]> jfr.im git - yt-dlp.git/blame - yt_dlp/YoutubeDL.py
[compat] Remove unnecessary code
[yt-dlp.git] / yt_dlp / YoutubeDL.py
CommitLineData
cc52de43 1#!/usr/bin/env python3
dcdb292f 2# coding: utf-8
8222d8de 3
6febd1c1 4from __future__ import absolute_import, unicode_literals
8222d8de 5
26e63931 6import collections
31bd3925 7import contextlib
317f7ab6 8import copy
9d2ecdbc 9import datetime
c1c9a79c 10import errno
31bd3925 11import fileinput
8222d8de 12import io
b82f815f 13import itertools
8694c600 14import json
62fec3b2 15import locale
083c9df9 16import operator
8222d8de 17import os
dca08720 18import platform
8222d8de
JMF
19import re
20import shutil
dca08720 21import subprocess
8222d8de 22import sys
21cd8fae 23import tempfile
8222d8de 24import time
67134eab 25import tokenize
8222d8de 26import traceback
75822ca7 27import random
8222d8de 28
961ea474 29from string import ascii_letters
e5813e53 30from zipimport import zipimporter
961ea474 31
8c25f81b 32from .compat import (
82d8a8b6 33 compat_basestring,
dca08720 34 compat_cookiejar,
003c69a8 35 compat_get_terminal_size,
4f026faf 36 compat_kwargs,
d0d9ade4 37 compat_numeric_types,
e9c0cdd3 38 compat_os_name,
ce02ed60 39 compat_str,
67134eab 40 compat_tokenize_tokenize,
ce02ed60
PH
41 compat_urllib_error,
42 compat_urllib_request,
8b172c2e 43 compat_urllib_request_DataHandler,
8c25f81b
PH
44)
45from .utils import (
eedb7ba5
S
46 age_restricted,
47 args_to_str,
ce02ed60
PH
48 ContentTooShortError,
49 date_from_str,
50 DateRange,
acd69589 51 DEFAULT_OUTTMPL,
ce02ed60 52 determine_ext,
b5559424 53 determine_protocol,
732044af 54 DOT_DESKTOP_LINK_TEMPLATE,
55 DOT_URL_LINK_TEMPLATE,
56 DOT_WEBLOC_LINK_TEMPLATE,
ce02ed60 57 DownloadError,
c0384f22 58 encode_compat_str,
ce02ed60 59 encodeFilename,
498f5606 60 EntryNotInPlaylist,
a06916d9 61 error_to_compat_str,
8b0d7497 62 ExistingVideoReached,
590bc6f6 63 expand_path,
ce02ed60 64 ExtractorError,
e29663c6 65 float_or_none,
02dbf93f 66 format_bytes,
76d321f6 67 format_field,
752cda38 68 STR_FORMAT_RE,
525ef922 69 formatSeconds,
773f291d 70 GeoRestrictedError,
b0249bca 71 HEADRequest,
c9969434 72 int_or_none,
732044af 73 iri_to_uri,
773f291d 74 ISO3166Utils,
56a8fb4f 75 LazyList,
ce02ed60 76 locked_file,
0202b52a 77 make_dir,
dca08720 78 make_HTTPS_handler,
ce02ed60 79 MaxDownloadsReached,
3158150c 80 network_exceptions,
cd6fc19e 81 orderedSet,
a06916d9 82 OUTTMPL_TYPES,
b7ab0590 83 PagedList,
083c9df9 84 parse_filesize,
91410c9b 85 PerRequestProxyHandler,
dca08720 86 platform_name,
eedb7ba5 87 PostProcessingError,
ce02ed60 88 preferredencoding,
eedb7ba5 89 prepend_extension,
a06916d9 90 process_communicate_or_kill,
51fb4995 91 register_socks_protocols,
a06916d9 92 RejectedVideoReached,
cfb56d1a 93 render_table,
eedb7ba5 94 replace_extension,
ce02ed60
PH
95 SameFileError,
96 sanitize_filename,
1bb5c511 97 sanitize_path,
dcf77cf1 98 sanitize_url,
67dda517 99 sanitized_Request,
e5660ee6 100 std_headers,
1211bb6d 101 str_or_none,
e29663c6 102 strftime_or_none,
ce02ed60 103 subtitles_filename,
51d9739f 104 ThrottledDownload,
732044af 105 to_high_limit_path,
324ad820 106 traverse_obj,
6033d980 107 try_get,
ce02ed60 108 UnavailableVideoError,
29eb5174 109 url_basename,
58b1f00d 110 version_tuple,
ce02ed60
PH
111 write_json_file,
112 write_string,
1bab3437 113 YoutubeDLCookieJar,
6a3f4c3f 114 YoutubeDLCookieProcessor,
dca08720 115 YoutubeDLHandler,
fca6dba8 116 YoutubeDLRedirectHandler,
ce02ed60 117)
a0e07d31 118from .cache import Cache
52a8a1e1 119from .extractor import (
120 gen_extractor_classes,
121 get_info_extractor,
122 _LAZY_LOADER,
123 _PLUGIN_CLASSES
124)
4c54b89e 125from .extractor.openload import PhantomJSwrapper
52a8a1e1 126from .downloader import (
127 get_suitable_downloader,
128 shorten_protocol_name
129)
4c83c967 130from .downloader.rtmp import rtmpdump_version
4f026faf 131from .postprocessor import (
e36d50c5 132 get_postprocessor,
133 FFmpegFixupDurationPP,
f17f8651 134 FFmpegFixupM3u8PP,
62cd676c 135 FFmpegFixupM4aPP,
6271f1ca 136 FFmpegFixupStretchedPP,
e36d50c5 137 FFmpegFixupTimestampPP,
4f026faf
PH
138 FFmpegMergerPP,
139 FFmpegPostProcessor,
0202b52a 140 MoveFilesAfterDownloadPP,
4f026faf 141)
dca08720 142from .version import __version__
8222d8de 143
e9c0cdd3
YCH
144if compat_os_name == 'nt':
145 import ctypes
146
2459b6e1 147
8222d8de
JMF
148class YoutubeDL(object):
149 """YoutubeDL class.
150
151 YoutubeDL objects are the ones responsible of downloading the
152 actual video file and writing it to disk if the user has requested
153 it, among some other tasks. In most cases there should be one per
154 program. As, given a video URL, the downloader doesn't know how to
155 extract all the needed information, task that InfoExtractors do, it
156 has to pass the URL to one of them.
157
158 For this, YoutubeDL objects have a method that allows
159 InfoExtractors to be registered in a given order. When it is passed
160 a URL, the YoutubeDL object handles it to the first InfoExtractor it
161 finds that reports being able to handle it. The InfoExtractor extracts
162 all the information about the video or videos the URL refers to, and
163 YoutubeDL process the extracted information, possibly using a File
164 Downloader to download the video.
165
166 YoutubeDL objects accept a lot of parameters. In order not to saturate
167 the object constructor with arguments, it receives a dictionary of
168 options instead. These options are available through the params
169 attribute for the InfoExtractors to use. The YoutubeDL also
170 registers itself as the downloader in charge for the InfoExtractors
171 that are added to it, so this is a "mutual registration".
172
173 Available options:
174
175 username: Username for authentication purposes.
176 password: Password for authentication purposes.
180940e0 177 videopassword: Password for accessing a video.
1da50aa3
S
178 ap_mso: Adobe Pass multiple-system operator identifier.
179 ap_username: Multiple-system operator account username.
180 ap_password: Multiple-system operator account password.
8222d8de
JMF
181 usenetrc: Use netrc for authentication instead.
182 verbose: Print additional info to stdout.
183 quiet: Do not print messages to stdout.
ad8915b7 184 no_warnings: Do not print out anything for warnings.
53c18592 185 forceprint: A list of templates to force print
186 forceurl: Force printing final URL. (Deprecated)
187 forcetitle: Force printing title. (Deprecated)
188 forceid: Force printing ID. (Deprecated)
189 forcethumbnail: Force printing thumbnail URL. (Deprecated)
190 forcedescription: Force printing description. (Deprecated)
191 forcefilename: Force printing final filename. (Deprecated)
192 forceduration: Force printing duration. (Deprecated)
8694c600 193 forcejson: Force printing info_dict as JSON.
63e0be34
PH
194 dump_single_json: Force printing the info_dict of the whole playlist
195 (or video) as a single JSON line.
c25228e5 196 force_write_download_archive: Force writing download archive regardless
197 of 'skip_download' or 'simulate'.
8222d8de 198 simulate: Do not download the video files.
eb8a4433 199 format: Video format code. see "FORMAT SELECTION" for more details.
63ad4d43 200 allow_unplayable_formats: Allow unplayable formats to be extracted and downloaded.
b7da73eb 201 ignore_no_formats_error: Ignore "No video formats" error. Usefull for
202 extracting metadata even if the video is not actually
203 available for download (experimental)
c25228e5 204 format_sort: How to sort the video formats. see "Sorting Formats"
205 for more details.
206 format_sort_force: Force the given format_sort. see "Sorting Formats"
207 for more details.
208 allow_multiple_video_streams: Allow multiple video streams to be merged
209 into a single file
210 allow_multiple_audio_streams: Allow multiple audio streams to be merged
211 into a single file
0ba692ac 212 check_formats Whether to test if the formats are downloadable.
213 Can be True (check all), False (check none)
214 or None (check only if requested by extractor)
4524baf0 215 paths: Dictionary of output paths. The allowed keys are 'home'
216 'temp' and the keys of OUTTMPL_TYPES (in utils.py)
de6000d9 217 outtmpl: Dictionary of templates for output names. Allowed keys
4524baf0 218 are 'default' and the keys of OUTTMPL_TYPES (in utils.py).
219 A string a also accepted for backward compatibility
a820dc72
RA
220 outtmpl_na_placeholder: Placeholder for unavailable meta fields.
221 restrictfilenames: Do not allow "&" and spaces in file names
222 trim_file_name: Limit length of filename (extension excluded)
4524baf0 223 windowsfilenames: Force the filenames to be windows compatible
a820dc72 224 ignoreerrors: Do not stop on download errors
7a5c1cfe 225 (Default True when running yt-dlp,
a820dc72 226 but False when directly accessing YoutubeDL class)
26e2805c 227 skip_playlist_after_errors: Number of allowed failures until the rest of
228 the playlist is skipped
d22dec74 229 force_generic_extractor: Force downloader to use the generic extractor
0c3d0f51 230 overwrites: Overwrite all video and metadata files if True,
231 overwrite only non-video files if None
232 and don't overwrite any file if False
8222d8de
JMF
233 playliststart: Playlist item to start at.
234 playlistend: Playlist item to end at.
c14e88f0 235 playlist_items: Specific indices of playlist to download.
ff815fe6 236 playlistreverse: Download playlist items in reverse order.
75822ca7 237 playlistrandom: Download playlist items in random order.
8222d8de
JMF
238 matchtitle: Download only matching titles.
239 rejecttitle: Reject downloads for matching titles.
8bf9319e 240 logger: Log messages to a logging.Logger instance.
8222d8de
JMF
241 logtostderr: Log messages to stderr instead of stdout.
242 writedescription: Write the video description to a .description file
243 writeinfojson: Write the video description to a .info.json file
75d43ca0 244 clean_infojson: Remove private fields from the infojson
06167fbb 245 writecomments: Extract video comments. This will not be written to disk
246 unless writeinfojson is also given
1fb07d10 247 writeannotations: Write the video annotations to a .annotations.xml file
8222d8de 248 writethumbnail: Write the thumbnail image to a file
c25228e5 249 allow_playlist_files: Whether to write playlists' description, infojson etc
250 also to disk when using the 'write*' options
ec82d85a 251 write_all_thumbnails: Write all thumbnail formats to files
732044af 252 writelink: Write an internet shortcut file, depending on the
253 current platform (.url/.webloc/.desktop)
254 writeurllink: Write a Windows internet shortcut file (.url)
255 writewebloclink: Write a macOS internet shortcut file (.webloc)
256 writedesktoplink: Write a Linux internet shortcut file (.desktop)
8222d8de 257 writesubtitles: Write the video subtitles to a file
741dd8ea 258 writeautomaticsub: Write the automatically generated subtitles to a file
c32b0aab 259 allsubtitles: Deprecated - Use subtitlelangs = ['all']
260 Downloads all the subtitles of the video
0b7f3118 261 (requires writesubtitles or writeautomaticsub)
8222d8de 262 listsubtitles: Lists all available subtitles for the video
a504ced0 263 subtitlesformat: The format code for subtitles
c32b0aab 264 subtitleslangs: List of languages of the subtitles to download (can be regex).
265 The list may contain "all" to refer to all the available
266 subtitles. The language can be prefixed with a "-" to
267 exclude it from the requested languages. Eg: ['all', '-live_chat']
8222d8de
JMF
268 keepvideo: Keep the video file after post-processing
269 daterange: A DateRange object, download only if the upload_date is in the range.
270 skip_download: Skip the actual download of the video file
c35f9e72 271 cachedir: Location of the cache files in the filesystem.
a0e07d31 272 False to disable filesystem cache.
47192f92 273 noplaylist: Download single video instead of a playlist if in doubt.
8dbe9899
PH
274 age_limit: An integer representing the user's age in years.
275 Unsuitable videos for the given age are skipped.
5fe18bdb
PH
276 min_views: An integer representing the minimum view count the video
277 must have in order to not be skipped.
278 Videos without view count information are always
279 downloaded. None for no limit.
280 max_views: An integer representing the maximum view count.
281 Videos that are more popular than that are not
282 downloaded.
283 Videos without view count information are always
284 downloaded. None for no limit.
285 download_archive: File name of a file where all downloads are recorded.
c1c9a79c
PH
286 Videos already present in the file are not downloaded
287 again.
8a51f564 288 break_on_existing: Stop the download process after attempting to download a
289 file that is in the archive.
290 break_on_reject: Stop the download process when encountering a video that
291 has been filtered out.
292 cookiefile: File name where cookies should be read from and dumped to
a1ee09e8 293 nocheckcertificate:Do not verify SSL certificates
7e8c0af0
PH
294 prefer_insecure: Use HTTP instead of HTTPS to retrieve information.
295 At the moment, this is only supported by YouTube.
a1ee09e8 296 proxy: URL of the proxy server to use
38cce791 297 geo_verification_proxy: URL of the proxy to use for IP address verification
504f20dd 298 on geo-restricted sites.
e344693b 299 socket_timeout: Time to wait for unresponsive hosts, in seconds
0783b09b
PH
300 bidi_workaround: Work around buggy terminals without bidirectional text
301 support, using fridibi
a0ddb8a2 302 debug_printtraffic:Print out sent and received HTTP traffic
7b0817e8 303 include_ads: Download ads as well
04b4d394
PH
304 default_search: Prepend this string if an input url is not valid.
305 'auto' for elaborate guessing
62fec3b2 306 encoding: Use this encoding instead of the system-specified.
e8ee972c 307 extract_flat: Do not resolve URLs, return the immediate result.
057a5206
PH
308 Pass in 'in_playlist' to only show this behavior for
309 playlist items.
4f026faf 310 postprocessors: A list of dictionaries, each with an entry
71b640cc 311 * key: The name of the postprocessor. See
7a5c1cfe 312 yt_dlp/postprocessor/__init__.py for a list.
56d868db 313 * when: When to run the postprocessor. Can be one of
314 pre_process|before_dl|post_process|after_move.
315 Assumed to be 'post_process' if not given
ab8e5e51
AM
316 post_hooks: A list of functions that get called as the final step
317 for each video file, after all postprocessors have been
318 called. The filename will be passed as the only argument.
71b640cc
PH
319 progress_hooks: A list of functions that get called on download
320 progress, with a dictionary with the entries
5cda4eda 321 * status: One of "downloading", "error", or "finished".
ee69b99a 322 Check this first and ignore unknown values.
71b640cc 323
5cda4eda 324 If status is one of "downloading", or "finished", the
ee69b99a
PH
325 following properties may also be present:
326 * filename: The final filename (always present)
5cda4eda 327 * tmpfilename: The filename we're currently writing to
71b640cc
PH
328 * downloaded_bytes: Bytes on disk
329 * total_bytes: Size of the whole file, None if unknown
5cda4eda
PH
330 * total_bytes_estimate: Guess of the eventual file size,
331 None if unavailable.
332 * elapsed: The number of seconds since download started.
71b640cc
PH
333 * eta: The estimated time in seconds, None if unknown
334 * speed: The download speed in bytes/second, None if
335 unknown
5cda4eda
PH
336 * fragment_index: The counter of the currently
337 downloaded video fragment.
338 * fragment_count: The number of fragments (= individual
339 files that will be merged)
71b640cc
PH
340
341 Progress hooks are guaranteed to be called at least once
342 (with status "finished") if the download is successful.
45598f15 343 merge_output_format: Extension to use when merging formats.
6b591b29 344 final_ext: Expected final extension; used to detect when the file was
345 already downloaded and converted. "merge_output_format" is
346 replaced by this extension when given
6271f1ca
PH
347 fixup: Automatically correct known faults of the file.
348 One of:
349 - "never": do nothing
350 - "warn": only emit a warning
351 - "detect_or_warn": check whether we can do anything
62cd676c 352 about it, warn otherwise (default)
504f20dd 353 source_address: Client-side IP address to bind to.
6ec6cb4e 354 call_home: Boolean, true iff we are allowed to contact the
7a5c1cfe 355 yt-dlp servers for debugging. (BROKEN)
1cf376f5 356 sleep_interval_requests: Number of seconds to sleep between requests
357 during extraction
7aa589a5
S
358 sleep_interval: Number of seconds to sleep before each download when
359 used alone or a lower bound of a range for randomized
360 sleep before each download (minimum possible number
361 of seconds to sleep) when used along with
362 max_sleep_interval.
363 max_sleep_interval:Upper bound of a range for randomized sleep before each
364 download (maximum possible number of seconds to sleep).
365 Must only be used along with sleep_interval.
366 Actual sleep time will be a random float from range
367 [sleep_interval; max_sleep_interval].
1cf376f5 368 sleep_interval_subtitles: Number of seconds to sleep before each subtitle download
cfb56d1a
PH
369 listformats: Print an overview of available video formats and exit.
370 list_thumbnails: Print a table of all thumbnails and exit.
347de493
PH
371 match_filter: A function that gets called with the info_dict of
372 every video.
373 If it returns a message, the video is ignored.
374 If it returns None, the video is downloaded.
375 match_filter_func in utils.py is one example for this.
7e5db8c9 376 no_color: Do not emit color codes in output.
0a840f58 377 geo_bypass: Bypass geographic restriction via faking X-Forwarded-For
504f20dd 378 HTTP header
0a840f58 379 geo_bypass_country:
773f291d
S
380 Two-letter ISO 3166-2 country code that will be used for
381 explicit geographic restriction bypassing via faking
504f20dd 382 X-Forwarded-For HTTP header
5f95927a
S
383 geo_bypass_ip_block:
384 IP range in CIDR notation that will be used similarly to
504f20dd 385 geo_bypass_country
71b640cc 386
85729c51 387 The following options determine which downloader is picked:
52a8a1e1 388 external_downloader: A dictionary of protocol keys and the executable of the
389 external downloader to use for it. The allowed protocols
390 are default|http|ftp|m3u8|dash|rtsp|rtmp|mms.
391 Set the value to 'native' to use the native downloader
392 hls_prefer_native: Deprecated - Use external_downloader = {'m3u8': 'native'}
393 or {'m3u8': 'ffmpeg'} instead.
394 Use the native HLS downloader instead of ffmpeg/avconv
bf09af3a
S
395 if True, otherwise use ffmpeg/avconv if False, otherwise
396 use downloader suggested by extractor if None.
53ed7066 397 compat_opts: Compatibility options. See "Differences in default behavior".
