]> jfr.im git - yt-dlp.git/blame - yt_dlp/YoutubeDL.py
[BravoTV] Improve metadata extraction (#483)
[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".
3acf6d38 395 The following options do not work when used through the API:
396 filename, abort-on-error, multistreams, no-live-chat,
397 no-playlist-metafiles. Refer __init__.py for their implementation
fe7e0c98 398
8222d8de 399 The following parameters are not used by YoutubeDL itself, they are used by
7a5c1cfe 400 the downloader (see yt_dlp/downloader/common.py):
51d9739f 401 nopart, updatetime, buffersize, ratelimit, throttledratelimit, min_filesize,
402 max_filesize, test, noresizebuffer, retries, continuedl, noprogress, consoletitle,
403 xattr_set_filesize, external_downloader_args, hls_use_mpegts, http_chunk_size.
76b1bd67
JMF
404
405 The following options are used by the post processors:
d4a24f40 406 prefer_ffmpeg: If False, use avconv instead of ffmpeg if both are available,
e4172ac9 407 otherwise prefer ffmpeg. (avconv support is deprecated)
c0b7d117
S
408 ffmpeg_location: Location of the ffmpeg/avconv binary; either the path
409 to the binary or its containing directory.
43820c03 410 postprocessor_args: A dictionary of postprocessor/executable keys (in lower case)
411 and a list of additional command-line arguments for the
412 postprocessor/executable. The dict can also have "PP+EXE" keys
413 which are used when the given exe is used by the given PP.
414 Use 'default' as the name for arguments to passed to all PP
e409895f 415
416 The following options are used by the extractors:
62bff2c1 417 extractor_retries: Number of times to retry for known errors
418 dynamic_mpd: Whether to process dynamic DASH manifests (default: True)
e409895f 419 hls_split_discontinuity: Split HLS playlists to different formats at
62bff2c1 420 discontinuities such as ad breaks (default: False)
5d3a0e79 421 extractor_args: A dictionary of arguments to be passed to the extractors.
422 See "EXTRACTOR ARGUMENTS" for details.
423 Eg: {'youtube': {'skip': ['dash', 'hls']}}
424 youtube_include_dash_manifest: Deprecated - Use extractor_args instead.
425 If True (default), DASH manifests and related
62bff2c1 426 data will be downloaded and processed by extractor.
427 You can reduce network I/O by disabling it if you don't
428 care about DASH. (only for youtube)
5d3a0e79 429 youtube_include_hls_manifest: Deprecated - Use extractor_args instead.
430 If True (default), HLS manifests and related
62bff2c1 431 data will be downloaded and processed by extractor.
432 You can reduce network I/O by disabling it if you don't
433 care about HLS. (only for youtube)
8222d8de
JMF
434 """
435
c9969434
S
436 _NUMERIC_FIELDS = set((
437 'width', 'height', 'tbr', 'abr', 'asr', 'vbr', 'fps', 'filesize', 'filesize_approx',
438 'timestamp', 'upload_year', 'upload_month', 'upload_day',
439 'duration', 'view_count', 'like_count', 'dislike_count', 'repost_count',
440 'average_rating', 'comment_count', 'age_limit',
441 'start_time', 'end_time',
442 'chapter_number', 'season_number', 'episode_number',
443 'track_number', 'disc_number', 'release_year',
444 'playlist_index',
445 ))
446
8222d8de
JMF
447 params = None
448 _ies = []
56d868db 449 _pps = {'pre_process': [], 'before_dl': [], 'after_move': [], 'post_process': []}
0202b52a 450 __prepare_filename_warned = False
1cf376f5 451 _first_webpage_request = True
8222d8de
JMF
452 _download_retcode = None
453 _num_downloads = None
30a074c2 454 _playlist_level = 0
455 _playlist_urls = set()
8222d8de
JMF
456 _screen_file = None
457
3511266b 458 def __init__(self, params=None, auto_init=True):
8222d8de 459 """Create a FileDownloader object with the given options."""
e9f9a10f
JMF
460 if params is None:
461 params = {}
8222d8de 462 self._ies = []
56c73665 463 self._ies_instances = {}
56d868db 464 self._pps = {'pre_process': [], 'before_dl': [], 'after_move': [], 'post_process': []}
0202b52a 465 self.__prepare_filename_warned = False
1cf376f5 466 self._first_webpage_request = True
ab8e5e51 467 self._post_hooks = []
933605d7 468 self._progress_hooks = []
8222d8de
JMF
469 self._download_retcode = 0
470 self._num_downloads = 0
471 self._screen_file = [sys.stdout, sys.stderr][params.get('logtostderr', False)]
0783b09b 472 self._err_file = sys.stderr
4abf617b
S
473 self.params = {
474 # Default parameters
475 'nocheckcertificate': False,
476 }
477 self.params.update(params)
a0e07d31 478 self.cache = Cache(self)
34308b30 479
a61f4b28 480 if sys.version_info < (3, 6):
481 self.report_warning(
0181adef 482 'Python version %d.%d is not supported! Please update to Python 3.6 or above' % sys.version_info[:2])
a61f4b28 483
be5df5ee
S
484 def check_deprecated(param, option, suggestion):
485 if self.params.get(param) is not None:
53ed7066 486 self.report_warning('%s is deprecated. Use %s instead' % (option, suggestion))
be5df5ee
S
487 return True
488 return False
489
490 if check_deprecated('cn_verification_proxy', '--cn-verification-proxy', '--geo-verification-proxy'):
38cce791
YCH
491 if self.params.get('geo_verification_proxy') is None:
492 self.params['geo_verification_proxy'] = self.params['cn_verification_proxy']
493
0d1bb027 494 check_deprecated('autonumber', '--auto-number', '-o "%(autonumber)s-%(title)s.%(ext)s"')
495 check_deprecated('usetitle', '--title', '-o "%(title)s-%(id)s.%(ext)s"')
53ed7066 496 check_deprecated('useid', '--id', '-o "%(id)s.%(ext)s"')
0d1bb027 497
498 for msg in self.params.get('warnings', []):
499 self.report_warning(msg)
500
6b591b29 501 if self.params.get('final_ext'):
502 if self.params.get('merge_output_format'):
503 self.report_warning('--merge-output-format will be ignored since --remux-video or --recode-video is given')
504 self.params['merge_output_format'] = self.params['final_ext']
505
b9d973be 506 if 'overwrites' in self.params and self.params['overwrites'] is None:
507 del self.params['overwrites']
508
0783b09b 509 if params.get('bidi_workaround', False):
1c088fa8
PH
510 try:
511 import pty
512 master, slave = pty.openpty()
003c69a8 513 width = compat_get_terminal_size().columns
1c088fa8
PH
514 if width is None:
515 width_args = []
516 else:
517 width_args = ['-w', str(width)]
5d681e96 518 sp_kwargs = dict(
1c088fa8
PH
519 stdin=subprocess.PIPE,
520 stdout=slave,
521 stderr=self._err_file)
5d681e96
PH
522 try:
523 self._output_process = subprocess.Popen(
524 ['bidiv'] + width_args, **sp_kwargs
525 )
526 except OSError:
5d681e96
PH
527 self._output_process = subprocess.Popen(
528 ['fribidi', '-c', 'UTF-8'] + width_args, **sp_kwargs)
529 self._output_channel = os.fdopen(master, 'rb')
1c088fa8 530 except OSError as ose:
66e7ace1 531 if ose.errno == errno.ENOENT:
6febd1c1 532 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
533 else:
534 raise
0783b09b 535
3089bc74
S
536 if (sys.platform != 'win32'
537 and sys.getfilesystemencoding() in ['ascii', 'ANSI_X3.4-1968']
538 and not params.get('restrictfilenames', False)):
e9137224 539 # Unicode filesystem API will throw errors (#1474, #13027)
34308b30 540 self.report_warning(
6febd1c1 541 'Assuming --restrict-filenames since file system encoding '
1b725173 542 'cannot encode all characters. '
6febd1c1 543 'Set the LC_ALL environment variable to fix this.')
