]> jfr.im git - yt-dlp.git/blame - yt_dlp/YoutubeDL.py
Option to choose different downloader for different protocols
[yt-dlp.git] / yt_dlp / 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 29from string import ascii_letters
e5813e53 30from zipimport import zipimporter
961ea474 31
8c25f81b 32from .compat import (
82d8a8b6 33 compat_basestring,
dca08720 34 compat_cookiejar,
003c69a8 35 compat_get_terminal_size,
ce02ed60 36 compat_http_client,
4f026faf 37 compat_kwargs,
d0d9ade4 38 compat_numeric_types,
e9c0cdd3 39 compat_os_name,
ce02ed60 40 compat_str,
67134eab 41 compat_tokenize_tokenize,
ce02ed60
PH
42 compat_urllib_error,
43 compat_urllib_request,
8b172c2e 44 compat_urllib_request_DataHandler,
8c25f81b
PH
45)
46from .utils import (
eedb7ba5
S
47 age_restricted,
48 args_to_str,
ce02ed60
PH
49 ContentTooShortError,
50 date_from_str,
51 DateRange,
acd69589 52 DEFAULT_OUTTMPL,
de6000d9 53 OUTTMPL_TYPES,
ce02ed60 54 determine_ext,
b5559424 55 determine_protocol,
732044af 56 DOT_DESKTOP_LINK_TEMPLATE,
57 DOT_URL_LINK_TEMPLATE,
58 DOT_WEBLOC_LINK_TEMPLATE,
ce02ed60 59 DownloadError,
c0384f22 60 encode_compat_str,
ce02ed60 61 encodeFilename,
9b9c5355 62 error_to_compat_str,
498f5606 63 EntryNotInPlaylist,
8b0d7497 64 ExistingVideoReached,
590bc6f6 65 expand_path,
ce02ed60 66 ExtractorError,
e29663c6 67 float_or_none,
02dbf93f 68 format_bytes,
76d321f6 69 format_field,
143db31d 70 FORMAT_RE,
525ef922 71 formatSeconds,
773f291d 72 GeoRestrictedError,
c9969434 73 int_or_none,
732044af 74 iri_to_uri,
773f291d 75 ISO3166Utils,
ce02ed60 76 locked_file,
0202b52a 77 make_dir,
dca08720 78 make_HTTPS_handler,
ce02ed60 79 MaxDownloadsReached,
cd6fc19e 80 orderedSet,
b7ab0590 81 PagedList,
083c9df9 82 parse_filesize,
91410c9b 83 PerRequestProxyHandler,
dca08720 84 platform_name,
eedb7ba5 85 PostProcessingError,
ce02ed60 86 preferredencoding,
eedb7ba5 87 prepend_extension,
51fb4995 88 register_socks_protocols,
cfb56d1a 89 render_table,
eedb7ba5 90 replace_extension,
8b0d7497 91 RejectedVideoReached,
ce02ed60
PH
92 SameFileError,
93 sanitize_filename,
1bb5c511 94 sanitize_path,
dcf77cf1 95 sanitize_url,
67dda517 96 sanitized_Request,
e5660ee6 97 std_headers,
1211bb6d 98 str_or_none,
e29663c6 99 strftime_or_none,
ce02ed60 100 subtitles_filename,
732044af 101 to_high_limit_path,
ce02ed60 102 UnavailableVideoError,
29eb5174 103 url_basename,
58b1f00d 104 version_tuple,
ce02ed60
PH
105 write_json_file,
106 write_string,
1bab3437 107 YoutubeDLCookieJar,
6a3f4c3f 108 YoutubeDLCookieProcessor,
dca08720 109 YoutubeDLHandler,
fca6dba8 110 YoutubeDLRedirectHandler,
f5b1bca9 111 process_communicate_or_kill,
ce02ed60 112)
a0e07d31 113from .cache import Cache
52a8a1e1 114from .extractor import (
115 gen_extractor_classes,
116 get_info_extractor,
117 _LAZY_LOADER,
118 _PLUGIN_CLASSES
119)
4c54b89e 120from .extractor.openload import PhantomJSwrapper
52a8a1e1 121from .downloader import (
122 get_suitable_downloader,
123 shorten_protocol_name
124)
4c83c967 125from .downloader.rtmp import rtmpdump_version
4f026faf 126from .postprocessor import (
f17f8651 127 FFmpegFixupM3u8PP,
62cd676c 128 FFmpegFixupM4aPP,
6271f1ca 129 FFmpegFixupStretchedPP,
4f026faf
PH
130 FFmpegMergerPP,
131 FFmpegPostProcessor,
0202b52a 132 # FFmpegSubtitlesConvertorPP,
4f026faf 133 get_postprocessor,
0202b52a 134 MoveFilesAfterDownloadPP,
4f026faf 135)
dca08720 136from .version import __version__
8222d8de 137
e9c0cdd3
YCH
138if compat_os_name == 'nt':
139 import ctypes
140
2459b6e1 141
8222d8de
JMF
142class YoutubeDL(object):
143 """YoutubeDL class.
144
145 YoutubeDL objects are the ones responsible of downloading the
146 actual video file and writing it to disk if the user has requested
147 it, among some other tasks. In most cases there should be one per
148 program. As, given a video URL, the downloader doesn't know how to
149 extract all the needed information, task that InfoExtractors do, it
150 has to pass the URL to one of them.
151
152 For this, YoutubeDL objects have a method that allows
153 InfoExtractors to be registered in a given order. When it is passed
154 a URL, the YoutubeDL object handles it to the first InfoExtractor it
155 finds that reports being able to handle it. The InfoExtractor extracts
156 all the information about the video or videos the URL refers to, and
157 YoutubeDL process the extracted information, possibly using a File
158 Downloader to download the video.
159
160 YoutubeDL objects accept a lot of parameters. In order not to saturate
161 the object constructor with arguments, it receives a dictionary of
162 options instead. These options are available through the params
163 attribute for the InfoExtractors to use. The YoutubeDL also
164 registers itself as the downloader in charge for the InfoExtractors
165 that are added to it, so this is a "mutual registration".
166
167 Available options:
168
169 username: Username for authentication purposes.
170 password: Password for authentication purposes.
180940e0 171 videopassword: Password for accessing a video.
1da50aa3
S
172 ap_mso: Adobe Pass multiple-system operator identifier.
173 ap_username: Multiple-system operator account username.
174 ap_password: Multiple-system operator account password.
8222d8de
JMF
175 usenetrc: Use netrc for authentication instead.
176 verbose: Print additional info to stdout.
177 quiet: Do not print messages to stdout.
ad8915b7 178 no_warnings: Do not print out anything for warnings.
8222d8de
JMF
179 forceurl: Force printing final URL.
180 forcetitle: Force printing title.
181 forceid: Force printing ID.
182 forcethumbnail: Force printing thumbnail URL.
183 forcedescription: Force printing description.
184 forcefilename: Force printing final filename.
525ef922 185 forceduration: Force printing duration.
8694c600 186 forcejson: Force printing info_dict as JSON.
63e0be34
PH
187 dump_single_json: Force printing the info_dict of the whole playlist
188 (or video) as a single JSON line.
c25228e5 189 force_write_download_archive: Force writing download archive regardless
190 of 'skip_download' or 'simulate'.
8222d8de 191 simulate: Do not download the video files.
eb8a4433 192 format: Video format code. see "FORMAT SELECTION" for more details.
63ad4d43 193 allow_unplayable_formats: Allow unplayable formats to be extracted and downloaded.
c25228e5 194 format_sort: How to sort the video formats. see "Sorting Formats"
195 for more details.
196 format_sort_force: Force the given format_sort. see "Sorting Formats"
197 for more details.
198 allow_multiple_video_streams: Allow multiple video streams to be merged
199 into a single file
200 allow_multiple_audio_streams: Allow multiple audio streams to be merged
201 into a single file
4524baf0 202 paths: Dictionary of output paths. The allowed keys are 'home'
203 'temp' and the keys of OUTTMPL_TYPES (in utils.py)
de6000d9 204 outtmpl: Dictionary of templates for output names. Allowed keys
4524baf0 205 are 'default' and the keys of OUTTMPL_TYPES (in utils.py).
206 A string a also accepted for backward compatibility
a820dc72
RA
207 outtmpl_na_placeholder: Placeholder for unavailable meta fields.
208 restrictfilenames: Do not allow "&" and spaces in file names
209 trim_file_name: Limit length of filename (extension excluded)
4524baf0 210 windowsfilenames: Force the filenames to be windows compatible
a820dc72 211 ignoreerrors: Do not stop on download errors
7a5c1cfe 212 (Default True when running yt-dlp,
a820dc72 213 but False when directly accessing YoutubeDL class)
d22dec74 214 force_generic_extractor: Force downloader to use the generic extractor
0c3d0f51 215 overwrites: Overwrite all video and metadata files if True,
216 overwrite only non-video files if None
217 and don't overwrite any file if False
8222d8de
JMF
218 playliststart: Playlist item to start at.
219 playlistend: Playlist item to end at.
c14e88f0 220 playlist_items: Specific indices of playlist to download.
ff815fe6 221 playlistreverse: Download playlist items in reverse order.
75822ca7 222 playlistrandom: Download playlist items in random order.
8222d8de
JMF
223 matchtitle: Download only matching titles.
224 rejecttitle: Reject downloads for matching titles.
8bf9319e 225 logger: Log messages to a logging.Logger instance.
8222d8de
JMF
226 logtostderr: Log messages to stderr instead of stdout.
227 writedescription: Write the video description to a .description file
228 writeinfojson: Write the video description to a .info.json file
75d43ca0 229 clean_infojson: Remove private fields from the infojson
06167fbb 230 writecomments: Extract video comments. This will not be written to disk
231 unless writeinfojson is also given
1fb07d10 232 writeannotations: Write the video annotations to a .annotations.xml file
8222d8de 233 writethumbnail: Write the thumbnail image to a file
c25228e5 234 allow_playlist_files: Whether to write playlists' description, infojson etc
235 also to disk when using the 'write*' options
ec82d85a 236 write_all_thumbnails: Write all thumbnail formats to files
732044af 237 writelink: Write an internet shortcut file, depending on the
238 current platform (.url/.webloc/.desktop)
239 writeurllink: Write a Windows internet shortcut file (.url)
240 writewebloclink: Write a macOS internet shortcut file (.webloc)
241 writedesktoplink: Write a Linux internet shortcut file (.desktop)
8222d8de 242 writesubtitles: Write the video subtitles to a file
741dd8ea 243 writeautomaticsub: Write the automatically generated subtitles to a file
8222d8de 244 allsubtitles: Downloads all the subtitles of the video
0b7f3118 245 (requires writesubtitles or writeautomaticsub)
8222d8de 246 listsubtitles: Lists all available subtitles for the video
a504ced0 247 subtitlesformat: The format code for subtitles
aa6a10c4 248 subtitleslangs: List of languages of the subtitles to download
8222d8de
JMF
249 keepvideo: Keep the video file after post-processing
250 daterange: A DateRange object, download only if the upload_date is in the range.
251 skip_download: Skip the actual download of the video file
c35f9e72 252 cachedir: Location of the cache files in the filesystem.
a0e07d31 253 False to disable filesystem cache.
47192f92 254 noplaylist: Download single video instead of a playlist if in doubt.
8dbe9899
PH
255 age_limit: An integer representing the user's age in years.
256 Unsuitable videos for the given age are skipped.
5fe18bdb
PH
257 min_views: An integer representing the minimum view count the video
258 must have in order to not be skipped.
259 Videos without view count information are always
260 downloaded. None for no limit.
261 max_views: An integer representing the maximum view count.
262 Videos that are more popular than that are not
263 downloaded.
264 Videos without view count information are always
265 downloaded. None for no limit.
266 download_archive: File name of a file where all downloads are recorded.
c1c9a79c
PH
267 Videos already present in the file are not downloaded
268 again.
8a51f564 269 break_on_existing: Stop the download process after attempting to download a
270 file that is in the archive.
271 break_on_reject: Stop the download process when encountering a video that
272 has been filtered out.
273 cookiefile: File name where cookies should be read from and dumped to
a1ee09e8 274 nocheckcertificate:Do not verify SSL certificates
7e8c0af0
PH
275 prefer_insecure: Use HTTP instead of HTTPS to retrieve information.
276 At the moment, this is only supported by YouTube.
a1ee09e8 277 proxy: URL of the proxy server to use
38cce791 278 geo_verification_proxy: URL of the proxy to use for IP address verification
504f20dd 279 on geo-restricted sites.
e344693b 280 socket_timeout: Time to wait for unresponsive hosts, in seconds
0783b09b
PH
281 bidi_workaround: Work around buggy terminals without bidirectional text
282 support, using fridibi
a0ddb8a2 283 debug_printtraffic:Print out sent and received HTTP traffic
7b0817e8 284 include_ads: Download ads as well
04b4d394
PH
285 default_search: Prepend this string if an input url is not valid.
286 'auto' for elaborate guessing
62fec3b2 287 encoding: Use this encoding instead of the system-specified.
e8ee972c 288 extract_flat: Do not resolve URLs, return the immediate result.
057a5206
PH
289 Pass in 'in_playlist' to only show this behavior for
290 playlist items.
4f026faf 291 postprocessors: A list of dictionaries, each with an entry
71b640cc 292 * key: The name of the postprocessor. See
7a5c1cfe 293 yt_dlp/postprocessor/__init__.py for a list.
0202b52a 294 * _after_move: Optional. If True, run this post_processor
295 after 'MoveFilesAfterDownload'
4f026faf
PH
296 as well as any further keyword arguments for the
297 postprocessor.
ab8e5e51
AM
298 post_hooks: A list of functions that get called as the final step
299 for each video file, after all postprocessors have been
300 called. The filename will be passed as the only argument.
71b640cc
PH
301 progress_hooks: A list of functions that get called on download
302 progress, with a dictionary with the entries
5cda4eda 303 * status: One of "downloading", "error", or "finished".
ee69b99a 304 Check this first and ignore unknown values.
71b640cc 305
5cda4eda 306 If status is one of "downloading", or "finished", the
ee69b99a
PH
307 following properties may also be present:
308 * filename: The final filename (always present)
5cda4eda 309 * tmpfilename: The filename we're currently writing to
71b640cc
PH
310 * downloaded_bytes: Bytes on disk
311 * total_bytes: Size of the whole file, None if unknown
5cda4eda
PH
312 * total_bytes_estimate: Guess of the eventual file size,
313 None if unavailable.
314 * elapsed: The number of seconds since download started.
71b640cc
PH
315 * eta: The estimated time in seconds, None if unknown
316 * speed: The download speed in bytes/second, None if
317 unknown
5cda4eda
PH
318 * fragment_index: The counter of the currently
319 downloaded video fragment.
320 * fragment_count: The number of fragments (= individual
321 files that will be merged)
71b640cc
PH
322
323 Progress hooks are guaranteed to be called at least once
324 (with status "finished") if the download is successful.
45598f15 325 merge_output_format: Extension to use when merging formats.
6b591b29 326 final_ext: Expected final extension; used to detect when the file was
327 already downloaded and converted. "merge_output_format" is
328 replaced by this extension when given
6271f1ca
PH
329 fixup: Automatically correct known faults of the file.
330 One of:
331 - "never": do nothing
332 - "warn": only emit a warning
333 - "detect_or_warn": check whether we can do anything
62cd676c 334 about it, warn otherwise (default)
504f20dd 335 source_address: Client-side IP address to bind to.
6ec6cb4e 336 call_home: Boolean, true iff we are allowed to contact the
7a5c1cfe 337 yt-dlp servers for debugging. (BROKEN)
1cf376f5 338 sleep_interval_requests: Number of seconds to sleep between requests
339 during extraction
7aa589a5
S
340 sleep_interval: Number of seconds to sleep before each download when
341 used alone or a lower bound of a range for randomized
342 sleep before each download (minimum possible number
343 of seconds to sleep) when used along with
344 max_sleep_interval.
345 max_sleep_interval:Upper bound of a range for randomized sleep before each
346 download (maximum possible number of seconds to sleep).
347 Must only be used along with sleep_interval.
348 Actual sleep time will be a random float from range
349 [sleep_interval; max_sleep_interval].
1cf376f5 350 sleep_interval_subtitles: Number of seconds to sleep before each subtitle download
cfb56d1a
PH
351 listformats: Print an overview of available video formats and exit.
352 list_thumbnails: Print a table of all thumbnails and exit.
347de493
PH
353 match_filter: A function that gets called with the info_dict of
354 every video.
355 If it returns a message, the video is ignored.
356 If it returns None, the video is downloaded.
357 match_filter_func in utils.py is one example for this.
7e5db8c9 358 no_color: Do not emit color codes in output.
0a840f58 359 geo_bypass: Bypass geographic restriction via faking X-Forwarded-For
504f20dd 360 HTTP header
0a840f58 361 geo_bypass_country:
773f291d
S
362 Two-letter ISO 3166-2 country code that will be used for
363 explicit geographic restriction bypassing via faking
504f20dd 364 X-Forwarded-For HTTP header
5f95927a
S
365 geo_bypass_ip_block:
366 IP range in CIDR notation that will be used similarly to
504f20dd 367 geo_bypass_country
71b640cc 368
85729c51 369 The following options determine which downloader is picked:
52a8a1e1 370 external_downloader: A dictionary of protocol keys and the executable of the
371 external downloader to use for it. The allowed protocols
372 are default|http|ftp|m3u8|dash|rtsp|rtmp|mms.
373 Set the value to 'native' to use the native downloader
374 hls_prefer_native: Deprecated - Use external_downloader = {'m3u8': 'native'}
375 or {'m3u8': 'ffmpeg'} instead.
376 Use the native HLS downloader instead of ffmpeg/avconv
bf09af3a
S
377 if True, otherwise use ffmpeg/avconv if False, otherwise
378 use downloader suggested by extractor if None.
fe7e0c98 379
8222d8de 380 The following parameters are not used by YoutubeDL itself, they are used by
7a5c1cfe 381 the downloader (see yt_dlp/downloader/common.py):
8222d8de 382 nopart, updatetime, buffersize, ratelimit, min_filesize, max_filesize, test,
881e6a1f 383 noresizebuffer, retries, continuedl, noprogress, consoletitle,
b54d4a5c 384 xattr_set_filesize, external_downloader_args, hls_use_mpegts,
e409895f 385 http_chunk_size.
