]> jfr.im git - yt-dlp.git/blame - yt_dlp/YoutubeDL.py
Fix `-J` when there are failed videos
[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
317f7ab6 8import copy
9d2ecdbc 9import datetime
c1c9a79c 10import errno
31bd3925 11import fileinput
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
8222d8de 28
961ea474 29from string import ascii_letters
e5813e53 30from zipimport import zipimporter
961ea474 31
8c25f81b 32from .compat import (
82d8a8b6 33 compat_basestring,
003c69a8 34 compat_get_terminal_size,
4f026faf 35 compat_kwargs,
d0d9ade4 36 compat_numeric_types,
e9c0cdd3 37 compat_os_name,
7d1eb38a 38 compat_shlex_quote,
ce02ed60 39 compat_str,
67134eab 40 compat_tokenize_tokenize,
ce02ed60
PH
41 compat_urllib_error,
42 compat_urllib_request,
8b172c2e 43 compat_urllib_request_DataHandler,
8c25f81b 44)
982ee69a 45from .cookies import load_cookies
8c25f81b 46from .utils import (
eedb7ba5
S
47 age_restricted,
48 args_to_str,
ce02ed60
PH
49 ContentTooShortError,
50 date_from_str,
51 DateRange,
acd69589 52 DEFAULT_OUTTMPL,
ce02ed60 53 determine_ext,
b5559424 54 determine_protocol,
732044af 55 DOT_DESKTOP_LINK_TEMPLATE,
56 DOT_URL_LINK_TEMPLATE,
57 DOT_WEBLOC_LINK_TEMPLATE,
ce02ed60 58 DownloadError,
c0384f22 59 encode_compat_str,
ce02ed60 60 encodeFilename,
498f5606 61 EntryNotInPlaylist,
a06916d9 62 error_to_compat_str,
8b0d7497 63 ExistingVideoReached,
590bc6f6 64 expand_path,
ce02ed60 65 ExtractorError,
e29663c6 66 float_or_none,
02dbf93f 67 format_bytes,
76d321f6 68 format_field,
901130bb 69 STR_FORMAT_RE_TMPL,
70 STR_FORMAT_TYPES,
525ef922 71 formatSeconds,
773f291d 72 GeoRestrictedError,
b0249bca 73 HEADRequest,
c9969434 74 int_or_none,
732044af 75 iri_to_uri,
773f291d 76 ISO3166Utils,
56a8fb4f 77 LazyList,
ce02ed60 78 locked_file,
0202b52a 79 make_dir,
dca08720 80 make_HTTPS_handler,
ce02ed60 81 MaxDownloadsReached,
3158150c 82 network_exceptions,
cd6fc19e 83 orderedSet,
a06916d9 84 OUTTMPL_TYPES,
b7ab0590 85 PagedList,
083c9df9 86 parse_filesize,
91410c9b 87 PerRequestProxyHandler,
dca08720 88 platform_name,
eedb7ba5 89 PostProcessingError,
ce02ed60 90 preferredencoding,
eedb7ba5 91 prepend_extension,
a06916d9 92 process_communicate_or_kill,
51fb4995 93 register_socks_protocols,
a06916d9 94 RejectedVideoReached,
cfb56d1a 95 render_table,
eedb7ba5 96 replace_extension,
ce02ed60
PH
97 SameFileError,
98 sanitize_filename,
1bb5c511 99 sanitize_path,
dcf77cf1 100 sanitize_url,
67dda517 101 sanitized_Request,
e5660ee6 102 std_headers,
1211bb6d 103 str_or_none,
e29663c6 104 strftime_or_none,
ce02ed60 105 subtitles_filename,
51d9739f 106 ThrottledDownload,
732044af 107 to_high_limit_path,
324ad820 108 traverse_obj,
6033d980 109 try_get,
ce02ed60 110 UnavailableVideoError,
29eb5174 111 url_basename,
7d1eb38a 112 variadic,
58b1f00d 113 version_tuple,
ce02ed60
PH
114 write_json_file,
115 write_string,
6a3f4c3f 116 YoutubeDLCookieProcessor,
dca08720 117 YoutubeDLHandler,
fca6dba8 118 YoutubeDLRedirectHandler,
ce02ed60 119)
a0e07d31 120from .cache import Cache
52a8a1e1 121from .extractor import (
122 gen_extractor_classes,
123 get_info_extractor,
124 _LAZY_LOADER,
125 _PLUGIN_CLASSES
126)
4c54b89e 127from .extractor.openload import PhantomJSwrapper
52a8a1e1 128from .downloader import (
dbf5416a 129 FFmpegFD,
52a8a1e1 130 get_suitable_downloader,
131 shorten_protocol_name
132)
4c83c967 133from .downloader.rtmp import rtmpdump_version
4f026faf 134from .postprocessor import (
e36d50c5 135 get_postprocessor,
136 FFmpegFixupDurationPP,
f17f8651 137 FFmpegFixupM3u8PP,
62cd676c 138 FFmpegFixupM4aPP,
6271f1ca 139 FFmpegFixupStretchedPP,
e36d50c5 140 FFmpegFixupTimestampPP,
4f026faf
PH
141 FFmpegMergerPP,
142 FFmpegPostProcessor,
0202b52a 143 MoveFilesAfterDownloadPP,
4f026faf 144)
dca08720 145from .version import __version__
8222d8de 146
e9c0cdd3
YCH
147if compat_os_name == 'nt':
148 import ctypes
149
2459b6e1 150
8222d8de
JMF
151class YoutubeDL(object):
152 """YoutubeDL class.
153
154 YoutubeDL objects are the ones responsible of downloading the
155 actual video file and writing it to disk if the user has requested
156 it, among some other tasks. In most cases there should be one per
157 program. As, given a video URL, the downloader doesn't know how to
158 extract all the needed information, task that InfoExtractors do, it
159 has to pass the URL to one of them.
160
161 For this, YoutubeDL objects have a method that allows
162 InfoExtractors to be registered in a given order. When it is passed
163 a URL, the YoutubeDL object handles it to the first InfoExtractor it
164 finds that reports being able to handle it. The InfoExtractor extracts
165 all the information about the video or videos the URL refers to, and
166 YoutubeDL process the extracted information, possibly using a File
167 Downloader to download the video.
168
169 YoutubeDL objects accept a lot of parameters. In order not to saturate
170 the object constructor with arguments, it receives a dictionary of
171 options instead. These options are available through the params
172 attribute for the InfoExtractors to use. The YoutubeDL also
173 registers itself as the downloader in charge for the InfoExtractors
174 that are added to it, so this is a "mutual registration".
175
176 Available options:
177
178 username: Username for authentication purposes.
179 password: Password for authentication purposes.
180940e0 180 videopassword: Password for accessing a video.
1da50aa3
S
181 ap_mso: Adobe Pass multiple-system operator identifier.
182 ap_username: Multiple-system operator account username.
183 ap_password: Multiple-system operator account password.
8222d8de
JMF
184 usenetrc: Use netrc for authentication instead.
185 verbose: Print additional info to stdout.
186 quiet: Do not print messages to stdout.
ad8915b7 187 no_warnings: Do not print out anything for warnings.
53c18592 188 forceprint: A list of templates to force print
189 forceurl: Force printing final URL. (Deprecated)
190 forcetitle: Force printing title. (Deprecated)
191 forceid: Force printing ID. (Deprecated)
192 forcethumbnail: Force printing thumbnail URL. (Deprecated)
193 forcedescription: Force printing description. (Deprecated)
194 forcefilename: Force printing final filename. (Deprecated)
195 forceduration: Force printing duration. (Deprecated)
8694c600 196 forcejson: Force printing info_dict as JSON.
63e0be34
PH
197 dump_single_json: Force printing the info_dict of the whole playlist
198 (or video) as a single JSON line.
c25228e5 199 force_write_download_archive: Force writing download archive regardless
200 of 'skip_download' or 'simulate'.
b7b04c78 201 simulate: Do not download the video files. If unset (or None),
202 simulate only if listsubtitles, listformats or list_thumbnails is used
eb8a4433 203 format: Video format code. see "FORMAT SELECTION" for more details.
63ad4d43 204 allow_unplayable_formats: Allow unplayable formats to be extracted and downloaded.
b7da73eb 205 ignore_no_formats_error: Ignore "No video formats" error. Usefull for
206 extracting metadata even if the video is not actually
207 available for download (experimental)
c25228e5 208 format_sort: How to sort the video formats. see "Sorting Formats"
209 for more details.
210 format_sort_force: Force the given format_sort. see "Sorting Formats"
211 for more details.
212 allow_multiple_video_streams: Allow multiple video streams to be merged
213 into a single file
214 allow_multiple_audio_streams: Allow multiple audio streams to be merged
215 into a single file
0ba692ac 216 check_formats Whether to test if the formats are downloadable.
217 Can be True (check all), False (check none)
218 or None (check only if requested by extractor)
4524baf0 219 paths: Dictionary of output paths. The allowed keys are 'home'
220 'temp' and the keys of OUTTMPL_TYPES (in utils.py)
de6000d9 221 outtmpl: Dictionary of templates for output names. Allowed keys
4524baf0 222 are 'default' and the keys of OUTTMPL_TYPES (in utils.py).
34488702 223 For compatibility with youtube-dl, a single string can also be used
a820dc72
RA
224 outtmpl_na_placeholder: Placeholder for unavailable meta fields.
225 restrictfilenames: Do not allow "&" and spaces in file names
226 trim_file_name: Limit length of filename (extension excluded)
4524baf0 227 windowsfilenames: Force the filenames to be windows compatible
a820dc72 228 ignoreerrors: Do not stop on download errors
7a5c1cfe 229 (Default True when running yt-dlp,
a820dc72 230 but False when directly accessing YoutubeDL class)
26e2805c 231 skip_playlist_after_errors: Number of allowed failures until the rest of
232 the playlist is skipped
d22dec74 233 force_generic_extractor: Force downloader to use the generic extractor
0c3d0f51 234 overwrites: Overwrite all video and metadata files if True,
235 overwrite only non-video files if None
236 and don't overwrite any file if False
34488702 237 For compatibility with youtube-dl,
238 "nooverwrites" may also be used instead
8222d8de
JMF
239 playliststart: Playlist item to start at.
240 playlistend: Playlist item to end at.
c14e88f0 241 playlist_items: Specific indices of playlist to download.
ff815fe6 242 playlistreverse: Download playlist items in reverse order.
75822ca7 243 playlistrandom: Download playlist items in random order.
8222d8de
JMF
244 matchtitle: Download only matching titles.
245 rejecttitle: Reject downloads for matching titles.
8bf9319e 246 logger: Log messages to a logging.Logger instance.
8222d8de
JMF
247 logtostderr: Log messages to stderr instead of stdout.
248 writedescription: Write the video description to a .description file
249 writeinfojson: Write the video description to a .info.json file
75d43ca0 250 clean_infojson: Remove private fields from the infojson
34488702 251 getcomments: Extract video comments. This will not be written to disk
06167fbb 252 unless writeinfojson is also given
1fb07d10 253 writeannotations: Write the video annotations to a .annotations.xml file
8222d8de 254 writethumbnail: Write the thumbnail image to a file
c25228e5 255 allow_playlist_files: Whether to write playlists' description, infojson etc
256 also to disk when using the 'write*' options
ec82d85a 257 write_all_thumbnails: Write all thumbnail formats to files
732044af 258 writelink: Write an internet shortcut file, depending on the
259 current platform (.url/.webloc/.desktop)
260 writeurllink: Write a Windows internet shortcut file (.url)
261 writewebloclink: Write a macOS internet shortcut file (.webloc)
262 writedesktoplink: Write a Linux internet shortcut file (.desktop)
8222d8de 263 writesubtitles: Write the video subtitles to a file
741dd8ea 264 writeautomaticsub: Write the automatically generated subtitles to a file
245524e6 265 allsubtitles: Deprecated - Use subtitleslangs = ['all']
c32b0aab 266 Downloads all the subtitles of the video
0b7f3118 267 (requires writesubtitles or writeautomaticsub)
8222d8de 268 listsubtitles: Lists all available subtitles for the video
a504ced0 269 subtitlesformat: The format code for subtitles
c32b0aab 270 subtitleslangs: List of languages of the subtitles to download (can be regex).
271 The list may contain "all" to refer to all the available
272 subtitles. The language can be prefixed with a "-" to
273 exclude it from the requested languages. Eg: ['all', '-live_chat']
8222d8de
JMF
274 keepvideo: Keep the video file after post-processing
275 daterange: A DateRange object, download only if the upload_date is in the range.
276 skip_download: Skip the actual download of the video file
c35f9e72 277 cachedir: Location of the cache files in the filesystem.
a0e07d31 278 False to disable filesystem cache.
47192f92 279 noplaylist: Download single video instead of a playlist if in doubt.
8dbe9899
PH
280 age_limit: An integer representing the user's age in years.
281 Unsuitable videos for the given age are skipped.
5fe18bdb
PH
282 min_views: An integer representing the minimum view count the video
283 must have in order to not be skipped.
284 Videos without view count information are always
285 downloaded. None for no limit.
286 max_views: An integer representing the maximum view count.
287 Videos that are more popular than that are not
288 downloaded.
289 Videos without view count information are always
290 downloaded. None for no limit.
291 download_archive: File name of a file where all downloads are recorded.
c1c9a79c
PH
292 Videos already present in the file are not downloaded
293 again.
8a51f564 294 break_on_existing: Stop the download process after attempting to download a
295 file that is in the archive.
296 break_on_reject: Stop the download process when encountering a video that
297 has been filtered out.
298 cookiefile: File name where cookies should be read from and dumped to
982ee69a
MB
299 cookiesfrombrowser: A tuple containing the name of the browser and the profile
300 name/path from where cookies are loaded.
301 Eg: ('chrome', ) or (vivaldi, 'default')
a1ee09e8 302 nocheckcertificate:Do not verify SSL certificates
7e8c0af0
PH
303 prefer_insecure: Use HTTP instead of HTTPS to retrieve information.
304 At the moment, this is only supported by YouTube.
a1ee09e8 305 proxy: URL of the proxy server to use
38cce791 306 geo_verification_proxy: URL of the proxy to use for IP address verification
504f20dd 307 on geo-restricted sites.
e344693b 308 socket_timeout: Time to wait for unresponsive hosts, in seconds
0783b09b
PH
309 bidi_workaround: Work around buggy terminals without bidirectional text
310 support, using fridibi
a0ddb8a2 311 debug_printtraffic:Print out sent and received HTTP traffic
7b0817e8 312 include_ads: Download ads as well
04b4d394
PH
313 default_search: Prepend this string if an input url is not valid.
314 'auto' for elaborate guessing
62fec3b2 315 encoding: Use this encoding instead of the system-specified.
e8ee972c 316 extract_flat: Do not resolve URLs, return the immediate result.
057a5206
PH
317 Pass in 'in_playlist' to only show this behavior for
318 playlist items.
4f026faf 319 postprocessors: A list of dictionaries, each with an entry
71b640cc 320 * key: The name of the postprocessor. See
7a5c1cfe 321 yt_dlp/postprocessor/__init__.py for a list.
56d868db 322 * when: When to run the postprocessor. Can be one of
323 pre_process|before_dl|post_process|after_move.
324 Assumed to be 'post_process' if not given
ab8e5e51
AM
325 post_hooks: A list of functions that get called as the final step
326 for each video file, after all postprocessors have been
327 called. The filename will be passed as the only argument.
71b640cc
PH
328 progress_hooks: A list of functions that get called on download
329 progress, with a dictionary with the entries
5cda4eda 330 * status: One of "downloading", "error", or "finished".
ee69b99a 331 Check this first and ignore unknown values.
3ba7740d 332 * info_dict: The extracted info_dict
71b640cc 333
5cda4eda 334 If status is one of "downloading", or "finished", the
ee69b99a
PH
335 following properties may also be present:
336 * filename: The final filename (always present)
5cda4eda 337 * tmpfilename: The filename we're currently writing to
71b640cc
PH
338 * downloaded_bytes: Bytes on disk
339 * total_bytes: Size of the whole file, None if unknown
5cda4eda
PH
340 * total_bytes_estimate: Guess of the eventual file size,
341 None if unavailable.
342 * elapsed: The number of seconds since download started.
71b640cc
PH
343 * eta: The estimated time in seconds, None if unknown
344 * speed: The download speed in bytes/second, None if
345 unknown
5cda4eda
PH
346 * fragment_index: The counter of the currently
347 downloaded video fragment.
348 * fragment_count: The number of fragments (= individual
349 files that will be merged)
71b640cc
PH
350
351 Progress hooks are guaranteed to be called at least once
352 (with status "finished") if the download is successful.
45598f15 353 merge_output_format: Extension to use when merging formats.
6b591b29 354 final_ext: Expected final extension; used to detect when the file was
355 already downloaded and converted. "merge_output_format" is
356 replaced by this extension when given
6271f1ca
PH
357 fixup: Automatically correct known faults of the file.
358 One of:
359 - "never": do nothing
360 - "warn": only emit a warning
361 - "detect_or_warn": check whether we can do anything
62cd676c 362 about it, warn otherwise (default)
504f20dd 363 source_address: Client-side IP address to bind to.
6ec6cb4e 364 call_home: Boolean, true iff we are allowed to contact the
7a5c1cfe 365 yt-dlp servers for debugging. (BROKEN)
1cf376f5 366 sleep_interval_requests: Number of seconds to sleep between requests
367 during extraction
7aa589a5
S
368 sleep_interval: Number of seconds to sleep before each download when
369 used alone or a lower bound of a range for randomized
370 sleep before each download (minimum possible number
371 of seconds to sleep) when used along with
372 max_sleep_interval.
373 max_sleep_interval:Upper bound of a range for randomized sleep before each
374 download (maximum possible number of seconds to sleep).
375 Must only be used along with sleep_interval.
376 Actual sleep time will be a random float from range
377 [sleep_interval; max_sleep_interval].
1cf376f5 378 sleep_interval_subtitles: Number of seconds to sleep before each subtitle download
cfb56d1a
PH
379 listformats: Print an overview of available video formats and exit.
380 list_thumbnails: Print a table of all thumbnails and exit.
347de493
PH
381 match_filter: A function that gets called with the info_dict of
382 every video.
383 If it returns a message, the video is ignored.
384 If it returns None, the video is downloaded.
385 match_filter_func in utils.py is one example for this.
7e5db8c9 386 no_color: Do not emit color codes in output.
0a840f58 387 geo_bypass: Bypass geographic restriction via faking X-Forwarded-For
504f20dd 388 HTTP header
0a840f58 389 geo_bypass_country:
773f291d
S
390 Two-letter ISO 3166-2 country code that will be used for
391 explicit geographic restriction bypassing via faking
504f20dd 392 X-Forwarded-For HTTP header
5f95927a
S
393 geo_bypass_ip_block:
394 IP range in CIDR notation that will be used similarly to
504f20dd 395 geo_bypass_country
71b640cc 396
85729c51 397 The following options determine which downloader is picked:
52a8a1e1 398 external_downloader: A dictionary of protocol keys and the executable of the
399 external downloader to use for it. The allowed protocols
400 are default|http|ftp|m3u8|dash|rtsp|rtmp|mms.
401 Set the value to 'native' to use the native downloader
402 hls_prefer_native: Deprecated - Use external_downloader = {'m3u8': 'native'}
403 or {'m3u8': 'ffmpeg'} instead.
404 Use the native HLS downloader instead of ffmpeg/avconv
bf09af3a
S
405 if True, otherwise use ffmpeg/avconv if False, otherwise
406 use downloader suggested by extractor if None.
53ed7066 407 compat_opts: Compatibility options. See "Differences in default behavior".
3acf6d38 408 The following options do not work when used through the API:
409 filename, abort-on-error, multistreams, no-live-chat,
b51d2ae3 410 no-clean-infojson, no-playlist-metafiles, no-keep-subs.
e4f02757 411 Refer __init__.py for their implementation
fe7e0c98 412
8222d8de 413 The following parameters are not used by YoutubeDL itself, they are used by
7a5c1cfe 414 the downloader (see yt_dlp/downloader/common.py):
51d9739f 415 nopart, updatetime, buffersize, ratelimit, throttledratelimit, min_filesize,
416 max_filesize, test, noresizebuffer, retries, continuedl, noprogress, consoletitle,
417 xattr_set_filesize, external_downloader_args, hls_use_mpegts, http_chunk_size.
76b1bd67
JMF
418
419 The following options are used by the post processors:
d4a24f40 420 prefer_ffmpeg: If False, use avconv instead of ffmpeg if both are available,
e4172ac9 421 otherwise prefer ffmpeg. (avconv support is deprecated)
c0b7d117
S
422 ffmpeg_location: Location of the ffmpeg/avconv binary; either the path
423 to the binary or its containing directory.
