]> jfr.im git - yt-dlp.git/blame - yt_dlp/YoutubeDL.py
[RCTIPlus] Add extractor (#443)
[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
4524baf0 212 paths: Dictionary of output paths. The allowed keys are 'home'
213 'temp' and the keys of OUTTMPL_TYPES (in utils.py)
de6000d9 214 outtmpl: Dictionary of templates for output names. Allowed keys
4524baf0 215 are 'default' and the keys of OUTTMPL_TYPES (in utils.py).
216 A string a also accepted for backward compatibility
a820dc72
RA
217 outtmpl_na_placeholder: Placeholder for unavailable meta fields.
218 restrictfilenames: Do not allow "&" and spaces in file names
219 trim_file_name: Limit length of filename (extension excluded)
4524baf0 220 windowsfilenames: Force the filenames to be windows compatible
a820dc72 221 ignoreerrors: Do not stop on download errors
7a5c1cfe 222 (Default True when running yt-dlp,
a820dc72 223 but False when directly accessing YoutubeDL class)
26e2805c 224 skip_playlist_after_errors: Number of allowed failures until the rest of
225 the playlist is skipped
d22dec74 226 force_generic_extractor: Force downloader to use the generic extractor
0c3d0f51 227 overwrites: Overwrite all video and metadata files if True,
228 overwrite only non-video files if None
229 and don't overwrite any file if False
8222d8de
JMF
230 playliststart: Playlist item to start at.
231 playlistend: Playlist item to end at.
c14e88f0 232 playlist_items: Specific indices of playlist to download.
ff815fe6 233 playlistreverse: Download playlist items in reverse order.
75822ca7 234 playlistrandom: Download playlist items in random order.
8222d8de
JMF
235 matchtitle: Download only matching titles.
236 rejecttitle: Reject downloads for matching titles.
8bf9319e 237 logger: Log messages to a logging.Logger instance.
8222d8de
JMF
238 logtostderr: Log messages to stderr instead of stdout.
239 writedescription: Write the video description to a .description file
240 writeinfojson: Write the video description to a .info.json file
75d43ca0 241 clean_infojson: Remove private fields from the infojson
06167fbb 242 writecomments: Extract video comments. This will not be written to disk
243 unless writeinfojson is also given
1fb07d10 244 writeannotations: Write the video annotations to a .annotations.xml file
8222d8de 245 writethumbnail: Write the thumbnail image to a file
c25228e5 246 allow_playlist_files: Whether to write playlists' description, infojson etc
247 also to disk when using the 'write*' options
ec82d85a 248 write_all_thumbnails: Write all thumbnail formats to files
732044af 249 writelink: Write an internet shortcut file, depending on the
250 current platform (.url/.webloc/.desktop)
251 writeurllink: Write a Windows internet shortcut file (.url)
252 writewebloclink: Write a macOS internet shortcut file (.webloc)
253 writedesktoplink: Write a Linux internet shortcut file (.desktop)
8222d8de 254 writesubtitles: Write the video subtitles to a file
741dd8ea 255 writeautomaticsub: Write the automatically generated subtitles to a file
c32b0aab 256 allsubtitles: Deprecated - Use subtitlelangs = ['all']
257 Downloads all the subtitles of the video
0b7f3118 258 (requires writesubtitles or writeautomaticsub)
8222d8de 259 listsubtitles: Lists all available subtitles for the video
a504ced0 260 subtitlesformat: The format code for subtitles
c32b0aab 261 subtitleslangs: List of languages of the subtitles to download (can be regex).
262 The list may contain "all" to refer to all the available
263 subtitles. The language can be prefixed with a "-" to
264 exclude it from the requested languages. Eg: ['all', '-live_chat']
8222d8de
JMF
265 keepvideo: Keep the video file after post-processing
266 daterange: A DateRange object, download only if the upload_date is in the range.
267 skip_download: Skip the actual download of the video file
c35f9e72 268 cachedir: Location of the cache files in the filesystem.
a0e07d31 269 False to disable filesystem cache.
47192f92 270 noplaylist: Download single video instead of a playlist if in doubt.
8dbe9899
PH
271 age_limit: An integer representing the user's age in years.
272 Unsuitable videos for the given age are skipped.
5fe18bdb
PH
273 min_views: An integer representing the minimum view count the video
274 must have in order to not be skipped.
275 Videos without view count information are always
276 downloaded. None for no limit.
277 max_views: An integer representing the maximum view count.
278 Videos that are more popular than that are not
279 downloaded.
280 Videos without view count information are always
281 downloaded. None for no limit.
282 download_archive: File name of a file where all downloads are recorded.
c1c9a79c
PH
283 Videos already present in the file are not downloaded
284 again.
8a51f564 285 break_on_existing: Stop the download process after attempting to download a
286 file that is in the archive.
287 break_on_reject: Stop the download process when encountering a video that
288 has been filtered out.
289 cookiefile: File name where cookies should be read from and dumped to
a1ee09e8 290 nocheckcertificate:Do not verify SSL certificates
7e8c0af0
PH
291 prefer_insecure: Use HTTP instead of HTTPS to retrieve information.
292 At the moment, this is only supported by YouTube.
a1ee09e8 293 proxy: URL of the proxy server to use
38cce791 294 geo_verification_proxy: URL of the proxy to use for IP address verification
504f20dd 295 on geo-restricted sites.
e344693b 296 socket_timeout: Time to wait for unresponsive hosts, in seconds
0783b09b
PH
297 bidi_workaround: Work around buggy terminals without bidirectional text
298 support, using fridibi
a0ddb8a2 299 debug_printtraffic:Print out sent and received HTTP traffic
7b0817e8 300 include_ads: Download ads as well
04b4d394
PH
301 default_search: Prepend this string if an input url is not valid.
302 'auto' for elaborate guessing
62fec3b2 303 encoding: Use this encoding instead of the system-specified.
e8ee972c 304 extract_flat: Do not resolve URLs, return the immediate result.
057a5206
PH
305 Pass in 'in_playlist' to only show this behavior for
306 playlist items.
4f026faf 307 postprocessors: A list of dictionaries, each with an entry
71b640cc 308 * key: The name of the postprocessor. See
7a5c1cfe 309 yt_dlp/postprocessor/__init__.py for a list.
56d868db 310 * when: When to run the postprocessor. Can be one of
311 pre_process|before_dl|post_process|after_move.
312 Assumed to be 'post_process' if not given
ab8e5e51
AM
313 post_hooks: A list of functions that get called as the final step
314 for each video file, after all postprocessors have been
315 called. The filename will be passed as the only argument.
71b640cc
PH
316 progress_hooks: A list of functions that get called on download
317 progress, with a dictionary with the entries
5cda4eda 318 * status: One of "downloading", "error", or "finished".
ee69b99a 319 Check this first and ignore unknown values.
71b640cc 320
5cda4eda 321 If status is one of "downloading", or "finished", the
ee69b99a
PH
322 following properties may also be present:
323 * filename: The final filename (always present)
5cda4eda 324 * tmpfilename: The filename we're currently writing to
71b640cc
PH
325 * downloaded_bytes: Bytes on disk
326 * total_bytes: Size of the whole file, None if unknown
5cda4eda
PH
327 * total_bytes_estimate: Guess of the eventual file size,
328 None if unavailable.
329 * elapsed: The number of seconds since download started.
71b640cc
PH
330 * eta: The estimated time in seconds, None if unknown
331 * speed: The download speed in bytes/second, None if
332 unknown
5cda4eda
PH
333 * fragment_index: The counter of the currently
334 downloaded video fragment.
335 * fragment_count: The number of fragments (= individual
336 files that will be merged)
71b640cc
PH
337
338 Progress hooks are guaranteed to be called at least once
339 (with status "finished") if the download is successful.
45598f15 340 merge_output_format: Extension to use when merging formats.
6b591b29 341 final_ext: Expected final extension; used to detect when the file was
342 already downloaded and converted. "merge_output_format" is
343 replaced by this extension when given
6271f1ca
PH
344 fixup: Automatically correct known faults of the file.
345 One of:
346 - "never": do nothing
347 - "warn": only emit a warning
348 - "detect_or_warn": check whether we can do anything
62cd676c 349 about it, warn otherwise (default)
504f20dd 350 source_address: Client-side IP address to bind to.
6ec6cb4e 351 call_home: Boolean, true iff we are allowed to contact the
7a5c1cfe 352 yt-dlp servers for debugging. (BROKEN)
1cf376f5 353 sleep_interval_requests: Number of seconds to sleep between requests
354 during extraction
7aa589a5
S
355 sleep_interval: Number of seconds to sleep before each download when
356 used alone or a lower bound of a range for randomized
357 sleep before each download (minimum possible number
358 of seconds to sleep) when used along with
359 max_sleep_interval.
360 max_sleep_interval:Upper bound of a range for randomized sleep before each
361 download (maximum possible number of seconds to sleep).
362 Must only be used along with sleep_interval.
363 Actual sleep time will be a random float from range
364 [sleep_interval; max_sleep_interval].
1cf376f5 365 sleep_interval_subtitles: Number of seconds to sleep before each subtitle download
cfb56d1a
PH
366 listformats: Print an overview of available video formats and exit.
367 list_thumbnails: Print a table of all thumbnails and exit.
347de493
PH
368 match_filter: A function that gets called with the info_dict of
369 every video.
370 If it returns a message, the video is ignored.
371 If it returns None, the video is downloaded.
372 match_filter_func in utils.py is one example for this.
7e5db8c9 373 no_color: Do not emit color codes in output.
0a840f58 374 geo_bypass: Bypass geographic restriction via faking X-Forwarded-For
504f20dd 375 HTTP header
0a840f58 376 geo_bypass_country:
773f291d
S
377 Two-letter ISO 3166-2 country code that will be used for
378 explicit geographic restriction bypassing via faking
504f20dd 379 X-Forwarded-For HTTP header
5f95927a
S
380 geo_bypass_ip_block:
381 IP range in CIDR notation that will be used similarly to
504f20dd 382 geo_bypass_country
71b640cc 383
85729c51 384 The following options determine which downloader is picked:
52a8a1e1 385 external_downloader: A dictionary of protocol keys and the executable of the
386 external downloader to use for it. The allowed protocols
387 are default|http|ftp|m3u8|dash|rtsp|rtmp|mms.
388 Set the value to 'native' to use the native downloader
389 hls_prefer_native: Deprecated - Use external_downloader = {'m3u8': 'native'}
390 or {'m3u8': 'ffmpeg'} instead.
391 Use the native HLS downloader instead of ffmpeg/avconv
bf09af3a
S
392 if True, otherwise use ffmpeg/avconv if False, otherwise
393 use downloader suggested by extractor if None.
53ed7066 394 compat_opts: Compatibility options. See "Differences in default behavior".
18e674b4 395 Note that only format-sort, format-spec, no-live-chat,
396 no-attach-info-json, playlist-index, list-formats,
e858a9d6 397 no-direct-merge, embed-thumbnail-atomicparsley,
398 no-youtube-unavailable-videos, no-youtube-channel-redirect,
399 works when used via the API
fe7e0c98 400
8222d8de 401 The following parameters are not used by YoutubeDL itself, they are used by
7a5c1cfe 402 the downloader (see yt_dlp/downloader/common.py):
51d9739f 403 nopart, updatetime, buffersize, ratelimit, throttledratelimit, min_filesize,
404 max_filesize, test, noresizebuffer, retries, continuedl, noprogress, consoletitle,
405 xattr_set_filesize, external_downloader_args, hls_use_mpegts, http_chunk_size.
76b1bd67
JMF
406
407 The following options are used by the post processors:
d4a24f40 408 prefer_ffmpeg: If False, use avconv instead of ffmpeg if both are available,
e4172ac9 409 otherwise prefer ffmpeg. (avconv support is deprecated)
c0b7d117
S
410 ffmpeg_location: Location of the ffmpeg/avconv binary; either the path
411 to the binary or its containing directory.
43820c03 412 postprocessor_args: A dictionary of postprocessor/executable keys (in lower case)
413 and a list of additional command-line arguments for the
414 postprocessor/executable. The dict can also have "PP+EXE" keys
415 which are used when the given exe is used by the given PP.
416 Use 'default' as the name for arguments to passed to all PP
e409895f 417
418 The following options are used by the extractors:
62bff2c1 419 extractor_retries: Number of times to retry for known errors
420 dynamic_mpd: Whether to process dynamic DASH manifests (default: True)
e409895f 421 hls_split_discontinuity: Split HLS playlists to different formats at
62bff2c1 422 discontinuities such as ad breaks (default: False)
5d3a0e79 423 extractor_args: A dictionary of arguments to be passed to the extractors.
424 See "EXTRACTOR ARGUMENTS" for details.
425 Eg: {'youtube': {'skip': ['dash', 'hls']}}
426 youtube_include_dash_manifest: Deprecated - Use extractor_args instead.
427 If True (default), DASH manifests and related
62bff2c1 428 data will be downloaded and processed by extractor.
429 You can reduce network I/O by disabling it if you don't
430 care about DASH. (only for youtube)
5d3a0e79 431 youtube_include_hls_manifest: Deprecated - Use extractor_args instead.
432 If True (default), HLS manifests and related
62bff2c1 433 data will be downloaded and processed by extractor.
434 You can reduce network I/O by disabling it if you don't
435 care about HLS. (only for youtube)
8222d8de
JMF
436 """
437
c9969434
S
438 _NUMERIC_FIELDS = set((
439 'width', 'height', 'tbr', 'abr', 'asr', 'vbr', 'fps', 'filesize', 'filesize_approx',
440 'timestamp', 'upload_year', 'upload_month', 'upload_day',
441 'duration', 'view_count', 'like_count', 'dislike_count', 'repost_count',
442 'average_rating', 'comment_count', 'age_limit',
443 'start_time', 'end_time',
444 'chapter_number', 'season_number', 'episode_number',
445 'track_number', 'disc_number', 'release_year',
446 'playlist_index',
447 ))
448
8222d8de
JMF
449 params = None
450 _ies = []
56d868db 451 _pps = {'pre_process': [], 'before_dl': [], 'after_move': [], 'post_process': []}
0202b52a 452 __prepare_filename_warned = False
1cf376f5 453 _first_webpage_request = True
8222d8de
JMF
454 _download_retcode = None
455 _num_downloads = None
30a074c2 456 _playlist_level = 0
457 _playlist_urls = set()
8222d8de
JMF
458 _screen_file = None
459
3511266b 460 def __init__(self, params=None, auto_init=True):
8222d8de 461 """Create a FileDownloader object with the given options."""
e9f9a10f
JMF
462 if params is None:
463 params = {}
8222d8de 464 self._ies = []
56c73665 465 self._ies_instances = {}
56d868db 466 self._pps = {'pre_process': [], 'before_dl': [], 'after_move': [], 'post_process': []}
0202b52a 467 self.__prepare_filename_warned = False
1cf376f5 468 self._first_webpage_request = True
ab8e5e51 469 self._post_hooks = []
933605d7 470 self._progress_hooks = []
8222d8de
JMF
471 self._download_retcode = 0
472 self._num_downloads = 0
473 self._screen_file = [sys.stdout, sys.stderr][params.get('logtostderr', False)]
0783b09b 474 self._err_file = sys.stderr
4abf617b
S
475 self.params = {
476 # Default parameters
477 'nocheckcertificate': False,
478 }
479 self.params.update(params)
a0e07d31 480 self.cache = Cache(self)
34308b30 481
a61f4b28 482 if sys.version_info < (3, 6):
483 self.report_warning(
0181adef 484 'Python version %d.%d is not supported! Please update to Python 3.6 or above' % sys.version_info[:2])
a61f4b28 485
be5df5ee
S
486 def check_deprecated(param, option, suggestion):
487 if self.params.get(param) is not None:
53ed7066 488 self.report_warning('%s is deprecated. Use %s instead' % (option, suggestion))
be5df5ee
S
489 return True
490 return False
491
492 if check_deprecated('cn_verification_proxy', '--cn-verification-proxy', '--geo-verification-proxy'):
38cce791
YCH
493 if self.params.get('geo_verification_proxy') is None:
494 self.params['geo_verification_proxy'] = self.params['cn_verification_proxy']
495
0d1bb027 496 check_deprecated('autonumber', '--auto-number', '-o "%(autonumber)s-%(title)s.%(ext)s"')
497 check_deprecated('usetitle', '--title', '-o "%(title)s-%(id)s.%(ext)s"')
53ed7066 498 check_deprecated('useid', '--id', '-o "%(id)s.%(ext)s"')
0d1bb027 499
500 for msg in self.params.get('warnings', []):
501 self.report_warning(msg)
502
6b591b29 503 if self.params.get('final_ext'):
504 if self.params.get('merge_output_format'):
505 self.report_warning('--merge-output-format will be ignored since --remux-video or --recode-video is given')
506 self.params['merge_output_format'] = self.params['final_ext']
507
b9d973be 508 if 'overwrites' in self.params and self.params['overwrites'] is None:
509 del self.params['overwrites']
510
0783b09b 511 if params.get('bidi_workaround', False):
1c088fa8
PH
512 try:
513 import pty
514 master, slave = pty.openpty()
003c69a8 515 width = compat_get_terminal_size().columns
1c088fa8
PH
516 if width is None:
517 width_args = []
518 else:
519 width_args = ['-w', str(width)]
5d681e96 520 sp_kwargs = dict(
1c088fa8
PH
521 stdin=subprocess.PIPE,
522 stdout=slave,
523 stderr=self._err_file)
5d681e96
PH
524 try:
525 self._output_process = subprocess.Popen(
526 ['bidiv'] + width_args, **sp_kwargs
527 )
528 except OSError:
5d681e96
PH
529 self._output_process = subprocess.Popen(
530 ['fribidi', '-c', 'UTF-8'] + width_args, **sp_kwargs)
531 self._output_channel = os.fdopen(master, 'rb')
1c088fa8 532 except OSError as ose:
66e7ace1 533 if ose.errno == errno.ENOENT:
6febd1c1 534 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
535 else:
536 raise
0783b09b 537
3089bc74
S
538 if (sys.platform != 'win32'
539 and sys.getfilesystemencoding() in ['ascii', 'ANSI_X3.4-1968']
540 and not params.get('restrictfilenames', False)):
e9137224 541 # Unicode filesystem API will throw errors (#1474, #13027)
34308b30 542 self.report_warning(
6febd1c1 543 'Assuming --restrict-filenames since file system encoding '
1b725173 544 'cannot encode all characters. '
6febd1c1 545 'Set the LC_ALL environment variable to fix this.')