76b1bd67
JMF
386
387 The following options are used by the post processors:
d4a24f40 388 prefer_ffmpeg: If False, use avconv instead of ffmpeg if both are available,
e4172ac9 389 otherwise prefer ffmpeg. (avconv support is deprecated)
c0b7d117
S
390 ffmpeg_location: Location of the ffmpeg/avconv binary; either the path
391 to the binary or its containing directory.
43820c03 392 postprocessor_args: A dictionary of postprocessor/executable keys (in lower case)
393 and a list of additional command-line arguments for the
394 postprocessor/executable. The dict can also have "PP+EXE" keys
395 which are used when the given exe is used by the given PP.
396 Use 'default' as the name for arguments to passed to all PP
e409895f 397
398 The following options are used by the extractors:
62bff2c1 399 extractor_retries: Number of times to retry for known errors
400 dynamic_mpd: Whether to process dynamic DASH manifests (default: True)
e409895f 401 hls_split_discontinuity: Split HLS playlists to different formats at
62bff2c1 402 discontinuities such as ad breaks (default: False)
3600fd59 403 youtube_include_dash_manifest: If True (default), DASH manifests and related
62bff2c1 404 data will be downloaded and processed by extractor.
405 You can reduce network I/O by disabling it if you don't
406 care about DASH. (only for youtube)
e409895f 407 youtube_include_hls_manifest: If True (default), HLS manifests and related
62bff2c1 408 data will be downloaded and processed by extractor.
409 You can reduce network I/O by disabling it if you don't
410 care about HLS. (only for youtube)
8222d8de
JMF
411 """
412
c9969434
S
413 _NUMERIC_FIELDS = set((
414 'width', 'height', 'tbr', 'abr', 'asr', 'vbr', 'fps', 'filesize', 'filesize_approx',
415 'timestamp', 'upload_year', 'upload_month', 'upload_day',
416 'duration', 'view_count', 'like_count', 'dislike_count', 'repost_count',
417 'average_rating', 'comment_count', 'age_limit',
418 'start_time', 'end_time',
419 'chapter_number', 'season_number', 'episode_number',
420 'track_number', 'disc_number', 'release_year',
421 'playlist_index',
422 ))
423
8222d8de
JMF
424 params = None
425 _ies = []
5bfa4862 426 _pps = {'beforedl': [], 'aftermove': [], 'normal': []}
0202b52a 427 __prepare_filename_warned = False
1cf376f5 428 _first_webpage_request = True
8222d8de
JMF
429 _download_retcode = None
430 _num_downloads = None
30a074c2 431 _playlist_level = 0
432 _playlist_urls = set()
8222d8de
JMF
433 _screen_file = None
434
3511266b 435 def __init__(self, params=None, auto_init=True):
8222d8de 436 """Create a FileDownloader object with the given options."""
e9f9a10f
JMF
437 if params is None:
438 params = {}
8222d8de 439 self._ies = []
56c73665 440 self._ies_instances = {}
5bfa4862 441 self._pps = {'beforedl': [], 'aftermove': [], 'normal': []}
0202b52a 442 self.__prepare_filename_warned = False
1cf376f5 443 self._first_webpage_request = True
ab8e5e51 444 self._post_hooks = []
933605d7 445 self._progress_hooks = []
8222d8de
JMF
446 self._download_retcode = 0
447 self._num_downloads = 0
448 self._screen_file = [sys.stdout, sys.stderr][params.get('logtostderr', False)]
0783b09b 449 self._err_file = sys.stderr
4abf617b
S
450 self.params = {
451 # Default parameters
452 'nocheckcertificate': False,
453 }
454 self.params.update(params)
a0e07d31 455 self.cache = Cache(self)
a45e8619 456 self.archive = set()
ecdec191
JB
457
458 """Preload the archive, if any is specified"""
459 def preload_download_archive(self):
460 fn = self.params.get('download_archive')
461 if fn is None:
462 return False
463 try:
464 with locked_file(fn, 'r', encoding='utf-8') as archive_file:
465 for line in archive_file:
a45e8619 466 self.archive.add(line.strip())
ecdec191
JB
467 except IOError as ioe:
468 if ioe.errno != errno.ENOENT:
469 raise
1d74d8d9 470 return False
ecdec191 471 return True
34308b30 472
be5df5ee
S
473 def check_deprecated(param, option, suggestion):
474 if self.params.get(param) is not None:
475 self.report_warning(
476 '%s is deprecated. Use %s instead.' % (option, suggestion))
477 return True
478 return False
479
1de7ea76
JB
480 if self.params.get('verbose'):
481 self.to_stdout('[debug] Loading archive file %r' % self.params.get('download_archive'))
482
ecdec191
JB
483 preload_download_archive(self)
484
be5df5ee 485 if check_deprecated('cn_verification_proxy', '--cn-verification-proxy', '--geo-verification-proxy'):
38cce791
YCH
486 if self.params.get('geo_verification_proxy') is None:
487 self.params['geo_verification_proxy'] = self.params['cn_verification_proxy']
488
6b591b29 489 if self.params.get('final_ext'):
490 if self.params.get('merge_output_format'):
491 self.report_warning('--merge-output-format will be ignored since --remux-video or --recode-video is given')
492 self.params['merge_output_format'] = self.params['final_ext']
493
b9d973be 494 if 'overwrites' in self.params and self.params['overwrites'] is None:
495 del self.params['overwrites']
496
be5df5ee
S
497 check_deprecated('autonumber_size', '--autonumber-size', 'output template with %(autonumber)0Nd, where N in the number of digits')
498 check_deprecated('autonumber', '--auto-number', '-o "%(autonumber)s-%(title)s.%(ext)s"')
499 check_deprecated('usetitle', '--title', '-o "%(title)s-%(id)s.%(ext)s"')
500
0783b09b 501 if params.get('bidi_workaround', False):
1c088fa8
PH
502 try:
503 import pty
504 master, slave = pty.openpty()
003c69a8 505 width = compat_get_terminal_size().columns
1c088fa8
PH
506 if width is None:
507 width_args = []
508 else:
509 width_args = ['-w', str(width)]
5d681e96 510 sp_kwargs = dict(
1c088fa8
PH
511 stdin=subprocess.PIPE,
512 stdout=slave,
513 stderr=self._err_file)
5d681e96
PH
514 try:
515 self._output_process = subprocess.Popen(
516 ['bidiv'] + width_args, **sp_kwargs
517 )
518 except OSError:
5d681e96
PH
519 self._output_process = subprocess.Popen(
520 ['fribidi', '-c', 'UTF-8'] + width_args, **sp_kwargs)
521 self._output_channel = os.fdopen(master, 'rb')
1c088fa8 522 except OSError as ose:
66e7ace1 523 if ose.errno == errno.ENOENT:
6febd1c1 524 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 $PATH.')
1c088fa8
PH
525 else:
526 raise
0783b09b 527
3089bc74
S
528 if (sys.platform != 'win32'
529 and sys.getfilesystemencoding() in ['ascii', 'ANSI_X3.4-1968']
530 and not params.get('restrictfilenames', False)):
e9137224 531 # Unicode filesystem API will throw errors (#1474, #13027)
34308b30 532 self.report_warning(
6febd1c1 533 'Assuming --restrict-filenames since file system encoding '
1b725173 534 'cannot encode all characters. '
6febd1c1 535 'Set the LC_ALL environment variable to fix this.')
4a98cdbf 536 self.params['restrictfilenames'] = True
34308b30 537
de6000d9 538 self.outtmpl_dict = self.parse_outtmpl()
486dd09e 539
dca08720
PH
540 self._setup_opener()
541
3511266b
PH
542 if auto_init:
543 self.print_debug_header()
544 self.add_default_info_extractors()
545
4f026faf
PH
546 for pp_def_raw in self.params.get('postprocessors', []):
547 pp_class = get_postprocessor(pp_def_raw['key'])
548 pp_def = dict(pp_def_raw)
549 del pp_def['key']
5bfa4862 550 if 'when' in pp_def:
551 when = pp_def['when']
552 del pp_def['when']
553 else:
554 when = 'normal'
4f026faf 555 pp = pp_class(self, **compat_kwargs(pp_def))
5bfa4862 556 self.add_post_processor(pp, when=when)
4f026faf 557
ab8e5e51
AM
558 for ph in self.params.get('post_hooks', []):
559 self.add_post_hook(ph)
560
71b640cc
PH
561 for ph in self.params.get('progress_hooks', []):
562 self.add_progress_hook(ph)
563
51fb4995
YCH
564 register_socks_protocols()
565
7d4111ed
PH
566 def warn_if_short_id(self, argv):
567 # short YouTube ID starting with dash?
568 idxs = [
569 i for i, a in enumerate(argv)
570 if re.match(r'^-[0-9A-Za-z_-]{10}$', a)]
571 if idxs:
572 correct_argv = (
7a5c1cfe 573 ['yt-dlp']
3089bc74
S
574 + [a for i, a in enumerate(argv) if i not in idxs]
575 + ['--'] + [argv[i] for i in idxs]
7d4111ed
PH
576 )
577 self.report_warning(
578 'Long argument string detected. '
579 'Use -- to separate parameters and URLs, like this:\n%s\n' %
580 args_to_str(correct_argv))
581
8222d8de
JMF
582 def add_info_extractor(self, ie):
583 """Add an InfoExtractor object to the end of the list."""
584 self._ies.append(ie)
e52d7f85
JMF
585 if not isinstance(ie, type):
586 self._ies_instances[ie.ie_key()] = ie
587 ie.set_downloader(self)
8222d8de 588
56c73665
JMF
589 def get_info_extractor(self, ie_key):
590 """
591 Get an instance of an IE with name ie_key, it will try to get one from
592 the _ies list, if there's no instance it will create a new one and add
593 it to the extractor list.
594 """
595 ie = self._ies_instances.get(ie_key)
596 if ie is None:
597 ie = get_info_extractor(ie_key)()
598 self.add_info_extractor(ie)
599 return ie
600
023fa8c4
JMF
601 def add_default_info_extractors(self):
602 """
603 Add the InfoExtractors returned by gen_extractors to the end of the list
604 """
e52d7f85 605 for ie in gen_extractor_classes():
023fa8c4
JMF
606 self.add_info_extractor(ie)
607
5bfa4862 608 def add_post_processor(self, pp, when='normal'):
8222d8de 609 """Add a PostProcessor object to the end of the chain."""
5bfa4862 610 self._pps[when].append(pp)
8222d8de
JMF
611 pp.set_downloader(self)
612
ab8e5e51
AM
613 def add_post_hook(self, ph):
614 """Add the post hook"""
615 self._post_hooks.append(ph)
616
933605d7
JMF
617 def add_progress_hook(self, ph):
618 """Add the progress hook (currently only for the file downloader)"""
619 self._progress_hooks.append(ph)
8ab470f1 620
1c088fa8 621 def _bidi_workaround(self, message):
5d681e96 622 if not hasattr(self, '_output_channel'):
1c088fa8
PH
623 return message
624
5d681e96 625 assert hasattr(self, '_output_process')
11b85ce6 626 assert isinstance(message, compat_str)
6febd1c1
PH
627 line_count = message.count('\n') + 1
628 self._output_process.stdin.write((message + '\n').encode('utf-8'))
5d681e96 629 self._output_process.stdin.flush()
6febd1c1 630 res = ''.join(self._output_channel.readline().decode('utf-8')
9e1a5b84 631 for _ in range(line_count))
6febd1c1 632 return res[:-len('\n')]
1c088fa8 633
8222d8de 634 def to_screen(self, message, skip_eol=False):
0783b09b
PH
635 """Print message to stdout if not in quiet mode."""
636 return self.to_stdout(message, skip_eol, check_quiet=True)
637
734f90bb 638 def _write_string(self, s, out=None):
b58ddb32 639 write_string(s, out=out, encoding=self.params.get('encoding'))
734f90bb 640
0783b09b 641 def to_stdout(self, message, skip_eol=False, check_quiet=False):
8222d8de 642 """Print message to stdout if not in quiet mode."""
8bf9319e 643 if self.params.get('logger'):
43afe285 644 self.params['logger'].debug(message)
0783b09b 645 elif not check_quiet or not self.params.get('quiet', False):
1c088fa8 646 message = self._bidi_workaround(message)
6febd1c1 647 terminator = ['\n', ''][skip_eol]
8222d8de 648 output = message + terminator
1c088fa8 649
734f90bb 650 self._write_string(output, self._screen_file)
8222d8de
JMF
651
652 def to_stderr(self, message):
653 """Print message to stderr."""
11b85ce6 654 assert isinstance(message, compat_str)
8bf9319e 655 if self.params.get('logger'):
43afe285
IB
656 self.params['logger'].error(message)
657 else:
1c088fa8 658 message = self._bidi_workaround(message)
6febd1c1 659 output = message + '\n'
734f90bb 660 self._write_string(output, self._err_file)
8222d8de 661
1e5b9a95
PH
662 def to_console_title(self, message):
663 if not self.params.get('consoletitle', False):
664 return
4bede0d8
C
665 if compat_os_name == 'nt':
666 if ctypes.windll.kernel32.GetConsoleWindow():
667 # c_wchar_p() might not be necessary if `message` is
668 # already of type unicode()
669 ctypes.windll.kernel32.SetConsoleTitleW(ctypes.c_wchar_p(message))
1e5b9a95 670 elif 'TERM' in os.environ:
b46696bd 671 self._write_string('\033]0;%s\007' % message, self._screen_file)
1e5b9a95 672
bdde425c
PH
673 def save_console_title(self):
674 if not self.params.get('consoletitle', False):
675 return
94c3442e
S
676 if self.params.get('simulate', False):
677 return
4bede0d8 678 if compat_os_name != 'nt' and 'TERM' in os.environ:
efd6c574 679 # Save the title on stack
734f90bb 680 self._write_string('\033[22;0t', self._screen_file)
bdde425c
PH
681
682 def restore_console_title(self):
683 if not self.params.get('consoletitle', False):
684 return
94c3442e
S
685 if self.params.get('simulate', False):
686 return
4bede0d8 687 if compat_os_name != 'nt' and 'TERM' in os.environ:
efd6c574 688 # Restore the title from stack
734f90bb 689 self._write_string('\033[23;0t', self._screen_file)
bdde425c
PH
690
691 def __enter__(self):
692 self.save_console_title()
693 return self
694
695 def __exit__(self, *args):
696 self.restore_console_title()
f89197d7 697
dca08720 698 if self.params.get('cookiefile') is not None:
1bab3437 699 self.cookiejar.save(ignore_discard=True, ignore_expires=True)
bdde425c 700
8222d8de
JMF
701 def trouble(self, message=None, tb=None):
702 """Determine action to take when a download problem appears.
703
704 Depending on if the downloader has been configured to ignore
705 download errors or not, this method may throw an exception or
706 not when errors are found, after printing the message.
707
708 tb, if given, is additional traceback information.
709 """
710 if message is not None:
711 self.to_stderr(message)
712 if self.params.get('verbose'):
713 if tb is None:
714 if sys.exc_info()[0]: # if .trouble has been called from an except block
6febd1c1 715 tb = ''
8222d8de 716 if hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
6febd1c1 717 tb += ''.join(traceback.format_exception(*sys.exc_info()[1].exc_info))
c0384f22 718 tb += encode_compat_str(traceback.format_exc())
8222d8de
JMF
719 else:
720 tb_data = traceback.format_list(traceback.extract_stack())
6febd1c1 721 tb = ''.join(tb_data)
8222d8de
JMF
722 self.to_stderr(tb)
723 if not self.params.get('ignoreerrors', False):
724 if sys.exc_info()[0] and hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
725 exc_info = sys.exc_info()[1].exc_info
726 else:
727 exc_info = sys.exc_info()
728 raise DownloadError(message, exc_info)
729 self._download_retcode = 1
730
731 def report_warning(self, message):
732 '''
733 Print the message to stderr, it will be prefixed with 'WARNING:'
734 If stderr is a tty file the 'WARNING:' will be colored
735 '''
6d07ce01
JMF
736 if self.params.get('logger') is not None:
737 self.params['logger'].warning(message)
8222d8de 738 else:
ad8915b7
PH
739 if self.params.get('no_warnings'):
740 return
e9c0cdd3 741 if not self.params.get('no_color') and self._err_file.isatty() and compat_os_name != 'nt':
6d07ce01
JMF
742 _msg_header = '\033[0;33mWARNING:\033[0m'
743 else:
744 _msg_header = 'WARNING:'
745 warning_message = '%s %s' % (_msg_header, message)
746 self.to_stderr(warning_message)
8222d8de
JMF
747
748 def report_error(self, message, tb=None):
749 '''
750 Do the same as trouble, but prefixes the message with 'ERROR:', colored
751 in red if stderr is a tty file.
752 '''
e9c0cdd3 753 if not self.params.get('no_color') and self._err_file.isatty() and compat_os_name != 'nt':
6febd1c1 754 _msg_header = '\033[0;31mERROR:\033[0m'
8222d8de 755 else:
6febd1c1
PH
756 _msg_header = 'ERROR:'
757 error_message = '%s %s' % (_msg_header, message)
8222d8de
JMF
758 self.trouble(error_message, tb)
759
8222d8de
JMF
760 def report_file_already_downloaded(self, file_name):
761 """Report file has already been fully downloaded."""
762 try:
6febd1c1 763 self.to_screen('[download] %s has already been downloaded' % file_name)
ce02ed60 764 except UnicodeEncodeError:
6febd1c1 765 self.to_screen('[download] The file has already been downloaded')
8222d8de 766
0c3d0f51 767 def report_file_delete(self, file_name):
768 """Report that existing file will be deleted."""
769 try:
c25228e5 770 self.to_screen('Deleting existing file %s' % file_name)
0c3d0f51 771 except UnicodeEncodeError:
c25228e5 772 self.to_screen('Deleting existing file')
0c3d0f51 773
de6000d9 774 def parse_outtmpl(self):
775 outtmpl_dict = self.params.get('outtmpl', {})
776 if not isinstance(outtmpl_dict, dict):
777 outtmpl_dict = {'default': outtmpl_dict}
778 outtmpl_dict.update({
779 k: v for k, v in DEFAULT_OUTTMPL.items()
780 if not outtmpl_dict.get(k)})
781 for key, val in outtmpl_dict.items():
782 if isinstance(val, bytes):
783 self.report_warning(
784 'Parameter outtmpl is bytes, but should be a unicode string. '
785 'Put from __future__ import unicode_literals at the top of your code file or consider switching to Python 3.x.')
