]> jfr.im git - yt-dlp.git/blame - youtube_dlc/YoutubeDL.py
[FormatSort] Allow user to prefer av01 over vp9
[yt-dlp.git] / youtube_dlc / YoutubeDL.py
CommitLineData
8222d8de 1#!/usr/bin/env python
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
JMF
22import socket
23import sys
24import time
67134eab 25import tokenize
8222d8de 26import traceback
75822ca7 27import random
8222d8de 28
961ea474
S
29from string import ascii_letters
30
8c25f81b 31from .compat import (
82d8a8b6 32 compat_basestring,
dca08720 33 compat_cookiejar,
003c69a8 34 compat_get_terminal_size,
ce02ed60 35 compat_http_client,
4f026faf 36 compat_kwargs,
d0d9ade4 37 compat_numeric_types,
e9c0cdd3 38 compat_os_name,
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
PH
44)
45from .utils import (
eedb7ba5
S
46 age_restricted,
47 args_to_str,
ce02ed60
PH
48 ContentTooShortError,
49 date_from_str,
50 DateRange,
acd69589 51 DEFAULT_OUTTMPL,
ce02ed60 52 determine_ext,
b5559424 53 determine_protocol,
732044af 54 DOT_DESKTOP_LINK_TEMPLATE,
55 DOT_URL_LINK_TEMPLATE,
56 DOT_WEBLOC_LINK_TEMPLATE,
ce02ed60 57 DownloadError,
c0384f22 58 encode_compat_str,
ce02ed60 59 encodeFilename,
9b9c5355 60 error_to_compat_str,
8b0d7497 61 ExistingVideoReached,
590bc6f6 62 expand_path,
ce02ed60 63 ExtractorError,
02dbf93f 64 format_bytes,
76d321f6 65 format_field,
525ef922 66 formatSeconds,
773f291d 67 GeoRestrictedError,
c9969434 68 int_or_none,
732044af 69 iri_to_uri,
773f291d 70 ISO3166Utils,
ce02ed60 71 locked_file,
0202b52a 72 make_dir,
dca08720 73 make_HTTPS_handler,
ce02ed60 74 MaxDownloadsReached,
cd6fc19e 75 orderedSet,
b7ab0590 76 PagedList,
083c9df9 77 parse_filesize,
91410c9b 78 PerRequestProxyHandler,
dca08720 79 platform_name,
eedb7ba5 80 PostProcessingError,
ce02ed60 81 preferredencoding,
eedb7ba5 82 prepend_extension,
51fb4995 83 register_socks_protocols,
cfb56d1a 84 render_table,
eedb7ba5 85 replace_extension,
8b0d7497 86 RejectedVideoReached,
ce02ed60
PH
87 SameFileError,
88 sanitize_filename,
1bb5c511 89 sanitize_path,
dcf77cf1 90 sanitize_url,
67dda517 91 sanitized_Request,
e5660ee6 92 std_headers,
1211bb6d 93 str_or_none,
ce02ed60 94 subtitles_filename,
732044af 95 to_high_limit_path,
ce02ed60 96 UnavailableVideoError,
29eb5174 97 url_basename,
58b1f00d 98 version_tuple,
ce02ed60
PH
99 write_json_file,
100 write_string,
1bab3437 101 YoutubeDLCookieJar,
6a3f4c3f 102 YoutubeDLCookieProcessor,
dca08720 103 YoutubeDLHandler,
fca6dba8 104 YoutubeDLRedirectHandler,
f5b1bca9 105 process_communicate_or_kill,
ce02ed60 106)
a0e07d31 107from .cache import Cache
f74980cb 108from .extractor import get_info_extractor, gen_extractor_classes, _LAZY_LOADER, _PLUGIN_CLASSES
4c54b89e 109from .extractor.openload import PhantomJSwrapper
3bc2ddcc 110from .downloader import get_suitable_downloader
4c83c967 111from .downloader.rtmp import rtmpdump_version
4f026faf 112from .postprocessor import (
f17f8651 113 FFmpegFixupM3u8PP,
62cd676c 114 FFmpegFixupM4aPP,
6271f1ca 115 FFmpegFixupStretchedPP,
4f026faf
PH
116 FFmpegMergerPP,
117 FFmpegPostProcessor,
0202b52a 118 # FFmpegSubtitlesConvertorPP,
4f026faf 119 get_postprocessor,
0202b52a 120 MoveFilesAfterDownloadPP,
4f026faf 121)
dca08720 122from .version import __version__
8222d8de 123
e9c0cdd3
YCH
124if compat_os_name == 'nt':
125 import ctypes
126
2459b6e1 127
8222d8de
JMF
128class YoutubeDL(object):
129 """YoutubeDL class.
130
131 YoutubeDL objects are the ones responsible of downloading the
132 actual video file and writing it to disk if the user has requested
133 it, among some other tasks. In most cases there should be one per
134 program. As, given a video URL, the downloader doesn't know how to
135 extract all the needed information, task that InfoExtractors do, it
136 has to pass the URL to one of them.
137
138 For this, YoutubeDL objects have a method that allows
139 InfoExtractors to be registered in a given order. When it is passed
140 a URL, the YoutubeDL object handles it to the first InfoExtractor it
141 finds that reports being able to handle it. The InfoExtractor extracts
142 all the information about the video or videos the URL refers to, and
143 YoutubeDL process the extracted information, possibly using a File
144 Downloader to download the video.
145
146 YoutubeDL objects accept a lot of parameters. In order not to saturate
147 the object constructor with arguments, it receives a dictionary of
148 options instead. These options are available through the params
149 attribute for the InfoExtractors to use. The YoutubeDL also
150 registers itself as the downloader in charge for the InfoExtractors
151 that are added to it, so this is a "mutual registration".
152
153 Available options:
154
155 username: Username for authentication purposes.
156 password: Password for authentication purposes.
180940e0 157 videopassword: Password for accessing a video.
1da50aa3
S
158 ap_mso: Adobe Pass multiple-system operator identifier.
159 ap_username: Multiple-system operator account username.
160 ap_password: Multiple-system operator account password.
8222d8de
JMF
161 usenetrc: Use netrc for authentication instead.
162 verbose: Print additional info to stdout.
163 quiet: Do not print messages to stdout.
ad8915b7 164 no_warnings: Do not print out anything for warnings.
8222d8de
JMF
165 forceurl: Force printing final URL.
166 forcetitle: Force printing title.
167 forceid: Force printing ID.
168 forcethumbnail: Force printing thumbnail URL.
169 forcedescription: Force printing description.
170 forcefilename: Force printing final filename.
525ef922 171 forceduration: Force printing duration.
8694c600 172 forcejson: Force printing info_dict as JSON.
63e0be34
PH
173 dump_single_json: Force printing the info_dict of the whole playlist
174 (or video) as a single JSON line.
2d30509f 175 force_write_download_archive: Force writing download archive regardless of
176 'skip_download' or 'simulate'.
8222d8de 177 simulate: Do not download the video files.
eb8a4433 178 format: Video format code. see "FORMAT SELECTION" for more details.
179 format_sort: How to sort the video formats. see "Sorting Formats" for more details.
180 format_sort_force: Force the given format_sort. see "Sorting Formats" for more details.
909d24dd 181 allow_multiple_video_streams: Allow multiple video streams to be merged into a single file
182 allow_multiple_audio_streams: Allow multiple audio streams to be merged into a single file
8222d8de 183 outtmpl: Template for output names.
a820dc72
RA
184 outtmpl_na_placeholder: Placeholder for unavailable meta fields.
185 restrictfilenames: Do not allow "&" and spaces in file names
186 trim_file_name: Limit length of filename (extension excluded)
187 ignoreerrors: Do not stop on download errors
188 (Default True when running youtube-dlc,
189 but False when directly accessing YoutubeDL class)
d22dec74 190 force_generic_extractor: Force downloader to use the generic extractor
0c3d0f51 191 overwrites: Overwrite all video and metadata files if True,
192 overwrite only non-video files if None
193 and don't overwrite any file if False
8222d8de
JMF
194 playliststart: Playlist item to start at.
195 playlistend: Playlist item to end at.
c14e88f0 196 playlist_items: Specific indices of playlist to download.
ff815fe6 197 playlistreverse: Download playlist items in reverse order.
75822ca7 198 playlistrandom: Download playlist items in random order.
8222d8de
JMF
199 matchtitle: Download only matching titles.
200 rejecttitle: Reject downloads for matching titles.
8bf9319e 201 logger: Log messages to a logging.Logger instance.
8222d8de
JMF
202 logtostderr: Log messages to stderr instead of stdout.
203 writedescription: Write the video description to a .description file
204 writeinfojson: Write the video description to a .info.json file
06167fbb 205 writecomments: Extract video comments. This will not be written to disk
206 unless writeinfojson is also given
1fb07d10 207 writeannotations: Write the video annotations to a .annotations.xml file
8222d8de 208 writethumbnail: Write the thumbnail image to a file
cac96421 209 allow_playlist_files: Also write playlists' description, infojson etc in a seperate file
ec82d85a 210 write_all_thumbnails: Write all thumbnail formats to files
732044af 211 writelink: Write an internet shortcut file, depending on the
212 current platform (.url/.webloc/.desktop)
213 writeurllink: Write a Windows internet shortcut file (.url)
214 writewebloclink: Write a macOS internet shortcut file (.webloc)
215 writedesktoplink: Write a Linux internet shortcut file (.desktop)
8222d8de 216 writesubtitles: Write the video subtitles to a file
741dd8ea 217 writeautomaticsub: Write the automatically generated subtitles to a file
8222d8de 218 allsubtitles: Downloads all the subtitles of the video
0b7f3118 219 (requires writesubtitles or writeautomaticsub)
8222d8de 220 listsubtitles: Lists all available subtitles for the video
a504ced0 221 subtitlesformat: The format code for subtitles
aa6a10c4 222 subtitleslangs: List of languages of the subtitles to download
8222d8de
JMF
223 keepvideo: Keep the video file after post-processing
224 daterange: A DateRange object, download only if the upload_date is in the range.
225 skip_download: Skip the actual download of the video file
c35f9e72 226 cachedir: Location of the cache files in the filesystem.
a0e07d31 227 False to disable filesystem cache.
47192f92 228 noplaylist: Download single video instead of a playlist if in doubt.
8dbe9899
PH
229 age_limit: An integer representing the user's age in years.
230 Unsuitable videos for the given age are skipped.
5fe18bdb
PH
231 min_views: An integer representing the minimum view count the video
232 must have in order to not be skipped.
233 Videos without view count information are always
234 downloaded. None for no limit.
235 max_views: An integer representing the maximum view count.
236 Videos that are more popular than that are not
237 downloaded.
238 Videos without view count information are always
239 downloaded. None for no limit.
240 download_archive: File name of a file where all downloads are recorded.
c1c9a79c
PH
241 Videos already present in the file are not downloaded
242 again.
8a51f564 243 break_on_existing: Stop the download process after attempting to download a
244 file that is in the archive.
245 break_on_reject: Stop the download process when encountering a video that
246 has been filtered out.
247 cookiefile: File name where cookies should be read from and dumped to
a1ee09e8 248 nocheckcertificate:Do not verify SSL certificates
7e8c0af0
PH
249 prefer_insecure: Use HTTP instead of HTTPS to retrieve information.
250 At the moment, this is only supported by YouTube.
a1ee09e8 251 proxy: URL of the proxy server to use
38cce791 252 geo_verification_proxy: URL of the proxy to use for IP address verification
504f20dd 253 on geo-restricted sites.
e344693b 254 socket_timeout: Time to wait for unresponsive hosts, in seconds
0783b09b
PH
255 bidi_workaround: Work around buggy terminals without bidirectional text
256 support, using fridibi
a0ddb8a2 257 debug_printtraffic:Print out sent and received HTTP traffic
7b0817e8 258 include_ads: Download ads as well
04b4d394
PH
259 default_search: Prepend this string if an input url is not valid.
260 'auto' for elaborate guessing
62fec3b2 261 encoding: Use this encoding instead of the system-specified.
e8ee972c 262 extract_flat: Do not resolve URLs, return the immediate result.
057a5206
PH
263 Pass in 'in_playlist' to only show this behavior for
264 playlist items.
4f026faf 265 postprocessors: A list of dictionaries, each with an entry
71b640cc 266 * key: The name of the postprocessor. See
cefecac1 267 youtube_dlc/postprocessor/__init__.py for a list.
0202b52a 268 * _after_move: Optional. If True, run this post_processor
269 after 'MoveFilesAfterDownload'
4f026faf
PH
270 as well as any further keyword arguments for the
271 postprocessor.
ab8e5e51
AM
272 post_hooks: A list of functions that get called as the final step
273 for each video file, after all postprocessors have been
274 called. The filename will be passed as the only argument.
71b640cc
PH
275 progress_hooks: A list of functions that get called on download
276 progress, with a dictionary with the entries
5cda4eda 277 * status: One of "downloading", "error", or "finished".
ee69b99a 278 Check this first and ignore unknown values.
71b640cc 279
5cda4eda 280 If status is one of "downloading", or "finished", the
ee69b99a
PH
281 following properties may also be present:
282 * filename: The final filename (always present)
5cda4eda 283 * tmpfilename: The filename we're currently writing to
71b640cc
PH
284 * downloaded_bytes: Bytes on disk
285 * total_bytes: Size of the whole file, None if unknown
5cda4eda
PH
286 * total_bytes_estimate: Guess of the eventual file size,
287 None if unavailable.
288 * elapsed: The number of seconds since download started.
71b640cc
PH
289 * eta: The estimated time in seconds, None if unknown
290 * speed: The download speed in bytes/second, None if
291 unknown
5cda4eda
PH
292 * fragment_index: The counter of the currently
293 downloaded video fragment.
294 * fragment_count: The number of fragments (= individual
295 files that will be merged)
71b640cc
PH
296
297 Progress hooks are guaranteed to be called at least once
298 (with status "finished") if the download is successful.
45598f15 299 merge_output_format: Extension to use when merging formats.
6b591b29 300 final_ext: Expected final extension; used to detect when the file was
301 already downloaded and converted. "merge_output_format" is
302 replaced by this extension when given
6271f1ca
PH
303 fixup: Automatically correct known faults of the file.
304 One of:
305 - "never": do nothing
306 - "warn": only emit a warning
307 - "detect_or_warn": check whether we can do anything
62cd676c 308 about it, warn otherwise (default)
504f20dd 309 source_address: Client-side IP address to bind to.
6ec6cb4e 310 call_home: Boolean, true iff we are allowed to contact the
cefecac1 311 youtube-dlc servers for debugging.
7aa589a5
S
312 sleep_interval: Number of seconds to sleep before each download when
313 used alone or a lower bound of a range for randomized
314 sleep before each download (minimum possible number
315 of seconds to sleep) when used along with
316 max_sleep_interval.
317 max_sleep_interval:Upper bound of a range for randomized sleep before each
318 download (maximum possible number of seconds to sleep).
319 Must only be used along with sleep_interval.
320 Actual sleep time will be a random float from range
321 [sleep_interval; max_sleep_interval].
cfb56d1a
PH
322 listformats: Print an overview of available video formats and exit.
323 list_thumbnails: Print a table of all thumbnails and exit.
347de493
PH
324 match_filter: A function that gets called with the info_dict of
325 every video.
326 If it returns a message, the video is ignored.
327 If it returns None, the video is downloaded.
328 match_filter_func in utils.py is one example for this.
7e5db8c9 329 no_color: Do not emit color codes in output.
0a840f58 330 geo_bypass: Bypass geographic restriction via faking X-Forwarded-For
504f20dd 331 HTTP header
0a840f58 332 geo_bypass_country:
773f291d
S
333 Two-letter ISO 3166-2 country code that will be used for
334 explicit geographic restriction bypassing via faking
504f20dd 335 X-Forwarded-For HTTP header
5f95927a
S
336 geo_bypass_ip_block:
337 IP range in CIDR notation that will be used similarly to
504f20dd 338 geo_bypass_country
71b640cc 339
85729c51
PH
340 The following options determine which downloader is picked:
341 external_downloader: Executable of the external downloader to call.
342 None or unset for standard (built-in) downloader.
bf09af3a
S
343 hls_prefer_native: Use the native HLS downloader instead of ffmpeg/avconv
344 if True, otherwise use ffmpeg/avconv if False, otherwise
345 use downloader suggested by extractor if None.
fe7e0c98 346
8222d8de 347 The following parameters are not used by YoutubeDL itself, they are used by
cefecac1 348 the downloader (see youtube_dlc/downloader/common.py):
8222d8de 349 nopart, updatetime, buffersize, ratelimit, min_filesize, max_filesize, test,
881e6a1f 350 noresizebuffer, retries, continuedl, noprogress, consoletitle,
b54d4a5c
S
351 xattr_set_filesize, external_downloader_args, hls_use_mpegts,
352 http_chunk_size.
76b1bd67
JMF
353
354 The following options are used by the post processors:
d4a24f40 355 prefer_ffmpeg: If False, use avconv instead of ffmpeg if both are available,
e4172ac9 356 otherwise prefer ffmpeg. (avconv support is deprecated)
c0b7d117
S
357 ffmpeg_location: Location of the ffmpeg/avconv binary; either the path
358 to the binary or its containing directory.