4a98cdbf 544 self.params['restrictfilenames'] = True
34308b30 545
de6000d9 546 self.outtmpl_dict = self.parse_outtmpl()
486dd09e 547
187986a8 548 # Creating format selector here allows us to catch syntax errors before the extraction
549 self.format_selector = (
550 None if self.params.get('format') is None
551 else self.build_format_selector(self.params['format']))
552
dca08720
PH
553 self._setup_opener()
554
4cd0a709 555 """Preload the archive, if any is specified"""
556 def preload_download_archive(fn):
557 if fn is None:
558 return False
0760b0a7 559 self.write_debug('Loading archive file %r\n' % fn)
4cd0a709 560 try:
561 with locked_file(fn, 'r', encoding='utf-8') as archive_file:
562 for line in archive_file:
563 self.archive.add(line.strip())
564 except IOError as ioe:
565 if ioe.errno != errno.ENOENT:
566 raise
567 return False
568 return True
569
570 self.archive = set()
571 preload_download_archive(self.params.get('download_archive'))
572
3511266b
PH
573 if auto_init:
574 self.print_debug_header()
575 self.add_default_info_extractors()
576
4f026faf 577 for pp_def_raw in self.params.get('postprocessors', []):
4f026faf 578 pp_def = dict(pp_def_raw)
fd7cfb64 579 when = pp_def.pop('when', 'post_process')
580 pp_class = get_postprocessor(pp_def.pop('key'))
4f026faf 581 pp = pp_class(self, **compat_kwargs(pp_def))
5bfa4862 582 self.add_post_processor(pp, when=when)
4f026faf 583
ab8e5e51
AM
584 for ph in self.params.get('post_hooks', []):
585 self.add_post_hook(ph)
586
71b640cc
PH
587 for ph in self.params.get('progress_hooks', []):
588 self.add_progress_hook(ph)
589
51fb4995
YCH
590 register_socks_protocols()
591
7d4111ed
PH
592 def warn_if_short_id(self, argv):
593 # short YouTube ID starting with dash?
594 idxs = [
595 i for i, a in enumerate(argv)
596 if re.match(r'^-[0-9A-Za-z_-]{10}$', a)]
597 if idxs:
598 correct_argv = (
7a5c1cfe 599 ['yt-dlp']
3089bc74
S
600 + [a for i, a in enumerate(argv) if i not in idxs]
601 + ['--'] + [argv[i] for i in idxs]
7d4111ed
PH
602 )
603 self.report_warning(
604 'Long argument string detected. '
605 'Use -- to separate parameters and URLs, like this:\n%s\n' %
606 args_to_str(correct_argv))
607
8222d8de
JMF
608 def add_info_extractor(self, ie):
609 """Add an InfoExtractor object to the end of the list."""
610 self._ies.append(ie)
e52d7f85
JMF
611 if not isinstance(ie, type):
612 self._ies_instances[ie.ie_key()] = ie
613 ie.set_downloader(self)
8222d8de 614
56c73665
JMF
615 def get_info_extractor(self, ie_key):
616 """
617 Get an instance of an IE with name ie_key, it will try to get one from
618 the _ies list, if there's no instance it will create a new one and add
619 it to the extractor list.
620 """
621 ie = self._ies_instances.get(ie_key)
622 if ie is None:
623 ie = get_info_extractor(ie_key)()
624 self.add_info_extractor(ie)
625 return ie
626
023fa8c4
JMF
627 def add_default_info_extractors(self):
628 """
629 Add the InfoExtractors returned by gen_extractors to the end of the list
630 """
e52d7f85 631 for ie in gen_extractor_classes():
023fa8c4
JMF
632 self.add_info_extractor(ie)
633
56d868db 634 def add_post_processor(self, pp, when='post_process'):
8222d8de 635 """Add a PostProcessor object to the end of the chain."""
5bfa4862 636 self._pps[when].append(pp)
8222d8de
JMF
637 pp.set_downloader(self)
638
ab8e5e51
AM
639 def add_post_hook(self, ph):
640 """Add the post hook"""
641 self._post_hooks.append(ph)
642
933605d7
JMF
643 def add_progress_hook(self, ph):
644 """Add the progress hook (currently only for the file downloader)"""
645 self._progress_hooks.append(ph)
8ab470f1 646
1c088fa8 647 def _bidi_workaround(self, message):
5d681e96 648 if not hasattr(self, '_output_channel'):
1c088fa8
PH
649 return message
650
5d681e96 651 assert hasattr(self, '_output_process')
11b85ce6 652 assert isinstance(message, compat_str)
6febd1c1
PH
653 line_count = message.count('\n') + 1
654 self._output_process.stdin.write((message + '\n').encode('utf-8'))
5d681e96 655 self._output_process.stdin.flush()
6febd1c1 656 res = ''.join(self._output_channel.readline().decode('utf-8')
9e1a5b84 657 for _ in range(line_count))
6febd1c1 658 return res[:-len('\n')]
1c088fa8 659
734f90bb 660 def _write_string(self, s, out=None):
b58ddb32 661 write_string(s, out=out, encoding=self.params.get('encoding'))
734f90bb 662
848887eb 663 def to_stdout(self, message, skip_eol=False, quiet=False):
0760b0a7 664 """Print message to stdout"""
8bf9319e 665 if self.params.get('logger'):
43afe285 666 self.params['logger'].debug(message)
835a1478 667 elif not quiet or self.params.get('verbose'):
668 self._write_string(
669 '%s%s' % (self._bidi_workaround(message), ('' if skip_eol else '\n')),
670 self._err_file if quiet else self._screen_file)
8222d8de
JMF
671
672 def to_stderr(self, message):
0760b0a7 673 """Print message to stderr"""
11b85ce6 674 assert isinstance(message, compat_str)
8bf9319e 675 if self.params.get('logger'):
43afe285
IB
676 self.params['logger'].error(message)
677 else:
835a1478 678 self._write_string('%s\n' % self._bidi_workaround(message), self._err_file)
8222d8de 679
1e5b9a95
PH
680 def to_console_title(self, message):
681 if not self.params.get('consoletitle', False):
682 return
4bede0d8
C
683 if compat_os_name == 'nt':
684 if ctypes.windll.kernel32.GetConsoleWindow():
685 # c_wchar_p() might not be necessary if `message` is
686 # already of type unicode()
687 ctypes.windll.kernel32.SetConsoleTitleW(ctypes.c_wchar_p(message))
1e5b9a95 688 elif 'TERM' in os.environ:
b46696bd 689 self._write_string('\033]0;%s\007' % message, self._screen_file)
1e5b9a95 690
bdde425c
PH
691 def save_console_title(self):
692 if not self.params.get('consoletitle', False):
693 return
94c3442e
S
694 if self.params.get('simulate', False):
695 return
4bede0d8 696 if compat_os_name != 'nt' and 'TERM' in os.environ:
efd6c574 697 # Save the title on stack
734f90bb 698 self._write_string('\033[22;0t', self._screen_file)
bdde425c
PH
699
700 def restore_console_title(self):
701 if not self.params.get('consoletitle', False):
702 return
94c3442e
S
703 if self.params.get('simulate', False):
704 return
4bede0d8 705 if compat_os_name != 'nt' and 'TERM' in os.environ:
efd6c574 706 # Restore the title from stack
734f90bb 707 self._write_string('\033[23;0t', self._screen_file)
bdde425c
PH
708
709 def __enter__(self):
710 self.save_console_title()
711 return self
712
713 def __exit__(self, *args):
714 self.restore_console_title()
f89197d7 715
dca08720 716 if self.params.get('cookiefile') is not None:
1bab3437 717 self.cookiejar.save(ignore_discard=True, ignore_expires=True)
bdde425c 718
8222d8de
JMF
719 def trouble(self, message=None, tb=None):
720 """Determine action to take when a download problem appears.
721
722 Depending on if the downloader has been configured to ignore
723 download errors or not, this method may throw an exception or
724 not when errors are found, after printing the message.
725
726 tb, if given, is additional traceback information.
727 """
728 if message is not None:
729 self.to_stderr(message)
730 if self.params.get('verbose'):
731 if tb is None:
732 if sys.exc_info()[0]: # if .trouble has been called from an except block
6febd1c1 733 tb = ''
8222d8de 734 if hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
6febd1c1 735 tb += ''.join(traceback.format_exception(*sys.exc_info()[1].exc_info))
c0384f22 736 tb += encode_compat_str(traceback.format_exc())
8222d8de
JMF
737 else:
738 tb_data = traceback.format_list(traceback.extract_stack())
6febd1c1 739 tb = ''.join(tb_data)
c19bc311 740 if tb:
741 self.to_stderr(tb)
8222d8de
JMF
742 if not self.params.get('ignoreerrors', False):
743 if sys.exc_info()[0] and hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
744 exc_info = sys.exc_info()[1].exc_info
745 else:
746 exc_info = sys.exc_info()
747 raise DownloadError(message, exc_info)
748 self._download_retcode = 1
749
0760b0a7 750 def to_screen(self, message, skip_eol=False):
751 """Print message to stdout if not in quiet mode"""
752 self.to_stdout(
753 message, skip_eol, quiet=self.params.get('quiet', False))
754
8222d8de
JMF
755 def report_warning(self, message):
756 '''
757 Print the message to stderr, it will be prefixed with 'WARNING:'
758 If stderr is a tty file the 'WARNING:' will be colored
759 '''
6d07ce01
JMF
760 if self.params.get('logger') is not None:
761 self.params['logger'].warning(message)
8222d8de 762 else:
ad8915b7
PH
763 if self.params.get('no_warnings'):
764 return
e9c0cdd3 765 if not self.params.get('no_color') and self._err_file.isatty() and compat_os_name != 'nt':
6d07ce01
JMF
766 _msg_header = '\033[0;33mWARNING:\033[0m'
767 else:
768 _msg_header = 'WARNING:'
769 warning_message = '%s %s' % (_msg_header, message)
770 self.to_stderr(warning_message)
8222d8de
JMF
771
772 def report_error(self, message, tb=None):
773 '''
774 Do the same as trouble, but prefixes the message with 'ERROR:', colored
775 in red if stderr is a tty file.