786 return outtmpl_dict
787
143db31d 788 def prepare_outtmpl(self, outtmpl, info_dict, sanitize=None):
789 """ Make the template and info_dict suitable for substitution (outtmpl % info_dict)"""
790 template_dict = dict(info_dict)
791
792 # duration_string
793 template_dict['duration_string'] = ( # %(duration>%H-%M-%S)s is wrong if duration > 24hrs
794 formatSeconds(info_dict['duration'], '-')
795 if info_dict.get('duration', None) is not None
796 else None)
797
798 # epoch
799 template_dict['epoch'] = int(time.time())
800
801 # autonumber
802 autonumber_size = self.params.get('autonumber_size')
803 if autonumber_size is None:
804 autonumber_size = 5
805 template_dict['autonumber'] = self.params.get('autonumber_start', 1) - 1 + self._num_downloads
806
807 # resolution if not defined
808 if template_dict.get('resolution') is None:
809 if template_dict.get('width') and template_dict.get('height'):
810 template_dict['resolution'] = '%dx%d' % (template_dict['width'], template_dict['height'])
811 elif template_dict.get('height'):
812 template_dict['resolution'] = '%sp' % template_dict['height']
813 elif template_dict.get('width'):
814 template_dict['resolution'] = '%dx?' % template_dict['width']
815
816 if sanitize is None:
817 sanitize = lambda k, v: v
818 template_dict = dict((k, v if isinstance(v, compat_numeric_types) else sanitize(k, v))
819 for k, v in template_dict.items()
820 if v is not None and not isinstance(v, (list, tuple, dict)))
821 na = self.params.get('outtmpl_na_placeholder', 'NA')
822 template_dict = collections.defaultdict(lambda: na, template_dict)
823
824 # For fields playlist_index and autonumber convert all occurrences
825 # of %(field)s to %(field)0Nd for backward compatibility
826 field_size_compat_map = {
827 'playlist_index': len(str(template_dict['n_entries'])),
828 'autonumber': autonumber_size,
829 }
830 FIELD_SIZE_COMPAT_RE = r'(?<!%)%\((?P<field>autonumber|playlist_index)\)s'
831 mobj = re.search(FIELD_SIZE_COMPAT_RE, outtmpl)
832 if mobj:
833 outtmpl = re.sub(
834 FIELD_SIZE_COMPAT_RE,
835 r'%%(\1)0%dd' % field_size_compat_map[mobj.group('field')],
836 outtmpl)
837
838 numeric_fields = list(self._NUMERIC_FIELDS)
839
840 # Format date
841 FORMAT_DATE_RE = FORMAT_RE.format(r'(?P<key>(?P<field>\w+)>(?P<format>.+?))')
842 for mobj in re.finditer(FORMAT_DATE_RE, outtmpl):
843 conv_type, field, frmt, key = mobj.group('type', 'field', 'format', 'key')
844 if key in template_dict:
845 continue
846 value = strftime_or_none(template_dict.get(field), frmt, na)
847 if conv_type in 'crs': # string
848 value = sanitize(field, value)
849 else: # number
850 numeric_fields.append(key)
851 value = float_or_none(value, default=None)
852 if value is not None:
853 template_dict[key] = value
854
855 # Missing numeric fields used together with integer presentation types
856 # in format specification will break the argument substitution since
857 # string NA placeholder is returned for missing fields. We will patch
858 # output template for missing fields to meet string presentation type.
859 for numeric_field in numeric_fields:
860 if numeric_field not in template_dict:
861 outtmpl = re.sub(
862 FORMAT_RE.format(re.escape(numeric_field)),
863 r'%({0})s'.format(numeric_field), outtmpl)
864
865 return outtmpl, template_dict
866
de6000d9 867 def _prepare_filename(self, info_dict, tmpl_type='default'):
8222d8de 868 try:
586a91b6 869 sanitize = lambda k, v: sanitize_filename(
45598aab 870 compat_str(v),
1bb5c511 871 restricted=self.params.get('restrictfilenames'),
40df485f 872 is_id=(k == 'id' or k.endswith('_id')))
de6000d9 873 outtmpl = self.outtmpl_dict.get(tmpl_type, self.outtmpl_dict['default'])
143db31d 874 outtmpl, template_dict = self.prepare_outtmpl(outtmpl, info_dict, sanitize)
d0d9ade4 875
15da37c7
S
876 # expand_path translates '%%' into '%' and '$$' into '$'
877 # correspondingly that is not what we want since we need to keep
878 # '%%' intact for template dict substitution step. Working around
879 # with boundary-alike separator hack.
961ea474 880 sep = ''.join([random.choice(ascii_letters) for _ in range(32)])
15da37c7
S
881 outtmpl = outtmpl.replace('%%', '%{0}%'.format(sep)).replace('$$', '${0}$'.format(sep))
882
883 # outtmpl should be expand_path'ed before template dict substitution
884 # because meta fields may contain env variables we don't want to
885 # be expanded. For example, for outtmpl "%(title)s.%(ext)s" and
886 # title "Hello $PATH", we don't want `$PATH` to be expanded.
887 filename = expand_path(outtmpl).replace(sep, '') % template_dict
888
143db31d 889 force_ext = OUTTMPL_TYPES.get(tmpl_type)
de6000d9 890 if force_ext is not None:
891 filename = replace_extension(filename, force_ext, template_dict.get('ext'))
892
bdc3fd2f
U
893 # https://github.com/blackjack4494/youtube-dlc/issues/85
894 trim_file_name = self.params.get('trim_file_name', False)
895 if trim_file_name:
896 fn_groups = filename.rsplit('.')
897 ext = fn_groups[-1]
898 sub_ext = ''
899 if len(fn_groups) > 2:
900 sub_ext = fn_groups[-2]
901 filename = '.'.join(filter(None, [fn_groups[0][:trim_file_name], sub_ext, ext]))
902
0202b52a 903 return filename
8222d8de 904 except ValueError as err:
6febd1c1 905 self.report_error('Error in output template: ' + str(err) + ' (encoding: ' + repr(preferredencoding()) + ')')
8222d8de
JMF
906 return None
907
de6000d9 908 def prepare_filename(self, info_dict, dir_type='', warn=False):
909 """Generate the output filename."""
0202b52a 910 paths = self.params.get('paths', {})
911 assert isinstance(paths, dict)
de6000d9 912 filename = self._prepare_filename(info_dict, dir_type or 'default')
913
914 if warn and not self.__prepare_filename_warned:
915 if not paths:
916 pass
917 elif filename == '-':
918 self.report_warning('--paths is ignored when an outputting to stdout')
919 elif os.path.isabs(filename):
920 self.report_warning('--paths is ignored since an absolute path is given in output template')
921 self.__prepare_filename_warned = True
922 if filename == '-' or not filename:
923 return filename
924
0202b52a 925 homepath = expand_path(paths.get('home', '').strip())
926 assert isinstance(homepath, compat_str)
927 subdir = expand_path(paths.get(dir_type, '').strip()) if dir_type else ''
928 assert isinstance(subdir, compat_str)
c2934512 929 path = os.path.join(homepath, subdir, filename)
930
931 # Temporary fix for #4787
932 # 'Treat' all problem characters by passing filename through preferredencoding
933 # to workaround encoding issues with subprocess on python2 @ Windows
934 if sys.version_info < (3, 0) and sys.platform == 'win32':
935 path = encodeFilename(path, True).decode(preferredencoding())
936 return sanitize_path(path, force=self.params.get('windowsfilenames'))
0202b52a 937
442c37b7 938 def _match_entry(self, info_dict, incomplete):
ecdec191 939 """ Returns None if the file should be downloaded """
8222d8de 940
8b0d7497 941 def check_filter():
942 video_title = info_dict.get('title', info_dict.get('id', 'video'))
943 if 'title' in info_dict:
944 # This can happen when we're just evaluating the playlist
945 title = info_dict['title']
946 matchtitle = self.params.get('matchtitle', False)
947 if matchtitle:
948 if not re.search(matchtitle, title, re.IGNORECASE):
949 return '"' + title + '" title did not match pattern "' + matchtitle + '"'
950 rejecttitle = self.params.get('rejecttitle', False)
951 if rejecttitle:
952 if re.search(rejecttitle, title, re.IGNORECASE):
953 return '"' + title + '" title matched reject pattern "' + rejecttitle + '"'
954 date = info_dict.get('upload_date')
955 if date is not None:
956 dateRange = self.params.get('daterange', DateRange())
957 if date not in dateRange:
958 return '%s upload date is not in range %s' % (date_from_str(date).isoformat(), dateRange)
959 view_count = info_dict.get('view_count')
960 if view_count is not None:
961 min_views = self.params.get('min_views')
962 if min_views is not None and view_count < min_views:
963 return 'Skipping %s, because it has not reached minimum view count (%d/%d)' % (video_title, view_count, min_views)
964 max_views = self.params.get('max_views')
965 if max_views is not None and view_count > max_views:
966 return 'Skipping %s, because it has exceeded the maximum view count (%d/%d)' % (video_title, view_count, max_views)
967 if age_restricted(info_dict.get('age_limit'), self.params.get('age_limit')):
968 return 'Skipping "%s" because it is age restricted' % video_title
969 if self.in_download_archive(info_dict):
970 return '%s has already been recorded in archive' % video_title
971
972 if not incomplete:
973 match_filter = self.params.get('match_filter')
974 if match_filter is not None:
975 ret = match_filter(info_dict)
976 if ret is not None:
977 return ret
978 return None
979
980 reason = check_filter()
981 if reason is not None:
982 self.to_screen('[download] ' + reason)
d83cb531 983 if reason.endswith('has already been recorded in the archive') and self.params.get('break_on_existing', False):
8b0d7497 984 raise ExistingVideoReached()
d83cb531 985 elif self.params.get('break_on_reject', False):
8b0d7497 986 raise RejectedVideoReached()
987 return reason
fe7e0c98 988
b6c45014
JMF
989 @staticmethod
990 def add_extra_info(info_dict, extra_info):
991 '''Set the keys from extra_info in info dict if they are missing'''
992 for key, value in extra_info.items():
993 info_dict.setdefault(key, value)
994
0704d222 995 def extract_info(self, url, download=True, ie_key=None, info_dict=None, extra_info={},
61aa5ba3 996 process=True, force_generic_extractor=False):
8222d8de
JMF
997 '''
998 Returns a list with a dictionary for each video we find.
999 If 'download', also downloads the videos.
1000 extra_info is a dict containing the extra values to add to each result
613b2d9d 1001 '''
fe7e0c98 1002
61aa5ba3 1003 if not ie_key and force_generic_extractor:
d22dec74
S
1004 ie_key = 'Generic'
1005
8222d8de 1006 if ie_key:
56c73665 1007 ies = [self.get_info_extractor(ie_key)]
8222d8de
JMF
1008 else:
1009 ies = self._ies
1010
1011 for ie in ies:
1012 if not ie.suitable(url):
1013 continue
1014
9a68de12 1015 ie_key = ie.ie_key()
1016 ie = self.get_info_extractor(ie_key)
8222d8de 1017 if not ie.working():
6febd1c1
PH
1018 self.report_warning('The program functionality for this site has been marked as broken, '
1019 'and will probably not work.')
8222d8de
JMF
1020
1021 try:
d0757229 1022 temp_id = str_or_none(
63be1aab 1023 ie.extract_id(url) if callable(getattr(ie, 'extract_id', None))
1024 else ie._match_id(url))
a0566bbf 1025 except (AssertionError, IndexError, AttributeError):
1026 temp_id = None
1027 if temp_id is not None and self.in_download_archive({'id': temp_id, 'ie_key': ie_key}):
1028 self.to_screen("[%s] %s: has already been recorded in archive" % (
1029 ie_key, temp_id))
1030 break
a0566bbf 1031 return self.__extract_info(url, ie, download, extra_info, process, info_dict)
a0566bbf 1032 else:
1033 self.report_error('no suitable InfoExtractor for URL %s' % url)
1034
1035 def __handle_extraction_exceptions(func):
1036 def wrapper(self, *args, **kwargs):
1037 try:
1038 return func(self, *args, **kwargs)
773f291d
S
1039 except GeoRestrictedError as e:
1040 msg = e.msg
1041 if e.countries:
1042 msg += '\nThis video is available in %s.' % ', '.join(
1043 map(ISO3166Utils.short2full, e.countries))
1044 msg += '\nYou might want to use a VPN or a proxy server (with --proxy) to workaround.'
1045 self.report_error(msg)
fb043a6e 1046 except ExtractorError as e: # An error we somewhat expected
2c74e6fa 1047 self.report_error(compat_str(e), e.format_traceback())
8b0d7497 1048 except (MaxDownloadsReached, ExistingVideoReached, RejectedVideoReached):
d3e5bbf4 1049 raise
8222d8de
JMF
1050 except Exception as e:
1051 if self.params.get('ignoreerrors', False):
9b9c5355 1052 self.report_error(error_to_compat_str(e), tb=encode_compat_str(traceback.format_exc()))
8222d8de
JMF
1053 else:
1054 raise
a0566bbf 1055 return wrapper
1056
1057 @__handle_extraction_exceptions
1058 def __extract_info(self, url, ie, download, extra_info, process, info_dict):
1059 ie_result = ie.extract(url)
1060 if ie_result is None: # Finished already (backwards compatibility; listformats and friends should be moved here)
1061 return
1062 if isinstance(ie_result, list):
1063 # Backwards compatibility: old IE result format
1064 ie_result = {
1065 '_type': 'compat_list',
1066 'entries': ie_result,
1067 }
1068 if info_dict:
1069 if info_dict.get('id'):
1070 ie_result['id'] = info_dict['id']
1071 if info_dict.get('title'):
1072 ie_result['title'] = info_dict['title']
1073 self.add_default_extra_info(ie_result, ie, url)
1074 if process:
1075 return self.process_ie_result(ie_result, download, extra_info)
8222d8de 1076 else:
a0566bbf 1077 return ie_result
fe7e0c98 1078
ea38e55f
PH
1079 def add_default_extra_info(self, ie_result, ie, url):
1080 self.add_extra_info(ie_result, {
1081 'extractor': ie.IE_NAME,
1082 'webpage_url': url,
1083 'webpage_url_basename': url_basename(url),
1084 'extractor_key': ie.ie_key(),
1085 })
1086
8222d8de
JMF
1087 def process_ie_result(self, ie_result, download=True, extra_info={}):
1088 """
1089 Take the result of the ie(may be modified) and resolve all unresolved
1090 references (URLs, playlist items).
1091
1092 It will also download the videos if 'download'.
1093 Returns the resolved ie_result.
1094 """
e8ee972c
PH
1095 result_type = ie_result.get('_type', 'video')
1096
057a5206 1097 if result_type in ('url', 'url_transparent'):
134c6ea8 1098 ie_result['url'] = sanitize_url(ie_result['url'])
057a5206 1099 extract_flat = self.params.get('extract_flat', False)
3089bc74
S
1100 if ((extract_flat == 'in_playlist' and 'playlist' in extra_info)
1101 or extract_flat is True):
de6000d9 1102 self.__forced_printings(ie_result, self.prepare_filename(ie_result), incomplete=True)
e8ee972c
PH
1103 return ie_result
1104
8222d8de 1105 if result_type == 'video':
b6c45014 1106 self.add_extra_info(ie_result, extra_info)
feee2ecf 1107 return self.process_video_result(ie_result, download=download)
8222d8de
JMF
1108 elif result_type == 'url':
1109 # We have to add extra_info to the results because it may be
1110 # contained in a playlist
1111 return self.extract_info(ie_result['url'],
0704d222 1112 download, info_dict=ie_result,
8222d8de
JMF
1113 ie_key=ie_result.get('ie_key'),
1114 extra_info=extra_info)
7fc3fa05
PH
1115 elif result_type == 'url_transparent':
1116 # Use the information from the embedding page
1117 info = self.extract_info(
1118 ie_result['url'], ie_key=ie_result.get('ie_key'),
1119 extra_info=extra_info, download=False, process=False)
1120
1640eb09
S
1121 # extract_info may return None when ignoreerrors is enabled and
1122 # extraction failed with an error, don't crash and return early
1123 # in this case
1124 if not info:
1125 return info
1126
412c617d
PH
1127 force_properties = dict(
1128 (k, v) for k, v in ie_result.items() if v is not None)
0396806f 1129 for f in ('_type', 'url', 'id', 'extractor', 'extractor_key', 'ie_key'):
412c617d
PH
1130 if f in force_properties:
1131 del force_properties[f]
1132 new_result = info.copy()
1133 new_result.update(force_properties)
7fc3fa05 1134
0563f7ac
S
1135 # Extracted info may not be a video result (i.e.
1136 # info.get('_type', 'video') != video) but rather an url or
1137 # url_transparent. In such cases outer metadata (from ie_result)
1138 # should be propagated to inner one (info). For this to happen
1139 # _type of info should be overridden with url_transparent. This
067aa17e 1140 # fixes issue from https://github.com/ytdl-org/youtube-dl/pull/11163.