43820c03 359 postprocessor_args: A dictionary of postprocessor/executable keys (in lower case)
360 and a list of additional command-line arguments for the
361 postprocessor/executable. The dict can also have "PP+EXE" keys
362 which are used when the given exe is used by the given PP.
363 Use 'default' as the name for arguments to passed to all PP
3600fd59
S
364 The following options are used by the Youtube extractor:
365 youtube_include_dash_manifest: If True (default), DASH manifests and related
366 data will be downloaded and processed by extractor.
367 You can reduce network I/O by disabling it if you don't
368 care about DASH.
8222d8de
JMF
369 """
370
c9969434
S
371 _NUMERIC_FIELDS = set((
372 'width', 'height', 'tbr', 'abr', 'asr', 'vbr', 'fps', 'filesize', 'filesize_approx',
373 'timestamp', 'upload_year', 'upload_month', 'upload_day',
374 'duration', 'view_count', 'like_count', 'dislike_count', 'repost_count',
375 'average_rating', 'comment_count', 'age_limit',
376 'start_time', 'end_time',
377 'chapter_number', 'season_number', 'episode_number',
378 'track_number', 'disc_number', 'release_year',
379 'playlist_index',
380 ))
381
8222d8de
JMF
382 params = None
383 _ies = []
5bfa4862 384 _pps = {'beforedl': [], 'aftermove': [], 'normal': []}
0202b52a 385 __prepare_filename_warned = False
8222d8de
JMF
386 _download_retcode = None
387 _num_downloads = None
30a074c2 388 _playlist_level = 0
389 _playlist_urls = set()
8222d8de
JMF
390 _screen_file = None
391
3511266b 392 def __init__(self, params=None, auto_init=True):
8222d8de 393 """Create a FileDownloader object with the given options."""
e9f9a10f
JMF
394 if params is None:
395 params = {}
8222d8de 396 self._ies = []
56c73665 397 self._ies_instances = {}
5bfa4862 398 self._pps = {'beforedl': [], 'aftermove': [], 'normal': []}
0202b52a 399 self.__prepare_filename_warned = False
ab8e5e51 400 self._post_hooks = []
933605d7 401 self._progress_hooks = []
8222d8de
JMF
402 self._download_retcode = 0
403 self._num_downloads = 0
404 self._screen_file = [sys.stdout, sys.stderr][params.get('logtostderr', False)]
0783b09b 405 self._err_file = sys.stderr
4abf617b
S
406 self.params = {
407 # Default parameters
408 'nocheckcertificate': False,
409 }
410 self.params.update(params)
a0e07d31 411 self.cache = Cache(self)
a45e8619 412 self.archive = set()
ecdec191
JB
413
414 """Preload the archive, if any is specified"""
415 def preload_download_archive(self):
416 fn = self.params.get('download_archive')
417 if fn is None:
418 return False
419 try:
420 with locked_file(fn, 'r', encoding='utf-8') as archive_file:
421 for line in archive_file:
a45e8619 422 self.archive.add(line.strip())
ecdec191
JB
423 except IOError as ioe:
424 if ioe.errno != errno.ENOENT:
425 raise
1d74d8d9 426 return False
ecdec191 427 return True
34308b30 428
be5df5ee
S
429 def check_deprecated(param, option, suggestion):
430 if self.params.get(param) is not None:
431 self.report_warning(
432 '%s is deprecated. Use %s instead.' % (option, suggestion))
433 return True
434 return False
435
1de7ea76
JB
436 if self.params.get('verbose'):
437 self.to_stdout('[debug] Loading archive file %r' % self.params.get('download_archive'))
438
ecdec191
JB
439 preload_download_archive(self)
440
be5df5ee 441 if check_deprecated('cn_verification_proxy', '--cn-verification-proxy', '--geo-verification-proxy'):
38cce791
YCH
442 if self.params.get('geo_verification_proxy') is None:
443 self.params['geo_verification_proxy'] = self.params['cn_verification_proxy']
444
6b591b29 445 if self.params.get('final_ext'):
446 if self.params.get('merge_output_format'):
447 self.report_warning('--merge-output-format will be ignored since --remux-video or --recode-video is given')
448 self.params['merge_output_format'] = self.params['final_ext']
449
b9d973be 450 if 'overwrites' in self.params and self.params['overwrites'] is None:
451 del self.params['overwrites']
452
be5df5ee
S
453 check_deprecated('autonumber_size', '--autonumber-size', 'output template with %(autonumber)0Nd, where N in the number of digits')
454 check_deprecated('autonumber', '--auto-number', '-o "%(autonumber)s-%(title)s.%(ext)s"')
455 check_deprecated('usetitle', '--title', '-o "%(title)s-%(id)s.%(ext)s"')
456
0783b09b 457 if params.get('bidi_workaround', False):
1c088fa8
PH
458 try:
459 import pty
460 master, slave = pty.openpty()
003c69a8 461 width = compat_get_terminal_size().columns
1c088fa8
PH
462 if width is None:
463 width_args = []
464 else:
465 width_args = ['-w', str(width)]
5d681e96 466 sp_kwargs = dict(
1c088fa8
PH
467 stdin=subprocess.PIPE,
468 stdout=slave,
469 stderr=self._err_file)
5d681e96
PH
470 try:
471 self._output_process = subprocess.Popen(
472 ['bidiv'] + width_args, **sp_kwargs
473 )
474 except OSError:
5d681e96
PH
475 self._output_process = subprocess.Popen(
476 ['fribidi', '-c', 'UTF-8'] + width_args, **sp_kwargs)
477 self._output_channel = os.fdopen(master, 'rb')
1c088fa8 478 except OSError as ose:
66e7ace1 479 if ose.errno == errno.ENOENT:
6febd1c1 480 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 $PATH.')
1c088fa8
PH
481 else:
482 raise
0783b09b 483
3089bc74
S
484 if (sys.platform != 'win32'
485 and sys.getfilesystemencoding() in ['ascii', 'ANSI_X3.4-1968']
486 and not params.get('restrictfilenames', False)):
e9137224 487 # Unicode filesystem API will throw errors (#1474, #13027)
34308b30 488 self.report_warning(
6febd1c1 489 'Assuming --restrict-filenames since file system encoding '
1b725173 490 'cannot encode all characters. '
6febd1c1 491 'Set the LC_ALL environment variable to fix this.')
4a98cdbf 492 self.params['restrictfilenames'] = True
34308b30 493
486dd09e
PH
494 if isinstance(params.get('outtmpl'), bytes):
495 self.report_warning(
496 'Parameter outtmpl is bytes, but should be a unicode string. '
497 'Put from __future__ import unicode_literals at the top of your code file or consider switching to Python 3.x.')
498
dca08720
PH
499 self._setup_opener()
500
3511266b
PH
501 if auto_init:
502 self.print_debug_header()
503 self.add_default_info_extractors()
504
4f026faf
PH
505 for pp_def_raw in self.params.get('postprocessors', []):
506 pp_class = get_postprocessor(pp_def_raw['key'])
507 pp_def = dict(pp_def_raw)
508 del pp_def['key']
5bfa4862 509 if 'when' in pp_def:
510 when = pp_def['when']
511 del pp_def['when']
512 else:
513 when = 'normal'
4f026faf 514 pp = pp_class(self, **compat_kwargs(pp_def))
5bfa4862 515 self.add_post_processor(pp, when=when)
4f026faf 516
ab8e5e51
AM
517 for ph in self.params.get('post_hooks', []):
518 self.add_post_hook(ph)
519
71b640cc
PH
520 for ph in self.params.get('progress_hooks', []):
521 self.add_progress_hook(ph)
522
51fb4995
YCH
523 register_socks_protocols()
524
7d4111ed
PH
525 def warn_if_short_id(self, argv):
526 # short YouTube ID starting with dash?
527 idxs = [
528 i for i, a in enumerate(argv)
529 if re.match(r'^-[0-9A-Za-z_-]{10}$', a)]
530 if idxs:
531 correct_argv = (
cefecac1 532 ['youtube-dlc']
3089bc74
S
533 + [a for i, a in enumerate(argv) if i not in idxs]
534 + ['--'] + [argv[i] for i in idxs]
7d4111ed
PH
535 )
536 self.report_warning(
537 'Long argument string detected. '
538 'Use -- to separate parameters and URLs, like this:\n%s\n' %
539 args_to_str(correct_argv))
540
8222d8de
JMF
541 def add_info_extractor(self, ie):
542 """Add an InfoExtractor object to the end of the list."""
543 self._ies.append(ie)
e52d7f85
JMF
544 if not isinstance(ie, type):
545 self._ies_instances[ie.ie_key()] = ie
546 ie.set_downloader(self)
8222d8de 547
56c73665
JMF
548 def get_info_extractor(self, ie_key):
549 """
550 Get an instance of an IE with name ie_key, it will try to get one from
551 the _ies list, if there's no instance it will create a new one and add
552 it to the extractor list.
553 """
554 ie = self._ies_instances.get(ie_key)
555 if ie is None:
556 ie = get_info_extractor(ie_key)()
557 self.add_info_extractor(ie)
558 return ie
559
023fa8c4
JMF
560 def add_default_info_extractors(self):
561 """
562 Add the InfoExtractors returned by gen_extractors to the end of the list
563 """
e52d7f85 564 for ie in gen_extractor_classes():
023fa8c4
JMF
565 self.add_info_extractor(ie)
566
5bfa4862 567 def add_post_processor(self, pp, when='normal'):
8222d8de 568 """Add a PostProcessor object to the end of the chain."""
5bfa4862 569 self._pps[when].append(pp)
8222d8de
JMF
570 pp.set_downloader(self)
571
ab8e5e51
AM
572 def add_post_hook(self, ph):
573 """Add the post hook"""
574 self._post_hooks.append(ph)
575
933605d7
JMF
576 def add_progress_hook(self, ph):
577 """Add the progress hook (currently only for the file downloader)"""
578 self._progress_hooks.append(ph)
8ab470f1 579
1c088fa8 580 def _bidi_workaround(self, message):
5d681e96 581 if not hasattr(self, '_output_channel'):
1c088fa8
PH
582 return message
583
5d681e96 584 assert hasattr(self, '_output_process')
11b85ce6 585 assert isinstance(message, compat_str)
6febd1c1
PH
586 line_count = message.count('\n') + 1
587 self._output_process.stdin.write((message + '\n').encode('utf-8'))
5d681e96 588 self._output_process.stdin.flush()
6febd1c1 589 res = ''.join(self._output_channel.readline().decode('utf-8')
9e1a5b84 590 for _ in range(line_count))
6febd1c1 591 return res[:-len('\n')]
1c088fa8 592
8222d8de 593 def to_screen(self, message, skip_eol=False):
0783b09b
PH
594 """Print message to stdout if not in quiet mode."""
595 return self.to_stdout(message, skip_eol, check_quiet=True)
596
734f90bb 597 def _write_string(self, s, out=None):
b58ddb32 598 write_string(s, out=out, encoding=self.params.get('encoding'))
734f90bb 599
0783b09b 600 def to_stdout(self, message, skip_eol=False, check_quiet=False):
8222d8de 601 """Print message to stdout if not in quiet mode."""
8bf9319e 602 if self.params.get('logger'):
43afe285 603 self.params['logger'].debug(message)
0783b09b 604 elif not check_quiet or not self.params.get('quiet', False):
1c088fa8 605 message = self._bidi_workaround(message)
6febd1c1 606 terminator = ['\n', ''][skip_eol]
8222d8de 607 output = message + terminator
1c088fa8 608
734f90bb 609 self._write_string(output, self._screen_file)
8222d8de
JMF
610
611 def to_stderr(self, message):
612 """Print message to stderr."""
11b85ce6 613 assert isinstance(message, compat_str)
8bf9319e 614 if self.params.get('logger'):
43afe285
IB
615 self.params['logger'].error(message)
616 else:
1c088fa8 617 message = self._bidi_workaround(message)
6febd1c1 618 output = message + '\n'
734f90bb 619 self._write_string(output, self._err_file)
8222d8de 620
1e5b9a95
PH
621 def to_console_title(self, message):
622 if not self.params.get('consoletitle', False):
623 return
4bede0d8
C
624 if compat_os_name == 'nt':
625 if ctypes.windll.kernel32.GetConsoleWindow():
626 # c_wchar_p() might not be necessary if `message` is
627 # already of type unicode()
628 ctypes.windll.kernel32.SetConsoleTitleW(ctypes.c_wchar_p(message))
1e5b9a95 629 elif 'TERM' in os.environ:
b46696bd 630 self._write_string('\033]0;%s\007' % message, self._screen_file)
1e5b9a95 631
bdde425c
PH
632 def save_console_title(self):
633 if not self.params.get('consoletitle', False):
634 return
94c3442e
S
635 if self.params.get('simulate', False):
636 return
4bede0d8 637 if compat_os_name != 'nt' and 'TERM' in os.environ:
efd6c574 638 # Save the title on stack
734f90bb 639 self._write_string('\033[22;0t', self._screen_file)
bdde425c
PH
640
641 def restore_console_title(self):
642 if not self.params.get('consoletitle', False):
643 return
94c3442e
S
644 if self.params.get('simulate', False):
645 return
4bede0d8 646 if compat_os_name != 'nt' and 'TERM' in os.environ:
efd6c574 647 # Restore the title from stack
734f90bb 648 self._write_string('\033[23;0t', self._screen_file)
bdde425c
PH
649
650 def __enter__(self):
651 self.save_console_title()
652 return self
653
654 def __exit__(self, *args):
655 self.restore_console_title()
f89197d7 656
dca08720 657 if self.params.get('cookiefile') is not None:
1bab3437 658 self.cookiejar.save(ignore_discard=True, ignore_expires=True)
bdde425c 659
8222d8de
JMF
660 def trouble(self, message=None, tb=None):
661 """Determine action to take when a download problem appears.
662
663 Depending on if the downloader has been configured to ignore
664 download errors or not, this method may throw an exception or
665 not when errors are found, after printing the message.
666
667 tb, if given, is additional traceback information.
668 """
669 if message is not None:
670 self.to_stderr(message)
671 if self.params.get('verbose'):
672 if tb is None:
673 if sys.exc_info()[0]: # if .trouble has been called from an except block
6febd1c1 674 tb = ''
8222d8de 675 if hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
6febd1c1 676 tb += ''.join(traceback.format_exception(*sys.exc_info()[1].exc_info))
c0384f22 677 tb += encode_compat_str(traceback.format_exc())
8222d8de
JMF
678 else:
679 tb_data = traceback.format_list(traceback.extract_stack())
6febd1c1 680 tb = ''.join(tb_data)
8222d8de
JMF
681 self.to_stderr(tb)
682 if not self.params.get('ignoreerrors', False):
683 if sys.exc_info()[0] and hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
684 exc_info = sys.exc_info()[1].exc_info
685 else:
686 exc_info = sys.exc_info()
687 raise DownloadError(message, exc_info)
688 self._download_retcode = 1
689
690 def report_warning(self, message):
691 '''
692 Print the message to stderr, it will be prefixed with 'WARNING:'
693 If stderr is a tty file the 'WARNING:' will be colored
694 '''
6d07ce01
JMF
695 if self.params.get('logger') is not None:
696 self.params['logger'].warning(message)
8222d8de 697 else:
ad8915b7
PH
698 if self.params.get('no_warnings'):
699 return
e9c0cdd3 700 if not self.params.get('no_color') and self._err_file.isatty() and compat_os_name != 'nt':
6d07ce01
JMF
701 _msg_header = '\033[0;33mWARNING:\033[0m'
702 else:
703 _msg_header = 'WARNING:'
704 warning_message = '%s %s' % (_msg_header, message)
705 self.to_stderr(warning_message)
8222d8de
JMF
706
707 def report_error(self, message, tb=None):
708 '''
709 Do the same as trouble, but prefixes the message with 'ERROR:', colored
710 in red if stderr is a tty file.
711 '''
e9c0cdd3 712 if not self.params.get('no_color') and self._err_file.isatty() and compat_os_name != 'nt':
6febd1c1 713 _msg_header = '\033[0;31mERROR:\033[0m'
8222d8de 714 else:
6febd1c1
PH
715 _msg_header = 'ERROR:'
716 error_message = '%s %s' % (_msg_header, message)
8222d8de
JMF
717 self.trouble(error_message, tb)
718
8222d8de
JMF
719 def report_file_already_downloaded(self, file_name):
720 """Report file has already been fully downloaded."""
721 try:
6febd1c1 722 self.to_screen('[download] %s has already been downloaded' % file_name)
ce02ed60 723 except UnicodeEncodeError:
6febd1c1 724 self.to_screen('[download] The file has already been downloaded')
8222d8de 725
0c3d0f51 726 def report_file_delete(self, file_name):
727 """Report that existing file will be deleted."""
728 try:
729 self.to_screen('Deleting already existent file %s' % file_name)
730 except UnicodeEncodeError:
731 self.to_screen('Deleting already existent file')
732
0202b52a 733 def prepare_filename(self, info_dict, warn=False):
8222d8de
JMF
734 """Generate the output filename."""