3acf6d38 398 The following options do not work when used through the API:
399 filename, abort-on-error, multistreams, no-live-chat,
400 no-playlist-metafiles. Refer __init__.py for their implementation
fe7e0c98 401
8222d8de 402 The following parameters are not used by YoutubeDL itself, they are used by
7a5c1cfe 403 the downloader (see yt_dlp/downloader/common.py):
51d9739f 404 nopart, updatetime, buffersize, ratelimit, throttledratelimit, min_filesize,
405 max_filesize, test, noresizebuffer, retries, continuedl, noprogress, consoletitle,
406 xattr_set_filesize, external_downloader_args, hls_use_mpegts, http_chunk_size.
76b1bd67
JMF
407
408 The following options are used by the post processors:
d4a24f40 409 prefer_ffmpeg: If False, use avconv instead of ffmpeg if both are available,
e4172ac9 410 otherwise prefer ffmpeg. (avconv support is deprecated)
c0b7d117
S
411 ffmpeg_location: Location of the ffmpeg/avconv binary; either the path
412 to the binary or its containing directory.
43820c03 413 postprocessor_args: A dictionary of postprocessor/executable keys (in lower case)
414 and a list of additional command-line arguments for the
415 postprocessor/executable. The dict can also have "PP+EXE" keys
416 which are used when the given exe is used by the given PP.
417 Use 'default' as the name for arguments to passed to all PP
e409895f 418
419 The following options are used by the extractors:
62bff2c1 420 extractor_retries: Number of times to retry for known errors
421 dynamic_mpd: Whether to process dynamic DASH manifests (default: True)
e409895f 422 hls_split_discontinuity: Split HLS playlists to different formats at
62bff2c1 423 discontinuities such as ad breaks (default: False)
5d3a0e79 424 extractor_args: A dictionary of arguments to be passed to the extractors.
425 See "EXTRACTOR ARGUMENTS" for details.
426 Eg: {'youtube': {'skip': ['dash', 'hls']}}
427 youtube_include_dash_manifest: Deprecated - Use extractor_args instead.
428 If True (default), DASH manifests and related
62bff2c1 429 data will be downloaded and processed by extractor.
430 You can reduce network I/O by disabling it if you don't
431 care about DASH. (only for youtube)
5d3a0e79 432 youtube_include_hls_manifest: Deprecated - Use extractor_args instead.
433 If True (default), HLS manifests and related
62bff2c1 434 data will be downloaded and processed by extractor.
435 You can reduce network I/O by disabling it if you don't
436 care about HLS. (only for youtube)
8222d8de
JMF
437 """
438
c9969434
S
439 _NUMERIC_FIELDS = set((
440 'width', 'height', 'tbr', 'abr', 'asr', 'vbr', 'fps', 'filesize', 'filesize_approx',
441 'timestamp', 'upload_year', 'upload_month', 'upload_day',
442 'duration', 'view_count', 'like_count', 'dislike_count', 'repost_count',
443 'average_rating', 'comment_count', 'age_limit',
444 'start_time', 'end_time',
445 'chapter_number', 'season_number', 'episode_number',
446 'track_number', 'disc_number', 'release_year',
447 'playlist_index',
448 ))
449
8222d8de
JMF
450 params = None
451 _ies = []
56d868db 452 _pps = {'pre_process': [], 'before_dl': [], 'after_move': [], 'post_process': []}
0202b52a 453 __prepare_filename_warned = False
1cf376f5 454 _first_webpage_request = True
8222d8de
JMF
455 _download_retcode = None
456 _num_downloads = None
30a074c2 457 _playlist_level = 0
458 _playlist_urls = set()
8222d8de
JMF
459 _screen_file = None
460
3511266b 461 def __init__(self, params=None, auto_init=True):
8222d8de 462 """Create a FileDownloader object with the given options."""
e9f9a10f
JMF
463 if params is None:
464 params = {}
8222d8de 465 self._ies = []
56c73665 466 self._ies_instances = {}
56d868db 467 self._pps = {'pre_process': [], 'before_dl': [], 'after_move': [], 'post_process': []}
0202b52a 468 self.__prepare_filename_warned = False
1cf376f5 469 self._first_webpage_request = True
ab8e5e51 470 self._post_hooks = []
933605d7 471 self._progress_hooks = []
8222d8de
JMF
472 self._download_retcode = 0
473 self._num_downloads = 0
474 self._screen_file = [sys.stdout, sys.stderr][params.get('logtostderr', False)]
0783b09b 475 self._err_file = sys.stderr
4abf617b
S
476 self.params = {
477 # Default parameters
478 'nocheckcertificate': False,
479 }
480 self.params.update(params)
a0e07d31 481 self.cache = Cache(self)
34308b30 482
a61f4b28 483 if sys.version_info < (3, 6):
484 self.report_warning(
0181adef 485 'Python version %d.%d is not supported! Please update to Python 3.6 or above' % sys.version_info[:2])
a61f4b28 486
be5df5ee
S
487 def check_deprecated(param, option, suggestion):
488 if self.params.get(param) is not None:
53ed7066 489 self.report_warning('%s is deprecated. Use %s instead' % (option, suggestion))
be5df5ee
S
490 return True
491 return False
492
493 if check_deprecated('cn_verification_proxy', '--cn-verification-proxy', '--geo-verification-proxy'):
38cce791
YCH
494 if self.params.get('geo_verification_proxy') is None:
495 self.params['geo_verification_proxy'] = self.params['cn_verification_proxy']
496
0d1bb027 497 check_deprecated('autonumber', '--auto-number', '-o "%(autonumber)s-%(title)s.%(ext)s"')
498 check_deprecated('usetitle', '--title', '-o "%(title)s-%(id)s.%(ext)s"')
53ed7066 499 check_deprecated('useid', '--id', '-o "%(id)s.%(ext)s"')
0d1bb027 500
501 for msg in self.params.get('warnings', []):
502 self.report_warning(msg)
503
6b591b29 504 if self.params.get('final_ext'):
505 if self.params.get('merge_output_format'):
506 self.report_warning('--merge-output-format will be ignored since --remux-video or --recode-video is given')
507 self.params['merge_output_format'] = self.params['final_ext']
508
b9d973be 509 if 'overwrites' in self.params and self.params['overwrites'] is None:
510 del self.params['overwrites']
511
0783b09b 512 if params.get('bidi_workaround', False):
1c088fa8
PH
513 try:
514 import pty
515 master, slave = pty.openpty()
003c69a8 516 width = compat_get_terminal_size().columns
1c088fa8
PH
517 if width is None:
518 width_args = []
519 else:
520 width_args = ['-w', str(width)]
5d681e96 521 sp_kwargs = dict(
1c088fa8
PH
522 stdin=subprocess.PIPE,
523 stdout=slave,
524 stderr=self._err_file)
5d681e96
PH
525 try:
526 self._output_process = subprocess.Popen(
527 ['bidiv'] + width_args, **sp_kwargs
528 )
529 except OSError:
5d681e96
PH
530 self._output_process = subprocess.Popen(
531 ['fribidi', '-c', 'UTF-8'] + width_args, **sp_kwargs)
532 self._output_channel = os.fdopen(master, 'rb')
1c088fa8 533 except OSError as ose:
66e7ace1 534 if ose.errno == errno.ENOENT:
6febd1c1 535 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
536 else:
537 raise
0783b09b 538
3089bc74
S
539 if (sys.platform != 'win32'
540 and sys.getfilesystemencoding() in ['ascii', 'ANSI_X3.4-1968']
541 and not params.get('restrictfilenames', False)):
e9137224 542 # Unicode filesystem API will throw errors (#1474, #13027)
34308b30 543 self.report_warning(
6febd1c1 544 'Assuming --restrict-filenames since file system encoding '
1b725173 545 'cannot encode all characters. '
6febd1c1 546 'Set the LC_ALL environment variable to fix this.')
4a98cdbf 547 self.params['restrictfilenames'] = True
34308b30 548
de6000d9 549 self.outtmpl_dict = self.parse_outtmpl()
486dd09e 550
187986a8 551 # Creating format selector here allows us to catch syntax errors before the extraction
552 self.format_selector = (
553 None if self.params.get('format') is None
554 else self.build_format_selector(self.params['format']))
555
dca08720
PH
556 self._setup_opener()
557
4cd0a709 558 """Preload the archive, if any is specified"""
559 def preload_download_archive(fn):
560 if fn is None:
561 return False
0760b0a7 562 self.write_debug('Loading archive file %r\n' % fn)
4cd0a709 563 try:
564 with locked_file(fn, 'r', encoding='utf-8') as archive_file:
565 for line in archive_file:
566 self.archive.add(line.strip())
567 except IOError as ioe:
568 if ioe.errno != errno.ENOENT:
569 raise
570 return False
571 return True
572
573 self.archive = set()
574 preload_download_archive(self.params.get('download_archive'))
575
3511266b
PH
576 if auto_init:
577 self.print_debug_header()
578 self.add_default_info_extractors()
579
4f026faf 580 for pp_def_raw in self.params.get('postprocessors', []):
4f026faf 581 pp_def = dict(pp_def_raw)
fd7cfb64 582 when = pp_def.pop('when', 'post_process')
583 pp_class = get_postprocessor(pp_def.pop('key'))
4f026faf 584 pp = pp_class(self, **compat_kwargs(pp_def))
5bfa4862 585 self.add_post_processor(pp, when=when)
4f026faf 586
ab8e5e51
AM
587 for ph in self.params.get('post_hooks', []):
588 self.add_post_hook(ph)
589
71b640cc
PH
590 for ph in self.params.get('progress_hooks', []):
591 self.add_progress_hook(ph)
592
51fb4995
YCH
593 register_socks_protocols()
594
7d4111ed
PH
595 def warn_if_short_id(self, argv):
596 # short YouTube ID starting with dash?
597 idxs = [
598 i for i, a in enumerate(argv)
599 if re.match(r'^-[0-9A-Za-z_-]{10}$', a)]
600 if idxs:
601 correct_argv = (
7a5c1cfe 602 ['yt-dlp']
3089bc74
S
603 + [a for i, a in enumerate(argv) if i not in idxs]
604 + ['--'] + [argv[i] for i in idxs]
7d4111ed
PH
605 )
606 self.report_warning(
607 'Long argument string detected. '
608 'Use -- to separate parameters and URLs, like this:\n%s\n' %
609 args_to_str(correct_argv))
610
8222d8de
JMF
611 def add_info_extractor(self, ie):
612 """Add an InfoExtractor object to the end of the list."""
613 self._ies.append(ie)
e52d7f85
JMF
614 if not isinstance(ie, type):
615 self._ies_instances[ie.ie_key()] = ie
616 ie.set_downloader(self)
8222d8de 617
56c73665
JMF
618 def get_info_extractor(self, ie_key):
619 """
620 Get an instance of an IE with name ie_key, it will try to get one from
621 the _ies list, if there's no instance it will create a new one and add
622 it to the extractor list.
623 """
624 ie = self._ies_instances.get(ie_key)
625 if ie is None:
626 ie = get_info_extractor(ie_key)()
627 self.add_info_extractor(ie)
628 return ie
629
023fa8c4
JMF
630 def add_default_info_extractors(self):
631 """
632 Add the InfoExtractors returned by gen_extractors to the end of the list
633 """
e52d7f85 634 for ie in gen_extractor_classes():
023fa8c4
JMF
635 self.add_info_extractor(ie)
636
56d868db 637 def add_post_processor(self, pp, when='post_process'):
8222d8de 638 """Add a PostProcessor object to the end of the chain."""
5bfa4862 639 self._pps[when].append(pp)
8222d8de
JMF
640 pp.set_downloader(self)
641
ab8e5e51
AM
642 def add_post_hook(self, ph):
643 """Add the post hook"""
644 self._post_hooks.append(ph)
645
933605d7
JMF
646 def add_progress_hook(self, ph):
647 """Add the progress hook (currently only for the file downloader)"""
648 self._progress_hooks.append(ph)
8ab470f1 649
1c088fa8 650 def _bidi_workaround(self, message):
5d681e96 651 if not hasattr(self, '_output_channel'):
1c088fa8
PH
652 return message
653
5d681e96 654 assert hasattr(self, '_output_process')
11b85ce6 655 assert isinstance(message, compat_str)
6febd1c1
PH
656 line_count = message.count('\n') + 1
657 self._output_process.stdin.write((message + '\n').encode('utf-8'))
5d681e96 658 self._output_process.stdin.flush()
6febd1c1 659 res = ''.join(self._output_channel.readline().decode('utf-8')
9e1a5b84 660 for _ in range(line_count))
6febd1c1 661 return res[:-len('\n')]
1c088fa8 662
734f90bb 663 def _write_string(self, s, out=None):
b58ddb32 664 write_string(s, out=out, encoding=self.params.get('encoding'))
734f90bb 665
848887eb 666 def to_stdout(self, message, skip_eol=False, quiet=False):
0760b0a7 667 """Print message to stdout"""
8bf9319e 668 if self.params.get('logger'):
43afe285 669 self.params['logger'].debug(message)
835a1478 670 elif not quiet or self.params.get('verbose'):
671 self._write_string(
672 '%s%s' % (self._bidi_workaround(message), ('' if skip_eol else '\n')),
673 self._err_file if quiet else self._screen_file)
8222d8de
JMF
674
675 def to_stderr(self, message):
0760b0a7 676 """Print message to stderr"""
11b85ce6 677 assert isinstance(message, compat_str)
8bf9319e 678 if self.params.get('logger'):
43afe285
IB
679 self.params['logger'].error(message)
680 else:
835a1478 681 self._write_string('%s\n' % self._bidi_workaround(message), self._err_file)
8222d8de 682
1e5b9a95
PH
683 def to_console_title(self, message):
684 if not self.params.get('consoletitle', False):
685 return
4bede0d8
C
686 if compat_os_name == 'nt':
687 if ctypes.windll.kernel32.GetConsoleWindow():
688 # c_wchar_p() might not be necessary if `message` is
689 # already of type unicode()
690 ctypes.windll.kernel32.SetConsoleTitleW(ctypes.c_wchar_p(message))
1e5b9a95 691 elif 'TERM' in os.environ:
b46696bd 692 self._write_string('\033]0;%s\007' % message, self._screen_file)
1e5b9a95 693
bdde425c
PH
694 def save_console_title(self):
695 if not self.params.get('consoletitle', False):
696 return
94c3442e
S
697 if self.params.get('simulate', False):
698 return
4bede0d8 699 if compat_os_name != 'nt' and 'TERM' in os.environ:
efd6c574 700 # Save the title on stack
734f90bb 701 self._write_string('\033[22;0t', self._screen_file)
bdde425c
PH
702
703 def restore_console_title(self):
704 if not self.params.get('consoletitle', False):
705 return
94c3442e
S
706 if self.params.get('simulate', False):
707 return
4bede0d8 708 if compat_os_name != 'nt' and 'TERM' in os.environ:
efd6c574 709 # Restore the title from stack
734f90bb 710 self._write_string('\033[23;0t', self._screen_file)
bdde425c
PH
711
712 def __enter__(self):
713 self.save_console_title()
714 return self
715
716 def __exit__(self, *args):
717 self.restore_console_title()
f89197d7 718
dca08720 719 if self.params.get('cookiefile') is not None:
1bab3437 720 self.cookiejar.save(ignore_discard=True, ignore_expires=True)
bdde425c 721
8222d8de
JMF
722 def trouble(self, message=None, tb=None):
723 """Determine action to take when a download problem appears.
724
725 Depending on if the downloader has been configured to ignore
726 download errors or not, this method may throw an exception or
727 not when errors are found, after printing the message.
728
729 tb, if given, is additional traceback information.
730 """
731 if message is not None:
732 self.to_stderr(message)
733 if self.params.get('verbose'):
734 if tb is None:
735 if sys.exc_info()[0]: # if .trouble has been called from an except block
6febd1c1 736 tb = ''
8222d8de 737 if hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
6febd1c1 738 tb += ''.join(traceback.format_exception(*sys.exc_info()[1].exc_info))
c0384f22 739 tb += encode_compat_str(traceback.format_exc())
8222d8de
JMF
740 else:
741 tb_data = traceback.format_list(traceback.extract_stack())
6febd1c1 742 tb = ''.join(tb_data)
c19bc311 743 if tb:
744 self.to_stderr(tb)
8222d8de
JMF
745 if not self.params.get('ignoreerrors', False):
746 if sys.exc_info()[0] and hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
747 exc_info = sys.exc_info()[1].exc_info
748 else:
749 exc_info = sys.exc_info()
750 raise DownloadError(message, exc_info)
751 self._download_retcode = 1
752
0760b0a7 753 def to_screen(self, message, skip_eol=False):
754 """Print message to stdout if not in quiet mode"""
755 self.to_stdout(
756 message, skip_eol, quiet=self.params.get('quiet', False))
757
8222d8de
JMF
758 def report_warning(self, message):
759 '''
760 Print the message to stderr, it will be prefixed with 'WARNING:'
761 If stderr is a tty file the 'WARNING:' will be colored
762 '''
6d07ce01
JMF
763 if self.params.get('logger') is not None:
764 self.params['logger'].warning(message)
8222d8de 765 else:
ad8915b7
PH
766 if self.params.get('no_warnings'):
767 return
e9c0cdd3 768 if not self.params.get('no_color') and self._err_file.isatty() and compat_os_name != 'nt':
6d07ce01
JMF
769 _msg_header = '\033[0;33mWARNING:\033[0m'
770 else:
771 _msg_header = 'WARNING:'
772 warning_message = '%s %s' % (_msg_header, message)
773 self.to_stderr(warning_message)
8222d8de
JMF
774
775 def report_error(self, message, tb=None):
776 '''
777 Do the same as trouble, but prefixes the message with 'ERROR:', colored
778 in red if stderr is a tty file.
779 '''
e9c0cdd3 780 if not self.params.get('no_color') and self._err_file.isatty() and compat_os_name != 'nt':
6febd1c1 781 _msg_header = '\033[0;31mERROR:\033[0m'
8222d8de 782 else:
6febd1c1
PH
783 _msg_header = 'ERROR:'
784 error_message = '%s %s' % (_msg_header, message)
8222d8de
JMF
785 self.trouble(error_message, tb)
786
0760b0a7 787 def write_debug(self, message):
788 '''Log debug message or Print message to stderr'''
789 if not self.params.get('verbose', False):
790 return
791 message = '[debug] %s' % message
792 if self.params.get('logger'):
793 self.params['logger'].debug(message)
794 else:
795 self._write_string('%s\n' % message)
796
8222d8de
JMF
797 def report_file_already_downloaded(self, file_name):
798 """Report file has already been fully downloaded."""
799 try:
6febd1c1 800 self.to_screen('[download] %s has already been downloaded' % file_name)
ce02ed60 801 except UnicodeEncodeError:
6febd1c1 802 self.to_screen('[download] The file has already been downloaded')
8222d8de 803
0c3d0f51 804 def report_file_delete(self, file_name):
805 """Report that existing file will be deleted."""
806 try:
c25228e5 807 self.to_screen('Deleting existing file %s' % file_name)
0c3d0f51 808 except UnicodeEncodeError:
c25228e5 809 self.to_screen('Deleting existing file')
0c3d0f51 810
de6000d9 811 def parse_outtmpl(self):
812 outtmpl_dict = self.params.get('outtmpl', {})
813 if not isinstance(outtmpl_dict, dict):
814 outtmpl_dict = {'default': outtmpl_dict}
815 outtmpl_dict.update({
816 k: v for k, v in DEFAULT_OUTTMPL.items()
817 if not outtmpl_dict.get(k)})
818 for key, val in outtmpl_dict.items():
819 if isinstance(val, bytes):
820 self.report_warning(
821 'Parameter outtmpl is bytes, but should be a unicode string. '
822 'Put from __future__ import unicode_literals at the top of your code file or consider switching to Python 3.x.')