0563f7ac
S
1141 if new_result.get('_type') == 'url':
1142 new_result['_type'] = 'url_transparent'
7fc3fa05
PH
1143
1144 return self.process_ie_result(
1145 new_result, download=download, extra_info=extra_info)
40fcba5e 1146 elif result_type in ('playlist', 'multi_video'):
30a074c2 1147 # Protect from infinite recursion due to recursively nested playlists
1148 # (see https://github.com/ytdl-org/youtube-dl/issues/27833)
1149 webpage_url = ie_result['webpage_url']
1150 if webpage_url in self._playlist_urls:
7e85e872 1151 self.to_screen(
30a074c2 1152 '[download] Skipping already downloaded playlist: %s'
1153 % ie_result.get('title') or ie_result.get('id'))
1154 return
7e85e872 1155
30a074c2 1156 self._playlist_level += 1
1157 self._playlist_urls.add(webpage_url)
1158 try:
1159 return self.__process_playlist(ie_result, download)
1160 finally:
1161 self._playlist_level -= 1
1162 if not self._playlist_level:
1163 self._playlist_urls.clear()
8222d8de 1164 elif result_type == 'compat_list':
c9bf4114
PH
1165 self.report_warning(
1166 'Extractor %s returned a compat_list result. '
1167 'It needs to be updated.' % ie_result.get('extractor'))
5f6a1245 1168
8222d8de 1169 def _fixup(r):
9e1a5b84
JW
1170 self.add_extra_info(
1171 r,
9103bbc5
JMF
1172 {
1173 'extractor': ie_result['extractor'],
1174 'webpage_url': ie_result['webpage_url'],
29eb5174 1175 'webpage_url_basename': url_basename(ie_result['webpage_url']),
be97abc2 1176 'extractor_key': ie_result['extractor_key'],
9e1a5b84
JW
1177 }
1178 )
8222d8de
JMF
1179 return r
1180 ie_result['entries'] = [
b6c45014 1181 self.process_ie_result(_fixup(r), download, extra_info)
8222d8de
JMF
1182 for r in ie_result['entries']
1183 ]
1184 return ie_result
1185 else:
1186 raise Exception('Invalid result type: %s' % result_type)
1187
e92caff5 1188 def _ensure_dir_exists(self, path):
1189 return make_dir(path, self.report_error)
1190
30a074c2 1191 def __process_playlist(self, ie_result, download):
1192 # We process each entry in the playlist
1193 playlist = ie_result.get('title') or ie_result.get('id')
1194 self.to_screen('[download] Downloading playlist: %s' % playlist)
1195
498f5606 1196 if 'entries' not in ie_result:
1197 raise EntryNotInPlaylist()
1198 incomplete_entries = bool(ie_result.get('requested_entries'))
1199 if incomplete_entries:
1200 def fill_missing_entries(entries, indexes):
1201 ret = [None] * max(*indexes)
1202 for i, entry in zip(indexes, entries):
1203 ret[i - 1] = entry
1204 return ret
1205 ie_result['entries'] = fill_missing_entries(ie_result['entries'], ie_result['requested_entries'])
02fd60d3 1206
30a074c2 1207 playlist_results = []
1208
1209 playliststart = self.params.get('playliststart', 1) - 1
1210 playlistend = self.params.get('playlistend')
1211 # For backwards compatibility, interpret -1 as whole list
1212 if playlistend == -1:
1213 playlistend = None
1214
1215 playlistitems_str = self.params.get('playlist_items')
1216 playlistitems = None
1217 if playlistitems_str is not None:
1218 def iter_playlistitems(format):
1219 for string_segment in format.split(','):
1220 if '-' in string_segment:
1221 start, end = string_segment.split('-')
1222 for item in range(int(start), int(end) + 1):
1223 yield int(item)
1224 else:
1225 yield int(string_segment)
1226 playlistitems = orderedSet(iter_playlistitems(playlistitems_str))
1227
1228 ie_entries = ie_result['entries']
1229
1230 def make_playlistitems_entries(list_ie_entries):
1231 num_entries = len(list_ie_entries)
498f5606 1232 for i in playlistitems:
1233 if -num_entries < i <= num_entries:
1234 yield list_ie_entries[i - 1]
1235 elif incomplete_entries:
1236 raise EntryNotInPlaylist()
30a074c2 1237
1238 if isinstance(ie_entries, list):
1239 n_all_entries = len(ie_entries)
1240 if playlistitems:
498f5606 1241 entries = list(make_playlistitems_entries(ie_entries))
30a074c2 1242 else:
1243 entries = ie_entries[playliststart:playlistend]
1244 n_entries = len(entries)
498f5606 1245 msg = 'Collected %d videos; downloading %d of them' % (n_all_entries, n_entries)
30a074c2 1246 elif isinstance(ie_entries, PagedList):
1247 if playlistitems:
1248 entries = []
1249 for item in playlistitems:
1250 entries.extend(ie_entries.getslice(
1251 item - 1, item
1252 ))
1253 else:
1254 entries = ie_entries.getslice(
1255 playliststart, playlistend)
1256 n_entries = len(entries)
498f5606 1257 msg = 'Downloading %d videos' % n_entries
30a074c2 1258 else: # iterable
1259 if playlistitems:
498f5606 1260 entries = list(make_playlistitems_entries(list(itertools.islice(
1261 ie_entries, 0, max(playlistitems)))))
30a074c2 1262 else:
1263 entries = list(itertools.islice(
1264 ie_entries, playliststart, playlistend))
1265 n_entries = len(entries)
498f5606 1266 msg = 'Downloading %d videos' % n_entries
1267
1268 if any((entry is None for entry in entries)):
1269 raise EntryNotInPlaylist()
1270 if not playlistitems and (playliststart or playlistend):
1271 playlistitems = list(range(1 + playliststart, 1 + playliststart + len(entries)))
1272 ie_result['entries'] = entries
1273 ie_result['requested_entries'] = playlistitems
1274
1275 if self.params.get('allow_playlist_files', True):
1276 ie_copy = {
1277 'playlist': playlist,
1278 'playlist_id': ie_result.get('id'),
1279 'playlist_title': ie_result.get('title'),
1280 'playlist_uploader': ie_result.get('uploader'),
1281 'playlist_uploader_id': ie_result.get('uploader_id'),
1282 'playlist_index': 0
1283 }
1284 ie_copy.update(dict(ie_result))
1285
1286 if self.params.get('writeinfojson', False):
1287 infofn = self.prepare_filename(ie_copy, 'pl_infojson')
1288 if not self._ensure_dir_exists(encodeFilename(infofn)):
1289 return
1290 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(infofn)):
1291 self.to_screen('[info] Playlist metadata is already present')
1292 else:
1293 self.to_screen('[info] Writing playlist metadata as JSON to: ' + infofn)
1294 try:
1295 write_json_file(self.filter_requested_info(ie_result, self.params.get('clean_infojson', True)), infofn)
1296 except (OSError, IOError):
1297 self.report_error('Cannot write playlist metadata to JSON file ' + infofn)
1298
1299 if self.params.get('writedescription', False):
1300 descfn = self.prepare_filename(ie_copy, 'pl_description')
1301 if not self._ensure_dir_exists(encodeFilename(descfn)):
1302 return
1303 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(descfn)):
1304 self.to_screen('[info] Playlist description is already present')
1305 elif ie_result.get('description') is None:
1306 self.report_warning('There\'s no playlist description to write.')
1307 else:
1308 try:
1309 self.to_screen('[info] Writing playlist description to: ' + descfn)
1310 with io.open(encodeFilename(descfn), 'w', encoding='utf-8') as descfile:
1311 descfile.write(ie_result['description'])
1312 except (OSError, IOError):
1313 self.report_error('Cannot write playlist description file ' + descfn)
1314 return
30a074c2 1315
1316 if self.params.get('playlistreverse', False):
1317 entries = entries[::-1]
30a074c2 1318 if self.params.get('playlistrandom', False):
1319 random.shuffle(entries)
1320
1321 x_forwarded_for = ie_result.get('__x_forwarded_for_ip')
1322
498f5606 1323 self.to_screen('[%s] playlist %s: %s' % (ie_result['extractor'], playlist, msg))
30a074c2 1324 for i, entry in enumerate(entries, 1):
1325 self.to_screen('[download] Downloading video %s of %s' % (i, n_entries))
1326 # This __x_forwarded_for_ip thing is a bit ugly but requires
1327 # minimal changes
1328 if x_forwarded_for:
1329 entry['__x_forwarded_for_ip'] = x_forwarded_for
1330 extra = {
1331 'n_entries': n_entries,
1332 'playlist': playlist,
1333 'playlist_id': ie_result.get('id'),
1334 'playlist_title': ie_result.get('title'),
1335 'playlist_uploader': ie_result.get('uploader'),
1336 'playlist_uploader_id': ie_result.get('uploader_id'),
498f5606 1337 'playlist_index': playlistitems[i - 1] if playlistitems else i,
30a074c2 1338 'extractor': ie_result['extractor'],
1339 'webpage_url': ie_result['webpage_url'],
1340 'webpage_url_basename': url_basename(ie_result['webpage_url']),
1341 'extractor_key': ie_result['extractor_key'],
1342 }
1343
1344 if self._match_entry(entry, incomplete=True) is not None:
1345 continue
1346
1347 entry_result = self.__process_iterable_entry(entry, download, extra)
1348 # TODO: skip failed (empty) entries?
1349 playlist_results.append(entry_result)
1350 ie_result['entries'] = playlist_results
1351 self.to_screen('[download] Finished downloading playlist: %s' % playlist)
1352 return ie_result
1353
a0566bbf 1354 @__handle_extraction_exceptions
1355 def __process_iterable_entry(self, entry, download, extra_info):
1356 return self.process_ie_result(
1357 entry, download=download, extra_info=extra_info)
1358
67134eab
JMF
1359 def _build_format_filter(self, filter_spec):
1360 " Returns a function to filter the formats according to the filter_spec "
083c9df9
PH
1361
1362 OPERATORS = {
1363 '<': operator.lt,
1364 '<=': operator.le,
1365 '>': operator.gt,
1366 '>=': operator.ge,
1367 '=': operator.eq,
1368 '!=': operator.ne,
1369 }
67134eab 1370 operator_rex = re.compile(r'''(?x)\s*
a03a3c80 1371 (?P<key>width|height|tbr|abr|vbr|asr|filesize|filesize_approx|fps)
083c9df9
PH
1372 \s*(?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
1373 (?P<value>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?)
67134eab 1374 $
083c9df9 1375 ''' % '|'.join(map(re.escape, OPERATORS.keys())))
67134eab 1376 m = operator_rex.search(filter_spec)
9ddb6925
S
1377 if m:
1378 try:
1379 comparison_value = int(m.group('value'))
1380 except ValueError:
1381 comparison_value = parse_filesize(m.group('value'))
1382 if comparison_value is None:
1383 comparison_value = parse_filesize(m.group('value') + 'B')
1384 if comparison_value is None:
1385 raise ValueError(
1386 'Invalid value %r in format specification %r' % (
67134eab 1387 m.group('value'), filter_spec))
9ddb6925
S
1388 op = OPERATORS[m.group('op')]
1389
083c9df9 1390 if not m:
9ddb6925
S
1391 STR_OPERATORS = {
1392 '=': operator.eq,
10d33b34
YCH
1393 '^=': lambda attr, value: attr.startswith(value),
1394 '$=': lambda attr, value: attr.endswith(value),
1395 '*=': lambda attr, value: value in attr,
9ddb6925 1396 }
67134eab 1397 str_operator_rex = re.compile(r'''(?x)
f96bff99 1398 \s*(?P<key>[a-zA-Z0-9._-]+)
2cc779f4 1399 \s*(?P<negation>!\s*)?(?P<op>%s)(?P<none_inclusive>\s*\?)?
b0df5223 1400 \s*(?P<value>[a-zA-Z0-9._-]+)
67134eab 1401 \s*$
9ddb6925 1402 ''' % '|'.join(map(re.escape, STR_OPERATORS.keys())))
67134eab 1403 m = str_operator_rex.search(filter_spec)
9ddb6925
S
1404 if m:
1405 comparison_value = m.group('value')
2cc779f4
S
1406 str_op = STR_OPERATORS[m.group('op')]
1407 if m.group('negation'):
e118a879 1408 op = lambda attr, value: not str_op(attr, value)
2cc779f4
S
1409 else:
1410 op = str_op
083c9df9 1411
9ddb6925 1412 if not m:
67134eab 1413 raise ValueError('Invalid filter specification %r' % filter_spec)
083c9df9
PH
1414
1415 def _filter(f):
1416 actual_value = f.get(m.group('key'))
1417 if actual_value is None:
1418 return m.group('none_inclusive')
1419 return op(actual_value, comparison_value)
67134eab
JMF
1420 return _filter
1421
0017d9ad 1422 def _default_format_spec(self, info_dict, download=True):
0017d9ad 1423
af0f7428
S
1424 def can_merge():
1425 merger = FFmpegMergerPP(self)
1426 return merger.available and merger.can_merge()
1427
91ebc640 1428 prefer_best = (
1429 not self.params.get('simulate', False)
1430 and download
1431 and (
1432 not can_merge()
19807826 1433 or info_dict.get('is_live', False)
de6000d9 1434 or self.outtmpl_dict['default'] == '-'))
91ebc640 1435
1436 return (
1437 'best/bestvideo+bestaudio'
1438 if prefer_best
1439 else 'bestvideo*+bestaudio/best'
19807826 1440 if not self.params.get('allow_multiple_audio_streams', False)
91ebc640 1441 else 'bestvideo+bestaudio/best')
0017d9ad 1442
67134eab
JMF
1443 def build_format_selector(self, format_spec):
1444 def syntax_error(note, start):
1445 message = (
1446 'Invalid format specification: '
1447 '{0}\n\t{1}\n\t{2}^'.format(note, format_spec, ' ' * start[1]))
1448 return SyntaxError(message)
1449
1450 PICKFIRST = 'PICKFIRST'
1451 MERGE = 'MERGE'
1452 SINGLE = 'SINGLE'
0130afb7 1453 GROUP = 'GROUP'
67134eab
JMF
1454 FormatSelector = collections.namedtuple('FormatSelector', ['type', 'selector', 'filters'])
1455
91ebc640 1456 allow_multiple_streams = {'audio': self.params.get('allow_multiple_audio_streams', False),
1457 'video': self.params.get('allow_multiple_video_streams', False)}
909d24dd 1458
67134eab
JMF
1459 def _parse_filter(tokens):
1460 filter_parts = []
1461 for type, string, start, _, _ in tokens:
1462 if type == tokenize.OP and string == ']':
1463 return ''.join(filter_parts)
1464 else:
1465 filter_parts.append(string)
1466
232541df 1467 def _remove_unused_ops(tokens):
17cc1534 1468 # Remove operators that we don't use and join them with the surrounding strings
232541df
JMF
1469 # for example: 'mp4' '-' 'baseline' '-' '16x9' is converted to 'mp4-baseline-16x9'
1470 ALLOWED_OPS = ('/', '+', ',', '(', ')')
1471 last_string, last_start, last_end, last_line = None, None, None, None
1472 for type, string, start, end, line in tokens:
1473 if type == tokenize.OP and string == '[':
1474 if last_string:
1475 yield tokenize.NAME, last_string, last_start, last_end, last_line
1476 last_string = None
1477 yield type, string, start, end, line
1478 # everything inside brackets will be handled by _parse_filter
1479 for type, string, start, end, line in tokens:
1480 yield type, string, start, end, line
1481 if type == tokenize.OP and string == ']':
1482 break
1483 elif type == tokenize.OP and string in ALLOWED_OPS:
1484 if last_string:
1485 yield tokenize.NAME, last_string, last_start, last_end, last_line
1486 last_string = None
1487 yield type, string, start, end, line
1488 elif type in [tokenize.NAME, tokenize.NUMBER, tokenize.OP]:
1489 if not last_string:
1490 last_string = string
1491 last_start = start
1492 last_end = end
1493 else:
1494 last_string += string
1495 if last_string:
1496 yield tokenize.NAME, last_string, last_start, last_end, last_line
1497
cf2ac6df 1498 def _parse_format_selection(tokens, inside_merge=False, inside_choice=False, inside_group=False):
67134eab
JMF
1499 selectors = []
1500 current_selector = None
1501 for type, string, start, _, _ in tokens:
1502 # ENCODING is only defined in python 3.x
1503 if type == getattr(tokenize, 'ENCODING', None):
1504 continue
1505 elif type in [tokenize.NAME, tokenize.NUMBER]:
1506 current_selector = FormatSelector(SINGLE, string, [])
1507 elif type == tokenize.OP:
cf2ac6df
JMF
1508 if string == ')':
1509 if not inside_group:
1510 # ')' will be handled by the parentheses group
1511 tokens.restore_last_token()
67134eab 1512 break
cf2ac6df 1513 elif inside_merge and string in ['/', ',']:
0130afb7
JMF
1514 tokens.restore_last_token()
1515 break
cf2ac6df
JMF
1516 elif inside_choice and string == ',':
1517 tokens.restore_last_token()
1518 break
1519 elif string == ',':
0a31a350
JMF
1520 if not current_selector:
1521 raise syntax_error('"," must follow a format selector', start)
67134eab
JMF
1522 selectors.append(current_selector)
1523 current_selector = None
1524 elif string == '/':
d96d604e
JMF
1525 if not current_selector:
1526 raise syntax_error('"/" must follow a format selector', start)
67134eab 1527 first_choice = current_selector
cf2ac6df 1528 second_choice = _parse_format_selection(tokens, inside_choice=True)
f5f4a27a 1529 current_selector = FormatSelector(PICKFIRST, (first_choice, second_choice), [])
67134eab
JMF
1530 elif string == '[':
1531 if not current_selector:
1532 current_selector = FormatSelector(SINGLE, 'best', [])
1533 format_filter = _parse_filter(tokens)
1534 current_selector.filters.append(format_filter)
0130afb7
JMF
1535 elif string == '(':
1536 if current_selector:
1537 raise syntax_error('Unexpected "("', start)
cf2ac6df
JMF
1538 group = _parse_format_selection(tokens, inside_group=True)
1539 current_selector = FormatSelector(GROUP, group, [])
67134eab 1540 elif string == '+':
d03cfdce 1541 if not current_selector:
1542 raise syntax_error('Unexpected "+"', start)
1543 selector_1 = current_selector
1544 selector_2 = _parse_format_selection(tokens, inside_merge=True)
1545 if not selector_2:
1546 raise syntax_error('Expected a selector', start)
1547 current_selector = FormatSelector(MERGE, (selector_1, selector_2), [])
67134eab
JMF
1548 else:
1549 raise syntax_error('Operator not recognized: "{0}"'.format(string), start)
1550 elif type == tokenize.ENDMARKER:
1551 break
1552 if current_selector:
1553 selectors.append(current_selector)
1554 return selectors
1555
f8d4ad9a 1556 def _merge(formats_pair):
1557 format_1, format_2 = formats_pair
1558
1559 formats_info = []