735 try:
736 template_dict = dict(info_dict)
737
738 template_dict['epoch'] = int(time.time())
739 autonumber_size = self.params.get('autonumber_size')
740 if autonumber_size is None:
741 autonumber_size = 5
89db639d 742 template_dict['autonumber'] = self.params.get('autonumber_start', 1) - 1 + self._num_downloads
17b75c0d
PH
743 if template_dict.get('resolution') is None:
744 if template_dict.get('width') and template_dict.get('height'):
745 template_dict['resolution'] = '%dx%d' % (template_dict['width'], template_dict['height'])
746 elif template_dict.get('height'):
805ef3c6 747 template_dict['resolution'] = '%sp' % template_dict['height']
17b75c0d 748 elif template_dict.get('width'):
51ce9117 749 template_dict['resolution'] = '%dx?' % template_dict['width']
8222d8de 750
586a91b6 751 sanitize = lambda k, v: sanitize_filename(
45598aab 752 compat_str(v),
1bb5c511 753 restricted=self.params.get('restrictfilenames'),
40df485f 754 is_id=(k == 'id' or k.endswith('_id')))
d0d9ade4 755 template_dict = dict((k, v if isinstance(v, compat_numeric_types) else sanitize(k, v))
45598aab 756 for k, v in template_dict.items()
f0e14fdd 757 if v is not None and not isinstance(v, (list, tuple, dict)))
a820dc72 758 template_dict = collections.defaultdict(lambda: self.params.get('outtmpl_na_placeholder', 'NA'), template_dict)
8222d8de 759
b3613d36 760 outtmpl = self.params.get('outtmpl', DEFAULT_OUTTMPL)
d0d9ade4 761
89db639d
S
762 # For fields playlist_index and autonumber convert all occurrences
763 # of %(field)s to %(field)0Nd for backward compatibility
764 field_size_compat_map = {
765 'playlist_index': len(str(template_dict['n_entries'])),
766 'autonumber': autonumber_size,
767 }
768 FIELD_SIZE_COMPAT_RE = r'(?<!%)%\((?P<field>autonumber|playlist_index)\)s'
769 mobj = re.search(FIELD_SIZE_COMPAT_RE, outtmpl)
770 if mobj:
771 outtmpl = re.sub(
772 FIELD_SIZE_COMPAT_RE,
773 r'%%(\1)0%dd' % field_size_compat_map[mobj.group('field')],
774 outtmpl)
775
d0d9ade4
S
776 # Missing numeric fields used together with integer presentation types
777 # in format specification will break the argument substitution since
a820dc72
RA
778 # string NA placeholder is returned for missing fields. We will patch
779 # output template for missing fields to meet string presentation type.
c9969434 780 for numeric_field in self._NUMERIC_FIELDS:
d0d9ade4
S
781 if numeric_field not in template_dict:
782 # As of [1] format syntax is:
783 # %[mapping_key][conversion_flags][minimum_width][.precision][length_modifier]type
784 # 1. https://docs.python.org/2/library/stdtypes.html#string-formatting
785 FORMAT_RE = r'''(?x)
786 (?<!%)
787 %
788 \({0}\) # mapping key
789 (?:[#0\-+ ]+)? # conversion flags (optional)
790 (?:\d+)? # minimum field width (optional)
791 (?:\.\d+)? # precision (optional)
792 [hlL]? # length modifier (optional)
793 [diouxXeEfFgGcrs%] # conversion type
794 '''
795 outtmpl = re.sub(
796 FORMAT_RE.format(numeric_field),
797 r'%({0})s'.format(numeric_field), outtmpl)
798
15da37c7
S
799 # expand_path translates '%%' into '%' and '$$' into '$'
800 # correspondingly that is not what we want since we need to keep
801 # '%%' intact for template dict substitution step. Working around
802 # with boundary-alike separator hack.
961ea474 803 sep = ''.join([random.choice(ascii_letters) for _ in range(32)])
15da37c7
S
804 outtmpl = outtmpl.replace('%%', '%{0}%'.format(sep)).replace('$$', '${0}$'.format(sep))
805
806 # outtmpl should be expand_path'ed before template dict substitution
807 # because meta fields may contain env variables we don't want to
808 # be expanded. For example, for outtmpl "%(title)s.%(ext)s" and
809 # title "Hello $PATH", we don't want `$PATH` to be expanded.
810 filename = expand_path(outtmpl).replace(sep, '') % template_dict
811
bdc3fd2f
U
812 # https://github.com/blackjack4494/youtube-dlc/issues/85
813 trim_file_name = self.params.get('trim_file_name', False)
814 if trim_file_name:
815 fn_groups = filename.rsplit('.')
816 ext = fn_groups[-1]
817 sub_ext = ''
818 if len(fn_groups) > 2:
819 sub_ext = fn_groups[-2]
820 filename = '.'.join(filter(None, [fn_groups[0][:trim_file_name], sub_ext, ext]))
821
3a0d2f52
S
822 # Temporary fix for #4787
823 # 'Treat' all problem characters by passing filename through preferredencoding
824 # to workaround encoding issues with subprocess on python2 @ Windows
825 if sys.version_info < (3, 0) and sys.platform == 'win32':
826 filename = encodeFilename(filename, True).decode(preferredencoding())
0202b52a 827 filename = sanitize_path(filename)
828
829 if warn and not self.__prepare_filename_warned:
830 if not self.params.get('paths'):
831 pass
832 elif filename == '-':
833 self.report_warning('--paths is ignored when an outputting to stdout')
834 elif os.path.isabs(filename):
835 self.report_warning('--paths is ignored since an absolute path is given in output template')
836 self.__prepare_filename_warned = True
837
838 return filename
8222d8de 839 except ValueError as err:
6febd1c1 840 self.report_error('Error in output template: ' + str(err) + ' (encoding: ' + repr(preferredencoding()) + ')')
8222d8de
JMF
841 return None
842
0202b52a 843 def prepare_filepath(self, filename, dir_type=''):
844 if filename == '-':
845 return filename
846 paths = self.params.get('paths', {})
847 assert isinstance(paths, dict)
848 homepath = expand_path(paths.get('home', '').strip())
849 assert isinstance(homepath, compat_str)
850 subdir = expand_path(paths.get(dir_type, '').strip()) if dir_type else ''
851 assert isinstance(subdir, compat_str)
852 return sanitize_path(os.path.join(homepath, subdir, filename))
853
442c37b7 854 def _match_entry(self, info_dict, incomplete):
ecdec191 855 """ Returns None if the file should be downloaded """
8222d8de 856
8b0d7497 857 def check_filter():
858 video_title = info_dict.get('title', info_dict.get('id', 'video'))
859 if 'title' in info_dict:
860 # This can happen when we're just evaluating the playlist
861 title = info_dict['title']
862 matchtitle = self.params.get('matchtitle', False)
863 if matchtitle:
864 if not re.search(matchtitle, title, re.IGNORECASE):
865 return '"' + title + '" title did not match pattern "' + matchtitle + '"'
866 rejecttitle = self.params.get('rejecttitle', False)
867 if rejecttitle:
868 if re.search(rejecttitle, title, re.IGNORECASE):
869 return '"' + title + '" title matched reject pattern "' + rejecttitle + '"'
870 date = info_dict.get('upload_date')
871 if date is not None:
872 dateRange = self.params.get('daterange', DateRange())
873 if date not in dateRange:
874 return '%s upload date is not in range %s' % (date_from_str(date).isoformat(), dateRange)
875 view_count = info_dict.get('view_count')
876 if view_count is not None:
877 min_views = self.params.get('min_views')
878 if min_views is not None and view_count < min_views:
879 return 'Skipping %s, because it has not reached minimum view count (%d/%d)' % (video_title, view_count, min_views)
880 max_views = self.params.get('max_views')
881 if max_views is not None and view_count > max_views:
882 return 'Skipping %s, because it has exceeded the maximum view count (%d/%d)' % (video_title, view_count, max_views)
883 if age_restricted(info_dict.get('age_limit'), self.params.get('age_limit')):
884 return 'Skipping "%s" because it is age restricted' % video_title
885 if self.in_download_archive(info_dict):
886 return '%s has already been recorded in archive' % video_title
887
888 if not incomplete:
889 match_filter = self.params.get('match_filter')
890 if match_filter is not None:
891 ret = match_filter(info_dict)
892 if ret is not None:
893 return ret
894 return None
895
896 reason = check_filter()
897 if reason is not None:
898 self.to_screen('[download] ' + reason)
d83cb531 899 if reason.endswith('has already been recorded in the archive') and self.params.get('break_on_existing', False):
8b0d7497 900 raise ExistingVideoReached()
d83cb531 901 elif self.params.get('break_on_reject', False):
8b0d7497 902 raise RejectedVideoReached()
903 return reason
fe7e0c98 904
b6c45014
JMF
905 @staticmethod
906 def add_extra_info(info_dict, extra_info):
907 '''Set the keys from extra_info in info dict if they are missing'''
908 for key, value in extra_info.items():
909 info_dict.setdefault(key, value)
910
0704d222 911 def extract_info(self, url, download=True, ie_key=None, info_dict=None, extra_info={},
61aa5ba3 912 process=True, force_generic_extractor=False):
8222d8de
JMF
913 '''
914 Returns a list with a dictionary for each video we find.
915 If 'download', also downloads the videos.
916 extra_info is a dict containing the extra values to add to each result
613b2d9d 917 '''
fe7e0c98 918
61aa5ba3 919 if not ie_key and force_generic_extractor:
d22dec74
S
920 ie_key = 'Generic'
921
8222d8de 922 if ie_key:
56c73665 923 ies = [self.get_info_extractor(ie_key)]
8222d8de
JMF
924 else:
925 ies = self._ies
926
927 for ie in ies:
928 if not ie.suitable(url):
929 continue
930
9a68de12 931 ie_key = ie.ie_key()
932 ie = self.get_info_extractor(ie_key)
8222d8de 933 if not ie.working():
6febd1c1
PH
934 self.report_warning('The program functionality for this site has been marked as broken, '
935 'and will probably not work.')
8222d8de
JMF
936
937 try:
d0757229 938 temp_id = str_or_none(
63be1aab 939 ie.extract_id(url) if callable(getattr(ie, 'extract_id', None))
940 else ie._match_id(url))
a0566bbf 941 except (AssertionError, IndexError, AttributeError):
942 temp_id = None
943 if temp_id is not None and self.in_download_archive({'id': temp_id, 'ie_key': ie_key}):
944 self.to_screen("[%s] %s: has already been recorded in archive" % (
945 ie_key, temp_id))
946 break
a0566bbf 947 return self.__extract_info(url, ie, download, extra_info, process, info_dict)
a0566bbf 948 else:
949 self.report_error('no suitable InfoExtractor for URL %s' % url)
950
951 def __handle_extraction_exceptions(func):
952 def wrapper(self, *args, **kwargs):
953 try:
954 return func(self, *args, **kwargs)
773f291d
S
955 except GeoRestrictedError as e:
956 msg = e.msg
957 if e.countries:
958 msg += '\nThis video is available in %s.' % ', '.join(
959 map(ISO3166Utils.short2full, e.countries))
960 msg += '\nYou might want to use a VPN or a proxy server (with --proxy) to workaround.'
961 self.report_error(msg)
fb043a6e 962 except ExtractorError as e: # An error we somewhat expected
2c74e6fa 963 self.report_error(compat_str(e), e.format_traceback())
8b0d7497 964 except (MaxDownloadsReached, ExistingVideoReached, RejectedVideoReached):
d3e5bbf4 965 raise
8222d8de
JMF
966 except Exception as e:
967 if self.params.get('ignoreerrors', False):
9b9c5355 968 self.report_error(error_to_compat_str(e), tb=encode_compat_str(traceback.format_exc()))
8222d8de
JMF
969 else:
970 raise
a0566bbf 971 return wrapper
972
973 @__handle_extraction_exceptions
974 def __extract_info(self, url, ie, download, extra_info, process, info_dict):
975 ie_result = ie.extract(url)
976 if ie_result is None: # Finished already (backwards compatibility; listformats and friends should be moved here)
977 return
978 if isinstance(ie_result, list):
979 # Backwards compatibility: old IE result format
980 ie_result = {
981 '_type': 'compat_list',
982 'entries': ie_result,
983 }
984 if info_dict:
985 if info_dict.get('id'):
986 ie_result['id'] = info_dict['id']
987 if info_dict.get('title'):
988 ie_result['title'] = info_dict['title']
989 self.add_default_extra_info(ie_result, ie, url)
990 if process:
991 return self.process_ie_result(ie_result, download, extra_info)
8222d8de 992 else:
a0566bbf 993 return ie_result
fe7e0c98 994
ea38e55f
PH
995 def add_default_extra_info(self, ie_result, ie, url):
996 self.add_extra_info(ie_result, {
997 'extractor': ie.IE_NAME,
998 'webpage_url': url,
dbbbe555 999 'duration_string': (
1000 formatSeconds(ie_result['duration'], '-')
1001 if ie_result.get('duration', None) is not None
1002 else None),
ea38e55f
PH
1003 'webpage_url_basename': url_basename(url),
1004 'extractor_key': ie.ie_key(),
1005 })
1006
8222d8de
JMF
1007 def process_ie_result(self, ie_result, download=True, extra_info={}):
1008 """
1009 Take the result of the ie(may be modified) and resolve all unresolved
1010 references (URLs, playlist items).
1011
1012 It will also download the videos if 'download'.
1013 Returns the resolved ie_result.
1014 """
e8ee972c
PH
1015 result_type = ie_result.get('_type', 'video')
1016
057a5206 1017 if result_type in ('url', 'url_transparent'):
134c6ea8 1018 ie_result['url'] = sanitize_url(ie_result['url'])
057a5206 1019 extract_flat = self.params.get('extract_flat', False)
3089bc74
S
1020 if ((extract_flat == 'in_playlist' and 'playlist' in extra_info)
1021 or extract_flat is True):
d06daf23 1022 self.__forced_printings(
0202b52a 1023 ie_result,
1024 self.prepare_filepath(self.prepare_filename(ie_result)),
d06daf23 1025 incomplete=True)
e8ee972c
PH
1026 return ie_result
1027
8222d8de 1028 if result_type == 'video':
b6c45014 1029 self.add_extra_info(ie_result, extra_info)
feee2ecf 1030 return self.process_video_result(ie_result, download=download)
8222d8de
JMF
1031 elif result_type == 'url':
1032 # We have to add extra_info to the results because it may be
1033 # contained in a playlist
1034 return self.extract_info(ie_result['url'],
0704d222 1035 download, info_dict=ie_result,
8222d8de
JMF
1036 ie_key=ie_result.get('ie_key'),
1037 extra_info=extra_info)
7fc3fa05
PH
1038 elif result_type == 'url_transparent':
1039 # Use the information from the embedding page
1040 info = self.extract_info(
1041 ie_result['url'], ie_key=ie_result.get('ie_key'),
1042 extra_info=extra_info, download=False, process=False)
1043
1640eb09
S
1044 # extract_info may return None when ignoreerrors is enabled and
1045 # extraction failed with an error, don't crash and return early
1046 # in this case
1047 if not info:
1048 return info
1049
412c617d
PH
1050 force_properties = dict(
1051 (k, v) for k, v in ie_result.items() if v is not None)
0396806f 1052 for f in ('_type', 'url', 'id', 'extractor', 'extractor_key', 'ie_key'):
412c617d
PH
1053 if f in force_properties:
1054 del force_properties[f]
1055 new_result = info.copy()
1056 new_result.update(force_properties)
7fc3fa05 1057
0563f7ac
S
1058 # Extracted info may not be a video result (i.e.
1059 # info.get('_type', 'video') != video) but rather an url or
1060 # url_transparent. In such cases outer metadata (from ie_result)
1061 # should be propagated to inner one (info). For this to happen
1062 # _type of info should be overridden with url_transparent. This
067aa17e 1063 # fixes issue from https://github.com/ytdl-org/youtube-dl/pull/11163.