43820c03 424 postprocessor_args: A dictionary of postprocessor/executable keys (in lower case)
34488702 425 and a list of additional command-line arguments for the
426 postprocessor/executable. The dict can also have "PP+EXE" keys
427 which are used when the given exe is used by the given PP.
428 Use 'default' as the name for arguments to passed to all PP
429 For compatibility with youtube-dl, a single list of args
430 can also be used
e409895f 431
432 The following options are used by the extractors:
62bff2c1 433 extractor_retries: Number of times to retry for known errors
434 dynamic_mpd: Whether to process dynamic DASH manifests (default: True)
e409895f 435 hls_split_discontinuity: Split HLS playlists to different formats at
62bff2c1 436 discontinuities such as ad breaks (default: False)
5d3a0e79 437 extractor_args: A dictionary of arguments to be passed to the extractors.
438 See "EXTRACTOR ARGUMENTS" for details.
439 Eg: {'youtube': {'skip': ['dash', 'hls']}}
440 youtube_include_dash_manifest: Deprecated - Use extractor_args instead.
441 If True (default), DASH manifests and related
62bff2c1 442 data will be downloaded and processed by extractor.
443 You can reduce network I/O by disabling it if you don't
444 care about DASH. (only for youtube)
5d3a0e79 445 youtube_include_hls_manifest: Deprecated - Use extractor_args instead.
446 If True (default), HLS manifests and related
62bff2c1 447 data will be downloaded and processed by extractor.
448 You can reduce network I/O by disabling it if you don't
449 care about HLS. (only for youtube)
8222d8de
JMF
450 """
451
c9969434
S
452 _NUMERIC_FIELDS = set((
453 'width', 'height', 'tbr', 'abr', 'asr', 'vbr', 'fps', 'filesize', 'filesize_approx',
454 'timestamp', 'upload_year', 'upload_month', 'upload_day',
455 'duration', 'view_count', 'like_count', 'dislike_count', 'repost_count',
456 'average_rating', 'comment_count', 'age_limit',
457 'start_time', 'end_time',
458 'chapter_number', 'season_number', 'episode_number',
459 'track_number', 'disc_number', 'release_year',
460 'playlist_index',
461 ))
462
8222d8de
JMF
463 params = None
464 _ies = []
56d868db 465 _pps = {'pre_process': [], 'before_dl': [], 'after_move': [], 'post_process': []}
b35496d8 466 _printed_messages = set()
1cf376f5 467 _first_webpage_request = True
8222d8de
JMF
468 _download_retcode = None
469 _num_downloads = None
30a074c2 470 _playlist_level = 0
471 _playlist_urls = set()
8222d8de
JMF
472 _screen_file = None
473
3511266b 474 def __init__(self, params=None, auto_init=True):
8222d8de 475 """Create a FileDownloader object with the given options."""
e9f9a10f
JMF
476 if params is None:
477 params = {}
8222d8de 478 self._ies = []
56c73665 479 self._ies_instances = {}
56d868db 480 self._pps = {'pre_process': [], 'before_dl': [], 'after_move': [], 'post_process': []}
b35496d8 481 self._printed_messages = set()
1cf376f5 482 self._first_webpage_request = True
ab8e5e51 483 self._post_hooks = []
933605d7 484 self._progress_hooks = []
8222d8de
JMF
485 self._download_retcode = 0
486 self._num_downloads = 0
487 self._screen_file = [sys.stdout, sys.stderr][params.get('logtostderr', False)]
0783b09b 488 self._err_file = sys.stderr
4abf617b
S
489 self.params = {
490 # Default parameters
491 'nocheckcertificate': False,
492 }
493 self.params.update(params)
a0e07d31 494 self.cache = Cache(self)
34308b30 495
a61f4b28 496 if sys.version_info < (3, 6):
497 self.report_warning(
0181adef 498 'Python version %d.%d is not supported! Please update to Python 3.6 or above' % sys.version_info[:2])
a61f4b28 499
be5df5ee
S
500 def check_deprecated(param, option, suggestion):
501 if self.params.get(param) is not None:
53ed7066 502 self.report_warning('%s is deprecated. Use %s instead' % (option, suggestion))
be5df5ee
S
503 return True
504 return False
505
506 if check_deprecated('cn_verification_proxy', '--cn-verification-proxy', '--geo-verification-proxy'):
38cce791
YCH
507 if self.params.get('geo_verification_proxy') is None:
508 self.params['geo_verification_proxy'] = self.params['cn_verification_proxy']
509
0d1bb027 510 check_deprecated('autonumber', '--auto-number', '-o "%(autonumber)s-%(title)s.%(ext)s"')
511 check_deprecated('usetitle', '--title', '-o "%(title)s-%(id)s.%(ext)s"')
53ed7066 512 check_deprecated('useid', '--id', '-o "%(id)s.%(ext)s"')
0d1bb027 513
514 for msg in self.params.get('warnings', []):
515 self.report_warning(msg)
516
6b591b29 517 if self.params.get('final_ext'):
518 if self.params.get('merge_output_format'):
519 self.report_warning('--merge-output-format will be ignored since --remux-video or --recode-video is given')
520 self.params['merge_output_format'] = self.params['final_ext']
521
b868936c 522 if self.params.get('overwrites') is None:
523 self.params.pop('overwrites', None)
524 elif self.params.get('nooverwrites') is not None:
525 # nooverwrites was unnecessarily changed to overwrites
526 # in 0c3d0f51778b153f65c21906031c2e091fcfb641
527 # This ensures compatibility with both keys
528 self.params['overwrites'] = not self.params['nooverwrites']
529 else:
530 self.params['nooverwrites'] = not self.params['overwrites']
b9d973be 531
0783b09b 532 if params.get('bidi_workaround', False):
1c088fa8
PH
533 try:
534 import pty
535 master, slave = pty.openpty()
003c69a8 536 width = compat_get_terminal_size().columns
1c088fa8
PH
537 if width is None:
538 width_args = []
539 else:
540 width_args = ['-w', str(width)]
5d681e96 541 sp_kwargs = dict(
1c088fa8
PH
542 stdin=subprocess.PIPE,
543 stdout=slave,
544 stderr=self._err_file)
5d681e96
PH
545 try:
546 self._output_process = subprocess.Popen(
547 ['bidiv'] + width_args, **sp_kwargs
548 )
549 except OSError:
5d681e96
PH
550 self._output_process = subprocess.Popen(
551 ['fribidi', '-c', 'UTF-8'] + width_args, **sp_kwargs)
552 self._output_channel = os.fdopen(master, 'rb')
1c088fa8 553 except OSError as ose:
66e7ace1 554 if ose.errno == errno.ENOENT:
6febd1c1 555 self.report_warning('Could not find fribidi executable, ignoring --bidi-workaround . Make sure that fribidi is an executable file in one of the directories in your $PATH.')
1c088fa8
PH
556 else:
557 raise
0783b09b 558
3089bc74
S
559 if (sys.platform != 'win32'
560 and sys.getfilesystemencoding() in ['ascii', 'ANSI_X3.4-1968']
561 and not params.get('restrictfilenames', False)):
e9137224 562 # Unicode filesystem API will throw errors (#1474, #13027)
34308b30 563 self.report_warning(
6febd1c1 564 'Assuming --restrict-filenames since file system encoding '
1b725173 565 'cannot encode all characters. '
6febd1c1 566 'Set the LC_ALL environment variable to fix this.')
4a98cdbf 567 self.params['restrictfilenames'] = True
34308b30 568
de6000d9 569 self.outtmpl_dict = self.parse_outtmpl()
486dd09e 570
187986a8 571 # Creating format selector here allows us to catch syntax errors before the extraction
572 self.format_selector = (
573 None if self.params.get('format') is None
574 else self.build_format_selector(self.params['format']))
575
dca08720
PH
576 self._setup_opener()
577
4cd0a709 578 """Preload the archive, if any is specified"""
579 def preload_download_archive(fn):
580 if fn is None:
581 return False
0760b0a7 582 self.write_debug('Loading archive file %r\n' % fn)
4cd0a709 583 try:
584 with locked_file(fn, 'r', encoding='utf-8') as archive_file:
585 for line in archive_file:
586 self.archive.add(line.strip())
587 except IOError as ioe:
588 if ioe.errno != errno.ENOENT:
589 raise
590 return False
591 return True
592
593 self.archive = set()
594 preload_download_archive(self.params.get('download_archive'))
595
3511266b
PH
596 if auto_init:
597 self.print_debug_header()
598 self.add_default_info_extractors()
599
4f026faf 600 for pp_def_raw in self.params.get('postprocessors', []):
4f026faf 601 pp_def = dict(pp_def_raw)
fd7cfb64 602 when = pp_def.pop('when', 'post_process')
603 pp_class = get_postprocessor(pp_def.pop('key'))
4f026faf 604 pp = pp_class(self, **compat_kwargs(pp_def))
5bfa4862 605 self.add_post_processor(pp, when=when)
4f026faf 606
ab8e5e51
AM
607 for ph in self.params.get('post_hooks', []):
608 self.add_post_hook(ph)
609
71b640cc
PH
610 for ph in self.params.get('progress_hooks', []):
611 self.add_progress_hook(ph)
612
51fb4995
YCH
613 register_socks_protocols()
614
7d4111ed
PH
615 def warn_if_short_id(self, argv):
616 # short YouTube ID starting with dash?
617 idxs = [
618 i for i, a in enumerate(argv)
619 if re.match(r'^-[0-9A-Za-z_-]{10}$', a)]
620 if idxs:
621 correct_argv = (
7a5c1cfe 622 ['yt-dlp']
3089bc74
S
623 + [a for i, a in enumerate(argv) if i not in idxs]
624 + ['--'] + [argv[i] for i in idxs]
7d4111ed
PH
625 )
626 self.report_warning(
627 'Long argument string detected. '
628 'Use -- to separate parameters and URLs, like this:\n%s\n' %
629 args_to_str(correct_argv))
630
8222d8de
JMF
631 def add_info_extractor(self, ie):
632 """Add an InfoExtractor object to the end of the list."""
633 self._ies.append(ie)
e52d7f85
JMF
634 if not isinstance(ie, type):
635 self._ies_instances[ie.ie_key()] = ie
636 ie.set_downloader(self)
8222d8de 637
56c73665
JMF
638 def get_info_extractor(self, ie_key):
639 """
640 Get an instance of an IE with name ie_key, it will try to get one from
641 the _ies list, if there's no instance it will create a new one and add
642 it to the extractor list.
643 """
644 ie = self._ies_instances.get(ie_key)
645 if ie is None:
646 ie = get_info_extractor(ie_key)()
647 self.add_info_extractor(ie)
648 return ie
649
023fa8c4
JMF
650 def add_default_info_extractors(self):
651 """
652 Add the InfoExtractors returned by gen_extractors to the end of the list
653 """
e52d7f85 654 for ie in gen_extractor_classes():
023fa8c4
JMF
655 self.add_info_extractor(ie)
656
56d868db 657 def add_post_processor(self, pp, when='post_process'):
8222d8de 658 """Add a PostProcessor object to the end of the chain."""
5bfa4862 659 self._pps[when].append(pp)
8222d8de
JMF
660 pp.set_downloader(self)
661
ab8e5e51
AM
662 def add_post_hook(self, ph):
663 """Add the post hook"""
664 self._post_hooks.append(ph)
665
933605d7
JMF
666 def add_progress_hook(self, ph):
667 """Add the progress hook (currently only for the file downloader)"""
668 self._progress_hooks.append(ph)
8ab470f1 669
1c088fa8 670 def _bidi_workaround(self, message):
5d681e96 671 if not hasattr(self, '_output_channel'):
1c088fa8
PH
672 return message
673
5d681e96 674 assert hasattr(self, '_output_process')
11b85ce6 675 assert isinstance(message, compat_str)
6febd1c1
PH
676 line_count = message.count('\n') + 1
677 self._output_process.stdin.write((message + '\n').encode('utf-8'))
5d681e96 678 self._output_process.stdin.flush()
6febd1c1 679 res = ''.join(self._output_channel.readline().decode('utf-8')
9e1a5b84 680 for _ in range(line_count))
6febd1c1 681 return res[:-len('\n')]
1c088fa8 682
b35496d8 683 def _write_string(self, message, out=None, only_once=False):
684 if only_once:
685 if message in self._printed_messages:
686 return
687 self._printed_messages.add(message)
688 write_string(message, out=out, encoding=self.params.get('encoding'))
734f90bb 689
848887eb 690 def to_stdout(self, message, skip_eol=False, quiet=False):
0760b0a7 691 """Print message to stdout"""
8bf9319e 692 if self.params.get('logger'):
43afe285 693 self.params['logger'].debug(message)
835a1478 694 elif not quiet or self.params.get('verbose'):
695 self._write_string(
696 '%s%s' % (self._bidi_workaround(message), ('' if skip_eol else '\n')),
697 self._err_file if quiet else self._screen_file)
8222d8de 698
b35496d8 699 def to_stderr(self, message, only_once=False):
0760b0a7 700 """Print message to stderr"""
11b85ce6 701 assert isinstance(message, compat_str)
8bf9319e 702 if self.params.get('logger'):
43afe285
IB
703 self.params['logger'].error(message)
704 else:
b35496d8 705 self._write_string('%s\n' % self._bidi_workaround(message), self._err_file, only_once=only_once)
8222d8de 706
1e5b9a95
PH
707 def to_console_title(self, message):
708 if not self.params.get('consoletitle', False):
709 return
4bede0d8
C
710 if compat_os_name == 'nt':
711 if ctypes.windll.kernel32.GetConsoleWindow():
712 # c_wchar_p() might not be necessary if `message` is
713 # already of type unicode()
714 ctypes.windll.kernel32.SetConsoleTitleW(ctypes.c_wchar_p(message))
1e5b9a95 715 elif 'TERM' in os.environ:
b46696bd 716 self._write_string('\033]0;%s\007' % message, self._screen_file)
1e5b9a95 717
bdde425c
PH
718 def save_console_title(self):
719 if not self.params.get('consoletitle', False):
720 return
b7b04c78 721 if self.params.get('simulate'):
94c3442e 722 return
4bede0d8 723 if compat_os_name != 'nt' and 'TERM' in os.environ:
efd6c574 724 # Save the title on stack
734f90bb 725 self._write_string('\033[22;0t', self._screen_file)
bdde425c
PH
726
727 def restore_console_title(self):
728 if not self.params.get('consoletitle', False):
729 return
b7b04c78 730 if self.params.get('simulate'):
94c3442e 731 return
4bede0d8 732 if compat_os_name != 'nt' and 'TERM' in os.environ:
efd6c574 733 # Restore the title from stack
734f90bb 734 self._write_string('\033[23;0t', self._screen_file)
bdde425c
PH
735
736 def __enter__(self):
737 self.save_console_title()
738 return self
739
740 def __exit__(self, *args):
741 self.restore_console_title()
f89197d7 742
dca08720 743 if self.params.get('cookiefile') is not None:
1bab3437 744 self.cookiejar.save(ignore_discard=True, ignore_expires=True)
bdde425c 745
8222d8de
JMF
746 def trouble(self, message=None, tb=None):
747 """Determine action to take when a download problem appears.
748
749 Depending on if the downloader has been configured to ignore
750 download errors or not, this method may throw an exception or
751 not when errors are found, after printing the message.
752
753 tb, if given, is additional traceback information.
754 """
755 if message is not None:
756 self.to_stderr(message)
757 if self.params.get('verbose'):
758 if tb is None:
759 if sys.exc_info()[0]: # if .trouble has been called from an except block
6febd1c1 760 tb = ''
8222d8de 761 if hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
6febd1c1 762 tb += ''.join(traceback.format_exception(*sys.exc_info()[1].exc_info))
c0384f22 763 tb += encode_compat_str(traceback.format_exc())
8222d8de
JMF
764 else:
765 tb_data = traceback.format_list(traceback.extract_stack())
6febd1c1 766 tb = ''.join(tb_data)
c19bc311 767 if tb:
768 self.to_stderr(tb)
8222d8de
JMF
769 if not self.params.get('ignoreerrors', False):
770 if sys.exc_info()[0] and hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
771 exc_info = sys.exc_info()[1].exc_info
772 else:
773 exc_info = sys.exc_info()
774 raise DownloadError(message, exc_info)
775 self._download_retcode = 1
776
0760b0a7 777 def to_screen(self, message, skip_eol=False):
778 """Print message to stdout if not in quiet mode"""
779 self.to_stdout(
780 message, skip_eol, quiet=self.params.get('quiet', False))
781
c84aeac6 782 def report_warning(self, message, only_once=False):
8222d8de
JMF
783 '''
784 Print the message to stderr, it will be prefixed with 'WARNING:'
785 If stderr is a tty file the 'WARNING:' will be colored
786 '''
6d07ce01
JMF
787 if self.params.get('logger') is not None:
788 self.params['logger'].warning(message)
8222d8de 789 else:
ad8915b7
PH
790 if self.params.get('no_warnings'):
791 return
e9c0cdd3 792 if not self.params.get('no_color') and self._err_file.isatty() and compat_os_name != 'nt':
6d07ce01
JMF
793 _msg_header = '\033[0;33mWARNING:\033[0m'
794 else:
795 _msg_header = 'WARNING:'
796 warning_message = '%s %s' % (_msg_header, message)
b35496d8 797 self.to_stderr(warning_message, only_once)
8222d8de
JMF
798
799 def report_error(self, message, tb=None):
800 '''
801 Do the same as trouble, but prefixes the message with 'ERROR:', colored
802 in red if stderr is a tty file.
803 '''
e9c0cdd3 804 if not self.params.get('no_color') and self._err_file.isatty() and compat_os_name != 'nt':
6febd1c1 805 _msg_header = '\033[0;31mERROR:\033[0m'
8222d8de 806 else:
6febd1c1
PH
807 _msg_header = 'ERROR:'
808 error_message = '%s %s' % (_msg_header, message)
8222d8de
JMF
809 self.trouble(error_message, tb)
810
b35496d8 811 def write_debug(self, message, only_once=False):
0760b0a7 812 '''Log debug message or Print message to stderr'''
813 if not self.params.get('verbose', False):
814 return
815 message = '[debug] %s' % message
816 if self.params.get('logger'):
817 self.params['logger'].debug(message)
818 else:
b35496d8 819 self.to_stderr(message, only_once)
0760b0a7 820
8222d8de
JMF
821 def report_file_already_downloaded(self, file_name):
822 """Report file has already been fully downloaded."""
823 try:
6febd1c1 824 self.to_screen('[download] %s has already been downloaded' % file_name)
ce02ed60 825 except UnicodeEncodeError:
6febd1c1 826 self.to_screen('[download] The file has already been downloaded')
8222d8de 827
0c3d0f51 828 def report_file_delete(self, file_name):
829 """Report that existing file will be deleted."""
830 try:
c25228e5 831 self.to_screen('Deleting existing file %s' % file_name)
0c3d0f51 832 except UnicodeEncodeError:
c25228e5 833 self.to_screen('Deleting existing file')
0c3d0f51 834
de6000d9 835 def parse_outtmpl(self):
836 outtmpl_dict = self.params.get('outtmpl', {})
837 if not isinstance(outtmpl_dict, dict):
838 outtmpl_dict = {'default': outtmpl_dict}
839 outtmpl_dict.update({
840 k: v for k, v in DEFAULT_OUTTMPL.items()
841 if not outtmpl_dict.get(k)})
842 for key, val in outtmpl_dict.items():
843 if isinstance(val, bytes):
844 self.report_warning(
845 'Parameter outtmpl is bytes, but should be a unicode string. '
846 'Put from __future__ import unicode_literals at the top of your code file or consider switching to Python 3.x.')
847 return outtmpl_dict
848
21cd8fae 849 def get_output_path(self, dir_type='', filename=None):
850 paths = self.params.get('paths', {})
851 assert isinstance(paths, dict)
852 path = os.path.join(
853 expand_path(paths.get('home', '').strip()),
854 expand_path(paths.get(dir_type, '').strip()) if dir_type else '',
855 filename or '')
856
857 # Temporary fix for #4787
858 # 'Treat' all problem characters by passing filename through preferredencoding
859 # to workaround encoding issues with subprocess on python2 @ Windows
860 if sys.version_info < (3, 0) and sys.platform == 'win32':
861 path = encodeFilename(path, True).decode(preferredencoding())
862 return sanitize_path(path, force=self.params.get('windowsfilenames'))
863
76a264ac 864 @staticmethod
901130bb 865 def _outtmpl_expandpath(outtmpl):
866 # expand_path translates '%%' into '%' and '$$' into '$'
867 # correspondingly that is not what we want since we need to keep
868 # '%%' intact for template dict substitution step. Working around
869 # with boundary-alike separator hack.