1560 formats_info.extend(format_1.get('requested_formats', (format_1,)))
1561 formats_info.extend(format_2.get('requested_formats', (format_2,)))
1562
1563 if not allow_multiple_streams['video'] or not allow_multiple_streams['audio']:
1564 get_no_more = {"video": False, "audio": False}
1565 for (i, fmt_info) in enumerate(formats_info):
1566 for aud_vid in ["audio", "video"]:
1567 if not allow_multiple_streams[aud_vid] and fmt_info.get(aud_vid[0] + 'codec') != 'none':
1568 if get_no_more[aud_vid]:
1569 formats_info.pop(i)
1570 get_no_more[aud_vid] = True
1571
1572 if len(formats_info) == 1:
1573 return formats_info[0]
1574
1575 video_fmts = [fmt_info for fmt_info in formats_info if fmt_info.get('vcodec') != 'none']
1576 audio_fmts = [fmt_info for fmt_info in formats_info if fmt_info.get('acodec') != 'none']
1577
1578 the_only_video = video_fmts[0] if len(video_fmts) == 1 else None
1579 the_only_audio = audio_fmts[0] if len(audio_fmts) == 1 else None
1580
1581 output_ext = self.params.get('merge_output_format')
1582 if not output_ext:
1583 if the_only_video:
1584 output_ext = the_only_video['ext']
1585 elif the_only_audio and not video_fmts:
1586 output_ext = the_only_audio['ext']
1587 else:
1588 output_ext = 'mkv'
1589
1590 new_dict = {
1591 'requested_formats': formats_info,
1592 'format': '+'.join(fmt_info.get('format') for fmt_info in formats_info),
1593 'format_id': '+'.join(fmt_info.get('format_id') for fmt_info in formats_info),
1594 'ext': output_ext,
1595 }
1596
1597 if the_only_video:
1598 new_dict.update({
1599 'width': the_only_video.get('width'),
1600 'height': the_only_video.get('height'),
1601 'resolution': the_only_video.get('resolution') or self.format_resolution(the_only_video),
1602 'fps': the_only_video.get('fps'),
1603 'vcodec': the_only_video.get('vcodec'),
1604 'vbr': the_only_video.get('vbr'),
1605 'stretched_ratio': the_only_video.get('stretched_ratio'),
1606 })
1607
1608 if the_only_audio:
1609 new_dict.update({
1610 'acodec': the_only_audio.get('acodec'),
1611 'abr': the_only_audio.get('abr'),
1612 })
1613
1614 return new_dict
1615
67134eab 1616 def _build_selector_function(selector):
909d24dd 1617 if isinstance(selector, list): # ,
67134eab
JMF
1618 fs = [_build_selector_function(s) for s in selector]
1619
317f7ab6 1620 def selector_function(ctx):
67134eab 1621 for f in fs:
317f7ab6 1622 for format in f(ctx):
67134eab
JMF
1623 yield format
1624 return selector_function
909d24dd 1625
1626 elif selector.type == GROUP: # ()
0130afb7 1627 selector_function = _build_selector_function(selector.selector)
909d24dd 1628
1629 elif selector.type == PICKFIRST: # /
67134eab
JMF
1630 fs = [_build_selector_function(s) for s in selector.selector]
1631
317f7ab6 1632 def selector_function(ctx):
67134eab 1633 for f in fs:
317f7ab6 1634 picked_formats = list(f(ctx))
67134eab
JMF
1635 if picked_formats:
1636 return picked_formats
1637 return []
67134eab 1638
909d24dd 1639 elif selector.type == SINGLE: # atom
f8d4ad9a 1640 format_spec = (selector.selector if selector.selector is not None else 'best').lower()
909d24dd 1641
f8d4ad9a 1642 # TODO: Add allvideo, allaudio etc by generalizing the code with best/worst selector
909d24dd 1643 if format_spec == 'all':
1644 def selector_function(ctx):
1645 formats = list(ctx['formats'])
1646 if formats:
1647 for f in formats:
1648 yield f
f8d4ad9a 1649 elif format_spec == 'mergeall':
1650 def selector_function(ctx):
1651 formats = list(ctx['formats'])
1652 merged_format = formats[0]
1653 for f in formats[1:]:
1654 merged_format = _merge((merged_format, f))
1655 yield merged_format
909d24dd 1656
1657 else:
1658 format_fallback = False
eff63539 1659 mobj = re.match(
1660 r'(?P<bw>best|worst|b|w)(?P<type>video|audio|v|a)?(?P<mod>\*)?(?:\.(?P<n>[1-9]\d*))?$',
1661 format_spec)
1662 if mobj is not None:
1663 format_idx = int_or_none(mobj.group('n'), default=1)
1664 format_idx = format_idx - 1 if mobj.group('bw')[0] == 'w' else -format_idx
1665 format_type = (mobj.group('type') or [None])[0]
1666 not_format_type = {'v': 'a', 'a': 'v'}.get(format_type)
1667 format_modified = mobj.group('mod') is not None
909d24dd 1668
1669 format_fallback = not format_type and not format_modified # for b, w
eff63539 1670 filter_f = (
1671 (lambda f: f.get('%scodec' % format_type) != 'none')
1672 if format_type and format_modified # bv*, ba*, wv*, wa*
1673 else (lambda f: f.get('%scodec' % not_format_type) == 'none')
1674 if format_type # bv, ba, wv, wa
1675 else (lambda f: f.get('vcodec') != 'none' and f.get('acodec') != 'none')
1676 if not format_modified # b, w
1677 else None) # b*, w*
67134eab 1678 else:
909d24dd 1679 format_idx = -1
1680 filter_f = ((lambda f: f.get('ext') == format_spec)
1681 if format_spec in ['mp4', 'flv', 'webm', '3gp', 'm4a', 'mp3', 'ogg', 'aac', 'wav'] # extension
1682 else (lambda f: f.get('format_id') == format_spec)) # id
1683
1684 def selector_function(ctx):
1685 formats = list(ctx['formats'])
1686 if not formats:
1687 return
1688 matches = list(filter(filter_f, formats)) if filter_f is not None else formats
eff63539 1689 n = len(matches)
1690 if -n <= format_idx < n:
909d24dd 1691 yield matches[format_idx]
eff63539 1692 elif format_fallback and ctx['incomplete_formats']:
909d24dd 1693 # for extractors with incomplete formats (audio only (soundcloud)
1694 # or video only (imgur)) best/worst will fallback to
1695 # best/worst {video,audio}-only format
eff63539 1696 n = len(formats)
1697 if -n <= format_idx < n:
1698 yield formats[format_idx]
909d24dd 1699
1700 elif selector.type == MERGE: # +
d03cfdce 1701 selector_1, selector_2 = map(_build_selector_function, selector.selector)
083c9df9 1702
317f7ab6
S
1703 def selector_function(ctx):
1704 for pair in itertools.product(
d03cfdce 1705 selector_1(copy.deepcopy(ctx)), selector_2(copy.deepcopy(ctx))):
67134eab 1706 yield _merge(pair)
083c9df9 1707
67134eab 1708 filters = [self._build_format_filter(f) for f in selector.filters]
083c9df9 1709
317f7ab6
S
1710 def final_selector(ctx):
1711 ctx_copy = copy.deepcopy(ctx)
67134eab 1712 for _filter in filters:
317f7ab6
S
1713 ctx_copy['formats'] = list(filter(_filter, ctx_copy['formats']))
1714 return selector_function(ctx_copy)
67134eab 1715 return final_selector
083c9df9 1716
67134eab 1717 stream = io.BytesIO(format_spec.encode('utf-8'))
0130afb7 1718 try:
232541df 1719 tokens = list(_remove_unused_ops(compat_tokenize_tokenize(stream.readline)))
0130afb7
JMF
1720 except tokenize.TokenError:
1721 raise syntax_error('Missing closing/opening brackets or parenthesis', (0, len(format_spec)))
1722
1723 class TokenIterator(object):
1724 def __init__(self, tokens):
1725 self.tokens = tokens
1726 self.counter = 0
1727
1728 def __iter__(self):
1729 return self
1730
1731 def __next__(self):
1732 if self.counter >= len(self.tokens):
1733 raise StopIteration()
1734 value = self.tokens[self.counter]
1735 self.counter += 1
1736 return value
1737
1738 next = __next__
1739
1740 def restore_last_token(self):
1741 self.counter -= 1
1742
1743 parsed_selector = _parse_format_selection(iter(TokenIterator(tokens)))
67134eab 1744 return _build_selector_function(parsed_selector)
a9c58ad9 1745
e5660ee6
JMF
1746 def _calc_headers(self, info_dict):
1747 res = std_headers.copy()
1748
1749 add_headers = info_dict.get('http_headers')
1750 if add_headers:
1751 res.update(add_headers)
1752
1753 cookies = self._calc_cookies(info_dict)
1754 if cookies:
1755 res['Cookie'] = cookies
1756
0016b84e
S
1757 if 'X-Forwarded-For' not in res:
1758 x_forwarded_for_ip = info_dict.get('__x_forwarded_for_ip')
1759 if x_forwarded_for_ip:
1760 res['X-Forwarded-For'] = x_forwarded_for_ip
1761
e5660ee6
JMF
1762 return res
1763
1764 def _calc_cookies(self, info_dict):
5c2266df 1765 pr = sanitized_Request(info_dict['url'])
e5660ee6 1766 self.cookiejar.add_cookie_header(pr)
662435f7 1767 return pr.get_header('Cookie')
e5660ee6 1768
dd82ffea
JMF
1769 def process_video_result(self, info_dict, download=True):
1770 assert info_dict.get('_type', 'video') == 'video'
1771
bec1fad2
PH
1772 if 'id' not in info_dict:
1773 raise ExtractorError('Missing "id" field in extractor result')
1774 if 'title' not in info_dict:
1775 raise ExtractorError('Missing "title" field in extractor result')
1776
c9969434
S
1777 def report_force_conversion(field, field_not, conversion):
1778 self.report_warning(
1779 '"%s" field is not %s - forcing %s conversion, there is an error in extractor'
1780 % (field, field_not, conversion))
1781
1782 def sanitize_string_field(info, string_field):
1783 field = info.get(string_field)
1784 if field is None or isinstance(field, compat_str):
1785 return
1786 report_force_conversion(string_field, 'a string', 'string')
1787 info[string_field] = compat_str(field)
1788
1789 def sanitize_numeric_fields(info):
1790 for numeric_field in self._NUMERIC_FIELDS:
1791 field = info.get(numeric_field)
1792 if field is None or isinstance(field, compat_numeric_types):
1793 continue
1794 report_force_conversion(numeric_field, 'numeric', 'int')
1795 info[numeric_field] = int_or_none(field)
1796
1797 sanitize_string_field(info_dict, 'id')
1798 sanitize_numeric_fields(info_dict)
be6217b2 1799
dd82ffea
JMF
1800 if 'playlist' not in info_dict:
1801 # It isn't part of a playlist
1802 info_dict['playlist'] = None
1803 info_dict['playlist_index'] = None
1804
d5519808 1805 thumbnails = info_dict.get('thumbnails')
cfb56d1a
PH
1806 if thumbnails is None:
1807 thumbnail = info_dict.get('thumbnail')
1808 if thumbnail:
a7a14d95 1809 info_dict['thumbnails'] = thumbnails = [{'url': thumbnail}]
d5519808 1810 if thumbnails:
be6d7229 1811 thumbnails.sort(key=lambda t: (
d37708fc
RA
1812 t.get('preference') if t.get('preference') is not None else -1,
1813 t.get('width') if t.get('width') is not None else -1,
1814 t.get('height') if t.get('height') is not None else -1,
1815 t.get('id') if t.get('id') is not None else '', t.get('url')))
f6c24009 1816 for i, t in enumerate(thumbnails):
dcf77cf1 1817 t['url'] = sanitize_url(t['url'])
9603e8a7 1818 if t.get('width') and t.get('height'):
d5519808 1819 t['resolution'] = '%dx%d' % (t['width'], t['height'])
f6c24009
PH
1820 if t.get('id') is None:
1821 t['id'] = '%d' % i
d5519808 1822
b7b72db9 1823 if self.params.get('list_thumbnails'):
1824 self.list_thumbnails(info_dict)
1825 return
1826
536a55da
S
1827 thumbnail = info_dict.get('thumbnail')
1828 if thumbnail:
1829 info_dict['thumbnail'] = sanitize_url(thumbnail)
1830 elif thumbnails:
d5519808
PH
1831 info_dict['thumbnail'] = thumbnails[-1]['url']
1832
c9ae7b95 1833 if 'display_id' not in info_dict and 'id' in info_dict:
0afef30b
PH
1834 info_dict['display_id'] = info_dict['id']
1835
10db0d2f 1836 for ts_key, date_key in (
1837 ('timestamp', 'upload_date'),
1838 ('release_timestamp', 'release_date'),
1839 ):
1840 if info_dict.get(date_key) is None and info_dict.get(ts_key) is not None:
1841 # Working around out-of-range timestamp values (e.g. negative ones on Windows,
1842 # see http://bugs.python.org/issue1646728)
1843 try:
1844 upload_date = datetime.datetime.utcfromtimestamp(info_dict[ts_key])
1845 info_dict[date_key] = upload_date.strftime('%Y%m%d')
1846 except (ValueError, OverflowError, OSError):
1847 pass
9d2ecdbc 1848
33d2fc2f
S
1849 # Auto generate title fields corresponding to the *_number fields when missing
1850 # in order to always have clean titles. This is very common for TV series.
1851 for field in ('chapter', 'season', 'episode'):
1852 if info_dict.get('%s_number' % field) is not None and not info_dict.get(field):
1853 info_dict[field] = '%s %d' % (field.capitalize(), info_dict['%s_number' % field])
1854
05108a49
S
1855 for cc_kind in ('subtitles', 'automatic_captions'):
1856 cc = info_dict.get(cc_kind)
1857 if cc:
1858 for _, subtitle in cc.items():
1859 for subtitle_format in subtitle:
1860 if subtitle_format.get('url'):
1861 subtitle_format['url'] = sanitize_url(subtitle_format['url'])
1862 if subtitle_format.get('ext') is None:
1863 subtitle_format['ext'] = determine_ext(subtitle_format['url']).lower()
1864
1865 automatic_captions = info_dict.get('automatic_captions')
4bba3716 1866 subtitles = info_dict.get('subtitles')
4bba3716 1867
a504ced0 1868 if self.params.get('listsubtitles', False):
360e1ca5 1869 if 'automatic_captions' in info_dict:
05108a49
S
1870 self.list_subtitles(
1871 info_dict['id'], automatic_captions, 'automatic captions')
4bba3716 1872 self.list_subtitles(info_dict['id'], subtitles, 'subtitles')
a504ced0 1873 return
05108a49 1874
360e1ca5 1875 info_dict['requested_subtitles'] = self.process_subtitles(
05108a49 1876 info_dict['id'], subtitles, automatic_captions)
a504ced0 1877
dd82ffea
JMF
1878 # We now pick which formats have to be downloaded
1879 if info_dict.get('formats') is None:
1880 # There's only one format available
1881 formats = [info_dict]
1882 else:
1883 formats = info_dict['formats']
1884
db95dc13
PH
1885 if not formats:
1886 raise ExtractorError('No video formats found!')
1887
73af5cc8
S
1888 def is_wellformed(f):
1889 url = f.get('url')
a5ac0c47 1890 if not url:
73af5cc8
S
1891 self.report_warning(
1892 '"url" field is missing or empty - skipping format, '
1893 'there is an error in extractor')
a5ac0c47
S
1894 return False
1895 if isinstance(url, bytes):
1896 sanitize_string_field(f, 'url')
1897 return True
73af5cc8
S
1898
1899 # Filter out malformed formats for better extraction robustness
1900 formats = list(filter(is_wellformed, formats))
1901
181c7053
S
1902 formats_dict = {}
1903
dd82ffea 1904 # We check that all the formats have the format and format_id fields
db95dc13 1905 for i, format in enumerate(formats):
c9969434
S
1906 sanitize_string_field(format, 'format_id')
1907 sanitize_numeric_fields(format)
dcf77cf1 1908 format['url'] = sanitize_url(format['url'])
e74e3b63 1909 if not format.get('format_id'):
8016c922 1910 format['format_id'] = compat_str(i)
e2effb08
S
1911 else:
1912 # Sanitize format_id from characters used in format selector expression
ec85ded8 1913 format['format_id'] = re.sub(r'[\s,/+\[\]()]', '_', format['format_id'])
181c7053
S
1914 format_id = format['format_id']
1915 if format_id not in formats_dict:
1916 formats_dict[format_id] = []
1917 formats_dict[format_id].append(format)
1918
1919 # Make sure all formats have unique format_id
1920 for format_id, ambiguous_formats in formats_dict.items():
1921 if len(ambiguous_formats) > 1:
1922 for i, format in enumerate(ambiguous_formats):
1923 format['format_id'] = '%s-%d' % (format_id, i)
1924
1925 for i, format in enumerate(formats):
8c51aa65 1926 if format.get('format') is None:
6febd1c1 1927 format['format'] = '{id} - {res}{note}'.format(
8c51aa65
JMF
1928 id=format['format_id'],
1929 res=self.format_resolution(format),
6febd1c1 1930 note=' ({0})'.format(format['format_note']) if format.get('format_note') is not None else '',
8c51aa65 1931 )
c1002e96 1932 # Automatically determine file extension if missing
5b1d8575 1933 if format.get('ext') is None:
cce929ea 1934 format['ext'] = determine_ext(format['url']).lower()
b5559424
S
1935 # Automatically determine protocol if missing (useful for format
1936 # selection purposes)
6f0be937 1937 if format.get('protocol') is None:
b5559424 1938 format['protocol'] = determine_protocol(format)
e5660ee6
JMF
1939 # Add HTTP headers, so that external programs can use them from the
1940 # json output
1941 full_format_info = info_dict.copy()
1942 full_format_info.update(format)
1943 format['http_headers'] = self._calc_headers(full_format_info)
0016b84e
S
1944 # Remove private housekeeping stuff
1945 if '__x_forwarded_for_ip' in info_dict:
1946 del info_dict['__x_forwarded_for_ip']
dd82ffea 1947
4bcc7bd1 1948 # TODO Central sorting goes here
99e206d5 1949
f89197d7 1950 if formats[0] is not info_dict:
b3d9ef88
JMF
1951 # only set the 'formats' fields if the original info_dict list them
1952 # otherwise we end up with a circular reference, the first (and unique)
f89197d7 1953 # element in the 'formats' field in info_dict is info_dict itself,
dfb1b146 1954 # which can't be exported to json
b3d9ef88 1955 info_dict['formats'] = formats
cfb56d1a 1956 if self.params.get('listformats'):
bfaae0a7 1957 self.list_formats(info_dict)
1958 return
1959
de3ef3ed 1960 req_format = self.params.get('format')
a9c58ad9 1961 if req_format is None:
0017d9ad
S
1962 req_format = self._default_format_spec(info_dict, download=download)
1963 if self.params.get('verbose'):
e8be92f9 1964 self.to_screen('[debug] Default format spec: %s' % req_format)
0017d9ad 1965
5acfa126 1966 format_selector = self.build_format_selector(req_format)
317f7ab6
S
1967
1968 # While in format selection we may need to have an access to the original
1969 # format set in order to calculate some metrics or do some processing.