823 return outtmpl_dict
824
21cd8fae 825 def get_output_path(self, dir_type='', filename=None):
826 paths = self.params.get('paths', {})
827 assert isinstance(paths, dict)
828 path = os.path.join(
829 expand_path(paths.get('home', '').strip()),
830 expand_path(paths.get(dir_type, '').strip()) if dir_type else '',
831 filename or '')
832
833 # Temporary fix for #4787
834 # 'Treat' all problem characters by passing filename through preferredencoding
835 # to workaround encoding issues with subprocess on python2 @ Windows
836 if sys.version_info < (3, 0) and sys.platform == 'win32':
837 path = encodeFilename(path, True).decode(preferredencoding())
838 return sanitize_path(path, force=self.params.get('windowsfilenames'))
839
76a264ac 840 @staticmethod
841 def validate_outtmpl(tmpl):
842 ''' @return None or Exception object '''
843 try:
844 re.sub(
845 STR_FORMAT_RE.format(''),
846 lambda mobj: ('%' if not mobj.group('has_key') else '') + mobj.group(0),
847 tmpl
848 ) % collections.defaultdict(int)
849 return None
850 except ValueError as err:
851 return err
852
143db31d 853 def prepare_outtmpl(self, outtmpl, info_dict, sanitize=None):
854 """ Make the template and info_dict suitable for substitution (outtmpl % info_dict)"""
752cda38 855 info_dict = dict(info_dict)
a439a3a4 856 na = self.params.get('outtmpl_na_placeholder', 'NA')
143db31d 857
752cda38 858 info_dict['duration_string'] = ( # %(duration>%H-%M-%S)s is wrong if duration > 24hrs
53c18592 859 formatSeconds(info_dict['duration'], '-' if sanitize else ':')
143db31d 860 if info_dict.get('duration', None) is not None
861 else None)
752cda38 862 info_dict['epoch'] = int(time.time())
863 info_dict['autonumber'] = self.params.get('autonumber_start', 1) - 1 + self._num_downloads
864 if info_dict.get('resolution') is None:
865 info_dict['resolution'] = self.format_resolution(info_dict, default=None)
143db31d 866
143db31d 867 # For fields playlist_index and autonumber convert all occurrences
868 # of %(field)s to %(field)0Nd for backward compatibility
869 field_size_compat_map = {
752cda38 870 'playlist_index': len(str(info_dict.get('_last_playlist_index') or '')),
871 'autonumber': self.params.get('autonumber_size') or 5,
143db31d 872 }
752cda38 873
385a27fa 874 TMPL_DICT = {}
875 EXTERNAL_FORMAT_RE = re.compile(STR_FORMAT_RE.format('[^)]*'))
876 MATH_FUNCTIONS = {
877 '+': float.__add__,
878 '-': float.__sub__,
879 }
e625be0d 880 # Field is of the form key1.key2...
881 # where keys (except first) can be string, int or slice
385a27fa 882 FIELD_RE = r'\w+(?:\.(?:\w+|{num}|{num}?(?::{num}?){{1,2}}))*'.format(num=r'(?:-?\d+)')
883 MATH_FIELD_RE = r'''{field}|{num}'''.format(field=FIELD_RE, num=r'-?\d+(?:.\d+)?')
884 MATH_OPERATORS_RE = r'(?:%s)' % '|'.join(map(re.escape, MATH_FUNCTIONS.keys()))
e625be0d 885 INTERNAL_FORMAT_RE = re.compile(r'''(?x)
886 (?P<negate>-)?
385a27fa 887 (?P<fields>{field})
888 (?P<maths>(?:{math_op}{math_field})*)
e625be0d 889 (?:>(?P<strf_format>.+?))?
890 (?:\|(?P<default>.*?))?
385a27fa 891 $'''.format(field=FIELD_RE, math_op=MATH_OPERATORS_RE, math_field=MATH_FIELD_RE))
752cda38 892
76a264ac 893 get_key = lambda k: traverse_obj(
894 info_dict, k.split('.'), is_user_input=True, traverse_string=True)
895
752cda38 896 def get_value(mdict):
897 # Object traversal
76a264ac 898 value = get_key(mdict['fields'])
752cda38 899 # Negative
900 if mdict['negate']:
901 value = float_or_none(value)
902 if value is not None:
903 value *= -1
904 # Do maths
385a27fa 905 offset_key = mdict['maths']
906 if offset_key:
752cda38 907 value = float_or_none(value)
908 operator = None
385a27fa 909 while offset_key:
910 item = re.match(
911 MATH_FIELD_RE if operator else MATH_OPERATORS_RE,
912 offset_key).group(0)
913 offset_key = offset_key[len(item):]
914 if operator is None:
752cda38 915 operator = MATH_FUNCTIONS[item]
385a27fa 916 continue
917 item, multiplier = (item[1:], -1) if item[0] == '-' else (item, 1)
918 offset = float_or_none(item)
919 if offset is None:
920 offset = float_or_none(get_key(item))
921 try:
922 value = operator(value, multiplier * offset)
923 except (TypeError, ZeroDivisionError):
924 return None
925 operator = None
752cda38 926 # Datetime formatting
927 if mdict['strf_format']:
928 value = strftime_or_none(value, mdict['strf_format'])
929
930 return value
931
932 def create_key(outer_mobj):
933 if not outer_mobj.group('has_key'):
934 return '%{}'.format(outer_mobj.group(0))
935
936 key = outer_mobj.group('key')
937 fmt = outer_mobj.group('format')
938 mobj = re.match(INTERNAL_FORMAT_RE, key)
939 if mobj is None:
9fea350f 940 value, default, mobj = None, na, {'fields': ''}
752cda38 941 else:
e625be0d 942 mobj = mobj.groupdict()
752cda38 943 default = mobj['default'] if mobj['default'] is not None else na
944 value = get_value(mobj)
945
946 if fmt == 's' and value is not None and key in field_size_compat_map.keys():
947 fmt = '0{:d}d'.format(field_size_compat_map[key])
948
949 value = default if value is None else value
752cda38 950
76a264ac 951 if fmt == 'c':
952 value = compat_str(value)
953 if value is None:
954 value, fmt = default, 's'
955 else:
956 value = value[0]
957 elif fmt[-1] not in 'rs': # numeric
a439a3a4 958 value = float_or_none(value)
752cda38 959 if value is None:
960 value, fmt = default, 's'
961 if sanitize:
962 if fmt[-1] == 'r':
963 # If value is an object, sanitize might convert it to a string
964 # So we convert it to repr first
965 value, fmt = repr(value), '%ss' % fmt[:-1]
639f1cea 966 if fmt[-1] in 'csr':
9fea350f 967 value = sanitize(mobj['fields'].split('.')[-1], value)
968 key += '\0%s' % fmt
385a27fa 969 TMPL_DICT[key] = value
752cda38 970 return '%({key}){fmt}'.format(key=key, fmt=fmt)
971
385a27fa 972 return EXTERNAL_FORMAT_RE.sub(create_key, outtmpl), TMPL_DICT
143db31d 973
de6000d9 974 def _prepare_filename(self, info_dict, tmpl_type='default'):
8222d8de 975 try:
586a91b6 976 sanitize = lambda k, v: sanitize_filename(
45598aab 977 compat_str(v),
1bb5c511 978 restricted=self.params.get('restrictfilenames'),
40df485f 979 is_id=(k == 'id' or k.endswith('_id')))
de6000d9 980 outtmpl = self.outtmpl_dict.get(tmpl_type, self.outtmpl_dict['default'])
143db31d 981 outtmpl, template_dict = self.prepare_outtmpl(outtmpl, info_dict, sanitize)
d0d9ade4 982
15da37c7
S
983 # expand_path translates '%%' into '%' and '$$' into '$'
984 # correspondingly that is not what we want since we need to keep
985 # '%%' intact for template dict substitution step. Working around
986 # with boundary-alike separator hack.
961ea474 987 sep = ''.join([random.choice(ascii_letters) for _ in range(32)])
15da37c7
S
988 outtmpl = outtmpl.replace('%%', '%{0}%'.format(sep)).replace('$$', '${0}$'.format(sep))
989
990 # outtmpl should be expand_path'ed before template dict substitution
991 # because meta fields may contain env variables we don't want to
992 # be expanded. For example, for outtmpl "%(title)s.%(ext)s" and
993 # title "Hello $PATH", we don't want `$PATH` to be expanded.
994 filename = expand_path(outtmpl).replace(sep, '') % template_dict
995
143db31d 996 force_ext = OUTTMPL_TYPES.get(tmpl_type)
de6000d9 997 if force_ext is not None:
752cda38 998 filename = replace_extension(filename, force_ext, info_dict.get('ext'))
de6000d9 999
bdc3fd2f
U
1000 # https://github.com/blackjack4494/youtube-dlc/issues/85
1001 trim_file_name = self.params.get('trim_file_name', False)
1002 if trim_file_name:
1003 fn_groups = filename.rsplit('.')
1004 ext = fn_groups[-1]
1005 sub_ext = ''
1006 if len(fn_groups) > 2:
1007 sub_ext = fn_groups[-2]
1008 filename = '.'.join(filter(None, [fn_groups[0][:trim_file_name], sub_ext, ext]))
1009
0202b52a 1010 return filename
8222d8de 1011 except ValueError as err:
6febd1c1 1012 self.report_error('Error in output template: ' + str(err) + ' (encoding: ' + repr(preferredencoding()) + ')')
8222d8de
JMF
1013 return None
1014
de6000d9 1015 def prepare_filename(self, info_dict, dir_type='', warn=False):
1016 """Generate the output filename."""
21cd8fae 1017
de6000d9 1018 filename = self._prepare_filename(info_dict, dir_type or 'default')
1019
1020 if warn and not self.__prepare_filename_warned:
21cd8fae 1021 if not self.params.get('paths'):
de6000d9 1022 pass
1023 elif filename == '-':
1024 self.report_warning('--paths is ignored when an outputting to stdout')
1025 elif os.path.isabs(filename):
1026 self.report_warning('--paths is ignored since an absolute path is given in output template')
1027 self.__prepare_filename_warned = True
1028 if filename == '-' or not filename:
1029 return filename
1030
21cd8fae 1031 return self.get_output_path(dir_type, filename)
0202b52a 1032
120fe513 1033 def _match_entry(self, info_dict, incomplete=False, silent=False):
ecdec191 1034 """ Returns None if the file should be downloaded """
8222d8de 1035
c77495e3 1036 video_title = info_dict.get('title', info_dict.get('id', 'video'))
1037
8b0d7497 1038 def check_filter():
8b0d7497 1039 if 'title' in info_dict:
1040 # This can happen when we're just evaluating the playlist
1041 title = info_dict['title']
1042 matchtitle = self.params.get('matchtitle', False)
1043 if matchtitle:
1044 if not re.search(matchtitle, title, re.IGNORECASE):
1045 return '"' + title + '" title did not match pattern "' + matchtitle + '"'
1046 rejecttitle = self.params.get('rejecttitle', False)
1047 if rejecttitle:
1048 if re.search(rejecttitle, title, re.IGNORECASE):
1049 return '"' + title + '" title matched reject pattern "' + rejecttitle + '"'
1050 date = info_dict.get('upload_date')
1051 if date is not None:
1052 dateRange = self.params.get('daterange', DateRange())
1053 if date not in dateRange:
1054 return '%s upload date is not in range %s' % (date_from_str(date).isoformat(), dateRange)
1055 view_count = info_dict.get('view_count')
1056 if view_count is not None:
1057 min_views = self.params.get('min_views')
1058 if min_views is not None and view_count < min_views:
1059 return 'Skipping %s, because it has not reached minimum view count (%d/%d)' % (video_title, view_count, min_views)
1060 max_views = self.params.get('max_views')
1061 if max_views is not None and view_count > max_views:
1062 return 'Skipping %s, because it has exceeded the maximum view count (%d/%d)' % (video_title, view_count, max_views)
1063 if age_restricted(info_dict.get('age_limit'), self.params.get('age_limit')):
1064 return 'Skipping "%s" because it is age restricted' % video_title
8b0d7497 1065
1066 if not incomplete:
1067 match_filter = self.params.get('match_filter')
1068 if match_filter is not None:
1069 ret = match_filter(info_dict)
1070 if ret is not None:
1071 return ret
1072 return None
1073
c77495e3 1074 if self.in_download_archive(info_dict):
1075 reason = '%s has already been recorded in the archive' % video_title
1076 break_opt, break_err = 'break_on_existing', ExistingVideoReached
1077 else:
1078 reason = check_filter()
1079 break_opt, break_err = 'break_on_reject', RejectedVideoReached
8b0d7497 1080 if reason is not None:
120fe513 1081 if not silent:
1082 self.to_screen('[download] ' + reason)
c77495e3 1083 if self.params.get(break_opt, False):
1084 raise break_err()
8b0d7497 1085 return reason
fe7e0c98 1086
b6c45014
JMF
1087 @staticmethod
1088 def add_extra_info(info_dict, extra_info):
1089 '''Set the keys from extra_info in info dict if they are missing'''
1090 for key, value in extra_info.items():
1091 info_dict.setdefault(key, value)
1092
58f197b7 1093 def extract_info(self, url, download=True, ie_key=None, extra_info={},
61aa5ba3 1094 process=True, force_generic_extractor=False):
41d1cca3 1095 """
1096 Return a list with a dictionary for each video extracted.
1097
1098 Arguments:
1099 url -- URL to extract
1100
1101 Keyword arguments:
1102 download -- whether to download videos during extraction
1103 ie_key -- extractor key hint
1104 extra_info -- dictionary containing the extra values to add to each result
1105 process -- whether to resolve all unresolved references (URLs, playlist items),
1106 must be True for download to work.
1107 force_generic_extractor -- force using the generic extractor
1108 """
fe7e0c98 1109
61aa5ba3 1110 if not ie_key and force_generic_extractor:
d22dec74
S
1111 ie_key = 'Generic'
1112
8222d8de 1113 if ie_key:
56c73665 1114 ies = [self.get_info_extractor(ie_key)]
8222d8de
JMF
1115 else:
1116 ies = self._ies
1117
1118 for ie in ies:
1119 if not ie.suitable(url):
1120 continue
1121
9a68de12 1122 ie_key = ie.ie_key()
1123 ie = self.get_info_extractor(ie_key)
8222d8de 1124 if not ie.working():
6febd1c1
PH
1125 self.report_warning('The program functionality for this site has been marked as broken, '
1126 'and will probably not work.')
8222d8de
JMF
1127
1128 try:
d0757229 1129 temp_id = str_or_none(
63be1aab 1130 ie.extract_id(url) if callable(getattr(ie, 'extract_id', None))
1131 else ie._match_id(url))
a0566bbf 1132 except (AssertionError, IndexError, AttributeError):
1133 temp_id = None
1134 if temp_id is not None and self.in_download_archive({'id': temp_id, 'ie_key': ie_key}):
1135 self.to_screen("[%s] %s: has already been recorded in archive" % (
1136 ie_key, temp_id))
1137 break
58f197b7 1138 return self.__extract_info(url, ie, download, extra_info, process)
a0566bbf 1139 else:
1140 self.report_error('no suitable InfoExtractor for URL %s' % url)
1141
1142 def __handle_extraction_exceptions(func):
1143 def wrapper(self, *args, **kwargs):
1144 try:
1145 return func(self, *args, **kwargs)
773f291d
S
1146 except GeoRestrictedError as e:
1147 msg = e.msg
1148 if e.countries:
1149 msg += '\nThis video is available in %s.' % ', '.join(
1150 map(ISO3166Utils.short2full, e.countries))
1151 msg += '\nYou might want to use a VPN or a proxy server (with --proxy) to workaround.'
1152 self.report_error(msg)
fb043a6e 1153 except ExtractorError as e: # An error we somewhat expected
2c74e6fa 1154 self.report_error(compat_str(e), e.format_traceback())
51d9739f 1155 except ThrottledDownload:
1156 self.to_stderr('\r')
1157 self.report_warning('The download speed is below throttle limit. Re-extracting data')
1158 return wrapper(self, *args, **kwargs)
8b0d7497 1159 except (MaxDownloadsReached, ExistingVideoReached, RejectedVideoReached):
d3e5bbf4 1160 raise
8222d8de
JMF
1161 except Exception as e:
1162 if self.params.get('ignoreerrors', False):
9b9c5355 1163 self.report_error(error_to_compat_str(e), tb=encode_compat_str(traceback.format_exc()))
8222d8de
JMF
1164 else:
1165 raise
a0566bbf 1166 return wrapper
1167
1168 @__handle_extraction_exceptions
58f197b7 1169 def __extract_info(self, url, ie, download, extra_info, process):
a0566bbf 1170 ie_result = ie.extract(url)
1171 if ie_result is None: # Finished already (backwards compatibility; listformats and friends should be moved here)
1172 return
1173 if isinstance(ie_result, list):
1174 # Backwards compatibility: old IE result format
1175 ie_result = {
1176 '_type': 'compat_list',
1177 'entries': ie_result,
1178 }
a0566bbf 1179 self.add_default_extra_info(ie_result, ie, url)
1180 if process:
1181 return self.process_ie_result(ie_result, download, extra_info)
8222d8de 1182 else:
a0566bbf 1183 return ie_result
fe7e0c98 1184
ea38e55f 1185 def add_default_extra_info(self, ie_result, ie, url):
6033d980 1186 if url is not None:
1187 self.add_extra_info(ie_result, {
1188 'webpage_url': url,
1189 'original_url': url,
1190 'webpage_url_basename': url_basename(url),
1191 })
1192 if ie is not None:
1193 self.add_extra_info(ie_result, {
1194 'extractor': ie.IE_NAME,
1195 'extractor_key': ie.ie_key(),
1196 })
ea38e55f 1197
8222d8de
JMF
1198 def process_ie_result(self, ie_result, download=True, extra_info={}):
1199 """
1200 Take the result of the ie(may be modified) and resolve all unresolved
1201 references (URLs, playlist items).
1202
1203 It will also download the videos if 'download'.
1204 Returns the resolved ie_result.
1205 """
e8ee972c
PH
1206 result_type = ie_result.get('_type', 'video')
1207
057a5206 1208 if result_type in ('url', 'url_transparent'):
134c6ea8 1209 ie_result['url'] = sanitize_url(ie_result['url'])
057a5206 1210 extract_flat = self.params.get('extract_flat', False)
3089bc74
S
1211 if ((extract_flat == 'in_playlist' and 'playlist' in extra_info)
1212 or extract_flat is True):
ecb54191 1213 info_copy = ie_result.copy()
1214 self.add_extra_info(info_copy, extra_info)
6033d980 1215 ie = try_get(ie_result.get('ie_key'), self.get_info_extractor)
1216 self.add_default_extra_info(info_copy, ie, ie_result['url'])
ecb54191 1217 self.__forced_printings(info_copy, self.prepare_filename(info_copy), incomplete=True)
e8ee972c
PH
1218 return ie_result
1219
8222d8de 1220 if result_type == 'video':
b6c45014 1221 self.add_extra_info(ie_result, extra_info)
9c2b75b5 1222 ie_result = self.process_video_result(ie_result, download=download)
28b0eb0f 1223 additional_urls = (ie_result or {}).get('additional_urls')
9c2b75b5 1224 if additional_urls:
1225 # TODO: Improve MetadataFromFieldPP to allow setting a list
1226 if isinstance(additional_urls, compat_str):
1227 additional_urls = [additional_urls]
1228 self.to_screen(
1229 '[info] %s: %d additional URL(s) requested' % (ie_result['id'], len(additional_urls)))
1230 self.write_debug('Additional URLs: "%s"' % '", "'.join(additional_urls))
1231 ie_result['additional_entries'] = [
1232 self.extract_info(
1233 url, download, extra_info,
1234 force_generic_extractor=self.params.get('force_generic_extractor'))
1235 for url in additional_urls
1236 ]
1237 return ie_result
8222d8de
JMF
1238 elif result_type == 'url':
1239 # We have to add extra_info to the results because it may be
1240 # contained in a playlist
07cce701 1241 return self.extract_info(
1242 ie_result['url'], download,
1243 ie_key=ie_result.get('ie_key'),
1244 extra_info=extra_info)
7fc3fa05
PH
1245 elif result_type == 'url_transparent':
1246 # Use the information from the embedding page
1247 info = self.extract_info(
1248 ie_result['url'], ie_key=ie_result.get('ie_key'),
1249 extra_info=extra_info, download=False, process=False)
1250
1640eb09
S
1251 # extract_info may return None when ignoreerrors is enabled and
1252 # extraction failed with an error, don't crash and return early
1253 # in this case
1254 if not info:
1255 return info
1256
412c617d
PH
1257 force_properties = dict(
1258 (k, v) for k, v in ie_result.items() if v is not None)
0396806f 1259 for f in ('_type', 'url', 'id', 'extractor', 'extractor_key', 'ie_key'):
412c617d
PH
1260 if f in force_properties:
1261 del force_properties[f]
1262 new_result = info.copy()
1263 new_result.update(force_properties)
7fc3fa05 1264
0563f7ac
S
1265 # Extracted info may not be a video result (i.e.