870 sep = ''.join([random.choice(ascii_letters) for _ in range(32)])
871 outtmpl = outtmpl.replace('%%', '%{0}%'.format(sep)).replace('$$', '${0}$'.format(sep))
872
873 # outtmpl should be expand_path'ed before template dict substitution
874 # because meta fields may contain env variables we don't want to
875 # be expanded. For example, for outtmpl "%(title)s.%(ext)s" and
876 # title "Hello $PATH", we don't want `$PATH` to be expanded.
877 return expand_path(outtmpl).replace(sep, '')
878
879 @staticmethod
880 def escape_outtmpl(outtmpl):
881 ''' Escape any remaining strings like %s, %abc% etc. '''
882 return re.sub(
883 STR_FORMAT_RE_TMPL.format('', '(?![%(\0])'),
884 lambda mobj: ('' if mobj.group('has_key') else '%') + mobj.group(0),
885 outtmpl)
886
887 @classmethod
888 def validate_outtmpl(cls, outtmpl):
76a264ac 889 ''' @return None or Exception object '''
7d1eb38a 890 outtmpl = re.sub(
891 STR_FORMAT_RE_TMPL.format('[^)]*', '[ljq]'),
892 lambda mobj: f'{mobj.group(0)[:-1]}s',
893 cls._outtmpl_expandpath(outtmpl))
76a264ac 894 try:
7d1eb38a 895 cls.escape_outtmpl(outtmpl) % collections.defaultdict(int)
76a264ac 896 return None
897 except ValueError as err:
898 return err
899
143db31d 900 def prepare_outtmpl(self, outtmpl, info_dict, sanitize=None):
901130bb 901 """ Make the template and info_dict suitable for substitution : ydl.outtmpl_escape(outtmpl) % info_dict """
6e84b215 902 info_dict.setdefault('epoch', int(time.time())) # keep epoch consistent once set
143db31d 903
6e84b215 904 info_dict = dict(info_dict) # Do not sanitize so as not to consume LazyList
905 for key in ('__original_infodict', '__postprocessors'):
906 info_dict.pop(key, None)
752cda38 907 info_dict['duration_string'] = ( # %(duration>%H-%M-%S)s is wrong if duration > 24hrs
53c18592 908 formatSeconds(info_dict['duration'], '-' if sanitize else ':')
143db31d 909 if info_dict.get('duration', None) is not None
910 else None)
752cda38 911 info_dict['autonumber'] = self.params.get('autonumber_start', 1) - 1 + self._num_downloads
912 if info_dict.get('resolution') is None:
913 info_dict['resolution'] = self.format_resolution(info_dict, default=None)
143db31d 914
143db31d 915 # For fields playlist_index and autonumber convert all occurrences
916 # of %(field)s to %(field)0Nd for backward compatibility
917 field_size_compat_map = {
752cda38 918 'playlist_index': len(str(info_dict.get('_last_playlist_index') or '')),
919 'autonumber': self.params.get('autonumber_size') or 5,
143db31d 920 }
752cda38 921
385a27fa 922 TMPL_DICT = {}
7d1eb38a 923 EXTERNAL_FORMAT_RE = re.compile(STR_FORMAT_RE_TMPL.format('[^)]*', f'[{STR_FORMAT_TYPES}ljq]'))
385a27fa 924 MATH_FUNCTIONS = {
925 '+': float.__add__,
926 '-': float.__sub__,
927 }
e625be0d 928 # Field is of the form key1.key2...
929 # where keys (except first) can be string, int or slice
2b8a2973 930 FIELD_RE = r'\w*(?:\.(?:\w+|{num}|{num}?(?::{num}?){{1,2}}))*'.format(num=r'(?:-?\d+)')
385a27fa 931 MATH_FIELD_RE = r'''{field}|{num}'''.format(field=FIELD_RE, num=r'-?\d+(?:.\d+)?')
932 MATH_OPERATORS_RE = r'(?:%s)' % '|'.join(map(re.escape, MATH_FUNCTIONS.keys()))
e625be0d 933 INTERNAL_FORMAT_RE = re.compile(r'''(?x)
934 (?P<negate>-)?
385a27fa 935 (?P<fields>{field})
936 (?P<maths>(?:{math_op}{math_field})*)
e625be0d 937 (?:>(?P<strf_format>.+?))?
938 (?:\|(?P<default>.*?))?
385a27fa 939 $'''.format(field=FIELD_RE, math_op=MATH_OPERATORS_RE, math_field=MATH_FIELD_RE))
752cda38 940
2b8a2973 941 def _traverse_infodict(k):
942 k = k.split('.')
943 if k[0] == '':
944 k.pop(0)
945 return traverse_obj(info_dict, k, is_user_input=True, traverse_string=True)
76a264ac 946
752cda38 947 def get_value(mdict):
948 # Object traversal
2b8a2973 949 value = _traverse_infodict(mdict['fields'])
752cda38 950 # Negative
951 if mdict['negate']:
952 value = float_or_none(value)
953 if value is not None:
954 value *= -1
955 # Do maths
385a27fa 956 offset_key = mdict['maths']
957 if offset_key:
752cda38 958 value = float_or_none(value)
959 operator = None
385a27fa 960 while offset_key:
961 item = re.match(
962 MATH_FIELD_RE if operator else MATH_OPERATORS_RE,
963 offset_key).group(0)
964 offset_key = offset_key[len(item):]
965 if operator is None:
752cda38 966 operator = MATH_FUNCTIONS[item]
385a27fa 967 continue
968 item, multiplier = (item[1:], -1) if item[0] == '-' else (item, 1)
969 offset = float_or_none(item)
970 if offset is None:
2b8a2973 971 offset = float_or_none(_traverse_infodict(item))
385a27fa 972 try:
973 value = operator(value, multiplier * offset)
974 except (TypeError, ZeroDivisionError):
975 return None
976 operator = None
752cda38 977 # Datetime formatting
978 if mdict['strf_format']:
979 value = strftime_or_none(value, mdict['strf_format'])
980
981 return value
982
b868936c 983 na = self.params.get('outtmpl_na_placeholder', 'NA')
984
6e84b215 985 def _dumpjson_default(obj):
986 if isinstance(obj, (set, LazyList)):
987 return list(obj)
988 raise TypeError(f'Object of type {type(obj).__name__} is not JSON serializable')
989
752cda38 990 def create_key(outer_mobj):
991 if not outer_mobj.group('has_key'):
901130bb 992 return f'%{outer_mobj.group(0)}'
752cda38 993 key = outer_mobj.group('key')
752cda38 994 mobj = re.match(INTERNAL_FORMAT_RE, key)
995 if mobj is None:
9fea350f 996 value, default, mobj = None, na, {'fields': ''}
752cda38 997 else:
e625be0d 998 mobj = mobj.groupdict()
752cda38 999 default = mobj['default'] if mobj['default'] is not None else na
1000 value = get_value(mobj)
1001
b868936c 1002 fmt = outer_mobj.group('format')
752cda38 1003 if fmt == 's' and value is not None and key in field_size_compat_map.keys():
1004 fmt = '0{:d}d'.format(field_size_compat_map[key])
1005
1006 value = default if value is None else value
752cda38 1007
7d1eb38a 1008 str_fmt = f'{fmt[:-1]}s'
1009 if fmt[-1] == 'l':
1010 value, fmt = ', '.join(variadic(value)), str_fmt
1011 elif fmt[-1] == 'j':
6e84b215 1012 value, fmt = json.dumps(value, default=_dumpjson_default), str_fmt
7d1eb38a 1013 elif fmt[-1] == 'q':
1014 value, fmt = compat_shlex_quote(str(value)), str_fmt
1015 elif fmt[-1] == 'c':
1016 value = str(value)
76a264ac 1017 if value is None:
1018 value, fmt = default, 's'
1019 else:
1020 value = value[0]
1021 elif fmt[-1] not in 'rs': # numeric
a439a3a4 1022 value = float_or_none(value)
752cda38 1023 if value is None:
1024 value, fmt = default, 's'
901130bb 1025
752cda38 1026 if sanitize:
1027 if fmt[-1] == 'r':
1028 # If value is an object, sanitize might convert it to a string
1029 # So we convert it to repr first
7d1eb38a 1030 value, fmt = repr(value), str_fmt
639f1cea 1031 if fmt[-1] in 'csr':
9fea350f 1032 value = sanitize(mobj['fields'].split('.')[-1], value)
901130bb 1033
b868936c 1034 key = '%s\0%s' % (key.replace('%', '%\0'), outer_mobj.group('format'))
385a27fa 1035 TMPL_DICT[key] = value
b868936c 1036 return '{prefix}%({key}){fmt}'.format(key=key, fmt=fmt, prefix=outer_mobj.group('prefix'))
752cda38 1037
385a27fa 1038 return EXTERNAL_FORMAT_RE.sub(create_key, outtmpl), TMPL_DICT
143db31d 1039
de6000d9 1040 def _prepare_filename(self, info_dict, tmpl_type='default'):
8222d8de 1041 try:
586a91b6 1042 sanitize = lambda k, v: sanitize_filename(
45598aab 1043 compat_str(v),
1bb5c511 1044 restricted=self.params.get('restrictfilenames'),
40df485f 1045 is_id=(k == 'id' or k.endswith('_id')))
de6000d9 1046 outtmpl = self.outtmpl_dict.get(tmpl_type, self.outtmpl_dict['default'])
143db31d 1047 outtmpl, template_dict = self.prepare_outtmpl(outtmpl, info_dict, sanitize)
901130bb 1048 outtmpl = self.escape_outtmpl(self._outtmpl_expandpath(outtmpl))
1049 filename = outtmpl % template_dict
15da37c7 1050
143db31d 1051 force_ext = OUTTMPL_TYPES.get(tmpl_type)
de6000d9 1052 if force_ext is not None:
752cda38 1053 filename = replace_extension(filename, force_ext, info_dict.get('ext'))
de6000d9 1054
bdc3fd2f
U
1055 # https://github.com/blackjack4494/youtube-dlc/issues/85
1056 trim_file_name = self.params.get('trim_file_name', False)
1057 if trim_file_name:
1058 fn_groups = filename.rsplit('.')
1059 ext = fn_groups[-1]
1060 sub_ext = ''
1061 if len(fn_groups) > 2:
1062 sub_ext = fn_groups[-2]
1063 filename = '.'.join(filter(None, [fn_groups[0][:trim_file_name], sub_ext, ext]))
1064
0202b52a 1065 return filename
8222d8de 1066 except ValueError as err:
6febd1c1 1067 self.report_error('Error in output template: ' + str(err) + ' (encoding: ' + repr(preferredencoding()) + ')')
8222d8de
JMF
1068 return None
1069
de6000d9 1070 def prepare_filename(self, info_dict, dir_type='', warn=False):
1071 """Generate the output filename."""
21cd8fae 1072
de6000d9 1073 filename = self._prepare_filename(info_dict, dir_type or 'default')
1074
c84aeac6 1075 if warn:
21cd8fae 1076 if not self.params.get('paths'):
de6000d9 1077 pass
1078 elif filename == '-':
c84aeac6 1079 self.report_warning('--paths is ignored when an outputting to stdout', only_once=True)
de6000d9 1080 elif os.path.isabs(filename):
c84aeac6 1081 self.report_warning('--paths is ignored since an absolute path is given in output template', only_once=True)
de6000d9 1082 if filename == '-' or not filename:
1083 return filename
1084
21cd8fae 1085 return self.get_output_path(dir_type, filename)
0202b52a 1086
120fe513 1087 def _match_entry(self, info_dict, incomplete=False, silent=False):
ecdec191 1088 """ Returns None if the file should be downloaded """
8222d8de 1089
c77495e3 1090 video_title = info_dict.get('title', info_dict.get('id', 'video'))
1091
8b0d7497 1092 def check_filter():
8b0d7497 1093 if 'title' in info_dict:
1094 # This can happen when we're just evaluating the playlist
1095 title = info_dict['title']
1096 matchtitle = self.params.get('matchtitle', False)
1097 if matchtitle:
1098 if not re.search(matchtitle, title, re.IGNORECASE):
1099 return '"' + title + '" title did not match pattern "' + matchtitle + '"'
1100 rejecttitle = self.params.get('rejecttitle', False)
1101 if rejecttitle:
1102 if re.search(rejecttitle, title, re.IGNORECASE):
1103 return '"' + title + '" title matched reject pattern "' + rejecttitle + '"'
1104 date = info_dict.get('upload_date')
1105 if date is not None:
1106 dateRange = self.params.get('daterange', DateRange())
1107 if date not in dateRange:
1108 return '%s upload date is not in range %s' % (date_from_str(date).isoformat(), dateRange)
1109 view_count = info_dict.get('view_count')
1110 if view_count is not None:
1111 min_views = self.params.get('min_views')
1112 if min_views is not None and view_count < min_views:
1113 return 'Skipping %s, because it has not reached minimum view count (%d/%d)' % (video_title, view_count, min_views)
1114 max_views = self.params.get('max_views')
1115 if max_views is not None and view_count > max_views:
1116 return 'Skipping %s, because it has exceeded the maximum view count (%d/%d)' % (video_title, view_count, max_views)
1117 if age_restricted(info_dict.get('age_limit'), self.params.get('age_limit')):
1118 return 'Skipping "%s" because it is age restricted' % video_title
8b0d7497 1119
1120 if not incomplete:
1121 match_filter = self.params.get('match_filter')
1122 if match_filter is not None:
1123 ret = match_filter(info_dict)
1124 if ret is not None:
1125 return ret
1126 return None
1127
c77495e3 1128 if self.in_download_archive(info_dict):
1129 reason = '%s has already been recorded in the archive' % video_title
1130 break_opt, break_err = 'break_on_existing', ExistingVideoReached
1131 else:
1132 reason = check_filter()
1133 break_opt, break_err = 'break_on_reject', RejectedVideoReached
8b0d7497 1134 if reason is not None:
120fe513 1135 if not silent:
1136 self.to_screen('[download] ' + reason)
c77495e3 1137 if self.params.get(break_opt, False):
1138 raise break_err()
8b0d7497 1139 return reason
fe7e0c98 1140
b6c45014
JMF
1141 @staticmethod
1142 def add_extra_info(info_dict, extra_info):
1143 '''Set the keys from extra_info in info dict if they are missing'''
1144 for key, value in extra_info.items():
1145 info_dict.setdefault(key, value)
1146
58f197b7 1147 def extract_info(self, url, download=True, ie_key=None, extra_info={},
61aa5ba3 1148 process=True, force_generic_extractor=False):
41d1cca3 1149 """
1150 Return a list with a dictionary for each video extracted.
1151
1152 Arguments:
1153 url -- URL to extract
1154
1155 Keyword arguments:
1156 download -- whether to download videos during extraction
1157 ie_key -- extractor key hint
1158 extra_info -- dictionary containing the extra values to add to each result
1159 process -- whether to resolve all unresolved references (URLs, playlist items),
1160 must be True for download to work.
1161 force_generic_extractor -- force using the generic extractor
1162 """
fe7e0c98 1163
61aa5ba3 1164 if not ie_key and force_generic_extractor:
d22dec74
S
1165 ie_key = 'Generic'
1166
8222d8de 1167 if ie_key:
56c73665 1168 ies = [self.get_info_extractor(ie_key)]
8222d8de
JMF
1169 else:
1170 ies = self._ies
1171
1172 for ie in ies:
1173 if not ie.suitable(url):
1174 continue
1175
9a68de12 1176 ie_key = ie.ie_key()
1177 ie = self.get_info_extractor(ie_key)
8222d8de 1178 if not ie.working():
6febd1c1
PH
1179 self.report_warning('The program functionality for this site has been marked as broken, '
1180 'and will probably not work.')
8222d8de
JMF
1181
1182 try:
d0757229 1183 temp_id = str_or_none(
63be1aab 1184 ie.extract_id(url) if callable(getattr(ie, 'extract_id', None))
1185 else ie._match_id(url))
a0566bbf 1186 except (AssertionError, IndexError, AttributeError):
1187 temp_id = None
1188 if temp_id is not None and self.in_download_archive({'id': temp_id, 'ie_key': ie_key}):
1189 self.to_screen("[%s] %s: has already been recorded in archive" % (
1190 ie_key, temp_id))
1191 break
58f197b7 1192 return self.__extract_info(url, ie, download, extra_info, process)
a0566bbf 1193 else:
1194 self.report_error('no suitable InfoExtractor for URL %s' % url)
1195
cc9d1493 1196 def __handle_extraction_exceptions(func, handle_all_errors=True):
a0566bbf 1197 def wrapper(self, *args, **kwargs):
1198 try:
1199 return func(self, *args, **kwargs)
773f291d
S
1200 except GeoRestrictedError as e:
1201 msg = e.msg
1202 if e.countries:
1203 msg += '\nThis video is available in %s.' % ', '.join(
1204 map(ISO3166Utils.short2full, e.countries))
1205 msg += '\nYou might want to use a VPN or a proxy server (with --proxy) to workaround.'
1206 self.report_error(msg)
fb043a6e 1207 except ExtractorError as e: # An error we somewhat expected
2c74e6fa 1208 self.report_error(compat_str(e), e.format_traceback())
51d9739f 1209 except ThrottledDownload:
1210 self.to_stderr('\r')
1211 self.report_warning('The download speed is below throttle limit. Re-extracting data')
1212 return wrapper(self, *args, **kwargs)
8b0d7497 1213 except (MaxDownloadsReached, ExistingVideoReached, RejectedVideoReached):
d3e5bbf4 1214 raise
8222d8de 1215 except Exception as e:
cc9d1493 1216 if handle_all_errors and self.params.get('ignoreerrors', False):
9b9c5355 1217 self.report_error(error_to_compat_str(e), tb=encode_compat_str(traceback.format_exc()))
8222d8de
JMF
1218 else:
1219 raise
a0566bbf 1220 return wrapper
1221
1222 @__handle_extraction_exceptions
58f197b7 1223 def __extract_info(self, url, ie, download, extra_info, process):
a0566bbf 1224 ie_result = ie.extract(url)
1225 if ie_result is None: # Finished already (backwards compatibility; listformats and friends should be moved here)
1226 return
1227 if isinstance(ie_result, list):
1228 # Backwards compatibility: old IE result format
1229 ie_result = {
1230 '_type': 'compat_list',
1231 'entries': ie_result,
1232 }
e37d0efb 1233 if extra_info.get('original_url'):
1234 ie_result.setdefault('original_url', extra_info['original_url'])
a0566bbf 1235 self.add_default_extra_info(ie_result, ie, url)
1236 if process:
1237 return self.process_ie_result(ie_result, download, extra_info)
8222d8de 1238 else:
a0566bbf 1239 return ie_result
fe7e0c98 1240
ea38e55f 1241 def add_default_extra_info(self, ie_result, ie, url):
6033d980 1242 if url is not None:
1243 self.add_extra_info(ie_result, {
1244 'webpage_url': url,
1245 'original_url': url,
1246 'webpage_url_basename': url_basename(url),
1247 })
1248 if ie is not None:
1249 self.add_extra_info(ie_result, {
1250 'extractor': ie.IE_NAME,
1251 'extractor_key': ie.ie_key(),
1252 })
ea38e55f 1253
8222d8de
JMF
1254 def process_ie_result(self, ie_result, download=True, extra_info={}):
1255 """
1256 Take the result of the ie(may be modified) and resolve all unresolved
1257 references (URLs, playlist items).