1970 # For now we need to be able to guess whether original formats provided
1971 # by extractor are incomplete or not (i.e. whether extractor provides only
1972 # video-only or audio-only formats) for proper formats selection for
1973 # extractors with such incomplete formats (see
067aa17e 1974 # https://github.com/ytdl-org/youtube-dl/pull/5556).
317f7ab6
S
1975 # Since formats may be filtered during format selection and may not match
1976 # the original formats the results may be incorrect. Thus original formats
1977 # or pre-calculated metrics should be passed to format selection routines
1978 # as well.
1979 # We will pass a context object containing all necessary additional data
1980 # instead of just formats.
1981 # This fixes incorrect format selection issue (see
067aa17e 1982 # https://github.com/ytdl-org/youtube-dl/issues/10083).
2e221ca3 1983 incomplete_formats = (
317f7ab6 1984 # All formats are video-only or
3089bc74 1985 all(f.get('vcodec') != 'none' and f.get('acodec') == 'none' for f in formats)
317f7ab6 1986 # all formats are audio-only
3089bc74 1987 or all(f.get('vcodec') == 'none' and f.get('acodec') != 'none' for f in formats))
317f7ab6
S
1988
1989 ctx = {
1990 'formats': formats,
1991 'incomplete_formats': incomplete_formats,
1992 }
1993
1994 formats_to_download = list(format_selector(ctx))
dd82ffea 1995 if not formats_to_download:
6febd1c1 1996 raise ExtractorError('requested format not available',
78a3a9f8 1997 expected=True)
dd82ffea
JMF
1998
1999 if download:
909d24dd 2000 self.to_screen('[info] Downloading format(s) %s' % ", ".join([f['format_id'] for f in formats_to_download]))
dd82ffea 2001 if len(formats_to_download) > 1:
6febd1c1 2002 self.to_screen('[info] %s: downloading video in %s formats' % (info_dict['id'], len(formats_to_download)))
dd82ffea
JMF
2003 for format in formats_to_download:
2004 new_info = dict(info_dict)
2005 new_info.update(format)
2006 self.process_info(new_info)
2007 # We update the info dict with the best quality format (backwards compatibility)
2008 info_dict.update(formats_to_download[-1])
2009 return info_dict
2010
98c70d6f 2011 def process_subtitles(self, video_id, normal_subtitles, automatic_captions):
a504ced0 2012 """Select the requested subtitles and their format"""
98c70d6f
JMF
2013 available_subs = {}
2014 if normal_subtitles and self.params.get('writesubtitles'):
2015 available_subs.update(normal_subtitles)
2016 if automatic_captions and self.params.get('writeautomaticsub'):
2017 for lang, cap_info in automatic_captions.items():
360e1ca5
JMF
2018 if lang not in available_subs:
2019 available_subs[lang] = cap_info
2020
4d171848
JMF
2021 if (not self.params.get('writesubtitles') and not
2022 self.params.get('writeautomaticsub') or not
2023 available_subs):
2024 return None
a504ced0
JMF
2025
2026 if self.params.get('allsubtitles', False):
2027 requested_langs = available_subs.keys()
2028 else:
2029 if self.params.get('subtitleslangs', False):
2030 requested_langs = self.params.get('subtitleslangs')
2031 elif 'en' in available_subs:
2032 requested_langs = ['en']
2033 else:
2034 requested_langs = [list(available_subs.keys())[0]]
2035
2036 formats_query = self.params.get('subtitlesformat', 'best')
2037 formats_preference = formats_query.split('/') if formats_query else []
2038 subs = {}
2039 for lang in requested_langs:
2040 formats = available_subs.get(lang)
2041 if formats is None:
2042 self.report_warning('%s subtitles not available for %s' % (lang, video_id))
2043 continue
a504ced0
JMF
2044 for ext in formats_preference:
2045 if ext == 'best':
2046 f = formats[-1]
2047 break
2048 matches = list(filter(lambda f: f['ext'] == ext, formats))
2049 if matches:
2050 f = matches[-1]
2051 break
2052 else:
2053 f = formats[-1]
2054 self.report_warning(
2055 'No subtitle format found matching "%s" for language %s, '
2056 'using %s' % (formats_query, lang, f['ext']))
2057 subs[lang] = f
2058 return subs
2059
d06daf23
S
2060 def __forced_printings(self, info_dict, filename, incomplete):
2061 def print_mandatory(field):
2062 if (self.params.get('force%s' % field, False)
2063 and (not incomplete or info_dict.get(field) is not None)):
2064 self.to_stdout(info_dict[field])
2065
2066 def print_optional(field):
2067 if (self.params.get('force%s' % field, False)
2068 and info_dict.get(field) is not None):
2069 self.to_stdout(info_dict[field])
2070
2071 print_mandatory('title')
2072 print_mandatory('id')
2073 if self.params.get('forceurl', False) and not incomplete:
2074 if info_dict.get('requested_formats') is not None:
2075 for f in info_dict['requested_formats']:
2076 self.to_stdout(f['url'] + f.get('play_path', ''))
2077 else:
2078 # For RTMP URLs, also include the playpath
2079 self.to_stdout(info_dict['url'] + info_dict.get('play_path', ''))
2080 print_optional('thumbnail')
2081 print_optional('description')
2082 if self.params.get('forcefilename', False) and filename is not None:
2083 self.to_stdout(filename)
2084 if self.params.get('forceduration', False) and info_dict.get('duration') is not None:
2085 self.to_stdout(formatSeconds(info_dict['duration']))
2086 print_mandatory('format')
2087 if self.params.get('forcejson', False):
277d6ff5 2088 self.post_extract(info_dict)
75d43ca0 2089 self.to_stdout(json.dumps(info_dict, default=repr))
d06daf23 2090
8222d8de
JMF
2091 def process_info(self, info_dict):
2092 """Process a single resolved IE result."""
2093
2094 assert info_dict.get('_type', 'video') == 'video'
fd288278 2095
0202b52a 2096 info_dict.setdefault('__postprocessors', [])
2097
fd288278
PH
2098 max_downloads = self.params.get('max_downloads')
2099 if max_downloads is not None:
2100 if self._num_downloads >= int(max_downloads):
2101 raise MaxDownloadsReached()
8222d8de 2102
d06daf23 2103 # TODO: backward compatibility, to be removed
8222d8de 2104 info_dict['fulltitle'] = info_dict['title']
8222d8de 2105
11b85ce6 2106 if 'format' not in info_dict:
8222d8de
JMF
2107 info_dict['format'] = info_dict['ext']
2108
8b0d7497 2109 if self._match_entry(info_dict, incomplete=False) is not None:
8222d8de
JMF
2110 return
2111
277d6ff5 2112 self.post_extract(info_dict)
fd288278 2113 self._num_downloads += 1
8222d8de 2114
5bfa4862 2115 info_dict = self.pre_process(info_dict)
2116
dcf64d43 2117 # info_dict['_filename'] needs to be set for backward compatibility
de6000d9 2118 info_dict['_filename'] = full_filename = self.prepare_filename(info_dict, warn=True)
2119 temp_filename = self.prepare_filename(info_dict, 'temp')
0202b52a 2120 files_to_move = {}
de6000d9 2121 skip_dl = self.params.get('skip_download', False)
8222d8de
JMF
2122
2123 # Forced printings
0202b52a 2124 self.__forced_printings(info_dict, full_filename, incomplete=False)
8222d8de 2125
8222d8de 2126 if self.params.get('simulate', False):
2d30509f 2127 if self.params.get('force_write_download_archive', False):
2128 self.record_download_archive(info_dict)
2129
2130 # Do nothing else if in simulate mode
8222d8de
JMF
2131 return
2132
de6000d9 2133 if full_filename is None:
8222d8de
JMF
2134 return
2135
e92caff5 2136 if not self._ensure_dir_exists(encodeFilename(full_filename)):
0202b52a 2137 return
e92caff5 2138 if not self._ensure_dir_exists(encodeFilename(temp_filename)):
8222d8de
JMF
2139 return
2140
2141 if self.params.get('writedescription', False):
de6000d9 2142 descfn = self.prepare_filename(info_dict, 'description')
e92caff5 2143 if not self._ensure_dir_exists(encodeFilename(descfn)):
0202b52a 2144 return
0c3d0f51 2145 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(descfn)):
6febd1c1 2146 self.to_screen('[info] Video description is already present')
f00fd51d
JMF
2147 elif info_dict.get('description') is None:
2148 self.report_warning('There\'s no description to write.')
7b6fefc9
PH
2149 else:
2150 try:
6febd1c1 2151 self.to_screen('[info] Writing video description to: ' + descfn)
7b6fefc9
PH
2152 with io.open(encodeFilename(descfn), 'w', encoding='utf-8') as descfile:
2153 descfile.write(info_dict['description'])
7b6fefc9 2154 except (OSError, IOError):
6febd1c1 2155 self.report_error('Cannot write description file ' + descfn)
7b6fefc9 2156 return
8222d8de 2157
1fb07d10 2158 if self.params.get('writeannotations', False):
de6000d9 2159 annofn = self.prepare_filename(info_dict, 'annotation')
e92caff5 2160 if not self._ensure_dir_exists(encodeFilename(annofn)):
0202b52a 2161 return
0c3d0f51 2162 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(annofn)):
6febd1c1 2163 self.to_screen('[info] Video annotations are already present')
ffddb112
RA
2164 elif not info_dict.get('annotations'):
2165 self.report_warning('There are no annotations to write.')
7b6fefc9
PH
2166 else:
2167 try:
6febd1c1 2168 self.to_screen('[info] Writing video annotations to: ' + annofn)
7b6fefc9
PH
2169 with io.open(encodeFilename(annofn), 'w', encoding='utf-8') as annofile:
2170 annofile.write(info_dict['annotations'])
2171 except (KeyError, TypeError):
6febd1c1 2172 self.report_warning('There are no annotations to write.')
7b6fefc9 2173 except (OSError, IOError):
6febd1c1 2174 self.report_error('Cannot write annotations file: ' + annofn)
7b6fefc9 2175 return
1fb07d10 2176
9f448fcb 2177 def dl(name, info, subtitle=False):
98b69821 2178 fd = get_suitable_downloader(info, self.params)(self, self.params)
2179 for ph in self._progress_hooks:
2180 fd.add_progress_hook(ph)
2181 if self.params.get('verbose'):
29f7c58a 2182 self.to_screen('[debug] Invoking downloader on %r' % info.get('url'))
46906886
DA
2183 new_info = dict(info)
2184 if new_info.get('http_headers') is None:
2185 new_info['http_headers'] = self._calc_headers(new_info)
2186 return fd.download(name, new_info, subtitle)
98b69821 2187
c4a91be7 2188 subtitles_are_requested = any([self.params.get('writesubtitles', False),
0b7f3118 2189 self.params.get('writeautomaticsub')])
c4a91be7 2190
c84dd8a9 2191 if subtitles_are_requested and info_dict.get('requested_subtitles'):
8222d8de
JMF
2192 # subtitles download errors are already managed as troubles in relevant IE
2193 # that way it will silently go on when used with unsupporting IE
c84dd8a9 2194 subtitles = info_dict['requested_subtitles']
fa57af1e 2195 # ie = self.get_info_extractor(info_dict['extractor_key'])
a504ced0
JMF
2196 for sub_lang, sub_info in subtitles.items():
2197 sub_format = sub_info['ext']
de6000d9 2198 sub_fn = self.prepare_filename(info_dict, 'subtitle')
2199 sub_filename = subtitles_filename(
0fd1a2b0 2200 temp_filename if not skip_dl else sub_fn,
0202b52a 2201 sub_lang, sub_format, info_dict.get('ext'))
de6000d9 2202 sub_filename_final = subtitles_filename(sub_fn, sub_lang, sub_format, info_dict.get('ext'))
0c3d0f51 2203 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(sub_filename)):
5ff1bc0c 2204 self.to_screen('[info] Video subtitle %s.%s is already present' % (sub_lang, sub_format))
dcf64d43 2205 sub_info['filepath'] = sub_filename
0202b52a 2206 files_to_move[sub_filename] = sub_filename_final
a504ced0 2207 else:
0c9df79e 2208 self.to_screen('[info] Writing video subtitles to: ' + sub_filename)
5ff1bc0c
RA
2209 if sub_info.get('data') is not None:
2210 try:
2211 # Use newline='' to prevent conversion of newline characters
067aa17e 2212 # See https://github.com/ytdl-org/youtube-dl/issues/10268
5ff1bc0c
RA
2213 with io.open(encodeFilename(sub_filename), 'w', encoding='utf-8', newline='') as subfile:
2214 subfile.write(sub_info['data'])
dcf64d43 2215 sub_info['filepath'] = sub_filename
0202b52a 2216 files_to_move[sub_filename] = sub_filename_final
5ff1bc0c
RA
2217 except (OSError, IOError):
2218 self.report_error('Cannot write subtitles file ' + sub_filename)
2219 return
7b6fefc9 2220 else:
5ff1bc0c 2221 try:
dcf64d43 2222 dl(sub_filename, sub_info.copy(), subtitle=True)
2223 sub_info['filepath'] = sub_filename
0202b52a 2224 files_to_move[sub_filename] = sub_filename_final
0c9df79e 2225 except (ExtractorError, IOError, OSError, ValueError, compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
5ff1bc0c
RA
2226 self.report_warning('Unable to download subtitle for "%s": %s' %
2227 (sub_lang, error_to_compat_str(err)))
2228 continue
8222d8de 2229
de6000d9 2230 if skip_dl:
57df9f53 2231 if self.params.get('convertsubtitles', False):
0202b52a 2232 # subconv = FFmpegSubtitlesConvertorPP(self, format=self.params.get('convertsubtitles'))
de6000d9 2233 filename_real_ext = os.path.splitext(full_filename)[1][1:]
57df9f53 2234 filename_wo_ext = (
0202b52a 2235 os.path.splitext(full_filename)[0]
57df9f53 2236 if filename_real_ext == info_dict['ext']
0202b52a 2237 else full_filename)
57df9f53 2238 afilename = '%s.%s' % (filename_wo_ext, self.params.get('convertsubtitles'))
0202b52a 2239 # if subconv.available:
2240 # info_dict['__postprocessors'].append(subconv)
57df9f53 2241 if os.path.exists(encodeFilename(afilename)):
f791b419
U
2242 self.to_screen(
2243 '[download] %s has already been downloaded and '
2244 'converted' % afilename)
57df9f53
U
2245 else:
2246 try:
0202b52a 2247 self.post_process(full_filename, info_dict, files_to_move)
af819c21 2248 except PostProcessingError as err:
2249 self.report_error('Postprocessing: %s' % str(err))
57df9f53
U
2250 return
2251
8222d8de 2252 if self.params.get('writeinfojson', False):
de6000d9 2253 infofn = self.prepare_filename(info_dict, 'infojson')
e92caff5 2254 if not self._ensure_dir_exists(encodeFilename(infofn)):
0202b52a 2255 return
0c3d0f51 2256 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(infofn)):
66c935fb 2257 self.to_screen('[info] Video metadata is already present')
7b6fefc9 2258 else:
66c935fb 2259 self.to_screen('[info] Writing video metadata as JSON to: ' + infofn)
7b6fefc9 2260 try:
75d43ca0 2261 write_json_file(self.filter_requested_info(info_dict, self.params.get('clean_infojson', True)), infofn)
7b6fefc9 2262 except (OSError, IOError):
66c935fb 2263 self.report_error('Cannot write video metadata to JSON file ' + infofn)
7b6fefc9 2264 return
de6000d9 2265 info_dict['__infojson_filename'] = infofn
8222d8de 2266
de6000d9 2267 thumbfn = self.prepare_filename(info_dict, 'thumbnail')
2268 thumb_fn_temp = temp_filename if not skip_dl else thumbfn
2269 for thumb_ext in self._write_thumbnails(info_dict, thumb_fn_temp):
2270 thumb_filename_temp = replace_extension(thumb_fn_temp, thumb_ext, info_dict.get('ext'))
2271 thumb_filename = replace_extension(thumbfn, thumb_ext, info_dict.get('ext'))
dcf64d43 2272 files_to_move[thumb_filename_temp] = thumb_filename
8222d8de 2273
732044af 2274 # Write internet shortcut files
2275 url_link = webloc_link = desktop_link = False
2276 if self.params.get('writelink', False):
2277 if sys.platform == "darwin": # macOS.