4a98cdbf 546 self.params['restrictfilenames'] = True
34308b30 547
de6000d9 548 self.outtmpl_dict = self.parse_outtmpl()
486dd09e 549
187986a8 550 # Creating format selector here allows us to catch syntax errors before the extraction
551 self.format_selector = (
552 None if self.params.get('format') is None
553 else self.build_format_selector(self.params['format']))
554
dca08720
PH
555 self._setup_opener()
556
4cd0a709 557 """Preload the archive, if any is specified"""
558 def preload_download_archive(fn):
559 if fn is None:
560 return False
0760b0a7 561 self.write_debug('Loading archive file %r\n' % fn)
4cd0a709 562 try:
563 with locked_file(fn, 'r', encoding='utf-8') as archive_file:
564 for line in archive_file:
565 self.archive.add(line.strip())
566 except IOError as ioe:
567 if ioe.errno != errno.ENOENT:
568 raise
569 return False
570 return True
571
572 self.archive = set()
573 preload_download_archive(self.params.get('download_archive'))
574
3511266b
PH
575 if auto_init:
576 self.print_debug_header()
577 self.add_default_info_extractors()
578
4f026faf 579 for pp_def_raw in self.params.get('postprocessors', []):
4f026faf 580 pp_def = dict(pp_def_raw)
fd7cfb64 581 when = pp_def.pop('when', 'post_process')
582 pp_class = get_postprocessor(pp_def.pop('key'))
4f026faf 583 pp = pp_class(self, **compat_kwargs(pp_def))
5bfa4862 584 self.add_post_processor(pp, when=when)
4f026faf 585
ab8e5e51
AM
586 for ph in self.params.get('post_hooks', []):
587 self.add_post_hook(ph)
588
71b640cc
PH
589 for ph in self.params.get('progress_hooks', []):
590 self.add_progress_hook(ph)
591
51fb4995
YCH
592 register_socks_protocols()
593
7d4111ed
PH
594 def warn_if_short_id(self, argv):
595 # short YouTube ID starting with dash?
596 idxs = [
597 i for i, a in enumerate(argv)
598 if re.match(r'^-[0-9A-Za-z_-]{10}$', a)]
599 if idxs:
600 correct_argv = (
7a5c1cfe 601 ['yt-dlp']
3089bc74
S
602 + [a for i, a in enumerate(argv) if i not in idxs]
603 + ['--'] + [argv[i] for i in idxs]
7d4111ed
PH
604 )
605 self.report_warning(
606 'Long argument string detected. '
607 'Use -- to separate parameters and URLs, like this:\n%s\n' %
608 args_to_str(correct_argv))
609
8222d8de
JMF
610 def add_info_extractor(self, ie):
611 """Add an InfoExtractor object to the end of the list."""
612 self._ies.append(ie)
e52d7f85
JMF
613 if not isinstance(ie, type):
614 self._ies_instances[ie.ie_key()] = ie
615 ie.set_downloader(self)
8222d8de 616
56c73665
JMF
617 def get_info_extractor(self, ie_key):
618 """
619 Get an instance of an IE with name ie_key, it will try to get one from
620 the _ies list, if there's no instance it will create a new one and add
621 it to the extractor list.
622 """
623 ie = self._ies_instances.get(ie_key)
624 if ie is None:
625 ie = get_info_extractor(ie_key)()
626 self.add_info_extractor(ie)
627 return ie
628
023fa8c4
JMF
629 def add_default_info_extractors(self):
630 """
631 Add the InfoExtractors returned by gen_extractors to the end of the list
632 """
e52d7f85 633 for ie in gen_extractor_classes():
023fa8c4
JMF
634 self.add_info_extractor(ie)
635
56d868db 636 def add_post_processor(self, pp, when='post_process'):
8222d8de 637 """Add a PostProcessor object to the end of the chain."""
5bfa4862 638 self._pps[when].append(pp)
8222d8de
JMF
639 pp.set_downloader(self)
640
ab8e5e51
AM
641 def add_post_hook(self, ph):
642 """Add the post hook"""
643 self._post_hooks.append(ph)
644
933605d7
JMF
645 def add_progress_hook(self, ph):
646 """Add the progress hook (currently only for the file downloader)"""
647 self._progress_hooks.append(ph)
8ab470f1 648
1c088fa8 649 def _bidi_workaround(self, message):
5d681e96 650 if not hasattr(self, '_output_channel'):
1c088fa8
PH
651 return message
652
5d681e96 653 assert hasattr(self, '_output_process')
11b85ce6 654 assert isinstance(message, compat_str)
6febd1c1
PH
655 line_count = message.count('\n') + 1
656 self._output_process.stdin.write((message + '\n').encode('utf-8'))
5d681e96 657 self._output_process.stdin.flush()
6febd1c1 658 res = ''.join(self._output_channel.readline().decode('utf-8')
9e1a5b84 659 for _ in range(line_count))
6febd1c1 660 return res[:-len('\n')]
1c088fa8 661
734f90bb 662 def _write_string(self, s, out=None):
b58ddb32 663 write_string(s, out=out, encoding=self.params.get('encoding'))
734f90bb 664
848887eb 665 def to_stdout(self, message, skip_eol=False, quiet=False):
0760b0a7 666 """Print message to stdout"""
8bf9319e 667 if self.params.get('logger'):
43afe285 668 self.params['logger'].debug(message)
835a1478 669 elif not quiet or self.params.get('verbose'):
670 self._write_string(
671 '%s%s' % (self._bidi_workaround(message), ('' if skip_eol else '\n')),
672 self._err_file if quiet else self._screen_file)
8222d8de
JMF
673
674 def to_stderr(self, message):
0760b0a7 675 """Print message to stderr"""
11b85ce6 676 assert isinstance(message, compat_str)
8bf9319e 677 if self.params.get('logger'):
43afe285
IB
678 self.params['logger'].error(message)
679 else:
835a1478 680 self._write_string('%s\n' % self._bidi_workaround(message), self._err_file)
8222d8de 681
1e5b9a95
PH
682 def to_console_title(self, message):
683 if not self.params.get('consoletitle', False):
684 return
4bede0d8
C
685 if compat_os_name == 'nt':
686 if ctypes.windll.kernel32.GetConsoleWindow():
687 # c_wchar_p() might not be necessary if `message` is
688 # already of type unicode()
689 ctypes.windll.kernel32.SetConsoleTitleW(ctypes.c_wchar_p(message))
1e5b9a95 690 elif 'TERM' in os.environ:
b46696bd 691 self._write_string('\033]0;%s\007' % message, self._screen_file)
1e5b9a95 692
bdde425c
PH
693 def save_console_title(self):
694 if not self.params.get('consoletitle', False):
695 return
94c3442e
S
696 if self.params.get('simulate', False):
697 return
4bede0d8 698 if compat_os_name != 'nt' and 'TERM' in os.environ:
efd6c574 699 # Save the title on stack
734f90bb 700 self._write_string('\033[22;0t', self._screen_file)
bdde425c
PH
701
702 def restore_console_title(self):
703 if not self.params.get('consoletitle', False):
704 return
94c3442e
S
705 if self.params.get('simulate', False):
706 return
4bede0d8 707 if compat_os_name != 'nt' and 'TERM' in os.environ:
efd6c574 708 # Restore the title from stack
734f90bb 709 self._write_string('\033[23;0t', self._screen_file)
bdde425c
PH
710
711 def __enter__(self):
712 self.save_console_title()
713 return self
714
715 def __exit__(self, *args):
716 self.restore_console_title()
f89197d7 717
dca08720 718 if self.params.get('cookiefile') is not None:
1bab3437 719 self.cookiejar.save(ignore_discard=True, ignore_expires=True)
bdde425c 720
8222d8de
JMF
721 def trouble(self, message=None, tb=None):
722 """Determine action to take when a download problem appears.
723
724 Depending on if the downloader has been configured to ignore
725 download errors or not, this method may throw an exception or
726 not when errors are found, after printing the message.
727
728 tb, if given, is additional traceback information.
729 """
730 if message is not None:
731 self.to_stderr(message)
732 if self.params.get('verbose'):
733 if tb is None:
734 if sys.exc_info()[0]: # if .trouble has been called from an except block
6febd1c1 735 tb = ''
8222d8de 736 if hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
6febd1c1 737 tb += ''.join(traceback.format_exception(*sys.exc_info()[1].exc_info))
c0384f22 738 tb += encode_compat_str(traceback.format_exc())
8222d8de
JMF
739 else:
740 tb_data = traceback.format_list(traceback.extract_stack())
6febd1c1 741 tb = ''.join(tb_data)
c19bc311 742 if tb:
743 self.to_stderr(tb)
8222d8de
JMF
744 if not self.params.get('ignoreerrors', False):
745 if sys.exc_info()[0] and hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
746 exc_info = sys.exc_info()[1].exc_info
747 else:
748 exc_info = sys.exc_info()
749 raise DownloadError(message, exc_info)
750 self._download_retcode = 1
751
0760b0a7 752 def to_screen(self, message, skip_eol=False):
753 """Print message to stdout if not in quiet mode"""
754 self.to_stdout(
755 message, skip_eol, quiet=self.params.get('quiet', False))
756
8222d8de
JMF
757 def report_warning(self, message):
758 '''
759 Print the message to stderr, it will be prefixed with 'WARNING:'
760 If stderr is a tty file the 'WARNING:' will be colored
761 '''
6d07ce01
JMF
762 if self.params.get('logger') is not None:
763 self.params['logger'].warning(message)
8222d8de 764 else:
ad8915b7
PH
765 if self.params.get('no_warnings'):
766 return
e9c0cdd3 767 if not self.params.get('no_color') and self._err_file.isatty() and compat_os_name != 'nt':
6d07ce01
JMF
768 _msg_header = '\033[0;33mWARNING:\033[0m'
769 else:
770 _msg_header = 'WARNING:'
771 warning_message = '%s %s' % (_msg_header, message)
772 self.to_stderr(warning_message)
8222d8de
JMF
773
774 def report_error(self, message, tb=None):
775 '''
776 Do the same as trouble, but prefixes the message with 'ERROR:', colored
777 in red if stderr is a tty file.
778 '''
e9c0cdd3 779 if not self.params.get('no_color') and self._err_file.isatty() and compat_os_name != 'nt':
6febd1c1 780 _msg_header = '\033[0;31mERROR:\033[0m'
8222d8de 781 else:
6febd1c1
PH
782 _msg_header = 'ERROR:'
783 error_message = '%s %s' % (_msg_header, message)
8222d8de
JMF
784 self.trouble(error_message, tb)
785
0760b0a7 786 def write_debug(self, message):
787 '''Log debug message or Print message to stderr'''
788 if not self.params.get('verbose', False):
789 return
790 message = '[debug] %s' % message
791 if self.params.get('logger'):
792 self.params['logger'].debug(message)
793 else:
794 self._write_string('%s\n' % message)
795
8222d8de
JMF
796 def report_file_already_downloaded(self, file_name):
797 """Report file has already been fully downloaded."""
798 try:
6febd1c1 799 self.to_screen('[download] %s has already been downloaded' % file_name)
ce02ed60 800 except UnicodeEncodeError:
6febd1c1 801 self.to_screen('[download] The file has already been downloaded')
8222d8de 802
0c3d0f51 803 def report_file_delete(self, file_name):
804 """Report that existing file will be deleted."""
805 try:
c25228e5 806 self.to_screen('Deleting existing file %s' % file_name)
0c3d0f51 807 except UnicodeEncodeError:
c25228e5 808 self.to_screen('Deleting existing file')
0c3d0f51 809
de6000d9 810 def parse_outtmpl(self):
811 outtmpl_dict = self.params.get('outtmpl', {})
812 if not isinstance(outtmpl_dict, dict):
813 outtmpl_dict = {'default': outtmpl_dict}
814 outtmpl_dict.update({
815 k: v for k, v in DEFAULT_OUTTMPL.items()
816 if not outtmpl_dict.get(k)})
817 for key, val in outtmpl_dict.items():
818 if isinstance(val, bytes):
819 self.report_warning(
820 'Parameter outtmpl is bytes, but should be a unicode string. '
821 'Put from __future__ import unicode_literals at the top of your code file or consider switching to Python 3.x.')
822 return outtmpl_dict
823
21cd8fae 824 def get_output_path(self, dir_type='', filename=None):
825 paths = self.params.get('paths', {})
826 assert isinstance(paths, dict)
827 path = os.path.join(
828 expand_path(paths.get('home', '').strip()),
829 expand_path(paths.get(dir_type, '').strip()) if dir_type else '',
830 filename or '')
831
832 # Temporary fix for #4787
833 # 'Treat' all problem characters by passing filename through preferredencoding
834 # to workaround encoding issues with subprocess on python2 @ Windows
835 if sys.version_info < (3, 0) and sys.platform == 'win32':
836 path = encodeFilename(path, True).decode(preferredencoding())
837 return sanitize_path(path, force=self.params.get('windowsfilenames'))
838
76a264ac 839 @staticmethod
840 def validate_outtmpl(tmpl):
841 ''' @return None or Exception object '''
842 try:
843 re.sub(
844 STR_FORMAT_RE.format(''),
845 lambda mobj: ('%' if not mobj.group('has_key') else '') + mobj.group(0),
846 tmpl
847 ) % collections.defaultdict(int)
848 return None
849 except ValueError as err:
850 return err
851
143db31d 852 def prepare_outtmpl(self, outtmpl, info_dict, sanitize=None):
853 """ Make the template and info_dict suitable for substitution (outtmpl % info_dict)"""
752cda38 854 info_dict = dict(info_dict)
a439a3a4 855 na = self.params.get('outtmpl_na_placeholder', 'NA')
143db31d 856
752cda38 857 info_dict['duration_string'] = ( # %(duration>%H-%M-%S)s is wrong if duration > 24hrs
53c18592 858 formatSeconds(info_dict['duration'], '-' if sanitize else ':')
143db31d 859 if info_dict.get('duration', None) is not None
860 else None)
752cda38 861 info_dict['epoch'] = int(time.time())
862 info_dict['autonumber'] = self.params.get('autonumber_start', 1) - 1 + self._num_downloads
863 if info_dict.get('resolution') is None:
864 info_dict['resolution'] = self.format_resolution(info_dict, default=None)
143db31d 865
143db31d 866 # For fields playlist_index and autonumber convert all occurrences
867 # of %(field)s to %(field)0Nd for backward compatibility
868 field_size_compat_map = {
752cda38 869 'playlist_index': len(str(info_dict.get('_last_playlist_index') or '')),
870 'autonumber': self.params.get('autonumber_size') or 5,
143db31d 871 }
752cda38 872
385a27fa 873 TMPL_DICT = {}
874 EXTERNAL_FORMAT_RE = re.compile(STR_FORMAT_RE.format('[^)]*'))
875 MATH_FUNCTIONS = {
876 '+': float.__add__,
877 '-': float.__sub__,
878 }
e625be0d 879 # Field is of the form key1.key2...
880 # where keys (except first) can be string, int or slice
385a27fa 881 FIELD_RE = r'\w+(?:\.(?:\w+|{num}|{num}?(?::{num}?){{1,2}}))*'.format(num=r'(?:-?\d+)')
882 MATH_FIELD_RE = r'''{field}|{num}'''.format(field=FIELD_RE, num=r'-?\d+(?:.\d+)?')
883 MATH_OPERATORS_RE = r'(?:%s)' % '|'.join(map(re.escape, MATH_FUNCTIONS.keys()))
e625be0d 884 INTERNAL_FORMAT_RE = re.compile(r'''(?x)
885 (?P<negate>-)?
385a27fa 886 (?P<fields>{field})
887 (?P<maths>(?:{math_op}{math_field})*)
e625be0d 888 (?:>(?P<strf_format>.+?))?
889 (?:\|(?P<default>.*?))?
385a27fa 890 $'''.format(field=FIELD_RE, math_op=MATH_OPERATORS_RE, math_field=MATH_FIELD_RE))
752cda38 891
76a264ac 892 get_key = lambda k: traverse_obj(
893 info_dict, k.split('.'), is_user_input=True, traverse_string=True)
894
752cda38 895 def get_value(mdict):
896 # Object traversal
76a264ac 897 value = get_key(mdict['fields'])
752cda38 898 # Negative
899 if mdict['negate']:
900 value = float_or_none(value)
901 if value is not None:
902 value *= -1
903 # Do maths
385a27fa 904 offset_key = mdict['maths']
905 if offset_key:
752cda38 906 value = float_or_none(value)
907 operator = None
385a27fa 908 while offset_key:
909 item = re.match(
910 MATH_FIELD_RE if operator else MATH_OPERATORS_RE,
911 offset_key).group(0)
912 offset_key = offset_key[len(item):]
913 if operator is None:
752cda38 914 operator = MATH_FUNCTIONS[item]
385a27fa 915 continue
916 item, multiplier = (item[1:], -1) if item[0] == '-' else (item, 1)
917 offset = float_or_none(item)
918 if offset is None:
919 offset = float_or_none(get_key(item))
920 try:
921 value = operator(value, multiplier * offset)
922 except (TypeError, ZeroDivisionError):
923 return None
924 operator = None
752cda38 925 # Datetime formatting
926 if mdict['strf_format']:
927 value = strftime_or_none(value, mdict['strf_format'])
928
929 return value
930
931 def create_key(outer_mobj):
932 if not outer_mobj.group('has_key'):
933 return '%{}'.format(outer_mobj.group(0))
934
935 key = outer_mobj.group('key')
936 fmt = outer_mobj.group('format')
937 mobj = re.match(INTERNAL_FORMAT_RE, key)
938 if mobj is None:
9fea350f 939 value, default, mobj = None, na, {'fields': ''}
752cda38 940 else:
e625be0d 941 mobj = mobj.groupdict()
752cda38 942 default = mobj['default'] if mobj['default'] is not None else na
943 value = get_value(mobj)
944
945 if fmt == 's' and value is not None and key in field_size_compat_map.keys():
946 fmt = '0{:d}d'.format(field_size_compat_map[key])
947
948 value = default if value is None else value
752cda38 949
76a264ac 950 if fmt == 'c':
951 value = compat_str(value)
952 if value is None:
953 value, fmt = default, 's'
954 else:
955 value = value[0]
956 elif fmt[-1] not in 'rs': # numeric
a439a3a4 957 value = float_or_none(value)
752cda38 958 if value is None:
959 value, fmt = default, 's'
960 if sanitize:
961 if fmt[-1] == 'r':
962 # If value is an object, sanitize might convert it to a string
963 # So we convert it to repr first
964 value, fmt = repr(value), '%ss' % fmt[:-1]
639f1cea 965 if fmt[-1] in 'csr':
9fea350f 966 value = sanitize(mobj['fields'].split('.')[-1], value)
967 key += '\0%s' % fmt
385a27fa 968 TMPL_DICT[key] = value
752cda38 969 return '%({key}){fmt}'.format(key=key, fmt=fmt)
970
385a27fa 971 return EXTERNAL_FORMAT_RE.sub(create_key, outtmpl), TMPL_DICT
143db31d 972
de6000d9 973 def _prepare_filename(self, info_dict, tmpl_type='default'):
8222d8de 974 try:
586a91b6 975 sanitize = lambda k, v: sanitize_filename(
45598aab 976 compat_str(v),
1bb5c511 977 restricted=self.params.get('restrictfilenames'),
40df485f 978 is_id=(k == 'id' or k.endswith('_id')))
de6000d9 979 outtmpl = self.outtmpl_dict.get(tmpl_type, self.outtmpl_dict['default'])
143db31d 980 outtmpl, template_dict = self.prepare_outtmpl(outtmpl, info_dict, sanitize)
d0d9ade4 981
15da37c7
S
982 # expand_path translates '%%' into '%' and '$$' into '$'
983 # correspondingly that is not what we want since we need to keep
984 # '%%' intact for template dict substitution step. Working around
985 # with boundary-alike separator hack.
