 sub_info['filepath'] = sub_filename
3857 ret.append((sub_filename, sub_filename_final))
6020e05d 3858 except (DownloadError, ExtractorError, IOError, OSError, ValueError) + network_exceptions as err:
3859 if self.params.get('ignoreerrors') is not True: # False or 'only_download'
3860 raise DownloadError(f'Unable to download video subtitles for {sub_lang!r}: {err}', err)
80c03fa9 3861 self.report_warning(f'Unable to download video subtitles for {sub_lang!r}: {err}')
519804a9 3862 return ret
80c03fa9 3863
3864 def _write_thumbnails(self, label, info_dict, filename, thumb_filename_base=None):
3865 ''' Write thumbnails to file and return list of (thumb_filename, final_thumb_filename) '''
6c4fd172 3866 write_all = self.params.get('write_all_thumbnails', False)
80c03fa9 3867 thumbnails, ret = [], []
6c4fd172 3868 if write_all or self.params.get('writethumbnail', False):
0202b52a 3869 thumbnails = info_dict.get('thumbnails') or []
6c4fd172 3870 multiple = write_all and len(thumbnails) > 1
ec82d85a 3871
80c03fa9 3872 if thumb_filename_base is None:
3873 thumb_filename_base = filename
3874 if thumbnails and not thumb_filename_base:
3875 self.write_debug(f'Skipping writing {label} thumbnail')
3876 return ret
3877
dd0228ce 3878 for idx, t in list(enumerate(thumbnails))[::-1]:
80c03fa9 3879 thumb_ext = (f'{t["id"]}.' if multiple else '') + determine_ext(t['url'], 'jpg')
aa9369a2 3880 thumb_display_id = f'{label} thumbnail {t["id"]}'
80c03fa9 3881 thumb_filename = replace_extension(filename, thumb_ext, info_dict.get('ext'))
3882 thumb_filename_final = replace_extension(thumb_filename_base, thumb_ext, info_dict.get('ext'))
ec82d85a 3883
e04938ab 3884 existing_thumb = self.existing_file((thumb_filename_final, thumb_filename))
3885 if existing_thumb:
aa9369a2 3886 self.to_screen('[info] %s is already present' % (
3887 thumb_display_id if multiple else f'{label} thumbnail').capitalize())
e04938ab 3888 t['filepath'] = existing_thumb
3889 ret.append((existing_thumb, thumb_filename_final))
ec82d85a 3890 else:
80c03fa9 3891 self.to_screen(f'[info] Downloading {thumb_display_id} ...')
ec82d85a 3892 try:
297e9952 3893 uf = self.urlopen(sanitized_Request(t['url'], headers=t.get('http_headers', {})))
80c03fa9 3894 self.to_screen(f'[info] Writing {thumb_display_id} to: {thumb_filename}')
d3d89c32 3895 with open(encodeFilename(thumb_filename), 'wb') as thumbf:
ec82d85a 3896 shutil.copyfileobj(uf, thumbf)
80c03fa9 3897 ret.append((thumb_filename, thumb_filename_final))
885cc0b7 3898 t['filepath'] = thumb_filename
3158150c 3899 except network_exceptions as err:
dd0228ce 3900 thumbnails.pop(idx)
80c03fa9 3901 self.report_warning(f'Unable to download {thumb_display_id}: {err}')
6c4fd172 3902 if ret and not write_all:
3903 break
0202b52a 3904 return ret