1266 # info.get('_type', 'video') != video) but rather an url or
1267 # url_transparent. In such cases outer metadata (from ie_result)
1268 # should be propagated to inner one (info). For this to happen
1269 # _type of info should be overridden with url_transparent. This
067aa17e 1270 # fixes issue from https://github.com/ytdl-org/youtube-dl/pull/11163.
0563f7ac
S
1271 if new_result.get('_type') == 'url':
1272 new_result['_type'] = 'url_transparent'
7fc3fa05
PH
1273
1274 return self.process_ie_result(
1275 new_result, download=download, extra_info=extra_info)
40fcba5e 1276 elif result_type in ('playlist', 'multi_video'):
30a074c2 1277 # Protect from infinite recursion due to recursively nested playlists
1278 # (see https://github.com/ytdl-org/youtube-dl/issues/27833)
1279 webpage_url = ie_result['webpage_url']
1280 if webpage_url in self._playlist_urls:
7e85e872 1281 self.to_screen(
30a074c2 1282 '[download] Skipping already downloaded playlist: %s'
1283 % ie_result.get('title') or ie_result.get('id'))
1284 return
7e85e872 1285
30a074c2 1286 self._playlist_level += 1
1287 self._playlist_urls.add(webpage_url)
bc516a3f 1288 self._sanitize_thumbnails(ie_result)
30a074c2 1289 try:
1290 return self.__process_playlist(ie_result, download)
1291 finally:
1292 self._playlist_level -= 1
1293 if not self._playlist_level:
1294 self._playlist_urls.clear()
8222d8de 1295 elif result_type == 'compat_list':
c9bf4114
PH
1296 self.report_warning(
1297 'Extractor %s returned a compat_list result. '
1298 'It needs to be updated.' % ie_result.get('extractor'))
5f6a1245 1299
8222d8de 1300 def _fixup(r):
9e1a5b84
JW
1301 self.add_extra_info(
1302 r,
9103bbc5
JMF
1303 {
1304 'extractor': ie_result['extractor'],
1305 'webpage_url': ie_result['webpage_url'],
29eb5174 1306 'webpage_url_basename': url_basename(ie_result['webpage_url']),
be97abc2 1307 'extractor_key': ie_result['extractor_key'],
9e1a5b84
JW
1308 }
1309 )
8222d8de
JMF
1310 return r
1311 ie_result['entries'] = [
b6c45014 1312 self.process_ie_result(_fixup(r), download, extra_info)
8222d8de
JMF
1313 for r in ie_result['entries']
1314 ]
1315 return ie_result
1316 else:
1317 raise Exception('Invalid result type: %s' % result_type)
1318
e92caff5 1319 def _ensure_dir_exists(self, path):
1320 return make_dir(path, self.report_error)
1321
30a074c2 1322 def __process_playlist(self, ie_result, download):
1323 # We process each entry in the playlist
1324 playlist = ie_result.get('title') or ie_result.get('id')
1325 self.to_screen('[download] Downloading playlist: %s' % playlist)
1326
498f5606 1327 if 'entries' not in ie_result:
1328 raise EntryNotInPlaylist()
1329 incomplete_entries = bool(ie_result.get('requested_entries'))
1330 if incomplete_entries:
1331 def fill_missing_entries(entries, indexes):
1332 ret = [None] * max(*indexes)
1333 for i, entry in zip(indexes, entries):
1334 ret[i - 1] = entry
1335 return ret
1336 ie_result['entries'] = fill_missing_entries(ie_result['entries'], ie_result['requested_entries'])
02fd60d3 1337
30a074c2 1338 playlist_results = []
1339
56a8fb4f 1340 playliststart = self.params.get('playliststart', 1)
30a074c2 1341 playlistend = self.params.get('playlistend')
1342 # For backwards compatibility, interpret -1 as whole list
1343 if playlistend == -1:
1344 playlistend = None
1345
1346 playlistitems_str = self.params.get('playlist_items')
1347 playlistitems = None
1348 if playlistitems_str is not None:
1349 def iter_playlistitems(format):
1350 for string_segment in format.split(','):
1351 if '-' in string_segment:
1352 start, end = string_segment.split('-')
1353 for item in range(int(start), int(end) + 1):
1354 yield int(item)
1355 else:
1356 yield int(string_segment)
1357 playlistitems = orderedSet(iter_playlistitems(playlistitems_str))
1358
1359 ie_entries = ie_result['entries']
56a8fb4f 1360 msg = (
1361 'Downloading %d videos' if not isinstance(ie_entries, list)
1362 else 'Collected %d videos; downloading %%d of them' % len(ie_entries))
1363 if not isinstance(ie_entries, (list, PagedList)):
1364 ie_entries = LazyList(ie_entries)
1365
1366 entries = []
1367 for i in playlistitems or itertools.count(playliststart):
1368 if playlistitems is None and playlistend is not None and playlistend < i:
1369 break
1370 entry = None
1371 try:
1372 entry = ie_entries[i - 1]
1373 if entry is None:
498f5606 1374 raise EntryNotInPlaylist()
56a8fb4f 1375 except (IndexError, EntryNotInPlaylist):
1376 if incomplete_entries:
1377 raise EntryNotInPlaylist()
1378 elif not playlistitems:
1379 break
1380 entries.append(entry)
120fe513 1381 try:
1382 if entry is not None:
1383 self._match_entry(entry, incomplete=True, silent=True)
1384 except (ExistingVideoReached, RejectedVideoReached):
1385 break
56a8fb4f 1386 ie_result['entries'] = entries
30a074c2 1387
56a8fb4f 1388 # Save playlist_index before re-ordering
1389 entries = [
1390 ((playlistitems[i - 1] if playlistitems else i), entry)
1391 for i, entry in enumerate(entries, 1)
1392 if entry is not None]
1393 n_entries = len(entries)
498f5606 1394
498f5606 1395 if not playlistitems and (playliststart or playlistend):
56a8fb4f 1396 playlistitems = list(range(playliststart, playliststart + n_entries))
498f5606 1397 ie_result['requested_entries'] = playlistitems
1398
1399 if self.params.get('allow_playlist_files', True):
1400 ie_copy = {
1401 'playlist': playlist,
1402 'playlist_id': ie_result.get('id'),
1403 'playlist_title': ie_result.get('title'),
1404 'playlist_uploader': ie_result.get('uploader'),
1405 'playlist_uploader_id': ie_result.get('uploader_id'),
71729754 1406 'playlist_index': 0,
498f5606 1407 }
1408 ie_copy.update(dict(ie_result))
1409
1410 if self.params.get('writeinfojson', False):
1411 infofn = self.prepare_filename(ie_copy, 'pl_infojson')
1412 if not self._ensure_dir_exists(encodeFilename(infofn)):
1413 return
1414 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(infofn)):
1415 self.to_screen('[info] Playlist metadata is already present')
1416 else:
1417 self.to_screen('[info] Writing playlist metadata as JSON to: ' + infofn)
1418 try:
1419 write_json_file(self.filter_requested_info(ie_result, self.params.get('clean_infojson', True)), infofn)
1420 except (OSError, IOError):
1421 self.report_error('Cannot write playlist metadata to JSON file ' + infofn)
1422
681de68e 1423 # TODO: This should be passed to ThumbnailsConvertor if necessary
1424 self._write_thumbnails(ie_copy, self.prepare_filename(ie_copy, 'pl_thumbnail'))
1425
498f5606 1426 if self.params.get('writedescription', False):
1427 descfn = self.prepare_filename(ie_copy, 'pl_description')
1428 if not self._ensure_dir_exists(encodeFilename(descfn)):
1429 return
1430 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(descfn)):
1431 self.to_screen('[info] Playlist description is already present')
1432 elif ie_result.get('description') is None:
1433 self.report_warning('There\'s no playlist description to write.')
1434 else:
1435 try:
1436 self.to_screen('[info] Writing playlist description to: ' + descfn)
1437 with io.open(encodeFilename(descfn), 'w', encoding='utf-8') as descfile:
1438 descfile.write(ie_result['description'])
1439 except (OSError, IOError):
1440 self.report_error('Cannot write playlist description file ' + descfn)
1441 return
30a074c2 1442
1443 if self.params.get('playlistreverse', False):
1444 entries = entries[::-1]
30a074c2 1445 if self.params.get('playlistrandom', False):
1446 random.shuffle(entries)
1447
1448 x_forwarded_for = ie_result.get('__x_forwarded_for_ip')
1449
56a8fb4f 1450 self.to_screen('[%s] playlist %s: %s' % (ie_result['extractor'], playlist, msg % n_entries))
26e2805c 1451 failures = 0
1452 max_failures = self.params.get('skip_playlist_after_errors') or float('inf')
71729754 1453 for i, entry_tuple in enumerate(entries, 1):
1454 playlist_index, entry = entry_tuple
53ed7066 1455 if 'playlist_index' in self.params.get('compat_options', []):
1456 playlist_index = playlistitems[i - 1] if playlistitems else i
30a074c2 1457 self.to_screen('[download] Downloading video %s of %s' % (i, n_entries))
1458 # This __x_forwarded_for_ip thing is a bit ugly but requires
1459 # minimal changes
1460 if x_forwarded_for:
1461 entry['__x_forwarded_for_ip'] = x_forwarded_for
1462 extra = {
1463 'n_entries': n_entries,
f59ae581 1464 '_last_playlist_index': max(playlistitems) if playlistitems else (playlistend or n_entries),
71729754 1465 'playlist_index': playlist_index,
1466 'playlist_autonumber': i,
30a074c2 1467 'playlist': playlist,
1468 'playlist_id': ie_result.get('id'),
1469 'playlist_title': ie_result.get('title'),
1470 'playlist_uploader': ie_result.get('uploader'),
1471 'playlist_uploader_id': ie_result.get('uploader_id'),
30a074c2 1472 'extractor': ie_result['extractor'],
1473 'webpage_url': ie_result['webpage_url'],
1474 'webpage_url_basename': url_basename(ie_result['webpage_url']),
1475 'extractor_key': ie_result['extractor_key'],
1476 }
1477
1478 if self._match_entry(entry, incomplete=True) is not None:
1479 continue
1480
1481 entry_result = self.__process_iterable_entry(entry, download, extra)
26e2805c 1482 if not entry_result:
1483 failures += 1
1484 if failures >= max_failures:
1485 self.report_error(
1486 'Skipping the remaining entries in playlist "%s" since %d items failed extraction' % (playlist, failures))
1487 break
30a074c2 1488 # TODO: skip failed (empty) entries?
1489 playlist_results.append(entry_result)
1490 ie_result['entries'] = playlist_results
1491 self.to_screen('[download] Finished downloading playlist: %s' % playlist)
1492 return ie_result
1493
a0566bbf 1494 @__handle_extraction_exceptions
1495 def __process_iterable_entry(self, entry, download, extra_info):
1496 return self.process_ie_result(
1497 entry, download=download, extra_info=extra_info)
1498
67134eab
JMF
1499 def _build_format_filter(self, filter_spec):
1500 " Returns a function to filter the formats according to the filter_spec "
083c9df9
PH
1501
1502 OPERATORS = {
1503 '<': operator.lt,
1504 '<=': operator.le,
1505 '>': operator.gt,
1506 '>=': operator.ge,
1507 '=': operator.eq,
1508 '!=': operator.ne,
1509 }
67134eab 1510 operator_rex = re.compile(r'''(?x)\s*
187986a8 1511 (?P<key>width|height|tbr|abr|vbr|asr|filesize|filesize_approx|fps)\s*
1512 (?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
1513 (?P<value>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?)\s*
083c9df9 1514 ''' % '|'.join(map(re.escape, OPERATORS.keys())))
187986a8 1515 m = operator_rex.fullmatch(filter_spec)
9ddb6925
S
1516 if m:
1517 try:
1518 comparison_value = int(m.group('value'))
1519 except ValueError:
1520 comparison_value = parse_filesize(m.group('value'))
1521 if comparison_value is None:
1522 comparison_value = parse_filesize(m.group('value') + 'B')
1523 if comparison_value is None:
1524 raise ValueError(
1525 'Invalid value %r in format specification %r' % (
67134eab 1526 m.group('value'), filter_spec))
9ddb6925
S
1527 op = OPERATORS[m.group('op')]
1528
083c9df9 1529 if not m:
9ddb6925
S
1530 STR_OPERATORS = {
1531 '=': operator.eq,
10d33b34
YCH
1532 '^=': lambda attr, value: attr.startswith(value),
1533 '$=': lambda attr, value: attr.endswith(value),
1534 '*=': lambda attr, value: value in attr,
9ddb6925 1535 }
187986a8 1536 str_operator_rex = re.compile(r'''(?x)\s*
1537 (?P<key>[a-zA-Z0-9._-]+)\s*
1538 (?P<negation>!\s*)?(?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
1539 (?P<value>[a-zA-Z0-9._-]+)\s*
9ddb6925 1540 ''' % '|'.join(map(re.escape, STR_OPERATORS.keys())))
187986a8 1541 m = str_operator_rex.fullmatch(filter_spec)
9ddb6925
S
1542 if m:
1543 comparison_value = m.group('value')
2cc779f4
S
1544 str_op = STR_OPERATORS[m.group('op')]
1545 if m.group('negation'):
e118a879 1546 op = lambda attr, value: not str_op(attr, value)
2cc779f4
S
1547 else:
1548 op = str_op
083c9df9 1549
9ddb6925 1550 if not m:
187986a8 1551 raise SyntaxError('Invalid filter specification %r' % filter_spec)
083c9df9
PH
1552
1553 def _filter(f):
1554 actual_value = f.get(m.group('key'))
1555 if actual_value is None:
1556 return m.group('none_inclusive')
1557 return op(actual_value, comparison_value)
67134eab
JMF
1558 return _filter
1559
0017d9ad 1560 def _default_format_spec(self, info_dict, download=True):
0017d9ad 1561
af0f7428
S
1562 def can_merge():
1563 merger = FFmpegMergerPP(self)
1564 return merger.available and merger.can_merge()
1565
91ebc640 1566 prefer_best = (
1567 not self.params.get('simulate', False)
1568 and download
1569 and (
1570 not can_merge()
19807826 1571 or info_dict.get('is_live', False)
de6000d9 1572 or self.outtmpl_dict['default'] == '-'))
53ed7066 1573 compat = (
1574 prefer_best
1575 or self.params.get('allow_multiple_audio_streams', False)
1576 or 'format-spec' in self.params.get('compat_opts', []))
91ebc640 1577
1578 return (
53ed7066 1579 'best/bestvideo+bestaudio' if prefer_best
1580 else 'bestvideo*+bestaudio/best' if not compat
91ebc640 1581 else 'bestvideo+bestaudio/best')
0017d9ad 1582
67134eab
JMF
1583 def build_format_selector(self, format_spec):
1584 def syntax_error(note, start):
1585 message = (
1586 'Invalid format specification: '
1587 '{0}\n\t{1}\n\t{2}^'.format(note, format_spec, ' ' * start[1]))
1588 return SyntaxError(message)
1589
1590 PICKFIRST = 'PICKFIRST'
1591 MERGE = 'MERGE'
1592 SINGLE = 'SINGLE'
0130afb7 1593 GROUP = 'GROUP'
67134eab
JMF
1594 FormatSelector = collections.namedtuple('FormatSelector', ['type', 'selector', 'filters'])
1595
91ebc640 1596 allow_multiple_streams = {'audio': self.params.get('allow_multiple_audio_streams', False),
1597 'video': self.params.get('allow_multiple_video_streams', False)}
909d24dd 1598
e8e73840 1599 check_formats = self.params.get('check_formats')
1600
67134eab
JMF
1601 def _parse_filter(tokens):
1602 filter_parts = []
1603 for type, string, start, _, _ in tokens:
1604 if type == tokenize.OP and string == ']':
1605 return ''.join(filter_parts)
1606 else:
1607 filter_parts.append(string)
1608
232541df 1609 def _remove_unused_ops(tokens):
17cc1534 1610 # Remove operators that we don't use and join them with the surrounding strings
232541df
JMF
1611 # for example: 'mp4' '-' 'baseline' '-' '16x9' is converted to 'mp4-baseline-16x9'
1612 ALLOWED_OPS = ('/', '+', ',', '(', ')')
1613 last_string, last_start, last_end, last_line = None, None, None, None
1614 for type, string, start, end, line in tokens:
1615 if type == tokenize.OP and string == '[':
1616 if last_string:
1617 yield tokenize.NAME, last_string, last_start, last_end, last_line
1618 last_string = None
1619 yield type, string, start, end, line
1620 # everything inside brackets will be handled by _parse_filter
1621 for type, string, start, end, line in tokens:
1622 yield type, string, start, end, line
1623 if type == tokenize.OP and string == ']':
1624 break
1625 elif type == tokenize.OP and string in ALLOWED_OPS:
1626 if last_string:
1627 yield tokenize.NAME, last_string, last_start, last_end, last_line
1628 last_string = None
1629 yield type, string, start, end, line
1630 elif type in [tokenize.NAME, tokenize.NUMBER, tokenize.OP]:
1631 if not last_string:
1632 last_string = string
1633 last_start = start
1634 last_end = end
1635 else:
1636 last_string += string
1637 if last_string:
1638 yield tokenize.NAME, last_string, last_start, last_end, last_line
1639
cf2ac6df 1640 def _parse_format_selection(tokens, inside_merge=False, inside_choice=False, inside_group=False):
67134eab
JMF
1641 selectors = []
1642 current_selector = None
1643 for type, string, start, _, _ in tokens:
1644 # ENCODING is only defined in python 3.x
1645 if type == getattr(tokenize, 'ENCODING', None):
1646 continue
1647 elif type in [tokenize.NAME, tokenize.NUMBER]:
1648 current_selector = FormatSelector(SINGLE, string, [])
1649 elif type == tokenize.OP:
cf2ac6df
JMF
1650 if string == ')':
1651 if not inside_group:
1652 # ')' will be handled by the parentheses group
1653 tokens.restore_last_token()
67134eab 1654 break
cf2ac6df 1655 elif inside_merge and string in ['/', ',']:
0130afb7
JMF
1656 tokens.restore_last_token()
1657 break
cf2ac6df
JMF
1658 elif inside_choice and string == ',':
1659 tokens.restore_last_token()
1660 break
1661 elif string == ',':
0a31a350
JMF
1662 if not current_selector:
1663 raise syntax_error('"," must follow a format selector', start)
67134eab
JMF