961ea474 986 sep = ''.join([random.choice(ascii_letters) for _ in range(32)])
15da37c7
S
987 outtmpl = outtmpl.replace('%%', '%{0}%'.format(sep)).replace('$$', '${0}$'.format(sep))
988
989 # outtmpl should be expand_path'ed before template dict substitution
990 # because meta fields may contain env variables we don't want to
991 # be expanded. For example, for outtmpl "%(title)s.%(ext)s" and
992 # title "Hello $PATH", we don't want `$PATH` to be expanded.
993 filename = expand_path(outtmpl).replace(sep, '') % template_dict
994
143db31d 995 force_ext = OUTTMPL_TYPES.get(tmpl_type)
de6000d9 996 if force_ext is not None:
752cda38 997 filename = replace_extension(filename, force_ext, info_dict.get('ext'))
de6000d9 998
bdc3fd2f
U
999 # https://github.com/blackjack4494/youtube-dlc/issues/85
1000 trim_file_name = self.params.get('trim_file_name', False)
1001 if trim_file_name:
1002 fn_groups = filename.rsplit('.')
1003 ext = fn_groups[-1]
1004 sub_ext = ''
1005 if len(fn_groups) > 2:
1006 sub_ext = fn_groups[-2]
1007 filename = '.'.join(filter(None, [fn_groups[0][:trim_file_name], sub_ext, ext]))
1008
0202b52a 1009 return filename
8222d8de 1010 except ValueError as err:
6febd1c1 1011 self.report_error('Error in output template: ' + str(err) + ' (encoding: ' + repr(preferredencoding()) + ')')
8222d8de
JMF
1012 return None
1013
de6000d9 1014 def prepare_filename(self, info_dict, dir_type='', warn=False):
1015 """Generate the output filename."""
21cd8fae 1016
de6000d9 1017 filename = self._prepare_filename(info_dict, dir_type or 'default')
1018
1019 if warn and not self.__prepare_filename_warned:
21cd8fae 1020 if not self.params.get('paths'):
de6000d9 1021 pass
1022 elif filename == '-':
1023 self.report_warning('--paths is ignored when an outputting to stdout')
1024 elif os.path.isabs(filename):
1025 self.report_warning('--paths is ignored since an absolute path is given in output template')
1026 self.__prepare_filename_warned = True
1027 if filename == '-' or not filename:
1028 return filename
1029
21cd8fae 1030 return self.get_output_path(dir_type, filename)
0202b52a 1031
120fe513 1032 def _match_entry(self, info_dict, incomplete=False, silent=False):
ecdec191 1033 """ Returns None if the file should be downloaded """
8222d8de 1034
c77495e3 1035 video_title = info_dict.get('title', info_dict.get('id', 'video'))
1036
8b0d7497 1037 def check_filter():
8b0d7497 1038 if 'title' in info_dict:
1039 # This can happen when we're just evaluating the playlist
1040 title = info_dict['title']
1041 matchtitle = self.params.get('matchtitle', False)
1042 if matchtitle:
1043 if not re.search(matchtitle, title, re.IGNORECASE):
1044 return '"' + title + '" title did not match pattern "' + matchtitle + '"'
1045 rejecttitle = self.params.get('rejecttitle', False)
1046 if rejecttitle:
1047 if re.search(rejecttitle, title, re.IGNORECASE):
1048 return '"' + title + '" title matched reject pattern "' + rejecttitle + '"'
1049 date = info_dict.get('upload_date')
1050 if date is not None:
1051 dateRange = self.params.get('daterange', DateRange())
1052 if date not in dateRange:
1053 return '%s upload date is not in range %s' % (date_from_str(date).isoformat(), dateRange)
1054 view_count = info_dict.get('view_count')
1055 if view_count is not None:
1056 min_views = self.params.get('min_views')
1057 if min_views is not None and view_count < min_views:
1058 return 'Skipping %s, because it has not reached minimum view count (%d/%d)' % (video_title, view_count, min_views)
1059 max_views = self.params.get('max_views')
1060 if max_views is not None and view_count > max_views:
1061 return 'Skipping %s, because it has exceeded the maximum view count (%d/%d)' % (video_title, view_count, max_views)
1062 if age_restricted(info_dict.get('age_limit'), self.params.get('age_limit')):
1063 return 'Skipping "%s" because it is age restricted' % video_title
8b0d7497 1064
1065 if not incomplete:
1066 match_filter = self.params.get('match_filter')
1067 if match_filter is not None:
1068 ret = match_filter(info_dict)
1069 if ret is not None:
1070 return ret
1071 return None
1072
c77495e3 1073 if self.in_download_archive(info_dict):
1074 reason = '%s has already been recorded in the archive' % video_title
1075 break_opt, break_err = 'break_on_existing', ExistingVideoReached
1076 else:
1077 reason = check_filter()
1078 break_opt, break_err = 'break_on_reject', RejectedVideoReached
8b0d7497 1079 if reason is not None:
120fe513 1080 if not silent:
1081 self.to_screen('[download] ' + reason)
c77495e3 1082 if self.params.get(break_opt, False):
1083 raise break_err()
8b0d7497 1084 return reason
fe7e0c98 1085
b6c45014
JMF
1086 @staticmethod
1087 def add_extra_info(info_dict, extra_info):
1088 '''Set the keys from extra_info in info dict if they are missing'''
1089 for key, value in extra_info.items():
1090 info_dict.setdefault(key, value)
1091
58f197b7 1092 def extract_info(self, url, download=True, ie_key=None, extra_info={},
61aa5ba3 1093 process=True, force_generic_extractor=False):
41d1cca3 1094 """
1095 Return a list with a dictionary for each video extracted.
1096
1097 Arguments:
1098 url -- URL to extract
1099
1100 Keyword arguments:
1101 download -- whether to download videos during extraction
1102 ie_key -- extractor key hint
1103 extra_info -- dictionary containing the extra values to add to each result
1104 process -- whether to resolve all unresolved references (URLs, playlist items),
1105 must be True for download to work.
1106 force_generic_extractor -- force using the generic extractor
1107 """
fe7e0c98 1108
61aa5ba3 1109 if not ie_key and force_generic_extractor:
d22dec74
S
1110 ie_key = 'Generic'
1111
8222d8de 1112 if ie_key:
56c73665 1113 ies = [self.get_info_extractor(ie_key)]
8222d8de
JMF
1114 else:
1115 ies = self._ies
1116
1117 for ie in ies:
1118 if not ie.suitable(url):
1119 continue
1120
9a68de12 1121 ie_key = ie.ie_key()
1122 ie = self.get_info_extractor(ie_key)
8222d8de 1123 if not ie.working():
6febd1c1
PH
1124 self.report_warning('The program functionality for this site has been marked as broken, '
1125 'and will probably not work.')
8222d8de
JMF
1126
1127 try:
d0757229 1128 temp_id = str_or_none(
63be1aab 1129 ie.extract_id(url) if callable(getattr(ie, 'extract_id', None))
1130 else ie._match_id(url))
a0566bbf 1131 except (AssertionError, IndexError, AttributeError):
1132 temp_id = None
1133 if temp_id is not None and self.in_download_archive({'id': temp_id, 'ie_key': ie_key}):
1134 self.to_screen("[%s] %s: has already been recorded in archive" % (
1135 ie_key, temp_id))
1136 break
58f197b7 1137 return self.__extract_info(url, ie, download, extra_info, process)
a0566bbf 1138 else:
1139 self.report_error('no suitable InfoExtractor for URL %s' % url)
1140
1141 def __handle_extraction_exceptions(func):
1142 def wrapper(self, *args, **kwargs):
1143 try:
1144 return func(self, *args, **kwargs)
773f291d
S
1145 except GeoRestrictedError as e:
1146 msg = e.msg
1147 if e.countries:
1148 msg += '\nThis video is available in %s.' % ', '.join(
1149 map(ISO3166Utils.short2full, e.countries))
1150 msg += '\nYou might want to use a VPN or a proxy server (with --proxy) to workaround.'
1151 self.report_error(msg)
fb043a6e 1152 except ExtractorError as e: # An error we somewhat expected
2c74e6fa 1153 self.report_error(compat_str(e), e.format_traceback())
51d9739f 1154 except ThrottledDownload:
1155 self.to_stderr('\r')
1156 self.report_warning('The download speed is below throttle limit. Re-extracting data')
1157 return wrapper(self, *args, **kwargs)
8b0d7497 1158 except (MaxDownloadsReached, ExistingVideoReached, RejectedVideoReached):
d3e5bbf4 1159 raise
8222d8de
JMF
1160 except Exception as e:
1161 if self.params.get('ignoreerrors', False):
9b9c5355 1162 self.report_error(error_to_compat_str(e), tb=encode_compat_str(traceback.format_exc()))
8222d8de
JMF
1163 else:
1164 raise
a0566bbf 1165 return wrapper
1166
1167 @__handle_extraction_exceptions
58f197b7 1168 def __extract_info(self, url, ie, download, extra_info, process):
a0566bbf 1169 ie_result = ie.extract(url)
1170 if ie_result is None: # Finished already (backwards compatibility; listformats and friends should be moved here)
1171 return
1172 if isinstance(ie_result, list):
1173 # Backwards compatibility: old IE result format
1174 ie_result = {
1175 '_type': 'compat_list',
1176 'entries': ie_result,
1177 }
a0566bbf 1178 self.add_default_extra_info(ie_result, ie, url)
1179 if process:
1180 return self.process_ie_result(ie_result, download, extra_info)
8222d8de 1181 else:
a0566bbf 1182 return ie_result
fe7e0c98 1183
ea38e55f 1184 def add_default_extra_info(self, ie_result, ie, url):
6033d980 1185 if url is not None:
1186 self.add_extra_info(ie_result, {
1187 'webpage_url': url,
1188 'original_url': url,
1189 'webpage_url_basename': url_basename(url),
1190 })
1191 if ie is not None:
1192 self.add_extra_info(ie_result, {
1193 'extractor': ie.IE_NAME,
1194 'extractor_key': ie.ie_key(),
1195 })
ea38e55f 1196
8222d8de
JMF
1197 def process_ie_result(self, ie_result, download=True, extra_info={}):
1198 """
1199 Take the result of the ie(may be modified) and resolve all unresolved
1200 references (URLs, playlist items).
1201
1202 It will also download the videos if 'download'.
1203 Returns the resolved ie_result.
1204 """
e8ee972c
PH
1205 result_type = ie_result.get('_type', 'video')
1206
057a5206 1207 if result_type in ('url', 'url_transparent'):
134c6ea8 1208 ie_result['url'] = sanitize_url(ie_result['url'])
057a5206 1209 extract_flat = self.params.get('extract_flat', False)
3089bc74
S
1210 if ((extract_flat == 'in_playlist' and 'playlist' in extra_info)
1211 or extract_flat is True):
ecb54191 1212 info_copy = ie_result.copy()
1213 self.add_extra_info(info_copy, extra_info)
6033d980 1214 ie = try_get(ie_result.get('ie_key'), self.get_info_extractor)
1215 self.add_default_extra_info(info_copy, ie, ie_result['url'])
ecb54191 1216 self.__forced_printings(info_copy, self.prepare_filename(info_copy), incomplete=True)
e8ee972c
PH
1217 return ie_result
1218
8222d8de 1219 if result_type == 'video':
b6c45014 1220 self.add_extra_info(ie_result, extra_info)
9c2b75b5 1221 ie_result = self.process_video_result(ie_result, download=download)
28b0eb0f 1222 additional_urls = (ie_result or {}).get('additional_urls')
9c2b75b5 1223 if additional_urls:
1224 # TODO: Improve MetadataFromFieldPP to allow setting a list
1225 if isinstance(additional_urls, compat_str):
1226 additional_urls = [additional_urls]
1227 self.to_screen(
1228 '[info] %s: %d additional URL(s) requested' % (ie_result['id'], len(additional_urls)))
1229 self.write_debug('Additional URLs: "%s"' % '", "'.join(additional_urls))
1230 ie_result['additional_entries'] = [
1231 self.extract_info(
1232 url, download, extra_info,
1233 force_generic_extractor=self.params.get('force_generic_extractor'))
1234 for url in additional_urls
1235 ]
1236 return ie_result
8222d8de
JMF
1237 elif result_type == 'url':
1238 # We have to add extra_info to the results because it may be
1239 # contained in a playlist
07cce701 1240 return self.extract_info(
1241 ie_result['url'], download,
1242 ie_key=ie_result.get('ie_key'),
1243 extra_info=extra_info)
7fc3fa05
PH
1244 elif result_type == 'url_transparent':
1245 # Use the information from the embedding page
1246 info = self.extract_info(
1247 ie_result['url'], ie_key=ie_result.get('ie_key'),
1248 extra_info=extra_info, download=False, process=False)
1249
1640eb09
S
1250 # extract_info may return None when ignoreerrors is enabled and
1251 # extraction failed with an error, don't crash and return early
1252 # in this case
1253 if not info:
1254 return info
1255
412c617d
PH
1256 force_properties = dict(
1257 (k, v) for k, v in ie_result.items() if v is not None)
0396806f 1258 for f in ('_type', 'url', 'id', 'extractor', 'extractor_key', 'ie_key'):
412c617d
PH
1259 if f in force_properties:
1260 del force_properties[f]
1261 new_result = info.copy()
1262 new_result.update(force_properties)
7fc3fa05 1263
0563f7ac
S
1264 # Extracted info may not be a video result (i.e.
1265 # info.get('_type', 'video') != video) but rather an url or
1266 # url_transparent. In such cases outer metadata (from ie_result)
1267 # should be propagated to inner one (info). For this to happen
1268 # _type of info should be overridden with url_transparent. This
067aa17e 1269 # fixes issue from https://github.com/ytdl-org/youtube-dl/pull/11163.
0563f7ac
S
1270 if new_result.get('_type') == 'url':
1271 new_result['_type'] = 'url_transparent'
7fc3fa05
PH
1272
1273 return self.process_ie_result(
1274 new_result, download=download, extra_info=extra_info)
40fcba5e 1275 elif result_type in ('playlist', 'multi_video'):
30a074c2 1276 # Protect from infinite recursion due to recursively nested playlists
1277 # (see https://github.com/ytdl-org/youtube-dl/issues/27833)
1278 webpage_url = ie_result['webpage_url']
1279 if webpage_url in self._playlist_urls:
7e85e872 1280 self.to_screen(
30a074c2 1281 '[download] Skipping already downloaded playlist: %s'
1282 % ie_result.get('title') or ie_result.get('id'))
1283 return
7e85e872 1284
30a074c2 1285 self._playlist_level += 1
1286 self._playlist_urls.add(webpage_url)
bc516a3f 1287 self._sanitize_thumbnails(ie_result)
30a074c2 1288 try:
1289 return self.__process_playlist(ie_result, download)
1290 finally:
1291 self._playlist_level -= 1
1292 if not self._playlist_level:
1293 self._playlist_urls.clear()
8222d8de 1294 elif result_type == 'compat_list':
c9bf4114
PH
1295 self.report_warning(
1296 'Extractor %s returned a compat_list result. '
1297 'It needs to be updated.' % ie_result.get('extractor'))
5f6a1245 1298
8222d8de 1299 def _fixup(r):
9e1a5b84
JW
1300 self.add_extra_info(
1301 r,
9103bbc5
JMF
1302 {
1303 'extractor': ie_result['extractor'],
1304 'webpage_url': ie_result['webpage_url'],
29eb5174 1305 'webpage_url_basename': url_basename(ie_result['webpage_url']),
be97abc2 1306 'extractor_key': ie_result['extractor_key'],
9e1a5b84
JW
1307 }
1308 )
8222d8de
JMF
1309 return r
1310 ie_result['entries'] = [
b6c45014 1311 self.process_ie_result(_fixup(r), download, extra_info)
8222d8de
JMF
1312 for r in ie_result['entries']
1313 ]
1314 return ie_result
1315 else:
1316 raise Exception('Invalid result type: %s' % result_type)
1317
e92caff5 1318 def _ensure_dir_exists(self, path):
1319 return make_dir(path, self.report_error)
1320
30a074c2 1321 def __process_playlist(self, ie_result, download):
1322 # We process each entry in the playlist
1323 playlist = ie_result.get('title') or ie_result.get('id')
1324 self.to_screen('[download] Downloading playlist: %s' % playlist)
1325
498f5606 1326 if 'entries' not in ie_result:
1327 raise EntryNotInPlaylist()
1328 incomplete_entries = bool(ie_result.get('requested_entries'))
1329 if incomplete_entries:
1330 def fill_missing_entries(entries, indexes):
1331 ret = [None] * max(*indexes)
1332 for i, entry in zip(indexes, entries):
1333 ret[i - 1] = entry
1334 return ret
1335 ie_result['entries'] = fill_missing_entries(ie_result['entries'], ie_result['requested_entries'])
02fd60d3 1336
30a074c2 1337 playlist_results = []
1338
56a8fb4f 1339 playliststart = self.params.get('playliststart', 1)
30a074c2 1340 playlistend = self.params.get('playlistend')
1341 # For backwards compatibility, interpret -1 as whole list
1342 if playlistend == -1:
1343 playlistend = None
1344
1345 playlistitems_str = self.params.get('playlist_items')
1346 playlistitems = None
1347 if playlistitems_str is not None:
1348 def iter_playlistitems(format):
1349 for string_segment in format.split(','):
1350 if '-' in string_segment:
1351 start, end = string_segment.split('-')
1352 for item in range(int(start), int(end) + 1):
1353 yield int(item)
1354 else:
1355 yield int(string_segment)
1356 playlistitems = orderedSet(iter_playlistitems(playlistitems_str))
1357
1358 ie_entries = ie_result['entries']
56a8fb4f 1359 msg = (
1360 'Downloading %d videos' if not isinstance(ie_entries, list)
1361 else 'Collected %d videos; downloading %%d of them' % len(ie_entries))
1362 if not isinstance(ie_entries, (list, PagedList)):
1363 ie_entries = LazyList(ie_entries)
1364
1365 entries = []
1366 for i in playlistitems or itertools.count(playliststart):
1367 if playlistitems is None and playlistend is not None and playlistend < i:
1368 break
1369 entry = None
1370 try:
1371 entry = ie_entries[i - 1]
1372 if entry is None:
498f5606 1373 raise EntryNotInPlaylist()
56a8fb4f 1374 except (IndexError, EntryNotInPlaylist):
1375 if incomplete_entries:
1376 raise EntryNotInPlaylist()
1377 elif not playlistitems:
1378 break
1379 entries.append(entry)
120fe513 1380 try:
1381 if entry is not None:
1382 self._match_entry(entry, incomplete=True, silent=True)
1383 except (ExistingVideoReached, RejectedVideoReached):
1384 break
56a8fb4f 1385 ie_result['entries'] = entries
30a074c2 1386
56a8fb4f 1387 # Save playlist_index before re-ordering
1388 entries = [
1389 ((playlistitems[i - 1] if playlistitems else i), entry)
1390 for i, entry in enumerate(entries, 1)
1391 if entry is not None]
1392 n_entries = len(entries)
498f5606 1393
498f5606 1394 if not playlistitems and (playliststart or playlistend):
56a8fb4f 1395 playlistitems = list(range(playliststart, playliststart + n_entries))
498f5606 1396 ie_result['requested_entries'] = playlistitems
1397
1398 if self.params.get('allow_playlist_files', True):
1399 ie_copy = {
1400 'playlist': playlist,
1401 'playlist_id': ie_result.get('id'),
1402 'playlist_title': ie_result.get('title'),
1403 'playlist_uploader': ie_result.get('uploader'),
1404 'playlist_uploader_id': ie_result.get('uploader_id'),
71729754 1405 'playlist_index': 0,
498f5606 1406 }
1407 ie_copy.update(dict(ie_result))
1408
1409 if self.params.get('writeinfojson', False):
1410 infofn = self.prepare_filename(ie_copy, 'pl_infojson')
1411 if not self._ensure_dir_exists(encodeFilename(infofn)):
1412 return
1413 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(infofn)):
1414 self.to_screen('[info] Playlist metadata is already present')
1415 else:
1416 self.to_screen('[info] Writing playlist metadata as JSON to: ' + infofn)
1417 try:
1418 write_json_file(self.filter_requested_info(ie_result, self.params.get('clean_infojson', True)), infofn)
1419 except (OSError, IOError):
1420 self.report_error('Cannot write playlist metadata to JSON file ' + infofn)
1421
681de68e 1422 # TODO: This should be passed to ThumbnailsConvertor if necessary
1423 self._write_thumbnails(ie_copy, self.prepare_filename(ie_copy, 'pl_thumbnail'))
1424
498f5606 1425 if self.params.get('writedescription', False):
1426 descfn = self.prepare_filename(ie_copy, 'pl_description')
1427 if not self._ensure_dir_exists(encodeFilename(descfn)):
1428 return
1429 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(descfn)):
1430 self.to_screen('[info] Playlist description is already present')
1431 elif ie_result.get('description') is None:
1432 self.report_warning('There\'s no playlist description to write.')