1258
1259 It will also download the videos if 'download'.
1260 Returns the resolved ie_result.
1261 """
e8ee972c
PH
1262 result_type = ie_result.get('_type', 'video')
1263
057a5206 1264 if result_type in ('url', 'url_transparent'):
134c6ea8 1265 ie_result['url'] = sanitize_url(ie_result['url'])
e37d0efb 1266 if ie_result.get('original_url'):
1267 extra_info.setdefault('original_url', ie_result['original_url'])
1268
057a5206 1269 extract_flat = self.params.get('extract_flat', False)
3089bc74
S
1270 if ((extract_flat == 'in_playlist' and 'playlist' in extra_info)
1271 or extract_flat is True):
ecb54191 1272 info_copy = ie_result.copy()
1273 self.add_extra_info(info_copy, extra_info)
6033d980 1274 ie = try_get(ie_result.get('ie_key'), self.get_info_extractor)
1275 self.add_default_extra_info(info_copy, ie, ie_result['url'])
ecb54191 1276 self.__forced_printings(info_copy, self.prepare_filename(info_copy), incomplete=True)
e8ee972c
PH
1277 return ie_result
1278
8222d8de 1279 if result_type == 'video':
b6c45014 1280 self.add_extra_info(ie_result, extra_info)
9c2b75b5 1281 ie_result = self.process_video_result(ie_result, download=download)
28b0eb0f 1282 additional_urls = (ie_result or {}).get('additional_urls')
9c2b75b5 1283 if additional_urls:
e9f4ccd1 1284 # TODO: Improve MetadataParserPP to allow setting a list
9c2b75b5 1285 if isinstance(additional_urls, compat_str):
1286 additional_urls = [additional_urls]
1287 self.to_screen(
1288 '[info] %s: %d additional URL(s) requested' % (ie_result['id'], len(additional_urls)))
1289 self.write_debug('Additional URLs: "%s"' % '", "'.join(additional_urls))
1290 ie_result['additional_entries'] = [
1291 self.extract_info(
1292 url, download, extra_info,
1293 force_generic_extractor=self.params.get('force_generic_extractor'))
1294 for url in additional_urls
1295 ]
1296 return ie_result
8222d8de
JMF
1297 elif result_type == 'url':
1298 # We have to add extra_info to the results because it may be
1299 # contained in a playlist
07cce701 1300 return self.extract_info(
1301 ie_result['url'], download,
1302 ie_key=ie_result.get('ie_key'),
1303 extra_info=extra_info)
7fc3fa05
PH
1304 elif result_type == 'url_transparent':
1305 # Use the information from the embedding page
1306 info = self.extract_info(
1307 ie_result['url'], ie_key=ie_result.get('ie_key'),
1308 extra_info=extra_info, download=False, process=False)
1309
1640eb09
S
1310 # extract_info may return None when ignoreerrors is enabled and
1311 # extraction failed with an error, don't crash and return early
1312 # in this case
1313 if not info:
1314 return info
1315
412c617d
PH
1316 force_properties = dict(
1317 (k, v) for k, v in ie_result.items() if v is not None)
0396806f 1318 for f in ('_type', 'url', 'id', 'extractor', 'extractor_key', 'ie_key'):
412c617d
PH
1319 if f in force_properties:
1320 del force_properties[f]
1321 new_result = info.copy()
1322 new_result.update(force_properties)
7fc3fa05 1323
0563f7ac
S
1324 # Extracted info may not be a video result (i.e.
1325 # info.get('_type', 'video') != video) but rather an url or
1326 # url_transparent. In such cases outer metadata (from ie_result)
1327 # should be propagated to inner one (info). For this to happen
1328 # _type of info should be overridden with url_transparent. This
067aa17e 1329 # fixes issue from https://github.com/ytdl-org/youtube-dl/pull/11163.
0563f7ac
S
1330 if new_result.get('_type') == 'url':
1331 new_result['_type'] = 'url_transparent'
7fc3fa05
PH
1332
1333 return self.process_ie_result(
1334 new_result, download=download, extra_info=extra_info)
40fcba5e 1335 elif result_type in ('playlist', 'multi_video'):
30a074c2 1336 # Protect from infinite recursion due to recursively nested playlists
1337 # (see https://github.com/ytdl-org/youtube-dl/issues/27833)
1338 webpage_url = ie_result['webpage_url']
1339 if webpage_url in self._playlist_urls:
7e85e872 1340 self.to_screen(
30a074c2 1341 '[download] Skipping already downloaded playlist: %s'
1342 % ie_result.get('title') or ie_result.get('id'))
1343 return
7e85e872 1344
30a074c2 1345 self._playlist_level += 1
1346 self._playlist_urls.add(webpage_url)
bc516a3f 1347 self._sanitize_thumbnails(ie_result)
30a074c2 1348 try:
1349 return self.__process_playlist(ie_result, download)
1350 finally:
1351 self._playlist_level -= 1
1352 if not self._playlist_level:
1353 self._playlist_urls.clear()
8222d8de 1354 elif result_type == 'compat_list':
c9bf4114
PH
1355 self.report_warning(
1356 'Extractor %s returned a compat_list result. '
1357 'It needs to be updated.' % ie_result.get('extractor'))
5f6a1245 1358
8222d8de 1359 def _fixup(r):
b868936c 1360 self.add_extra_info(r, {
1361 'extractor': ie_result['extractor'],
1362 'webpage_url': ie_result['webpage_url'],
1363 'webpage_url_basename': url_basename(ie_result['webpage_url']),
1364 'extractor_key': ie_result['extractor_key'],
1365 })
8222d8de
JMF
1366 return r
1367 ie_result['entries'] = [
b6c45014 1368 self.process_ie_result(_fixup(r), download, extra_info)
8222d8de
JMF
1369 for r in ie_result['entries']
1370 ]
1371 return ie_result
1372 else:
1373 raise Exception('Invalid result type: %s' % result_type)
1374
e92caff5 1375 def _ensure_dir_exists(self, path):
1376 return make_dir(path, self.report_error)
1377
30a074c2 1378 def __process_playlist(self, ie_result, download):
1379 # We process each entry in the playlist
1380 playlist = ie_result.get('title') or ie_result.get('id')
1381 self.to_screen('[download] Downloading playlist: %s' % playlist)
1382
498f5606 1383 if 'entries' not in ie_result:
1384 raise EntryNotInPlaylist()
1385 incomplete_entries = bool(ie_result.get('requested_entries'))
1386 if incomplete_entries:
1387 def fill_missing_entries(entries, indexes):
1388 ret = [None] * max(*indexes)
1389 for i, entry in zip(indexes, entries):
1390 ret[i - 1] = entry
1391 return ret
1392 ie_result['entries'] = fill_missing_entries(ie_result['entries'], ie_result['requested_entries'])
02fd60d3 1393
30a074c2 1394 playlist_results = []
1395
56a8fb4f 1396 playliststart = self.params.get('playliststart', 1)
30a074c2 1397 playlistend = self.params.get('playlistend')
1398 # For backwards compatibility, interpret -1 as whole list
1399 if playlistend == -1:
1400 playlistend = None
1401
1402 playlistitems_str = self.params.get('playlist_items')
1403 playlistitems = None
1404 if playlistitems_str is not None:
1405 def iter_playlistitems(format):
1406 for string_segment in format.split(','):
1407 if '-' in string_segment:
1408 start, end = string_segment.split('-')
1409 for item in range(int(start), int(end) + 1):
1410 yield int(item)
1411 else:
1412 yield int(string_segment)
1413 playlistitems = orderedSet(iter_playlistitems(playlistitems_str))
1414
1415 ie_entries = ie_result['entries']
56a8fb4f 1416 msg = (
1417 'Downloading %d videos' if not isinstance(ie_entries, list)
1418 else 'Collected %d videos; downloading %%d of them' % len(ie_entries))
1419 if not isinstance(ie_entries, (list, PagedList)):
1420 ie_entries = LazyList(ie_entries)
1421
50fed816 1422 def get_entry(i):
1423 return YoutubeDL.__handle_extraction_exceptions(
cc9d1493 1424 lambda self, i: ie_entries[i - 1],
1425 False
50fed816 1426 )(self, i)
1427
56a8fb4f 1428 entries = []
1429 for i in playlistitems or itertools.count(playliststart):
1430 if playlistitems is None and playlistend is not None and playlistend < i:
1431 break
1432 entry = None
1433 try:
50fed816 1434 entry = get_entry(i)
56a8fb4f 1435 if entry is None:
498f5606 1436 raise EntryNotInPlaylist()
56a8fb4f 1437 except (IndexError, EntryNotInPlaylist):
1438 if incomplete_entries:
1439 raise EntryNotInPlaylist()
1440 elif not playlistitems:
1441 break
1442 entries.append(entry)
120fe513 1443 try:
1444 if entry is not None:
1445 self._match_entry(entry, incomplete=True, silent=True)
1446 except (ExistingVideoReached, RejectedVideoReached):
1447 break
56a8fb4f 1448 ie_result['entries'] = entries
30a074c2 1449
56a8fb4f 1450 # Save playlist_index before re-ordering
1451 entries = [
1452 ((playlistitems[i - 1] if playlistitems else i), entry)
1453 for i, entry in enumerate(entries, 1)
1454 if entry is not None]
1455 n_entries = len(entries)
498f5606 1456
498f5606 1457 if not playlistitems and (playliststart or playlistend):
56a8fb4f 1458 playlistitems = list(range(playliststart, playliststart + n_entries))
498f5606 1459 ie_result['requested_entries'] = playlistitems
1460
1461 if self.params.get('allow_playlist_files', True):
1462 ie_copy = {
1463 'playlist': playlist,
1464 'playlist_id': ie_result.get('id'),
1465 'playlist_title': ie_result.get('title'),
1466 'playlist_uploader': ie_result.get('uploader'),
1467 'playlist_uploader_id': ie_result.get('uploader_id'),
71729754 1468 'playlist_index': 0,
498f5606 1469 }
1470 ie_copy.update(dict(ie_result))
1471
1472 if self.params.get('writeinfojson', False):
1473 infofn = self.prepare_filename(ie_copy, 'pl_infojson')
1474 if not self._ensure_dir_exists(encodeFilename(infofn)):
1475 return
1476 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(infofn)):
1477 self.to_screen('[info] Playlist metadata is already present')
1478 else:
1479 self.to_screen('[info] Writing playlist metadata as JSON to: ' + infofn)
1480 try:
8012d892 1481 write_json_file(self.sanitize_info(ie_result, self.params.get('clean_infojson', True)), infofn)
498f5606 1482 except (OSError, IOError):
1483 self.report_error('Cannot write playlist metadata to JSON file ' + infofn)
1484
681de68e 1485 # TODO: This should be passed to ThumbnailsConvertor if necessary
1486 self._write_thumbnails(ie_copy, self.prepare_filename(ie_copy, 'pl_thumbnail'))
1487
498f5606 1488 if self.params.get('writedescription', False):
1489 descfn = self.prepare_filename(ie_copy, 'pl_description')
1490 if not self._ensure_dir_exists(encodeFilename(descfn)):
1491 return
1492 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(descfn)):
1493 self.to_screen('[info] Playlist description is already present')
1494 elif ie_result.get('description') is None:
1495 self.report_warning('There\'s no playlist description to write.')
1496 else:
1497 try:
1498 self.to_screen('[info] Writing playlist description to: ' + descfn)
1499 with io.open(encodeFilename(descfn), 'w', encoding='utf-8') as descfile:
1500 descfile.write(ie_result['description'])
1501 except (OSError, IOError):
1502 self.report_error('Cannot write playlist description file ' + descfn)
1503 return
30a074c2 1504
1505 if self.params.get('playlistreverse', False):
1506 entries = entries[::-1]
30a074c2 1507 if self.params.get('playlistrandom', False):
1508 random.shuffle(entries)
1509
1510 x_forwarded_for = ie_result.get('__x_forwarded_for_ip')
1511
56a8fb4f 1512 self.to_screen('[%s] playlist %s: %s' % (ie_result['extractor'], playlist, msg % n_entries))
26e2805c 1513 failures = 0
1514 max_failures = self.params.get('skip_playlist_after_errors') or float('inf')
71729754 1515 for i, entry_tuple in enumerate(entries, 1):
1516 playlist_index, entry = entry_tuple
53ed7066 1517 if 'playlist_index' in self.params.get('compat_options', []):
1518 playlist_index = playlistitems[i - 1] if playlistitems else i
30a074c2 1519 self.to_screen('[download] Downloading video %s of %s' % (i, n_entries))
1520 # This __x_forwarded_for_ip thing is a bit ugly but requires
1521 # minimal changes
1522 if x_forwarded_for:
1523 entry['__x_forwarded_for_ip'] = x_forwarded_for
1524 extra = {
1525 'n_entries': n_entries,
f59ae581 1526 '_last_playlist_index': max(playlistitems) if playlistitems else (playlistend or n_entries),
71729754 1527 'playlist_index': playlist_index,
1528 'playlist_autonumber': i,
30a074c2 1529 'playlist': playlist,
1530 'playlist_id': ie_result.get('id'),
1531 'playlist_title': ie_result.get('title'),
1532 'playlist_uploader': ie_result.get('uploader'),
1533 'playlist_uploader_id': ie_result.get('uploader_id'),
30a074c2 1534 'extractor': ie_result['extractor'],
1535 'webpage_url': ie_result['webpage_url'],
1536 'webpage_url_basename': url_basename(ie_result['webpage_url']),
1537 'extractor_key': ie_result['extractor_key'],
1538 }
1539
1540 if self._match_entry(entry, incomplete=True) is not None:
1541 continue
1542
1543 entry_result = self.__process_iterable_entry(entry, download, extra)
26e2805c 1544 if not entry_result:
1545 failures += 1
1546 if failures >= max_failures:
1547 self.report_error(
1548 'Skipping the remaining entries in playlist "%s" since %d items failed extraction' % (playlist, failures))
1549 break
30a074c2 1550 # TODO: skip failed (empty) entries?
1551 playlist_results.append(entry_result)
1552 ie_result['entries'] = playlist_results
1553 self.to_screen('[download] Finished downloading playlist: %s' % playlist)
1554 return ie_result
1555
a0566bbf 1556 @__handle_extraction_exceptions
1557 def __process_iterable_entry(self, entry, download, extra_info):
1558 return self.process_ie_result(
1559 entry, download=download, extra_info=extra_info)
1560
67134eab
JMF
1561 def _build_format_filter(self, filter_spec):
1562 " Returns a function to filter the formats according to the filter_spec "
083c9df9
PH
1563
1564 OPERATORS = {
1565 '<': operator.lt,
1566 '<=': operator.le,
1567 '>': operator.gt,
1568 '>=': operator.ge,
1569 '=': operator.eq,
1570 '!=': operator.ne,
1571 }
67134eab 1572 operator_rex = re.compile(r'''(?x)\s*
187986a8 1573 (?P<key>width|height|tbr|abr|vbr|asr|filesize|filesize_approx|fps)\s*
1574 (?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
1575 (?P<value>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?)\s*
083c9df9 1576 ''' % '|'.join(map(re.escape, OPERATORS.keys())))
187986a8 1577 m = operator_rex.fullmatch(filter_spec)
9ddb6925
S
1578 if m:
1579 try:
1580 comparison_value = int(m.group('value'))
1581 except ValueError:
1582 comparison_value = parse_filesize(m.group('value'))
1583 if comparison_value is None:
1584 comparison_value = parse_filesize(m.group('value') + 'B')
1585 if comparison_value is None:
1586 raise ValueError(
1587 'Invalid value %r in format specification %r' % (
67134eab 1588 m.group('value'), filter_spec))
9ddb6925
S
1589 op = OPERATORS[m.group('op')]
1590
083c9df9 1591 if not m:
9ddb6925
S
1592 STR_OPERATORS = {
1593 '=': operator.eq,
10d33b34
YCH
1594 '^=': lambda attr, value: attr.startswith(value),
1595 '$=': lambda attr, value: attr.endswith(value),
1596 '*=': lambda attr, value: value in attr,
9ddb6925 1597 }
187986a8 1598 str_operator_rex = re.compile(r'''(?x)\s*
1599 (?P<key>[a-zA-Z0-9._-]+)\s*
1600 (?P<negation>!\s*)?(?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
1601 (?P<value>[a-zA-Z0-9._-]+)\s*
9ddb6925 1602 ''' % '|'.join(map(re.escape, STR_OPERATORS.keys())))
187986a8 1603 m = str_operator_rex.fullmatch(filter_spec)
9ddb6925
S
1604 if m:
1605 comparison_value = m.group('value')
2cc779f4
S
1606 str_op = STR_OPERATORS[m.group('op')]
1607 if m.group('negation'):
e118a879 1608 op = lambda attr, value: not str_op(attr, value)
2cc779f4
S
1609 else:
1610 op = str_op
083c9df9 1611
9ddb6925 1612 if not m:
187986a8 1613 raise SyntaxError('Invalid filter specification %r' % filter_spec)
083c9df9
PH
1614
1615 def _filter(f):
1616 actual_value = f.get(m.group('key'))
1617 if actual_value is None:
1618 return m.group('none_inclusive')
1619 return op(actual_value, comparison_value)
67134eab
JMF
1620 return _filter
1621
0017d9ad 1622 def _default_format_spec(self, info_dict, download=True):
0017d9ad 1623
af0f7428
S
1624 def can_merge():
1625 merger = FFmpegMergerPP(self)
1626 return merger.available and merger.can_merge()
1627
91ebc640 1628 prefer_best = (
b7b04c78 1629 not self.params.get('simulate')
91ebc640 1630 and download
1631 and (
1632 not can_merge()
19807826 1633 or info_dict.get('is_live', False)
de6000d9 1634 or self.outtmpl_dict['default'] == '-'))
53ed7066 1635 compat = (
1636 prefer_best
1637 or self.params.get('allow_multiple_audio_streams', False)
1638 or 'format-spec' in self.params.get('compat_opts', []))
91ebc640 1639
1640 return (
53ed7066 1641 'best/bestvideo+bestaudio' if prefer_best
1642 else 'bestvideo*+bestaudio/best' if not compat
91ebc640 1643 else 'bestvideo+bestaudio/best')
0017d9ad 1644
67134eab
JMF
1645 def build_format_selector(self, format_spec):
1646 def syntax_error(note, start):
1647 message = (
1648 'Invalid format specification: '
1649 '{0}\n\t{1}\n\t{2}^'.format(note, format_spec, ' ' * start[1]))
1650 return SyntaxError(message)
1651
1652 PICKFIRST = 'PICKFIRST'
1653 MERGE = 'MERGE'
1654 SINGLE = 'SINGLE'
0130afb7 1655 GROUP = 'GROUP'
67134eab
JMF
1656 FormatSelector = collections.namedtuple('FormatSelector', ['type', 'selector', 'filters'])
1657
91ebc640 1658 allow_multiple_streams = {'audio': self.params.get('allow_multiple_audio_streams', False),
1659 'video': self.params.get('allow_multiple_video_streams', False)}
909d24dd 1660
e8e73840 1661 check_formats = self.params.get('check_formats')
1662
67134eab
JMF
1663 def _parse_filter(tokens):
1664 filter_parts = []
1665 for type, string, start, _, _ in tokens:
1666 if type == tokenize.OP and string == ']':
1667 return ''.join(filter_parts)
1668 else:
1669 filter_parts.append(string)
1670
232541df 1671 def _remove_unused_ops(tokens):
17cc1534 1672 # Remove operators that we don't use and join them with the surrounding strings
232541df
JMF
1673 # for example: 'mp4' '-' 'baseline' '-' '16x9' is converted to 'mp4-baseline-16x9'
1674 ALLOWED_OPS = ('/', '+', ',', '(', ')')
1675 last_string, last_start, last_end, last_line = None, None, None, None
1676 for type, string, start, end, line in tokens:
1677 if type == tokenize.OP and string == '[':
1678 if last_string:
1679 yield tokenize.NAME, last_string, last_start, last_end, last_line
1680 last_string = None
1681 yield type, string, start, end, line
1682 # everything inside brackets will be handled by _parse_filter
1683 for type, string, start, end, line in tokens:
1684 yield type, string, start, end, line
1685 if type == tokenize.OP and string == ']':
1686 break
1687 elif type == tokenize.OP and string in ALLOWED_OPS:
1688 if last_string:
1689 yield tokenize.NAME, last_string, last_start, last_end, last_line
1690 last_string = None
1691 yield type, string, start, end, line
1692 elif type in [tokenize.NAME, tokenize.NUMBER, tokenize.OP]:
1693 if not last_string:
1694 last_string = string
1695 last_start = start
1696 last_end = end
1697 else:
1698 last_string += string
1699 if last_string:
1700 yield tokenize.NAME, last_string, last_start, last_end, last_line
1701
cf2ac6df 1702 def _parse_format_selection(tokens, inside_merge=False, inside_choice=False, inside_group=False):
67134eab
JMF
1703 selectors = []
1704 current_selector = None
1705 for type, string, start, _, _ in tokens:
1706 # ENCODING is only defined in python 3.x
1707 if type == getattr(tokenize, 'ENCODING', None):
1708 continue
1709 elif type in [tokenize.NAME, tokenize.NUMBER]:
1710 current_selector = FormatSelector(SINGLE, string, [])
1711 elif type == tokenize.OP:
cf2ac6df
JMF
1712 if string == ')':
1713 if not inside_group:
1714 # ')' will be handled by the parentheses group
1715 tokens.restore_last_token()
67134eab 1716 break
cf2ac6df 1717 elif inside_merge and string in ['/', ',']:
0130afb7
JMF
1718 tokens.restore_last_token()
1719 break
cf2ac6df
JMF
1720 elif inside_choice and string == ',':
1721 tokens.restore_last_token()
1722 break
1723 elif string == ',':
0a31a350
JMF
1724 if not current_selector:
1725 raise syntax_error('"," must follow a format selector', start)
67134eab
JMF
1726 selectors.append(current_selector)
1727 current_selector = None
1728 elif string == '/':
d96d604e
JMF
1729 if not current_selector:
1730 raise syntax_error('"/" must follow a format selector', start)
67134eab 1731 first_choice = current_selector
cf2ac6df 1732 second_choice = _parse_format_selection(tokens, inside_choice=True)
f5f4a27a 1733 current_selector = FormatSelector(PICKFIRST, (first_choice, second_choice), [])
67134eab
JMF
1734 elif string == '[':
1735 if not current_selector:
1736 current_selector = FormatSelector(SINGLE, 'best', [])
1737 format_filter = _parse_filter(tokens)
1738 current_selector.filters.append(format_filter)
0130afb7
JMF
1739 elif string == '(':
1740 if current_selector:
1741 raise syntax_error('Unexpected "("', start)
cf2ac6df
JMF
1742 group = _parse_format_selection(tokens, inside_group=True)
1743 current_selector = FormatSelector(GROUP, group, [])
67134eab 1744 elif string == '+':
d03cfdce 1745 if not current_selector:
1746 raise syntax_error('Unexpected "+"', start)
1747 selector_1 = current_selector
1748 selector_2 = _parse_format_selection(tokens, inside_merge=True)
1749 if not selector_2:
1750 raise syntax_error('Expected a selector', start)
1751 current_selector = FormatSelector(MERGE, (selector_1, selector_2), [])
67134eab
JMF
1752 else:
1753 raise syntax_error('Operator not recognized: "{0}"'.format(string), start)
1754 elif type == tokenize.ENDMARKER:
1755 break
1756 if current_selector:
1757 selectors.append(current_selector)
1758 return selectors
1759
f8d4ad9a 1760 def _merge(formats_pair):
1761 format_1, format_2 = formats_pair
1762
1763 formats_info = []
1764 formats_info.extend(format_1.get('requested_formats', (format_1,)))
1765 formats_info.extend(format_2.get('requested_formats', (format_2,)))
1766
1767 if not allow_multiple_streams['video'] or not allow_multiple_streams['audio']:
551f9388 1768 get_no_more = {'video': False, 'audio': False}
f8d4ad9a 1769 for (i, fmt_info) in enumerate(formats_info):
551f9388 1770 if fmt_info.get('acodec') == fmt_info.get('vcodec') == 'none':
1771 formats_info.pop(i)
1772 continue
1773 for aud_vid in ['audio', 'video']:
f8d4ad9a 1774 if not allow_multiple_streams[aud_vid] and fmt_info.get(aud_vid[0] + 'codec') != 'none':
1775 if get_no_more[aud_vid]:
1776 formats_info.pop(i)
f5510afe 1777 break
f8d4ad9a 1778 get_no_more[aud_vid] = True
1779
1780 if len(formats_info) == 1:
1781 return formats_info[0]
1782
1783 video_fmts = [fmt_info for fmt_info in formats_info if fmt_info.get('vcodec') != 'none']
1784 audio_fmts = [fmt_info for fmt_info in formats_info if fmt_info.get('acodec') != 'none']
1785
1786 the_only_video = video_fmts[0] if len(video_fmts) == 1 else None
1787 the_only_audio = audio_fmts[0] if len(audio_fmts) == 1 else None
1788
1789 output_ext = self.params.get('merge_output_format')
1790 if not output_ext:
1791 if the_only_video:
1792 output_ext = the_only_video['ext']
1793 elif the_only_audio and not video_fmts:
1794 output_ext = the_only_audio['ext']
1795 else:
1796 output_ext = 'mkv'
1797
1798 new_dict = {
1799 'requested_formats': formats_info,
1800 'format': '+'.join(fmt_info.get('format') for fmt_info in formats_info),
1801 'format_id': '+'.join(fmt_info.get('format_id') for fmt_info in formats_info),
1802 'ext': output_ext,
1803 }
1804
1805 if the_only_video:
1806 new_dict.update({
1807 'width': the_only_video.get('width'),
1808 'height': the_only_video.get('height'),
1809 'resolution': the_only_video.get('resolution') or self.format_resolution(the_only_video),
1810 'fps': the_only_video.get('fps'),
1811 'vcodec': the_only_video.get('vcodec'),
1812 'vbr': the_only_video.get('vbr'),
1813 'stretched_ratio': the_only_video.get('stretched_ratio'),
1814 })
1815
1816 if the_only_audio:
1817 new_dict.update({
1818 'acodec': the_only_audio.get('acodec'),
1819 'abr': the_only_audio.get('abr'),
1820 })
1821
1822 return new_dict
1823
e8e73840 1824 def _check_formats(formats):
981052c9 1825 if not check_formats:
1826 yield from formats
b5ac45b1 1827 return
e8e73840 1828 for f in formats:
1829 self.to_screen('[info] Testing format %s' % f['format_id'])
21cd8fae 1830 temp_file = tempfile.NamedTemporaryFile(
1831 suffix='.tmp', delete=False,
1832 dir=self.get_output_path('temp') or None)
1833 temp_file.close()
fe346461 1834 try:
981052c9 1835 success, _ = self.dl(temp_file.name, f, test=True)
1836 except (DownloadError, IOError, OSError, ValueError) + network_exceptions:
1837 success = False
fe346461 1838 finally:
21cd8fae 1839 if os.path.exists(temp_file.name):
1840 try:
1841 os.remove(temp_file.name)
1842 except OSError:
1843 self.report_warning('Unable to delete temporary file "%s"' % temp_file.name)
981052c9 1844 if success:
e8e73840 1845 yield f
1846 else:
1847 self.to_screen('[info] Unable to download format %s. Skipping...' % f['format_id'])
1848
67134eab 1849 def _build_selector_function(selector):
909d24dd 1850 if isinstance(selector, list): # ,
67134eab
JMF
1851 fs = [_build_selector_function(s) for s in selector]
1852
317f7ab6 1853 def selector_function(ctx):
67134eab 1854 for f in fs:
981052c9 1855 yield from f(ctx)
67134eab 1856 return selector_function
909d24dd 1857
1858 elif selector.type == GROUP: # ()
0130afb7 1859 selector_function = _build_selector_function(selector.selector)
909d24dd 1860
1861 elif selector.type == PICKFIRST: # /
67134eab
JMF
1862 fs = [_build_selector_function(s) for s in selector.selector]
1863
317f7ab6 1864 def selector_function(ctx):
67134eab 1865 for f in fs:
317f7ab6 1866 picked_formats = list(f(ctx))
67134eab
JMF
1867 if picked_formats:
1868 return picked_formats
1869 return []
67134eab 1870
981052c9 1871 elif selector.type == MERGE: # +
1872 selector_1, selector_2 = map(_build_selector_function, selector.selector)
1873
1874 def selector_function(ctx):
1875 for pair in itertools.product(
1876 selector_1(copy.deepcopy(ctx)), selector_2(copy.deepcopy(ctx))):
1877 yield _merge(pair)
1878
909d24dd 1879 elif selector.type == SINGLE: # atom
598d185d 1880 format_spec = selector.selector or 'best'
909d24dd 1881
f8d4ad9a 1882 # TODO: Add allvideo, allaudio etc by generalizing the code with best/worst selector
909d24dd 1883 if format_spec == 'all':
1884 def selector_function(ctx):
981052c9 1885 yield from _check_formats(ctx['formats'])
f8d4ad9a 1886 elif format_spec == 'mergeall':
1887 def selector_function(ctx):
981052c9 1888 formats = list(_check_formats(ctx['formats']))
e01d6aa4 1889 if not formats:
1890 return
921b76ca 1891 merged_format = formats[-1]
1892 for f in formats[-2::-1]:
f8d4ad9a 1893 merged_format = _merge((merged_format, f))
1894 yield merged_format
909d24dd 1895
1896 else:
e8e73840 1897 format_fallback, format_reverse, format_idx = False, True, 1
eff63539 1898 mobj = re.match(
1899 r'(?P<bw>best|worst|b|w)(?P<type>video|audio|v|a)?(?P<mod>\*)?(?:\.(?P<n>[1-9]\d*))?$',
1900 format_spec)
1901 if mobj is not None:
1902 format_idx = int_or_none(mobj.group('n'), default=1)
e8e73840 1903 format_reverse = mobj.group('bw')[0] == 'b'
eff63539 1904 format_type = (mobj.group('type') or [None])[0]
1905 not_format_type = {'v': 'a', 'a': 'v'}.get(format_type)
1906 format_modified = mobj.group('mod') is not None
909d24dd 1907
1908 format_fallback = not format_type and not format_modified # for b, w
8326b00a 1909 _filter_f = (
eff63539 1910 (lambda f: f.get('%scodec' % format_type) != 'none')
1911 if format_type and format_modified # bv*, ba*, wv*, wa*
1912 else (lambda f: f.get('%scodec' % not_format_type) == 'none')
1913 if format_type # bv, ba, wv, wa
1914 else (lambda f: f.get('vcodec') != 'none' and f.get('acodec') != 'none')
1915 if not format_modified # b, w
8326b00a 1916 else lambda f: True) # b*, w*
1917 filter_f = lambda f: _filter_f(f) and (
1918 f.get('vcodec') != 'none' or f.get('acodec') != 'none')
67134eab 1919 else:
909d24dd 1920 filter_f = ((lambda f: f.get('ext') == format_spec)
1921 if format_spec in ['mp4', 'flv', 'webm', '3gp', 'm4a', 'mp3', 'ogg', 'aac', 'wav'] # extension
1922 else (lambda f: f.get('format_id') == format_spec)) # id
1923
1924 def selector_function(ctx):
1925 formats = list(ctx['formats'])
909d24dd 1926 matches = list(filter(filter_f, formats)) if filter_f is not None else formats
e8e73840 1927 if format_fallback and ctx['incomplete_formats'] and not matches:
909d24dd 1928 # for extractors with incomplete formats (audio only (soundcloud)
1929 # or video only (imgur)) best/worst will fallback to
1930 # best/worst {video,audio}-only format
e8e73840 1931 matches = formats
981052c9 1932 matches = LazyList(_check_formats(matches[::-1 if format_reverse else 1]))
1933 try:
e8e73840 1934 yield matches[format_idx - 1]
981052c9 1935 except IndexError:
1936 return
083c9df9 1937
67134eab 1938 filters = [self._build_format_filter(f) for f in selector.filters]
083c9df9 1939
317f7ab6
S
1940 def final_selector(ctx):
1941 ctx_copy = copy.deepcopy(ctx)
67134eab 1942 for _filter in filters:
317f7ab6
S
1943 ctx_copy['formats'] = list(filter(_filter, ctx_copy['formats']))
1944 return selector_function(ctx_copy)
67134eab 1945 return final_selector
083c9df9 1946
67134eab 1947 stream = io.BytesIO(format_spec.encode('utf-8'))
0130afb7 1948 try:
232541df 1949 tokens = list(_remove_unused_ops(compat_tokenize_tokenize(stream.readline)))
0130afb7
JMF
1950 except tokenize.TokenError:
1951 raise syntax_error('Missing closing/opening brackets or parenthesis', (0, len(format_spec)))
1952
1953 class TokenIterator(object):
1954 def __init__(self, tokens):
1955 self.tokens = tokens
1956 self.counter = 0
1957
1958 def __iter__(self):
1959 return self
1960
1961 def __next__(self):
1962 if self.counter >= len(self.tokens):
1963 raise StopIteration()
1964 value = self.tokens[self.counter]
1965 self.counter += 1
1966 return value
1967
1968 next = __next__
1969
1970 def restore_last_token(self):
1971 self.counter -= 1
1972
1973 parsed_selector = _parse_format_selection(iter(TokenIterator(tokens)))
67134eab 1974 return _build_selector_function(parsed_selector)
a9c58ad9 1975
e5660ee6
JMF
1976 def _calc_headers(self, info_dict):
1977 res = std_headers.copy()
1978
1979 add_headers = info_dict.get('http_headers')
1980 if add_headers:
1981 res.update(add_headers)
1982
1983 cookies = self._calc_cookies(info_dict)
1984 if cookies:
1985 res['Cookie'] = cookies
1986
0016b84e
S
1987 if 'X-Forwarded-For' not in res:
1988 x_forwarded_for_ip = info_dict.get('__x_forwarded_for_ip')
1989 if x_forwarded_for_ip:
1990 res['X-Forwarded-For'] = x_forwarded_for_ip
1991
e5660ee6
JMF
1992 return res
1993
1994 def _calc_cookies(self, info_dict):
5c2266df 1995 pr = sanitized_Request(info_dict['url'])
e5660ee6 1996 self.cookiejar.add_cookie_header(pr)
662435f7 1997 return pr.get_header('Cookie')
e5660ee6 1998
b0249bca 1999 def _sanitize_thumbnails(self, info_dict):
bc516a3f 2000 thumbnails = info_dict.get('thumbnails')
2001 if thumbnails is None:
2002 thumbnail = info_dict.get('thumbnail')
2003 if thumbnail:
2004 info_dict['thumbnails'] = thumbnails = [{'url': thumbnail}]
2005 if thumbnails:
2006 thumbnails.sort(key=lambda t: (
2007 t.get('preference') if t.get('preference') is not None else -1,
2008 t.get('width') if t.get('width') is not None else -1,
2009 t.get('height') if t.get('height') is not None else -1,
2010 t.get('id') if t.get('id') is not None else '',
2011 t.get('url')))
b0249bca 2012
0ba692ac 2013 def thumbnail_tester():
2014 if self.params.get('check_formats'):
cca80fe6 2015 test_all = True
2016 to_screen = lambda msg: self.to_screen(f'[info] {msg}')
0ba692ac 2017 else:
cca80fe6 2018 test_all = False
0ba692ac 2019 to_screen = self.write_debug
2020
2021 def test_thumbnail(t):
cca80fe6 2022 if not test_all and not t.get('_test_url'):
2023 return True
0ba692ac 2024 to_screen('Testing thumbnail %s' % t['id'])
2025 try:
2026 self.urlopen(HEADRequest(t['url']))
2027 except network_exceptions as err:
2028 to_screen('Unable to connect to thumbnail %s URL "%s" - %s. Skipping...' % (
2029 t['id'], t['url'], error_to_compat_str(err)))
2030 return False
2031 return True
2032
2033 return test_thumbnail
b0249bca 2034
bc516a3f 2035 for i, t in enumerate(thumbnails):
bc516a3f 2036 if t.get('id') is None:
2037 t['id'] = '%d' % i
b0249bca 2038 if t.get('width') and t.get('height'):
2039 t['resolution'] = '%dx%d' % (t['width'], t['height'])
2040 t['url'] = sanitize_url(t['url'])
0ba692ac 2041
2042 if self.params.get('check_formats') is not False:
2043 info_dict['thumbnails'] = LazyList(filter(thumbnail_tester(), thumbnails[::-1])).reverse()
2044 else:
2045 info_dict['thumbnails'] = thumbnails
bc516a3f 2046
dd82ffea
JMF
2047 def process_video_result(self, info_dict, download=True):
2048 assert info_dict.get('_type', 'video') == 'video'
2049
bec1fad2
PH
2050 if 'id' not in info_dict:
2051 raise ExtractorError('Missing "id" field in extractor result')
2052 if 'title' not in info_dict:
2053 raise ExtractorError('Missing "title" field in extractor result')
2054
c9969434
S
2055 def report_force_conversion(field, field_not, conversion):
2056 self.report_warning(
2057 '"%s" field is not %s - forcing %s conversion, there is an error in extractor'
2058 % (field, field_not, conversion))
2059
2060 def sanitize_string_field(info, string_field):
2061 field = info.get(string_field)
2062 if field is None or isinstance(field, compat_str):
2063 return
2064 report_force_conversion(string_field, 'a string', 'string')
2065 info[string_field] = compat_str(field)
2066
2067 def sanitize_numeric_fields(info):
2068 for numeric_field in self._NUMERIC_FIELDS:
2069 field = info.get(numeric_field)
2070 if field is None or isinstance(field, compat_numeric_types):
2071 continue
2072 report_force_conversion(numeric_field, 'numeric', 'int')
2073 info[numeric_field] = int_or_none(field)
2074
2075 sanitize_string_field(info_dict, 'id')
2076 sanitize_numeric_fields(info_dict)
be6217b2 2077
dd82ffea
JMF
2078 if 'playlist' not in info_dict:
2079 # It isn't part of a playlist
2080 info_dict['playlist'] = None
2081 info_dict['playlist_index'] = None
2082
bc516a3f 2083 self._sanitize_thumbnails(info_dict)
d5519808 2084
536a55da 2085 thumbnail = info_dict.get('thumbnail')
bc516a3f 2086 thumbnails = info_dict.get('thumbnails')
536a55da
S
2087 if thumbnail:
2088 info_dict['thumbnail'] = sanitize_url(thumbnail)
2089 elif thumbnails:
d5519808
PH
2090 info_dict['thumbnail'] = thumbnails[-1]['url']
2091
ae30b840 2092 if info_dict.get('display_id') is None and 'id' in info_dict:
0afef30b
PH
2093 info_dict['display_id'] = info_dict['id']
2094
10db0d2f 2095 for ts_key, date_key in (
2096 ('timestamp', 'upload_date'),
2097 ('release_timestamp', 'release_date'),
2098 ):
2099 if info_dict.get(date_key) is None and info_dict.get(ts_key) is not None:
2100 # Working around out-of-range timestamp values (e.g. negative ones on Windows,
2101 # see http://bugs.python.org/issue1646728)
2102 try:
2103 upload_date = datetime.datetime.utcfromtimestamp(info_dict[ts_key])
2104 info_dict[date_key] = upload_date.strftime('%Y%m%d')
2105 except (ValueError, OverflowError, OSError):
2106 pass
9d2ecdbc 2107
ae30b840 2108 live_keys = ('is_live', 'was_live')
2109 live_status = info_dict.get('live_status')
2110 if live_status is None:
2111 for key in live_keys:
2112 if info_dict.get(key) is False:
2113 continue
2114 if info_dict.get(key):
2115 live_status = key
2116 break
2117 if all(info_dict.get(key) is False for key in live_keys):
2118 live_status = 'not_live'
2119 if live_status:
2120 info_dict['live_status'] = live_status
2121 for key in live_keys:
2122 if info_dict.get(key) is None:
2123 info_dict[key] = (live_status == key)
2124
33d2fc2f
S
2125 # Auto generate title fields corresponding to the *_number fields when missing
2126 # in order to always have clean titles. This is very common for TV series.
2127 for field in ('chapter', 'season', 'episode'):
2128 if info_dict.get('%s_number' % field) is not None and not info_dict.get(field):
2129 info_dict[field] = '%s %d' % (field.capitalize(), info_dict['%s_number' % field])
2130
05108a49
S
2131 for cc_kind in ('subtitles', 'automatic_captions'):
2132 cc = info_dict.get(cc_kind)
2133 if cc:
2134 for _, subtitle in cc.items():
2135 for subtitle_format in subtitle:
2136 if subtitle_format.get('url'):
2137 subtitle_format['url'] = sanitize_url(subtitle_format['url'])
2138 if subtitle_format.get('ext') is None:
2139 subtitle_format['ext'] = determine_ext(subtitle_format['url']).lower()
2140
2141 automatic_captions = info_dict.get('automatic_captions')
4bba3716 2142 subtitles = info_dict.get('subtitles')
4bba3716 2143
360e1ca5 2144 info_dict['requested_subtitles'] = self.process_subtitles(
05108a49 2145 info_dict['id'], subtitles, automatic_captions)
a504ced0 2146
dd82ffea
JMF
2147 # We now pick which formats have to be downloaded
2148 if info_dict.get('formats') is None:
2149 # There's only one format available
2150 formats = [info_dict]
2151 else:
2152 formats = info_dict['formats']
2153
db95dc13 2154 if not formats:
b7da73eb 2155 if not self.params.get('ignore_no_formats_error'):
2156 raise ExtractorError('No video formats found!')