1664 selectors.append(current_selector)
1665 current_selector = None
1666 elif string == '/':
d96d604e
JMF
1667 if not current_selector:
1668 raise syntax_error('"/" must follow a format selector', start)
67134eab 1669 first_choice = current_selector
cf2ac6df 1670 second_choice = _parse_format_selection(tokens, inside_choice=True)
f5f4a27a 1671 current_selector = FormatSelector(PICKFIRST, (first_choice, second_choice), [])
67134eab
JMF
1672 elif string == '[':
1673 if not current_selector:
1674 current_selector = FormatSelector(SINGLE, 'best', [])
1675 format_filter = _parse_filter(tokens)
1676 current_selector.filters.append(format_filter)
0130afb7
JMF
1677 elif string == '(':
1678 if current_selector:
1679 raise syntax_error('Unexpected "("', start)
cf2ac6df
JMF
1680 group = _parse_format_selection(tokens, inside_group=True)
1681 current_selector = FormatSelector(GROUP, group, [])
67134eab 1682 elif string == '+':
d03cfdce 1683 if not current_selector:
1684 raise syntax_error('Unexpected "+"', start)
1685 selector_1 = current_selector
1686 selector_2 = _parse_format_selection(tokens, inside_merge=True)
1687 if not selector_2:
1688 raise syntax_error('Expected a selector', start)
1689 current_selector = FormatSelector(MERGE, (selector_1, selector_2), [])
67134eab
JMF
1690 else:
1691 raise syntax_error('Operator not recognized: "{0}"'.format(string), start)
1692 elif type == tokenize.ENDMARKER:
1693 break
1694 if current_selector:
1695 selectors.append(current_selector)
1696 return selectors
1697
f8d4ad9a 1698 def _merge(formats_pair):
1699 format_1, format_2 = formats_pair
1700
1701 formats_info = []
1702 formats_info.extend(format_1.get('requested_formats', (format_1,)))
1703 formats_info.extend(format_2.get('requested_formats', (format_2,)))
1704
1705 if not allow_multiple_streams['video'] or not allow_multiple_streams['audio']:
551f9388 1706 get_no_more = {'video': False, 'audio': False}
f8d4ad9a 1707 for (i, fmt_info) in enumerate(formats_info):
551f9388 1708 if fmt_info.get('acodec') == fmt_info.get('vcodec') == 'none':
1709 formats_info.pop(i)
1710 continue
1711 for aud_vid in ['audio', 'video']:
f8d4ad9a 1712 if not allow_multiple_streams[aud_vid] and fmt_info.get(aud_vid[0] + 'codec') != 'none':
1713 if get_no_more[aud_vid]:
1714 formats_info.pop(i)
1715 get_no_more[aud_vid] = True
1716
1717 if len(formats_info) == 1:
1718 return formats_info[0]
1719
1720 video_fmts = [fmt_info for fmt_info in formats_info if fmt_info.get('vcodec') != 'none']
1721 audio_fmts = [fmt_info for fmt_info in formats_info if fmt_info.get('acodec') != 'none']
1722
1723 the_only_video = video_fmts[0] if len(video_fmts) == 1 else None
1724 the_only_audio = audio_fmts[0] if len(audio_fmts) == 1 else None
1725
1726 output_ext = self.params.get('merge_output_format')
1727 if not output_ext:
1728 if the_only_video:
1729 output_ext = the_only_video['ext']
1730 elif the_only_audio and not video_fmts:
1731 output_ext = the_only_audio['ext']
1732 else:
1733 output_ext = 'mkv'
1734
1735 new_dict = {
1736 'requested_formats': formats_info,
1737 'format': '+'.join(fmt_info.get('format') for fmt_info in formats_info),
1738 'format_id': '+'.join(fmt_info.get('format_id') for fmt_info in formats_info),
1739 'ext': output_ext,
1740 }
1741
1742 if the_only_video:
1743 new_dict.update({
1744 'width': the_only_video.get('width'),
1745 'height': the_only_video.get('height'),
1746 'resolution': the_only_video.get('resolution') or self.format_resolution(the_only_video),
1747 'fps': the_only_video.get('fps'),
1748 'vcodec': the_only_video.get('vcodec'),
1749 'vbr': the_only_video.get('vbr'),
1750 'stretched_ratio': the_only_video.get('stretched_ratio'),
1751 })
1752
1753 if the_only_audio:
1754 new_dict.update({
1755 'acodec': the_only_audio.get('acodec'),
1756 'abr': the_only_audio.get('abr'),
1757 })
1758
1759 return new_dict
1760
e8e73840 1761 def _check_formats(formats):
981052c9 1762 if not check_formats:
1763 yield from formats
b5ac45b1 1764 return
e8e73840 1765 for f in formats:
1766 self.to_screen('[info] Testing format %s' % f['format_id'])
21cd8fae 1767 temp_file = tempfile.NamedTemporaryFile(
1768 suffix='.tmp', delete=False,
1769 dir=self.get_output_path('temp') or None)
1770 temp_file.close()
fe346461 1771 try:
981052c9 1772 success, _ = self.dl(temp_file.name, f, test=True)
1773 except (DownloadError, IOError, OSError, ValueError) + network_exceptions:
1774 success = False
fe346461 1775 finally:
21cd8fae 1776 if os.path.exists(temp_file.name):
1777 try:
1778 os.remove(temp_file.name)
1779 except OSError:
1780 self.report_warning('Unable to delete temporary file "%s"' % temp_file.name)
981052c9 1781 if success:
e8e73840 1782 yield f
1783 else:
1784 self.to_screen('[info] Unable to download format %s. Skipping...' % f['format_id'])
1785
67134eab 1786 def _build_selector_function(selector):
909d24dd 1787 if isinstance(selector, list): # ,
67134eab
JMF
1788 fs = [_build_selector_function(s) for s in selector]
1789
317f7ab6 1790 def selector_function(ctx):
67134eab 1791 for f in fs:
981052c9 1792 yield from f(ctx)
67134eab 1793 return selector_function
909d24dd 1794
1795 elif selector.type == GROUP: # ()
0130afb7 1796 selector_function = _build_selector_function(selector.selector)
909d24dd 1797
1798 elif selector.type == PICKFIRST: # /
67134eab
JMF
1799 fs = [_build_selector_function(s) for s in selector.selector]
1800
317f7ab6 1801 def selector_function(ctx):
67134eab 1802 for f in fs:
317f7ab6 1803 picked_formats = list(f(ctx))
67134eab
JMF
1804 if picked_formats:
1805 return picked_formats
1806 return []
67134eab 1807
981052c9 1808 elif selector.type == MERGE: # +
1809 selector_1, selector_2 = map(_build_selector_function, selector.selector)
1810
1811 def selector_function(ctx):
1812 for pair in itertools.product(
1813 selector_1(copy.deepcopy(ctx)), selector_2(copy.deepcopy(ctx))):
1814 yield _merge(pair)
1815
909d24dd 1816 elif selector.type == SINGLE: # atom
598d185d 1817 format_spec = selector.selector or 'best'
909d24dd 1818
f8d4ad9a 1819 # TODO: Add allvideo, allaudio etc by generalizing the code with best/worst selector
909d24dd 1820 if format_spec == 'all':
1821 def selector_function(ctx):
981052c9 1822 yield from _check_formats(ctx['formats'])
f8d4ad9a 1823 elif format_spec == 'mergeall':
1824 def selector_function(ctx):
981052c9 1825 formats = list(_check_formats(ctx['formats']))
e01d6aa4 1826 if not formats:
1827 return
921b76ca 1828 merged_format = formats[-1]
1829 for f in formats[-2::-1]:
f8d4ad9a 1830 merged_format = _merge((merged_format, f))
1831 yield merged_format
909d24dd 1832
1833 else:
e8e73840 1834 format_fallback, format_reverse, format_idx = False, True, 1
eff63539 1835 mobj = re.match(
1836 r'(?P<bw>best|worst|b|w)(?P<type>video|audio|v|a)?(?P<mod>\*)?(?:\.(?P<n>[1-9]\d*))?$',
1837 format_spec)
1838 if mobj is not None:
1839 format_idx = int_or_none(mobj.group('n'), default=1)
e8e73840 1840 format_reverse = mobj.group('bw')[0] == 'b'
eff63539 1841 format_type = (mobj.group('type') or [None])[0]
1842 not_format_type = {'v': 'a', 'a': 'v'}.get(format_type)
1843 format_modified = mobj.group('mod') is not None
909d24dd 1844
1845 format_fallback = not format_type and not format_modified # for b, w
8326b00a 1846 _filter_f = (
eff63539 1847 (lambda f: f.get('%scodec' % format_type) != 'none')
1848 if format_type and format_modified # bv*, ba*, wv*, wa*
1849 else (lambda f: f.get('%scodec' % not_format_type) == 'none')
1850 if format_type # bv, ba, wv, wa
1851 else (lambda f: f.get('vcodec') != 'none' and f.get('acodec') != 'none')
1852 if not format_modified # b, w
8326b00a 1853 else lambda f: True) # b*, w*
1854 filter_f = lambda f: _filter_f(f) and (
1855 f.get('vcodec') != 'none' or f.get('acodec') != 'none')
67134eab 1856 else:
909d24dd 1857 filter_f = ((lambda f: f.get('ext') == format_spec)
1858 if format_spec in ['mp4', 'flv', 'webm', '3gp', 'm4a', 'mp3', 'ogg', 'aac', 'wav'] # extension
1859 else (lambda f: f.get('format_id') == format_spec)) # id
1860
1861 def selector_function(ctx):
1862 formats = list(ctx['formats'])
909d24dd 1863 matches = list(filter(filter_f, formats)) if filter_f is not None else formats
e8e73840 1864 if format_fallback and ctx['incomplete_formats'] and not matches:
909d24dd 1865 # for extractors with incomplete formats (audio only (soundcloud)
1866 # or video only (imgur)) best/worst will fallback to
1867 # best/worst {video,audio}-only format
e8e73840 1868 matches = formats
981052c9 1869 matches = LazyList(_check_formats(matches[::-1 if format_reverse else 1]))
1870 try:
e8e73840 1871 yield matches[format_idx - 1]
981052c9 1872 except IndexError:
1873 return
083c9df9 1874
67134eab 1875 filters = [self._build_format_filter(f) for f in selector.filters]
083c9df9 1876
317f7ab6
S
1877 def final_selector(ctx):
1878 ctx_copy = copy.deepcopy(ctx)
67134eab 1879 for _filter in filters:
317f7ab6
S
1880 ctx_copy['formats'] = list(filter(_filter, ctx_copy['formats']))
1881 return selector_function(ctx_copy)
67134eab 1882 return final_selector
083c9df9 1883
67134eab 1884 stream = io.BytesIO(format_spec.encode('utf-8'))
0130afb7 1885 try:
232541df 1886 tokens = list(_remove_unused_ops(compat_tokenize_tokenize(stream.readline)))
0130afb7
JMF
1887 except tokenize.TokenError:
1888 raise syntax_error('Missing closing/opening brackets or parenthesis', (0, len(format_spec)))
1889
1890 class TokenIterator(object):
1891 def __init__(self, tokens):
1892 self.tokens = tokens
1893 self.counter = 0
1894
1895 def __iter__(self):
1896 return self
1897
1898 def __next__(self):
1899 if self.counter >= len(self.tokens):
1900 raise StopIteration()
1901 value = self.tokens[self.counter]
1902 self.counter += 1
1903 return value
1904
1905 next = __next__
1906
1907 def restore_last_token(self):
1908 self.counter -= 1
1909
1910 parsed_selector = _parse_format_selection(iter(TokenIterator(tokens)))
67134eab 1911 return _build_selector_function(parsed_selector)
a9c58ad9 1912
e5660ee6
JMF
1913 def _calc_headers(self, info_dict):
1914 res = std_headers.copy()
1915
1916 add_headers = info_dict.get('http_headers')
1917 if add_headers:
1918 res.update(add_headers)
1919
1920 cookies = self._calc_cookies(info_dict)
1921 if cookies:
1922 res['Cookie'] = cookies
1923
0016b84e
S
1924 if 'X-Forwarded-For' not in res:
1925 x_forwarded_for_ip = info_dict.get('__x_forwarded_for_ip')
1926 if x_forwarded_for_ip:
1927 res['X-Forwarded-For'] = x_forwarded_for_ip
1928
e5660ee6
JMF
1929 return res
1930
1931 def _calc_cookies(self, info_dict):
5c2266df 1932 pr = sanitized_Request(info_dict['url'])
e5660ee6 1933 self.cookiejar.add_cookie_header(pr)
662435f7 1934 return pr.get_header('Cookie')
e5660ee6 1935
b0249bca 1936 def _sanitize_thumbnails(self, info_dict):
bc516a3f 1937 thumbnails = info_dict.get('thumbnails')
1938 if thumbnails is None:
1939 thumbnail = info_dict.get('thumbnail')
1940 if thumbnail:
1941 info_dict['thumbnails'] = thumbnails = [{'url': thumbnail}]
1942 if thumbnails:
1943 thumbnails.sort(key=lambda t: (
1944 t.get('preference') if t.get('preference') is not None else -1,
1945 t.get('width') if t.get('width') is not None else -1,
1946 t.get('height') if t.get('height') is not None else -1,
1947 t.get('id') if t.get('id') is not None else '',
1948 t.get('url')))
b0249bca 1949
0ba692ac 1950 def thumbnail_tester():
1951 if self.params.get('check_formats'):
1952 def to_screen(msg):
1953 return self.to_screen(f'[info] {msg}')
1954 else:
1955 to_screen = self.write_debug
1956
1957 def test_thumbnail(t):
1958 to_screen('Testing thumbnail %s' % t['id'])
1959 try:
1960 self.urlopen(HEADRequest(t['url']))
1961 except network_exceptions as err:
1962 to_screen('Unable to connect to thumbnail %s URL "%s" - %s. Skipping...' % (
1963 t['id'], t['url'], error_to_compat_str(err)))
1964 return False
1965 return True
1966
1967 return test_thumbnail
b0249bca 1968
bc516a3f 1969 for i, t in enumerate(thumbnails):
bc516a3f 1970 if t.get('id') is None:
1971 t['id'] = '%d' % i
b0249bca 1972 if t.get('width') and t.get('height'):
1973 t['resolution'] = '%dx%d' % (t['width'], t['height'])
1974 t['url'] = sanitize_url(t['url'])
0ba692ac 1975
1976 if self.params.get('check_formats') is not False:
1977 info_dict['thumbnails'] = LazyList(filter(thumbnail_tester(), thumbnails[::-1])).reverse()
1978 else:
1979 info_dict['thumbnails'] = thumbnails
bc516a3f 1980
dd82ffea
JMF
1981 def process_video_result(self, info_dict, download=True):
1982 assert info_dict.get('_type', 'video') == 'video'
1983
bec1fad2
PH
1984 if 'id' not in info_dict:
1985 raise ExtractorError('Missing "id" field in extractor result')
1986 if 'title' not in info_dict:
1987 raise ExtractorError('Missing "title" field in extractor result')
1988
c9969434
S
1989 def report_force_conversion(field, field_not, conversion):
1990 self.report_warning(
1991 '"%s" field is not %s - forcing %s conversion, there is an error in extractor'
1992 % (field, field_not, conversion))
1993
1994 def sanitize_string_field(info, string_field):
1995 field = info.get(string_field)
1996 if field is None or isinstance(field, compat_str):
1997 return
1998 report_force_conversion(string_field, 'a string', 'string')
1999 info[string_field] = compat_str(field)
2000
2001 def sanitize_numeric_fields(info):
2002 for numeric_field in self._NUMERIC_FIELDS:
2003 field = info.get(numeric_field)
2004 if field is None or isinstance(field, compat_numeric_types):
2005 continue
2006 report_force_conversion(numeric_field, 'numeric', 'int')
2007 info[numeric_field] = int_or_none(field)
2008
2009 sanitize_string_field(info_dict, 'id')
2010 sanitize_numeric_fields(info_dict)
be6217b2 2011
dd82ffea
JMF
2012 if 'playlist' not in info_dict:
2013 # It isn't part of a playlist
2014 info_dict['playlist'] = None
2015 info_dict['playlist_index'] = None
2016
bc516a3f 2017 self._sanitize_thumbnails(info_dict)
d5519808 2018
536a55da 2019 thumbnail = info_dict.get('thumbnail')
bc516a3f 2020 thumbnails = info_dict.get('thumbnails')
536a55da
S
2021 if thumbnail:
2022 info_dict['thumbnail'] = sanitize_url(thumbnail)
2023 elif thumbnails:
d5519808
PH
2024 info_dict['thumbnail'] = thumbnails[-1]['url']
2025
c9ae7b95 2026 if 'display_id' not in info_dict and 'id' in info_dict:
0afef30b
PH
2027 info_dict['display_id'] = info_dict['id']
2028
10db0d2f 2029 for ts_key, date_key in (
2030 ('timestamp', 'upload_date'),
2031 ('release_timestamp', 'release_date'),
2032 ):
2033 if info_dict.get(date_key) is None and info_dict.get(ts_key) is not None:
2034 # Working around out-of-range timestamp values (e.g. negative ones on Windows,
2035 # see http://bugs.python.org/issue1646728)
2036 try:
2037 upload_date = datetime.datetime.utcfromtimestamp(info_dict[ts_key])
2038 info_dict[date_key] = upload_date.strftime('%Y%m%d')
2039 except (ValueError, OverflowError, OSError):
2040 pass
9d2ecdbc 2041
33d2fc2f
S
2042 # Auto generate title fields corresponding to the *_number fields when missing
2043 # in order to always have clean titles. This is very common for TV series.
2044 for field in ('chapter', 'season', 'episode'):
2045 if info_dict.get('%s_number' % field) is not None and not info_dict.get(field):
2046 info_dict[field] = '%s %d' % (field.capitalize(), info_dict['%s_number' % field])
2047
05108a49
S
2048 for cc_kind in ('subtitles', 'automatic_captions'):
2049 cc = info_dict.get(cc_kind)
2050 if cc:
2051 for _, subtitle in cc.items():
2052 for subtitle_format in subtitle:
2053 if subtitle_format.get('url'):
2054 subtitle_format['url'] = sanitize_url(subtitle_format['url'])
2055 if subtitle_format.get('ext') is None:
2056 subtitle_format['ext'] = determine_ext(subtitle_format['url']).lower()
2057
2058 automatic_captions = info_dict.get('automatic_captions')
4bba3716 2059 subtitles = info_dict.get('subtitles')
4bba3716 2060
360e1ca5 2061 info_dict['requested_subtitles'] = self.process_subtitles(
05108a49 2062 info_dict['id'], subtitles, automatic_captions)
a504ced0 2063
dd82ffea
JMF
2064 # We now pick which formats have to be downloaded
2065 if info_dict.get('formats') is None:
2066 # There's only one format available
2067 formats = [info_dict]
2068 else:
2069 formats = info_dict['formats']
2070
db95dc13 2071 if not formats:
b7da73eb 2072 if not self.params.get('ignore_no_formats_error'):
2073 raise ExtractorError('No video formats found!')