1433 else:
1434 try:
1435 self.to_screen('[info] Writing playlist description to: ' + descfn)
1436 with io.open(encodeFilename(descfn), 'w', encoding='utf-8') as descfile:
1437 descfile.write(ie_result['description'])
1438 except (OSError, IOError):
1439 self.report_error('Cannot write playlist description file ' + descfn)
1440 return
30a074c2 1441
1442 if self.params.get('playlistreverse', False):
1443 entries = entries[::-1]
30a074c2 1444 if self.params.get('playlistrandom', False):
1445 random.shuffle(entries)
1446
1447 x_forwarded_for = ie_result.get('__x_forwarded_for_ip')
1448
56a8fb4f 1449 self.to_screen('[%s] playlist %s: %s' % (ie_result['extractor'], playlist, msg % n_entries))
26e2805c 1450 failures = 0
1451 max_failures = self.params.get('skip_playlist_after_errors') or float('inf')
71729754 1452 for i, entry_tuple in enumerate(entries, 1):
1453 playlist_index, entry = entry_tuple
53ed7066 1454 if 'playlist_index' in self.params.get('compat_options', []):
1455 playlist_index = playlistitems[i - 1] if playlistitems else i
30a074c2 1456 self.to_screen('[download] Downloading video %s of %s' % (i, n_entries))
1457 # This __x_forwarded_for_ip thing is a bit ugly but requires
1458 # minimal changes
1459 if x_forwarded_for:
1460 entry['__x_forwarded_for_ip'] = x_forwarded_for
1461 extra = {
1462 'n_entries': n_entries,
f59ae581 1463 '_last_playlist_index': max(playlistitems) if playlistitems else (playlistend or n_entries),
71729754 1464 'playlist_index': playlist_index,
1465 'playlist_autonumber': i,
30a074c2 1466 'playlist': playlist,
1467 'playlist_id': ie_result.get('id'),
1468 'playlist_title': ie_result.get('title'),
1469 'playlist_uploader': ie_result.get('uploader'),
1470 'playlist_uploader_id': ie_result.get('uploader_id'),
30a074c2 1471 'extractor': ie_result['extractor'],
1472 'webpage_url': ie_result['webpage_url'],
1473 'webpage_url_basename': url_basename(ie_result['webpage_url']),
1474 'extractor_key': ie_result['extractor_key'],
1475 }
1476
1477 if self._match_entry(entry, incomplete=True) is not None:
1478 continue
1479
1480 entry_result = self.__process_iterable_entry(entry, download, extra)
26e2805c 1481 if not entry_result:
1482 failures += 1
1483 if failures >= max_failures:
1484 self.report_error(
1485 'Skipping the remaining entries in playlist "%s" since %d items failed extraction' % (playlist, failures))
1486 break
30a074c2 1487 # TODO: skip failed (empty) entries?
1488 playlist_results.append(entry_result)
1489 ie_result['entries'] = playlist_results
1490 self.to_screen('[download] Finished downloading playlist: %s' % playlist)
1491 return ie_result
1492
a0566bbf 1493 @__handle_extraction_exceptions
1494 def __process_iterable_entry(self, entry, download, extra_info):
1495 return self.process_ie_result(
1496 entry, download=download, extra_info=extra_info)
1497
67134eab
JMF
1498 def _build_format_filter(self, filter_spec):
1499 " Returns a function to filter the formats according to the filter_spec "
083c9df9
PH
1500
1501 OPERATORS = {
1502 '<': operator.lt,
1503 '<=': operator.le,
1504 '>': operator.gt,
1505 '>=': operator.ge,
1506 '=': operator.eq,
1507 '!=': operator.ne,
1508 }
67134eab 1509 operator_rex = re.compile(r'''(?x)\s*
187986a8 1510 (?P<key>width|height|tbr|abr|vbr|asr|filesize|filesize_approx|fps)\s*
1511 (?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
1512 (?P<value>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?)\s*
083c9df9 1513 ''' % '|'.join(map(re.escape, OPERATORS.keys())))
187986a8 1514 m = operator_rex.fullmatch(filter_spec)
9ddb6925
S
1515 if m:
1516 try:
1517 comparison_value = int(m.group('value'))
1518 except ValueError:
1519 comparison_value = parse_filesize(m.group('value'))
1520 if comparison_value is None:
1521 comparison_value = parse_filesize(m.group('value') + 'B')
1522 if comparison_value is None:
1523 raise ValueError(
1524 'Invalid value %r in format specification %r' % (
67134eab 1525 m.group('value'), filter_spec))
9ddb6925
S
1526 op = OPERATORS[m.group('op')]
1527
083c9df9 1528 if not m:
9ddb6925
S
1529 STR_OPERATORS = {
1530 '=': operator.eq,
10d33b34
YCH
1531 '^=': lambda attr, value: attr.startswith(value),
1532 '$=': lambda attr, value: attr.endswith(value),
1533 '*=': lambda attr, value: value in attr,
9ddb6925 1534 }
187986a8 1535 str_operator_rex = re.compile(r'''(?x)\s*
1536 (?P<key>[a-zA-Z0-9._-]+)\s*
1537 (?P<negation>!\s*)?(?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
1538 (?P<value>[a-zA-Z0-9._-]+)\s*
9ddb6925 1539 ''' % '|'.join(map(re.escape, STR_OPERATORS.keys())))
187986a8 1540 m = str_operator_rex.fullmatch(filter_spec)
9ddb6925
S
1541 if m:
1542 comparison_value = m.group('value')
2cc779f4
S
1543 str_op = STR_OPERATORS[m.group('op')]
1544 if m.group('negation'):
e118a879 1545 op = lambda attr, value: not str_op(attr, value)
2cc779f4
S
1546 else:
1547 op = str_op
083c9df9 1548
9ddb6925 1549 if not m:
187986a8 1550 raise SyntaxError('Invalid filter specification %r' % filter_spec)
083c9df9
PH
1551
1552 def _filter(f):
1553 actual_value = f.get(m.group('key'))
1554 if actual_value is None:
1555 return m.group('none_inclusive')
1556 return op(actual_value, comparison_value)
67134eab
JMF
1557 return _filter
1558
0017d9ad 1559 def _default_format_spec(self, info_dict, download=True):
0017d9ad 1560
af0f7428
S
1561 def can_merge():
1562 merger = FFmpegMergerPP(self)
1563 return merger.available and merger.can_merge()
1564
91ebc640 1565 prefer_best = (
1566 not self.params.get('simulate', False)
1567 and download
1568 and (
1569 not can_merge()
19807826 1570 or info_dict.get('is_live', False)
de6000d9 1571 or self.outtmpl_dict['default'] == '-'))
53ed7066 1572 compat = (
1573 prefer_best
1574 or self.params.get('allow_multiple_audio_streams', False)
1575 or 'format-spec' in self.params.get('compat_opts', []))
91ebc640 1576
1577 return (
53ed7066 1578 'best/bestvideo+bestaudio' if prefer_best
1579 else 'bestvideo*+bestaudio/best' if not compat
91ebc640 1580 else 'bestvideo+bestaudio/best')
0017d9ad 1581
67134eab
JMF
1582 def build_format_selector(self, format_spec):
1583 def syntax_error(note, start):
1584 message = (
1585 'Invalid format specification: '
1586 '{0}\n\t{1}\n\t{2}^'.format(note, format_spec, ' ' * start[1]))
1587 return SyntaxError(message)
1588
1589 PICKFIRST = 'PICKFIRST'
1590 MERGE = 'MERGE'
1591 SINGLE = 'SINGLE'
0130afb7 1592 GROUP = 'GROUP'
67134eab
JMF
1593 FormatSelector = collections.namedtuple('FormatSelector', ['type', 'selector', 'filters'])
1594
91ebc640 1595 allow_multiple_streams = {'audio': self.params.get('allow_multiple_audio_streams', False),
1596 'video': self.params.get('allow_multiple_video_streams', False)}
909d24dd 1597
e8e73840 1598 check_formats = self.params.get('check_formats')
1599
67134eab
JMF
1600 def _parse_filter(tokens):
1601 filter_parts = []
1602 for type, string, start, _, _ in tokens:
1603 if type == tokenize.OP and string == ']':
1604 return ''.join(filter_parts)
1605 else:
1606 filter_parts.append(string)
1607
232541df 1608 def _remove_unused_ops(tokens):
17cc1534 1609 # Remove operators that we don't use and join them with the surrounding strings
232541df
JMF
1610 # for example: 'mp4' '-' 'baseline' '-' '16x9' is converted to 'mp4-baseline-16x9'
1611 ALLOWED_OPS = ('/', '+', ',', '(', ')')
1612 last_string, last_start, last_end, last_line = None, None, None, None
1613 for type, string, start, end, line in tokens:
1614 if type == tokenize.OP and string == '[':
1615 if last_string:
1616 yield tokenize.NAME, last_string, last_start, last_end, last_line
1617 last_string = None
1618 yield type, string, start, end, line
1619 # everything inside brackets will be handled by _parse_filter
1620 for type, string, start, end, line in tokens:
1621 yield type, string, start, end, line
1622 if type == tokenize.OP and string == ']':
1623 break
1624 elif type == tokenize.OP and string in ALLOWED_OPS:
1625 if last_string:
1626 yield tokenize.NAME, last_string, last_start, last_end, last_line
1627 last_string = None
1628 yield type, string, start, end, line
1629 elif type in [tokenize.NAME, tokenize.NUMBER, tokenize.OP]:
1630 if not last_string:
1631 last_string = string
1632 last_start = start
1633 last_end = end
1634 else:
1635 last_string += string
1636 if last_string:
1637 yield tokenize.NAME, last_string, last_start, last_end, last_line
1638
cf2ac6df 1639 def _parse_format_selection(tokens, inside_merge=False, inside_choice=False, inside_group=False):
67134eab
JMF
1640 selectors = []
1641 current_selector = None
1642 for type, string, start, _, _ in tokens:
1643 # ENCODING is only defined in python 3.x
1644 if type == getattr(tokenize, 'ENCODING', None):
1645 continue
1646 elif type in [tokenize.NAME, tokenize.NUMBER]:
1647 current_selector = FormatSelector(SINGLE, string, [])
1648 elif type == tokenize.OP:
cf2ac6df
JMF
1649 if string == ')':
1650 if not inside_group:
1651 # ')' will be handled by the parentheses group
1652 tokens.restore_last_token()
67134eab 1653 break
cf2ac6df 1654 elif inside_merge and string in ['/', ',']:
0130afb7
JMF
1655 tokens.restore_last_token()
1656 break
cf2ac6df
JMF
1657 elif inside_choice and string == ',':
1658 tokens.restore_last_token()
1659 break
1660 elif string == ',':
0a31a350
JMF
1661 if not current_selector:
1662 raise syntax_error('"," must follow a format selector', start)
67134eab
JMF
1663 selectors.append(current_selector)
1664 current_selector = None
1665 elif string == '/':
d96d604e
JMF
1666 if not current_selector:
1667 raise syntax_error('"/" must follow a format selector', start)
67134eab 1668 first_choice = current_selector
cf2ac6df 1669 second_choice = _parse_format_selection(tokens, inside_choice=True)
f5f4a27a 1670 current_selector = FormatSelector(PICKFIRST, (first_choice, second_choice), [])
67134eab
JMF
1671 elif string == '[':
1672 if not current_selector:
1673 current_selector = FormatSelector(SINGLE, 'best', [])
1674 format_filter = _parse_filter(tokens)
1675 current_selector.filters.append(format_filter)
0130afb7
JMF
1676 elif string == '(':
1677 if current_selector:
1678 raise syntax_error('Unexpected "("', start)
cf2ac6df
JMF
1679 group = _parse_format_selection(tokens, inside_group=True)
1680 current_selector = FormatSelector(GROUP, group, [])
67134eab 1681 elif string == '+':
d03cfdce 1682 if not current_selector:
1683 raise syntax_error('Unexpected "+"', start)
1684 selector_1 = current_selector
1685 selector_2 = _parse_format_selection(tokens, inside_merge=True)
1686 if not selector_2:
1687 raise syntax_error('Expected a selector', start)
1688 current_selector = FormatSelector(MERGE, (selector_1, selector_2), [])
67134eab
JMF
1689 else:
1690 raise syntax_error('Operator not recognized: "{0}"'.format(string), start)
1691 elif type == tokenize.ENDMARKER:
1692 break
1693 if current_selector:
1694 selectors.append(current_selector)
1695 return selectors
1696
f8d4ad9a 1697 def _merge(formats_pair):
1698 format_1, format_2 = formats_pair
1699
1700 formats_info = []
1701 formats_info.extend(format_1.get('requested_formats', (format_1,)))
1702 formats_info.extend(format_2.get('requested_formats', (format_2,)))
1703
1704 if not allow_multiple_streams['video'] or not allow_multiple_streams['audio']:
551f9388 1705 get_no_more = {'video': False, 'audio': False}
f8d4ad9a 1706 for (i, fmt_info) in enumerate(formats_info):
551f9388 1707 if fmt_info.get('acodec') == fmt_info.get('vcodec') == 'none':
1708 formats_info.pop(i)
1709 continue
1710 for aud_vid in ['audio', 'video']:
f8d4ad9a 1711 if not allow_multiple_streams[aud_vid] and fmt_info.get(aud_vid[0] + 'codec') != 'none':
1712 if get_no_more[aud_vid]:
1713 formats_info.pop(i)
1714 get_no_more[aud_vid] = True
1715
1716 if len(formats_info) == 1:
1717 return formats_info[0]
1718
1719 video_fmts = [fmt_info for fmt_info in formats_info if fmt_info.get('vcodec') != 'none']
1720 audio_fmts = [fmt_info for fmt_info in formats_info if fmt_info.get('acodec') != 'none']
1721
1722 the_only_video = video_fmts[0] if len(video_fmts) == 1 else None
1723 the_only_audio = audio_fmts[0] if len(audio_fmts) == 1 else None
1724
1725 output_ext = self.params.get('merge_output_format')
1726 if not output_ext:
1727 if the_only_video:
1728 output_ext = the_only_video['ext']
1729 elif the_only_audio and not video_fmts:
1730 output_ext = the_only_audio['ext']
1731 else:
1732 output_ext = 'mkv'
1733
1734 new_dict = {
1735 'requested_formats': formats_info,
1736 'format': '+'.join(fmt_info.get('format') for fmt_info in formats_info),
1737 'format_id': '+'.join(fmt_info.get('format_id') for fmt_info in formats_info),
1738 'ext': output_ext,
1739 }
1740
1741 if the_only_video:
1742 new_dict.update({
1743 'width': the_only_video.get('width'),
1744 'height': the_only_video.get('height'),
1745 'resolution': the_only_video.get('resolution') or self.format_resolution(the_only_video),
1746 'fps': the_only_video.get('fps'),
1747 'vcodec': the_only_video.get('vcodec'),
1748 'vbr': the_only_video.get('vbr'),
1749 'stretched_ratio': the_only_video.get('stretched_ratio'),
1750 })
1751
1752 if the_only_audio:
1753 new_dict.update({
1754 'acodec': the_only_audio.get('acodec'),
1755 'abr': the_only_audio.get('abr'),
1756 })
1757
1758 return new_dict
1759
e8e73840 1760 def _check_formats(formats):
981052c9 1761 if not check_formats:
1762 yield from formats
e8e73840 1763 for f in formats:
1764 self.to_screen('[info] Testing format %s' % f['format_id'])
21cd8fae 1765 temp_file = tempfile.NamedTemporaryFile(
1766 suffix='.tmp', delete=False,
1767 dir=self.get_output_path('temp') or None)
1768 temp_file.close()
fe346461 1769 try:
981052c9 1770 success, _ = self.dl(temp_file.name, f, test=True)
1771 except (DownloadError, IOError, OSError, ValueError) + network_exceptions:
1772 success = False
fe346461 1773 finally:
21cd8fae 1774 if os.path.exists(temp_file.name):
1775 try:
1776 os.remove(temp_file.name)
1777 except OSError:
1778 self.report_warning('Unable to delete temporary file "%s"' % temp_file.name)
981052c9 1779 if success:
e8e73840 1780 yield f
1781 else:
1782 self.to_screen('[info] Unable to download format %s. Skipping...' % f['format_id'])
1783
67134eab 1784 def _build_selector_function(selector):
909d24dd 1785 if isinstance(selector, list): # ,
67134eab
JMF
1786 fs = [_build_selector_function(s) for s in selector]
1787
317f7ab6 1788 def selector_function(ctx):
67134eab 1789 for f in fs:
981052c9 1790 yield from f(ctx)
67134eab 1791 return selector_function
909d24dd 1792
1793 elif selector.type == GROUP: # ()
0130afb7 1794 selector_function = _build_selector_function(selector.selector)
909d24dd 1795
1796 elif selector.type == PICKFIRST: # /
67134eab
JMF
1797 fs = [_build_selector_function(s) for s in selector.selector]
1798
317f7ab6 1799 def selector_function(ctx):
67134eab 1800 for f in fs:
317f7ab6 1801 picked_formats = list(f(ctx))
67134eab
JMF
1802 if picked_formats:
1803 return picked_formats
1804 return []
67134eab 1805
981052c9 1806 elif selector.type == MERGE: # +
1807 selector_1, selector_2 = map(_build_selector_function, selector.selector)