0563f7ac
S
1064 if new_result.get('_type') == 'url':
1065 new_result['_type'] = 'url_transparent'
7fc3fa05
PH
1066
1067 return self.process_ie_result(
1068 new_result, download=download, extra_info=extra_info)
40fcba5e 1069 elif result_type in ('playlist', 'multi_video'):
30a074c2 1070 # Protect from infinite recursion due to recursively nested playlists
1071 # (see https://github.com/ytdl-org/youtube-dl/issues/27833)
1072 webpage_url = ie_result['webpage_url']
1073 if webpage_url in self._playlist_urls:
7e85e872 1074 self.to_screen(
30a074c2 1075 '[download] Skipping already downloaded playlist: %s'
1076 % ie_result.get('title') or ie_result.get('id'))
1077 return
7e85e872 1078
30a074c2 1079 self._playlist_level += 1
1080 self._playlist_urls.add(webpage_url)
1081 try:
1082 return self.__process_playlist(ie_result, download)
1083 finally:
1084 self._playlist_level -= 1
1085 if not self._playlist_level:
1086 self._playlist_urls.clear()
8222d8de 1087 elif result_type == 'compat_list':
c9bf4114
PH
1088 self.report_warning(
1089 'Extractor %s returned a compat_list result. '
1090 'It needs to be updated.' % ie_result.get('extractor'))
5f6a1245 1091
8222d8de 1092 def _fixup(r):
9e1a5b84
JW
1093 self.add_extra_info(
1094 r,
9103bbc5
JMF
1095 {
1096 'extractor': ie_result['extractor'],
1097 'webpage_url': ie_result['webpage_url'],
29eb5174 1098 'webpage_url_basename': url_basename(ie_result['webpage_url']),
be97abc2 1099 'extractor_key': ie_result['extractor_key'],
9e1a5b84
JW
1100 }
1101 )
8222d8de
JMF
1102 return r
1103 ie_result['entries'] = [
b6c45014 1104 self.process_ie_result(_fixup(r), download, extra_info)
8222d8de
JMF
1105 for r in ie_result['entries']
1106 ]
1107 return ie_result
1108 else:
1109 raise Exception('Invalid result type: %s' % result_type)
1110
30a074c2 1111 def __process_playlist(self, ie_result, download):
1112 # We process each entry in the playlist
1113 playlist = ie_result.get('title') or ie_result.get('id')
1114 self.to_screen('[download] Downloading playlist: %s' % playlist)
1115
cac96421 1116 if self.params.get('allow_playlist_files', True):
1117 ie_copy = {
1118 'playlist': playlist,
1119 'playlist_id': ie_result.get('id'),
1120 'playlist_title': ie_result.get('title'),
1121 'playlist_uploader': ie_result.get('uploader'),
1122 'playlist_uploader_id': ie_result.get('uploader_id'),
1123 'playlist_index': 0
1124 }
1125 ie_copy.update(dict(ie_result))
02fd60d3 1126
cac96421 1127 def ensure_dir_exists(path):
1128 return make_dir(path, self.report_error)
1129
1130 if self.params.get('writeinfojson', False):
1131 infofn = replace_extension(
1132 self.prepare_filepath(self.prepare_filename(ie_copy), 'infojson'),
1133 'info.json', ie_result.get('ext'))
1134 if not ensure_dir_exists(encodeFilename(infofn)):
02fd60d3 1135 return
b9d973be 1136 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(infofn)):
cac96421 1137 self.to_screen('[info] Playlist metadata is already present')
1138 else:
1139 self.to_screen('[info] Writing playlist metadata as JSON to: ' + infofn)
1140 playlist_info = dict(ie_result)
1141 playlist_info.pop('entries')
1142 try:
1143 write_json_file(self.filter_requested_info(playlist_info), infofn)
1144 except (OSError, IOError):
1145 self.report_error('Cannot write playlist metadata to JSON file ' + infofn)
1146
1147 if self.params.get('writedescription', False):
1148 descfn = replace_extension(
1149 self.prepare_filepath(self.prepare_filename(ie_copy), 'description'),
1150 'description', ie_result.get('ext'))
1151 if not ensure_dir_exists(encodeFilename(descfn)):
1152 return
1153 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(descfn)):
1154 self.to_screen('[info] Playlist description is already present')
1155 elif ie_result.get('description') is None:
1156 self.report_warning('There\'s no playlist description to write.')
1157 else:
1158 try:
1159 self.to_screen('[info] Writing playlist description to: ' + descfn)
1160 with io.open(encodeFilename(descfn), 'w', encoding='utf-8') as descfile:
1161 descfile.write(ie_result['description'])
1162 except (OSError, IOError):
1163 self.report_error('Cannot write playlist description file ' + descfn)
1164 return
02fd60d3 1165
30a074c2 1166 playlist_results = []
1167
1168 playliststart = self.params.get('playliststart', 1) - 1
1169 playlistend = self.params.get('playlistend')
1170 # For backwards compatibility, interpret -1 as whole list
1171 if playlistend == -1:
1172 playlistend = None
1173
1174 playlistitems_str = self.params.get('playlist_items')
1175 playlistitems = None
1176 if playlistitems_str is not None:
1177 def iter_playlistitems(format):
1178 for string_segment in format.split(','):
1179 if '-' in string_segment:
1180 start, end = string_segment.split('-')
1181 for item in range(int(start), int(end) + 1):
1182 yield int(item)
1183 else:
1184 yield int(string_segment)
1185 playlistitems = orderedSet(iter_playlistitems(playlistitems_str))
1186
1187 ie_entries = ie_result['entries']
1188
1189 def make_playlistitems_entries(list_ie_entries):
1190 num_entries = len(list_ie_entries)
1191 return [
1192 list_ie_entries[i - 1] for i in playlistitems
1193 if -num_entries <= i - 1 < num_entries]
1194
1195 def report_download(num_entries):
1196 self.to_screen(
1197 '[%s] playlist %s: Downloading %d videos' %
1198 (ie_result['extractor'], playlist, num_entries))
1199
1200 if isinstance(ie_entries, list):
1201 n_all_entries = len(ie_entries)
1202 if playlistitems:
1203 entries = make_playlistitems_entries(ie_entries)
1204 else:
1205 entries = ie_entries[playliststart:playlistend]
1206 n_entries = len(entries)
1207 self.to_screen(
1208 '[%s] playlist %s: Collected %d video ids (downloading %d of them)' %
1209 (ie_result['extractor'], playlist, n_all_entries, n_entries))
1210 elif isinstance(ie_entries, PagedList):
1211 if playlistitems:
1212 entries = []
1213 for item in playlistitems:
1214 entries.extend(ie_entries.getslice(
1215 item - 1, item
1216 ))
1217 else:
1218 entries = ie_entries.getslice(
1219 playliststart, playlistend)
1220 n_entries = len(entries)
1221 report_download(n_entries)
1222 else: # iterable
1223 if playlistitems:
1224 entries = make_playlistitems_entries(list(itertools.islice(
1225 ie_entries, 0, max(playlistitems))))
1226 else:
1227 entries = list(itertools.islice(
1228 ie_entries, playliststart, playlistend))
1229 n_entries = len(entries)
1230 report_download(n_entries)
1231
1232 if self.params.get('playlistreverse', False):
1233 entries = entries[::-1]
1234
1235 if self.params.get('playlistrandom', False):
1236 random.shuffle(entries)
1237
1238 x_forwarded_for = ie_result.get('__x_forwarded_for_ip')
1239
1240 for i, entry in enumerate(entries, 1):
1241 self.to_screen('[download] Downloading video %s of %s' % (i, n_entries))
1242 # This __x_forwarded_for_ip thing is a bit ugly but requires
1243 # minimal changes
1244 if x_forwarded_for:
1245 entry['__x_forwarded_for_ip'] = x_forwarded_for
1246 extra = {
1247 'n_entries': n_entries,
1248 'playlist': playlist,
1249 'playlist_id': ie_result.get('id'),
1250 'playlist_title': ie_result.get('title'),
1251 'playlist_uploader': ie_result.get('uploader'),
1252 'playlist_uploader_id': ie_result.get('uploader_id'),
1253 'playlist_index': playlistitems[i - 1] if playlistitems else i + playliststart,
1254 'extractor': ie_result['extractor'],
1255 'webpage_url': ie_result['webpage_url'],
1256 'webpage_url_basename': url_basename(ie_result['webpage_url']),
1257 'extractor_key': ie_result['extractor_key'],
1258 }
1259
1260 if self._match_entry(entry, incomplete=True) is not None:
1261 continue
1262
1263 entry_result = self.__process_iterable_entry(entry, download, extra)
1264 # TODO: skip failed (empty) entries?
1265 playlist_results.append(entry_result)
1266 ie_result['entries'] = playlist_results
1267 self.to_screen('[download] Finished downloading playlist: %s' % playlist)
1268 return ie_result
1269
a0566bbf 1270 @__handle_extraction_exceptions
1271 def __process_iterable_entry(self, entry, download, extra_info):
1272 return self.process_ie_result(
1273 entry, download=download, extra_info=extra_info)
1274
67134eab
JMF
1275 def _build_format_filter(self, filter_spec):
1276 " Returns a function to filter the formats according to the filter_spec "
083c9df9
PH
1277
1278 OPERATORS = {
1279 '<': operator.lt,
1280 '<=': operator.le,
1281 '>': operator.gt,
1282 '>=': operator.ge,
1283 '=': operator.eq,
1284 '!=': operator.ne,
1285 }
67134eab 1286 operator_rex = re.compile(r'''(?x)\s*
a03a3c80 1287 (?P<key>width|height|tbr|abr|vbr|asr|filesize|filesize_approx|fps)
083c9df9
PH
1288 \s*(?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
1289 (?P<value>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?)
67134eab 1290 $
083c9df9 1291 ''' % '|'.join(map(re.escape, OPERATORS.keys())))
67134eab 1292 m = operator_rex.search(filter_spec)
9ddb6925
S
1293 if m:
1294 try:
1295 comparison_value = int(m.group('value'))
1296 except ValueError:
1297 comparison_value = parse_filesize(m.group('value'))
1298 if comparison_value is None:
1299 comparison_value = parse_filesize(m.group('value') + 'B')
1300 if comparison_value is None:
1301 raise ValueError(
1302 'Invalid value %r in format specification %r' % (
67134eab 1303 m.group('value'), filter_spec))
9ddb6925
S
1304 op = OPERATORS[m.group('op')]
1305
083c9df9 1306 if not m:
9ddb6925
S
1307 STR_OPERATORS = {
1308 '=': operator.eq,
10d33b34
YCH
1309 '^=': lambda attr, value: attr.startswith(value),
1310 '$=': lambda attr, value: attr.endswith(value),
1311 '*=': lambda attr, value: value in attr,
9ddb6925 1312 }
67134eab 1313 str_operator_rex = re.compile(r'''(?x)
f96bff99 1314 \s*(?P<key>[a-zA-Z0-9._-]+)
2cc779f4 1315 \s*(?P<negation>!\s*)?(?P<op>%s)(?P<none_inclusive>\s*\?)?
b0df5223 1316 \s*(?P<value>[a-zA-Z0-9._-]+)
67134eab 1317 \s*$
9ddb6925 1318 ''' % '|'.join(map(re.escape, STR_OPERATORS.keys())))
67134eab 1319 m = str_operator_rex.search(filter_spec)
9ddb6925
S
1320 if m:
1321 comparison_value = m.group('value')
2cc779f4
S
1322 str_op = STR_OPERATORS[m.group('op')]
1323 if m.group('negation'):
e118a879 1324 op = lambda attr, value: not str_op(attr, value)
2cc779f4
S
1325 else:
1326 op = str_op
083c9df9 1327
9ddb6925 1328 if not m:
67134eab 1329 raise ValueError('Invalid filter specification %r' % filter_spec)
083c9df9
PH
1330
1331 def _filter(f):
1332 actual_value = f.get(m.group('key'))
1333 if actual_value is None:
1334 return m.group('none_inclusive')
1335 return op(actual_value, comparison_value)
67134eab
JMF
1336 return _filter
1337
0017d9ad 1338 def _default_format_spec(self, info_dict, download=True):
0017d9ad 1339
af0f7428
S
1340 def can_merge():
1341 merger = FFmpegMergerPP(self)
1342 return merger.available and merger.can_merge()
1343
91ebc640 1344 prefer_best = (
1345 not self.params.get('simulate', False)
1346 and download
1347 and (
1348 not can_merge()
19807826 1349 or info_dict.get('is_live', False)
91ebc640 1350 or self.params.get('outtmpl', DEFAULT_OUTTMPL) == '-'))
1351
1352 return (
1353 'best/bestvideo+bestaudio'
1354 if prefer_best
1355 else 'bestvideo*+bestaudio/best'
19807826 1356 if not self.params.get('allow_multiple_audio_streams', False)
91ebc640 1357 else 'bestvideo+bestaudio/best')
0017d9ad 1358
67134eab
JMF
1359 def build_format_selector(self, format_spec):
1360 def syntax_error(note, start):
1361 message = (
1362 'Invalid format specification: '
1363 '{0}\n\t{1}\n\t{2}^'.format(note, format_spec, ' ' * start[1]))
1364 return SyntaxError(message)
1365
1366 PICKFIRST = 'PICKFIRST'
1367 MERGE = 'MERGE'
1368 SINGLE = 'SINGLE'
0130afb7 1369 GROUP = 'GROUP'
67134eab
JMF
1370 FormatSelector = collections.namedtuple('FormatSelector', ['type', 'selector', 'filters'])
1371
91ebc640 1372 allow_multiple_streams = {'audio': self.params.get('allow_multiple_audio_streams', False),
1373 'video': self.params.get('allow_multiple_video_streams', False)}
909d24dd 1374
67134eab
JMF
1375 def _parse_filter(tokens):
1376 filter_parts = []
1377 for type, string, start, _, _ in tokens:
1378 if type == tokenize.OP and string == ']':
1379 return ''.join(filter_parts)
1380 else:
1381 filter_parts.append(string)
1382
232541df 1383 def _remove_unused_ops(tokens):
17cc1534 1384 # Remove operators that we don't use and join them with the surrounding strings
232541df
JMF
1385 # for example: 'mp4' '-' 'baseline' '-' '16x9' is converted to 'mp4-baseline-16x9'
1386 ALLOWED_OPS = ('/', '+', ',', '(', ')')
1387 last_string, last_start, last_end, last_line = None, None, None, None
1388 for type, string, start, end, line in tokens:
1389 if type == tokenize.OP and string == '[':
1390 if last_string:
1391 yield tokenize.NAME, last_string, last_start, last_end, last_line
1392 last_string = None
1393 yield type, string, start, end, line
1394 # everything inside brackets will be handled by _parse_filter
1395 for type, string, start, end, line in tokens:
1396 yield type, string, start, end, line
1397 if type == tokenize.OP and string == ']':
1398 break
1399 elif type == tokenize.OP and string in ALLOWED_OPS:
1400 if last_string:
1401 yield tokenize.NAME, last_string, last_start, last_end, last_line
1402 last_string = None
1403 yield type, string, start, end, line
1404 elif type in [tokenize.NAME, tokenize.NUMBER, tokenize.OP]:
1405 if not last_string:
1406 last_string = string
1407 last_start = start
1408 last_end = end
1409 else:
1410 last_string += string
1411 if last_string:
1412 yield tokenize.NAME, last_string, last_start, last_end, last_line
1413
cf2ac6df 1414 def _parse_format_selection(tokens, inside_merge=False, inside_choice=False, inside_group=False):
67134eab
JMF
1415 selectors = []
1416 current_selector = None
1417 for type, string, start, _, _ in tokens:
1418 # ENCODING is only defined in python 3.x
1419 if type == getattr(tokenize, 'ENCODING', None):
1420 continue
1421 elif type in [tokenize.NAME, tokenize.NUMBER]:
1422 current_selector = FormatSelector(SINGLE, string, [])
1423 elif type == tokenize.OP:
cf2ac6df
JMF
1424 if string == ')':
1425 if not inside_group:
1426 # ')' will be handled by the parentheses group
1427 tokens.restore_last_token()
67134eab 1428 break
cf2ac6df 1429 elif inside_merge and string in ['/', ',']:
0130afb7
JMF
1430 tokens.restore_last_token()
1431 break
cf2ac6df
JMF
1432 elif inside_choice and string == ',':
1433 tokens.restore_last_token()
1434 break
1435 elif string == ',':
0a31a350
JMF
1436 if not current_selector:
1437 raise syntax_error('"," must follow a format selector', start)
67134eab
JMF
1438 selectors.append(current_selector)
1439 current_selector = None
1440 elif string == '/':
d96d604e
JMF
1441 if not current_selector:
1442 raise syntax_error('"/" must follow a format selector', start)
67134eab 1443 first_choice = current_selector
cf2ac6df 1444 second_choice = _parse_format_selection(tokens, inside_choice=True)
f5f4a27a 1445 current_selector = FormatSelector(PICKFIRST, (first_choice, second_choice), [])
67134eab
JMF
1446 elif string == '[':
1447 if not current_selector:
1448 current_selector = FormatSelector(SINGLE, 'best', [])
1449 format_filter = _parse_filter(tokens)
1450 current_selector.filters.append(format_filter)
0130afb7
JMF
1451 elif string == '(':
1452 if current_selector:
1453 raise syntax_error('Unexpected "("', start)
cf2ac6df
JMF
1454 group = _parse_format_selection(tokens, inside_group=True)
1455 current_selector = FormatSelector(GROUP, group, [])
67134eab 1456 elif string == '+':
d03cfdce 1457 if not current_selector:
1458 raise syntax_error('Unexpected "+"', start)
1459 selector_1 = current_selector
1460 selector_2 = _parse_format_selection(tokens, inside_merge=True)
1461 if not selector_2:
1462 raise syntax_error('Expected a selector', start)
1463 current_selector = FormatSelector(MERGE, (selector_1, selector_2), [])
67134eab
JMF
1464 else:
1465 raise syntax_error('Operator not recognized: "{0}"'.format(string), start)
1466 elif type == tokenize.ENDMARKER:
1467 break
1468 if current_selector:
1469 selectors.append(current_selector)
1470 return selectors
1471
1472 def _build_selector_function(selector):