2074 else:
2075 self.report_warning('No video formats found!')
db95dc13 2076
73af5cc8
S
2077 def is_wellformed(f):
2078 url = f.get('url')
a5ac0c47 2079 if not url:
73af5cc8
S
2080 self.report_warning(
2081 '"url" field is missing or empty - skipping format, '
2082 'there is an error in extractor')
a5ac0c47
S
2083 return False
2084 if isinstance(url, bytes):
2085 sanitize_string_field(f, 'url')
2086 return True
73af5cc8
S
2087
2088 # Filter out malformed formats for better extraction robustness
2089 formats = list(filter(is_wellformed, formats))
2090
181c7053
S
2091 formats_dict = {}
2092
dd82ffea 2093 # We check that all the formats have the format and format_id fields
db95dc13 2094 for i, format in enumerate(formats):
c9969434
S
2095 sanitize_string_field(format, 'format_id')
2096 sanitize_numeric_fields(format)
dcf77cf1 2097 format['url'] = sanitize_url(format['url'])
e74e3b63 2098 if not format.get('format_id'):
8016c922 2099 format['format_id'] = compat_str(i)
e2effb08
S
2100 else:
2101 # Sanitize format_id from characters used in format selector expression
ec85ded8 2102 format['format_id'] = re.sub(r'[\s,/+\[\]()]', '_', format['format_id'])
181c7053
S
2103 format_id = format['format_id']
2104 if format_id not in formats_dict:
2105 formats_dict[format_id] = []
2106 formats_dict[format_id].append(format)
2107
2108 # Make sure all formats have unique format_id
2109 for format_id, ambiguous_formats in formats_dict.items():
2110 if len(ambiguous_formats) > 1:
2111 for i, format in enumerate(ambiguous_formats):
2112 format['format_id'] = '%s-%d' % (format_id, i)
2113
2114 for i, format in enumerate(formats):
8c51aa65 2115 if format.get('format') is None:
6febd1c1 2116 format['format'] = '{id} - {res}{note}'.format(
8c51aa65
JMF
2117 id=format['format_id'],
2118 res=self.format_resolution(format),
6febd1c1 2119 note=' ({0})'.format(format['format_note']) if format.get('format_note') is not None else '',
8c51aa65 2120 )
c1002e96 2121 # Automatically determine file extension if missing
5b1d8575 2122 if format.get('ext') is None:
cce929ea 2123 format['ext'] = determine_ext(format['url']).lower()
b5559424
S
2124 # Automatically determine protocol if missing (useful for format
2125 # selection purposes)
6f0be937 2126 if format.get('protocol') is None:
b5559424 2127 format['protocol'] = determine_protocol(format)
e5660ee6
JMF
2128 # Add HTTP headers, so that external programs can use them from the
2129 # json output
2130 full_format_info = info_dict.copy()
2131 full_format_info.update(format)
2132 format['http_headers'] = self._calc_headers(full_format_info)
0016b84e
S
2133 # Remove private housekeeping stuff
2134 if '__x_forwarded_for_ip' in info_dict:
2135 del info_dict['__x_forwarded_for_ip']
dd82ffea 2136
4bcc7bd1 2137 # TODO Central sorting goes here
99e206d5 2138
b7da73eb 2139 if formats and formats[0] is not info_dict:
b3d9ef88
JMF
2140 # only set the 'formats' fields if the original info_dict list them
2141 # otherwise we end up with a circular reference, the first (and unique)
f89197d7 2142 # element in the 'formats' field in info_dict is info_dict itself,
dfb1b146 2143 # which can't be exported to json
b3d9ef88 2144 info_dict['formats'] = formats
4ec82a72 2145
2146 info_dict, _ = self.pre_process(info_dict)
2147
169dbde9 2148 list_only = self.params.get('list_thumbnails') or self.params.get('listformats') or self.params.get('listsubtitles')
2149 if list_only:
2150 self.__forced_printings(info_dict, self.prepare_filename(info_dict), incomplete=True)
2151 if self.params.get('list_thumbnails'):
2152 self.list_thumbnails(info_dict)
2153 if self.params.get('listformats'):
2154 if not info_dict.get('formats'):
2155 raise ExtractorError('No video formats found', expected=True)
2156 self.list_formats(info_dict)
2157 if self.params.get('listsubtitles'):
2158 if 'automatic_captions' in info_dict:
2159 self.list_subtitles(
2160 info_dict['id'], automatic_captions, 'automatic captions')
2161 self.list_subtitles(info_dict['id'], subtitles, 'subtitles')
bfaae0a7 2162 return
2163
187986a8 2164 format_selector = self.format_selector
2165 if format_selector is None:
0017d9ad 2166 req_format = self._default_format_spec(info_dict, download=download)
0760b0a7 2167 self.write_debug('Default format spec: %s' % req_format)
187986a8 2168 format_selector = self.build_format_selector(req_format)
317f7ab6
S
2169
2170 # While in format selection we may need to have an access to the original
2171 # format set in order to calculate some metrics or do some processing.
2172 # For now we need to be able to guess whether original formats provided
2173 # by extractor are incomplete or not (i.e. whether extractor provides only
2174 # video-only or audio-only formats) for proper formats selection for
2175 # extractors with such incomplete formats (see
067aa17e 2176 # https://github.com/ytdl-org/youtube-dl/pull/5556).
317f7ab6
S
2177 # Since formats may be filtered during format selection and may not match
2178 # the original formats the results may be incorrect. Thus original formats
2179 # or pre-calculated metrics should be passed to format selection routines
2180 # as well.
2181 # We will pass a context object containing all necessary additional data
2182 # instead of just formats.
2183 # This fixes incorrect format selection issue (see
067aa17e 2184 # https://github.com/ytdl-org/youtube-dl/issues/10083).
2e221ca3 2185 incomplete_formats = (
317f7ab6 2186 # All formats are video-only or
3089bc74 2187 all(f.get('vcodec') != 'none' and f.get('acodec') == 'none' for f in formats)
317f7ab6 2188 # all formats are audio-only
3089bc74 2189 or all(f.get('vcodec') == 'none' and f.get('acodec') != 'none' for f in formats))
317f7ab6
S
2190
2191 ctx = {
2192 'formats': formats,
2193 'incomplete_formats': incomplete_formats,
2194 }
2195
2196 formats_to_download = list(format_selector(ctx))
dd82ffea 2197 if not formats_to_download:
b7da73eb 2198 if not self.params.get('ignore_no_formats_error'):
2199 raise ExtractorError('Requested format is not available', expected=True)
2200 else:
2201 self.report_warning('Requested format is not available')
4513a41a
A
2202 # Process what we can, even without any available formats.
2203 self.process_info(dict(info_dict))
b7da73eb 2204 elif download:
2205 self.to_screen(
07cce701 2206 '[info] %s: Downloading %d format(s): %s' % (
2207 info_dict['id'], len(formats_to_download),
2208 ", ".join([f['format_id'] for f in formats_to_download])))
b7da73eb 2209 for fmt in formats_to_download:
dd82ffea 2210 new_info = dict(info_dict)
4ec82a72 2211 # Save a reference to the original info_dict so that it can be modified in process_info if needed
2212 new_info['__original_infodict'] = info_dict
b7da73eb 2213 new_info.update(fmt)
dd82ffea
JMF
2214 self.process_info(new_info)
2215 # We update the info dict with the best quality format (backwards compatibility)
b7da73eb 2216 if formats_to_download:
2217 info_dict.update(formats_to_download[-1])
dd82ffea
JMF
2218 return info_dict
2219
98c70d6f 2220 def process_subtitles(self, video_id, normal_subtitles, automatic_captions):
a504ced0 2221 """Select the requested subtitles and their format"""
98c70d6f
JMF
2222 available_subs = {}
2223 if normal_subtitles and self.params.get('writesubtitles'):
2224 available_subs.update(normal_subtitles)
2225 if automatic_captions and self.params.get('writeautomaticsub'):
2226 for lang, cap_info in automatic_captions.items():
360e1ca5
JMF
2227 if lang not in available_subs:
2228 available_subs[lang] = cap_info
2229
4d171848
JMF
2230 if (not self.params.get('writesubtitles') and not
2231 self.params.get('writeautomaticsub') or not
2232 available_subs):
2233 return None
a504ced0 2234
c32b0aab 2235 all_sub_langs = available_subs.keys()
a504ced0 2236 if self.params.get('allsubtitles', False):
c32b0aab 2237 requested_langs = all_sub_langs
2238 elif self.params.get('subtitleslangs', False):
2239 requested_langs = set()
2240 for lang in self.params.get('subtitleslangs'):
2241 if lang == 'all':
2242 requested_langs.update(all_sub_langs)
2243 continue
2244 discard = lang[0] == '-'
2245 if discard:
2246 lang = lang[1:]
2247 current_langs = filter(re.compile(lang + '$').match, all_sub_langs)
2248 if discard:
2249 for lang in current_langs:
2250 requested_langs.discard(lang)
2251 else:
2252 requested_langs.update(current_langs)
2253 elif 'en' in available_subs:
2254 requested_langs = ['en']
a504ced0 2255 else:
c32b0aab 2256 requested_langs = [list(all_sub_langs)[0]]
e632bce2 2257 self.write_debug('Downloading subtitles: %s' % ', '.join(requested_langs))
a504ced0
JMF
2258
2259 formats_query = self.params.get('subtitlesformat', 'best')
2260 formats_preference = formats_query.split('/') if formats_query else []
2261 subs = {}
2262 for lang in requested_langs:
2263 formats = available_subs.get(lang)
2264 if formats is None:
2265 self.report_warning('%s subtitles not available for %s' % (lang, video_id))
2266 continue
a504ced0
JMF
2267 for ext in formats_preference:
2268 if ext == 'best':
2269 f = formats[-1]
2270 break
2271 matches = list(filter(lambda f: f['ext'] == ext, formats))
2272 if matches:
2273 f = matches[-1]
2274 break
2275 else:
2276 f = formats[-1]
2277 self.report_warning(
2278 'No subtitle format found matching "%s" for language %s, '
2279 'using %s' % (formats_query, lang, f['ext']))
2280 subs[lang] = f
2281 return subs
2282
d06daf23 2283 def __forced_printings(self, info_dict, filename, incomplete):
53c18592 2284 def print_mandatory(field, actual_field=None):
2285 if actual_field is None:
2286 actual_field = field
d06daf23 2287 if (self.params.get('force%s' % field, False)
53c18592 2288 and (not incomplete or info_dict.get(actual_field) is not None)):
2289 self.to_stdout(info_dict[actual_field])
d06daf23
S
2290
2291 def print_optional(field):
2292 if (self.params.get('force%s' % field, False)
2293 and info_dict.get(field) is not None):
2294 self.to_stdout(info_dict[field])
2295
53c18592 2296 info_dict = info_dict.copy()
2297 if filename is not None:
2298 info_dict['filename'] = filename
2299 if info_dict.get('requested_formats') is not None:
2300 # For RTMP URLs, also include the playpath
2301 info_dict['urls'] = '\n'.join(f['url'] + f.get('play_path', '') for f in info_dict['requested_formats'])
2302 elif 'url' in info_dict:
2303 info_dict['urls'] = info_dict['url'] + info_dict.get('play_path', '')
2304
2305 for tmpl in self.params.get('forceprint', []):
2306 if re.match(r'\w+$', tmpl):
2307 tmpl = '%({})s'.format(tmpl)
2308 tmpl, info_copy = self.prepare_outtmpl(tmpl, info_dict)
2309 self.to_stdout(tmpl % info_copy)
2310
d06daf23
S
2311 print_mandatory('title')
2312 print_mandatory('id')
53c18592 2313 print_mandatory('url', 'urls')
d06daf23
S
2314 print_optional('thumbnail')
2315 print_optional('description')
53c18592 2316 print_optional('filename')
d06daf23
S
2317 if self.params.get('forceduration', False) and info_dict.get('duration') is not None:
2318 self.to_stdout(formatSeconds(info_dict['duration']))
2319 print_mandatory('format')
53c18592 2320
d06daf23 2321 if self.params.get('forcejson', False):
277d6ff5 2322 self.post_extract(info_dict)
75d43ca0 2323 self.to_stdout(json.dumps(info_dict, default=repr))
d06daf23 2324
e8e73840 2325 def dl(self, name, info, subtitle=False, test=False):
2326
2327 if test:
2328 verbose = self.params.get('verbose')
2329 params = {
2330 'test': True,
2331 'quiet': not verbose,
2332 'verbose': verbose,
2333 'noprogress': not verbose,
2334 'nopart': True,
2335 'skip_unavailable_fragments': False,
2336 'keep_fragments': False,
2337 'overwrites': True,
2338 '_no_ytdl_file': True,
2339 }
2340 else:
2341 params = self.params
2342 fd = get_suitable_downloader(info, params)(self, params)
2343 if not test:
2344 for ph in self._progress_hooks:
2345 fd.add_progress_hook(ph)
18e674b4 2346 urls = '", "'.join([f['url'] for f in info.get('requested_formats', [])] or [info['url']])
2347 self.write_debug('Invoking downloader on "%s"' % urls)
e8e73840 2348 new_info = dict(info)
2349 if new_info.get('http_headers') is None:
2350 new_info['http_headers'] = self._calc_headers(new_info)
2351 return fd.download(name, new_info, subtitle)
2352
8222d8de
JMF
2353 def process_info(self, info_dict):
2354 """Process a single resolved IE result."""
2355
2356 assert info_dict.get('_type', 'video') == 'video'
fd288278 2357
0202b52a 2358 info_dict.setdefault('__postprocessors', [])
2359
fd288278
PH
2360 max_downloads = self.params.get('max_downloads')
2361 if max_downloads is not None:
2362 if self._num_downloads >= int(max_downloads):
2363 raise MaxDownloadsReached()
8222d8de 2364
d06daf23 2365 # TODO: backward compatibility, to be removed
8222d8de 2366 info_dict['fulltitle'] = info_dict['title']
8222d8de 2367
4513a41a 2368 if 'format' not in info_dict and 'ext' in info_dict:
8222d8de
JMF
2369 info_dict['format'] = info_dict['ext']
2370
c77495e3 2371 if self._match_entry(info_dict) is not None:
8222d8de
JMF
2372 return
2373
277d6ff5 2374 self.post_extract(info_dict)
fd288278 2375 self._num_downloads += 1
8222d8de 2376
dcf64d43 2377 # info_dict['_filename'] needs to be set for backward compatibility
de6000d9 2378 info_dict['_filename'] = full_filename = self.prepare_filename(info_dict, warn=True)
2379 temp_filename = self.prepare_filename(info_dict, 'temp')
0202b52a 2380 files_to_move = {}
8222d8de
JMF
2381
2382 # Forced printings
4513a41a 2383 self.__forced_printings(info_dict, full_filename, incomplete=('format' not in info_dict))
8222d8de 2384
8222d8de 2385 if self.params.get('simulate', False):
2d30509f 2386 if self.params.get('force_write_download_archive', False):
2387 self.record_download_archive(info_dict)
2388
2389 # Do nothing else if in simulate mode
8222d8de
JMF
2390 return
2391
de6000d9 2392 if full_filename is None:
8222d8de
JMF
2393 return
2394
e92caff5 2395 if not self._ensure_dir_exists(encodeFilename(full_filename)):
0202b52a 2396 return
e92caff5 2397 if not self._ensure_dir_exists(encodeFilename(temp_filename)):
8222d8de
JMF
2398 return
2399
2400 if self.params.get('writedescription', False):
de6000d9 2401 descfn = self.prepare_filename(info_dict, 'description')
e92caff5 2402 if not self._ensure_dir_exists(encodeFilename(descfn)):
0202b52a 2403 return
0c3d0f51 2404 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(descfn)):
6febd1c1 2405 self.to_screen('[info] Video description is already present')
f00fd51d
JMF
2406 elif info_dict.get('description') is None:
2407 self.report_warning('There\'s no description to write.')
7b6fefc9
PH
2408 else:
2409 try:
6febd1c1 2410 self.to_screen('[info] Writing video description to: ' + descfn)
7b6fefc9
PH
2411 with io.open(encodeFilename(descfn), 'w', encoding='utf-8') as descfile:
2412 descfile.write(info_dict['description'])
7b6fefc9 2413 except (OSError, IOError):
6febd1c1 2414 self.report_error('Cannot write description file ' + descfn)
7b6fefc9 2415 return
8222d8de 2416
1fb07d10 2417 if self.params.get('writeannotations', False):
de6000d9 2418 annofn = self.prepare_filename(info_dict, 'annotation')
e92caff5 2419 if not self._ensure_dir_exists(encodeFilename(annofn)):
0202b52a 2420 return
0c3d0f51 2421 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(annofn)):
6febd1c1 2422 self.to_screen('[info] Video annotations are already present')
ffddb112
RA
2423 elif not info_dict.get('annotations'):
2424 self.report_warning('There are no annotations to write.')
7b6fefc9
PH
2425 else:
2426 try:
6febd1c1 2427 self.to_screen('[info] Writing video annotations to: ' + annofn)
7b6fefc9
PH
2428 with io.open(encodeFilename(annofn), 'w', encoding='utf-8') as annofile:
2429 annofile.write(info_dict['annotations'])
2430 except (KeyError, TypeError):
6febd1c1 2431 self.report_warning('There are no annotations to write.')
7b6fefc9 2432 except (OSError, IOError):
6febd1c1 2433 self.report_error('Cannot write annotations file: ' + annofn)
7b6fefc9 2434 return
1fb07d10 2435
c4a91be7 2436 subtitles_are_requested = any([self.params.get('writesubtitles', False),
0b7f3118 2437 self.params.get('writeautomaticsub')])
c4a91be7 2438
c84dd8a9 2439 if subtitles_are_requested and info_dict.get('requested_subtitles'):
8222d8de
JMF
2440 # subtitles download errors are already managed as troubles in relevant IE
2441 # that way it will silently go on when used with unsupporting IE
c84dd8a9 2442 subtitles = info_dict['requested_subtitles']
fa57af1e 2443 # ie = self.get_info_extractor(info_dict['extractor_key'])
a504ced0
JMF
2444 for sub_lang, sub_info in subtitles.items():
2445 sub_format = sub_info['ext']
56d868db 2446 sub_filename = subtitles_filename(temp_filename, sub_lang, sub_format, info_dict.get('ext'))
2447 sub_filename_final = subtitles_filename(
2448 self.prepare_filename(info_dict, 'subtitle'), sub_lang, sub_format, info_dict.get('ext'))
0c3d0f51 2449 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(sub_filename)):
5ff1bc0c 2450 self.to_screen('[info] Video subtitle %s.%s is already present' % (sub_lang, sub_format))
dcf64d43 2451 sub_info['filepath'] = sub_filename
0202b52a 2452 files_to_move[sub_filename] = sub_filename_final
a504ced0 2453 else:
0c9df79e 2454 self.to_screen('[info] Writing video subtitles to: ' + sub_filename)
5ff1bc0c
RA
2455 if sub_info.get('data') is not None:
2456 try:
2457 # Use newline='' to prevent conversion of newline characters
067aa17e 2458 # See https://github.com/ytdl-org/youtube-dl/issues/10268
5ff1bc0c
RA
2459 with io.open(encodeFilename(sub_filename), 'w', encoding='utf-8', newline='') as subfile:
2460 subfile.write(sub_info['data'])
dcf64d43 2461 sub_info['filepath'] = sub_filename
0202b52a 2462 files_to_move[sub_filename] = sub_filename_final
5ff1bc0c
RA
2463 except (OSError, IOError):
2464 self.report_error('Cannot write subtitles file ' + sub_filename)
2465 return
7b6fefc9 2466 else:
5ff1bc0c 2467 try:
e8e73840 2468 self.dl(sub_filename, sub_info.copy(), subtitle=True)
dcf64d43 2469 sub_info['filepath'] = sub_filename
0202b52a 2470 files_to_move[sub_filename] = sub_filename_final
fe346461 2471 except (ExtractorError, IOError, OSError, ValueError) + network_exceptions as err:
5ff1bc0c
RA
2472 self.report_warning('Unable to download subtitle for "%s": %s' %
2473 (sub_lang, error_to_compat_str(err)))
2474 continue
8222d8de 2475
8222d8de 2476 if self.params.get('writeinfojson', False):
de6000d9 2477 infofn = self.prepare_filename(info_dict, 'infojson')
e92caff5 2478 if not self._ensure_dir_exists(encodeFilename(infofn)):
0202b52a 2479 return
0c3d0f51 2480 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(infofn)):
66c935fb 2481 self.to_screen('[info] Video metadata is already present')
7b6fefc9 2482 else:
66c935fb 2483 self.to_screen('[info] Writing video metadata as JSON to: ' + infofn)
7b6fefc9 2484 try:
75d43ca0 2485 write_json_file(self.filter_requested_info(info_dict, self.params.get('clean_infojson', True)), infofn)
7b6fefc9 2486 except (OSError, IOError):
66c935fb 2487 self.report_error('Cannot write video metadata to JSON file ' + infofn)
7b6fefc9 2488 return
de6000d9 2489 info_dict['__infojson_filename'] = infofn
8222d8de 2490
56d868db 2491 for thumb_ext in self._write_thumbnails(info_dict, temp_filename):
2492 thumb_filename_temp = replace_extension(temp_filename, thumb_ext, info_dict.get('ext'))
2493 thumb_filename = replace_extension(
2494 self.prepare_filename(info_dict, 'thumbnail'), thumb_ext, info_dict.get('ext'))
dcf64d43 2495 files_to_move[thumb_filename_temp] = thumb_filename
8222d8de 2496
732044af 2497 # Write internet shortcut files
2498 url_link = webloc_link = desktop_link = False
2499 if self.params.get('writelink', False):
2500 if sys.platform == "darwin": # macOS.