1808
1809 def selector_function(ctx):
1810 for pair in itertools.product(
1811 selector_1(copy.deepcopy(ctx)), selector_2(copy.deepcopy(ctx))):
1812 yield _merge(pair)
1813
909d24dd 1814 elif selector.type == SINGLE: # atom
598d185d 1815 format_spec = selector.selector or 'best'
909d24dd 1816
f8d4ad9a 1817 # TODO: Add allvideo, allaudio etc by generalizing the code with best/worst selector
909d24dd 1818 if format_spec == 'all':
1819 def selector_function(ctx):
981052c9 1820 yield from _check_formats(ctx['formats'])
f8d4ad9a 1821 elif format_spec == 'mergeall':
1822 def selector_function(ctx):
981052c9 1823 formats = list(_check_formats(ctx['formats']))
e01d6aa4 1824 if not formats:
1825 return
921b76ca 1826 merged_format = formats[-1]
1827 for f in formats[-2::-1]:
f8d4ad9a 1828 merged_format = _merge((merged_format, f))
1829 yield merged_format
909d24dd 1830
1831 else:
e8e73840 1832 format_fallback, format_reverse, format_idx = False, True, 1
eff63539 1833 mobj = re.match(
1834 r'(?P<bw>best|worst|b|w)(?P<type>video|audio|v|a)?(?P<mod>\*)?(?:\.(?P<n>[1-9]\d*))?$',
1835 format_spec)
1836 if mobj is not None:
1837 format_idx = int_or_none(mobj.group('n'), default=1)
e8e73840 1838 format_reverse = mobj.group('bw')[0] == 'b'
eff63539 1839 format_type = (mobj.group('type') or [None])[0]
1840 not_format_type = {'v': 'a', 'a': 'v'}.get(format_type)
1841 format_modified = mobj.group('mod') is not None
909d24dd 1842
1843 format_fallback = not format_type and not format_modified # for b, w
8326b00a 1844 _filter_f = (
eff63539 1845 (lambda f: f.get('%scodec' % format_type) != 'none')
1846 if format_type and format_modified # bv*, ba*, wv*, wa*
1847 else (lambda f: f.get('%scodec' % not_format_type) == 'none')
1848 if format_type # bv, ba, wv, wa
1849 else (lambda f: f.get('vcodec') != 'none' and f.get('acodec') != 'none')
1850 if not format_modified # b, w
8326b00a 1851 else lambda f: True) # b*, w*
1852 filter_f = lambda f: _filter_f(f) and (
1853 f.get('vcodec') != 'none' or f.get('acodec') != 'none')
67134eab 1854 else:
909d24dd 1855 filter_f = ((lambda f: f.get('ext') == format_spec)
1856 if format_spec in ['mp4', 'flv', 'webm', '3gp', 'm4a', 'mp3', 'ogg', 'aac', 'wav'] # extension
1857 else (lambda f: f.get('format_id') == format_spec)) # id
1858
1859 def selector_function(ctx):
1860 formats = list(ctx['formats'])
909d24dd 1861 matches = list(filter(filter_f, formats)) if filter_f is not None else formats
e8e73840 1862 if format_fallback and ctx['incomplete_formats'] and not matches:
909d24dd 1863 # for extractors with incomplete formats (audio only (soundcloud)
1864 # or video only (imgur)) best/worst will fallback to
1865 # best/worst {video,audio}-only format
e8e73840 1866 matches = formats
981052c9 1867 matches = LazyList(_check_formats(matches[::-1 if format_reverse else 1]))
1868 try:
e8e73840 1869 yield matches[format_idx - 1]
981052c9 1870 except IndexError:
1871 return
083c9df9 1872
67134eab 1873 filters = [self._build_format_filter(f) for f in selector.filters]
083c9df9 1874
317f7ab6
S
1875 def final_selector(ctx):
1876 ctx_copy = copy.deepcopy(ctx)
67134eab 1877 for _filter in filters:
317f7ab6
S
1878 ctx_copy['formats'] = list(filter(_filter, ctx_copy['formats']))
1879 return selector_function(ctx_copy)
67134eab 1880 return final_selector
083c9df9 1881
67134eab 1882 stream = io.BytesIO(format_spec.encode('utf-8'))
0130afb7 1883 try:
232541df 1884 tokens = list(_remove_unused_ops(compat_tokenize_tokenize(stream.readline)))
0130afb7
JMF
1885 except tokenize.TokenError:
1886 raise syntax_error('Missing closing/opening brackets or parenthesis', (0, len(format_spec)))
1887
1888 class TokenIterator(object):
1889 def __init__(self, tokens):
1890 self.tokens = tokens
1891 self.counter = 0
1892
1893 def __iter__(self):
1894 return self
1895
1896 def __next__(self):
1897 if self.counter >= len(self.tokens):
1898 raise StopIteration()
1899 value = self.tokens[self.counter]
1900 self.counter += 1
1901 return value
1902
1903 next = __next__
1904
1905 def restore_last_token(self):
1906 self.counter -= 1
1907
1908 parsed_selector = _parse_format_selection(iter(TokenIterator(tokens)))
67134eab 1909 return _build_selector_function(parsed_selector)
a9c58ad9 1910
e5660ee6
JMF
1911 def _calc_headers(self, info_dict):
1912 res = std_headers.copy()
1913
1914 add_headers = info_dict.get('http_headers')
1915 if add_headers:
1916 res.update(add_headers)
1917
1918 cookies = self._calc_cookies(info_dict)
1919 if cookies:
1920 res['Cookie'] = cookies
1921
0016b84e
S
1922 if 'X-Forwarded-For' not in res:
1923 x_forwarded_for_ip = info_dict.get('__x_forwarded_for_ip')
1924 if x_forwarded_for_ip:
1925 res['X-Forwarded-For'] = x_forwarded_for_ip
1926
e5660ee6
JMF
1927 return res
1928
1929 def _calc_cookies(self, info_dict):
5c2266df 1930 pr = sanitized_Request(info_dict['url'])
e5660ee6 1931 self.cookiejar.add_cookie_header(pr)
662435f7 1932 return pr.get_header('Cookie')
e5660ee6 1933
b0249bca 1934 def _sanitize_thumbnails(self, info_dict):
bc516a3f 1935 thumbnails = info_dict.get('thumbnails')
1936 if thumbnails is None:
1937 thumbnail = info_dict.get('thumbnail')
1938 if thumbnail:
1939 info_dict['thumbnails'] = thumbnails = [{'url': thumbnail}]
1940 if thumbnails:
1941 thumbnails.sort(key=lambda t: (
1942 t.get('preference') if t.get('preference') is not None else -1,
1943 t.get('width') if t.get('width') is not None else -1,
1944 t.get('height') if t.get('height') is not None else -1,
1945 t.get('id') if t.get('id') is not None else '',
1946 t.get('url')))
b0249bca 1947
1948 def test_thumbnail(t):
1949 self.to_screen('[info] Testing thumbnail %s' % t['id'])
1950 try:
1951 self.urlopen(HEADRequest(t['url']))
1952 except network_exceptions as err:
1953 self.to_screen('[info] Unable to connect to thumbnail %s URL "%s" - %s. Skipping...' % (
1954 t['id'], t['url'], error_to_compat_str(err)))
1955 return False
1956 return True
1957
bc516a3f 1958 for i, t in enumerate(thumbnails):
bc516a3f 1959 if t.get('id') is None:
1960 t['id'] = '%d' % i
b0249bca 1961 if t.get('width') and t.get('height'):
1962 t['resolution'] = '%dx%d' % (t['width'], t['height'])
1963 t['url'] = sanitize_url(t['url'])
1964 if self.params.get('check_formats'):
981052c9 1965 info_dict['thumbnails'] = LazyList(filter(test_thumbnail, thumbnails[::-1])).reverse()
bc516a3f 1966
dd82ffea
JMF
1967 def process_video_result(self, info_dict, download=True):
1968 assert info_dict.get('_type', 'video') == 'video'
1969
bec1fad2
PH
1970 if 'id' not in info_dict:
1971 raise ExtractorError('Missing "id" field in extractor result')
1972 if 'title' not in info_dict:
1973 raise ExtractorError('Missing "title" field in extractor result')
1974
c9969434
S
1975 def report_force_conversion(field, field_not, conversion):
1976 self.report_warning(
1977 '"%s" field is not %s - forcing %s conversion, there is an error in extractor'
1978 % (field, field_not, conversion))
1979
1980 def sanitize_string_field(info, string_field):
1981 field = info.get(string_field)
1982 if field is None or isinstance(field, compat_str):
1983 return
1984 report_force_conversion(string_field, 'a string', 'string')
1985 info[string_field] = compat_str(field)
1986
1987 def sanitize_numeric_fields(info):
1988 for numeric_field in self._NUMERIC_FIELDS:
1989 field = info.get(numeric_field)
1990 if field is None or isinstance(field, compat_numeric_types):
1991 continue
1992 report_force_conversion(numeric_field, 'numeric', 'int')
1993 info[numeric_field] = int_or_none(field)
1994
1995 sanitize_string_field(info_dict, 'id')
1996 sanitize_numeric_fields(info_dict)
be6217b2 1997
dd82ffea
JMF
1998 if 'playlist' not in info_dict:
1999 # It isn't part of a playlist
2000 info_dict['playlist'] = None
2001 info_dict['playlist_index'] = None
2002
bc516a3f 2003 self._sanitize_thumbnails(info_dict)
d5519808 2004
b7b72db9 2005 if self.params.get('list_thumbnails'):
2006 self.list_thumbnails(info_dict)
2007 return
2008
536a55da 2009 thumbnail = info_dict.get('thumbnail')
bc516a3f 2010 thumbnails = info_dict.get('thumbnails')
536a55da
S
2011 if thumbnail:
2012 info_dict['thumbnail'] = sanitize_url(thumbnail)
2013 elif thumbnails:
d5519808
PH
2014 info_dict['thumbnail'] = thumbnails[-1]['url']
2015
c9ae7b95 2016 if 'display_id' not in info_dict and 'id' in info_dict:
0afef30b
PH
2017 info_dict['display_id'] = info_dict['id']
2018
10db0d2f 2019 for ts_key, date_key in (
2020 ('timestamp', 'upload_date'),
2021 ('release_timestamp', 'release_date'),
2022 ):
2023 if info_dict.get(date_key) is None and info_dict.get(ts_key) is not None:
2024 # Working around out-of-range timestamp values (e.g. negative ones on Windows,
2025 # see http://bugs.python.org/issue1646728)
2026 try:
2027 upload_date = datetime.datetime.utcfromtimestamp(info_dict[ts_key])
2028 info_dict[date_key] = upload_date.strftime('%Y%m%d')
2029 except (ValueError, OverflowError, OSError):
2030 pass
9d2ecdbc 2031
33d2fc2f
S
2032 # Auto generate title fields corresponding to the *_number fields when missing
2033 # in order to always have clean titles. This is very common for TV series.
2034 for field in ('chapter', 'season', 'episode'):
2035 if info_dict.get('%s_number' % field) is not None and not info_dict.get(field):
2036 info_dict[field] = '%s %d' % (field.capitalize(), info_dict['%s_number' % field])
2037
05108a49
S
2038 for cc_kind in ('subtitles', 'automatic_captions'):
2039 cc = info_dict.get(cc_kind)
2040 if cc:
2041 for _, subtitle in cc.items():
2042 for subtitle_format in subtitle:
2043 if subtitle_format.get('url'):
2044 subtitle_format['url'] = sanitize_url(subtitle_format['url'])
2045 if subtitle_format.get('ext') is None:
2046 subtitle_format['ext'] = determine_ext(subtitle_format['url']).lower()
2047
2048 automatic_captions = info_dict.get('automatic_captions')
4bba3716 2049 subtitles = info_dict.get('subtitles')
4bba3716 2050
a504ced0 2051 if self.params.get('listsubtitles', False):
360e1ca5 2052 if 'automatic_captions' in info_dict:
05108a49
S
2053 self.list_subtitles(
2054 info_dict['id'], automatic_captions, 'automatic captions')
4bba3716 2055 self.list_subtitles(info_dict['id'], subtitles, 'subtitles')
a504ced0 2056 return
05108a49 2057
360e1ca5 2058 info_dict['requested_subtitles'] = self.process_subtitles(
05108a49 2059 info_dict['id'], subtitles, automatic_captions)
a504ced0 2060
dd82ffea
JMF
2061 # We now pick which formats have to be downloaded
2062 if info_dict.get('formats') is None:
2063 # There's only one format available
2064 formats = [info_dict]
2065 else:
2066 formats = info_dict['formats']
2067
db95dc13 2068 if not formats:
b7da73eb 2069 if not self.params.get('ignore_no_formats_error'):
2070 raise ExtractorError('No video formats found!')
2071 else:
2072 self.report_warning('No video formats found!')
db95dc13 2073
73af5cc8
S
2074 def is_wellformed(f):
2075 url = f.get('url')
a5ac0c47 2076 if not url:
73af5cc8
S
2077 self.report_warning(
2078 '"url" field is missing or empty - skipping format, '
2079 'there is an error in extractor')
a5ac0c47
S
2080 return False
2081 if isinstance(url, bytes):
2082 sanitize_string_field(f, 'url')
2083 return True
73af5cc8
S
2084
2085 # Filter out malformed formats for better extraction robustness
2086 formats = list(filter(is_wellformed, formats))
2087
181c7053
S
2088 formats_dict = {}
2089
dd82ffea 2090 # We check that all the formats have the format and format_id fields
db95dc13 2091 for i, format in enumerate(formats):
c9969434
S
2092 sanitize_string_field(format, 'format_id')
2093 sanitize_numeric_fields(format)
dcf77cf1 2094 format['url'] = sanitize_url(format['url'])
e74e3b63 2095 if not format.get('format_id'):
8016c922 2096 format['format_id'] = compat_str(i)
e2effb08
S
2097 else:
2098 # Sanitize format_id from characters used in format selector expression
ec85ded8 2099 format['format_id'] = re.sub(r'[\s,/+\[\]()]', '_', format['format_id'])
181c7053
S
2100 format_id = format['format_id']
2101 if format_id not in formats_dict:
2102 formats_dict[format_id] = []
2103 formats_dict[format_id].append(format)
2104
2105 # Make sure all formats have unique format_id
2106 for format_id, ambiguous_formats in formats_dict.items():
2107 if len(ambiguous_formats) > 1:
2108 for i, format in enumerate(ambiguous_formats):
2109 format['format_id'] = '%s-%d' % (format_id, i)
2110
2111 for i, format in enumerate(formats):
8c51aa65 2112 if format.get('format') is None:
6febd1c1 2113 format['format'] = '{id} - {res}{note}'.format(
8c51aa65
JMF
2114 id=format['format_id'],
2115 res=self.format_resolution(format),
6febd1c1 2116 note=' ({0})'.format(format['format_note']) if format.get('format_note') is not None else '',
8c51aa65 2117 )
c1002e96 2118 # Automatically determine file extension if missing
5b1d8575 2119 if format.get('ext') is None:
cce929ea 2120 format['ext'] = determine_ext(format['url']).lower()
b5559424
S
2121 # Automatically determine protocol if missing (useful for format
2122 # selection purposes)
6f0be937 2123 if format.get('protocol') is None:
b5559424 2124 format['protocol'] = determine_protocol(format)
e5660ee6
JMF
2125 # Add HTTP headers, so that external programs can use them from the
2126 # json output
2127 full_format_info = info_dict.copy()
2128 full_format_info.update(format)
2129 format['http_headers'] = self._calc_headers(full_format_info)
0016b84e
S
2130 # Remove private housekeeping stuff
2131 if '__x_forwarded_for_ip' in info_dict:
2132 del info_dict['__x_forwarded_for_ip']
dd82ffea 2133
4bcc7bd1 2134 # TODO Central sorting goes here
99e206d5 2135
b7da73eb 2136 if formats and formats[0] is not info_dict:
b3d9ef88
JMF
2137 # only set the 'formats' fields if the original info_dict list them
2138 # otherwise we end up with a circular reference, the first (and unique)
f89197d7 2139 # element in the 'formats' field in info_dict is info_dict itself,
dfb1b146 2140 # which can't be exported to json
b3d9ef88 2141 info_dict['formats'] = formats
4ec82a72 2142
2143 info_dict, _ = self.pre_process(info_dict)
2144
cfb56d1a 2145 if self.params.get('listformats'):
b7da73eb 2146 if not info_dict.get('formats'):
2147 raise ExtractorError('No video formats found', expected=True)
bfaae0a7 2148 self.list_formats(info_dict)
2149 return
2150
187986a8 2151 format_selector = self.format_selector
2152 if format_selector is None:
0017d9ad 2153 req_format = self._default_format_spec(info_dict, download=download)
0760b0a7 2154 self.write_debug('Default format spec: %s' % req_format)
187986a8 2155 format_selector = self.build_format_selector(req_format)
317f7ab6
S
2156
2157 # While in format selection we may need to have an access to the original
2158 # format set in order to calculate some metrics or do some processing.
2159 # For now we need to be able to guess whether original formats provided
2160 # by extractor are incomplete or not (i.e. whether extractor provides only
2161 # video-only or audio-only formats) for proper formats selection for
2162 # extractors with such incomplete formats (see
067aa17e 2163 # https://github.com/ytdl-org/youtube-dl/pull/5556).