2278 webloc_link = True
2279 elif sys.platform.startswith("linux"):
2280 desktop_link = True
2281 else: # if sys.platform in ['win32', 'cygwin']:
2282 url_link = True
2283 if self.params.get('writeurllink', False):
2284 url_link = True
2285 if self.params.get('writewebloclink', False):
2286 webloc_link = True
2287 if self.params.get('writedesktoplink', False):
2288 desktop_link = True
2289
2290 if url_link or webloc_link or desktop_link:
2291 if 'webpage_url' not in info_dict:
2292 self.report_error('Cannot write internet shortcut file because the "webpage_url" field is missing in the media information')
2293 return
2294 ascii_url = iri_to_uri(info_dict['webpage_url'])
2295
2296 def _write_link_file(extension, template, newline, embed_filename):
0202b52a 2297 linkfn = replace_extension(full_filename, extension, info_dict.get('ext'))
10e3742e 2298 if self.params.get('overwrites', True) and os.path.exists(encodeFilename(linkfn)):
732044af 2299 self.to_screen('[info] Internet shortcut is already present')
2300 else:
2301 try:
2302 self.to_screen('[info] Writing internet shortcut to: ' + linkfn)
2303 with io.open(encodeFilename(to_high_limit_path(linkfn)), 'w', encoding='utf-8', newline=newline) as linkfile:
2304 template_vars = {'url': ascii_url}
2305 if embed_filename:
2306 template_vars['filename'] = linkfn[:-(len(extension) + 1)]
2307 linkfile.write(template % template_vars)
2308 except (OSError, IOError):
2309 self.report_error('Cannot write internet shortcut ' + linkfn)
2310 return False
2311 return True
2312
2313 if url_link:
2314 if not _write_link_file('url', DOT_URL_LINK_TEMPLATE, '\r\n', embed_filename=False):
2315 return
2316 if webloc_link:
2317 if not _write_link_file('webloc', DOT_WEBLOC_LINK_TEMPLATE, '\n', embed_filename=False):
2318 return
2319 if desktop_link:
2320 if not _write_link_file('desktop', DOT_DESKTOP_LINK_TEMPLATE, '\n', embed_filename=True):
2321 return
2322
2323 # Download
2324 must_record_download_archive = False
de6000d9 2325 if not skip_dl:
4340deca 2326 try:
0202b52a 2327
6b591b29 2328 def existing_file(*filepaths):
2329 ext = info_dict.get('ext')
2330 final_ext = self.params.get('final_ext', ext)
2331 existing_files = []
2332 for file in orderedSet(filepaths):
2333 if final_ext != ext:
2334 converted = replace_extension(file, final_ext, ext)
2335 if os.path.exists(encodeFilename(converted)):
2336 existing_files.append(converted)
2337 if os.path.exists(encodeFilename(file)):
2338 existing_files.append(file)
2339
2340 if not existing_files or self.params.get('overwrites', False):
2341 for file in orderedSet(existing_files):
2342 self.report_file_delete(file)
2343 os.remove(encodeFilename(file))
2344 return None
2345
2346 self.report_file_already_downloaded(existing_files[0])
2347 info_dict['ext'] = os.path.splitext(existing_files[0])[1][1:]
2348 return existing_files[0]
0202b52a 2349
2350 success = True
4340deca
P
2351 if info_dict.get('requested_formats') is not None:
2352 downloaded = []
d47aeb22 2353 merger = FFmpegMergerPP(self)
63ad4d43 2354 if self.params.get('allow_unplayable_formats'):
2355 self.report_warning(
2356 'You have requested merging of multiple formats '
2357 'while also allowing unplayable formats to be downloaded. '
2358 'The formats won\'t be merged to prevent data corruption.')
2359 elif not merger.available:
2360 self.report_warning(
2361 'You have requested merging of multiple formats but ffmpeg is not installed. '
2362 'The formats won\'t be merged.')
81cd954a
S
2363
2364 def compatible_formats(formats):
d03cfdce 2365 # TODO: some formats actually allow this (mkv, webm, ogg, mp4), but not all of them.
2366 video_formats = [format for format in formats if format.get('vcodec') != 'none']
2367 audio_formats = [format for format in formats if format.get('acodec') != 'none']
2368 if len(video_formats) > 2 or len(audio_formats) > 2:
2369 return False
2370
81cd954a 2371 # Check extension
d03cfdce 2372 exts = set(format.get('ext') for format in formats)
2373 COMPATIBLE_EXTS = (
2374 set(('mp3', 'mp4', 'm4a', 'm4p', 'm4b', 'm4r', 'm4v', 'ismv', 'isma')),
2375 set(('webm',)),
2376 )
2377 for ext_sets in COMPATIBLE_EXTS:
2378 if ext_sets.issuperset(exts):
2379 return True
81cd954a
S
2380 # TODO: Check acodec/vcodec
2381 return False
2382
2383 requested_formats = info_dict['requested_formats']
0202b52a 2384 old_ext = info_dict['ext']
4d971a16 2385 if self.params.get('merge_output_format') is None:
2386 if not compatible_formats(requested_formats):
2387 info_dict['ext'] = 'mkv'
2388 self.report_warning(
2389 'Requested formats are incompatible for merge and will be merged into mkv.')
2390 if (info_dict['ext'] == 'webm'
2391 and self.params.get('writethumbnail', False)
2392 and info_dict.get('thumbnails')):
2393 info_dict['ext'] = 'mkv'
2394 self.report_warning(
2395 'webm doesn\'t support embedding a thumbnail, mkv will be used.')
0202b52a 2396
2397 def correct_ext(filename):
2398 filename_real_ext = os.path.splitext(filename)[1][1:]
2399 filename_wo_ext = (
2400 os.path.splitext(filename)[0]
2401 if filename_real_ext == old_ext
2402 else filename)
2403 return '%s.%s' % (filename_wo_ext, info_dict['ext'])
2404
38c6902b 2405 # Ensure filename always has a correct extension for successful merge
0202b52a 2406 full_filename = correct_ext(full_filename)
2407 temp_filename = correct_ext(temp_filename)
2408 dl_filename = existing_file(full_filename, temp_filename)
1ea24129 2409 info_dict['__real_download'] = False
0202b52a 2410 if dl_filename is None:
81cd954a 2411 for f in requested_formats:
5b5fbc08
JMF
2412 new_info = dict(info_dict)
2413 new_info.update(f)
c5c9bf0c 2414 fname = prepend_extension(
de6000d9 2415 self.prepare_filename(new_info, 'temp'),
c5c9bf0c 2416 'f%s' % f['format_id'], new_info['ext'])
e92caff5 2417 if not self._ensure_dir_exists(fname):
c5c9bf0c 2418 return
5b5fbc08 2419 downloaded.append(fname)
a9e7f546 2420 partial_success, real_download = dl(fname, new_info)
1ea24129 2421 info_dict['__real_download'] = info_dict['__real_download'] or real_download
5b5fbc08 2422 success = success and partial_success
63ad4d43 2423 if merger.available and not self.params.get('allow_unplayable_formats'):
efabc161 2424 info_dict['__postprocessors'].append(merger)
1ea24129 2425 info_dict['__files_to_merge'] = downloaded
2426 # Even if there were no downloads, it is being merged only now
2427 info_dict['__real_download'] = True
42bb0c59 2428 else:
2429 for file in downloaded:
2430 files_to_move[file] = None
4340deca
P
2431 else:
2432 # Just a single file
0202b52a 2433 dl_filename = existing_file(full_filename, temp_filename)
2434 if dl_filename is None:
2435 success, real_download = dl(temp_filename, info_dict)
2436 info_dict['__real_download'] = real_download
2437
0202b52a 2438 dl_filename = dl_filename or temp_filename
c571435f 2439 info_dict['__finaldir'] = os.path.dirname(os.path.abspath(encodeFilename(full_filename)))
0202b52a 2440
4340deca 2441 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
7960b056 2442 self.report_error('unable to download video data: %s' % error_to_compat_str(err))
4340deca
P
2443 return
2444 except (OSError, IOError) as err:
2445 raise UnavailableVideoError(err)
2446 except (ContentTooShortError, ) as err:
2447 self.report_error('content too short (expected %s bytes and served %s)' % (err.expected, err.downloaded))
2448 return
8222d8de 2449
de6000d9 2450 if success and full_filename != '-':
6271f1ca 2451 # Fixup content
62cd676c
PH
2452 fixup_policy = self.params.get('fixup')
2453 if fixup_policy is None:
2454 fixup_policy = 'detect_or_warn'
2455
e4172ac9 2456 INSTALL_FFMPEG_MESSAGE = 'Install ffmpeg to fix this automatically.'
d1e4a464 2457
6271f1ca
PH
2458 stretched_ratio = info_dict.get('stretched_ratio')
2459 if stretched_ratio is not None and stretched_ratio != 1:
6271f1ca
PH
2460 if fixup_policy == 'warn':
2461 self.report_warning('%s: Non-uniform pixel ratio (%s)' % (
2462 info_dict['id'], stretched_ratio))
2463 elif fixup_policy == 'detect_or_warn':
2464 stretched_pp = FFmpegFixupStretchedPP(self)
2465 if stretched_pp.available:
6271f1ca
PH
2466 info_dict['__postprocessors'].append(stretched_pp)
2467 else:
2468 self.report_warning(
d1e4a464
S
2469 '%s: Non-uniform pixel ratio (%s). %s'
2470 % (info_dict['id'], stretched_ratio, INSTALL_FFMPEG_MESSAGE))
6271f1ca 2471 else:
62cd676c
PH
2472 assert fixup_policy in ('ignore', 'never')
2473
3089bc74 2474 if (info_dict.get('requested_formats') is None
6b591b29 2475 and info_dict.get('container') == 'm4a_dash'
2476 and info_dict.get('ext') == 'm4a'):
62cd676c 2477 if fixup_policy == 'warn':
d1e4a464
S
2478 self.report_warning(
2479 '%s: writing DASH m4a. '
2480 'Only some players support this container.'
2481 % info_dict['id'])
62cd676c
PH
2482 elif fixup_policy == 'detect_or_warn':
2483 fixup_pp = FFmpegFixupM4aPP(self)
2484 if fixup_pp.available:
62cd676c
PH
2485 info_dict['__postprocessors'].append(fixup_pp)
2486 else:
2487 self.report_warning(
d1e4a464
S
2488 '%s: writing DASH m4a. '
2489 'Only some players support this container. %s'
2490 % (info_dict['id'], INSTALL_FFMPEG_MESSAGE))
62cd676c
PH
2491 else:
2492 assert fixup_policy in ('ignore', 'never')
6271f1ca 2493
0a473f2f 2494 if ('protocol' in info_dict
2495 and get_suitable_downloader(info_dict, self.params).__name__ == 'HlsFD'):
f17f8651 2496 if fixup_policy == 'warn':
a02682fd 2497 self.report_warning('%s: malformed AAC bitstream detected.' % (
f17f8651 2498 info_dict['id']))
2499 elif fixup_policy == 'detect_or_warn':
2500 fixup_pp = FFmpegFixupM3u8PP(self)
2501 if fixup_pp.available:
f17f8651 2502 info_dict['__postprocessors'].append(fixup_pp)
2503 else:
2504 self.report_warning(
a02682fd 2505 '%s: malformed AAC bitstream detected. %s'
d1e4a464 2506 % (info_dict['id'], INSTALL_FFMPEG_MESSAGE))
f17f8651 2507 else:
2508 assert fixup_policy in ('ignore', 'never')
2509
8222d8de 2510 try:
23c1a667 2511 info_dict = self.post_process(dl_filename, info_dict, files_to_move)
af819c21 2512 except PostProcessingError as err:
2513 self.report_error('Postprocessing: %s' % str(err))
8222d8de 2514 return
ab8e5e51
AM
2515 try:
2516 for ph in self._post_hooks:
23c1a667 2517 ph(info_dict['filepath'])
ab8e5e51
AM
2518 except Exception as err:
2519 self.report_error('post hooks: %s' % str(err))
2520 return
2d30509f 2521 must_record_download_archive = True
2522
2523 if must_record_download_archive or self.params.get('force_write_download_archive', False):
2524 self.record_download_archive(info_dict)
c3e6ffba 2525 max_downloads = self.params.get('max_downloads')
2526 if max_downloads is not None and self._num_downloads >= int(max_downloads):
2527 raise MaxDownloadsReached()
8222d8de
JMF
2528
2529 def download(self, url_list):
2530 """Download a given list of URLs."""
de6000d9 2531 outtmpl = self.outtmpl_dict['default']
3089bc74
S
2532 if (len(url_list) > 1
2533 and outtmpl != '-'
2534 and '%' not in outtmpl
2535 and self.params.get('max_downloads') != 1):
acd69589 2536 raise SameFileError(outtmpl)
8222d8de
JMF
2537
2538 for url in url_list:
2539 try:
5f6a1245 2540 # It also downloads the videos
61aa5ba3
S
2541 res = self.extract_info(
2542 url, force_generic_extractor=self.params.get('force_generic_extractor', False))
8222d8de 2543 except UnavailableVideoError:
6febd1c1 2544 self.report_error('unable to download video')
8222d8de 2545 except MaxDownloadsReached:
8b0d7497 2546 self.to_screen('[info] Maximum number of downloaded files reached')
2547 raise
2548 except ExistingVideoReached:
d83cb531 2549 self.to_screen('[info] Encountered a file that is already in the archive, stopping due to --break-on-existing')
8b0d7497 2550 raise
2551 except RejectedVideoReached:
d83cb531 2552 self.to_screen('[info] Encountered a file that did not match filter, stopping due to --break-on-reject')
8222d8de 2553 raise
63e0be34
PH
2554 else:
2555 if self.params.get('dump_single_json', False):
277d6ff5 2556 self.post_extract(res)
75d43ca0 2557 self.to_stdout(json.dumps(res, default=repr))
8222d8de
JMF
2558
2559 return self._download_retcode
2560
1dcc4c0c 2561 def download_with_info_file(self, info_filename):
31bd3925
JMF
2562 with contextlib.closing(fileinput.FileInput(
2563 [info_filename], mode='r',
2564 openhook=fileinput.hook_encoded('utf-8'))) as f:
2565 # FileInput doesn't have a read method, we can't call json.load
498f5606 2566 info = self.filter_requested_info(json.loads('\n'.join(f)), self.params.get('clean_infojson', True))
d4943898
JMF
2567 try:
2568 self.process_ie_result(info, download=True)
498f5606 2569 except (DownloadError, EntryNotInPlaylist):
d4943898
JMF
2570 webpage_url = info.get('webpage_url')
2571 if webpage_url is not None:
6febd1c1 2572 self.report_warning('The info failed to download, trying with "%s"' % webpage_url)
d4943898
JMF
2573 return self.download([webpage_url])
2574 else:
2575 raise
2576 return self._download_retcode
1dcc4c0c 2577
cb202fd2 2578 @staticmethod
75d43ca0 2579 def filter_requested_info(info_dict, actually_filter=True):
2580 if not actually_filter:
394dcd44 2581 info_dict['epoch'] = int(time.time())
75d43ca0 2582 return info_dict
5226731e 2583 exceptions = {
498f5606 2584 'remove': ['requested_formats', 'requested_subtitles', 'requested_entries', 'filepath', 'entries'],
5226731e 2585 'keep': ['_type'],
2586 }
2587 keep_key = lambda k: k in exceptions['keep'] or not (k.startswith('_') or k in exceptions['remove'])
2588 filter_fn = lambda obj: (
a515a78d 2589 list(map(filter_fn, obj)) if isinstance(obj, (list, tuple))
2590 else obj if not isinstance(obj, dict)
2591 else dict((k, filter_fn(v)) for k, v in obj.items() if keep_key(k)))
5226731e 2592 return filter_fn(info_dict)
cb202fd2 2593
dcf64d43 2594 def run_pp(self, pp, infodict):
5bfa4862 2595 files_to_delete = []
dcf64d43 2596 if '__files_to_move' not in infodict:
2597 infodict['__files_to_move'] = {}
af819c21 2598 files_to_delete, infodict = pp.run(infodict)
5bfa4862 2599 if not files_to_delete:
dcf64d43 2600 return infodict
5bfa4862 2601
2602 if self.params.get('keepvideo', False):
2603 for f in files_to_delete:
dcf64d43 2604 infodict['__files_to_move'].setdefault(f, '')
5bfa4862 2605 else:
2606 for old_filename in set(files_to_delete):
2607 self.to_screen('Deleting original file %s (pass -k to keep)' % old_filename)
2608 try:
2609 os.remove(encodeFilename(old_filename))
2610 except (IOError, OSError):
2611 self.report_warning('Unable to remove downloaded original file')
dcf64d43 2612 if old_filename in infodict['__files_to_move']:
2613 del infodict['__files_to_move'][old_filename]
2614 return infodict
5bfa4862 2615
277d6ff5 2616 @staticmethod
2617 def post_extract(info_dict):
2618 def actual_post_extract(info_dict):
2619 if info_dict.get('_type') in ('playlist', 'multi_video'):
2620 for video_dict in info_dict.get('entries', {}):
b050d210 2621 actual_post_extract(video_dict or {})
277d6ff5 2622 return
2623
2624 if '__post_extractor' not in info_dict:
2625 return
2626 post_extractor = info_dict['__post_extractor']
2627 if post_extractor:
2628 info_dict.update(post_extractor().items())
2629 del info_dict['__post_extractor']
2630 return
2631
b050d210 2632 actual_post_extract(info_dict or {})
277d6ff5 2633
5bfa4862 2634 def pre_process(self, ie_info):
2635 info = dict(ie_info)
2636 for pp in self._pps['beforedl']:
dcf64d43 2637 info = self.run_pp(pp, info)
5bfa4862 2638 return info
2639
dcf64d43 2640 def post_process(self, filename, ie_info, files_to_move=None):
8222d8de
JMF
2641 """Run all the postprocessors on the given file."""