2501 webloc_link = True
2502 elif sys.platform.startswith("linux"):
2503 desktop_link = True
2504 else: # if sys.platform in ['win32', 'cygwin']:
2505 url_link = True
2506 if self.params.get('writeurllink', False):
2507 url_link = True
2508 if self.params.get('writewebloclink', False):
2509 webloc_link = True
2510 if self.params.get('writedesktoplink', False):
2511 desktop_link = True
2512
2513 if url_link or webloc_link or desktop_link:
2514 if 'webpage_url' not in info_dict:
2515 self.report_error('Cannot write internet shortcut file because the "webpage_url" field is missing in the media information')
2516 return
2517 ascii_url = iri_to_uri(info_dict['webpage_url'])
2518
2519 def _write_link_file(extension, template, newline, embed_filename):
0202b52a 2520 linkfn = replace_extension(full_filename, extension, info_dict.get('ext'))
10e3742e 2521 if self.params.get('overwrites', True) and os.path.exists(encodeFilename(linkfn)):
732044af 2522 self.to_screen('[info] Internet shortcut is already present')
2523 else:
2524 try:
2525 self.to_screen('[info] Writing internet shortcut to: ' + linkfn)
2526 with io.open(encodeFilename(to_high_limit_path(linkfn)), 'w', encoding='utf-8', newline=newline) as linkfile:
2527 template_vars = {'url': ascii_url}
2528 if embed_filename:
2529 template_vars['filename'] = linkfn[:-(len(extension) + 1)]
2530 linkfile.write(template % template_vars)
2531 except (OSError, IOError):
2532 self.report_error('Cannot write internet shortcut ' + linkfn)
2533 return False
2534 return True
2535
2536 if url_link:
2537 if not _write_link_file('url', DOT_URL_LINK_TEMPLATE, '\r\n', embed_filename=False):
2538 return
2539 if webloc_link:
2540 if not _write_link_file('webloc', DOT_WEBLOC_LINK_TEMPLATE, '\n', embed_filename=False):
2541 return
2542 if desktop_link:
2543 if not _write_link_file('desktop', DOT_DESKTOP_LINK_TEMPLATE, '\n', embed_filename=True):
2544 return
2545
56d868db 2546 try:
2547 info_dict, files_to_move = self.pre_process(info_dict, 'before_dl', files_to_move)
2548 except PostProcessingError as err:
2549 self.report_error('Preprocessing: %s' % str(err))
2550 return
2551
732044af 2552 must_record_download_archive = False
56d868db 2553 if self.params.get('skip_download', False):
2554 info_dict['filepath'] = temp_filename
2555 info_dict['__finaldir'] = os.path.dirname(os.path.abspath(encodeFilename(full_filename)))
2556 info_dict['__files_to_move'] = files_to_move
2557 info_dict = self.run_pp(MoveFilesAfterDownloadPP(self, False), info_dict)
2558 else:
2559 # Download
4340deca 2560 try:
0202b52a 2561
6b591b29 2562 def existing_file(*filepaths):
2563 ext = info_dict.get('ext')
2564 final_ext = self.params.get('final_ext', ext)
2565 existing_files = []
2566 for file in orderedSet(filepaths):
2567 if final_ext != ext:
2568 converted = replace_extension(file, final_ext, ext)
2569 if os.path.exists(encodeFilename(converted)):
2570 existing_files.append(converted)
2571 if os.path.exists(encodeFilename(file)):
2572 existing_files.append(file)
2573
2574 if not existing_files or self.params.get('overwrites', False):
2575 for file in orderedSet(existing_files):
2576 self.report_file_delete(file)
2577 os.remove(encodeFilename(file))
2578 return None
2579
2580 self.report_file_already_downloaded(existing_files[0])
2581 info_dict['ext'] = os.path.splitext(existing_files[0])[1][1:]
2582 return existing_files[0]
0202b52a 2583
2584 success = True
4340deca 2585 if info_dict.get('requested_formats') is not None:
81cd954a
S
2586
2587 def compatible_formats(formats):
d03cfdce 2588 # TODO: some formats actually allow this (mkv, webm, ogg, mp4), but not all of them.
2589 video_formats = [format for format in formats if format.get('vcodec') != 'none']
2590 audio_formats = [format for format in formats if format.get('acodec') != 'none']
2591 if len(video_formats) > 2 or len(audio_formats) > 2:
2592 return False
2593
81cd954a 2594 # Check extension
d03cfdce 2595 exts = set(format.get('ext') for format in formats)
2596 COMPATIBLE_EXTS = (
2597 set(('mp3', 'mp4', 'm4a', 'm4p', 'm4b', 'm4r', 'm4v', 'ismv', 'isma')),
2598 set(('webm',)),
2599 )
2600 for ext_sets in COMPATIBLE_EXTS:
2601 if ext_sets.issuperset(exts):
2602 return True
81cd954a
S
2603 # TODO: Check acodec/vcodec
2604 return False
2605
2606 requested_formats = info_dict['requested_formats']
0202b52a 2607 old_ext = info_dict['ext']
3b297919 2608 if self.params.get('merge_output_format') is None and not compatible_formats(requested_formats):
2609 info_dict['ext'] = 'mkv'
2610 self.report_warning(
2611 'Requested formats are incompatible for merge and will be merged into mkv.')
0202b52a 2612
2613 def correct_ext(filename):
2614 filename_real_ext = os.path.splitext(filename)[1][1:]
2615 filename_wo_ext = (
2616 os.path.splitext(filename)[0]
2617 if filename_real_ext == old_ext
2618 else filename)
2619 return '%s.%s' % (filename_wo_ext, info_dict['ext'])
2620
38c6902b 2621 # Ensure filename always has a correct extension for successful merge
0202b52a 2622 full_filename = correct_ext(full_filename)
2623 temp_filename = correct_ext(temp_filename)
2624 dl_filename = existing_file(full_filename, temp_filename)
1ea24129 2625 info_dict['__real_download'] = False
18e674b4 2626
2627 _protocols = set(determine_protocol(f) for f in requested_formats)
2628 if len(_protocols) == 1:
2629 info_dict['protocol'] = _protocols.pop()
2630 directly_mergable = (
2631 'no-direct-merge' not in self.params.get('compat_opts', [])
2632 and info_dict.get('protocol') is not None # All requested formats have same protocol
2633 and not self.params.get('allow_unplayable_formats')
2634 and get_suitable_downloader(info_dict, self.params).__name__ == 'FFmpegFD')
2635 if directly_mergable:
2636 info_dict['url'] = requested_formats[0]['url']
2637 # Treat it as a single download
2638 dl_filename = existing_file(full_filename, temp_filename)
2639 if dl_filename is None:
2640 success, real_download = self.dl(temp_filename, info_dict)
2641 info_dict['__real_download'] = real_download
2642 else:
2643 downloaded = []
2644 merger = FFmpegMergerPP(self)
2645 if self.params.get('allow_unplayable_formats'):
2646 self.report_warning(
2647 'You have requested merging of multiple formats '
2648 'while also allowing unplayable formats to be downloaded. '
2649 'The formats won\'t be merged to prevent data corruption.')
2650 elif not merger.available:
2651 self.report_warning(
2652 'You have requested merging of multiple formats but ffmpeg is not installed. '
2653 'The formats won\'t be merged.')
2654
2655 if dl_filename is None:
2656 for f in requested_formats:
2657 new_info = dict(info_dict)
2658 del new_info['requested_formats']
2659 new_info.update(f)
2660 fname = prepend_extension(
2661 self.prepare_filename(new_info, 'temp'),
2662 'f%s' % f['format_id'], new_info['ext'])
2663 if not self._ensure_dir_exists(fname):
2664 return
2665 downloaded.append(fname)
2666 partial_success, real_download = self.dl(fname, new_info)
2667 info_dict['__real_download'] = info_dict['__real_download'] or real_download
2668 success = success and partial_success
2669 if merger.available and not self.params.get('allow_unplayable_formats'):
2670 info_dict['__postprocessors'].append(merger)
2671 info_dict['__files_to_merge'] = downloaded
2672 # Even if there were no downloads, it is being merged only now
2673 info_dict['__real_download'] = True
2674 else:
2675 for file in downloaded:
2676 files_to_move[file] = None
4340deca
P
2677 else:
2678 # Just a single file
0202b52a 2679 dl_filename = existing_file(full_filename, temp_filename)
2680 if dl_filename is None:
e8e73840 2681 success, real_download = self.dl(temp_filename, info_dict)
0202b52a 2682 info_dict['__real_download'] = real_download
2683
0202b52a 2684 dl_filename = dl_filename or temp_filename
c571435f 2685 info_dict['__finaldir'] = os.path.dirname(os.path.abspath(encodeFilename(full_filename)))
0202b52a 2686
3158150c 2687 except network_exceptions as err:
7960b056 2688 self.report_error('unable to download video data: %s' % error_to_compat_str(err))
4340deca
P
2689 return
2690 except (OSError, IOError) as err:
2691 raise UnavailableVideoError(err)
2692 except (ContentTooShortError, ) as err:
2693 self.report_error('content too short (expected %s bytes and served %s)' % (err.expected, err.downloaded))
2694 return
8222d8de 2695
de6000d9 2696 if success and full_filename != '-':
f17f8651 2697
fd7cfb64 2698 def fixup():
2699 do_fixup = True
2700 fixup_policy = self.params.get('fixup')
2701 vid = info_dict['id']
2702
2703 if fixup_policy in ('ignore', 'never'):
2704 return
2705 elif fixup_policy == 'warn':
2706 do_fixup = False
f89b3e2d 2707 elif fixup_policy != 'force':
2708 assert fixup_policy in ('detect_or_warn', None)
2709 if not info_dict.get('__real_download'):
2710 do_fixup = False
fd7cfb64 2711
2712 def ffmpeg_fixup(cndn, msg, cls):
2713 if not cndn:
2714 return
2715 if not do_fixup:
2716 self.report_warning(f'{vid}: {msg}')
2717 return
2718 pp = cls(self)
2719 if pp.available:
2720 info_dict['__postprocessors'].append(pp)
2721 else:
2722 self.report_warning(f'{vid}: {msg}. Install ffmpeg to fix this automatically')
2723
2724 stretched_ratio = info_dict.get('stretched_ratio')
2725 ffmpeg_fixup(
2726 stretched_ratio not in (1, None),
2727 f'Non-uniform pixel ratio {stretched_ratio}',
2728 FFmpegFixupStretchedPP)
2729
2730 ffmpeg_fixup(
2731 (info_dict.get('requested_formats') is None
2732 and info_dict.get('container') == 'm4a_dash'
2733 and info_dict.get('ext') == 'm4a'),
2734 'writing DASH m4a. Only some players support this container',
2735 FFmpegFixupM4aPP)
2736
2737 downloader = (get_suitable_downloader(info_dict, self.params).__name__
2738 if 'protocol' in info_dict else None)
2739 ffmpeg_fixup(downloader == 'HlsFD', 'malformed AAC bitstream detected', FFmpegFixupM3u8PP)
e36d50c5 2740 ffmpeg_fixup(downloader == 'WebSocketFragmentFD', 'malformed timestamps detected', FFmpegFixupTimestampPP)
2741 ffmpeg_fixup(downloader == 'WebSocketFragmentFD', 'malformed duration detected', FFmpegFixupDurationPP)
fd7cfb64 2742
2743 fixup()
8222d8de 2744 try:
23c1a667 2745 info_dict = self.post_process(dl_filename, info_dict, files_to_move)
af819c21 2746 except PostProcessingError as err:
2747 self.report_error('Postprocessing: %s' % str(err))
8222d8de 2748 return
ab8e5e51
AM
2749 try:
2750 for ph in self._post_hooks:
23c1a667 2751 ph(info_dict['filepath'])
ab8e5e51
AM
2752 except Exception as err:
2753 self.report_error('post hooks: %s' % str(err))
2754 return
2d30509f 2755 must_record_download_archive = True
2756
2757 if must_record_download_archive or self.params.get('force_write_download_archive', False):
2758 self.record_download_archive(info_dict)
c3e6ffba 2759 max_downloads = self.params.get('max_downloads')
2760 if max_downloads is not None and self._num_downloads >= int(max_downloads):
2761 raise MaxDownloadsReached()
8222d8de
JMF
2762
2763 def download(self, url_list):
2764 """Download a given list of URLs."""
de6000d9 2765 outtmpl = self.outtmpl_dict['default']
3089bc74
S
2766 if (len(url_list) > 1
2767 and outtmpl != '-'
2768 and '%' not in outtmpl
2769 and self.params.get('max_downloads') != 1):
acd69589 2770 raise SameFileError(outtmpl)
8222d8de
JMF
2771
2772 for url in url_list:
2773 try:
5f6a1245 2774 # It also downloads the videos
61aa5ba3
S
2775 res = self.extract_info(
2776 url, force_generic_extractor=self.params.get('force_generic_extractor', False))
8222d8de 2777 except UnavailableVideoError:
6febd1c1 2778 self.report_error('unable to download video')
8222d8de 2779 except MaxDownloadsReached:
8b0d7497 2780 self.to_screen('[info] Maximum number of downloaded files reached')
2781 raise
2782 except ExistingVideoReached:
d83cb531 2783 self.to_screen('[info] Encountered a file that is already in the archive, stopping due to --break-on-existing')
8b0d7497 2784 raise
2785 except RejectedVideoReached:
d83cb531 2786 self.to_screen('[info] Encountered a file that did not match filter, stopping due to --break-on-reject')
8222d8de 2787 raise
63e0be34
PH
2788 else:
2789 if self.params.get('dump_single_json', False):
277d6ff5 2790 self.post_extract(res)
75d43ca0 2791 self.to_stdout(json.dumps(res, default=repr))
8222d8de
JMF
2792
2793 return self._download_retcode
2794
1dcc4c0c 2795 def download_with_info_file(self, info_filename):
31bd3925
JMF
2796 with contextlib.closing(fileinput.FileInput(
2797 [info_filename], mode='r',
2798 openhook=fileinput.hook_encoded('utf-8'))) as f:
2799 # FileInput doesn't have a read method, we can't call json.load
498f5606 2800 info = self.filter_requested_info(json.loads('\n'.join(f)), self.params.get('clean_infojson', True))
d4943898
JMF
2801 try:
2802 self.process_ie_result(info, download=True)
d3f62c19 2803 except (DownloadError, EntryNotInPlaylist, ThrottledDownload):
d4943898
JMF
2804 webpage_url = info.get('webpage_url')
2805 if webpage_url is not None:
6febd1c1 2806 self.report_warning('The info failed to download, trying with "%s"' % webpage_url)
d4943898
JMF
2807 return self.download([webpage_url])
2808 else:
2809 raise
2810 return self._download_retcode
1dcc4c0c 2811
cb202fd2 2812 @staticmethod
75d43ca0 2813 def filter_requested_info(info_dict, actually_filter=True):
ae8f99e6 2814 remove_keys = ['__original_infodict'] # Always remove this since this may contain a copy of the entire dict
2815 keep_keys = ['_type'], # Always keep this to facilitate load-info-json
2816 if actually_filter:
bd99f6e6 2817 remove_keys += ('requested_formats', 'requested_subtitles', 'requested_entries', 'filepath', 'entries', 'original_url')
ae8f99e6 2818 empty_values = (None, {}, [], set(), tuple())
2819 reject = lambda k, v: k not in keep_keys and (
2820 k.startswith('_') or k in remove_keys or v in empty_values)
2821 else:
394dcd44 2822 info_dict['epoch'] = int(time.time())
ae8f99e6 2823 reject = lambda k, v: k in remove_keys
5226731e 2824 filter_fn = lambda obj: (
b0249bca 2825 list(map(filter_fn, obj)) if isinstance(obj, (LazyList, list, tuple, set))
a515a78d 2826 else obj if not isinstance(obj, dict)
ae8f99e6 2827 else dict((k, filter_fn(v)) for k, v in obj.items() if not reject(k, v)))
5226731e 2828 return filter_fn(info_dict)
cb202fd2 2829
dcf64d43 2830 def run_pp(self, pp, infodict):
5bfa4862 2831 files_to_delete = []
dcf64d43 2832 if '__files_to_move' not in infodict:
2833 infodict['__files_to_move'] = {}
af819c21 2834 files_to_delete, infodict = pp.run(infodict)
5bfa4862 2835 if not files_to_delete:
dcf64d43 2836 return infodict
5bfa4862 2837
2838 if self.params.get('keepvideo', False):
2839 for f in files_to_delete:
dcf64d43 2840 infodict['__files_to_move'].setdefault(f, '')
5bfa4862 2841 else:
2842 for old_filename in set(files_to_delete):
2843 self.to_screen('Deleting original file %s (pass -k to keep)' % old_filename)
2844 try:
2845 os.remove(encodeFilename(old_filename))
2846 except (IOError, OSError):
2847 self.report_warning('Unable to remove downloaded original file')
dcf64d43 2848 if old_filename in infodict['__files_to_move']:
2849 del infodict['__files_to_move'][old_filename]
2850 return infodict
5bfa4862 2851
277d6ff5 2852 @staticmethod
2853 def post_extract(info_dict):
2854 def actual_post_extract(info_dict):
2855 if info_dict.get('_type') in ('playlist', 'multi_video'):
2856 for video_dict in info_dict.get('entries', {}):
b050d210 2857 actual_post_extract(video_dict or {})
277d6ff5 2858 return
2859
07cce701 2860 post_extractor = info_dict.get('__post_extractor') or (lambda: {})
4ec82a72 2861 extra = post_extractor().items()
2862 info_dict.update(extra)
07cce701 2863 info_dict.pop('__post_extractor', None)
277d6ff5 2864
4ec82a72 2865 original_infodict = info_dict.get('__original_infodict') or {}
2866 original_infodict.update(extra)
2867 original_infodict.pop('__post_extractor', None)
2868
b050d210 2869 actual_post_extract(info_dict or {})
277d6ff5 2870
56d868db 2871 def pre_process(self, ie_info, key='pre_process', files_to_move=None):
5bfa4862 2872 info = dict(ie_info)
56d868db 2873 info['__files_to_move'] = files_to_move or {}
2874 for pp in self._pps[key]:
dcf64d43 2875 info = self.run_pp(pp, info)
56d868db 2876 return info, info.pop('__files_to_move', None)
5bfa4862 2877
dcf64d43 2878 def post_process(self, filename, ie_info, files_to_move=None):
8222d8de
JMF
2879 """Run all the postprocessors on the given file."""