2157 else:
2158 self.report_warning('No video formats found!')
db95dc13 2159
73af5cc8
S
2160 def is_wellformed(f):
2161 url = f.get('url')
a5ac0c47 2162 if not url:
73af5cc8
S
2163 self.report_warning(
2164 '"url" field is missing or empty - skipping format, '
2165 'there is an error in extractor')
a5ac0c47
S
2166 return False
2167 if isinstance(url, bytes):
2168 sanitize_string_field(f, 'url')
2169 return True
73af5cc8
S
2170
2171 # Filter out malformed formats for better extraction robustness
2172 formats = list(filter(is_wellformed, formats))
2173
181c7053
S
2174 formats_dict = {}
2175
dd82ffea 2176 # We check that all the formats have the format and format_id fields
db95dc13 2177 for i, format in enumerate(formats):
c9969434
S
2178 sanitize_string_field(format, 'format_id')
2179 sanitize_numeric_fields(format)
dcf77cf1 2180 format['url'] = sanitize_url(format['url'])
e74e3b63 2181 if not format.get('format_id'):
8016c922 2182 format['format_id'] = compat_str(i)
e2effb08
S
2183 else:
2184 # Sanitize format_id from characters used in format selector expression
ec85ded8 2185 format['format_id'] = re.sub(r'[\s,/+\[\]()]', '_', format['format_id'])
181c7053
S
2186 format_id = format['format_id']
2187 if format_id not in formats_dict:
2188 formats_dict[format_id] = []
2189 formats_dict[format_id].append(format)
2190
2191 # Make sure all formats have unique format_id
2192 for format_id, ambiguous_formats in formats_dict.items():
2193 if len(ambiguous_formats) > 1:
2194 for i, format in enumerate(ambiguous_formats):
2195 format['format_id'] = '%s-%d' % (format_id, i)
2196
2197 for i, format in enumerate(formats):
8c51aa65 2198 if format.get('format') is None:
6febd1c1 2199 format['format'] = '{id} - {res}{note}'.format(
8c51aa65
JMF
2200 id=format['format_id'],
2201 res=self.format_resolution(format),
b868936c 2202 note=format_field(format, 'format_note', ' (%s)'),
8c51aa65 2203 )
c1002e96 2204 # Automatically determine file extension if missing
5b1d8575 2205 if format.get('ext') is None:
cce929ea 2206 format['ext'] = determine_ext(format['url']).lower()
b5559424
S
2207 # Automatically determine protocol if missing (useful for format
2208 # selection purposes)
6f0be937 2209 if format.get('protocol') is None:
b5559424 2210 format['protocol'] = determine_protocol(format)
e5660ee6
JMF
2211 # Add HTTP headers, so that external programs can use them from the
2212 # json output
2213 full_format_info = info_dict.copy()
2214 full_format_info.update(format)
2215 format['http_headers'] = self._calc_headers(full_format_info)
0016b84e
S
2216 # Remove private housekeeping stuff
2217 if '__x_forwarded_for_ip' in info_dict:
2218 del info_dict['__x_forwarded_for_ip']
dd82ffea 2219
4bcc7bd1 2220 # TODO Central sorting goes here
99e206d5 2221
b7da73eb 2222 if formats and formats[0] is not info_dict:
b3d9ef88
JMF
2223 # only set the 'formats' fields if the original info_dict list them
2224 # otherwise we end up with a circular reference, the first (and unique)
f89197d7 2225 # element in the 'formats' field in info_dict is info_dict itself,
dfb1b146 2226 # which can't be exported to json
b3d9ef88 2227 info_dict['formats'] = formats
4ec82a72 2228
2229 info_dict, _ = self.pre_process(info_dict)
2230
b7b04c78 2231 if self.params.get('list_thumbnails'):
2232 self.list_thumbnails(info_dict)
2233 if self.params.get('listformats'):
86c66b2d 2234 if not info_dict.get('formats') and not info_dict.get('url'):
b7b04c78 2235 raise ExtractorError('No video formats found', expected=True)
2236 self.list_formats(info_dict)
2237 if self.params.get('listsubtitles'):
2238 if 'automatic_captions' in info_dict:
2239 self.list_subtitles(
2240 info_dict['id'], automatic_captions, 'automatic captions')
2241 self.list_subtitles(info_dict['id'], subtitles, 'subtitles')
2242 list_only = self.params.get('simulate') is None and (
2243 self.params.get('list_thumbnails') or self.params.get('listformats') or self.params.get('listsubtitles'))
169dbde9 2244 if list_only:
b7b04c78 2245 # Without this printing, -F --print-json will not work
169dbde9 2246 self.__forced_printings(info_dict, self.prepare_filename(info_dict), incomplete=True)
bfaae0a7 2247 return
2248
187986a8 2249 format_selector = self.format_selector
2250 if format_selector is None:
0017d9ad 2251 req_format = self._default_format_spec(info_dict, download=download)
0760b0a7 2252 self.write_debug('Default format spec: %s' % req_format)
187986a8 2253 format_selector = self.build_format_selector(req_format)
317f7ab6
S
2254
2255 # While in format selection we may need to have an access to the original
2256 # format set in order to calculate some metrics or do some processing.
2257 # For now we need to be able to guess whether original formats provided
2258 # by extractor are incomplete or not (i.e. whether extractor provides only
2259 # video-only or audio-only formats) for proper formats selection for
2260 # extractors with such incomplete formats (see
067aa17e 2261 # https://github.com/ytdl-org/youtube-dl/pull/5556).
317f7ab6
S
2262 # Since formats may be filtered during format selection and may not match
2263 # the original formats the results may be incorrect. Thus original formats
2264 # or pre-calculated metrics should be passed to format selection routines
2265 # as well.
2266 # We will pass a context object containing all necessary additional data
2267 # instead of just formats.
2268 # This fixes incorrect format selection issue (see
067aa17e 2269 # https://github.com/ytdl-org/youtube-dl/issues/10083).
2e221ca3 2270 incomplete_formats = (
317f7ab6 2271 # All formats are video-only or
3089bc74 2272 all(f.get('vcodec') != 'none' and f.get('acodec') == 'none' for f in formats)
317f7ab6 2273 # all formats are audio-only
3089bc74 2274 or all(f.get('vcodec') == 'none' and f.get('acodec') != 'none' for f in formats))
317f7ab6
S
2275
2276 ctx = {
2277 'formats': formats,
2278 'incomplete_formats': incomplete_formats,
2279 }
2280
2281 formats_to_download = list(format_selector(ctx))
dd82ffea 2282 if not formats_to_download:
b7da73eb 2283 if not self.params.get('ignore_no_formats_error'):
2284 raise ExtractorError('Requested format is not available', expected=True)
2285 else:
2286 self.report_warning('Requested format is not available')
4513a41a
A
2287 # Process what we can, even without any available formats.
2288 self.process_info(dict(info_dict))
b7da73eb 2289 elif download:
2290 self.to_screen(
07cce701 2291 '[info] %s: Downloading %d format(s): %s' % (
2292 info_dict['id'], len(formats_to_download),
2293 ", ".join([f['format_id'] for f in formats_to_download])))
b7da73eb 2294 for fmt in formats_to_download:
dd82ffea 2295 new_info = dict(info_dict)
4ec82a72 2296 # Save a reference to the original info_dict so that it can be modified in process_info if needed
2297 new_info['__original_infodict'] = info_dict
b7da73eb 2298 new_info.update(fmt)
dd82ffea
JMF
2299 self.process_info(new_info)
2300 # We update the info dict with the best quality format (backwards compatibility)
b7da73eb 2301 if formats_to_download:
2302 info_dict.update(formats_to_download[-1])
dd82ffea
JMF
2303 return info_dict
2304
98c70d6f 2305 def process_subtitles(self, video_id, normal_subtitles, automatic_captions):
a504ced0 2306 """Select the requested subtitles and their format"""
98c70d6f
JMF
2307 available_subs = {}
2308 if normal_subtitles and self.params.get('writesubtitles'):
2309 available_subs.update(normal_subtitles)
2310 if automatic_captions and self.params.get('writeautomaticsub'):
2311 for lang, cap_info in automatic_captions.items():
360e1ca5
JMF
2312 if lang not in available_subs:
2313 available_subs[lang] = cap_info
2314
4d171848
JMF
2315 if (not self.params.get('writesubtitles') and not
2316 self.params.get('writeautomaticsub') or not
2317 available_subs):
2318 return None
a504ced0 2319
c32b0aab 2320 all_sub_langs = available_subs.keys()
a504ced0 2321 if self.params.get('allsubtitles', False):
c32b0aab 2322 requested_langs = all_sub_langs
2323 elif self.params.get('subtitleslangs', False):
2324 requested_langs = set()
2325 for lang in self.params.get('subtitleslangs'):
2326 if lang == 'all':
2327 requested_langs.update(all_sub_langs)
2328 continue
2329 discard = lang[0] == '-'
2330 if discard:
2331 lang = lang[1:]
2332 current_langs = filter(re.compile(lang + '$').match, all_sub_langs)
2333 if discard:
2334 for lang in current_langs:
2335 requested_langs.discard(lang)
2336 else:
2337 requested_langs.update(current_langs)
2338 elif 'en' in available_subs:
2339 requested_langs = ['en']
a504ced0 2340 else:
c32b0aab 2341 requested_langs = [list(all_sub_langs)[0]]
ad3dc496 2342 if requested_langs:
2343 self.write_debug('Downloading subtitles: %s' % ', '.join(requested_langs))
a504ced0
JMF
2344
2345 formats_query = self.params.get('subtitlesformat', 'best')
2346 formats_preference = formats_query.split('/') if formats_query else []
2347 subs = {}
2348 for lang in requested_langs:
2349 formats = available_subs.get(lang)
2350 if formats is None:
2351 self.report_warning('%s subtitles not available for %s' % (lang, video_id))
2352 continue
a504ced0
JMF
2353 for ext in formats_preference:
2354 if ext == 'best':
2355 f = formats[-1]
2356 break
2357 matches = list(filter(lambda f: f['ext'] == ext, formats))
2358 if matches:
2359 f = matches[-1]
2360 break
2361 else:
2362 f = formats[-1]
2363 self.report_warning(
2364 'No subtitle format found matching "%s" for language %s, '
2365 'using %s' % (formats_query, lang, f['ext']))
2366 subs[lang] = f
2367 return subs
2368
d06daf23 2369 def __forced_printings(self, info_dict, filename, incomplete):
53c18592 2370 def print_mandatory(field, actual_field=None):
2371 if actual_field is None:
2372 actual_field = field
d06daf23 2373 if (self.params.get('force%s' % field, False)
53c18592 2374 and (not incomplete or info_dict.get(actual_field) is not None)):
2375 self.to_stdout(info_dict[actual_field])
d06daf23
S
2376
2377 def print_optional(field):
2378 if (self.params.get('force%s' % field, False)
2379 and info_dict.get(field) is not None):
2380 self.to_stdout(info_dict[field])
2381
53c18592 2382 info_dict = info_dict.copy()
2383 if filename is not None:
2384 info_dict['filename'] = filename
2385 if info_dict.get('requested_formats') is not None:
2386 # For RTMP URLs, also include the playpath
2387 info_dict['urls'] = '\n'.join(f['url'] + f.get('play_path', '') for f in info_dict['requested_formats'])
2388 elif 'url' in info_dict:
2389 info_dict['urls'] = info_dict['url'] + info_dict.get('play_path', '')
2390
2b8a2973 2391 if self.params.get('forceprint') or self.params.get('forcejson'):
2392 self.post_extract(info_dict)
53c18592 2393 for tmpl in self.params.get('forceprint', []):
2394 if re.match(r'\w+$', tmpl):
2395 tmpl = '%({})s'.format(tmpl)
2396 tmpl, info_copy = self.prepare_outtmpl(tmpl, info_dict)
901130bb 2397 self.to_stdout(self.escape_outtmpl(tmpl) % info_copy)
53c18592 2398
d06daf23
S
2399 print_mandatory('title')
2400 print_mandatory('id')
53c18592 2401 print_mandatory('url', 'urls')
d06daf23
S
2402 print_optional('thumbnail')
2403 print_optional('description')
53c18592 2404 print_optional('filename')
b868936c 2405 if self.params.get('forceduration') and info_dict.get('duration') is not None:
d06daf23
S
2406 self.to_stdout(formatSeconds(info_dict['duration']))
2407 print_mandatory('format')
53c18592 2408
2b8a2973 2409 if self.params.get('forcejson'):
6e84b215 2410 self.to_stdout(json.dumps(self.sanitize_info(info_dict)))
d06daf23 2411
e8e73840 2412 def dl(self, name, info, subtitle=False, test=False):
2413
2414 if test:
2415 verbose = self.params.get('verbose')
2416 params = {
2417 'test': True,
2418 'quiet': not verbose,
2419 'verbose': verbose,
2420 'noprogress': not verbose,
2421 'nopart': True,
2422 'skip_unavailable_fragments': False,
2423 'keep_fragments': False,
2424 'overwrites': True,
2425 '_no_ytdl_file': True,
2426 }
2427 else:
2428 params = self.params
96fccc10 2429 fd = get_suitable_downloader(info, params, to_stdout=(name == '-'))(self, params)
e8e73840 2430 if not test:
2431 for ph in self._progress_hooks:
2432 fd.add_progress_hook(ph)
18e674b4 2433 urls = '", "'.join([f['url'] for f in info.get('requested_formats', [])] or [info['url']])
2434 self.write_debug('Invoking downloader on "%s"' % urls)
e8e73840 2435 new_info = dict(info)
2436 if new_info.get('http_headers') is None:
2437 new_info['http_headers'] = self._calc_headers(new_info)
2438 return fd.download(name, new_info, subtitle)
2439
8222d8de
JMF
2440 def process_info(self, info_dict):
2441 """Process a single resolved IE result."""
2442
2443 assert info_dict.get('_type', 'video') == 'video'
fd288278
PH
2444
2445 max_downloads = self.params.get('max_downloads')
2446 if max_downloads is not None:
2447 if self._num_downloads >= int(max_downloads):
2448 raise MaxDownloadsReached()
8222d8de 2449
d06daf23 2450 # TODO: backward compatibility, to be removed
8222d8de 2451 info_dict['fulltitle'] = info_dict['title']
8222d8de 2452
4513a41a 2453 if 'format' not in info_dict and 'ext' in info_dict:
8222d8de
JMF
2454 info_dict['format'] = info_dict['ext']
2455
c77495e3 2456 if self._match_entry(info_dict) is not None:
8222d8de
JMF
2457 return
2458
277d6ff5 2459 self.post_extract(info_dict)
fd288278 2460 self._num_downloads += 1
8222d8de 2461
dcf64d43 2462 # info_dict['_filename'] needs to be set for backward compatibility
de6000d9 2463 info_dict['_filename'] = full_filename = self.prepare_filename(info_dict, warn=True)
2464 temp_filename = self.prepare_filename(info_dict, 'temp')
0202b52a 2465 files_to_move = {}
8222d8de
JMF
2466
2467 # Forced printings
4513a41a 2468 self.__forced_printings(info_dict, full_filename, incomplete=('format' not in info_dict))
8222d8de 2469
b7b04c78 2470 if self.params.get('simulate'):
2d30509f 2471 if self.params.get('force_write_download_archive', False):
2472 self.record_download_archive(info_dict)
2473
2474 # Do nothing else if in simulate mode
8222d8de
JMF
2475 return
2476
de6000d9 2477 if full_filename is None:
8222d8de
JMF
2478 return
2479
e92caff5 2480 if not self._ensure_dir_exists(encodeFilename(full_filename)):
0202b52a 2481 return
e92caff5 2482 if not self._ensure_dir_exists(encodeFilename(temp_filename)):
8222d8de
JMF
2483 return
2484
2485 if self.params.get('writedescription', False):
de6000d9 2486 descfn = self.prepare_filename(info_dict, 'description')
e92caff5 2487 if not self._ensure_dir_exists(encodeFilename(descfn)):
0202b52a 2488 return
0c3d0f51 2489 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(descfn)):
6febd1c1 2490 self.to_screen('[info] Video description is already present')
f00fd51d
JMF
2491 elif info_dict.get('description') is None:
2492 self.report_warning('There\'s no description to write.')
7b6fefc9
PH
2493 else:
2494 try:
6febd1c1 2495 self.to_screen('[info] Writing video description to: ' + descfn)
7b6fefc9
PH
2496 with io.open(encodeFilename(descfn), 'w', encoding='utf-8') as descfile:
2497 descfile.write(info_dict['description'])
7b6fefc9 2498 except (OSError, IOError):
6febd1c1 2499 self.report_error('Cannot write description file ' + descfn)
7b6fefc9 2500 return
8222d8de 2501
1fb07d10 2502 if self.params.get('writeannotations', False):
de6000d9 2503 annofn = self.prepare_filename(info_dict, 'annotation')
e92caff5 2504 if not self._ensure_dir_exists(encodeFilename(annofn)):
0202b52a 2505 return
0c3d0f51 2506 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(annofn)):
6febd1c1 2507 self.to_screen('[info] Video annotations are already present')
ffddb112
RA
2508 elif not info_dict.get('annotations'):
2509 self.report_warning('There are no annotations to write.')
7b6fefc9
PH
2510 else:
2511 try:
6febd1c1 2512 self.to_screen('[info] Writing video annotations to: ' + annofn)
7b6fefc9
PH
2513 with io.open(encodeFilename(annofn), 'w', encoding='utf-8') as annofile:
2514 annofile.write(info_dict['annotations'])
2515 except (KeyError, TypeError):
6febd1c1 2516 self.report_warning('There are no annotations to write.')
7b6fefc9 2517 except (OSError, IOError):
6febd1c1 2518 self.report_error('Cannot write annotations file: ' + annofn)
7b6fefc9 2519 return
1fb07d10 2520
c4a91be7 2521 subtitles_are_requested = any([self.params.get('writesubtitles', False),
0b7f3118 2522 self.params.get('writeautomaticsub')])
c4a91be7 2523
c84dd8a9 2524 if subtitles_are_requested and info_dict.get('requested_subtitles'):
8222d8de
JMF
2525 # subtitles download errors are already managed as troubles in relevant IE
2526 # that way it will silently go on when used with unsupporting IE
c84dd8a9 2527 subtitles = info_dict['requested_subtitles']
fa57af1e 2528 # ie = self.get_info_extractor(info_dict['extractor_key'])
a504ced0
JMF
2529 for sub_lang, sub_info in subtitles.items():
2530 sub_format = sub_info['ext']
56d868db 2531 sub_filename = subtitles_filename(temp_filename, sub_lang, sub_format, info_dict.get('ext'))
2532 sub_filename_final = subtitles_filename(
2533 self.prepare_filename(info_dict, 'subtitle'), sub_lang, sub_format, info_dict.get('ext'))
0c3d0f51 2534 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(sub_filename)):
5ff1bc0c 2535 self.to_screen('[info] Video subtitle %s.%s is already present' % (sub_lang, sub_format))
dcf64d43 2536 sub_info['filepath'] = sub_filename
0202b52a 2537 files_to_move[sub_filename] = sub_filename_final
a504ced0 2538 else:
0c9df79e 2539 self.to_screen('[info] Writing video subtitles to: ' + sub_filename)
5ff1bc0c
RA
2540 if sub_info.get('data') is not None:
2541 try:
2542 # Use newline='' to prevent conversion of newline characters
067aa17e 2543 # See https://github.com/ytdl-org/youtube-dl/issues/10268
5ff1bc0c
RA
2544 with io.open(encodeFilename(sub_filename), 'w', encoding='utf-8', newline='') as subfile:
2545 subfile.write(sub_info['data'])
dcf64d43 2546 sub_info['filepath'] = sub_filename
0202b52a 2547 files_to_move[sub_filename] = sub_filename_final
5ff1bc0c
RA
2548 except (OSError, IOError):
2549 self.report_error('Cannot write subtitles file ' + sub_filename)
2550 return
7b6fefc9 2551 else:
5ff1bc0c 2552 try:
e8e73840 2553 self.dl(sub_filename, sub_info.copy(), subtitle=True)
dcf64d43 2554 sub_info['filepath'] = sub_filename
0202b52a 2555 files_to_move[sub_filename] = sub_filename_final
fe346461 2556 except (ExtractorError, IOError, OSError, ValueError) + network_exceptions as err:
5ff1bc0c
RA
2557 self.report_warning('Unable to download subtitle for "%s": %s' %
2558 (sub_lang, error_to_compat_str(err)))
2559 continue
8222d8de 2560
8222d8de 2561 if self.params.get('writeinfojson', False):
de6000d9 2562 infofn = self.prepare_filename(info_dict, 'infojson')
e92caff5 2563 if not self._ensure_dir_exists(encodeFilename(infofn)):
0202b52a 2564 return
0c3d0f51 2565 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(infofn)):
66c935fb 2566 self.to_screen('[info] Video metadata is already present')
7b6fefc9 2567 else:
66c935fb 2568 self.to_screen('[info] Writing video metadata as JSON to: ' + infofn)
7b6fefc9 2569 try:
8012d892 2570 write_json_file(self.sanitize_info(info_dict, self.params.get('clean_infojson', True)), infofn)
7b6fefc9 2571 except (OSError, IOError):
66c935fb 2572 self.report_error('Cannot write video metadata to JSON file ' + infofn)
7b6fefc9 2573 return
de6000d9 2574 info_dict['__infojson_filename'] = infofn
8222d8de 2575
56d868db 2576 for thumb_ext in self._write_thumbnails(info_dict, temp_filename):
2577 thumb_filename_temp = replace_extension(temp_filename, thumb_ext, info_dict.get('ext'))
2578 thumb_filename = replace_extension(
2579 self.prepare_filename(info_dict, 'thumbnail'), thumb_ext, info_dict.get('ext'))
dcf64d43 2580 files_to_move[thumb_filename_temp] = thumb_filename
8222d8de 2581
732044af 2582 # Write internet shortcut files
2583 url_link = webloc_link = desktop_link = False
2584 if self.params.get('writelink', False):
2585 if sys.platform == "darwin": # macOS.