776 '''
e9c0cdd3 777 if not self.params.get('no_color') and self._err_file.isatty() and compat_os_name != 'nt':
6febd1c1 778 _msg_header = '\033[0;31mERROR:\033[0m'
8222d8de 779 else:
6febd1c1
PH
780 _msg_header = 'ERROR:'
781 error_message = '%s %s' % (_msg_header, message)
8222d8de
JMF
782 self.trouble(error_message, tb)
783
0760b0a7 784 def write_debug(self, message):
785 '''Log debug message or Print message to stderr'''
786 if not self.params.get('verbose', False):
787 return
788 message = '[debug] %s' % message
789 if self.params.get('logger'):
790 self.params['logger'].debug(message)
791 else:
792 self._write_string('%s\n' % message)
793
8222d8de
JMF
794 def report_file_already_downloaded(self, file_name):
795 """Report file has already been fully downloaded."""
796 try:
6febd1c1 797 self.to_screen('[download] %s has already been downloaded' % file_name)
ce02ed60 798 except UnicodeEncodeError:
6febd1c1 799 self.to_screen('[download] The file has already been downloaded')
8222d8de 800
0c3d0f51 801 def report_file_delete(self, file_name):
802 """Report that existing file will be deleted."""
803 try:
c25228e5 804 self.to_screen('Deleting existing file %s' % file_name)
0c3d0f51 805 except UnicodeEncodeError:
c25228e5 806 self.to_screen('Deleting existing file')
0c3d0f51 807
de6000d9 808 def parse_outtmpl(self):
809 outtmpl_dict = self.params.get('outtmpl', {})
810 if not isinstance(outtmpl_dict, dict):
811 outtmpl_dict = {'default': outtmpl_dict}
812 outtmpl_dict.update({
813 k: v for k, v in DEFAULT_OUTTMPL.items()
814 if not outtmpl_dict.get(k)})
815 for key, val in outtmpl_dict.items():
816 if isinstance(val, bytes):
817 self.report_warning(
818 'Parameter outtmpl is bytes, but should be a unicode string. '
819 'Put from __future__ import unicode_literals at the top of your code file or consider switching to Python 3.x.')
820 return outtmpl_dict
821
21cd8fae 822 def get_output_path(self, dir_type='', filename=None):
823 paths = self.params.get('paths', {})
824 assert isinstance(paths, dict)
825 path = os.path.join(
826 expand_path(paths.get('home', '').strip()),
827 expand_path(paths.get(dir_type, '').strip()) if dir_type else '',
828 filename or '')
829
830 # Temporary fix for #4787
831 # 'Treat' all problem characters by passing filename through preferredencoding
832 # to workaround encoding issues with subprocess on python2 @ Windows
833 if sys.version_info < (3, 0) and sys.platform == 'win32':
834 path = encodeFilename(path, True).decode(preferredencoding())
835 return sanitize_path(path, force=self.params.get('windowsfilenames'))
836
76a264ac 837 @staticmethod
838 def validate_outtmpl(tmpl):
839 ''' @return None or Exception object '''
840 try:
841 re.sub(
842 STR_FORMAT_RE.format(''),
843 lambda mobj: ('%' if not mobj.group('has_key') else '') + mobj.group(0),
844 tmpl
845 ) % collections.defaultdict(int)
846 return None
847 except ValueError as err:
848 return err
849
143db31d 850 def prepare_outtmpl(self, outtmpl, info_dict, sanitize=None):
851 """ Make the template and info_dict suitable for substitution (outtmpl % info_dict)"""
752cda38 852 info_dict = dict(info_dict)
a439a3a4 853 na = self.params.get('outtmpl_na_placeholder', 'NA')
143db31d 854
752cda38 855 info_dict['duration_string'] = ( # %(duration>%H-%M-%S)s is wrong if duration > 24hrs
53c18592 856 formatSeconds(info_dict['duration'], '-' if sanitize else ':')
143db31d 857 if info_dict.get('duration', None) is not None
858 else None)
752cda38 859 info_dict['epoch'] = int(time.time())
860 info_dict['autonumber'] = self.params.get('autonumber_start', 1) - 1 + self._num_downloads
861 if info_dict.get('resolution') is None:
862 info_dict['resolution'] = self.format_resolution(info_dict, default=None)
143db31d 863
143db31d 864 # For fields playlist_index and autonumber convert all occurrences
865 # of %(field)s to %(field)0Nd for backward compatibility
866 field_size_compat_map = {
752cda38 867 'playlist_index': len(str(info_dict.get('_last_playlist_index') or '')),
868 'autonumber': self.params.get('autonumber_size') or 5,
143db31d 869 }
752cda38 870
385a27fa 871 TMPL_DICT = {}
872 EXTERNAL_FORMAT_RE = re.compile(STR_FORMAT_RE.format('[^)]*'))
873 MATH_FUNCTIONS = {
874 '+': float.__add__,
875 '-': float.__sub__,
876 }
e625be0d 877 # Field is of the form key1.key2...
878 # where keys (except first) can be string, int or slice
385a27fa 879 FIELD_RE = r'\w+(?:\.(?:\w+|{num}|{num}?(?::{num}?){{1,2}}))*'.format(num=r'(?:-?\d+)')
880 MATH_FIELD_RE = r'''{field}|{num}'''.format(field=FIELD_RE, num=r'-?\d+(?:.\d+)?')
881 MATH_OPERATORS_RE = r'(?:%s)' % '|'.join(map(re.escape, MATH_FUNCTIONS.keys()))
e625be0d 882 INTERNAL_FORMAT_RE = re.compile(r'''(?x)
883 (?P<negate>-)?
385a27fa 884 (?P<fields>{field})
885 (?P<maths>(?:{math_op}{math_field})*)
e625be0d 886 (?:>(?P<strf_format>.+?))?
887 (?:\|(?P<default>.*?))?
385a27fa 888 $'''.format(field=FIELD_RE, math_op=MATH_OPERATORS_RE, math_field=MATH_FIELD_RE))
752cda38 889
76a264ac 890 get_key = lambda k: traverse_obj(
891 info_dict, k.split('.'), is_user_input=True, traverse_string=True)
892
752cda38 893 def get_value(mdict):
894 # Object traversal
76a264ac 895 value = get_key(mdict['fields'])
752cda38 896 # Negative
897 if mdict['negate']:
898 value = float_or_none(value)
899 if value is not None:
900 value *= -1
901 # Do maths
385a27fa 902 offset_key = mdict['maths']
903 if offset_key:
752cda38 904 value = float_or_none(value)
905 operator = None
385a27fa 906 while offset_key:
907 item = re.match(
908 MATH_FIELD_RE if operator else MATH_OPERATORS_RE,
909 offset_key).group(0)
910 offset_key = offset_key[len(item):]
911 if operator is None:
752cda38 912 operator = MATH_FUNCTIONS[item]
385a27fa 913 continue
914 item, multiplier = (item[1:], -1) if item[0] == '-' else (item, 1)
915 offset = float_or_none(item)
916 if offset is None:
917 offset = float_or_none(get_key(item))
918 try:
919 value = operator(value, multiplier * offset)
920 except (TypeError, ZeroDivisionError):
921 return None
922 operator = None
752cda38 923 # Datetime formatting
924 if mdict['strf_format']:
925 value = strftime_or_none(value, mdict['strf_format'])
926
927 return value
928
929 def create_key(outer_mobj):
930 if not outer_mobj.group('has_key'):
931 return '%{}'.format(outer_mobj.group(0))
932
933 key = outer_mobj.group('key')
934 fmt = outer_mobj.group('format')
935 mobj = re.match(INTERNAL_FORMAT_RE, key)
936 if mobj is None:
9fea350f 937 value, default, mobj = None, na, {'fields': ''}
752cda38 938 else:
e625be0d 939 mobj = mobj.groupdict()
752cda38 940 default = mobj['default'] if mobj['default'] is not None else na
941 value = get_value(mobj)
942
943 if fmt == 's' and value is not None and key in field_size_compat_map.keys():
944 fmt = '0{:d}d'.format(field_size_compat_map[key])
945
946 value = default if value is None else value
752cda38 947
76a264ac 948 if fmt == 'c':
949 value = compat_str(value)
950 if value is None:
951 value, fmt = default, 's'
952 else:
953 value = value[0]
954 elif fmt[-1] not in 'rs': # numeric
a439a3a4 955 value = float_or_none(value)
752cda38 956 if value is None:
957 value, fmt = default, 's'
958 if sanitize:
959 if fmt[-1] == 'r':
960 # If value is an object, sanitize might convert it to a string
961 # So we convert it to repr first
962 value, fmt = repr(value), '%ss' % fmt[:-1]
639f1cea 963 if fmt[-1] in 'csr':
9fea350f 964 value = sanitize(mobj['fields'].split('.')[-1], value)
965 key += '\0%s' % fmt
385a27fa 966 TMPL_DICT[key] = value
752cda38 967 return '%({key}){fmt}'.format(key=key, fmt=fmt)
968
385a27fa 969 return EXTERNAL_FORMAT_RE.sub(create_key, outtmpl), TMPL_DICT
143db31d 970
de6000d9 971 def _prepare_filename(self, info_dict, tmpl_type='default'):
8222d8de 972 try:
586a91b6 973 sanitize = lambda k, v: sanitize_filename(
45598aab 974 compat_str(v),
1bb5c511 975 restricted=self.params.get('restrictfilenames'),
40df485f 976 is_id=(k == 'id' or k.endswith('_id')))
de6000d9 977 outtmpl = self.outtmpl_dict.get(tmpl_type, self.outtmpl_dict['default'])
143db31d 978 outtmpl, template_dict = self.prepare_outtmpl(outtmpl, info_dict, sanitize)
d0d9ade4 979
15da37c7
S
980 # expand_path translates '%%' into '%' and '$$' into '$'
981 # correspondingly that is not what we want since we need to keep
982 # '%%' intact for template dict substitution step. Working around
983 # with boundary-alike separator hack.