317f7ab6
S
2164 # Since formats may be filtered during format selection and may not match
2165 # the original formats the results may be incorrect. Thus original formats
2166 # or pre-calculated metrics should be passed to format selection routines
2167 # as well.
2168 # We will pass a context object containing all necessary additional data
2169 # instead of just formats.
2170 # This fixes incorrect format selection issue (see
067aa17e 2171 # https://github.com/ytdl-org/youtube-dl/issues/10083).
2e221ca3 2172 incomplete_formats = (
317f7ab6 2173 # All formats are video-only or
3089bc74 2174 all(f.get('vcodec') != 'none' and f.get('acodec') == 'none' for f in formats)
317f7ab6 2175 # all formats are audio-only
3089bc74 2176 or all(f.get('vcodec') == 'none' and f.get('acodec') != 'none' for f in formats))
317f7ab6
S
2177
2178 ctx = {
2179 'formats': formats,
2180 'incomplete_formats': incomplete_formats,
2181 }
2182
2183 formats_to_download = list(format_selector(ctx))
dd82ffea 2184 if not formats_to_download:
b7da73eb 2185 if not self.params.get('ignore_no_formats_error'):
2186 raise ExtractorError('Requested format is not available', expected=True)
2187 else:
2188 self.report_warning('Requested format is not available')
4513a41a
A
2189 # Process what we can, even without any available formats.
2190 self.process_info(dict(info_dict))
b7da73eb 2191 elif download:
2192 self.to_screen(
07cce701 2193 '[info] %s: Downloading %d format(s): %s' % (
2194 info_dict['id'], len(formats_to_download),
2195 ", ".join([f['format_id'] for f in formats_to_download])))
b7da73eb 2196 for fmt in formats_to_download:
dd82ffea 2197 new_info = dict(info_dict)
4ec82a72 2198 # Save a reference to the original info_dict so that it can be modified in process_info if needed
2199 new_info['__original_infodict'] = info_dict
b7da73eb 2200 new_info.update(fmt)
dd82ffea
JMF
2201 self.process_info(new_info)
2202 # We update the info dict with the best quality format (backwards compatibility)
b7da73eb 2203 if formats_to_download:
2204 info_dict.update(formats_to_download[-1])
dd82ffea
JMF
2205 return info_dict
2206
98c70d6f 2207 def process_subtitles(self, video_id, normal_subtitles, automatic_captions):
a504ced0 2208 """Select the requested subtitles and their format"""
98c70d6f
JMF
2209 available_subs = {}
2210 if normal_subtitles and self.params.get('writesubtitles'):
2211 available_subs.update(normal_subtitles)
2212 if automatic_captions and self.params.get('writeautomaticsub'):
2213 for lang, cap_info in automatic_captions.items():
360e1ca5
JMF
2214 if lang not in available_subs:
2215 available_subs[lang] = cap_info
2216
4d171848
JMF
2217 if (not self.params.get('writesubtitles') and not
2218 self.params.get('writeautomaticsub') or not
2219 available_subs):
2220 return None
a504ced0 2221
c32b0aab 2222 all_sub_langs = available_subs.keys()
a504ced0 2223 if self.params.get('allsubtitles', False):
c32b0aab 2224 requested_langs = all_sub_langs
2225 elif self.params.get('subtitleslangs', False):
2226 requested_langs = set()
2227 for lang in self.params.get('subtitleslangs'):
2228 if lang == 'all':
2229 requested_langs.update(all_sub_langs)
2230 continue
2231 discard = lang[0] == '-'
2232 if discard:
2233 lang = lang[1:]
2234 current_langs = filter(re.compile(lang + '$').match, all_sub_langs)
2235 if discard:
2236 for lang in current_langs:
2237 requested_langs.discard(lang)
2238 else:
2239 requested_langs.update(current_langs)
2240 elif 'en' in available_subs:
2241 requested_langs = ['en']
a504ced0 2242 else:
c32b0aab 2243 requested_langs = [list(all_sub_langs)[0]]
e632bce2 2244 self.write_debug('Downloading subtitles: %s' % ', '.join(requested_langs))
a504ced0
JMF
2245
2246 formats_query = self.params.get('subtitlesformat', 'best')
2247 formats_preference = formats_query.split('/') if formats_query else []
2248 subs = {}
2249 for lang in requested_langs:
2250 formats = available_subs.get(lang)
2251 if formats is None:
2252 self.report_warning('%s subtitles not available for %s' % (lang, video_id))
2253 continue
a504ced0
JMF
2254 for ext in formats_preference:
2255 if ext == 'best':
2256 f = formats[-1]
2257 break
2258 matches = list(filter(lambda f: f['ext'] == ext, formats))
2259 if matches:
2260 f = matches[-1]
2261 break
2262 else:
2263 f = formats[-1]
2264 self.report_warning(
2265 'No subtitle format found matching "%s" for language %s, '
2266 'using %s' % (formats_query, lang, f['ext']))
2267 subs[lang] = f
2268 return subs
2269
d06daf23 2270 def __forced_printings(self, info_dict, filename, incomplete):
53c18592 2271 def print_mandatory(field, actual_field=None):
2272 if actual_field is None:
2273 actual_field = field
d06daf23 2274 if (self.params.get('force%s' % field, False)
53c18592 2275 and (not incomplete or info_dict.get(actual_field) is not None)):
2276 self.to_stdout(info_dict[actual_field])
d06daf23
S
2277
2278 def print_optional(field):
2279 if (self.params.get('force%s' % field, False)
2280 and info_dict.get(field) is not None):
2281 self.to_stdout(info_dict[field])
2282
53c18592 2283 info_dict = info_dict.copy()
2284 if filename is not None:
2285 info_dict['filename'] = filename
2286 if info_dict.get('requested_formats') is not None:
2287 # For RTMP URLs, also include the playpath
2288 info_dict['urls'] = '\n'.join(f['url'] + f.get('play_path', '') for f in info_dict['requested_formats'])
2289 elif 'url' in info_dict:
2290 info_dict['urls'] = info_dict['url'] + info_dict.get('play_path', '')
2291
2292 for tmpl in self.params.get('forceprint', []):
2293 if re.match(r'\w+$', tmpl):
2294 tmpl = '%({})s'.format(tmpl)
2295 tmpl, info_copy = self.prepare_outtmpl(tmpl, info_dict)
2296 self.to_stdout(tmpl % info_copy)
2297
d06daf23
S
2298 print_mandatory('title')
2299 print_mandatory('id')
53c18592 2300 print_mandatory('url', 'urls')
d06daf23
S
2301 print_optional('thumbnail')
2302 print_optional('description')
53c18592 2303 print_optional('filename')
d06daf23
S
2304 if self.params.get('forceduration', False) and info_dict.get('duration') is not None:
2305 self.to_stdout(formatSeconds(info_dict['duration']))
2306 print_mandatory('format')
53c18592 2307
d06daf23 2308 if self.params.get('forcejson', False):
277d6ff5 2309 self.post_extract(info_dict)
75d43ca0 2310 self.to_stdout(json.dumps(info_dict, default=repr))
d06daf23 2311
e8e73840 2312 def dl(self, name, info, subtitle=False, test=False):
2313
2314 if test:
2315 verbose = self.params.get('verbose')
2316 params = {
2317 'test': True,
2318 'quiet': not verbose,
2319 'verbose': verbose,
2320 'noprogress': not verbose,
2321 'nopart': True,
2322 'skip_unavailable_fragments': False,
2323 'keep_fragments': False,
2324 'overwrites': True,
2325 '_no_ytdl_file': True,
2326 }
2327 else:
2328 params = self.params
2329 fd = get_suitable_downloader(info, params)(self, params)
2330 if not test:
2331 for ph in self._progress_hooks:
2332 fd.add_progress_hook(ph)
18e674b4 2333 urls = '", "'.join([f['url'] for f in info.get('requested_formats', [])] or [info['url']])
2334 self.write_debug('Invoking downloader on "%s"' % urls)
e8e73840 2335 new_info = dict(info)
2336 if new_info.get('http_headers') is None:
2337 new_info['http_headers'] = self._calc_headers(new_info)
2338 return fd.download(name, new_info, subtitle)
2339
8222d8de
JMF
2340 def process_info(self, info_dict):
2341 """Process a single resolved IE result."""
2342
2343 assert info_dict.get('_type', 'video') == 'video'
fd288278 2344
0202b52a 2345 info_dict.setdefault('__postprocessors', [])
2346
fd288278
PH
2347 max_downloads = self.params.get('max_downloads')
2348 if max_downloads is not None:
2349 if self._num_downloads >= int(max_downloads):
2350 raise MaxDownloadsReached()
8222d8de 2351
d06daf23 2352 # TODO: backward compatibility, to be removed
8222d8de 2353 info_dict['fulltitle'] = info_dict['title']
8222d8de 2354
4513a41a 2355 if 'format' not in info_dict and 'ext' in info_dict:
8222d8de
JMF
2356 info_dict['format'] = info_dict['ext']
2357
c77495e3 2358 if self._match_entry(info_dict) is not None:
8222d8de
JMF
2359 return
2360
277d6ff5 2361 self.post_extract(info_dict)
fd288278 2362 self._num_downloads += 1
8222d8de 2363
dcf64d43 2364 # info_dict['_filename'] needs to be set for backward compatibility
de6000d9 2365 info_dict['_filename'] = full_filename = self.prepare_filename(info_dict, warn=True)
2366 temp_filename = self.prepare_filename(info_dict, 'temp')
0202b52a 2367 files_to_move = {}
8222d8de
JMF
2368
2369 # Forced printings
4513a41a 2370 self.__forced_printings(info_dict, full_filename, incomplete=('format' not in info_dict))
8222d8de 2371
8222d8de 2372 if self.params.get('simulate', False):
2d30509f 2373 if self.params.get('force_write_download_archive', False):
2374 self.record_download_archive(info_dict)
2375
2376 # Do nothing else if in simulate mode
8222d8de
JMF
2377 return
2378
de6000d9 2379 if full_filename is None:
8222d8de
JMF
2380 return
2381
e92caff5 2382 if not self._ensure_dir_exists(encodeFilename(full_filename)):
0202b52a 2383 return
e92caff5 2384 if not self._ensure_dir_exists(encodeFilename(temp_filename)):
8222d8de
JMF
2385 return
2386
2387 if self.params.get('writedescription', False):
de6000d9 2388 descfn = self.prepare_filename(info_dict, 'description')
e92caff5 2389 if not self._ensure_dir_exists(encodeFilename(descfn)):
0202b52a 2390 return
0c3d0f51 2391 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(descfn)):
6febd1c1 2392 self.to_screen('[info] Video description is already present')
f00fd51d
JMF
2393 elif info_dict.get('description') is None:
2394 self.report_warning('There\'s no description to write.')
7b6fefc9
PH
2395 else:
2396 try:
6febd1c1 2397 self.to_screen('[info] Writing video description to: ' + descfn)
7b6fefc9
PH
2398 with io.open(encodeFilename(descfn), 'w', encoding='utf-8') as descfile:
2399 descfile.write(info_dict['description'])
7b6fefc9 2400 except (OSError, IOError):
6febd1c1 2401 self.report_error('Cannot write description file ' + descfn)
7b6fefc9 2402 return
8222d8de 2403
1fb07d10 2404 if self.params.get('writeannotations', False):
de6000d9 2405 annofn = self.prepare_filename(info_dict, 'annotation')
e92caff5 2406 if not self._ensure_dir_exists(encodeFilename(annofn)):
0202b52a 2407 return
0c3d0f51 2408 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(annofn)):
6febd1c1 2409 self.to_screen('[info] Video annotations are already present')
ffddb112
RA
2410 elif not info_dict.get('annotations'):
2411 self.report_warning('There are no annotations to write.')
7b6fefc9
PH
2412 else:
2413 try:
6febd1c1 2414 self.to_screen('[info] Writing video annotations to: ' + annofn)
7b6fefc9
PH
2415 with io.open(encodeFilename(annofn), 'w', encoding='utf-8') as annofile:
2416 annofile.write(info_dict['annotations'])
2417 except (KeyError, TypeError):
6febd1c1 2418 self.report_warning('There are no annotations to write.')
7b6fefc9 2419 except (OSError, IOError):
6febd1c1 2420 self.report_error('Cannot write annotations file: ' + annofn)
7b6fefc9 2421 return
1fb07d10 2422
c4a91be7 2423 subtitles_are_requested = any([self.params.get('writesubtitles', False),
0b7f3118 2424 self.params.get('writeautomaticsub')])
c4a91be7 2425
c84dd8a9 2426 if subtitles_are_requested and info_dict.get('requested_subtitles'):
8222d8de
JMF
2427 # subtitles download errors are already managed as troubles in relevant IE
2428 # that way it will silently go on when used with unsupporting IE
c84dd8a9 2429 subtitles = info_dict['requested_subtitles']
fa57af1e 2430 # ie = self.get_info_extractor(info_dict['extractor_key'])
a504ced0
JMF
2431 for sub_lang, sub_info in subtitles.items():
2432 sub_format = sub_info['ext']
56d868db 2433 sub_filename = subtitles_filename(temp_filename, sub_lang, sub_format, info_dict.get('ext'))
2434 sub_filename_final = subtitles_filename(
2435 self.prepare_filename(info_dict, 'subtitle'), sub_lang, sub_format, info_dict.get('ext'))
0c3d0f51 2436 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(sub_filename)):
5ff1bc0c 2437 self.to_screen('[info] Video subtitle %s.%s is already present' % (sub_lang, sub_format))
dcf64d43 2438 sub_info['filepath'] = sub_filename
0202b52a 2439 files_to_move[sub_filename] = sub_filename_final
a504ced0 2440 else:
0c9df79e 2441 self.to_screen('[info] Writing video subtitles to: ' + sub_filename)
5ff1bc0c
RA
2442 if sub_info.get('data') is not None:
2443 try:
2444 # Use newline='' to prevent conversion of newline characters
067aa17e 2445 # See https://github.com/ytdl-org/youtube-dl/issues/10268
5ff1bc0c
RA
2446 with io.open(encodeFilename(sub_filename), 'w', encoding='utf-8', newline='') as subfile:
2447 subfile.write(sub_info['data'])
dcf64d43 2448 sub_info['filepath'] = sub_filename
0202b52a 2449 files_to_move[sub_filename] = sub_filename_final
5ff1bc0c
RA
2450 except (OSError, IOError):
2451 self.report_error('Cannot write subtitles file ' + sub_filename)
2452 return
7b6fefc9 2453 else:
5ff1bc0c 2454 try:
e8e73840 2455 self.dl(sub_filename, sub_info.copy(), subtitle=True)
dcf64d43 2456 sub_info['filepath'] = sub_filename
0202b52a 2457 files_to_move[sub_filename] = sub_filename_final
fe346461 2458 except (ExtractorError, IOError, OSError, ValueError) + network_exceptions as err:
5ff1bc0c
RA
2459 self.report_warning('Unable to download subtitle for "%s": %s' %
2460 (sub_lang, error_to_compat_str(err)))
2461 continue
8222d8de 2462
8222d8de 2463 if self.params.get('writeinfojson', False):
de6000d9 2464 infofn = self.prepare_filename(info_dict, 'infojson')
e92caff5 2465 if not self._ensure_dir_exists(encodeFilename(infofn)):
0202b52a 2466 return
0c3d0f51 2467 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(infofn)):
66c935fb 2468 self.to_screen('[info] Video metadata is already present')
7b6fefc9 2469 else:
66c935fb 2470 self.to_screen('[info] Writing video metadata as JSON to: ' + infofn)
7b6fefc9 2471 try:
75d43ca0 2472 write_json_file(self.filter_requested_info(info_dict, self.params.get('clean_infojson', True)), infofn)
7b6fefc9 2473 except (OSError, IOError):
66c935fb 2474 self.report_error('Cannot write video metadata to JSON file ' + infofn)
7b6fefc9 2475 return
de6000d9 2476 info_dict['__infojson_filename'] = infofn
8222d8de 2477
56d868db 2478 for thumb_ext in self._write_thumbnails(info_dict, temp_filename):
2479 thumb_filename_temp = replace_extension(temp_filename, thumb_ext, info_dict.get('ext'))
2480 thumb_filename = replace_extension(
2481 self.prepare_filename(info_dict, 'thumbnail'), thumb_ext, info_dict.get('ext'))
dcf64d43 2482 files_to_move[thumb_filename_temp] = thumb_filename
8222d8de 2483
732044af 2484 # Write internet shortcut files
2485 url_link = webloc_link = desktop_link = False
2486 if self.params.get('writelink', False):
2487 if sys.platform == "darwin": # macOS.
2488 webloc_link = True
2489 elif sys.platform.startswith("linux"):
2490 desktop_link = True
2491 else: # if sys.platform in ['win32', 'cygwin']:
2492 url_link = True
2493 if self.params.get('writeurllink', False):
2494 url_link = True
2495 if self.params.get('writewebloclink', False):
2496 webloc_link = True
2497 if self.params.get('writedesktoplink', False):
2498 desktop_link = True
2499
2500 if url_link or webloc_link or desktop_link:
2501 if 'webpage_url' not in info_dict:
2502 self.report_error('Cannot write internet shortcut file because the "webpage_url" field is missing in the media information')
2503 return
2504 ascii_url = iri_to_uri(info_dict['webpage_url'])
2505
2506 def _write_link_file(extension, template, newline, embed_filename):
0202b52a 2507 linkfn = replace_extension(full_filename, extension, info_dict.get('ext'))
10e3742e 2508 if self.params.get('overwrites', True) and os.path.exists(encodeFilename(linkfn)):
732044af 2509 self.to_screen('[info] Internet shortcut is already present')
2510 else:
2511 try:
2512 self.to_screen('[info] Writing internet shortcut to: ' + linkfn)
2513 with io.open(encodeFilename(to_high_limit_path(linkfn)), 'w', encoding='utf-8', newline=newline) as linkfile:
2514 template_vars = {'url': ascii_url}
2515 if embed_filename:
2516 template_vars['filename'] = linkfn[:-(len(extension) + 1)]
2517 linkfile.write(template % template_vars)
2518 except (OSError, IOError):
2519 self.report_error('Cannot write internet shortcut ' + linkfn)
2520 return False
2521 return True
2522
2523 if url_link:
2524 if not _write_link_file('url', DOT_URL_LINK_TEMPLATE, '\r\n', embed_filename=False):
2525 return
2526 if webloc_link:
2527 if not _write_link_file('webloc', DOT_WEBLOC_LINK_TEMPLATE, '\n', embed_filename=False):
2528 return
2529 if desktop_link:
2530 if not _write_link_file('desktop', DOT_DESKTOP_LINK_TEMPLATE, '\n', embed_filename=True):
2531 return
2532
56d868db 2533 try:
2534 info_dict, files_to_move = self.pre_process(info_dict, 'before_dl', files_to_move)
2535 except PostProcessingError as err:
2536 self.report_error('Preprocessing: %s' % str(err))
2537 return
2538
732044af 2539 must_record_download_archive = False
56d868db 2540 if self.params.get('skip_download', False):
2541 info_dict['filepath'] = temp_filename
2542 info_dict['__finaldir'] = os.path.dirname(os.path.abspath(encodeFilename(full_filename)))
2543 info_dict['__files_to_move'] = files_to_move
2544 info_dict = self.run_pp(MoveFilesAfterDownloadPP(self, False), info_dict)
2545 else:
2546 # Download
4340deca 2547 try:
0202b52a 2548
6b591b29 2549 def existing_file(*filepaths):
2550 ext = info_dict.get('ext')
2551 final_ext = self.params.get('final_ext', ext)
2552 existing_files = []
2553 for file in orderedSet(filepaths):
2554 if final_ext != ext:
2555 converted = replace_extension(file, final_ext, ext)
2556 if os.path.exists(encodeFilename(converted)):
2557 existing_files.append(converted)
2558 if os.path.exists(encodeFilename(file)):
2559 existing_files.append(file)
2560
2561 if not existing_files or self.params.get('overwrites', False):
2562 for file in orderedSet(existing_files):
2563 self.report_file_delete(file)
2564 os.remove(encodeFilename(file))
2565 return None
2566
2567 self.report_file_already_downloaded(existing_files[0])
2568 info_dict['ext'] = os.path.splitext(existing_files[0])[1][1:]
2569 return existing_files[0]
0202b52a 2570
2571 success = True
4340deca 2572 if info_dict.get('requested_formats') is not None:
81cd954a
S
2573
2574 def compatible_formats(formats):
d03cfdce 2575 # TODO: some formats actually allow this (mkv, webm, ogg, mp4), but not all of them.