2586 webloc_link = True
2587 elif sys.platform.startswith("linux"):
2588 desktop_link = True
2589 else: # if sys.platform in ['win32', 'cygwin']:
2590 url_link = True
2591 if self.params.get('writeurllink', False):
2592 url_link = True
2593 if self.params.get('writewebloclink', False):
2594 webloc_link = True
2595 if self.params.get('writedesktoplink', False):
2596 desktop_link = True
2597
2598 if url_link or webloc_link or desktop_link:
2599 if 'webpage_url' not in info_dict:
2600 self.report_error('Cannot write internet shortcut file because the "webpage_url" field is missing in the media information')
2601 return
2602 ascii_url = iri_to_uri(info_dict['webpage_url'])
2603
2604 def _write_link_file(extension, template, newline, embed_filename):
0202b52a 2605 linkfn = replace_extension(full_filename, extension, info_dict.get('ext'))
10e3742e 2606 if self.params.get('overwrites', True) and os.path.exists(encodeFilename(linkfn)):
732044af 2607 self.to_screen('[info] Internet shortcut is already present')
2608 else:
2609 try:
2610 self.to_screen('[info] Writing internet shortcut to: ' + linkfn)
2611 with io.open(encodeFilename(to_high_limit_path(linkfn)), 'w', encoding='utf-8', newline=newline) as linkfile:
2612 template_vars = {'url': ascii_url}
2613 if embed_filename:
2614 template_vars['filename'] = linkfn[:-(len(extension) + 1)]
2615 linkfile.write(template % template_vars)
2616 except (OSError, IOError):
2617 self.report_error('Cannot write internet shortcut ' + linkfn)
2618 return False
2619 return True
2620
2621 if url_link:
2622 if not _write_link_file('url', DOT_URL_LINK_TEMPLATE, '\r\n', embed_filename=False):
2623 return
2624 if webloc_link:
2625 if not _write_link_file('webloc', DOT_WEBLOC_LINK_TEMPLATE, '\n', embed_filename=False):
2626 return
2627 if desktop_link:
2628 if not _write_link_file('desktop', DOT_DESKTOP_LINK_TEMPLATE, '\n', embed_filename=True):
2629 return
2630
56d868db 2631 try:
2632 info_dict, files_to_move = self.pre_process(info_dict, 'before_dl', files_to_move)
2633 except PostProcessingError as err:
2634 self.report_error('Preprocessing: %s' % str(err))
2635 return
2636
732044af 2637 must_record_download_archive = False
56d868db 2638 if self.params.get('skip_download', False):
2639 info_dict['filepath'] = temp_filename
2640 info_dict['__finaldir'] = os.path.dirname(os.path.abspath(encodeFilename(full_filename)))
2641 info_dict['__files_to_move'] = files_to_move
2642 info_dict = self.run_pp(MoveFilesAfterDownloadPP(self, False), info_dict)
2643 else:
2644 # Download
b868936c 2645 info_dict.setdefault('__postprocessors', [])
4340deca 2646 try:
0202b52a 2647
6b591b29 2648 def existing_file(*filepaths):
2649 ext = info_dict.get('ext')
2650 final_ext = self.params.get('final_ext', ext)
2651 existing_files = []
2652 for file in orderedSet(filepaths):
2653 if final_ext != ext:
2654 converted = replace_extension(file, final_ext, ext)
2655 if os.path.exists(encodeFilename(converted)):
2656 existing_files.append(converted)
2657 if os.path.exists(encodeFilename(file)):
2658 existing_files.append(file)
2659
2660 if not existing_files or self.params.get('overwrites', False):
2661 for file in orderedSet(existing_files):
2662 self.report_file_delete(file)
2663 os.remove(encodeFilename(file))
2664 return None
2665
6b591b29 2666 info_dict['ext'] = os.path.splitext(existing_files[0])[1][1:]
2667 return existing_files[0]
0202b52a 2668
2669 success = True
4340deca 2670 if info_dict.get('requested_formats') is not None:
81cd954a
S
2671
2672 def compatible_formats(formats):
d03cfdce 2673 # TODO: some formats actually allow this (mkv, webm, ogg, mp4), but not all of them.
2674 video_formats = [format for format in formats if format.get('vcodec') != 'none']
2675 audio_formats = [format for format in formats if format.get('acodec') != 'none']
2676 if len(video_formats) > 2 or len(audio_formats) > 2:
2677 return False
2678
81cd954a 2679 # Check extension
d03cfdce 2680 exts = set(format.get('ext') for format in formats)
2681 COMPATIBLE_EXTS = (
2682 set(('mp3', 'mp4', 'm4a', 'm4p', 'm4b', 'm4r', 'm4v', 'ismv', 'isma')),
2683 set(('webm',)),
2684 )
2685 for ext_sets in COMPATIBLE_EXTS:
2686 if ext_sets.issuperset(exts):
2687 return True
81cd954a
S
2688 # TODO: Check acodec/vcodec
2689 return False
2690
2691 requested_formats = info_dict['requested_formats']
0202b52a 2692 old_ext = info_dict['ext']
3b297919 2693 if self.params.get('merge_output_format') is None and not compatible_formats(requested_formats):
2694 info_dict['ext'] = 'mkv'
2695 self.report_warning(
2696 'Requested formats are incompatible for merge and will be merged into mkv.')
124bc071 2697 new_ext = info_dict['ext']
0202b52a 2698
124bc071 2699 def correct_ext(filename, ext=new_ext):
96fccc10 2700 if filename == '-':
2701 return filename
0202b52a 2702 filename_real_ext = os.path.splitext(filename)[1][1:]
2703 filename_wo_ext = (
2704 os.path.splitext(filename)[0]
124bc071 2705 if filename_real_ext in (old_ext, new_ext)
0202b52a 2706 else filename)
124bc071 2707 return '%s.%s' % (filename_wo_ext, ext)
0202b52a 2708
38c6902b 2709 # Ensure filename always has a correct extension for successful merge
0202b52a 2710 full_filename = correct_ext(full_filename)
2711 temp_filename = correct_ext(temp_filename)
2712 dl_filename = existing_file(full_filename, temp_filename)
1ea24129 2713 info_dict['__real_download'] = False
18e674b4 2714
2715 _protocols = set(determine_protocol(f) for f in requested_formats)
dbf5416a 2716 if len(_protocols) == 1: # All requested formats have same protocol
18e674b4 2717 info_dict['protocol'] = _protocols.pop()
dbf5416a 2718 directly_mergable = FFmpegFD.can_merge_formats(info_dict)
2719 if dl_filename is not None:
6c7274ec 2720 self.report_file_already_downloaded(dl_filename)
96fccc10 2721 elif (directly_mergable and get_suitable_downloader(
a46a815b 2722 info_dict, self.params, to_stdout=(temp_filename == '-')) == FFmpegFD):
dbf5416a 2723 info_dict['url'] = '\n'.join(f['url'] for f in requested_formats)
2724 success, real_download = self.dl(temp_filename, info_dict)
2725 info_dict['__real_download'] = real_download
18e674b4 2726 else:
2727 downloaded = []
2728 merger = FFmpegMergerPP(self)
2729 if self.params.get('allow_unplayable_formats'):
2730 self.report_warning(
2731 'You have requested merging of multiple formats '
2732 'while also allowing unplayable formats to be downloaded. '
2733 'The formats won\'t be merged to prevent data corruption.')
2734 elif not merger.available:
2735 self.report_warning(
2736 'You have requested merging of multiple formats but ffmpeg is not installed. '
2737 'The formats won\'t be merged.')
2738
96fccc10 2739 if temp_filename == '-':
2740 reason = ('using a downloader other than ffmpeg' if directly_mergable
2741 else 'but the formats are incompatible for simultaneous download' if merger.available
2742 else 'but ffmpeg is not installed')
2743 self.report_warning(
2744 f'You have requested downloading multiple formats to stdout {reason}. '
2745 'The formats will be streamed one after the other')
2746 fname = temp_filename
dbf5416a 2747 for f in requested_formats:
2748 new_info = dict(info_dict)
2749 del new_info['requested_formats']
2750 new_info.update(f)
96fccc10 2751 if temp_filename != '-':
124bc071 2752 fname = prepend_extension(
2753 correct_ext(temp_filename, new_info['ext']),
2754 'f%s' % f['format_id'], new_info['ext'])
96fccc10 2755 if not self._ensure_dir_exists(fname):
2756 return
2757 downloaded.append(fname)
dbf5416a 2758 partial_success, real_download = self.dl(fname, new_info)
2759 info_dict['__real_download'] = info_dict['__real_download'] or real_download
2760 success = success and partial_success
2761 if merger.available and not self.params.get('allow_unplayable_formats'):
2762 info_dict['__postprocessors'].append(merger)
2763 info_dict['__files_to_merge'] = downloaded
2764 # Even if there were no downloads, it is being merged only now
2765 info_dict['__real_download'] = True
2766 else:
2767 for file in downloaded:
2768 files_to_move[file] = None
4340deca
P
2769 else:
2770 # Just a single file
0202b52a 2771 dl_filename = existing_file(full_filename, temp_filename)
6c7274ec 2772 if dl_filename is None or dl_filename == temp_filename:
2773 # dl_filename == temp_filename could mean that the file was partially downloaded with --no-part.
2774 # So we should try to resume the download
e8e73840 2775 success, real_download = self.dl(temp_filename, info_dict)
0202b52a 2776 info_dict['__real_download'] = real_download
6c7274ec 2777 else:
2778 self.report_file_already_downloaded(dl_filename)
0202b52a 2779
0202b52a 2780 dl_filename = dl_filename or temp_filename
c571435f 2781 info_dict['__finaldir'] = os.path.dirname(os.path.abspath(encodeFilename(full_filename)))
0202b52a 2782
3158150c 2783 except network_exceptions as err:
7960b056 2784 self.report_error('unable to download video data: %s' % error_to_compat_str(err))
4340deca
P
2785 return
2786 except (OSError, IOError) as err:
2787 raise UnavailableVideoError(err)
2788 except (ContentTooShortError, ) as err:
2789 self.report_error('content too short (expected %s bytes and served %s)' % (err.expected, err.downloaded))
2790 return
8222d8de 2791
de6000d9 2792 if success and full_filename != '-':
f17f8651 2793
fd7cfb64 2794 def fixup():
2795 do_fixup = True
2796 fixup_policy = self.params.get('fixup')
2797 vid = info_dict['id']
2798
2799 if fixup_policy in ('ignore', 'never'):
2800 return
2801 elif fixup_policy == 'warn':
2802 do_fixup = False
f89b3e2d 2803 elif fixup_policy != 'force':
2804 assert fixup_policy in ('detect_or_warn', None)
2805 if not info_dict.get('__real_download'):
2806 do_fixup = False
fd7cfb64 2807
2808 def ffmpeg_fixup(cndn, msg, cls):
2809 if not cndn:
2810 return
2811 if not do_fixup:
2812 self.report_warning(f'{vid}: {msg}')
2813 return
2814 pp = cls(self)
2815 if pp.available:
2816 info_dict['__postprocessors'].append(pp)
2817 else:
2818 self.report_warning(f'{vid}: {msg}. Install ffmpeg to fix this automatically')
2819
2820 stretched_ratio = info_dict.get('stretched_ratio')
2821 ffmpeg_fixup(
2822 stretched_ratio not in (1, None),
2823 f'Non-uniform pixel ratio {stretched_ratio}',
2824 FFmpegFixupStretchedPP)
2825
2826 ffmpeg_fixup(
2827 (info_dict.get('requested_formats') is None
2828 and info_dict.get('container') == 'm4a_dash'
2829 and info_dict.get('ext') == 'm4a'),
2830 'writing DASH m4a. Only some players support this container',
2831 FFmpegFixupM4aPP)
2832
2833 downloader = (get_suitable_downloader(info_dict, self.params).__name__
2834 if 'protocol' in info_dict else None)
2835 ffmpeg_fixup(downloader == 'HlsFD', 'malformed AAC bitstream detected', FFmpegFixupM3u8PP)
e36d50c5 2836 ffmpeg_fixup(downloader == 'WebSocketFragmentFD', 'malformed timestamps detected', FFmpegFixupTimestampPP)
2837 ffmpeg_fixup(downloader == 'WebSocketFragmentFD', 'malformed duration detected', FFmpegFixupDurationPP)
fd7cfb64 2838
2839 fixup()
8222d8de 2840 try:
23c1a667 2841 info_dict = self.post_process(dl_filename, info_dict, files_to_move)
af819c21 2842 except PostProcessingError as err:
2843 self.report_error('Postprocessing: %s' % str(err))
8222d8de 2844 return
ab8e5e51
AM
2845 try:
2846 for ph in self._post_hooks:
23c1a667 2847 ph(info_dict['filepath'])
ab8e5e51
AM
2848 except Exception as err:
2849 self.report_error('post hooks: %s' % str(err))
2850 return
2d30509f 2851 must_record_download_archive = True
2852
2853 if must_record_download_archive or self.params.get('force_write_download_archive', False):
2854 self.record_download_archive(info_dict)
c3e6ffba 2855 max_downloads = self.params.get('max_downloads')
2856 if max_downloads is not None and self._num_downloads >= int(max_downloads):
2857 raise MaxDownloadsReached()
8222d8de
JMF
2858
2859 def download(self, url_list):
2860 """Download a given list of URLs."""
de6000d9 2861 outtmpl = self.outtmpl_dict['default']
3089bc74
S
2862 if (len(url_list) > 1
2863 and outtmpl != '-'
2864 and '%' not in outtmpl
2865 and self.params.get('max_downloads') != 1):
acd69589 2866 raise SameFileError(outtmpl)
8222d8de
JMF
2867
2868 for url in url_list:
2869 try:
5f6a1245 2870 # It also downloads the videos
61aa5ba3
S
2871 res = self.extract_info(
2872 url, force_generic_extractor=self.params.get('force_generic_extractor', False))
8222d8de 2873 except UnavailableVideoError:
6febd1c1 2874 self.report_error('unable to download video')
8222d8de 2875 except MaxDownloadsReached:
8b0d7497 2876 self.to_screen('[info] Maximum number of downloaded files reached')
2877 raise
2878 except ExistingVideoReached:
d83cb531 2879 self.to_screen('[info] Encountered a file that is already in the archive, stopping due to --break-on-existing')
8b0d7497 2880 raise
2881 except RejectedVideoReached:
d83cb531 2882 self.to_screen('[info] Encountered a file that did not match filter, stopping due to --break-on-reject')
8222d8de 2883 raise
63e0be34
PH
2884 else:
2885 if self.params.get('dump_single_json', False):
277d6ff5 2886 self.post_extract(res)
6e84b215 2887 self.to_stdout(json.dumps(self.sanitize_info(res)))
8222d8de
JMF
2888
2889 return self._download_retcode
2890
1dcc4c0c 2891 def download_with_info_file(self, info_filename):
31bd3925
JMF
2892 with contextlib.closing(fileinput.FileInput(
2893 [info_filename], mode='r',
2894 openhook=fileinput.hook_encoded('utf-8'))) as f:
2895 # FileInput doesn't have a read method, we can't call json.load
8012d892 2896 info = self.sanitize_info(json.loads('\n'.join(f)), self.params.get('clean_infojson', True))
d4943898
JMF
2897 try:
2898 self.process_ie_result(info, download=True)
d3f62c19 2899 except (DownloadError, EntryNotInPlaylist, ThrottledDownload):
d4943898
JMF
2900 webpage_url = info.get('webpage_url')
2901 if webpage_url is not None:
6febd1c1 2902 self.report_warning('The info failed to download, trying with "%s"' % webpage_url)
d4943898
JMF
2903 return self.download([webpage_url])
2904 else:
2905 raise
2906 return self._download_retcode
1dcc4c0c 2907
cb202fd2 2908 @staticmethod
8012d892 2909 def sanitize_info(info_dict, remove_private_keys=False):
2910 ''' Sanitize the infodict for converting to json '''
3ad56b42 2911 if info_dict is None:
2912 return info_dict
6e84b215 2913 info_dict.setdefault('epoch', int(time.time()))
2914 remove_keys = {'__original_infodict'} # Always remove this since this may contain a copy of the entire dict
ae8f99e6 2915 keep_keys = ['_type'], # Always keep this to facilitate load-info-json
8012d892 2916 if remove_private_keys:
6e84b215 2917 remove_keys |= {
2918 'requested_formats', 'requested_subtitles', 'requested_entries',
2919 'filepath', 'entries', 'original_url', 'playlist_autonumber',
2920 }
ae8f99e6 2921 empty_values = (None, {}, [], set(), tuple())
2922 reject = lambda k, v: k not in keep_keys and (
2923 k.startswith('_') or k in remove_keys or v in empty_values)
2924 else:
ae8f99e6 2925 reject = lambda k, v: k in remove_keys
5226731e 2926 filter_fn = lambda obj: (
b0249bca 2927 list(map(filter_fn, obj)) if isinstance(obj, (LazyList, list, tuple, set))
a515a78d 2928 else obj if not isinstance(obj, dict)
ae8f99e6 2929 else dict((k, filter_fn(v)) for k, v in obj.items() if not reject(k, v)))
5226731e 2930 return filter_fn(info_dict)
cb202fd2 2931
8012d892 2932 @staticmethod
2933 def filter_requested_info(info_dict, actually_filter=True):
2934 ''' Alias of sanitize_info for backward compatibility '''
2935 return YoutubeDL.sanitize_info(info_dict, actually_filter)
2936
dcf64d43 2937 def run_pp(self, pp, infodict):
5bfa4862 2938 files_to_delete = []
dcf64d43 2939 if '__files_to_move' not in infodict:
2940 infodict['__files_to_move'] = {}
af819c21 2941 files_to_delete, infodict = pp.run(infodict)
5bfa4862 2942 if not files_to_delete:
dcf64d43 2943 return infodict
5bfa4862 2944
2945 if self.params.get('keepvideo', False):
2946 for f in files_to_delete:
dcf64d43 2947 infodict['__files_to_move'].setdefault(f, '')
5bfa4862 2948 else:
2949 for old_filename in set(files_to_delete):
2950 self.to_screen('Deleting original file %s (pass -k to keep)' % old_filename)
2951 try:
2952 os.remove(encodeFilename(old_filename))
2953 except (IOError, OSError):
2954 self.report_warning('Unable to remove downloaded original file')
dcf64d43 2955 if old_filename in infodict['__files_to_move']:
2956 del infodict['__files_to_move'][old_filename]
2957 return infodict
5bfa4862 2958
277d6ff5 2959 @staticmethod
2960 def post_extract(info_dict):
2961 def actual_post_extract(info_dict):
2962 if info_dict.get('_type') in ('playlist', 'multi_video'):
2963 for video_dict in info_dict.get('entries', {}):
b050d210 2964 actual_post_extract(video_dict or {})
277d6ff5 2965 return
2966
07cce701 2967 post_extractor = info_dict.get('__post_extractor') or (lambda: {})
4ec82a72 2968 extra = post_extractor().items()
2969 info_dict.update(extra)
07cce701 2970 info_dict.pop('__post_extractor', None)
277d6ff5 2971
4ec82a72 2972 original_infodict = info_dict.get('__original_infodict') or {}
2973 original_infodict.update(extra)
2974 original_infodict.pop('__post_extractor', None)
2975
b050d210 2976 actual_post_extract(info_dict or {})
277d6ff5 2977
56d868db 2978 def pre_process(self, ie_info, key='pre_process', files_to_move=None):
5bfa4862 2979 info = dict(ie_info)
56d868db 2980 info['__files_to_move'] = files_to_move or {}
2981 for pp in self._pps[key]:
dcf64d43 2982 info = self.run_pp(pp, info)
56d868db 2983 return info, info.pop('__files_to_move', None)
5bfa4862 2984
dcf64d43 2985 def post_process(self, filename, ie_info, files_to_move=None):
8222d8de
JMF
2986 """Run all the postprocessors on the given file."""