909d24dd 1473 if isinstance(selector, list): # ,
67134eab
JMF
1474 fs = [_build_selector_function(s) for s in selector]
1475
317f7ab6 1476 def selector_function(ctx):
67134eab 1477 for f in fs:
317f7ab6 1478 for format in f(ctx):
67134eab
JMF
1479 yield format
1480 return selector_function
909d24dd 1481
1482 elif selector.type == GROUP: # ()
0130afb7 1483 selector_function = _build_selector_function(selector.selector)
909d24dd 1484
1485 elif selector.type == PICKFIRST: # /
67134eab
JMF
1486 fs = [_build_selector_function(s) for s in selector.selector]
1487
317f7ab6 1488 def selector_function(ctx):
67134eab 1489 for f in fs:
317f7ab6 1490 picked_formats = list(f(ctx))
67134eab
JMF
1491 if picked_formats:
1492 return picked_formats
1493 return []
67134eab 1494
909d24dd 1495 elif selector.type == SINGLE: # atom
1496 format_spec = selector.selector if selector.selector is not None else 'best'
1497
1498 if format_spec == 'all':
1499 def selector_function(ctx):
1500 formats = list(ctx['formats'])
1501 if formats:
1502 for f in formats:
1503 yield f
1504
1505 else:
1506 format_fallback = False
1507 format_spec_obj = re.match(r'(best|worst|b|w)(video|audio|v|a)?(\*)?$', format_spec)
1508 if format_spec_obj is not None:
1509 format_idx = 0 if format_spec_obj.group(1)[0] == 'w' else -1
1510 format_type = format_spec_obj.group(2)[0] if format_spec_obj.group(2) else False
1511 not_format_type = 'v' if format_type == 'a' else 'a'
1512 format_modified = format_spec_obj.group(3) is not None
1513
1514 format_fallback = not format_type and not format_modified # for b, w
1515 filter_f = ((lambda f: f.get(format_type + 'codec') != 'none')
1516 if format_type and format_modified # bv*, ba*, wv*, wa*
1517 else (lambda f: f.get(not_format_type + 'codec') == 'none')
1518 if format_type # bv, ba, wv, wa
1519 else (lambda f: f.get('vcodec') != 'none' and f.get('acodec') != 'none')
1520 if not format_modified # b, w
1521 else None) # b*, w*
67134eab 1522 else:
909d24dd 1523 format_idx = -1
1524 filter_f = ((lambda f: f.get('ext') == format_spec)
1525 if format_spec in ['mp4', 'flv', 'webm', '3gp', 'm4a', 'mp3', 'ogg', 'aac', 'wav'] # extension
1526 else (lambda f: f.get('format_id') == format_spec)) # id
1527
1528 def selector_function(ctx):
1529 formats = list(ctx['formats'])
1530 if not formats:
1531 return
1532 matches = list(filter(filter_f, formats)) if filter_f is not None else formats
67134eab 1533 if matches:
909d24dd 1534 yield matches[format_idx]
1535 elif format_fallback == 'force' or (format_fallback and ctx['incomplete_formats']):
1536 # for extractors with incomplete formats (audio only (soundcloud)
1537 # or video only (imgur)) best/worst will fallback to
1538 # best/worst {video,audio}-only format
1539 yield formats[format_idx]
1540
1541 elif selector.type == MERGE: # +
d03cfdce 1542 def _merge(formats_pair):
1543 format_1, format_2 = formats_pair
1544
1545 formats_info = []
1546 formats_info.extend(format_1.get('requested_formats', (format_1,)))
1547 formats_info.extend(format_2.get('requested_formats', (format_2,)))
1548
909d24dd 1549 if not allow_multiple_streams['video'] or not allow_multiple_streams['audio']:
1550 get_no_more = {"video": False, "audio": False}
1551 for (i, fmt_info) in enumerate(formats_info):
1552 for aud_vid in ["audio", "video"]:
1553 if not allow_multiple_streams[aud_vid] and fmt_info.get(aud_vid[0] + 'codec') != 'none':
1554 if get_no_more[aud_vid]:
1555 formats_info.pop(i)
1556 get_no_more[aud_vid] = True
1557
1558 if len(formats_info) == 1:
1559 return formats_info[0]
1560
d03cfdce 1561 video_fmts = [fmt_info for fmt_info in formats_info if fmt_info.get('vcodec') != 'none']
1562 audio_fmts = [fmt_info for fmt_info in formats_info if fmt_info.get('acodec') != 'none']
1563
1564 the_only_video = video_fmts[0] if len(video_fmts) == 1 else None
1565 the_only_audio = audio_fmts[0] if len(audio_fmts) == 1 else None
1566
1567 output_ext = self.params.get('merge_output_format')
1568 if not output_ext:
1569 if the_only_video:
1570 output_ext = the_only_video['ext']
1571 elif the_only_audio and not video_fmts:
1572 output_ext = the_only_audio['ext']
1573 else:
1574 output_ext = 'mkv'
1575
1576 new_dict = {
67134eab 1577 'requested_formats': formats_info,
d03cfdce 1578 'format': '+'.join(fmt_info.get('format') for fmt_info in formats_info),
1579 'format_id': '+'.join(fmt_info.get('format_id') for fmt_info in formats_info),
67134eab
JMF
1580 'ext': output_ext,
1581 }
d03cfdce 1582
1583 if the_only_video:
1584 new_dict.update({
1585 'width': the_only_video.get('width'),
1586 'height': the_only_video.get('height'),
1587 'resolution': the_only_video.get('resolution'),
1588 'fps': the_only_video.get('fps'),
1589 'vcodec': the_only_video.get('vcodec'),
1590 'vbr': the_only_video.get('vbr'),
1591 'stretched_ratio': the_only_video.get('stretched_ratio'),
1592 })
1593
1594 if the_only_audio:
1595 new_dict.update({
1596 'acodec': the_only_audio.get('acodec'),
1597 'abr': the_only_audio.get('abr'),
1598 })
1599
1600 return new_dict
1601
1602 selector_1, selector_2 = map(_build_selector_function, selector.selector)
083c9df9 1603
317f7ab6
S
1604 def selector_function(ctx):
1605 for pair in itertools.product(
d03cfdce 1606 selector_1(copy.deepcopy(ctx)), selector_2(copy.deepcopy(ctx))):
67134eab 1607 yield _merge(pair)
083c9df9 1608
67134eab 1609 filters = [self._build_format_filter(f) for f in selector.filters]
083c9df9 1610
317f7ab6
S
1611 def final_selector(ctx):
1612 ctx_copy = copy.deepcopy(ctx)
67134eab 1613 for _filter in filters:
317f7ab6
S
1614 ctx_copy['formats'] = list(filter(_filter, ctx_copy['formats']))
1615 return selector_function(ctx_copy)
67134eab 1616 return final_selector
083c9df9 1617
67134eab 1618 stream = io.BytesIO(format_spec.encode('utf-8'))
0130afb7 1619 try:
232541df 1620 tokens = list(_remove_unused_ops(compat_tokenize_tokenize(stream.readline)))
0130afb7
JMF
1621 except tokenize.TokenError:
1622 raise syntax_error('Missing closing/opening brackets or parenthesis', (0, len(format_spec)))
1623
1624 class TokenIterator(object):
1625 def __init__(self, tokens):
1626 self.tokens = tokens
1627 self.counter = 0
1628
1629 def __iter__(self):
1630 return self
1631
1632 def __next__(self):
1633 if self.counter >= len(self.tokens):
1634 raise StopIteration()
1635 value = self.tokens[self.counter]
1636 self.counter += 1
1637 return value
1638
1639 next = __next__
1640
1641 def restore_last_token(self):
1642 self.counter -= 1
1643
1644 parsed_selector = _parse_format_selection(iter(TokenIterator(tokens)))
67134eab 1645 return _build_selector_function(parsed_selector)
a9c58ad9 1646
e5660ee6
JMF
1647 def _calc_headers(self, info_dict):
1648 res = std_headers.copy()
1649
1650 add_headers = info_dict.get('http_headers')
1651 if add_headers:
1652 res.update(add_headers)
1653
1654 cookies = self._calc_cookies(info_dict)
1655 if cookies:
1656 res['Cookie'] = cookies
1657
0016b84e
S
1658 if 'X-Forwarded-For' not in res:
1659 x_forwarded_for_ip = info_dict.get('__x_forwarded_for_ip')
1660 if x_forwarded_for_ip:
1661 res['X-Forwarded-For'] = x_forwarded_for_ip
1662
e5660ee6
JMF
1663 return res
1664
1665 def _calc_cookies(self, info_dict):
5c2266df 1666 pr = sanitized_Request(info_dict['url'])
e5660ee6 1667 self.cookiejar.add_cookie_header(pr)
662435f7 1668 return pr.get_header('Cookie')
e5660ee6 1669
dd82ffea
JMF
1670 def process_video_result(self, info_dict, download=True):
1671 assert info_dict.get('_type', 'video') == 'video'
1672
bec1fad2
PH
1673 if 'id' not in info_dict:
1674 raise ExtractorError('Missing "id" field in extractor result')
1675 if 'title' not in info_dict:
1676 raise ExtractorError('Missing "title" field in extractor result')
1677
c9969434
S
1678 def report_force_conversion(field, field_not, conversion):
1679 self.report_warning(
1680 '"%s" field is not %s - forcing %s conversion, there is an error in extractor'
1681 % (field, field_not, conversion))
1682
1683 def sanitize_string_field(info, string_field):
1684 field = info.get(string_field)
1685 if field is None or isinstance(field, compat_str):
1686 return
1687 report_force_conversion(string_field, 'a string', 'string')
1688 info[string_field] = compat_str(field)
1689
1690 def sanitize_numeric_fields(info):
1691 for numeric_field in self._NUMERIC_FIELDS:
1692 field = info.get(numeric_field)
1693 if field is None or isinstance(field, compat_numeric_types):
1694 continue
1695 report_force_conversion(numeric_field, 'numeric', 'int')
1696 info[numeric_field] = int_or_none(field)
1697
1698 sanitize_string_field(info_dict, 'id')
1699 sanitize_numeric_fields(info_dict)
be6217b2 1700
dd82ffea
JMF
1701 if 'playlist' not in info_dict:
1702 # It isn't part of a playlist
1703 info_dict['playlist'] = None
1704 info_dict['playlist_index'] = None
1705
d5519808 1706 thumbnails = info_dict.get('thumbnails')
cfb56d1a
PH
1707 if thumbnails is None:
1708 thumbnail = info_dict.get('thumbnail')
1709 if thumbnail:
a7a14d95 1710 info_dict['thumbnails'] = thumbnails = [{'url': thumbnail}]
d5519808 1711 if thumbnails:
be6d7229 1712 thumbnails.sort(key=lambda t: (
d37708fc
RA
1713 t.get('preference') if t.get('preference') is not None else -1,
1714 t.get('width') if t.get('width') is not None else -1,
1715 t.get('height') if t.get('height') is not None else -1,
1716 t.get('id') if t.get('id') is not None else '', t.get('url')))
f6c24009 1717 for i, t in enumerate(thumbnails):
dcf77cf1 1718 t['url'] = sanitize_url(t['url'])
9603e8a7 1719 if t.get('width') and t.get('height'):
d5519808 1720 t['resolution'] = '%dx%d' % (t['width'], t['height'])
f6c24009
PH
1721 if t.get('id') is None:
1722 t['id'] = '%d' % i
d5519808 1723
b7b72db9 1724 if self.params.get('list_thumbnails'):
1725 self.list_thumbnails(info_dict)
1726 return
1727
536a55da
S
1728 thumbnail = info_dict.get('thumbnail')
1729 if thumbnail:
1730 info_dict['thumbnail'] = sanitize_url(thumbnail)
1731 elif thumbnails:
d5519808
PH
1732 info_dict['thumbnail'] = thumbnails[-1]['url']
1733
c9ae7b95 1734 if 'display_id' not in info_dict and 'id' in info_dict:
0afef30b
PH
1735 info_dict['display_id'] = info_dict['id']
1736
955c4514 1737 if info_dict.get('upload_date') is None and info_dict.get('timestamp') is not None:
a55e36f4
S
1738 # Working around out-of-range timestamp values (e.g. negative ones on Windows,
1739 # see http://bugs.python.org/issue1646728)
1740 try:
1741 upload_date = datetime.datetime.utcfromtimestamp(info_dict['timestamp'])
1742 info_dict['upload_date'] = upload_date.strftime('%Y%m%d')
1743 except (ValueError, OverflowError, OSError):
1744 pass
9d2ecdbc 1745
33d2fc2f
S
1746 # Auto generate title fields corresponding to the *_number fields when missing
1747 # in order to always have clean titles. This is very common for TV series.
1748 for field in ('chapter', 'season', 'episode'):
1749 if info_dict.get('%s_number' % field) is not None and not info_dict.get(field):
1750 info_dict[field] = '%s %d' % (field.capitalize(), info_dict['%s_number' % field])
1751
05108a49
S
1752 for cc_kind in ('subtitles', 'automatic_captions'):
1753 cc = info_dict.get(cc_kind)
1754 if cc:
1755 for _, subtitle in cc.items():
1756 for subtitle_format in subtitle:
1757 if subtitle_format.get('url'):
1758 subtitle_format['url'] = sanitize_url(subtitle_format['url'])
1759 if subtitle_format.get('ext') is None:
1760 subtitle_format['ext'] = determine_ext(subtitle_format['url']).lower()
1761
1762 automatic_captions = info_dict.get('automatic_captions')
4bba3716 1763 subtitles = info_dict.get('subtitles')
4bba3716 1764
a504ced0 1765 if self.params.get('listsubtitles', False):
360e1ca5 1766 if 'automatic_captions' in info_dict:
05108a49
S
1767 self.list_subtitles(
1768 info_dict['id'], automatic_captions, 'automatic captions')
4bba3716 1769 self.list_subtitles(info_dict['id'], subtitles, 'subtitles')
a504ced0 1770 return
05108a49 1771
360e1ca5 1772 info_dict['requested_subtitles'] = self.process_subtitles(
05108a49 1773 info_dict['id'], subtitles, automatic_captions)
a504ced0 1774
dd82ffea
JMF
1775 # We now pick which formats have to be downloaded
1776 if info_dict.get('formats') is None:
1777 # There's only one format available
1778 formats = [info_dict]
1779 else:
1780 formats = info_dict['formats']
1781
db95dc13
PH
1782 if not formats:
1783 raise ExtractorError('No video formats found!')
1784
73af5cc8
S
1785 def is_wellformed(f):
1786 url = f.get('url')
a5ac0c47 1787 if not url:
73af5cc8
S
1788 self.report_warning(
1789 '"url" field is missing or empty - skipping format, '
1790 'there is an error in extractor')
a5ac0c47
S
1791 return False
1792 if isinstance(url, bytes):
1793 sanitize_string_field(f, 'url')
1794 return True
73af5cc8
S
1795
1796 # Filter out malformed formats for better extraction robustness
1797 formats = list(filter(is_wellformed, formats))
1798
181c7053
S
1799 formats_dict = {}
1800
dd82ffea 1801 # We check that all the formats have the format and format_id fields
db95dc13 1802 for i, format in enumerate(formats):
c9969434
S
1803 sanitize_string_field(format, 'format_id')
1804 sanitize_numeric_fields(format)
dcf77cf1 1805 format['url'] = sanitize_url(format['url'])
e74e3b63 1806 if not format.get('format_id'):
8016c922 1807 format['format_id'] = compat_str(i)
e2effb08
S
1808 else:
1809 # Sanitize format_id from characters used in format selector expression
ec85ded8 1810 format['format_id'] = re.sub(r'[\s,/+\[\]()]', '_', format['format_id'])
181c7053
S
1811 format_id = format['format_id']
1812 if format_id not in formats_dict:
1813 formats_dict[format_id] = []
1814 formats_dict[format_id].append(format)
1815
1816 # Make sure all formats have unique format_id
1817 for format_id, ambiguous_formats in formats_dict.items():
1818 if len(ambiguous_formats) > 1:
1819 for i, format in enumerate(ambiguous_formats):
1820 format['format_id'] = '%s-%d' % (format_id, i)
1821
1822 for i, format in enumerate(formats):
8c51aa65 1823 if format.get('format') is None:
6febd1c1 1824 format['format'] = '{id} - {res}{note}'.format(
8c51aa65
JMF
1825 id=format['format_id'],
1826 res=self.format_resolution(format),
6febd1c1 1827 note=' ({0})'.format(format['format_note']) if format.get('format_note') is not None else '',
8c51aa65 1828 )
c1002e96 1829 # Automatically determine file extension if missing
5b1d8575 1830 if format.get('ext') is None:
cce929ea 1831 format['ext'] = determine_ext(format['url']).lower()
b5559424
S
1832 # Automatically determine protocol if missing (useful for format
1833 # selection purposes)
6f0be937 1834 if format.get('protocol') is None:
b5559424 1835 format['protocol'] = determine_protocol(format)
e5660ee6
JMF
1836 # Add HTTP headers, so that external programs can use them from the
1837 # json output
1838 full_format_info = info_dict.copy()
1839 full_format_info.update(format)
1840 format['http_headers'] = self._calc_headers(full_format_info)
0016b84e
S
1841 # Remove private housekeeping stuff
1842 if '__x_forwarded_for_ip' in info_dict:
1843 del info_dict['__x_forwarded_for_ip']
dd82ffea 1844
4bcc7bd1 1845 # TODO Central sorting goes here
99e206d5 1846
f89197d7 1847 if formats[0] is not info_dict:
b3d9ef88
JMF
1848 # only set the 'formats' fields if the original info_dict list them
1849 # otherwise we end up with a circular reference, the first (and unique)
f89197d7 1850 # element in the 'formats' field in info_dict is info_dict itself,
dfb1b146 1851 # which can't be exported to json
b3d9ef88 1852 info_dict['formats'] = formats
cfb56d1a 1853 if self.params.get('listformats'):
bfaae0a7 1854 self.list_formats(info_dict)
1855 return
1856
de3ef3ed 1857 req_format = self.params.get('format')
a9c58ad9 1858 if req_format is None:
0017d9ad
S
1859 req_format = self._default_format_spec(info_dict, download=download)
1860 if self.params.get('verbose'):
e8be92f9 1861 self.to_screen('[debug] Default format spec: %s' % req_format)
0017d9ad 1862
5acfa126 1863 format_selector = self.build_format_selector(req_format)
317f7ab6
S
1864
1865 # While in format selection we may need to have an access to the original
1866 # format set in order to calculate some metrics or do some processing.