961ea474 984 sep = ''.join([random.choice(ascii_letters) for _ in range(32)])
15da37c7
S
985 outtmpl = outtmpl.replace('%%', '%{0}%'.format(sep)).replace('$$', '${0}$'.format(sep))
986
987 # outtmpl should be expand_path'ed before template dict substitution
988 # because meta fields may contain env variables we don't want to
989 # be expanded. For example, for outtmpl "%(title)s.%(ext)s" and
990 # title "Hello $PATH", we don't want `$PATH` to be expanded.
991 filename = expand_path(outtmpl).replace(sep, '') % template_dict
992
143db31d 993 force_ext = OUTTMPL_TYPES.get(tmpl_type)
de6000d9 994 if force_ext is not None:
752cda38 995 filename = replace_extension(filename, force_ext, info_dict.get('ext'))
de6000d9 996
bdc3fd2f
U
997 # https://github.com/blackjack4494/youtube-dlc/issues/85
998 trim_file_name = self.params.get('trim_file_name', False)
999 if trim_file_name:
1000 fn_groups = filename.rsplit('.')
1001 ext = fn_groups[-1]
1002 sub_ext = ''
1003 if len(fn_groups) > 2:
1004 sub_ext = fn_groups[-2]
1005 filename = '.'.join(filter(None, [fn_groups[0][:trim_file_name], sub_ext, ext]))
1006
0202b52a 1007 return filename
8222d8de 1008 except ValueError as err:
6febd1c1 1009 self.report_error('Error in output template: ' + str(err) + ' (encoding: ' + repr(preferredencoding()) + ')')
8222d8de
JMF
1010 return None
1011
de6000d9 1012 def prepare_filename(self, info_dict, dir_type='', warn=False):
1013 """Generate the output filename."""
21cd8fae 1014
de6000d9 1015 filename = self._prepare_filename(info_dict, dir_type or 'default')
1016
1017 if warn and not self.__prepare_filename_warned:
21cd8fae 1018 if not self.params.get('paths'):
de6000d9 1019 pass
1020 elif filename == '-':
1021 self.report_warning('--paths is ignored when an outputting to stdout')
1022 elif os.path.isabs(filename):
1023 self.report_warning('--paths is ignored since an absolute path is given in output template')
1024 self.__prepare_filename_warned = True
1025 if filename == '-' or not filename:
1026 return filename
1027
21cd8fae 1028 return self.get_output_path(dir_type, filename)
0202b52a 1029
120fe513 1030 def _match_entry(self, info_dict, incomplete=False, silent=False):
ecdec191 1031 """ Returns None if the file should be downloaded """
8222d8de 1032
c77495e3 1033 video_title = info_dict.get('title', info_dict.get('id', 'video'))
1034
8b0d7497 1035 def check_filter():
8b0d7497 1036 if 'title' in info_dict:
1037 # This can happen when we're just evaluating the playlist
1038 title = info_dict['title']
1039 matchtitle = self.params.get('matchtitle', False)
1040 if matchtitle:
1041 if not re.search(matchtitle, title, re.IGNORECASE):
1042 return '"' + title + '" title did not match pattern "' + matchtitle + '"'
1043 rejecttitle = self.params.get('rejecttitle', False)
1044 if rejecttitle:
1045 if re.search(rejecttitle, title, re.IGNORECASE):
1046 return '"' + title + '" title matched reject pattern "' + rejecttitle + '"'
1047 date = info_dict.get('upload_date')
1048 if date is not None:
1049 dateRange = self.params.get('daterange', DateRange())
1050 if date not in dateRange:
1051 return '%s upload date is not in range %s' % (date_from_str(date).isoformat(), dateRange)
1052 view_count = info_dict.get('view_count')
1053 if view_count is not None:
1054 min_views = self.params.get('min_views')
1055 if min_views is not None and view_count < min_views:
1056 return 'Skipping %s, because it has not reached minimum view count (%d/%d)' % (video_title, view_count, min_views)
1057 max_views = self.params.get('max_views')
1058 if max_views is not None and view_count > max_views:
1059 return 'Skipping %s, because it has exceeded the maximum view count (%d/%d)' % (video_title, view_count, max_views)
1060 if age_restricted(info_dict.get('age_limit'), self.params.get('age_limit')):
1061 return 'Skipping "%s" because it is age restricted' % video_title
8b0d7497 1062
1063 if not incomplete:
1064 match_filter = self.params.get('match_filter')
1065 if match_filter is not None:
1066 ret = match_filter(info_dict)
1067 if ret is not None:
1068 return ret
1069 return None
1070
c77495e3 1071 if self.in_download_archive(info_dict):
1072 reason = '%s has already been recorded in the archive' % video_title
1073 break_opt, break_err = 'break_on_existing', ExistingVideoReached
1074 else:
1075 reason = check_filter()
1076 break_opt, break_err = 'break_on_reject', RejectedVideoReached
8b0d7497 1077 if reason is not None:
120fe513 1078 if not silent:
1079 self.to_screen('[download] ' + reason)
c77495e3 1080 if self.params.get(break_opt, False):
1081 raise break_err()
8b0d7497 1082 return reason
fe7e0c98 1083
b6c45014
JMF
1084 @staticmethod
1085 def add_extra_info(info_dict, extra_info):
1086 '''Set the keys from extra_info in info dict if they are missing'''
1087 for key, value in extra_info.items():
1088 info_dict.setdefault(key, value)
1089
58f197b7 1090 def extract_info(self, url, download=True, ie_key=None, extra_info={},
61aa5ba3 1091 process=True, force_generic_extractor=False):
41d1cca3 1092 """
1093 Return a list with a dictionary for each video extracted.
1094
1095 Arguments:
1096 url -- URL to extract
1097
1098 Keyword arguments:
1099 download -- whether to download videos during extraction
1100 ie_key -- extractor key hint
1101 extra_info -- dictionary containing the extra values to add to each result
1102 process -- whether to resolve all unresolved references (URLs, playlist items),
1103 must be True for download to work.
1104 force_generic_extractor -- force using the generic extractor
1105 """
fe7e0c98 1106
61aa5ba3 1107 if not ie_key and force_generic_extractor:
d22dec74
S
1108 ie_key = 'Generic'
1109
8222d8de 1110 if ie_key:
56c73665 1111 ies = [self.get_info_extractor(ie_key)]
8222d8de
JMF
1112 else:
1113 ies = self._ies
1114
1115 for ie in ies:
1116 if not ie.suitable(url):
1117 continue
1118
9a68de12 1119 ie_key = ie.ie_key()
1120 ie = self.get_info_extractor(ie_key)
8222d8de 1121 if not ie.working():
6febd1c1
PH
1122 self.report_warning('The program functionality for this site has been marked as broken, '
1123 'and will probably not work.')