2987 info = dict(ie_info)
2988 info['filepath'] = filename
dcf64d43 2989 info['__files_to_move'] = files_to_move or {}
0202b52a 2990
56d868db 2991 for pp in ie_info.get('__postprocessors', []) + self._pps['post_process']:
dcf64d43 2992 info = self.run_pp(pp, info)
2993 info = self.run_pp(MoveFilesAfterDownloadPP(self), info)
2994 del info['__files_to_move']
56d868db 2995 for pp in self._pps['after_move']:
dcf64d43 2996 info = self.run_pp(pp, info)
23c1a667 2997 return info
c1c9a79c 2998
5db07df6 2999 def _make_archive_id(self, info_dict):
e9fef7ee
S
3000 video_id = info_dict.get('id')
3001 if not video_id:
3002 return
5db07df6
PH
3003 # Future-proof against any change in case
3004 # and backwards compatibility with prior versions
e9fef7ee 3005 extractor = info_dict.get('extractor_key') or info_dict.get('ie_key') # key in a playlist
7012b23c 3006 if extractor is None:
1211bb6d
S
3007 url = str_or_none(info_dict.get('url'))
3008 if not url:
3009 return
e9fef7ee
S
3010 # Try to find matching extractor for the URL and take its ie_key
3011 for ie in self._ies:
1211bb6d 3012 if ie.suitable(url):
e9fef7ee
S
3013 extractor = ie.ie_key()
3014 break
3015 else:
3016 return
d0757229 3017 return '%s %s' % (extractor.lower(), video_id)
5db07df6
PH
3018
3019 def in_download_archive(self, info_dict):
3020 fn = self.params.get('download_archive')
3021 if fn is None:
3022 return False
3023
3024 vid_id = self._make_archive_id(info_dict)
e9fef7ee 3025 if not vid_id:
7012b23c 3026 return False # Incomplete video information
5db07df6 3027
a45e8619 3028 return vid_id in self.archive
c1c9a79c
PH
3029
3030 def record_download_archive(self, info_dict):
3031 fn = self.params.get('download_archive')
3032 if fn is None:
3033 return
5db07df6
PH
3034 vid_id = self._make_archive_id(info_dict)
3035 assert vid_id
c1c9a79c 3036 with locked_file(fn, 'a', encoding='utf-8') as archive_file:
6febd1c1 3037 archive_file.write(vid_id + '\n')
a45e8619 3038 self.archive.add(vid_id)
dd82ffea 3039
8c51aa65 3040 @staticmethod
8abeeb94 3041 def format_resolution(format, default='unknown'):
fb04e403 3042 if format.get('vcodec') == 'none':
8326b00a 3043 if format.get('acodec') == 'none':
3044 return 'images'
fb04e403 3045 return 'audio only'
f49d89ee
PH
3046 if format.get('resolution') is not None:
3047 return format['resolution']
35615307
DA
3048 if format.get('width') and format.get('height'):
3049 res = '%dx%d' % (format['width'], format['height'])
3050 elif format.get('height'):
3051 res = '%sp' % format['height']
3052 elif format.get('width'):
388ae76b 3053 res = '%dx?' % format['width']
8c51aa65 3054 else:
8abeeb94 3055 res = default
8c51aa65
JMF
3056 return res
3057
c57f7757
PH
3058 def _format_note(self, fdict):
3059 res = ''
3060 if fdict.get('ext') in ['f4f', 'f4m']:
3061 res += '(unsupported) '
32f90364
PH
3062 if fdict.get('language'):
3063 if res:
3064 res += ' '
9016d76f 3065 res += '[%s] ' % fdict['language']
c57f7757
PH
3066 if fdict.get('format_note') is not None:
3067 res += fdict['format_note'] + ' '
3068 if fdict.get('tbr') is not None:
3069 res += '%4dk ' % fdict['tbr']
3070 if fdict.get('container') is not None:
3071 if res:
3072 res += ', '
3073 res += '%s container' % fdict['container']
3089bc74
S
3074 if (fdict.get('vcodec') is not None
3075 and fdict.get('vcodec') != 'none'):
c57f7757
PH
3076 if res:
3077 res += ', '
3078 res += fdict['vcodec']
91c7271a 3079 if fdict.get('vbr') is not None:
c57f7757
PH
3080 res += '@'
3081 elif fdict.get('vbr') is not None and fdict.get('abr') is not None:
3082 res += 'video@'
3083 if fdict.get('vbr') is not None:
3084 res += '%4dk' % fdict['vbr']
fbb21cf5 3085 if fdict.get('fps') is not None:
5d583bdf
S
3086 if res:
3087 res += ', '
3088 res += '%sfps' % fdict['fps']
c57f7757
PH
3089 if fdict.get('acodec') is not None:
3090 if res:
3091 res += ', '
3092 if fdict['acodec'] == 'none':
3093 res += 'video only'
3094 else:
3095 res += '%-5s' % fdict['acodec']
3096 elif fdict.get('abr') is not None:
3097 if res:
3098 res += ', '
3099 res += 'audio'
3100 if fdict.get('abr') is not None:
3101 res += '@%3dk' % fdict['abr']
3102 if fdict.get('asr') is not None:
3103 res += ' (%5dHz)' % fdict['asr']
3104 if fdict.get('filesize') is not None:
3105 if res:
3106 res += ', '
3107 res += format_bytes(fdict['filesize'])
9732d77e
PH
3108 elif fdict.get('filesize_approx') is not None:
3109 if res:
3110 res += ', '
3111 res += '~' + format_bytes(fdict['filesize_approx'])
c57f7757 3112 return res
91c7271a 3113
c57f7757 3114 def list_formats(self, info_dict):
94badb25 3115 formats = info_dict.get('formats', [info_dict])
53ed7066 3116 new_format = (
3117 'list-formats' not in self.params.get('compat_opts', [])
169dbde9 3118 and self.params.get('listformats_table', True) is not False)
76d321f6 3119 if new_format:
3120 table = [
3121 [
3122 format_field(f, 'format_id'),
3123 format_field(f, 'ext'),
3124 self.format_resolution(f),
3125 format_field(f, 'fps', '%d'),
3126 '|',
3127 format_field(f, 'filesize', ' %s', func=format_bytes) + format_field(f, 'filesize_approx', '~%s', func=format_bytes),
3128 format_field(f, 'tbr', '%4dk'),
52a8a1e1 3129 shorten_protocol_name(f.get('protocol', '').replace("native", "n")),
76d321f6 3130 '|',
3131 format_field(f, 'vcodec', default='unknown').replace('none', ''),
3132 format_field(f, 'vbr', '%4dk'),
3133 format_field(f, 'acodec', default='unknown').replace('none', ''),
3134 format_field(f, 'abr', '%3dk'),
3135 format_field(f, 'asr', '%5dHz'),
3f698246 3136 ', '.join(filter(None, (
3137 'UNSUPPORTED' if f.get('ext') in ('f4f', 'f4m') else '',
3138 format_field(f, 'language', '[%s]'),
3139 format_field(f, 'format_note'),
3140 format_field(f, 'container', ignore=(None, f.get('ext'))),
ea05b302 3141 ))),
3f698246 3142 ] for f in formats if f.get('preference') is None or f['preference'] >= -1000]
76d321f6 3143 header_line = ['ID', 'EXT', 'RESOLUTION', 'FPS', '|', ' FILESIZE', ' TBR', 'PROTO',
3f698246 3144 '|', 'VCODEC', ' VBR', 'ACODEC', ' ABR', ' ASR', 'MORE INFO']
76d321f6 3145 else:
3146 table = [
3147 [
3148 format_field(f, 'format_id'),
3149 format_field(f, 'ext'),
3150 self.format_resolution(f),
3151 self._format_note(f)]
3152 for f in formats
3153 if f.get('preference') is None or f['preference'] >= -1000]
3154 header_line = ['format code', 'extension', 'resolution', 'note']
57dd9a8f 3155
cfb56d1a 3156 self.to_screen(
169dbde9 3157 '[info] Available formats for %s:' % info_dict['id'])
3158 self.to_stdout(render_table(
bc97cdae 3159 header_line, table, delim=new_format, extraGap=(0 if new_format else 1), hideEmpty=new_format))
cfb56d1a
PH
3160
3161 def list_thumbnails(self, info_dict):
b0249bca 3162 thumbnails = list(info_dict.get('thumbnails'))
cfb56d1a 3163 if not thumbnails:
b7b72db9 3164 self.to_screen('[info] No thumbnails present for %s' % info_dict['id'])
3165 return
cfb56d1a
PH
3166
3167 self.to_screen(
3168 '[info] Thumbnails for %s:' % info_dict['id'])
169dbde9 3169 self.to_stdout(render_table(
cfb56d1a
PH
3170 ['ID', 'width', 'height', 'URL'],
3171 [[t['id'], t.get('width', 'unknown'), t.get('height', 'unknown'), t['url']] for t in thumbnails]))
dca08720 3172
360e1ca5 3173 def list_subtitles(self, video_id, subtitles, name='subtitles'):
a504ced0 3174 if not subtitles:
360e1ca5 3175 self.to_screen('%s has no %s' % (video_id, name))
a504ced0 3176 return
a504ced0 3177 self.to_screen(
edab9dbf 3178 'Available %s for %s:' % (name, video_id))
2412044c 3179
3180 def _row(lang, formats):
49c258e1 3181 exts, names = zip(*((f['ext'], f.get('name') or 'unknown') for f in reversed(formats)))
2412044c 3182 if len(set(names)) == 1:
7aee40c1 3183 names = [] if names[0] == 'unknown' else names[:1]
2412044c 3184 return [lang, ', '.join(names), ', '.join(exts)]
3185
169dbde9 3186 self.to_stdout(render_table(
2412044c 3187 ['Language', 'Name', 'Formats'],
3188 [_row(lang, formats) for lang, formats in subtitles.items()],
3189 hideEmpty=True))
a504ced0 3190
dca08720
PH
3191 def urlopen(self, req):
3192 """ Start an HTTP download """
82d8a8b6 3193 if isinstance(req, compat_basestring):
67dda517 3194 req = sanitized_Request(req)
19a41fc6 3195 return self._opener.open(req, timeout=self._socket_timeout)
dca08720
PH
3196
3197 def print_debug_header(self):
3198 if not self.params.get('verbose'):
3199 return
62fec3b2 3200
c6afed48
PH
3201 stdout_encoding = getattr(
3202 sys.stdout, 'encoding', 'missing (%s)' % type(sys.stdout).__name__)
b0472057 3203 encoding_str = (
734f90bb
PH
3204 '[debug] Encodings: locale %s, fs %s, out %s, pref %s\n' % (
3205 locale.getpreferredencoding(),
3206 sys.getfilesystemencoding(),
c6afed48 3207 stdout_encoding,
b0472057 3208 self.get_encoding()))
4192b51c 3209 write_string(encoding_str, encoding=None)
734f90bb 3210
e5813e53 3211 source = (
3212 '(exe)' if hasattr(sys, 'frozen')
3213 else '(zip)' if isinstance(globals().get('__loader__'), zipimporter)
3214 else '(source)' if os.path.basename(sys.argv[0]) == '__main__.py'
3215 else '')
3216 self._write_string('[debug] yt-dlp version %s %s\n' % (__version__, source))
e0986e31 3217 if _LAZY_LOADER:
f74980cb 3218 self._write_string('[debug] Lazy loading extractors enabled\n')
3219 if _PLUGIN_CLASSES:
3220 self._write_string(
3221 '[debug] Plugin Extractors: %s\n' % [ie.ie_key() for ie in _PLUGIN_CLASSES])
53ed7066 3222 if self.params.get('compat_opts'):
3223 self._write_string(
3224 '[debug] Compatibility options: %s\n' % ', '.join(self.params.get('compat_opts')))
dca08720
PH
3225 try:
3226 sp = subprocess.Popen(
3227 ['git', 'rev-parse', '--short', 'HEAD'],
3228 stdout=subprocess.PIPE, stderr=subprocess.PIPE,
3229 cwd=os.path.dirname(os.path.abspath(__file__)))
f5b1bca9 3230 out, err = process_communicate_or_kill(sp)
dca08720
PH
3231 out = out.decode().strip()
3232 if re.match('[0-9a-f]+', out):
f74980cb 3233 self._write_string('[debug] Git HEAD: %s\n' % out)
70a1165b 3234 except Exception:
dca08720
PH
3235 try:
3236 sys.exc_clear()
70a1165b 3237 except Exception:
dca08720 3238 pass
b300cda4
S
3239
3240 def python_implementation():
3241 impl_name = platform.python_implementation()
3242 if impl_name == 'PyPy' and hasattr(sys, 'pypy_version_info'):
3243 return impl_name + ' version %d.%d.%d' % sys.pypy_version_info[:3]
3244 return impl_name
3245
e5813e53 3246 self._write_string('[debug] Python version %s (%s %s) - %s\n' % (
3247 platform.python_version(),
3248 python_implementation(),
3249 platform.architecture()[0],
b300cda4 3250 platform_name()))
d28b5171 3251
73fac4e9 3252 exe_versions = FFmpegPostProcessor.get_versions(self)
4c83c967 3253 exe_versions['rtmpdump'] = rtmpdump_version()
feee8d32 3254 exe_versions['phantomjs'] = PhantomJSwrapper._version()
d28b5171 3255 exe_str = ', '.join(
2831b468 3256 f'{exe} {v}' for exe, v in sorted(exe_versions.items()) if v
3257 ) or 'none'
d28b5171 3258 self._write_string('[debug] exe versions: %s\n' % exe_str)
dca08720 3259
2831b468 3260 from .downloader.fragment import can_decrypt_frag
3261 from .downloader.websocket import has_websockets
3262 from .postprocessor.embedthumbnail import has_mutagen
3263 from .cookies import SQLITE_AVAILABLE, KEYRING_AVAILABLE
3264
ad3dc496 3265 lib_str = ', '.join(sorted(filter(None, (
2831b468 3266 can_decrypt_frag and 'pycryptodome',
3267 has_websockets and 'websockets',
3268 has_mutagen and 'mutagen',
3269 SQLITE_AVAILABLE and 'sqlite',
3270 KEYRING_AVAILABLE and 'keyring',
ad3dc496 3271 )))) or 'none'
2831b468 3272 self._write_string('[debug] Optional libraries: %s\n' % lib_str)
3273
dca08720
PH
3274 proxy_map = {}
3275 for handler in self._opener.handlers:
3276 if hasattr(handler, 'proxies'):
3277 proxy_map.update(handler.proxies)
734f90bb 3278 self._write_string('[debug] Proxy map: ' + compat_str(proxy_map) + '\n')
dca08720 3279
58b1f00d
PH
3280 if self.params.get('call_home', False):
3281 ipaddr = self.urlopen('https://yt-dl.org/ip').read().decode('utf-8')
3282 self._write_string('[debug] Public IP address: %s\n' % ipaddr)
f5546c0b 3283 return
58b1f00d
PH
3284 latest_version = self.urlopen(
3285 'https://yt-dl.org/latest/version').read().decode('utf-8')
3286 if version_tuple(latest_version) > version_tuple(__version__):
3287 self.report_warning(
3288 'You are using an outdated version (newest version: %s)! '
3289 'See https://yt-dl.org/update if you need help updating.' %
3290 latest_version)
3291
e344693b 3292 def _setup_opener(self):
6ad14cab 3293 timeout_val = self.params.get('socket_timeout')
19a41fc6 3294 self._socket_timeout = 600 if timeout_val is None else float(timeout_val)
6ad14cab 3295
982ee69a 3296 opts_cookiesfrombrowser = self.params.get('cookiesfrombrowser')
dca08720
PH
3297 opts_cookiefile = self.params.get('cookiefile')
3298 opts_proxy = self.params.get('proxy')
3299
982ee69a 3300 self.cookiejar = load_cookies(opts_cookiefile, opts_cookiesfrombrowser, self)
dca08720 3301
6a3f4c3f 3302 cookie_processor = YoutubeDLCookieProcessor(self.cookiejar)
dca08720
PH
3303 if opts_proxy is not None:
3304 if opts_proxy == '':
3305 proxies = {}
3306 else:
3307 proxies = {'http': opts_proxy, 'https': opts_proxy}
3308 else:
3309 proxies = compat_urllib_request.getproxies()
067aa17e 3310 # Set HTTPS proxy to HTTP one if given (https://github.com/ytdl-org/youtube-dl/issues/805)
dca08720
PH
3311 if 'http' in proxies and 'https' not in proxies:
3312 proxies['https'] = proxies['http']
91410c9b 3313 proxy_handler = PerRequestProxyHandler(proxies)
a0ddb8a2
PH
3314
3315 debuglevel = 1 if self.params.get('debug_printtraffic') else 0
be4a824d
PH
3316 https_handler = make_HTTPS_handler(self.params, debuglevel=debuglevel)
3317 ydlh = YoutubeDLHandler(self.params, debuglevel=debuglevel)
fca6dba8 3318 redirect_handler = YoutubeDLRedirectHandler()
8b172c2e 3319 data_handler = compat_urllib_request_DataHandler()
6240b0a2
JMF
3320
3321 # When passing our own FileHandler instance, build_opener won't add the
3322 # default FileHandler and allows us to disable the file protocol, which
3323 # can be used for malicious purposes (see
067aa17e 3324 # https://github.com/ytdl-org/youtube-dl/issues/8227)
6240b0a2
JMF
3325 file_handler = compat_urllib_request.FileHandler()
3326
3327 def file_open(*args, **kwargs):
7a5c1cfe 3328 raise compat_urllib_error.URLError('file:// scheme is explicitly disabled in yt-dlp for security reasons')
6240b0a2
JMF
3329 file_handler.file_open = file_open
3330
3331 opener = compat_urllib_request.build_opener(
fca6dba8 3332 proxy_handler, https_handler, cookie_processor, ydlh, redirect_handler, data_handler, file_handler)
2461f79d 3333
dca08720
PH
3334 # Delete the default user-agent header, which would otherwise apply in
3335 # cases where our custom HTTP handler doesn't come into play
067aa17e 3336 # (See https://github.com/ytdl-org/youtube-dl/issues/1309 for details)
dca08720
PH
3337 opener.addheaders = []
3338 self._opener = opener
62fec3b2
PH
3339
3340 def encode(self, s):
3341 if isinstance(s, bytes):
3342 return s # Already encoded
3343
3344 try:
3345 return s.encode(self.get_encoding())
3346 except UnicodeEncodeError as err:
3347 err.reason = err.reason + '. Check your system encoding configuration or use the --encoding option.'
3348 raise
3349
3350 def get_encoding(self):
3351 encoding = self.params.get('encoding')
3352 if encoding is None:
3353 encoding = preferredencoding()
3354 return encoding
ec82d85a 3355
de6000d9 3356 def _write_thumbnails(self, info_dict, filename): # return the extensions
6c4fd172 3357 write_all = self.params.get('write_all_thumbnails', False)
3358 thumbnails = []
3359 if write_all or self.params.get('writethumbnail', False):
0202b52a 3360 thumbnails = info_dict.get('thumbnails') or []
6c4fd172 3361 multiple = write_all and len(thumbnails) > 1
ec82d85a 3362
0202b52a 3363 ret = []
981052c9 3364 for t in thumbnails[::-1]:
ec82d85a 3365 thumb_ext = determine_ext(t['url'], 'jpg')
6c4fd172 3366 suffix = '%s.' % t['id'] if multiple else ''
3367 thumb_display_id = '%s ' % t['id'] if multiple else ''
885cc0b7 3368 thumb_filename = replace_extension(filename, suffix + thumb_ext, info_dict.get('ext'))
ec82d85a 3369
0c3d0f51 3370 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(thumb_filename)):
de6000d9 3371 ret.append(suffix + thumb_ext)
8ba87148 3372 t['filepath'] = thumb_filename
ec82d85a
PH
3373 self.to_screen('[%s] %s: Thumbnail %sis already present' %
3374 (info_dict['extractor'], info_dict['id'], thumb_display_id))
3375 else:
5ef7d9bd 3376 self.to_screen('[%s] %s: Downloading thumbnail %s ...' %
ec82d85a
PH
3377 (info_dict['extractor'], info_dict['id'], thumb_display_id))
3378 try:
3379 uf = self.urlopen(t['url'])
d3d89c32 3380 with open(encodeFilename(thumb_filename), 'wb') as thumbf:
ec82d85a 3381 shutil.copyfileobj(uf, thumbf)
de6000d9 3382 ret.append(suffix + thumb_ext)
ec82d85a
PH
3383 self.to_screen('[%s] %s: Writing thumbnail %sto: %s' %
3384 (info_dict['extractor'], info_dict['id'], thumb_display_id, thumb_filename))
885cc0b7 3385 t['filepath'] = thumb_filename
3158150c 3386 except network_exceptions as err:
ec82d85a 3387 self.report_warning('Unable to download thumbnail "%s": %s' %
9b9c5355 3388 (t['url'], error_to_compat_str(err)))
6c4fd172 3389 if ret and not write_all:
3390 break
0202b52a 3391 return ret