1867 # For now we need to be able to guess whether original formats provided
1868 # by extractor are incomplete or not (i.e. whether extractor provides only
1869 # video-only or audio-only formats) for proper formats selection for
1870 # extractors with such incomplete formats (see
067aa17e 1871 # https://github.com/ytdl-org/youtube-dl/pull/5556).
317f7ab6
S
1872 # Since formats may be filtered during format selection and may not match
1873 # the original formats the results may be incorrect. Thus original formats
1874 # or pre-calculated metrics should be passed to format selection routines
1875 # as well.
1876 # We will pass a context object containing all necessary additional data
1877 # instead of just formats.
1878 # This fixes incorrect format selection issue (see
067aa17e 1879 # https://github.com/ytdl-org/youtube-dl/issues/10083).
2e221ca3 1880 incomplete_formats = (
317f7ab6 1881 # All formats are video-only or
3089bc74 1882 all(f.get('vcodec') != 'none' and f.get('acodec') == 'none' for f in formats)
317f7ab6 1883 # all formats are audio-only
3089bc74 1884 or all(f.get('vcodec') == 'none' and f.get('acodec') != 'none' for f in formats))
317f7ab6
S
1885
1886 ctx = {
1887 'formats': formats,
1888 'incomplete_formats': incomplete_formats,
1889 }
1890
1891 formats_to_download = list(format_selector(ctx))
dd82ffea 1892 if not formats_to_download:
6febd1c1 1893 raise ExtractorError('requested format not available',
78a3a9f8 1894 expected=True)
dd82ffea
JMF
1895
1896 if download:
909d24dd 1897 self.to_screen('[info] Downloading format(s) %s' % ", ".join([f['format_id'] for f in formats_to_download]))
dd82ffea 1898 if len(formats_to_download) > 1:
6febd1c1 1899 self.to_screen('[info] %s: downloading video in %s formats' % (info_dict['id'], len(formats_to_download)))
dd82ffea
JMF
1900 for format in formats_to_download:
1901 new_info = dict(info_dict)
1902 new_info.update(format)
1903 self.process_info(new_info)
1904 # We update the info dict with the best quality format (backwards compatibility)
1905 info_dict.update(formats_to_download[-1])
1906 return info_dict
1907
98c70d6f 1908 def process_subtitles(self, video_id, normal_subtitles, automatic_captions):
a504ced0 1909 """Select the requested subtitles and their format"""
98c70d6f
JMF
1910 available_subs = {}
1911 if normal_subtitles and self.params.get('writesubtitles'):
1912 available_subs.update(normal_subtitles)
1913 if automatic_captions and self.params.get('writeautomaticsub'):
1914 for lang, cap_info in automatic_captions.items():
360e1ca5
JMF
1915 if lang not in available_subs:
1916 available_subs[lang] = cap_info
1917
4d171848
JMF
1918 if (not self.params.get('writesubtitles') and not
1919 self.params.get('writeautomaticsub') or not
1920 available_subs):
1921 return None
a504ced0
JMF
1922
1923 if self.params.get('allsubtitles', False):
1924 requested_langs = available_subs.keys()
1925 else:
1926 if self.params.get('subtitleslangs', False):
1927 requested_langs = self.params.get('subtitleslangs')
1928 elif 'en' in available_subs:
1929 requested_langs = ['en']
1930 else:
1931 requested_langs = [list(available_subs.keys())[0]]
1932
1933 formats_query = self.params.get('subtitlesformat', 'best')
1934 formats_preference = formats_query.split('/') if formats_query else []
1935 subs = {}
1936 for lang in requested_langs:
1937 formats = available_subs.get(lang)
1938 if formats is None:
1939 self.report_warning('%s subtitles not available for %s' % (lang, video_id))
1940 continue
a504ced0
JMF
1941 for ext in formats_preference:
1942 if ext == 'best':
1943 f = formats[-1]
1944 break
1945 matches = list(filter(lambda f: f['ext'] == ext, formats))
1946 if matches:
1947 f = matches[-1]
1948 break
1949 else:
1950 f = formats[-1]
1951 self.report_warning(
1952 'No subtitle format found matching "%s" for language %s, '
1953 'using %s' % (formats_query, lang, f['ext']))
1954 subs[lang] = f
1955 return subs
1956
d06daf23
S
1957 def __forced_printings(self, info_dict, filename, incomplete):
1958 def print_mandatory(field):
1959 if (self.params.get('force%s' % field, False)
1960 and (not incomplete or info_dict.get(field) is not None)):
1961 self.to_stdout(info_dict[field])
1962
1963 def print_optional(field):
1964 if (self.params.get('force%s' % field, False)
1965 and info_dict.get(field) is not None):
1966 self.to_stdout(info_dict[field])
1967
1968 print_mandatory('title')
1969 print_mandatory('id')
1970 if self.params.get('forceurl', False) and not incomplete:
1971 if info_dict.get('requested_formats') is not None:
1972 for f in info_dict['requested_formats']:
1973 self.to_stdout(f['url'] + f.get('play_path', ''))
1974 else:
1975 # For RTMP URLs, also include the playpath
1976 self.to_stdout(info_dict['url'] + info_dict.get('play_path', ''))
1977 print_optional('thumbnail')
1978 print_optional('description')
1979 if self.params.get('forcefilename', False) and filename is not None:
1980 self.to_stdout(filename)
1981 if self.params.get('forceduration', False) and info_dict.get('duration') is not None:
1982 self.to_stdout(formatSeconds(info_dict['duration']))
1983 print_mandatory('format')
1984 if self.params.get('forcejson', False):
1985 self.to_stdout(json.dumps(info_dict))
1986
8222d8de
JMF
1987 def process_info(self, info_dict):
1988 """Process a single resolved IE result."""
1989
1990 assert info_dict.get('_type', 'video') == 'video'
fd288278 1991
0202b52a 1992 info_dict.setdefault('__postprocessors', [])
1993
fd288278
PH
1994 max_downloads = self.params.get('max_downloads')
1995 if max_downloads is not None:
1996 if self._num_downloads >= int(max_downloads):
1997 raise MaxDownloadsReached()
8222d8de 1998
d06daf23 1999 # TODO: backward compatibility, to be removed
8222d8de 2000 info_dict['fulltitle'] = info_dict['title']
8222d8de 2001
11b85ce6 2002 if 'format' not in info_dict:
8222d8de
JMF
2003 info_dict['format'] = info_dict['ext']
2004
8b0d7497 2005 if self._match_entry(info_dict, incomplete=False) is not None:
8222d8de
JMF
2006 return
2007
fd288278 2008 self._num_downloads += 1
8222d8de 2009
5bfa4862 2010 info_dict = self.pre_process(info_dict)
2011
0202b52a 2012 filename = self.prepare_filename(info_dict, warn=True)
2013 info_dict['_filename'] = full_filename = self.prepare_filepath(filename)
2014 temp_filename = self.prepare_filepath(filename, 'temp')
2015 files_to_move = {}
8222d8de
JMF
2016
2017 # Forced printings
0202b52a 2018 self.__forced_printings(info_dict, full_filename, incomplete=False)
8222d8de 2019
8222d8de 2020 if self.params.get('simulate', False):
2d30509f 2021 if self.params.get('force_write_download_archive', False):
2022 self.record_download_archive(info_dict)
2023
2024 # Do nothing else if in simulate mode
8222d8de
JMF
2025 return
2026
2027 if filename is None:
2028 return
2029
c5c9bf0c 2030 def ensure_dir_exists(path):
0202b52a 2031 return make_dir(path, self.report_error)
c5c9bf0c 2032
0202b52a 2033 if not ensure_dir_exists(encodeFilename(full_filename)):
2034 return
2035 if not ensure_dir_exists(encodeFilename(temp_filename)):
8222d8de
JMF
2036 return
2037
2038 if self.params.get('writedescription', False):
0202b52a 2039 descfn = replace_extension(
2040 self.prepare_filepath(filename, 'description'),
2041 'description', info_dict.get('ext'))
2042 if not ensure_dir_exists(encodeFilename(descfn)):
2043 return
0c3d0f51 2044 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(descfn)):
6febd1c1 2045 self.to_screen('[info] Video description is already present')
f00fd51d
JMF
2046 elif info_dict.get('description') is None:
2047 self.report_warning('There\'s no description to write.')
7b6fefc9
PH
2048 else:
2049 try:
6febd1c1 2050 self.to_screen('[info] Writing video description to: ' + descfn)
7b6fefc9
PH
2051 with io.open(encodeFilename(descfn), 'w', encoding='utf-8') as descfile:
2052 descfile.write(info_dict['description'])
7b6fefc9 2053 except (OSError, IOError):
6febd1c1 2054 self.report_error('Cannot write description file ' + descfn)
7b6fefc9 2055 return
8222d8de 2056
1fb07d10 2057 if self.params.get('writeannotations', False):
0202b52a 2058 annofn = replace_extension(
2059 self.prepare_filepath(filename, 'annotation'),
2060 'annotations.xml', info_dict.get('ext'))
2061 if not ensure_dir_exists(encodeFilename(annofn)):
2062 return
0c3d0f51 2063 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(annofn)):
6febd1c1 2064 self.to_screen('[info] Video annotations are already present')
ffddb112
RA
2065 elif not info_dict.get('annotations'):
2066 self.report_warning('There are no annotations to write.')
7b6fefc9
PH
2067 else:
2068 try:
6febd1c1 2069 self.to_screen('[info] Writing video annotations to: ' + annofn)
7b6fefc9
PH
2070 with io.open(encodeFilename(annofn), 'w', encoding='utf-8') as annofile:
2071 annofile.write(info_dict['annotations'])
2072 except (KeyError, TypeError):
6febd1c1 2073 self.report_warning('There are no annotations to write.')
7b6fefc9 2074 except (OSError, IOError):
6febd1c1 2075 self.report_error('Cannot write annotations file: ' + annofn)
7b6fefc9 2076 return
1fb07d10 2077
9f448fcb 2078 def dl(name, info, subtitle=False):
98b69821 2079 fd = get_suitable_downloader(info, self.params)(self, self.params)
2080 for ph in self._progress_hooks:
2081 fd.add_progress_hook(ph)
2082 if self.params.get('verbose'):
29f7c58a 2083 self.to_screen('[debug] Invoking downloader on %r' % info.get('url'))
9f448fcb 2084 return fd.download(name, info, subtitle)
98b69821 2085
c4a91be7 2086 subtitles_are_requested = any([self.params.get('writesubtitles', False),
0b7f3118 2087 self.params.get('writeautomaticsub')])
c4a91be7 2088
c84dd8a9 2089 if subtitles_are_requested and info_dict.get('requested_subtitles'):
8222d8de
JMF
2090 # subtitles download errors are already managed as troubles in relevant IE
2091 # that way it will silently go on when used with unsupporting IE
c84dd8a9 2092 subtitles = info_dict['requested_subtitles']
fa57af1e 2093 # ie = self.get_info_extractor(info_dict['extractor_key'])
a504ced0
JMF
2094 for sub_lang, sub_info in subtitles.items():
2095 sub_format = sub_info['ext']
0202b52a 2096 sub_filename = subtitles_filename(temp_filename, sub_lang, sub_format, info_dict.get('ext'))
2097 sub_filename_final = subtitles_filename(
2098 self.prepare_filepath(filename, 'subtitle'),
2099 sub_lang, sub_format, info_dict.get('ext'))
0c3d0f51 2100 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(sub_filename)):
5ff1bc0c 2101 self.to_screen('[info] Video subtitle %s.%s is already present' % (sub_lang, sub_format))
0202b52a 2102 files_to_move[sub_filename] = sub_filename_final
a504ced0 2103 else:
0c9df79e 2104 self.to_screen('[info] Writing video subtitles to: ' + sub_filename)
5ff1bc0c
RA
2105 if sub_info.get('data') is not None:
2106 try:
2107 # Use newline='' to prevent conversion of newline characters
067aa17e 2108 # See https://github.com/ytdl-org/youtube-dl/issues/10268
5ff1bc0c
RA
2109 with io.open(encodeFilename(sub_filename), 'w', encoding='utf-8', newline='') as subfile:
2110 subfile.write(sub_info['data'])
0202b52a 2111 files_to_move[sub_filename] = sub_filename_final
5ff1bc0c
RA
2112 except (OSError, IOError):
2113 self.report_error('Cannot write subtitles file ' + sub_filename)
2114 return
7b6fefc9 2115 else:
5ff1bc0c 2116 try:
9f448fcb
U
2117 dl(sub_filename, sub_info, subtitle=True)
2118 '''
0c9df79e
U
2119 if self.params.get('sleep_interval_subtitles', False):
2120 dl(sub_filename, sub_info)
2121 else:
2122 sub_data = ie._request_webpage(
2123 sub_info['url'], info_dict['id'], note=False).read()
2124 with io.open(encodeFilename(sub_filename), 'wb') as subfile:
2125 subfile.write(sub_data)
9f448fcb 2126 '''
0202b52a 2127 files_to_move[sub_filename] = sub_filename_final
0c9df79e 2128 except (ExtractorError, IOError, OSError, ValueError, compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
5ff1bc0c
RA
2129 self.report_warning('Unable to download subtitle for "%s": %s' %
2130 (sub_lang, error_to_compat_str(err)))
2131 continue
8222d8de 2132
57df9f53
U
2133 if self.params.get('skip_download', False):
2134 if self.params.get('convertsubtitles', False):
0202b52a 2135 # subconv = FFmpegSubtitlesConvertorPP(self, format=self.params.get('convertsubtitles'))
57df9f53
U
2136 filename_real_ext = os.path.splitext(filename)[1][1:]
2137 filename_wo_ext = (
0202b52a 2138 os.path.splitext(full_filename)[0]
57df9f53 2139 if filename_real_ext == info_dict['ext']
0202b52a 2140 else full_filename)
57df9f53 2141 afilename = '%s.%s' % (filename_wo_ext, self.params.get('convertsubtitles'))
0202b52a 2142 # if subconv.available:
2143 # info_dict['__postprocessors'].append(subconv)
57df9f53 2144 if os.path.exists(encodeFilename(afilename)):
f791b419
U
2145 self.to_screen(
2146 '[download] %s has already been downloaded and '
2147 'converted' % afilename)
57df9f53
U
2148 else:
2149 try:
0202b52a 2150 self.post_process(full_filename, info_dict, files_to_move)
af819c21 2151 except PostProcessingError as err:
2152 self.report_error('Postprocessing: %s' % str(err))
57df9f53
U
2153 return
2154
8222d8de 2155 if self.params.get('writeinfojson', False):
0202b52a 2156 infofn = replace_extension(
2157 self.prepare_filepath(filename, 'infojson'),
2158 'info.json', info_dict.get('ext'))
2159 if not ensure_dir_exists(encodeFilename(infofn)):
2160 return
0c3d0f51 2161 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(infofn)):
66c935fb 2162 self.to_screen('[info] Video metadata is already present')
7b6fefc9 2163 else:
66c935fb 2164 self.to_screen('[info] Writing video metadata as JSON to: ' + infofn)
7b6fefc9 2165 try:
cb202fd2 2166 write_json_file(self.filter_requested_info(info_dict), infofn)
7b6fefc9 2167 except (OSError, IOError):
66c935fb 2168 self.report_error('Cannot write video metadata to JSON file ' + infofn)
7b6fefc9 2169 return
06167fbb 2170 info_dict['__infojson_filepath'] = infofn
8222d8de 2171
0202b52a 2172 thumbdir = os.path.dirname(self.prepare_filepath(filename, 'thumbnail'))
2173 for thumbfn in self._write_thumbnails(info_dict, temp_filename):
2174 files_to_move[thumbfn] = os.path.join(thumbdir, os.path.basename(thumbfn))
8222d8de 2175
732044af 2176 # Write internet shortcut files
2177 url_link = webloc_link = desktop_link = False
2178 if self.params.get('writelink', False):
2179 if sys.platform == "darwin": # macOS.