8222d8de
JMF
1124
1125 try:
d0757229 1126 temp_id = str_or_none(
63be1aab 1127 ie.extract_id(url) if callable(getattr(ie, 'extract_id', None))
1128 else ie._match_id(url))
a0566bbf 1129 except (AssertionError, IndexError, AttributeError):
1130 temp_id = None
1131 if temp_id is not None and self.in_download_archive({'id': temp_id, 'ie_key': ie_key}):
1132 self.to_screen("[%s] %s: has already been recorded in archive" % (
1133 ie_key, temp_id))
1134 break
58f197b7 1135 return self.__extract_info(url, ie, download, extra_info, process)
a0566bbf 1136 else:
1137 self.report_error('no suitable InfoExtractor for URL %s' % url)
1138
1139 def __handle_extraction_exceptions(func):
1140 def wrapper(self, *args, **kwargs):
1141 try:
1142 return func(self, *args, **kwargs)
773f291d
S
1143 except GeoRestrictedError as e:
1144 msg = e.msg
1145 if e.countries:
1146 msg += '\nThis video is available in %s.' % ', '.join(
1147 map(ISO3166Utils.short2full, e.countries))
1148 msg += '\nYou might want to use a VPN or a proxy server (with --proxy) to workaround.'
1149 self.report_error(msg)
fb043a6e 1150 except ExtractorError as e: # An error we somewhat expected
2c74e6fa 1151 self.report_error(compat_str(e), e.format_traceback())
51d9739f 1152 except ThrottledDownload:
1153 self.to_stderr('\r')
1154 self.report_warning('The download speed is below throttle limit. Re-extracting data')
1155 return wrapper(self, *args, **kwargs)
8b0d7497 1156 except (MaxDownloadsReached, ExistingVideoReached, RejectedVideoReached):
d3e5bbf4 1157 raise
8222d8de
JMF
1158 except Exception as e:
1159 if self.params.get('ignoreerrors', False):
9b9c5355 1160 self.report_error(error_to_compat_str(e), tb=encode_compat_str(traceback.format_exc()))
8222d8de
JMF
1161 else:
1162 raise
a0566bbf 1163 return wrapper
1164
1165 @__handle_extraction_exceptions
58f197b7 1166 def __extract_info(self, url, ie, download, extra_info, process):
a0566bbf 1167 ie_result = ie.extract(url)
1168 if ie_result is None: # Finished already (backwards compatibility; listformats and friends should be moved here)
1169 return
1170 if isinstance(ie_result, list):
1171 # Backwards compatibility: old IE result format
1172 ie_result = {
1173 '_type': 'compat_list',
1174 'entries': ie_result,
1175 }
a0566bbf 1176 self.add_default_extra_info(ie_result, ie, url)
1177 if process:
1178 return self.process_ie_result(ie_result, download, extra_info)
8222d8de 1179 else:
a0566bbf 1180 return ie_result
fe7e0c98 1181
ea38e55f 1182 def add_default_extra_info(self, ie_result, ie, url):
6033d980 1183 if url is not None:
1184 self.add_extra_info(ie_result, {
1185 'webpage_url': url,
1186 'original_url': url,
1187 'webpage_url_basename': url_basename(url),
1188 })
1189 if ie is not None:
1190 self.add_extra_info(ie_result, {
1191 'extractor': ie.IE_NAME,
1192 'extractor_key': ie.ie_key(),
1193 })
ea38e55f 1194
8222d8de
JMF
1195 def process_ie_result(self, ie_result, download=True, extra_info={}):
1196 """
1197 Take the result of the ie(may be modified) and resolve all unresolved
1198 references (URLs, playlist items).
1199
1200 It will also download the videos if 'download'.
1201 Returns the resolved ie_result.
1202 """
e8ee972c
PH
1203 result_type = ie_result.get('_type', 'video')
1204
057a5206 1205 if result_type in ('url', 'url_transparent'):
134c6ea8 1206 ie_result['url'] = sanitize_url(ie_result['url'])
057a5206 1207 extract_flat = self.params.get('extract_flat', False)
3089bc74
S
1208 if ((extract_flat == 'in_playlist' and 'playlist' in extra_info)
1209 or extract_flat is True):
ecb54191 1210 info_copy = ie_result.copy()
1211 self.add_extra_info(info_copy, extra_info)
6033d980 1212 ie = try_get(ie_result.get('ie_key'), self.get_info_extractor)
1213 self.add_default_extra_info(info_copy, ie, ie_result['url'])
ecb54191 1214 self.__forced_printings(info_copy, self.prepare_filename(info_copy), incomplete=True)
e8ee972c
PH
1215 return ie_result
1216
8222d8de 1217 if result_type == 'video':
b6c45014 1218 self.add_extra_info(ie_result, extra_info)
9c2b75b5 1219 ie_result = self.process_video_result(ie_result, download=download)
28b0eb0f 1220 additional_urls = (ie_result or {}).get('additional_urls')
9c2b75b5 1221 if additional_urls:
1222 # TODO: Improve MetadataFromFieldPP to allow setting a list
1223 if isinstance(additional_urls, compat_str):
1224 additional_urls = [additional_urls]
1225 self.to_screen(
1226 '[info] %s: %d additional URL(s) requested' % (ie_result['id'], len(additional_urls)))
1227 self.write_debug('Additional URLs: "%s"' % '", "'.join(additional_urls))
1228 ie_result['additional_entries'] = [
1229 self.extract_info(
1230 url, download, extra_info,
1231 force_generic_extractor=self.params.get('force_generic_extractor'))
1232 for url in additional_urls
1233 ]
1234 return ie_result
8222d8de
JMF
1235 elif result_type == 'url':
1236 # We have to add extra_info to the results because it may be
1237 # contained in a playlist
07cce701 1238 return self.extract_info(
1239 ie_result['url'], download,
1240 ie_key=ie_result.get('ie_key'),
1241 extra_info=extra_info)
7fc3fa05
PH
1242 elif result_type == 'url_transparent':
1243 # Use the information from the embedding page
1244 info = self.extract_info(
1245 ie_result['url'], ie_key=ie_result.get('ie_key'),
1246 extra_info=extra_info, download=False, process=False)
1247
1640eb09
S
1248 # extract_info may return None when ignoreerrors is enabled and
1249 # extraction failed with an error, don't crash and return early
1250 # in this case
1251 if not info:
1252 return info
1253
412c617d
PH
1254 force_properties = dict(
1255 (k, v) for k, v in ie_result.items() if v is not None)
0396806f 1256 for f in ('_type', 'url', 'id', 'extractor', 'extractor_key', 'ie_key'):
412c617d
PH
1257 if f in force_properties:
1258 del force_properties[f]
1259 new_result = info.copy()
1260 new_result.update(force_properties)
7fc3fa05 1261
0563f7ac
S
1262 # Extracted info may not be a video result (i.e.
1263 # info.get('_type', 'video') != video) but rather an url or
1264 # url_transparent. In such cases outer metadata (from ie_result)
1265 # should be propagated to inner one (info). For this to happen
1266 # _type of info should be overridden with url_transparent. This
067aa17e 1267 # fixes issue from https://github.com/ytdl-org/youtube-dl/pull/11163.