2576 video_formats = [format for format in formats if format.get('vcodec') != 'none']
2577 audio_formats = [format for format in formats if format.get('acodec') != 'none']
2578 if len(video_formats) > 2 or len(audio_formats) > 2:
2579 return False
2580
81cd954a 2581 # Check extension
d03cfdce 2582 exts = set(format.get('ext') for format in formats)
2583 COMPATIBLE_EXTS = (
2584 set(('mp3', 'mp4', 'm4a', 'm4p', 'm4b', 'm4r', 'm4v', 'ismv', 'isma')),
2585 set(('webm',)),
2586 )
2587 for ext_sets in COMPATIBLE_EXTS:
2588 if ext_sets.issuperset(exts):
2589 return True
81cd954a
S
2590 # TODO: Check acodec/vcodec
2591 return False
2592
2593 requested_formats = info_dict['requested_formats']
0202b52a 2594 old_ext = info_dict['ext']
4d971a16 2595 if self.params.get('merge_output_format') is None:
2596 if not compatible_formats(requested_formats):
2597 info_dict['ext'] = 'mkv'
2598 self.report_warning(
2599 'Requested formats are incompatible for merge and will be merged into mkv.')
2600 if (info_dict['ext'] == 'webm'
2601 and self.params.get('writethumbnail', False)
2602 and info_dict.get('thumbnails')):
2603 info_dict['ext'] = 'mkv'
2604 self.report_warning(
2605 'webm doesn\'t support embedding a thumbnail, mkv will be used.')
0202b52a 2606
2607 def correct_ext(filename):
2608 filename_real_ext = os.path.splitext(filename)[1][1:]
2609 filename_wo_ext = (
2610 os.path.splitext(filename)[0]
2611 if filename_real_ext == old_ext
2612 else filename)
2613 return '%s.%s' % (filename_wo_ext, info_dict['ext'])
2614
38c6902b 2615 # Ensure filename always has a correct extension for successful merge
0202b52a 2616 full_filename = correct_ext(full_filename)
2617 temp_filename = correct_ext(temp_filename)
2618 dl_filename = existing_file(full_filename, temp_filename)
1ea24129 2619 info_dict['__real_download'] = False
18e674b4 2620
2621 _protocols = set(determine_protocol(f) for f in requested_formats)
2622 if len(_protocols) == 1:
2623 info_dict['protocol'] = _protocols.pop()
2624 directly_mergable = (
2625 'no-direct-merge' not in self.params.get('compat_opts', [])
2626 and info_dict.get('protocol') is not None # All requested formats have same protocol
2627 and not self.params.get('allow_unplayable_formats')
2628 and get_suitable_downloader(info_dict, self.params).__name__ == 'FFmpegFD')
2629 if directly_mergable:
2630 info_dict['url'] = requested_formats[0]['url']
2631 # Treat it as a single download
2632 dl_filename = existing_file(full_filename, temp_filename)
2633 if dl_filename is None:
2634 success, real_download = self.dl(temp_filename, info_dict)
2635 info_dict['__real_download'] = real_download
2636 else:
2637 downloaded = []
2638 merger = FFmpegMergerPP(self)
2639 if self.params.get('allow_unplayable_formats'):
2640 self.report_warning(
2641 'You have requested merging of multiple formats '
2642 'while also allowing unplayable formats to be downloaded. '
2643 'The formats won\'t be merged to prevent data corruption.')
2644 elif not merger.available:
2645 self.report_warning(
2646 'You have requested merging of multiple formats but ffmpeg is not installed. '
2647 'The formats won\'t be merged.')
2648
2649 if dl_filename is None:
2650 for f in requested_formats:
2651 new_info = dict(info_dict)
2652 del new_info['requested_formats']
2653 new_info.update(f)
2654 fname = prepend_extension(
2655 self.prepare_filename(new_info, 'temp'),
2656 'f%s' % f['format_id'], new_info['ext'])
2657 if not self._ensure_dir_exists(fname):
2658 return
2659 downloaded.append(fname)
2660 partial_success, real_download = self.dl(fname, new_info)
2661 info_dict['__real_download'] = info_dict['__real_download'] or real_download
2662 success = success and partial_success
2663 if merger.available and not self.params.get('allow_unplayable_formats'):
2664 info_dict['__postprocessors'].append(merger)
2665 info_dict['__files_to_merge'] = downloaded
2666 # Even if there were no downloads, it is being merged only now
2667 info_dict['__real_download'] = True
2668 else:
2669 for file in downloaded:
2670 files_to_move[file] = None
4340deca
P
2671 else:
2672 # Just a single file
0202b52a 2673 dl_filename = existing_file(full_filename, temp_filename)
2674 if dl_filename is None:
e8e73840 2675 success, real_download = self.dl(temp_filename, info_dict)
0202b52a 2676 info_dict['__real_download'] = real_download
2677
0202b52a 2678 dl_filename = dl_filename or temp_filename
c571435f 2679 info_dict['__finaldir'] = os.path.dirname(os.path.abspath(encodeFilename(full_filename)))
0202b52a 2680
3158150c 2681 except network_exceptions as err:
7960b056 2682 self.report_error('unable to download video data: %s' % error_to_compat_str(err))
4340deca
P
2683 return
2684 except (OSError, IOError) as err:
2685 raise UnavailableVideoError(err)
2686 except (ContentTooShortError, ) as err:
2687 self.report_error('content too short (expected %s bytes and served %s)' % (err.expected, err.downloaded))
2688 return
8222d8de 2689
de6000d9 2690 if success and full_filename != '-':
f17f8651 2691
fd7cfb64 2692 def fixup():
2693 do_fixup = True
2694 fixup_policy = self.params.get('fixup')
2695 vid = info_dict['id']
2696
2697 if fixup_policy in ('ignore', 'never'):
2698 return
2699 elif fixup_policy == 'warn':
2700 do_fixup = False
f89b3e2d 2701 elif fixup_policy != 'force':
2702 assert fixup_policy in ('detect_or_warn', None)
2703 if not info_dict.get('__real_download'):
2704 do_fixup = False
fd7cfb64 2705
2706 def ffmpeg_fixup(cndn, msg, cls):
2707 if not cndn:
2708 return
2709 if not do_fixup:
2710 self.report_warning(f'{vid}: {msg}')
2711 return
2712 pp = cls(self)
2713 if pp.available:
2714 info_dict['__postprocessors'].append(pp)
2715 else:
2716 self.report_warning(f'{vid}: {msg}. Install ffmpeg to fix this automatically')
2717
2718 stretched_ratio = info_dict.get('stretched_ratio')
2719 ffmpeg_fixup(
2720 stretched_ratio not in (1, None),
2721 f'Non-uniform pixel ratio {stretched_ratio}',
2722 FFmpegFixupStretchedPP)
2723
2724 ffmpeg_fixup(
2725 (info_dict.get('requested_formats') is None
2726 and info_dict.get('container') == 'm4a_dash'
2727 and info_dict.get('ext') == 'm4a'),
2728 'writing DASH m4a. Only some players support this container',
2729 FFmpegFixupM4aPP)
2730
2731 downloader = (get_suitable_downloader(info_dict, self.params).__name__
2732 if 'protocol' in info_dict else None)
2733 ffmpeg_fixup(downloader == 'HlsFD', 'malformed AAC bitstream detected', FFmpegFixupM3u8PP)
e36d50c5 2734 ffmpeg_fixup(downloader == 'WebSocketFragmentFD', 'malformed timestamps detected', FFmpegFixupTimestampPP)
2735 ffmpeg_fixup(downloader == 'WebSocketFragmentFD', 'malformed duration detected', FFmpegFixupDurationPP)
fd7cfb64 2736
2737 fixup()
8222d8de 2738 try:
23c1a667 2739 info_dict = self.post_process(dl_filename, info_dict, files_to_move)
af819c21 2740 except PostProcessingError as err:
2741 self.report_error('Postprocessing: %s' % str(err))
8222d8de 2742 return
ab8e5e51
AM
2743 try:
2744 for ph in self._post_hooks:
23c1a667 2745 ph(info_dict['filepath'])
ab8e5e51
AM
2746 except Exception as err:
2747 self.report_error('post hooks: %s' % str(err))
2748 return
2d30509f 2749 must_record_download_archive = True
2750
2751 if must_record_download_archive or self.params.get('force_write_download_archive', False):
2752 self.record_download_archive(info_dict)
c3e6ffba 2753 max_downloads = self.params.get('max_downloads')
2754 if max_downloads is not None and self._num_downloads >= int(max_downloads):
2755 raise MaxDownloadsReached()
8222d8de
JMF
2756
2757 def download(self, url_list):
2758 """Download a given list of URLs."""
de6000d9 2759 outtmpl = self.outtmpl_dict['default']
3089bc74
S
2760 if (len(url_list) > 1
2761 and outtmpl != '-'
2762 and '%' not in outtmpl
2763 and self.params.get('max_downloads') != 1):
acd69589 2764 raise SameFileError(outtmpl)
8222d8de
JMF
2765
2766 for url in url_list:
2767 try:
5f6a1245 2768 # It also downloads the videos
61aa5ba3
S
2769 res = self.extract_info(
2770 url, force_generic_extractor=self.params.get('force_generic_extractor', False))
8222d8de 2771 except UnavailableVideoError:
6febd1c1 2772 self.report_error('unable to download video')
8222d8de 2773 except MaxDownloadsReached:
8b0d7497 2774 self.to_screen('[info] Maximum number of downloaded files reached')
2775 raise
2776 except ExistingVideoReached:
d83cb531 2777 self.to_screen('[info] Encountered a file that is already in the archive, stopping due to --break-on-existing')
8b0d7497 2778 raise
2779 except RejectedVideoReached:
d83cb531 2780 self.to_screen('[info] Encountered a file that did not match filter, stopping due to --break-on-reject')
8222d8de 2781 raise
63e0be34
PH
2782 else:
2783 if self.params.get('dump_single_json', False):
277d6ff5 2784 self.post_extract(res)
75d43ca0 2785 self.to_stdout(json.dumps(res, default=repr))
8222d8de
JMF
2786
2787 return self._download_retcode
2788
1dcc4c0c 2789 def download_with_info_file(self, info_filename):
31bd3925
JMF
2790 with contextlib.closing(fileinput.FileInput(
2791 [info_filename], mode='r',
2792 openhook=fileinput.hook_encoded('utf-8'))) as f:
2793 # FileInput doesn't have a read method, we can't call json.load
498f5606 2794 info = self.filter_requested_info(json.loads('\n'.join(f)), self.params.get('clean_infojson', True))
d4943898
JMF
2795 try:
2796 self.process_ie_result(info, download=True)
d3f62c19 2797 except (DownloadError, EntryNotInPlaylist, ThrottledDownload):
d4943898
JMF
2798 webpage_url = info.get('webpage_url')
2799 if webpage_url is not None:
6febd1c1 2800 self.report_warning('The info failed to download, trying with "%s"' % webpage_url)
d4943898
JMF
2801 return self.download([webpage_url])
2802 else:
2803 raise
2804 return self._download_retcode
1dcc4c0c 2805
cb202fd2 2806 @staticmethod
75d43ca0 2807 def filter_requested_info(info_dict, actually_filter=True):
ae8f99e6 2808 remove_keys = ['__original_infodict'] # Always remove this since this may contain a copy of the entire dict
2809 keep_keys = ['_type'], # Always keep this to facilitate load-info-json
2810 if actually_filter:
bd99f6e6 2811 remove_keys += ('requested_formats', 'requested_subtitles', 'requested_entries', 'filepath', 'entries', 'original_url')
ae8f99e6 2812 empty_values = (None, {}, [], set(), tuple())
2813 reject = lambda k, v: k not in keep_keys and (
2814 k.startswith('_') or k in remove_keys or v in empty_values)
2815 else:
394dcd44 2816 info_dict['epoch'] = int(time.time())
ae8f99e6 2817 reject = lambda k, v: k in remove_keys
5226731e 2818 filter_fn = lambda obj: (
b0249bca 2819 list(map(filter_fn, obj)) if isinstance(obj, (LazyList, list, tuple, set))
a515a78d 2820 else obj if not isinstance(obj, dict)
ae8f99e6 2821 else dict((k, filter_fn(v)) for k, v in obj.items() if not reject(k, v)))
5226731e 2822 return filter_fn(info_dict)
cb202fd2 2823
dcf64d43 2824 def run_pp(self, pp, infodict):
5bfa4862 2825 files_to_delete = []
dcf64d43 2826 if '__files_to_move' not in infodict:
2827 infodict['__files_to_move'] = {}
af819c21 2828 files_to_delete, infodict = pp.run(infodict)
5bfa4862 2829 if not files_to_delete:
dcf64d43 2830 return infodict
5bfa4862 2831
2832 if self.params.get('keepvideo', False):
2833 for f in files_to_delete:
dcf64d43 2834 infodict['__files_to_move'].setdefault(f, '')
5bfa4862 2835 else:
2836 for old_filename in set(files_to_delete):
2837 self.to_screen('Deleting original file %s (pass -k to keep)' % old_filename)
2838 try:
2839 os.remove(encodeFilename(old_filename))
2840 except (IOError, OSError):
2841 self.report_warning('Unable to remove downloaded original file')
dcf64d43 2842 if old_filename in infodict['__files_to_move']:
2843 del infodict['__files_to_move'][old_filename]
2844 return infodict
5bfa4862 2845
277d6ff5 2846 @staticmethod
2847 def post_extract(info_dict):
2848 def actual_post_extract(info_dict):
2849 if info_dict.get('_type') in ('playlist', 'multi_video'):
2850 for video_dict in info_dict.get('entries', {}):
b050d210 2851 actual_post_extract(video_dict or {})
277d6ff5 2852 return
2853
07cce701 2854 post_extractor = info_dict.get('__post_extractor') or (lambda: {})
4ec82a72 2855 extra = post_extractor().items()
2856 info_dict.update(extra)
07cce701 2857 info_dict.pop('__post_extractor', None)
277d6ff5 2858
4ec82a72 2859 original_infodict = info_dict.get('__original_infodict') or {}
2860 original_infodict.update(extra)
2861 original_infodict.pop('__post_extractor', None)
2862
b050d210 2863 actual_post_extract(info_dict or {})
277d6ff5 2864
56d868db 2865 def pre_process(self, ie_info, key='pre_process', files_to_move=None):
5bfa4862 2866 info = dict(ie_info)
56d868db 2867 info['__files_to_move'] = files_to_move or {}
2868 for pp in self._pps[key]:
dcf64d43 2869 info = self.run_pp(pp, info)
56d868db 2870 return info, info.pop('__files_to_move', None)
5bfa4862 2871
dcf64d43 2872 def post_process(self, filename, ie_info, files_to_move=None):
8222d8de
JMF
2873 """Run all the postprocessors on the given file."""