2880 info = dict(ie_info)
2881 info['filepath'] = filename
dcf64d43 2882 info['__files_to_move'] = files_to_move or {}
0202b52a 2883
56d868db 2884 for pp in ie_info.get('__postprocessors', []) + self._pps['post_process']:
dcf64d43 2885 info = self.run_pp(pp, info)
2886 info = self.run_pp(MoveFilesAfterDownloadPP(self), info)
2887 del info['__files_to_move']
56d868db 2888 for pp in self._pps['after_move']:
dcf64d43 2889 info = self.run_pp(pp, info)
23c1a667 2890 return info
c1c9a79c 2891
5db07df6 2892 def _make_archive_id(self, info_dict):
e9fef7ee
S
2893 video_id = info_dict.get('id')
2894 if not video_id:
2895 return
5db07df6
PH
2896 # Future-proof against any change in case
2897 # and backwards compatibility with prior versions
e9fef7ee 2898 extractor = info_dict.get('extractor_key') or info_dict.get('ie_key') # key in a playlist
7012b23c 2899 if extractor is None:
1211bb6d
S
2900 url = str_or_none(info_dict.get('url'))
2901 if not url:
2902 return
e9fef7ee
S
2903 # Try to find matching extractor for the URL and take its ie_key
2904 for ie in self._ies:
1211bb6d 2905 if ie.suitable(url):
e9fef7ee
S
2906 extractor = ie.ie_key()
2907 break
2908 else:
2909 return
d0757229 2910 return '%s %s' % (extractor.lower(), video_id)
5db07df6
PH
2911
2912 def in_download_archive(self, info_dict):
2913 fn = self.params.get('download_archive')
2914 if fn is None:
2915 return False
2916
2917 vid_id = self._make_archive_id(info_dict)
e9fef7ee 2918 if not vid_id:
7012b23c 2919 return False # Incomplete video information
5db07df6 2920
a45e8619 2921 return vid_id in self.archive
c1c9a79c
PH
2922
2923 def record_download_archive(self, info_dict):
2924 fn = self.params.get('download_archive')
2925 if fn is None:
2926 return
5db07df6
PH
2927 vid_id = self._make_archive_id(info_dict)
2928 assert vid_id
c1c9a79c 2929 with locked_file(fn, 'a', encoding='utf-8') as archive_file:
6febd1c1 2930 archive_file.write(vid_id + '\n')
a45e8619 2931 self.archive.add(vid_id)
dd82ffea 2932
8c51aa65 2933 @staticmethod
8abeeb94 2934 def format_resolution(format, default='unknown'):
fb04e403 2935 if format.get('vcodec') == 'none':
8326b00a 2936 if format.get('acodec') == 'none':
2937 return 'images'
fb04e403 2938 return 'audio only'
f49d89ee
PH
2939 if format.get('resolution') is not None:
2940 return format['resolution']
35615307
DA
2941 if format.get('width') and format.get('height'):
2942 res = '%dx%d' % (format['width'], format['height'])
2943 elif format.get('height'):
2944 res = '%sp' % format['height']
2945 elif format.get('width'):
388ae76b 2946 res = '%dx?' % format['width']
8c51aa65 2947 else:
8abeeb94 2948 res = default
8c51aa65
JMF
2949 return res
2950
c57f7757
PH
2951 def _format_note(self, fdict):
2952 res = ''
2953 if fdict.get('ext') in ['f4f', 'f4m']:
2954 res += '(unsupported) '
32f90364
PH
2955 if fdict.get('language'):
2956 if res:
2957 res += ' '
9016d76f 2958 res += '[%s] ' % fdict['language']
c57f7757
PH
2959 if fdict.get('format_note') is not None:
2960 res += fdict['format_note'] + ' '
2961 if fdict.get('tbr') is not None:
2962 res += '%4dk ' % fdict['tbr']
2963 if fdict.get('container') is not None:
2964 if res:
2965 res += ', '
2966 res += '%s container' % fdict['container']
3089bc74
S
2967 if (fdict.get('vcodec') is not None
2968 and fdict.get('vcodec') != 'none'):
c57f7757
PH
2969 if res:
2970 res += ', '
2971 res += fdict['vcodec']
91c7271a 2972 if fdict.get('vbr') is not None:
c57f7757
PH
2973 res += '@'
2974 elif fdict.get('vbr') is not None and fdict.get('abr') is not None:
2975 res += 'video@'
2976 if fdict.get('vbr') is not None:
2977 res += '%4dk' % fdict['vbr']
fbb21cf5 2978 if fdict.get('fps') is not None:
5d583bdf
S
2979 if res:
2980 res += ', '
2981 res += '%sfps' % fdict['fps']
c57f7757
PH
2982 if fdict.get('acodec') is not None:
2983 if res:
2984 res += ', '
2985 if fdict['acodec'] == 'none':
2986 res += 'video only'
2987 else:
2988 res += '%-5s' % fdict['acodec']
2989 elif fdict.get('abr') is not None:
2990 if res:
2991 res += ', '
2992 res += 'audio'
2993 if fdict.get('abr') is not None:
2994 res += '@%3dk' % fdict['abr']
2995 if fdict.get('asr') is not None:
2996 res += ' (%5dHz)' % fdict['asr']
2997 if fdict.get('filesize') is not None:
2998 if res:
2999 res += ', '
3000 res += format_bytes(fdict['filesize'])
9732d77e
PH
3001 elif fdict.get('filesize_approx') is not None:
3002 if res:
3003 res += ', '
3004 res += '~' + format_bytes(fdict['filesize_approx'])
c57f7757 3005 return res
91c7271a 3006
76d321f6 3007 def _format_note_table(self, f):
3008 def join_fields(*vargs):
3009 return ', '.join((val for val in vargs if val != ''))
3010
3011 return join_fields(
3012 'UNSUPPORTED' if f.get('ext') in ('f4f', 'f4m') else '',
3013 format_field(f, 'language', '[%s]'),
3014 format_field(f, 'format_note'),
3015 format_field(f, 'container', ignore=(None, f.get('ext'))),
3016 format_field(f, 'asr', '%5dHz'))
3017
c57f7757 3018 def list_formats(self, info_dict):
94badb25 3019 formats = info_dict.get('formats', [info_dict])
53ed7066 3020 new_format = (
3021 'list-formats' not in self.params.get('compat_opts', [])
169dbde9 3022 and self.params.get('listformats_table', True) is not False)
76d321f6 3023 if new_format:
3024 table = [
3025 [
3026 format_field(f, 'format_id'),
3027 format_field(f, 'ext'),
3028 self.format_resolution(f),
3029 format_field(f, 'fps', '%d'),
3030 '|',
3031 format_field(f, 'filesize', ' %s', func=format_bytes) + format_field(f, 'filesize_approx', '~%s', func=format_bytes),
3032 format_field(f, 'tbr', '%4dk'),
52a8a1e1 3033 shorten_protocol_name(f.get('protocol', '').replace("native", "n")),
76d321f6 3034 '|',
3035 format_field(f, 'vcodec', default='unknown').replace('none', ''),
3036 format_field(f, 'vbr', '%4dk'),
3037 format_field(f, 'acodec', default='unknown').replace('none', ''),
3038 format_field(f, 'abr', '%3dk'),
3039 format_field(f, 'asr', '%5dHz'),
3040 self._format_note_table(f)]
3041 for f in formats
3042 if f.get('preference') is None or f['preference'] >= -1000]
3043 header_line = ['ID', 'EXT', 'RESOLUTION', 'FPS', '|', ' FILESIZE', ' TBR', 'PROTO',
3044 '|', 'VCODEC', ' VBR', 'ACODEC', ' ABR', ' ASR', 'NOTE']
3045 else:
3046 table = [
3047 [
3048 format_field(f, 'format_id'),
3049 format_field(f, 'ext'),
3050 self.format_resolution(f),
3051 self._format_note(f)]
3052 for f in formats
3053 if f.get('preference') is None or f['preference'] >= -1000]
3054 header_line = ['format code', 'extension', 'resolution', 'note']
57dd9a8f 3055
cfb56d1a 3056 self.to_screen(
169dbde9 3057 '[info] Available formats for %s:' % info_dict['id'])
3058 self.to_stdout(render_table(
bc97cdae 3059 header_line, table, delim=new_format, extraGap=(0 if new_format else 1), hideEmpty=new_format))
cfb56d1a
PH
3060
3061 def list_thumbnails(self, info_dict):
b0249bca 3062 thumbnails = list(info_dict.get('thumbnails'))
cfb56d1a 3063 if not thumbnails:
b7b72db9 3064 self.to_screen('[info] No thumbnails present for %s' % info_dict['id'])
3065 return
cfb56d1a
PH
3066
3067 self.to_screen(
3068 '[info] Thumbnails for %s:' % info_dict['id'])
169dbde9 3069 self.to_stdout(render_table(
cfb56d1a
PH
3070 ['ID', 'width', 'height', 'URL'],
3071 [[t['id'], t.get('width', 'unknown'), t.get('height', 'unknown'), t['url']] for t in thumbnails]))
dca08720 3072
360e1ca5 3073 def list_subtitles(self, video_id, subtitles, name='subtitles'):
a504ced0 3074 if not subtitles:
360e1ca5 3075 self.to_screen('%s has no %s' % (video_id, name))
a504ced0 3076 return
a504ced0 3077 self.to_screen(
edab9dbf 3078 'Available %s for %s:' % (name, video_id))
2412044c 3079
3080 def _row(lang, formats):
49c258e1 3081 exts, names = zip(*((f['ext'], f.get('name') or 'unknown') for f in reversed(formats)))
2412044c 3082 if len(set(names)) == 1:
7aee40c1 3083 names = [] if names[0] == 'unknown' else names[:1]
2412044c 3084 return [lang, ', '.join(names), ', '.join(exts)]
3085
169dbde9 3086 self.to_stdout(render_table(
2412044c 3087 ['Language', 'Name', 'Formats'],
3088 [_row(lang, formats) for lang, formats in subtitles.items()],
3089 hideEmpty=True))
a504ced0 3090
dca08720
PH
3091 def urlopen(self, req):
3092 """ Start an HTTP download """
82d8a8b6 3093 if isinstance(req, compat_basestring):
67dda517 3094 req = sanitized_Request(req)
19a41fc6 3095 return self._opener.open(req, timeout=self._socket_timeout)
dca08720
PH
3096
3097 def print_debug_header(self):
3098 if not self.params.get('verbose'):
3099 return
62fec3b2 3100
4192b51c 3101 if type('') is not compat_str:
067aa17e 3102 # Python 2.6 on SLES11 SP1 (https://github.com/ytdl-org/youtube-dl/issues/3326)
4192b51c
PH
3103 self.report_warning(
3104 'Your Python is broken! Update to a newer and supported version')
3105
c6afed48
PH
3106 stdout_encoding = getattr(
3107 sys.stdout, 'encoding', 'missing (%s)' % type(sys.stdout).__name__)
b0472057 3108 encoding_str = (
734f90bb
PH
3109 '[debug] Encodings: locale %s, fs %s, out %s, pref %s\n' % (
3110 locale.getpreferredencoding(),
3111 sys.getfilesystemencoding(),
c6afed48 3112 stdout_encoding,
b0472057 3113 self.get_encoding()))
4192b51c 3114 write_string(encoding_str, encoding=None)
734f90bb 3115
e5813e53 3116 source = (
3117 '(exe)' if hasattr(sys, 'frozen')
3118 else '(zip)' if isinstance(globals().get('__loader__'), zipimporter)
3119 else '(source)' if os.path.basename(sys.argv[0]) == '__main__.py'
3120 else '')
3121 self._write_string('[debug] yt-dlp version %s %s\n' % (__version__, source))
e0986e31 3122 if _LAZY_LOADER:
f74980cb 3123 self._write_string('[debug] Lazy loading extractors enabled\n')
3124 if _PLUGIN_CLASSES:
3125 self._write_string(
3126 '[debug] Plugin Extractors: %s\n' % [ie.ie_key() for ie in _PLUGIN_CLASSES])
53ed7066 3127 if self.params.get('compat_opts'):
3128 self._write_string(
3129 '[debug] Compatibility options: %s\n' % ', '.join(self.params.get('compat_opts')))
dca08720
PH
3130 try:
3131 sp = subprocess.Popen(
3132 ['git', 'rev-parse', '--short', 'HEAD'],
3133 stdout=subprocess.PIPE, stderr=subprocess.PIPE,
3134 cwd=os.path.dirname(os.path.abspath(__file__)))
f5b1bca9 3135 out, err = process_communicate_or_kill(sp)
dca08720
PH
3136 out = out.decode().strip()
3137 if re.match('[0-9a-f]+', out):
f74980cb 3138 self._write_string('[debug] Git HEAD: %s\n' % out)
70a1165b 3139 except Exception:
dca08720
PH
3140 try:
3141 sys.exc_clear()
70a1165b 3142 except Exception:
dca08720 3143 pass
b300cda4
S
3144
3145 def python_implementation():
3146 impl_name = platform.python_implementation()
3147 if impl_name == 'PyPy' and hasattr(sys, 'pypy_version_info'):
3148 return impl_name + ' version %d.%d.%d' % sys.pypy_version_info[:3]
3149 return impl_name
3150
e5813e53 3151 self._write_string('[debug] Python version %s (%s %s) - %s\n' % (
3152 platform.python_version(),
3153 python_implementation(),
3154 platform.architecture()[0],
b300cda4 3155 platform_name()))
d28b5171 3156
73fac4e9 3157 exe_versions = FFmpegPostProcessor.get_versions(self)
4c83c967 3158 exe_versions['rtmpdump'] = rtmpdump_version()
feee8d32 3159 exe_versions['phantomjs'] = PhantomJSwrapper._version()
d28b5171
PH
3160 exe_str = ', '.join(
3161 '%s %s' % (exe, v)
3162 for exe, v in sorted(exe_versions.items())
3163 if v
3164 )
3165 if not exe_str:
3166 exe_str = 'none'
3167 self._write_string('[debug] exe versions: %s\n' % exe_str)
dca08720
PH
3168
3169 proxy_map = {}
3170 for handler in self._opener.handlers:
3171 if hasattr(handler, 'proxies'):
3172 proxy_map.update(handler.proxies)
734f90bb 3173 self._write_string('[debug] Proxy map: ' + compat_str(proxy_map) + '\n')
dca08720 3174
58b1f00d
PH
3175 if self.params.get('call_home', False):
3176 ipaddr = self.urlopen('https://yt-dl.org/ip').read().decode('utf-8')
3177 self._write_string('[debug] Public IP address: %s\n' % ipaddr)
f5546c0b 3178 return
58b1f00d
PH
3179 latest_version = self.urlopen(
3180 'https://yt-dl.org/latest/version').read().decode('utf-8')
3181 if version_tuple(latest_version) > version_tuple(__version__):
3182 self.report_warning(
3183 'You are using an outdated version (newest version: %s)! '
3184 'See https://yt-dl.org/update if you need help updating.' %
3185 latest_version)
3186
e344693b 3187 def _setup_opener(self):
6ad14cab 3188 timeout_val = self.params.get('socket_timeout')
19a41fc6 3189 self._socket_timeout = 600 if timeout_val is None else float(timeout_val)
6ad14cab 3190
dca08720
PH
3191 opts_cookiefile = self.params.get('cookiefile')
3192 opts_proxy = self.params.get('proxy')
3193
3194 if opts_cookiefile is None:
3195 self.cookiejar = compat_cookiejar.CookieJar()
3196 else:
590bc6f6 3197 opts_cookiefile = expand_path(opts_cookiefile)
1bab3437 3198 self.cookiejar = YoutubeDLCookieJar(opts_cookiefile)
dca08720 3199 if os.access(opts_cookiefile, os.R_OK):
1d88b3e6 3200 self.cookiejar.load(ignore_discard=True, ignore_expires=True)
dca08720 3201
6a3f4c3f 3202 cookie_processor = YoutubeDLCookieProcessor(self.cookiejar)
dca08720
PH
3203 if opts_proxy is not None:
3204 if opts_proxy == '':
3205 proxies = {}
3206 else:
3207 proxies = {'http': opts_proxy, 'https': opts_proxy}
3208 else:
3209 proxies = compat_urllib_request.getproxies()
067aa17e 3210 # Set HTTPS proxy to HTTP one if given (https://github.com/ytdl-org/youtube-dl/issues/805)
dca08720
PH
3211 if 'http' in proxies and 'https' not in proxies:
3212 proxies['https'] = proxies['http']
91410c9b 3213 proxy_handler = PerRequestProxyHandler(proxies)
a0ddb8a2
PH
3214
3215 debuglevel = 1 if self.params.get('debug_printtraffic') else 0
be4a824d
PH
3216 https_handler = make_HTTPS_handler(self.params, debuglevel=debuglevel)
3217 ydlh = YoutubeDLHandler(self.params, debuglevel=debuglevel)
fca6dba8 3218 redirect_handler = YoutubeDLRedirectHandler()
8b172c2e 3219 data_handler = compat_urllib_request_DataHandler()
6240b0a2
JMF
3220
3221 # When passing our own FileHandler instance, build_opener won't add the
3222 # default FileHandler and allows us to disable the file protocol, which
3223 # can be used for malicious purposes (see
067aa17e 3224 # https://github.com/ytdl-org/youtube-dl/issues/8227)
6240b0a2
JMF
3225 file_handler = compat_urllib_request.FileHandler()
3226
3227 def file_open(*args, **kwargs):
7a5c1cfe 3228 raise compat_urllib_error.URLError('file:// scheme is explicitly disabled in yt-dlp for security reasons')
6240b0a2
JMF
3229 file_handler.file_open = file_open
3230
3231 opener = compat_urllib_request.build_opener(
fca6dba8 3232 proxy_handler, https_handler, cookie_processor, ydlh, redirect_handler, data_handler, file_handler)
2461f79d 3233
dca08720
PH
3234 # Delete the default user-agent header, which would otherwise apply in
3235 # cases where our custom HTTP handler doesn't come into play
067aa17e 3236 # (See https://github.com/ytdl-org/youtube-dl/issues/1309 for details)
dca08720
PH
3237 opener.addheaders = []
3238 self._opener = opener
62fec3b2
PH
3239
3240 def encode(self, s):
3241 if isinstance(s, bytes):
3242 return s # Already encoded
3243
3244 try:
3245 return s.encode(self.get_encoding())
3246 except UnicodeEncodeError as err:
3247 err.reason = err.reason + '. Check your system encoding configuration or use the --encoding option.'
3248 raise
3249
3250 def get_encoding(self):
3251 encoding = self.params.get('encoding')
3252 if encoding is None:
3253 encoding = preferredencoding()
3254 return encoding
ec82d85a 3255
de6000d9 3256 def _write_thumbnails(self, info_dict, filename): # return the extensions
6c4fd172 3257 write_all = self.params.get('write_all_thumbnails', False)
3258 thumbnails = []
3259 if write_all or self.params.get('writethumbnail', False):
0202b52a 3260 thumbnails = info_dict.get('thumbnails') or []
6c4fd172 3261 multiple = write_all and len(thumbnails) > 1
ec82d85a 3262
0202b52a 3263 ret = []
981052c9 3264 for t in thumbnails[::-1]:
ec82d85a 3265 thumb_ext = determine_ext(t['url'], 'jpg')
6c4fd172 3266 suffix = '%s.' % t['id'] if multiple else ''
3267 thumb_display_id = '%s ' % t['id'] if multiple else ''
885cc0b7 3268 thumb_filename = replace_extension(filename, suffix + thumb_ext, info_dict.get('ext'))
ec82d85a 3269
0c3d0f51 3270 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(thumb_filename)):
de6000d9 3271 ret.append(suffix + thumb_ext)
8ba87148 3272 t['filepath'] = thumb_filename
ec82d85a
PH
3273 self.to_screen('[%s] %s: Thumbnail %sis already present' %
3274 (info_dict['extractor'], info_dict['id'], thumb_display_id))
3275 else:
5ef7d9bd 3276 self.to_screen('[%s] %s: Downloading thumbnail %s ...' %
ec82d85a
PH
3277 (info_dict['extractor'], info_dict['id'], thumb_display_id))
3278 try:
3279 uf = self.urlopen(t['url'])
d3d89c32 3280 with open(encodeFilename(thumb_filename), 'wb') as thumbf:
ec82d85a 3281 shutil.copyfileobj(uf, thumbf)
de6000d9 3282 ret.append(suffix + thumb_ext)
ec82d85a
PH
3283 self.to_screen('[%s] %s: Writing thumbnail %sto: %s' %
3284 (info_dict['extractor'], info_dict['id'], thumb_display_id, thumb_filename))
885cc0b7 3285 t['filepath'] = thumb_filename
3158150c 3286 except network_exceptions as err:
ec82d85a 3287 self.report_warning('Unable to download thumbnail "%s": %s' %
9b9c5355 3288 (t['url'], error_to_compat_str(err)))
6c4fd172 3289 if ret and not write_all:
3290 break
0202b52a 3291 return ret