0563f7ac
S
1268 if new_result.get('_type') == 'url':
1269 new_result['_type'] = 'url_transparent'
7fc3fa05
PH
1270
1271 return self.process_ie_result(
1272 new_result, download=download, extra_info=extra_info)
40fcba5e 1273 elif result_type in ('playlist', 'multi_video'):
30a074c2 1274 # Protect from infinite recursion due to recursively nested playlists
1275 # (see https://github.com/ytdl-org/youtube-dl/issues/27833)
1276 webpage_url = ie_result['webpage_url']
1277 if webpage_url in self._playlist_urls:
7e85e872 1278 self.to_screen(
30a074c2 1279 '[download] Skipping already downloaded playlist: %s'
1280 % ie_result.get('title') or ie_result.get('id'))
1281 return
7e85e872 1282
30a074c2 1283 self._playlist_level += 1
1284 self._playlist_urls.add(webpage_url)
bc516a3f 1285 self._sanitize_thumbnails(ie_result)
30a074c2 1286 try:
1287 return self.__process_playlist(ie_result, download)
1288 finally:
1289 self._playlist_level -= 1
1290 if not self._playlist_level:
1291 self._playlist_urls.clear()
8222d8de 1292 elif result_type == 'compat_list':
c9bf4114
PH
1293 self.report_warning(
1294 'Extractor %s returned a compat_list result. '
1295 'It needs to be updated.' % ie_result.get('extractor'))
5f6a1245 1296
8222d8de 1297 def _fixup(r):
9e1a5b84
JW
1298 self.add_extra_info(
1299 r,
9103bbc5
JMF
1300 {
1301 'extractor': ie_result['extractor'],
1302 'webpage_url': ie_result['webpage_url'],
29eb5174 1303 'webpage_url_basename': url_basename(ie_result['webpage_url']),
be97abc2 1304 'extractor_key': ie_result['extractor_key'],
9e1a5b84
JW
1305 }
1306 )
8222d8de
JMF
1307 return r
1308 ie_result['entries'] = [
b6c45014 1309 self.process_ie_result(_fixup(r), download, extra_info)
8222d8de
JMF
1310 for r in ie_result['entries']
1311 ]
1312 return ie_result
1313 else:
1314 raise Exception('Invalid result type: %s' % result_type)
1315
e92caff5 1316 def _ensure_dir_exists(self, path):
1317 return make_dir(path, self.report_error)
1318
30a074c2 1319 def __process_playlist(self, ie_result, download):
1320 # We process each entry in the playlist
1321 playlist = ie_result.get('title') or ie_result.get('id')
1322 self.to_screen('[download] Downloading playlist: %s' % playlist)
1323
498f5606 1324 if 'entries' not in ie_result:
1325 raise EntryNotInPlaylist()
1326 incomplete_entries = bool(ie_result.get('requested_entries'))
1327 if incomplete_entries:
1328 def fill_missing_entries(entries, indexes):
1329 ret = [None] * max(*indexes)
1330 for i, entry in zip(indexes, entries):
1331 ret[i - 1] = entry
1332 return ret
1333 ie_result['entries'] = fill_missing_entries(ie_result['entries'], ie_result['requested_entries'])
02fd60d3 1334
30a074c2 1335 playlist_results = []
1336
56a8fb4f 1337 playliststart = self.params.get('playliststart', 1)
30a074c2 1338 playlistend = self.params.get('playlistend')
1339 # For backwards compatibility, interpret -1 as whole list
1340 if playlistend == -1:
1341 playlistend = None
1342
1343 playlistitems_str = self.params.get('playlist_items')
1344 playlistitems = None
1345 if playlistitems_str is not None:
1346 def iter_playlistitems(format):
1347 for string_segment in format.split(','):
1348 if '-' in string_segment:
1349 start, end = string_segment.split('-')
1350 for item in range(int(start), int(end) + 1):
1351 yield int(item)
1352 else:
1353 yield int(string_segment)
1354 playlistitems = orderedSet(iter_playlistitems(playlistitems_str))
1355
1356 ie_entries = ie_result['entries']
56a8fb4f 1357 msg = (
1358 'Downloading %d videos' if not isinstance(ie_entries, list)
1359 else 'Collected %d videos; downloading %%d of them' % len(ie_entries))
1360 if not isinstance(ie_entries, (list, PagedList)):
1361 ie_entries = LazyList(ie_entries)
1362
1363 entries = []
1364 for i in playlistitems or itertools.count(playliststart):
1365 if playlistitems is None and playlistend is not None and playlistend < i:
1366 break
1367 entry = None
1368 try:
1369 entry = ie_entries[i - 1]
1370 if entry is None:
498f5606 1371 raise EntryNotInPlaylist()
56a8fb4f 1372 except (IndexError, EntryNotInPlaylist):
1373 if incomplete_entries:
1374 raise EntryNotInPlaylist()
1375 elif not playlistitems:
1376 break
1377 entries.append(entry)
120fe513 1378 try:
1379 if entry is not None:
1380 self._match_entry(entry, incomplete=True, silent=True)
1381 except (ExistingVideoReached, RejectedVideoReached):
1382 break
56a8fb4f 1383 ie_result['entries'] = entries
30a074c2 1384
56a8fb4f 1385 # Save playlist_index before re-ordering
1386 entries = [
1387 ((playlistitems[i - 1] if playlistitems else i), entry)
1388 for i, entry in enumerate(entries, 1)
1389 if entry is not None]
1390 n_entries = len(entries)
498f5606 1391
498f5606 1392 if not playlistitems and (playliststart or playlistend):
56a8fb4f 1393 playlistitems = list(range(playliststart, playliststart + n_entries))
498f5606 1394 ie_result['requested_entries'] = playlistitems
1395
1396 if self.params.get('allow_playlist_files', True):
1397 ie_copy = {
1398 'playlist': playlist,
1399 'playlist_id': ie_result.get('id'),
1400 'playlist_title': ie_result.get('title'),
1401 'playlist_uploader': ie_result.get('uploader'),
1402 'playlist_uploader_id': ie_result.get('uploader_id'),
71729754 1403 'playlist_index': 0,
498f5606 1404 }
1405 ie_copy.update(dict(ie_result))
1406
1407 if self.params.get('writeinfojson', False):
1408 infofn = self.prepare_filename(ie_copy, 'pl_infojson')
1409 if not self._ensure_dir_exists(encodeFilename(infofn)):
1410 return
1411 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(infofn)):
1412 self.to_screen('[info] Playlist metadata is already present')
1413 else:
1414 self.to_screen('[info] Writing playlist metadata as JSON to: ' + infofn)
1415 try:
1416 write_json_file(self.filter_requested_info(ie_result, self.params.get('clean_infojson', True)), infofn)
1417 except (OSError, IOError):
1418 self.report_error('Cannot write playlist metadata to JSON file ' + infofn)
1419
681de68e 1420 # TODO: This should be passed to ThumbnailsConvertor if necessary
1421 self._write_thumbnails(ie_copy, self.prepare_filename(ie_copy, 'pl_thumbnail'))
1422
498f5606 1423 if self.params.get('writedescription', False):
1424 descfn = self.prepare_filename(ie_copy, 'pl_description')
1425 if not self._ensure_dir_exists(encodeFilename(descfn)):
1426 return
1427 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(descfn)):
1428 self.to_screen('[info] Playlist description is already present')
1429 elif ie_result.get('description') is None:
1430 self.report_warning('There\'s no playlist description to write.')
1431 else:
1432 try:
1433 self.to_screen('[info] Writing playlist description to: ' + descfn)
1434 with io.open(encodeFilename(descfn), 'w', encoding='utf-8') as descfile:
1435 descfile.write(ie_result['description'])
1436 except (OSError, IOError):
1437 self.report_error('Cannot write playlist description file ' + descfn)
1438 return
30a074c2 1439
1440 if self.params.get('playlistreverse', False):
1441 entries = entries[::-1]
30a074c2 1442 if self.params.get('playlistrandom', False):
1443 random.shuffle(entries)
1444
1445 x_forwarded_for = ie_result.get('__x_forwarded_for_ip')
1446
56a8fb4f 1447 self.to_screen('[%s] playlist %s: %s' % (ie_result['extractor'], playlist, msg % n_entries))
26e2805c 1448 failures = 0
1449 max_failures = self.params.get('skip_playlist_after_errors') or float('inf')
71729754 1450 for i, entry_tuple in enumerate(entries, 1):
1451 playlist_index, entry = entry_tuple
53ed7066 1452 if 'playlist_index' in self.params.get('compat_options', []):
1453 playlist_index = playlistitems[i - 1] if playlistitems else i
30a074c2 1454 self.to_screen('[download] Downloading video %s of %s' % (i, n_entries))
1455 # This __x_forwarded_for_ip thing is a bit ugly but requires
1456 # minimal changes
1457 if x_forwarded_for:
1458 entry['__x_forwarded_for_ip'] = x_forwarded_for
1459 extra = {
1460 'n_entries': n_entries,
f59ae581 1461 '_last_playlist_index': max(playlistitems) if playlistitems else (playlistend or n_entries),
71729754 1462 'playlist_index': playlist_index,
1463 'playlist_autonumber': i,
30a074c2 1464 'playlist': playlist,
1465 'playlist_id': ie_result.get('id'),
1466 'playlist_title': ie_result.get('title'),
1467 'playlist_uploader': ie_result.get('uploader'),
1468 'playlist_uploader_id': ie_result.get('uploader_id'),
30a074c2 1469 'extractor': ie_result['extractor'],
1470 'webpage_url': ie_result['webpage_url'],
1471 'webpage_url_basename': url_basename(ie_result['webpage_url']),
1472 'extractor_key': ie_result['extractor_key'],
1473 }
1474
1475 if self._match_entry(entry, incomplete=True) is not None:
1476 continue
1477
1478 entry_result = self.__process_iterable_entry(entry, download, extra)
26e2805c 1479 if not entry_result:
1480 failures += 1
1481 if failures >= max_failures:
1482 self.report_error(
1483 'Skipping the remaining entries in playlist "%s" since %d items failed extraction' % (playlist, failures))
1484 break
30a074c2 1485 # TODO: skip failed (empty) entries?