2874 info = dict(ie_info)
2875 info['filepath'] = filename
dcf64d43 2876 info['__files_to_move'] = files_to_move or {}
0202b52a 2877
56d868db 2878 for pp in ie_info.get('__postprocessors', []) + self._pps['post_process']:
dcf64d43 2879 info = self.run_pp(pp, info)
2880 info = self.run_pp(MoveFilesAfterDownloadPP(self), info)
2881 del info['__files_to_move']
56d868db 2882 for pp in self._pps['after_move']:
dcf64d43 2883 info = self.run_pp(pp, info)
23c1a667 2884 return info
c1c9a79c 2885
5db07df6 2886 def _make_archive_id(self, info_dict):
e9fef7ee
S
2887 video_id = info_dict.get('id')
2888 if not video_id:
2889 return
5db07df6
PH
2890 # Future-proof against any change in case
2891 # and backwards compatibility with prior versions
e9fef7ee 2892 extractor = info_dict.get('extractor_key') or info_dict.get('ie_key') # key in a playlist
7012b23c 2893 if extractor is None:
1211bb6d
S
2894 url = str_or_none(info_dict.get('url'))
2895 if not url:
2896 return
e9fef7ee
S
2897 # Try to find matching extractor for the URL and take its ie_key
2898 for ie in self._ies:
1211bb6d 2899 if ie.suitable(url):
e9fef7ee
S
2900 extractor = ie.ie_key()
2901 break
2902 else:
2903 return
d0757229 2904 return '%s %s' % (extractor.lower(), video_id)
5db07df6
PH
2905
2906 def in_download_archive(self, info_dict):
2907 fn = self.params.get('download_archive')
2908 if fn is None:
2909 return False
2910
2911 vid_id = self._make_archive_id(info_dict)
e9fef7ee 2912 if not vid_id:
7012b23c 2913 return False # Incomplete video information
5db07df6 2914
a45e8619 2915 return vid_id in self.archive
c1c9a79c
PH
2916
2917 def record_download_archive(self, info_dict):
2918 fn = self.params.get('download_archive')
2919 if fn is None:
2920 return
5db07df6
PH
2921 vid_id = self._make_archive_id(info_dict)
2922 assert vid_id
c1c9a79c 2923 with locked_file(fn, 'a', encoding='utf-8') as archive_file:
6febd1c1 2924 archive_file.write(vid_id + '\n')
a45e8619 2925 self.archive.add(vid_id)
dd82ffea 2926
8c51aa65 2927 @staticmethod
8abeeb94 2928 def format_resolution(format, default='unknown'):
fb04e403 2929 if format.get('vcodec') == 'none':
8326b00a 2930 if format.get('acodec') == 'none':
2931 return 'images'
fb04e403 2932 return 'audio only'
f49d89ee
PH
2933 if format.get('resolution') is not None:
2934 return format['resolution']
35615307
DA
2935 if format.get('width') and format.get('height'):
2936 res = '%dx%d' % (format['width'], format['height'])
2937 elif format.get('height'):
2938 res = '%sp' % format['height']
2939 elif format.get('width'):
388ae76b 2940 res = '%dx?' % format['width']
8c51aa65 2941 else:
8abeeb94 2942 res = default
8c51aa65
JMF
2943 return res
2944
c57f7757
PH
2945 def _format_note(self, fdict):
2946 res = ''
2947 if fdict.get('ext') in ['f4f', 'f4m']:
2948 res += '(unsupported) '
32f90364
PH
2949 if fdict.get('language'):
2950 if res:
2951 res += ' '
9016d76f 2952 res += '[%s] ' % fdict['language']
c57f7757
PH
2953 if fdict.get('format_note') is not None:
2954 res += fdict['format_note'] + ' '
2955 if fdict.get('tbr') is not None:
2956 res += '%4dk ' % fdict['tbr']
2957 if fdict.get('container') is not None:
2958 if res:
2959 res += ', '
2960 res += '%s container' % fdict['container']
3089bc74
S
2961 if (fdict.get('vcodec') is not None
2962 and fdict.get('vcodec') != 'none'):
c57f7757
PH
2963 if res:
2964 res += ', '
2965 res += fdict['vcodec']
91c7271a 2966 if fdict.get('vbr') is not None:
c57f7757
PH
2967 res += '@'
2968 elif fdict.get('vbr') is not None and fdict.get('abr') is not None:
2969 res += 'video@'
2970 if fdict.get('vbr') is not None:
2971 res += '%4dk' % fdict['vbr']
fbb21cf5 2972 if fdict.get('fps') is not None:
5d583bdf
S
2973 if res:
2974 res += ', '
2975 res += '%sfps' % fdict['fps']
c57f7757
PH
2976 if fdict.get('acodec') is not None:
2977 if res:
2978 res += ', '
2979 if fdict['acodec'] == 'none':
2980 res += 'video only'
2981 else:
2982 res += '%-5s' % fdict['acodec']
2983 elif fdict.get('abr') is not None:
2984 if res:
2985 res += ', '
2986 res += 'audio'
2987 if fdict.get('abr') is not None:
2988 res += '@%3dk' % fdict['abr']
2989 if fdict.get('asr') is not None:
2990 res += ' (%5dHz)' % fdict['asr']
2991 if fdict.get('filesize') is not None:
2992 if res:
2993 res += ', '
2994 res += format_bytes(fdict['filesize'])
9732d77e
PH
2995 elif fdict.get('filesize_approx') is not None:
2996 if res:
2997 res += ', '
2998 res += '~' + format_bytes(fdict['filesize_approx'])
c57f7757 2999 return res
91c7271a 3000
76d321f6 3001 def _format_note_table(self, f):
3002 def join_fields(*vargs):
3003 return ', '.join((val for val in vargs if val != ''))
3004
3005 return join_fields(
3006 'UNSUPPORTED' if f.get('ext') in ('f4f', 'f4m') else '',
3007 format_field(f, 'language', '[%s]'),
3008 format_field(f, 'format_note'),
3009 format_field(f, 'container', ignore=(None, f.get('ext'))),
3010 format_field(f, 'asr', '%5dHz'))
3011
c57f7757 3012 def list_formats(self, info_dict):
94badb25 3013 formats = info_dict.get('formats', [info_dict])
53ed7066 3014 new_format = (
3015 'list-formats' not in self.params.get('compat_opts', [])
3016 and self.params.get('list_formats_as_table', True) is not False)
76d321f6 3017 if new_format:
3018 table = [
3019 [
3020 format_field(f, 'format_id'),
3021 format_field(f, 'ext'),
3022 self.format_resolution(f),
3023 format_field(f, 'fps', '%d'),
3024 '|',
3025 format_field(f, 'filesize', ' %s', func=format_bytes) + format_field(f, 'filesize_approx', '~%s', func=format_bytes),
3026 format_field(f, 'tbr', '%4dk'),
52a8a1e1 3027 shorten_protocol_name(f.get('protocol', '').replace("native", "n")),
76d321f6 3028 '|',
3029 format_field(f, 'vcodec', default='unknown').replace('none', ''),
3030 format_field(f, 'vbr', '%4dk'),
3031 format_field(f, 'acodec', default='unknown').replace('none', ''),
3032 format_field(f, 'abr', '%3dk'),
3033 format_field(f, 'asr', '%5dHz'),
3034 self._format_note_table(f)]
3035 for f in formats
3036 if f.get('preference') is None or f['preference'] >= -1000]
3037 header_line = ['ID', 'EXT', 'RESOLUTION', 'FPS', '|', ' FILESIZE', ' TBR', 'PROTO',
3038 '|', 'VCODEC', ' VBR', 'ACODEC', ' ABR', ' ASR', 'NOTE']
3039 else:
3040 table = [
3041 [
3042 format_field(f, 'format_id'),
3043 format_field(f, 'ext'),
3044 self.format_resolution(f),
3045 self._format_note(f)]
3046 for f in formats
3047 if f.get('preference') is None or f['preference'] >= -1000]
3048 header_line = ['format code', 'extension', 'resolution', 'note']
57dd9a8f 3049
cfb56d1a 3050 self.to_screen(
76d321f6 3051 '[info] Available formats for %s:\n%s' % (info_dict['id'], render_table(
3052 header_line,
3053 table,
3054 delim=new_format,
3055 extraGap=(0 if new_format else 1),
3056 hideEmpty=new_format)))
cfb56d1a
PH
3057
3058 def list_thumbnails(self, info_dict):
b0249bca 3059 thumbnails = list(info_dict.get('thumbnails'))
cfb56d1a 3060 if not thumbnails:
b7b72db9 3061 self.to_screen('[info] No thumbnails present for %s' % info_dict['id'])
3062 return
cfb56d1a
PH
3063
3064 self.to_screen(
3065 '[info] Thumbnails for %s:' % info_dict['id'])
3066 self.to_screen(render_table(
3067 ['ID', 'width', 'height', 'URL'],
3068 [[t['id'], t.get('width', 'unknown'), t.get('height', 'unknown'), t['url']] for t in thumbnails]))
dca08720 3069
360e1ca5 3070 def list_subtitles(self, video_id, subtitles, name='subtitles'):
a504ced0 3071 if not subtitles:
360e1ca5 3072 self.to_screen('%s has no %s' % (video_id, name))
a504ced0 3073 return
a504ced0 3074 self.to_screen(
edab9dbf 3075 'Available %s for %s:' % (name, video_id))
2412044c 3076
3077 def _row(lang, formats):
49c258e1 3078 exts, names = zip(*((f['ext'], f.get('name') or 'unknown') for f in reversed(formats)))
2412044c 3079 if len(set(names)) == 1:
7aee40c1 3080 names = [] if names[0] == 'unknown' else names[:1]
2412044c 3081 return [lang, ', '.join(names), ', '.join(exts)]
3082
edab9dbf 3083 self.to_screen(render_table(
2412044c 3084 ['Language', 'Name', 'Formats'],
3085 [_row(lang, formats) for lang, formats in subtitles.items()],
3086 hideEmpty=True))
a504ced0 3087
dca08720
PH
3088 def urlopen(self, req):
3089 """ Start an HTTP download """
82d8a8b6 3090 if isinstance(req, compat_basestring):
67dda517 3091 req = sanitized_Request(req)
19a41fc6 3092 return self._opener.open(req, timeout=self._socket_timeout)
dca08720
PH
3093
3094 def print_debug_header(self):
3095 if not self.params.get('verbose'):
3096 return
62fec3b2 3097
4192b51c 3098 if type('') is not compat_str:
067aa17e 3099 # Python 2.6 on SLES11 SP1 (https://github.com/ytdl-org/youtube-dl/issues/3326)
4192b51c
PH
3100 self.report_warning(
3101 'Your Python is broken! Update to a newer and supported version')
3102
c6afed48
PH
3103 stdout_encoding = getattr(
3104 sys.stdout, 'encoding', 'missing (%s)' % type(sys.stdout).__name__)
b0472057 3105 encoding_str = (
734f90bb
PH
3106 '[debug] Encodings: locale %s, fs %s, out %s, pref %s\n' % (
3107 locale.getpreferredencoding(),
3108 sys.getfilesystemencoding(),
c6afed48 3109 stdout_encoding,
b0472057 3110 self.get_encoding()))
4192b51c 3111 write_string(encoding_str, encoding=None)
734f90bb 3112
e5813e53 3113 source = (
3114 '(exe)' if hasattr(sys, 'frozen')
3115 else '(zip)' if isinstance(globals().get('__loader__'), zipimporter)
3116 else '(source)' if os.path.basename(sys.argv[0]) == '__main__.py'
3117 else '')
3118 self._write_string('[debug] yt-dlp version %s %s\n' % (__version__, source))
e0986e31 3119 if _LAZY_LOADER:
f74980cb 3120 self._write_string('[debug] Lazy loading extractors enabled\n')
3121 if _PLUGIN_CLASSES:
3122 self._write_string(
3123 '[debug] Plugin Extractors: %s\n' % [ie.ie_key() for ie in _PLUGIN_CLASSES])
53ed7066 3124 if self.params.get('compat_opts'):
3125 self._write_string(
3126 '[debug] Compatibility options: %s\n' % ', '.join(self.params.get('compat_opts')))
dca08720
PH
3127 try:
3128 sp = subprocess.Popen(
3129 ['git', 'rev-parse', '--short', 'HEAD'],
3130 stdout=subprocess.PIPE, stderr=subprocess.PIPE,
3131 cwd=os.path.dirname(os.path.abspath(__file__)))
f5b1bca9 3132 out, err = process_communicate_or_kill(sp)
dca08720
PH
3133 out = out.decode().strip()
3134 if re.match('[0-9a-f]+', out):
f74980cb 3135 self._write_string('[debug] Git HEAD: %s\n' % out)
70a1165b 3136 except Exception:
dca08720
PH
3137 try:
3138 sys.exc_clear()
70a1165b 3139 except Exception:
dca08720 3140 pass
b300cda4
S
3141
3142 def python_implementation():
3143 impl_name = platform.python_implementation()
3144 if impl_name == 'PyPy' and hasattr(sys, 'pypy_version_info'):
3145 return impl_name + ' version %d.%d.%d' % sys.pypy_version_info[:3]
3146 return impl_name
3147
e5813e53 3148 self._write_string('[debug] Python version %s (%s %s) - %s\n' % (
3149 platform.python_version(),
3150 python_implementation(),
3151 platform.architecture()[0],
b300cda4 3152 platform_name()))
d28b5171 3153
73fac4e9 3154 exe_versions = FFmpegPostProcessor.get_versions(self)
4c83c967 3155 exe_versions['rtmpdump'] = rtmpdump_version()
feee8d32 3156 exe_versions['phantomjs'] = PhantomJSwrapper._version()
d28b5171
PH
3157 exe_str = ', '.join(
3158 '%s %s' % (exe, v)
3159 for exe, v in sorted(exe_versions.items())
3160 if v
3161 )
3162 if not exe_str:
3163 exe_str = 'none'
3164 self._write_string('[debug] exe versions: %s\n' % exe_str)
dca08720
PH
3165
3166 proxy_map = {}
3167 for handler in self._opener.handlers:
3168 if hasattr(handler, 'proxies'):
3169 proxy_map.update(handler.proxies)
734f90bb 3170 self._write_string('[debug] Proxy map: ' + compat_str(proxy_map) + '\n')
dca08720 3171
58b1f00d
PH
3172 if self.params.get('call_home', False):
3173 ipaddr = self.urlopen('https://yt-dl.org/ip').read().decode('utf-8')
3174 self._write_string('[debug] Public IP address: %s\n' % ipaddr)
f5546c0b 3175 return
58b1f00d
PH
3176 latest_version = self.urlopen(
3177 'https://yt-dl.org/latest/version').read().decode('utf-8')
3178 if version_tuple(latest_version) > version_tuple(__version__):
3179 self.report_warning(
3180 'You are using an outdated version (newest version: %s)! '
3181 'See https://yt-dl.org/update if you need help updating.' %
3182 latest_version)
3183
e344693b 3184 def _setup_opener(self):
6ad14cab 3185 timeout_val = self.params.get('socket_timeout')
19a41fc6 3186 self._socket_timeout = 600 if timeout_val is None else float(timeout_val)
6ad14cab 3187
dca08720
PH
3188 opts_cookiefile = self.params.get('cookiefile')
3189 opts_proxy = self.params.get('proxy')
3190
3191 if opts_cookiefile is None:
3192 self.cookiejar = compat_cookiejar.CookieJar()
3193 else:
590bc6f6 3194 opts_cookiefile = expand_path(opts_cookiefile)
1bab3437 3195 self.cookiejar = YoutubeDLCookieJar(opts_cookiefile)
dca08720 3196 if os.access(opts_cookiefile, os.R_OK):
1d88b3e6 3197 self.cookiejar.load(ignore_discard=True, ignore_expires=True)
dca08720 3198
6a3f4c3f 3199 cookie_processor = YoutubeDLCookieProcessor(self.cookiejar)
dca08720
PH
3200 if opts_proxy is not None:
3201 if opts_proxy == '':
3202 proxies = {}
3203 else:
3204 proxies = {'http': opts_proxy, 'https': opts_proxy}
3205 else:
3206 proxies = compat_urllib_request.getproxies()
067aa17e 3207 # Set HTTPS proxy to HTTP one if given (https://github.com/ytdl-org/youtube-dl/issues/805)
dca08720
PH
3208 if 'http' in proxies and 'https' not in proxies:
3209 proxies['https'] = proxies['http']
91410c9b 3210 proxy_handler = PerRequestProxyHandler(proxies)
a0ddb8a2
PH
3211
3212 debuglevel = 1 if self.params.get('debug_printtraffic') else 0
be4a824d
PH
3213 https_handler = make_HTTPS_handler(self.params, debuglevel=debuglevel)
3214 ydlh = YoutubeDLHandler(self.params, debuglevel=debuglevel)
fca6dba8 3215 redirect_handler = YoutubeDLRedirectHandler()
8b172c2e 3216 data_handler = compat_urllib_request_DataHandler()
6240b0a2
JMF
3217
3218 # When passing our own FileHandler instance, build_opener won't add the
3219 # default FileHandler and allows us to disable the file protocol, which
3220 # can be used for malicious purposes (see
067aa17e 3221 # https://github.com/ytdl-org/youtube-dl/issues/8227)
6240b0a2
JMF
3222 file_handler = compat_urllib_request.FileHandler()
3223
3224 def file_open(*args, **kwargs):
7a5c1cfe 3225 raise compat_urllib_error.URLError('file:// scheme is explicitly disabled in yt-dlp for security reasons')
6240b0a2
JMF
3226 file_handler.file_open = file_open
3227
3228 opener = compat_urllib_request.build_opener(
fca6dba8 3229 proxy_handler, https_handler, cookie_processor, ydlh, redirect_handler, data_handler, file_handler)
2461f79d 3230
dca08720
PH
3231 # Delete the default user-agent header, which would otherwise apply in
3232 # cases where our custom HTTP handler doesn't come into play
067aa17e 3233 # (See https://github.com/ytdl-org/youtube-dl/issues/1309 for details)
dca08720
PH
3234 opener.addheaders = []
3235 self._opener = opener
62fec3b2
PH
3236
3237 def encode(self, s):
3238 if isinstance(s, bytes):
3239 return s # Already encoded
3240
3241 try:
3242 return s.encode(self.get_encoding())
3243 except UnicodeEncodeError as err:
3244 err.reason = err.reason + '. Check your system encoding configuration or use the --encoding option.'
3245 raise
3246
3247 def get_encoding(self):
3248 encoding = self.params.get('encoding')
3249 if encoding is None:
3250 encoding = preferredencoding()
3251 return encoding
ec82d85a 3252
de6000d9 3253 def _write_thumbnails(self, info_dict, filename): # return the extensions
6c4fd172 3254 write_all = self.params.get('write_all_thumbnails', False)
3255 thumbnails = []
3256 if write_all or self.params.get('writethumbnail', False):
0202b52a 3257 thumbnails = info_dict.get('thumbnails') or []
6c4fd172 3258 multiple = write_all and len(thumbnails) > 1
ec82d85a 3259
0202b52a 3260 ret = []
981052c9 3261 for t in thumbnails[::-1]:
ec82d85a 3262 thumb_ext = determine_ext(t['url'], 'jpg')
6c4fd172 3263 suffix = '%s.' % t['id'] if multiple else ''
3264 thumb_display_id = '%s ' % t['id'] if multiple else ''
885cc0b7 3265 thumb_filename = replace_extension(filename, suffix + thumb_ext, info_dict.get('ext'))
ec82d85a 3266
0c3d0f51 3267 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(thumb_filename)):
de6000d9 3268 ret.append(suffix + thumb_ext)
8ba87148 3269 t['filepath'] = thumb_filename
ec82d85a
PH
3270 self.to_screen('[%s] %s: Thumbnail %sis already present' %
3271 (info_dict['extractor'], info_dict['id'], thumb_display_id))
3272 else:
5ef7d9bd 3273 self.to_screen('[%s] %s: Downloading thumbnail %s ...' %
ec82d85a
PH
3274 (info_dict['extractor'], info_dict['id'], thumb_display_id))
3275 try:
3276 uf = self.urlopen(t['url'])
d3d89c32 3277 with open(encodeFilename(thumb_filename), 'wb') as thumbf:
ec82d85a 3278 shutil.copyfileobj(uf, thumbf)
de6000d9 3279 ret.append(suffix + thumb_ext)
ec82d85a
PH
3280 self.to_screen('[%s] %s: Writing thumbnail %sto: %s' %
3281 (info_dict['extractor'], info_dict['id'], thumb_display_id, thumb_filename))
885cc0b7 3282 t['filepath'] = thumb_filename
3158150c 3283 except network_exceptions as err:
ec82d85a 3284 self.report_warning('Unable to download thumbnail "%s": %s' %
9b9c5355 3285 (t['url'], error_to_compat_str(err)))
6c4fd172 3286 if ret and not write_all:
3287 break
0202b52a 3288 return ret