2180 webloc_link = True
2181 elif sys.platform.startswith("linux"):
2182 desktop_link = True
2183 else: # if sys.platform in ['win32', 'cygwin']:
2184 url_link = True
2185 if self.params.get('writeurllink', False):
2186 url_link = True
2187 if self.params.get('writewebloclink', False):
2188 webloc_link = True
2189 if self.params.get('writedesktoplink', False):
2190 desktop_link = True
2191
2192 if url_link or webloc_link or desktop_link:
2193 if 'webpage_url' not in info_dict:
2194 self.report_error('Cannot write internet shortcut file because the "webpage_url" field is missing in the media information')
2195 return
2196 ascii_url = iri_to_uri(info_dict['webpage_url'])
2197
2198 def _write_link_file(extension, template, newline, embed_filename):
0202b52a 2199 linkfn = replace_extension(full_filename, extension, info_dict.get('ext'))
10e3742e 2200 if self.params.get('overwrites', True) and os.path.exists(encodeFilename(linkfn)):
732044af 2201 self.to_screen('[info] Internet shortcut is already present')
2202 else:
2203 try:
2204 self.to_screen('[info] Writing internet shortcut to: ' + linkfn)
2205 with io.open(encodeFilename(to_high_limit_path(linkfn)), 'w', encoding='utf-8', newline=newline) as linkfile:
2206 template_vars = {'url': ascii_url}
2207 if embed_filename:
2208 template_vars['filename'] = linkfn[:-(len(extension) + 1)]
2209 linkfile.write(template % template_vars)
2210 except (OSError, IOError):
2211 self.report_error('Cannot write internet shortcut ' + linkfn)
2212 return False
2213 return True
2214
2215 if url_link:
2216 if not _write_link_file('url', DOT_URL_LINK_TEMPLATE, '\r\n', embed_filename=False):
2217 return
2218 if webloc_link:
2219 if not _write_link_file('webloc', DOT_WEBLOC_LINK_TEMPLATE, '\n', embed_filename=False):
2220 return
2221 if desktop_link:
2222 if not _write_link_file('desktop', DOT_DESKTOP_LINK_TEMPLATE, '\n', embed_filename=True):
2223 return
2224
2225 # Download
2226 must_record_download_archive = False
8222d8de 2227 if not self.params.get('skip_download', False):
4340deca 2228 try:
0202b52a 2229
6b591b29 2230 def existing_file(*filepaths):
2231 ext = info_dict.get('ext')
2232 final_ext = self.params.get('final_ext', ext)
2233 existing_files = []
2234 for file in orderedSet(filepaths):
2235 if final_ext != ext:
2236 converted = replace_extension(file, final_ext, ext)
2237 if os.path.exists(encodeFilename(converted)):
2238 existing_files.append(converted)
2239 if os.path.exists(encodeFilename(file)):
2240 existing_files.append(file)
2241
2242 if not existing_files or self.params.get('overwrites', False):
2243 for file in orderedSet(existing_files):
2244 self.report_file_delete(file)
2245 os.remove(encodeFilename(file))
2246 return None
2247
2248 self.report_file_already_downloaded(existing_files[0])
2249 info_dict['ext'] = os.path.splitext(existing_files[0])[1][1:]
2250 return existing_files[0]
0202b52a 2251
2252 success = True
4340deca
P
2253 if info_dict.get('requested_formats') is not None:
2254 downloaded = []
d47aeb22 2255 merger = FFmpegMergerPP(self)
f740fae2 2256 if not merger.available:
4340deca
P
2257 postprocessors = []
2258 self.report_warning('You have requested multiple '
e4172ac9 2259 'formats but ffmpeg is not installed.'
4a5a898a 2260 ' The formats won\'t be merged.')
6350728b 2261 else:
4340deca 2262 postprocessors = [merger]
81cd954a
S
2263
2264 def compatible_formats(formats):
d03cfdce 2265 # TODO: some formats actually allow this (mkv, webm, ogg, mp4), but not all of them.
2266 video_formats = [format for format in formats if format.get('vcodec') != 'none']
2267 audio_formats = [format for format in formats if format.get('acodec') != 'none']
2268 if len(video_formats) > 2 or len(audio_formats) > 2:
2269 return False
2270
81cd954a 2271 # Check extension
d03cfdce 2272 exts = set(format.get('ext') for format in formats)
2273 COMPATIBLE_EXTS = (
2274 set(('mp3', 'mp4', 'm4a', 'm4p', 'm4b', 'm4r', 'm4v', 'ismv', 'isma')),
2275 set(('webm',)),
2276 )
2277 for ext_sets in COMPATIBLE_EXTS:
2278 if ext_sets.issuperset(exts):
2279 return True
81cd954a
S
2280 # TODO: Check acodec/vcodec
2281 return False
2282
2283 requested_formats = info_dict['requested_formats']
0202b52a 2284 old_ext = info_dict['ext']
c0dea0a7 2285 if self.params.get('merge_output_format') is None and not compatible_formats(requested_formats):
38c6902b 2286 info_dict['ext'] = 'mkv'
4a5a898a
S
2287 self.report_warning(
2288 'Requested formats are incompatible for merge and will be merged into mkv.')
0202b52a 2289
2290 def correct_ext(filename):
2291 filename_real_ext = os.path.splitext(filename)[1][1:]
2292 filename_wo_ext = (
2293 os.path.splitext(filename)[0]
2294 if filename_real_ext == old_ext
2295 else filename)
2296 return '%s.%s' % (filename_wo_ext, info_dict['ext'])
2297
38c6902b 2298 # Ensure filename always has a correct extension for successful merge
0202b52a 2299 full_filename = correct_ext(full_filename)
2300 temp_filename = correct_ext(temp_filename)
2301 dl_filename = existing_file(full_filename, temp_filename)
2302 if dl_filename is None:
81cd954a 2303 for f in requested_formats:
5b5fbc08
JMF
2304 new_info = dict(info_dict)
2305 new_info.update(f)
c5c9bf0c 2306 fname = prepend_extension(
0202b52a 2307 self.prepare_filepath(self.prepare_filename(new_info), 'temp'),
c5c9bf0c
S
2308 'f%s' % f['format_id'], new_info['ext'])
2309 if not ensure_dir_exists(fname):
2310 return
5b5fbc08 2311 downloaded.append(fname)
a9e7f546 2312 partial_success, real_download = dl(fname, new_info)
5b5fbc08
JMF
2313 success = success and partial_success
2314 info_dict['__postprocessors'] = postprocessors
2315 info_dict['__files_to_merge'] = downloaded
a9e7f546 2316 # Even if there were no downloads, it is being merged only now
2317 info_dict['__real_download'] = True
4340deca
P
2318 else:
2319 # Just a single file
0202b52a 2320 dl_filename = existing_file(full_filename, temp_filename)
2321 if dl_filename is None:
2322 success, real_download = dl(temp_filename, info_dict)
2323 info_dict['__real_download'] = real_download
2324
0202b52a 2325 dl_filename = dl_filename or temp_filename
c571435f 2326 info_dict['__finaldir'] = os.path.dirname(os.path.abspath(encodeFilename(full_filename)))
0202b52a 2327
4340deca 2328 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
7960b056 2329 self.report_error('unable to download video data: %s' % error_to_compat_str(err))
4340deca
P
2330 return
2331 except (OSError, IOError) as err:
2332 raise UnavailableVideoError(err)
2333 except (ContentTooShortError, ) as err:
2334 self.report_error('content too short (expected %s bytes and served %s)' % (err.expected, err.downloaded))
2335 return
8222d8de 2336
e38cafe9 2337 if success and filename != '-':
6271f1ca 2338 # Fixup content
62cd676c
PH
2339 fixup_policy = self.params.get('fixup')
2340 if fixup_policy is None:
2341 fixup_policy = 'detect_or_warn'
2342
e4172ac9 2343 INSTALL_FFMPEG_MESSAGE = 'Install ffmpeg to fix this automatically.'
d1e4a464 2344
6271f1ca
PH
2345 stretched_ratio = info_dict.get('stretched_ratio')
2346 if stretched_ratio is not None and stretched_ratio != 1:
6271f1ca
PH
2347 if fixup_policy == 'warn':
2348 self.report_warning('%s: Non-uniform pixel ratio (%s)' % (
2349 info_dict['id'], stretched_ratio))
2350 elif fixup_policy == 'detect_or_warn':
2351 stretched_pp = FFmpegFixupStretchedPP(self)
2352 if stretched_pp.available:
6271f1ca
PH
2353 info_dict['__postprocessors'].append(stretched_pp)
2354 else:
2355 self.report_warning(
d1e4a464
S
2356 '%s: Non-uniform pixel ratio (%s). %s'
2357 % (info_dict['id'], stretched_ratio, INSTALL_FFMPEG_MESSAGE))
6271f1ca 2358 else:
62cd676c
PH
2359 assert fixup_policy in ('ignore', 'never')
2360
3089bc74 2361 if (info_dict.get('requested_formats') is None
6b591b29 2362 and info_dict.get('container') == 'm4a_dash'
2363 and info_dict.get('ext') == 'm4a'):
62cd676c 2364 if fixup_policy == 'warn':
d1e4a464
S
2365 self.report_warning(
2366 '%s: writing DASH m4a. '
2367 'Only some players support this container.'
2368 % info_dict['id'])
62cd676c
PH
2369 elif fixup_policy == 'detect_or_warn':
2370 fixup_pp = FFmpegFixupM4aPP(self)
2371 if fixup_pp.available:
62cd676c
PH
2372 info_dict['__postprocessors'].append(fixup_pp)
2373 else:
2374 self.report_warning(
d1e4a464
S
2375 '%s: writing DASH m4a. '
2376 'Only some players support this container. %s'
2377 % (info_dict['id'], INSTALL_FFMPEG_MESSAGE))
62cd676c
PH
2378 else:
2379 assert fixup_policy in ('ignore', 'never')
6271f1ca 2380
3089bc74
S
2381 if (info_dict.get('protocol') == 'm3u8_native'
2382 or info_dict.get('protocol') == 'm3u8'
2383 and self.params.get('hls_prefer_native')):
f17f8651 2384 if fixup_policy == 'warn':
a02682fd 2385 self.report_warning('%s: malformed AAC bitstream detected.' % (
f17f8651 2386 info_dict['id']))
2387 elif fixup_policy == 'detect_or_warn':
2388 fixup_pp = FFmpegFixupM3u8PP(self)
2389 if fixup_pp.available:
f17f8651 2390 info_dict['__postprocessors'].append(fixup_pp)
2391 else:
2392 self.report_warning(
a02682fd 2393 '%s: malformed AAC bitstream detected. %s'
d1e4a464 2394 % (info_dict['id'], INSTALL_FFMPEG_MESSAGE))
f17f8651 2395 else:
2396 assert fixup_policy in ('ignore', 'never')
2397
8222d8de 2398 try:
0202b52a 2399 self.post_process(dl_filename, info_dict, files_to_move)
af819c21 2400 except PostProcessingError as err:
2401 self.report_error('Postprocessing: %s' % str(err))
8222d8de 2402 return
ab8e5e51
AM
2403 try:
2404 for ph in self._post_hooks:
0202b52a 2405 ph(full_filename)
ab8e5e51
AM
2406 except Exception as err:
2407 self.report_error('post hooks: %s' % str(err))
2408 return
2d30509f 2409 must_record_download_archive = True
2410
2411 if must_record_download_archive or self.params.get('force_write_download_archive', False):
2412 self.record_download_archive(info_dict)
c3e6ffba 2413 max_downloads = self.params.get('max_downloads')
2414 if max_downloads is not None and self._num_downloads >= int(max_downloads):
2415 raise MaxDownloadsReached()
8222d8de
JMF
2416
2417 def download(self, url_list):
2418 """Download a given list of URLs."""
acd69589 2419 outtmpl = self.params.get('outtmpl', DEFAULT_OUTTMPL)
3089bc74
S
2420 if (len(url_list) > 1
2421 and outtmpl != '-'
2422 and '%' not in outtmpl
2423 and self.params.get('max_downloads') != 1):
acd69589 2424 raise SameFileError(outtmpl)
8222d8de
JMF
2425
2426 for url in url_list:
2427 try:
5f6a1245 2428 # It also downloads the videos
61aa5ba3
S
2429 res = self.extract_info(
2430 url, force_generic_extractor=self.params.get('force_generic_extractor', False))
8222d8de 2431 except UnavailableVideoError:
6febd1c1 2432 self.report_error('unable to download video')
8222d8de 2433 except MaxDownloadsReached:
8b0d7497 2434 self.to_screen('[info] Maximum number of downloaded files reached')
2435 raise
2436 except ExistingVideoReached:
d83cb531 2437 self.to_screen('[info] Encountered a file that is already in the archive, stopping due to --break-on-existing')
8b0d7497 2438 raise
2439 except RejectedVideoReached:
d83cb531 2440 self.to_screen('[info] Encountered a file that did not match filter, stopping due to --break-on-reject')
8222d8de 2441 raise
63e0be34
PH
2442 else:
2443 if self.params.get('dump_single_json', False):
2444 self.to_stdout(json.dumps(res))
8222d8de
JMF
2445
2446 return self._download_retcode
2447
1dcc4c0c 2448 def download_with_info_file(self, info_filename):
31bd3925
JMF
2449 with contextlib.closing(fileinput.FileInput(
2450 [info_filename], mode='r',
2451 openhook=fileinput.hook_encoded('utf-8'))) as f:
2452 # FileInput doesn't have a read method, we can't call json.load
cb202fd2 2453 info = self.filter_requested_info(json.loads('\n'.join(f)))
d4943898
JMF
2454 try:
2455 self.process_ie_result(info, download=True)
2456 except DownloadError:
2457 webpage_url = info.get('webpage_url')
2458 if webpage_url is not None:
6febd1c1 2459 self.report_warning('The info failed to download, trying with "%s"' % webpage_url)
d4943898
JMF
2460 return self.download([webpage_url])
2461 else:
2462 raise
2463 return self._download_retcode
1dcc4c0c 2464
cb202fd2
S
2465 @staticmethod
2466 def filter_requested_info(info_dict):
2467 return dict(
2468 (k, v) for k, v in info_dict.items()
2469 if k not in ['requested_formats', 'requested_subtitles'])
2470
5bfa4862 2471 def run_pp(self, pp, infodict, files_to_move={}):
2472 files_to_delete = []
af819c21 2473 files_to_delete, infodict = pp.run(infodict)
5bfa4862 2474 if not files_to_delete:
2475 return files_to_move, infodict
2476
2477 if self.params.get('keepvideo', False):
2478 for f in files_to_delete:
2479 files_to_move.setdefault(f, '')
2480 else:
2481 for old_filename in set(files_to_delete):
2482 self.to_screen('Deleting original file %s (pass -k to keep)' % old_filename)
2483 try:
2484 os.remove(encodeFilename(old_filename))
2485 except (IOError, OSError):
2486 self.report_warning('Unable to remove downloaded original file')
2487 if old_filename in files_to_move:
2488 del files_to_move[old_filename]
2489 return files_to_move, infodict
2490
2491 def pre_process(self, ie_info):
2492 info = dict(ie_info)
2493 for pp in self._pps['beforedl']:
2494 info = self.run_pp(pp, info)[1]
2495 return info
2496
0202b52a 2497 def post_process(self, filename, ie_info, files_to_move={}):
8222d8de
JMF
2498 """Run all the postprocessors on the given file."""