1486 playlist_results.append(entry_result)
1487 ie_result['entries'] = playlist_results
1488 self.to_screen('[download] Finished downloading playlist: %s' % playlist)
1489 return ie_result
1490
a0566bbf 1491 @__handle_extraction_exceptions
1492 def __process_iterable_entry(self, entry, download, extra_info):
1493 return self.process_ie_result(
1494 entry, download=download, extra_info=extra_info)
1495
67134eab
JMF
1496 def _build_format_filter(self, filter_spec):
1497 " Returns a function to filter the formats according to the filter_spec "
083c9df9
PH
1498
1499 OPERATORS = {
1500 '<': operator.lt,
1501 '<=': operator.le,
1502 '>': operator.gt,
1503 '>=': operator.ge,
1504 '=': operator.eq,
1505 '!=': operator.ne,
1506 }
67134eab 1507 operator_rex = re.compile(r'''(?x)\s*
187986a8 1508 (?P<key>width|height|tbr|abr|vbr|asr|filesize|filesize_approx|fps)\s*
1509 (?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
1510 (?P<value>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?)\s*
083c9df9 1511 ''' % '|'.join(map(re.escape, OPERATORS.keys())))
187986a8 1512 m = operator_rex.fullmatch(filter_spec)
9ddb6925
S
1513 if m:
1514 try:
1515 comparison_value = int(m.group('value'))
1516 except ValueError:
1517 comparison_value = parse_filesize(m.group('value'))
1518 if comparison_value is None:
1519 comparison_value = parse_filesize(m.group('value') + 'B')
1520 if comparison_value is None:
1521 raise ValueError(
1522 'Invalid value %r in format specification %r' % (
67134eab 1523 m.group('value'), filter_spec))
9ddb6925
S
1524 op = OPERATORS[m.group('op')]
1525
083c9df9 1526 if not m:
9ddb6925
S
1527 STR_OPERATORS = {
1528 '=': operator.eq,
10d33b34
YCH
1529 '^=': lambda attr, value: attr.startswith(value),
1530 '$=': lambda attr, value: attr.endswith(value),
1531 '*=': lambda attr, value: value in attr,
9ddb6925 1532 }
187986a8 1533 str_operator_rex = re.compile(r'''(?x)\s*
1534 (?P<key>[a-zA-Z0-9._-]+)\s*
1535 (?P<negation>!\s*)?(?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
1536 (?P<value>[a-zA-Z0-9._-]+)\s*
9ddb6925 1537 ''' % '|'.join(map(re.escape, STR_OPERATORS.keys())))
187986a8 1538 m = str_operator_rex.fullmatch(filter_spec)
9ddb6925
S
1539 if m:
1540 comparison_value = m.group('value')
2cc779f4
S
1541 str_op = STR_OPERATORS[m.group('op')]
1542 if m.group('negation'):
e118a879 1543 op = lambda attr, value: not str_op(attr, value)
2cc779f4
S
1544 else:
1545 op = str_op
083c9df9 1546
9ddb6925 1547 if not m:
187986a8 1548 raise SyntaxError('Invalid filter specification %r' % filter_spec)
083c9df9
PH
1549
1550 def _filter(f):
1551 actual_value = f.get(m.group('key'))
1552 if actual_value is None:
1553 return m.group('none_inclusive')
1554 return op(actual_value, comparison_value)
67134eab
JMF
1555 return _filter
1556
0017d9ad 1557 def _default_format_spec(self, info_dict, download=True):
0017d9ad 1558
af0f7428
S
1559 def can_merge():
1560 merger = FFmpegMergerPP(self)
1561 return merger.available and merger.can_merge()
1562
91ebc640 1563 prefer_best = (
1564 not self.params.get('simulate', False)
1565 and download
1566 and (
1567 not can_merge()
19807826 1568 or info_dict.get('is_live', False)
de6000d9 1569 or self.outtmpl_dict['default'] == '-'))
53ed7066 1570 compat = (
1571 prefer_best
1572 or self.params.get('allow_multiple_audio_streams', False)
1573 or 'format-spec' in self.params.get('compat_opts', []))
91ebc640 1574
1575 return (
53ed7066 1576 'best/bestvideo+bestaudio' if prefer_best
1577 else 'bestvideo*+bestaudio/best' if not compat
91ebc640 1578 else 'bestvideo+bestaudio/best')
0017d9ad 1579
67134eab
JMF
1580 def build_format_selector(self, format_spec):
1581 def syntax_error(note, start):
1582 message = (
1583 'Invalid format specification: '
1584 '{0}\n\t{1}\n\t{2}^'.format(note, format_spec, ' ' * start[1]))
1585 return SyntaxError(message)
1586
1587 PICKFIRST = 'PICKFIRST'
1588 MERGE = 'MERGE'
1589 SINGLE = 'SINGLE'
0130afb7 1590 GROUP = 'GROUP'
67134eab
JMF
1591 FormatSelector = collections.namedtuple('FormatSelector', ['type', 'selector', 'filters'])
1592
91ebc640 1593 allow_multiple_streams = {'audio': self.params.get('allow_multiple_audio_streams', False),
1594 'video': self.params.get('allow_multiple_video_streams', False)}
909d24dd 1595
e8e73840 1596 check_formats = self.params.get('check_formats')
1597
67134eab
JMF
1598 def _parse_filter(tokens):
1599 filter_parts = []
1600 for type, string, start, _, _ in tokens:
1601 if type == tokenize.OP and string == ']':
1602 return ''.join(filter_parts)
1603 else:
1604 filter_parts.append(string)
1605
232541df 1606 def _remove_unused_ops(tokens):
17cc1534 1607 # Remove operators that we don't use and join them with the surrounding strings
232541df
JMF
1608 # for example: 'mp4' '-' 'baseline' '-' '16x9' is converted to 'mp4-baseline-16x9'
1609 ALLOWED_OPS = ('/', '+', ',', '(', ')')
1610 last_string, last_start, last_end, last_line = None, None, None, None
1611 for type, string, start, end, line in tokens:
1612 if type == tokenize.OP and string == '[':
1613 if last_string:
1614 yield tokenize.NAME, last_string, last_start, last_end, last_line
1615 last_string = None
1616 yield type, string, start, end, line
1617 # everything inside brackets will be handled by _parse_filter
1618 for type, string, start, end, line in tokens:
1619 yield type, string, start, end, line
1620 if type == tokenize.OP and string == ']':
1621 break
1622 elif type == tokenize.OP and string in ALLOWED_OPS:
1623 if last_string:
1624 yield tokenize.NAME, last_string, last_start, last_end, last_line
1625 last_string = None
1626 yield type, string, start, end, line
1627 elif type in [tokenize.NAME, tokenize.NUMBER, tokenize.OP]:
1628 if not last_string:
1629 last_string = string
1630 last_start = start
1631 last_end = end
1632 else:
1633 last_string += string
1634 if last_string:
1635 yield tokenize.NAME, last_string, last_start, last_end, last_line
1636
cf2ac6df 1637 def _parse_format_selection(tokens, inside_merge=False, inside_choice=False, inside_group=False):
67134eab
JMF
1638 selectors = []
1639 current_selector = None
1640 for type, string, start, _, _ in tokens:
1641 # ENCODING is only defined in python 3.x
1642 if type == getattr(tokenize, 'ENCODING', None):
1643 continue
1644 elif type in [tokenize.NAME, tokenize.NUMBER]:
1645 current_selector = FormatSelector(SINGLE, string, [])
1646 elif type == tokenize.OP:
cf2ac6df
JMF
1647 if string == ')':
1648 if not inside_group:
1649 # ')' will be handled by the parentheses group
1650 tokens.restore_last_token()
67134eab 1651 break
cf2ac6df 1652 elif inside_merge and string in ['/', ',']:
0130afb7
JMF
1653 tokens.restore_last_token()
1654 break
cf2ac6df
JMF
1655 elif inside_choice and string == ',':
1656 tokens.restore_last_token()
1657 break
1658 elif string == ',':
0a31a350
JMF
1659 if not current_selector:
1660 raise syntax_error('"," must follow a format selector', start)
67134eab
JMF
1661 selectors.append(current_selector)
1662 current_selector = None
1663 elif string == '/':
d96d604e
JMF
1664 if not current_selector:
1665 raise syntax_error('"/" must follow a format selector', start)
67134eab 1666 first_choice = current_selector
cf2ac6df 1667 second_choice = _parse_format_selection(tokens, inside_choice=True)
f5f4a27a 1668 current_selector = FormatSelector(PICKFIRST, (first_choice, second_choice), [])
67134eab
JMF
1669 elif string == '[':
1670 if not current_selector:
1671 current_selector = FormatSelector(SINGLE, 'best', [])