2642 info = dict(ie_info)
2643 info['filepath'] = filename
dcf64d43 2644 info['__files_to_move'] = files_to_move or {}
0202b52a 2645
5bfa4862 2646 for pp in ie_info.get('__postprocessors', []) + self._pps['normal']:
dcf64d43 2647 info = self.run_pp(pp, info)
2648 info = self.run_pp(MoveFilesAfterDownloadPP(self), info)
2649 del info['__files_to_move']
5bfa4862 2650 for pp in self._pps['aftermove']:
dcf64d43 2651 info = self.run_pp(pp, info)
23c1a667 2652 return info
c1c9a79c 2653
5db07df6 2654 def _make_archive_id(self, info_dict):
e9fef7ee
S
2655 video_id = info_dict.get('id')
2656 if not video_id:
2657 return
5db07df6
PH
2658 # Future-proof against any change in case
2659 # and backwards compatibility with prior versions
e9fef7ee 2660 extractor = info_dict.get('extractor_key') or info_dict.get('ie_key') # key in a playlist
7012b23c 2661 if extractor is None:
1211bb6d
S
2662 url = str_or_none(info_dict.get('url'))
2663 if not url:
2664 return
e9fef7ee
S
2665 # Try to find matching extractor for the URL and take its ie_key
2666 for ie in self._ies:
1211bb6d 2667 if ie.suitable(url):
e9fef7ee
S
2668 extractor = ie.ie_key()
2669 break
2670 else:
2671 return
d0757229 2672 return '%s %s' % (extractor.lower(), video_id)
5db07df6
PH
2673
2674 def in_download_archive(self, info_dict):
2675 fn = self.params.get('download_archive')
2676 if fn is None:
2677 return False
2678
2679 vid_id = self._make_archive_id(info_dict)
e9fef7ee 2680 if not vid_id:
7012b23c 2681 return False # Incomplete video information
5db07df6 2682
a45e8619 2683 return vid_id in self.archive
c1c9a79c
PH
2684
2685 def record_download_archive(self, info_dict):
2686 fn = self.params.get('download_archive')
2687 if fn is None:
2688 return
5db07df6
PH
2689 vid_id = self._make_archive_id(info_dict)
2690 assert vid_id
c1c9a79c 2691 with locked_file(fn, 'a', encoding='utf-8') as archive_file:
6febd1c1 2692 archive_file.write(vid_id + '\n')
a45e8619 2693 self.archive.add(vid_id)
dd82ffea 2694
8c51aa65 2695 @staticmethod
8abeeb94 2696 def format_resolution(format, default='unknown'):
fb04e403
PH
2697 if format.get('vcodec') == 'none':
2698 return 'audio only'
f49d89ee
PH
2699 if format.get('resolution') is not None:
2700 return format['resolution']
35615307
DA
2701 if format.get('width') and format.get('height'):
2702 res = '%dx%d' % (format['width'], format['height'])
2703 elif format.get('height'):
2704 res = '%sp' % format['height']
2705 elif format.get('width'):
388ae76b 2706 res = '%dx?' % format['width']
8c51aa65 2707 else:
8abeeb94 2708 res = default
8c51aa65
JMF
2709 return res
2710
c57f7757
PH
2711 def _format_note(self, fdict):
2712 res = ''
2713 if fdict.get('ext') in ['f4f', 'f4m']:
2714 res += '(unsupported) '
32f90364
PH
2715 if fdict.get('language'):
2716 if res:
2717 res += ' '
9016d76f 2718 res += '[%s] ' % fdict['language']
c57f7757
PH
2719 if fdict.get('format_note') is not None:
2720 res += fdict['format_note'] + ' '
2721 if fdict.get('tbr') is not None:
2722 res += '%4dk ' % fdict['tbr']
2723 if fdict.get('container') is not None:
2724 if res:
2725 res += ', '
2726 res += '%s container' % fdict['container']
3089bc74
S
2727 if (fdict.get('vcodec') is not None
2728 and fdict.get('vcodec') != 'none'):
c57f7757
PH
2729 if res:
2730 res += ', '
2731 res += fdict['vcodec']
91c7271a 2732 if fdict.get('vbr') is not None:
c57f7757
PH
2733 res += '@'
2734 elif fdict.get('vbr') is not None and fdict.get('abr') is not None:
2735 res += 'video@'
2736 if fdict.get('vbr') is not None:
2737 res += '%4dk' % fdict['vbr']
fbb21cf5 2738 if fdict.get('fps') is not None:
5d583bdf
S
2739 if res:
2740 res += ', '
2741 res += '%sfps' % fdict['fps']
c57f7757
PH
2742 if fdict.get('acodec') is not None:
2743 if res:
2744 res += ', '
2745 if fdict['acodec'] == 'none':
2746 res += 'video only'
2747 else:
2748 res += '%-5s' % fdict['acodec']
2749 elif fdict.get('abr') is not None:
2750 if res:
2751 res += ', '
2752 res += 'audio'
2753 if fdict.get('abr') is not None:
2754 res += '@%3dk' % fdict['abr']
2755 if fdict.get('asr') is not None:
2756 res += ' (%5dHz)' % fdict['asr']
2757 if fdict.get('filesize') is not None:
2758 if res:
2759 res += ', '
2760 res += format_bytes(fdict['filesize'])
9732d77e
PH
2761 elif fdict.get('filesize_approx') is not None:
2762 if res:
2763 res += ', '
2764 res += '~' + format_bytes(fdict['filesize_approx'])
c57f7757 2765 return res
91c7271a 2766
76d321f6 2767 def _format_note_table(self, f):
2768 def join_fields(*vargs):
2769 return ', '.join((val for val in vargs if val != ''))
2770
2771 return join_fields(
2772 'UNSUPPORTED' if f.get('ext') in ('f4f', 'f4m') else '',
2773 format_field(f, 'language', '[%s]'),
2774 format_field(f, 'format_note'),
2775 format_field(f, 'container', ignore=(None, f.get('ext'))),
2776 format_field(f, 'asr', '%5dHz'))
2777
c57f7757 2778 def list_formats(self, info_dict):
94badb25 2779 formats = info_dict.get('formats', [info_dict])
76d321f6 2780 new_format = self.params.get('listformats_table', False)
2781 if new_format:
2782 table = [
2783 [
2784 format_field(f, 'format_id'),
2785 format_field(f, 'ext'),
2786 self.format_resolution(f),
2787 format_field(f, 'fps', '%d'),
2788 '|',
2789 format_field(f, 'filesize', ' %s', func=format_bytes) + format_field(f, 'filesize_approx', '~%s', func=format_bytes),
2790 format_field(f, 'tbr', '%4dk'),
52a8a1e1 2791 shorten_protocol_name(f.get('protocol', '').replace("native", "n")),
76d321f6 2792 '|',
2793 format_field(f, 'vcodec', default='unknown').replace('none', ''),
2794 format_field(f, 'vbr', '%4dk'),
2795 format_field(f, 'acodec', default='unknown').replace('none', ''),
2796 format_field(f, 'abr', '%3dk'),
2797 format_field(f, 'asr', '%5dHz'),
2798 self._format_note_table(f)]
2799 for f in formats
2800 if f.get('preference') is None or f['preference'] >= -1000]
2801 header_line = ['ID', 'EXT', 'RESOLUTION', 'FPS', '|', ' FILESIZE', ' TBR', 'PROTO',
2802 '|', 'VCODEC', ' VBR', 'ACODEC', ' ABR', ' ASR', 'NOTE']
2803 else:
2804 table = [
2805 [
2806 format_field(f, 'format_id'),
2807 format_field(f, 'ext'),
2808 self.format_resolution(f),
2809 self._format_note(f)]
2810 for f in formats
2811 if f.get('preference') is None or f['preference'] >= -1000]
2812 header_line = ['format code', 'extension', 'resolution', 'note']
57dd9a8f 2813
cfb56d1a 2814 self.to_screen(
76d321f6 2815 '[info] Available formats for %s:\n%s' % (info_dict['id'], render_table(
2816 header_line,
2817 table,
2818 delim=new_format,
2819 extraGap=(0 if new_format else 1),
2820 hideEmpty=new_format)))
cfb56d1a
PH
2821
2822 def list_thumbnails(self, info_dict):
2823 thumbnails = info_dict.get('thumbnails')
2824 if not thumbnails:
b7b72db9 2825 self.to_screen('[info] No thumbnails present for %s' % info_dict['id'])
2826 return
cfb56d1a
PH
2827
2828 self.to_screen(
2829 '[info] Thumbnails for %s:' % info_dict['id'])
2830 self.to_screen(render_table(
2831 ['ID', 'width', 'height', 'URL'],
2832 [[t['id'], t.get('width', 'unknown'), t.get('height', 'unknown'), t['url']] for t in thumbnails]))
dca08720 2833
360e1ca5 2834 def list_subtitles(self, video_id, subtitles, name='subtitles'):
a504ced0 2835 if not subtitles:
360e1ca5 2836 self.to_screen('%s has no %s' % (video_id, name))
a504ced0 2837 return
a504ced0 2838 self.to_screen(
edab9dbf
JMF
2839 'Available %s for %s:' % (name, video_id))
2840 self.to_screen(render_table(
2841 ['Language', 'formats'],
2842 [[lang, ', '.join(f['ext'] for f in reversed(formats))]
2843 for lang, formats in subtitles.items()]))
a504ced0 2844
dca08720
PH
2845 def urlopen(self, req):
2846 """ Start an HTTP download """
82d8a8b6 2847 if isinstance(req, compat_basestring):
67dda517 2848 req = sanitized_Request(req)
19a41fc6 2849 return self._opener.open(req, timeout=self._socket_timeout)
dca08720
PH
2850
2851 def print_debug_header(self):
2852 if not self.params.get('verbose'):
2853 return
62fec3b2 2854
4192b51c 2855 if type('') is not compat_str:
067aa17e 2856 # Python 2.6 on SLES11 SP1 (https://github.com/ytdl-org/youtube-dl/issues/3326)
4192b51c
PH
2857 self.report_warning(
2858 'Your Python is broken! Update to a newer and supported version')
2859
c6afed48
PH
2860 stdout_encoding = getattr(
2861 sys.stdout, 'encoding', 'missing (%s)' % type(sys.stdout).__name__)
b0472057 2862 encoding_str = (
734f90bb
PH
2863 '[debug] Encodings: locale %s, fs %s, out %s, pref %s\n' % (
2864 locale.getpreferredencoding(),
2865 sys.getfilesystemencoding(),
c6afed48 2866 stdout_encoding,
b0472057 2867 self.get_encoding()))
4192b51c 2868 write_string(encoding_str, encoding=None)
734f90bb 2869
e5813e53 2870 source = (
2871 '(exe)' if hasattr(sys, 'frozen')
2872 else '(zip)' if isinstance(globals().get('__loader__'), zipimporter)
2873 else '(source)' if os.path.basename(sys.argv[0]) == '__main__.py'
2874 else '')
2875 self._write_string('[debug] yt-dlp version %s %s\n' % (__version__, source))
e0986e31 2876 if _LAZY_LOADER:
f74980cb 2877 self._write_string('[debug] Lazy loading extractors enabled\n')
2878 if _PLUGIN_CLASSES:
2879 self._write_string(
2880 '[debug] Plugin Extractors: %s\n' % [ie.ie_key() for ie in _PLUGIN_CLASSES])
dca08720
PH
2881 try:
2882 sp = subprocess.Popen(
2883 ['git', 'rev-parse', '--short', 'HEAD'],
2884 stdout=subprocess.PIPE, stderr=subprocess.PIPE,
2885 cwd=os.path.dirname(os.path.abspath(__file__)))
f5b1bca9 2886 out, err = process_communicate_or_kill(sp)
dca08720
PH
2887 out = out.decode().strip()
2888 if re.match('[0-9a-f]+', out):
f74980cb 2889 self._write_string('[debug] Git HEAD: %s\n' % out)
70a1165b 2890 except Exception:
dca08720
PH
2891 try:
2892 sys.exc_clear()
70a1165b 2893 except Exception:
dca08720 2894 pass
b300cda4
S
2895
2896 def python_implementation():
2897 impl_name = platform.python_implementation()
2898 if impl_name == 'PyPy' and hasattr(sys, 'pypy_version_info'):
2899 return impl_name + ' version %d.%d.%d' % sys.pypy_version_info[:3]
2900 return impl_name
2901
e5813e53 2902 self._write_string('[debug] Python version %s (%s %s) - %s\n' % (
2903 platform.python_version(),
2904 python_implementation(),
2905 platform.architecture()[0],
b300cda4 2906 platform_name()))
d28b5171 2907
73fac4e9 2908 exe_versions = FFmpegPostProcessor.get_versions(self)
4c83c967 2909 exe_versions['rtmpdump'] = rtmpdump_version()
feee8d32 2910 exe_versions['phantomjs'] = PhantomJSwrapper._version()
d28b5171
PH
2911 exe_str = ', '.join(
2912 '%s %s' % (exe, v)
2913 for exe, v in sorted(exe_versions.items())
2914 if v
2915 )
2916 if not exe_str:
2917 exe_str = 'none'
2918 self._write_string('[debug] exe versions: %s\n' % exe_str)
dca08720
PH
2919
2920 proxy_map = {}
2921 for handler in self._opener.handlers:
2922 if hasattr(handler, 'proxies'):
2923 proxy_map.update(handler.proxies)
734f90bb 2924 self._write_string('[debug] Proxy map: ' + compat_str(proxy_map) + '\n')
dca08720 2925
58b1f00d
PH
2926 if self.params.get('call_home', False):
2927 ipaddr = self.urlopen('https://yt-dl.org/ip').read().decode('utf-8')
2928 self._write_string('[debug] Public IP address: %s\n' % ipaddr)
f5546c0b 2929 return
58b1f00d
PH
2930 latest_version = self.urlopen(
2931 'https://yt-dl.org/latest/version').read().decode('utf-8')
2932 if version_tuple(latest_version) > version_tuple(__version__):
2933 self.report_warning(
2934 'You are using an outdated version (newest version: %s)! '
2935 'See https://yt-dl.org/update if you need help updating.' %
2936 latest_version)
2937
e344693b 2938 def _setup_opener(self):
6ad14cab 2939 timeout_val = self.params.get('socket_timeout')
19a41fc6 2940 self._socket_timeout = 600 if timeout_val is None else float(timeout_val)
6ad14cab 2941
dca08720
PH
2942 opts_cookiefile = self.params.get('cookiefile')
2943 opts_proxy = self.params.get('proxy')
2944
2945 if opts_cookiefile is None:
2946 self.cookiejar = compat_cookiejar.CookieJar()
2947 else:
590bc6f6 2948 opts_cookiefile = expand_path(opts_cookiefile)
1bab3437 2949 self.cookiejar = YoutubeDLCookieJar(opts_cookiefile)
dca08720 2950 if os.access(opts_cookiefile, os.R_OK):
1d88b3e6 2951 self.cookiejar.load(ignore_discard=True, ignore_expires=True)
dca08720 2952
6a3f4c3f 2953 cookie_processor = YoutubeDLCookieProcessor(self.cookiejar)
dca08720
PH
2954 if opts_proxy is not None:
2955 if opts_proxy == '':
2956 proxies = {}
2957 else:
2958 proxies = {'http': opts_proxy, 'https': opts_proxy}
2959 else:
2960 proxies = compat_urllib_request.getproxies()
067aa17e 2961 # Set HTTPS proxy to HTTP one if given (https://github.com/ytdl-org/youtube-dl/issues/805)
dca08720
PH
2962 if 'http' in proxies and 'https' not in proxies:
2963 proxies['https'] = proxies['http']
91410c9b 2964 proxy_handler = PerRequestProxyHandler(proxies)
a0ddb8a2
PH
2965
2966 debuglevel = 1 if self.params.get('debug_printtraffic') else 0
be4a824d
PH
2967 https_handler = make_HTTPS_handler(self.params, debuglevel=debuglevel)
2968 ydlh = YoutubeDLHandler(self.params, debuglevel=debuglevel)
fca6dba8 2969 redirect_handler = YoutubeDLRedirectHandler()
8b172c2e 2970 data_handler = compat_urllib_request_DataHandler()
6240b0a2
JMF
2971
2972 # When passing our own FileHandler instance, build_opener won't add the
2973 # default FileHandler and allows us to disable the file protocol, which
2974 # can be used for malicious purposes (see
067aa17e 2975 # https://github.com/ytdl-org/youtube-dl/issues/8227)
6240b0a2
JMF
2976 file_handler = compat_urllib_request.FileHandler()
2977
2978 def file_open(*args, **kwargs):
7a5c1cfe 2979 raise compat_urllib_error.URLError('file:// scheme is explicitly disabled in yt-dlp for security reasons')
6240b0a2
JMF
2980 file_handler.file_open = file_open
2981
2982 opener = compat_urllib_request.build_opener(
fca6dba8 2983 proxy_handler, https_handler, cookie_processor, ydlh, redirect_handler, data_handler, file_handler)
2461f79d 2984
dca08720
PH
2985 # Delete the default user-agent header, which would otherwise apply in
2986 # cases where our custom HTTP handler doesn't come into play
067aa17e 2987 # (See https://github.com/ytdl-org/youtube-dl/issues/1309 for details)
dca08720
PH
2988 opener.addheaders = []
2989 self._opener = opener
62fec3b2
PH
2990
2991 def encode(self, s):
2992 if isinstance(s, bytes):
2993 return s # Already encoded
2994
2995 try:
2996 return s.encode(self.get_encoding())
2997 except UnicodeEncodeError as err:
2998 err.reason = err.reason + '. Check your system encoding configuration or use the --encoding option.'
2999 raise
3000
3001 def get_encoding(self):
3002 encoding = self.params.get('encoding')
3003 if encoding is None:
3004 encoding = preferredencoding()
3005 return encoding
ec82d85a 3006
de6000d9 3007 def _write_thumbnails(self, info_dict, filename): # return the extensions
6c4fd172 3008 write_all = self.params.get('write_all_thumbnails', False)
3009 thumbnails = []
3010 if write_all or self.params.get('writethumbnail', False):
0202b52a 3011 thumbnails = info_dict.get('thumbnails') or []
6c4fd172 3012 multiple = write_all and len(thumbnails) > 1
ec82d85a 3013
0202b52a 3014 ret = []
6c4fd172 3015 for t in thumbnails[::1 if write_all else -1]:
ec82d85a 3016 thumb_ext = determine_ext(t['url'], 'jpg')
6c4fd172 3017 suffix = '%s.' % t['id'] if multiple else ''
3018 thumb_display_id = '%s ' % t['id'] if multiple else ''
dcf64d43 3019 t['filepath'] = thumb_filename = replace_extension(filename, suffix + thumb_ext, info_dict.get('ext'))
ec82d85a 3020
0c3d0f51 3021 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(thumb_filename)):
de6000d9 3022 ret.append(suffix + thumb_ext)
ec82d85a
PH
3023 self.to_screen('[%s] %s: Thumbnail %sis already present' %
3024 (info_dict['extractor'], info_dict['id'], thumb_display_id))
3025 else:
5ef7d9bd 3026 self.to_screen('[%s] %s: Downloading thumbnail %s ...' %
ec82d85a
PH
3027 (info_dict['extractor'], info_dict['id'], thumb_display_id))
3028 try:
3029 uf = self.urlopen(t['url'])
d3d89c32 3030 with open(encodeFilename(thumb_filename), 'wb') as thumbf:
ec82d85a 3031 shutil.copyfileobj(uf, thumbf)
de6000d9 3032 ret.append(suffix + thumb_ext)
ec82d85a
PH
3033 self.to_screen('[%s] %s: Writing thumbnail %sto: %s' %
3034 (info_dict['extractor'], info_dict['id'], thumb_display_id, thumb_filename))
3035 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
3036 self.report_warning('Unable to download thumbnail "%s": %s' %
9b9c5355 3037 (t['url'], error_to_compat_str(err)))
6c4fd172 3038 if ret and not write_all:
3039 break
0202b52a 3040 return ret