2499 info = dict(ie_info)
2500 info['filepath'] = filename
0202b52a 2501
5bfa4862 2502 for pp in ie_info.get('__postprocessors', []) + self._pps['normal']:
2503 files_to_move, info = self.run_pp(pp, info, files_to_move)
2504 info = self.run_pp(MoveFilesAfterDownloadPP(self, files_to_move), info, files_to_move)[1]
2505 for pp in self._pps['aftermove']:
2506 files_to_move, info = self.run_pp(pp, info, {})
c1c9a79c 2507
5db07df6 2508 def _make_archive_id(self, info_dict):
e9fef7ee
S
2509 video_id = info_dict.get('id')
2510 if not video_id:
2511 return
5db07df6
PH
2512 # Future-proof against any change in case
2513 # and backwards compatibility with prior versions
e9fef7ee 2514 extractor = info_dict.get('extractor_key') or info_dict.get('ie_key') # key in a playlist
7012b23c 2515 if extractor is None:
1211bb6d
S
2516 url = str_or_none(info_dict.get('url'))
2517 if not url:
2518 return
e9fef7ee
S
2519 # Try to find matching extractor for the URL and take its ie_key
2520 for ie in self._ies:
1211bb6d 2521 if ie.suitable(url):
e9fef7ee
S
2522 extractor = ie.ie_key()
2523 break
2524 else:
2525 return
d0757229 2526 return '%s %s' % (extractor.lower(), video_id)
5db07df6
PH
2527
2528 def in_download_archive(self, info_dict):
2529 fn = self.params.get('download_archive')
2530 if fn is None:
2531 return False
2532
2533 vid_id = self._make_archive_id(info_dict)
e9fef7ee 2534 if not vid_id:
7012b23c 2535 return False # Incomplete video information
5db07df6 2536
a45e8619 2537 return vid_id in self.archive
c1c9a79c
PH
2538
2539 def record_download_archive(self, info_dict):
2540 fn = self.params.get('download_archive')
2541 if fn is None:
2542 return
5db07df6
PH
2543 vid_id = self._make_archive_id(info_dict)
2544 assert vid_id
c1c9a79c 2545 with locked_file(fn, 'a', encoding='utf-8') as archive_file:
6febd1c1 2546 archive_file.write(vid_id + '\n')
a45e8619 2547 self.archive.add(vid_id)
dd82ffea 2548
8c51aa65 2549 @staticmethod
8abeeb94 2550 def format_resolution(format, default='unknown'):
fb04e403
PH
2551 if format.get('vcodec') == 'none':
2552 return 'audio only'
f49d89ee
PH
2553 if format.get('resolution') is not None:
2554 return format['resolution']
8c51aa65
JMF
2555 if format.get('height') is not None:
2556 if format.get('width') is not None:
6febd1c1 2557 res = '%sx%s' % (format['width'], format['height'])
8c51aa65 2558 else:
6febd1c1 2559 res = '%sp' % format['height']
f49d89ee 2560 elif format.get('width') is not None:
388ae76b 2561 res = '%dx?' % format['width']
8c51aa65 2562 else:
8abeeb94 2563 res = default
8c51aa65
JMF
2564 return res
2565
c57f7757
PH
2566 def _format_note(self, fdict):
2567 res = ''
2568 if fdict.get('ext') in ['f4f', 'f4m']:
2569 res += '(unsupported) '
32f90364
PH
2570 if fdict.get('language'):
2571 if res:
2572 res += ' '
9016d76f 2573 res += '[%s] ' % fdict['language']
c57f7757
PH
2574 if fdict.get('format_note') is not None:
2575 res += fdict['format_note'] + ' '
2576 if fdict.get('tbr') is not None:
2577 res += '%4dk ' % fdict['tbr']
2578 if fdict.get('container') is not None:
2579 if res:
2580 res += ', '
2581 res += '%s container' % fdict['container']
3089bc74
S
2582 if (fdict.get('vcodec') is not None
2583 and fdict.get('vcodec') != 'none'):
c57f7757
PH
2584 if res:
2585 res += ', '
2586 res += fdict['vcodec']
91c7271a 2587 if fdict.get('vbr') is not None:
c57f7757
PH
2588 res += '@'
2589 elif fdict.get('vbr') is not None and fdict.get('abr') is not None:
2590 res += 'video@'
2591 if fdict.get('vbr') is not None:
2592 res += '%4dk' % fdict['vbr']
fbb21cf5 2593 if fdict.get('fps') is not None:
5d583bdf
S
2594 if res:
2595 res += ', '
2596 res += '%sfps' % fdict['fps']
c57f7757
PH
2597 if fdict.get('acodec') is not None:
2598 if res:
2599 res += ', '
2600 if fdict['acodec'] == 'none':
2601 res += 'video only'
2602 else:
2603 res += '%-5s' % fdict['acodec']
2604 elif fdict.get('abr') is not None:
2605 if res:
2606 res += ', '
2607 res += 'audio'
2608 if fdict.get('abr') is not None:
2609 res += '@%3dk' % fdict['abr']
2610 if fdict.get('asr') is not None:
2611 res += ' (%5dHz)' % fdict['asr']
2612 if fdict.get('filesize') is not None:
2613 if res:
2614 res += ', '
2615 res += format_bytes(fdict['filesize'])
9732d77e
PH
2616 elif fdict.get('filesize_approx') is not None:
2617 if res:
2618 res += ', '
2619 res += '~' + format_bytes(fdict['filesize_approx'])
c57f7757 2620 return res
91c7271a 2621
76d321f6 2622 def _format_note_table(self, f):
2623 def join_fields(*vargs):
2624 return ', '.join((val for val in vargs if val != ''))
2625
2626 return join_fields(
2627 'UNSUPPORTED' if f.get('ext') in ('f4f', 'f4m') else '',
2628 format_field(f, 'language', '[%s]'),
2629 format_field(f, 'format_note'),
2630 format_field(f, 'container', ignore=(None, f.get('ext'))),
2631 format_field(f, 'asr', '%5dHz'))
2632
c57f7757 2633 def list_formats(self, info_dict):
94badb25 2634 formats = info_dict.get('formats', [info_dict])
76d321f6 2635 new_format = self.params.get('listformats_table', False)
2636 if new_format:
2637 table = [
2638 [
2639 format_field(f, 'format_id'),
2640 format_field(f, 'ext'),
2641 self.format_resolution(f),
2642 format_field(f, 'fps', '%d'),
2643 '|',
2644 format_field(f, 'filesize', ' %s', func=format_bytes) + format_field(f, 'filesize_approx', '~%s', func=format_bytes),
2645 format_field(f, 'tbr', '%4dk'),
2646 f.get('protocol').replace('http_dash_segments', 'dash').replace("native", "n"),
2647 '|',
2648 format_field(f, 'vcodec', default='unknown').replace('none', ''),
2649 format_field(f, 'vbr', '%4dk'),
2650 format_field(f, 'acodec', default='unknown').replace('none', ''),
2651 format_field(f, 'abr', '%3dk'),
2652 format_field(f, 'asr', '%5dHz'),
2653 self._format_note_table(f)]
2654 for f in formats
2655 if f.get('preference') is None or f['preference'] >= -1000]
2656 header_line = ['ID', 'EXT', 'RESOLUTION', 'FPS', '|', ' FILESIZE', ' TBR', 'PROTO',
2657 '|', 'VCODEC', ' VBR', 'ACODEC', ' ABR', ' ASR', 'NOTE']
2658 else:
2659 table = [
2660 [
2661 format_field(f, 'format_id'),
2662 format_field(f, 'ext'),
2663 self.format_resolution(f),
2664 self._format_note(f)]
2665 for f in formats
2666 if f.get('preference') is None or f['preference'] >= -1000]
2667 header_line = ['format code', 'extension', 'resolution', 'note']
57dd9a8f 2668
76d321f6 2669 # if len(formats) > 1:
2670 # table[-1][-1] += (' ' if table[-1][-1] else '') + '(best)'
cfb56d1a 2671 self.to_screen(
76d321f6 2672 '[info] Available formats for %s:\n%s' % (info_dict['id'], render_table(
2673 header_line,
2674 table,
2675 delim=new_format,
2676 extraGap=(0 if new_format else 1),
2677 hideEmpty=new_format)))
cfb56d1a
PH
2678
2679 def list_thumbnails(self, info_dict):
2680 thumbnails = info_dict.get('thumbnails')
2681 if not thumbnails:
b7b72db9 2682 self.to_screen('[info] No thumbnails present for %s' % info_dict['id'])
2683 return
cfb56d1a
PH
2684
2685 self.to_screen(
2686 '[info] Thumbnails for %s:' % info_dict['id'])
2687 self.to_screen(render_table(
2688 ['ID', 'width', 'height', 'URL'],
2689 [[t['id'], t.get('width', 'unknown'), t.get('height', 'unknown'), t['url']] for t in thumbnails]))
dca08720 2690
360e1ca5 2691 def list_subtitles(self, video_id, subtitles, name='subtitles'):
a504ced0 2692 if not subtitles:
360e1ca5 2693 self.to_screen('%s has no %s' % (video_id, name))
a504ced0 2694 return
a504ced0 2695 self.to_screen(
edab9dbf
JMF
2696 'Available %s for %s:' % (name, video_id))
2697 self.to_screen(render_table(
2698 ['Language', 'formats'],
2699 [[lang, ', '.join(f['ext'] for f in reversed(formats))]
2700 for lang, formats in subtitles.items()]))
a504ced0 2701
dca08720
PH
2702 def urlopen(self, req):
2703 """ Start an HTTP download """
82d8a8b6 2704 if isinstance(req, compat_basestring):
67dda517 2705 req = sanitized_Request(req)
19a41fc6 2706 return self._opener.open(req, timeout=self._socket_timeout)
dca08720
PH
2707
2708 def print_debug_header(self):
2709 if not self.params.get('verbose'):
2710 return
62fec3b2 2711
4192b51c 2712 if type('') is not compat_str:
067aa17e 2713 # Python 2.6 on SLES11 SP1 (https://github.com/ytdl-org/youtube-dl/issues/3326)
4192b51c
PH
2714 self.report_warning(
2715 'Your Python is broken! Update to a newer and supported version')
2716
c6afed48
PH
2717 stdout_encoding = getattr(
2718 sys.stdout, 'encoding', 'missing (%s)' % type(sys.stdout).__name__)
b0472057 2719 encoding_str = (
734f90bb
PH
2720 '[debug] Encodings: locale %s, fs %s, out %s, pref %s\n' % (
2721 locale.getpreferredencoding(),
2722 sys.getfilesystemencoding(),
c6afed48 2723 stdout_encoding,
b0472057 2724 self.get_encoding()))
4192b51c 2725 write_string(encoding_str, encoding=None)
734f90bb 2726
f74980cb 2727 self._write_string('[debug] yt-dlp version %s\n' % __version__)
e0986e31 2728 if _LAZY_LOADER:
f74980cb 2729 self._write_string('[debug] Lazy loading extractors enabled\n')
2730 if _PLUGIN_CLASSES:
2731 self._write_string(
2732 '[debug] Plugin Extractors: %s\n' % [ie.ie_key() for ie in _PLUGIN_CLASSES])
dca08720
PH
2733 try:
2734 sp = subprocess.Popen(
2735 ['git', 'rev-parse', '--short', 'HEAD'],
2736 stdout=subprocess.PIPE, stderr=subprocess.PIPE,
2737 cwd=os.path.dirname(os.path.abspath(__file__)))
f5b1bca9 2738 out, err = process_communicate_or_kill(sp)
dca08720
PH
2739 out = out.decode().strip()
2740 if re.match('[0-9a-f]+', out):
f74980cb 2741 self._write_string('[debug] Git HEAD: %s\n' % out)
70a1165b 2742 except Exception:
dca08720
PH
2743 try:
2744 sys.exc_clear()
70a1165b 2745 except Exception:
dca08720 2746 pass
b300cda4
S
2747
2748 def python_implementation():
2749 impl_name = platform.python_implementation()
2750 if impl_name == 'PyPy' and hasattr(sys, 'pypy_version_info'):
2751 return impl_name + ' version %d.%d.%d' % sys.pypy_version_info[:3]
2752 return impl_name
2753
2754 self._write_string('[debug] Python version %s (%s) - %s\n' % (
2755 platform.python_version(), python_implementation(),
2756 platform_name()))
d28b5171 2757
73fac4e9 2758 exe_versions = FFmpegPostProcessor.get_versions(self)
4c83c967 2759 exe_versions['rtmpdump'] = rtmpdump_version()
feee8d32 2760 exe_versions['phantomjs'] = PhantomJSwrapper._version()
d28b5171
PH
2761 exe_str = ', '.join(
2762 '%s %s' % (exe, v)
2763 for exe, v in sorted(exe_versions.items())
2764 if v
2765 )
2766 if not exe_str:
2767 exe_str = 'none'
2768 self._write_string('[debug] exe versions: %s\n' % exe_str)
dca08720
PH
2769
2770 proxy_map = {}
2771 for handler in self._opener.handlers:
2772 if hasattr(handler, 'proxies'):
2773 proxy_map.update(handler.proxies)
734f90bb 2774 self._write_string('[debug] Proxy map: ' + compat_str(proxy_map) + '\n')
dca08720 2775
58b1f00d
PH
2776 if self.params.get('call_home', False):
2777 ipaddr = self.urlopen('https://yt-dl.org/ip').read().decode('utf-8')
2778 self._write_string('[debug] Public IP address: %s\n' % ipaddr)
f5546c0b 2779 return
58b1f00d
PH
2780 latest_version = self.urlopen(
2781 'https://yt-dl.org/latest/version').read().decode('utf-8')
2782 if version_tuple(latest_version) > version_tuple(__version__):
2783 self.report_warning(
2784 'You are using an outdated version (newest version: %s)! '
2785 'See https://yt-dl.org/update if you need help updating.' %
2786 latest_version)
2787
e344693b 2788 def _setup_opener(self):
6ad14cab 2789 timeout_val = self.params.get('socket_timeout')
19a41fc6 2790 self._socket_timeout = 600 if timeout_val is None else float(timeout_val)
6ad14cab 2791
dca08720
PH
2792 opts_cookiefile = self.params.get('cookiefile')
2793 opts_proxy = self.params.get('proxy')
2794
2795 if opts_cookiefile is None:
2796 self.cookiejar = compat_cookiejar.CookieJar()
2797 else:
590bc6f6 2798 opts_cookiefile = expand_path(opts_cookiefile)
1bab3437 2799 self.cookiejar = YoutubeDLCookieJar(opts_cookiefile)
dca08720 2800 if os.access(opts_cookiefile, os.R_OK):
1d88b3e6 2801 self.cookiejar.load(ignore_discard=True, ignore_expires=True)
dca08720 2802
6a3f4c3f 2803 cookie_processor = YoutubeDLCookieProcessor(self.cookiejar)
dca08720
PH
2804 if opts_proxy is not None:
2805 if opts_proxy == '':
2806 proxies = {}
2807 else:
2808 proxies = {'http': opts_proxy, 'https': opts_proxy}
2809 else:
2810 proxies = compat_urllib_request.getproxies()
067aa17e 2811 # Set HTTPS proxy to HTTP one if given (https://github.com/ytdl-org/youtube-dl/issues/805)
dca08720
PH
2812 if 'http' in proxies and 'https' not in proxies:
2813 proxies['https'] = proxies['http']
91410c9b 2814 proxy_handler = PerRequestProxyHandler(proxies)
a0ddb8a2
PH
2815
2816 debuglevel = 1 if self.params.get('debug_printtraffic') else 0
be4a824d
PH
2817 https_handler = make_HTTPS_handler(self.params, debuglevel=debuglevel)
2818 ydlh = YoutubeDLHandler(self.params, debuglevel=debuglevel)
fca6dba8 2819 redirect_handler = YoutubeDLRedirectHandler()
8b172c2e 2820 data_handler = compat_urllib_request_DataHandler()
6240b0a2
JMF
2821
2822 # When passing our own FileHandler instance, build_opener won't add the
2823 # default FileHandler and allows us to disable the file protocol, which
2824 # can be used for malicious purposes (see
067aa17e 2825 # https://github.com/ytdl-org/youtube-dl/issues/8227)
6240b0a2
JMF
2826 file_handler = compat_urllib_request.FileHandler()
2827
2828 def file_open(*args, **kwargs):
cefecac1 2829 raise compat_urllib_error.URLError('file:// scheme is explicitly disabled in youtube-dlc for security reasons')
6240b0a2
JMF
2830 file_handler.file_open = file_open
2831
2832 opener = compat_urllib_request.build_opener(
fca6dba8 2833 proxy_handler, https_handler, cookie_processor, ydlh, redirect_handler, data_handler, file_handler)
2461f79d 2834
dca08720
PH
2835 # Delete the default user-agent header, which would otherwise apply in
2836 # cases where our custom HTTP handler doesn't come into play
067aa17e 2837 # (See https://github.com/ytdl-org/youtube-dl/issues/1309 for details)
dca08720
PH
2838 opener.addheaders = []
2839 self._opener = opener
62fec3b2
PH
2840
2841 def encode(self, s):
2842 if isinstance(s, bytes):
2843 return s # Already encoded
2844
2845 try:
2846 return s.encode(self.get_encoding())
2847 except UnicodeEncodeError as err:
2848 err.reason = err.reason + '. Check your system encoding configuration or use the --encoding option.'
2849 raise
2850
2851 def get_encoding(self):
2852 encoding = self.params.get('encoding')
2853 if encoding is None:
2854 encoding = preferredencoding()
2855 return encoding
ec82d85a
PH
2856
2857 def _write_thumbnails(self, info_dict, filename):
2858 if self.params.get('writethumbnail', False):
2859 thumbnails = info_dict.get('thumbnails')
2860 if thumbnails:
2861 thumbnails = [thumbnails[-1]]
2862 elif self.params.get('write_all_thumbnails', False):
0202b52a 2863 thumbnails = info_dict.get('thumbnails') or []
ec82d85a 2864 else:
0202b52a 2865 thumbnails = []
ec82d85a 2866
0202b52a 2867 ret = []
ec82d85a
PH
2868 for t in thumbnails:
2869 thumb_ext = determine_ext(t['url'], 'jpg')
2870 suffix = '_%s' % t['id'] if len(thumbnails) > 1 else ''
2871 thumb_display_id = '%s ' % t['id'] if len(thumbnails) > 1 else ''
29f7c58a 2872 t['filename'] = thumb_filename = replace_extension(filename + suffix, thumb_ext, info_dict.get('ext'))
ec82d85a 2873
0c3d0f51 2874 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(thumb_filename)):
0202b52a 2875 ret.append(thumb_filename)
ec82d85a
PH
2876 self.to_screen('[%s] %s: Thumbnail %sis already present' %
2877 (info_dict['extractor'], info_dict['id'], thumb_display_id))
2878 else:
2879 self.to_screen('[%s] %s: Downloading thumbnail %s...' %
2880 (info_dict['extractor'], info_dict['id'], thumb_display_id))
2881 try:
2882 uf = self.urlopen(t['url'])
d3d89c32 2883 with open(encodeFilename(thumb_filename), 'wb') as thumbf:
ec82d85a 2884 shutil.copyfileobj(uf, thumbf)
0202b52a 2885 ret.append(thumb_filename)
ec82d85a
PH
2886 self.to_screen('[%s] %s: Writing thumbnail %sto: %s' %
2887 (info_dict['extractor'], info_dict['id'], thumb_display_id, thumb_filename))
2888 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
2889 self.report_warning('Unable to download thumbnail "%s": %s' %
9b9c5355 2890 (t['url'], error_to_compat_str(err)))
0202b52a 2891 return ret