1672 format_filter = _parse_filter(tokens)
1673 current_selector.filters.append(format_filter)
0130afb7
JMF
1674 elif string == '(':
1675 if current_selector:
1676 raise syntax_error('Unexpected "("', start)
cf2ac6df
JMF
1677 group = _parse_format_selection(tokens, inside_group=True)
1678 current_selector = FormatSelector(GROUP, group, [])
67134eab 1679 elif string == '+':
d03cfdce 1680 if not current_selector:
1681 raise syntax_error('Unexpected "+"', start)
1682 selector_1 = current_selector
1683 selector_2 = _parse_format_selection(tokens, inside_merge=True)
1684 if not selector_2:
1685 raise syntax_error('Expected a selector', start)
1686 current_selector = FormatSelector(MERGE, (selector_1, selector_2), [])
67134eab
JMF
1687 else:
1688 raise syntax_error('Operator not recognized: "{0}"'.format(string), start)
1689 elif type == tokenize.ENDMARKER:
1690 break
1691 if current_selector:
1692 selectors.append(current_selector)
1693 return selectors
1694
f8d4ad9a 1695 def _merge(formats_pair):
1696 format_1, format_2 = formats_pair
1697
1698 formats_info = []
1699 formats_info.extend(format_1.get('requested_formats', (format_1,)))
1700 formats_info.extend(format_2.get('requested_formats', (format_2,)))
1701
1702 if not allow_multiple_streams['video'] or not allow_multiple_streams['audio']:
551f9388 1703 get_no_more = {'video': False, 'audio': False}
f8d4ad9a 1704 for (i, fmt_info) in enumerate(formats_info):
551f9388 1705 if fmt_info.get('acodec') == fmt_info.get('vcodec') == 'none':
1706 formats_info.pop(i)
1707 continue
1708 for aud_vid in ['audio', 'video']:
f8d4ad9a 1709 if not allow_multiple_streams[aud_vid] and fmt_info.get(aud_vid[0] + 'codec') != 'none':
1710 if get_no_more[aud_vid]:
1711 formats_info.pop(i)
1712 get_no_more[aud_vid] = True
1713
1714 if len(formats_info) == 1:
1715 return formats_info[0]
1716
1717 video_fmts = [fmt_info for fmt_info in formats_info if fmt_info.get('vcodec') != 'none']
1718 audio_fmts = [fmt_info for fmt_info in formats_info if fmt_info.get('acodec') != 'none']
1719
1720 the_only_video = video_fmts[0] if len(video_fmts) == 1 else None
1721 the_only_audio = audio_fmts[0] if len(audio_fmts) == 1 else None
1722
1723 output_ext = self.params.get('merge_output_format')
1724 if not output_ext:
1725 if the_only_video:
1726 output_ext = the_only_video['ext']
1727 elif the_only_audio and not video_fmts:
1728 output_ext = the_only_audio['ext']
1729 else:
1730 output_ext = 'mkv'
1731
1732 new_dict = {
1733 'requested_formats': formats_info,
1734 'format': '+'.join(fmt_info.get('format') for fmt_info in formats_info),
1735 'format_id': '+'.join(fmt_info.get('format_id') for fmt_info in formats_info),
1736 'ext': output_ext,
1737 }
1738
1739 if the_only_video:
1740 new_dict.update({
1741 'width': the_only_video.get('width'),
1742 'height': the_only_video.get('height'),
1743 'resolution': the_only_video.get('resolution') or self.format_resolution(the_only_video),
1744 'fps': the_only_video.get('fps'),
1745 'vcodec': the_only_video.get('vcodec'),
1746 'vbr': the_only_video.get('vbr'),
1747 'stretched_ratio': the_only_video.get('stretched_ratio'),
1748 })
1749
1750 if the_only_audio:
1751 new_dict.update({
1752 'acodec': the_only_audio.get('acodec'),
1753 'abr': the_only_audio.get('abr'),
1754 })
1755
1756 return new_dict
1757
e8e73840 1758 def _check_formats(formats):
981052c9 1759 if not check_formats:
1760 yield from formats
b5ac45b1 1761 return
e8e73840 1762 for f in formats:
1763 self.to_screen('[info] Testing format %s' % f['format_id'])
21cd8fae 1764 temp_file = tempfile.NamedTemporaryFile(
1765 suffix='.tmp', delete=False,
1766 dir=self.get_output_path('temp') or None)
1767 temp_file.close()
fe346461 1768 try:
981052c9 1769 success, _ = self.dl(temp_file.name, f, test=True)
1770 except (DownloadError, IOError, OSError, ValueError) + network_exceptions:
1771 success = False
fe346461 1772 finally:
21cd8fae 1773 if os.path.exists(temp_file.name):
1774 try:
1775 os.remove(temp_file.name)
1776 except OSError:
1777 self.report_warning('Unable to delete temporary file "%s"' % temp_file.name)
981052c9 1778 if success:
e8e73840 1779 yield f
1780 else:
1781 self.to_screen('[info] Unable to download format %s. Skipping...' % f['format_id'])
1782
67134eab 1783 def _build_selector_function(selector):
909d24dd 1784 if isinstance(selector, list): # ,
67134eab
JMF
1785 fs = [_build_selector_function(s) for s in selector]
1786
317f7ab6 1787 def selector_function(ctx):
67134eab 1788 for f in fs:
981052c9 1789 yield from f(ctx)
67134eab 1790 return selector_function
909d24dd 1791
1792 elif selector.type == GROUP: # ()
0130afb7 1793 selector_function = _build_selector_function(selector.selector)
909d24dd 1794
1795 elif selector.type == PICKFIRST: # /
67134eab
JMF
1796 fs = [_build_selector_function(s) for s in selector.selector]
1797
317f7ab6 1798 def selector_function(ctx):
67134eab 1799 for f in fs:
317f7ab6 1800 picked_formats = list(f(ctx))
67134eab
JMF
1801 if picked_formats:
1802 return picked_formats
1803 return []
67134eab 1804
981052c9 1805 elif selector.type == MERGE: # +
1806 selector_1, selector_2 = map(_build_selector_function, selector.selector)
1807
1808 def selector_function(ctx):
1809 for pair in itertools.product(
1810 selector_1(copy.deepcopy(ctx)), selector_2(copy.deepcopy(ctx))):
1811 yield _merge(pair)
1812
909d24dd 1813 elif selector.type == SINGLE: # atom
598d185d 1814 format_spec = selector.selector or 'best'
909d24dd 1815
f8d4ad9a 1816 # TODO: Add allvideo, allaudio etc by generalizing the code with best/worst selector
909d24dd 1817 if format_spec == 'all':
1818 def selector_function(ctx):
981052c9 1819 yield from _check_formats(ctx['formats'])
f8d4ad9a 1820 elif format_spec == 'mergeall':
1821 def selector_function(ctx):
981052c9 1822 formats = list(_check_formats(ctx['formats']))
e01d6aa4 1823 if not formats:
1824 return
921b76ca 1825 merged_format = formats[-1]
1826 for f in formats[-2::-1]:
f8d4ad9a 1827 merged_format = _merge((merged_format, f))
1828 yield merged_format
909d24dd 1829
1830 else:
e8e73840 1831 format_fallback, format_reverse, format_idx = False, True, 1
eff63539 1832 mobj = re.match(
1833 r'(?P<bw>best|worst|b|w)(?P<type>video|audio|v|a)?(?P<mod>\*)?(?:\.(?P<n>[1-9]\d*))?$',
1834 format_spec)
1835 if mobj is not None:
1836 format_idx = int_or_none(mobj.group('n'), default=1)
e8e73840 1837 format_reverse = mobj.group('bw')[0] == 'b'
eff63539 1838 format_type = (mobj.group('type') or [None])[0]
1839 not_format_type = {'v': 'a', 'a': 'v'}.get(format_type)
1840 format_modified = mobj.group('mod') is not None
909d24dd 1841
1842 format_fallback = not format_type and not format_modified # for b, w
8326b00a 1843 _filter_f = (
eff63539 1844 (lambda f: f.get('%scodec' % format_type) != 'none')
1845 if format_type and format_modified # bv*, ba*, wv*, wa*
1846 else (lambda f: f.get('%scodec' % not_format_type) == 'none')
1847 if format_type # bv, ba, wv, wa
1848 else (lambda f: f.get('vcodec') != 'none' and f.get('acodec') != 'none')
1849 if not format_modified # b, w
8326b00a 1850 else lambda f: True) # b*, w*
1851 filter_f = lambda f: _filter_f(f) and (
1852 f.get('vcodec') != 'none' or f.get('acodec') != 'none')
67134eab 1853 else:
909d24dd 1854 filter_f = ((lambda f: f.get('ext') == format_spec)
1855 if format_spec in ['mp4', 'flv', 'webm', '3gp', 'm4a', 'mp3', 'ogg', 'aac', 'wav'] # extension
1856 else (lambda f: f.get('format_id') == format_spec)) # id
1857
1858 def selector_function(ctx):
1859 formats = list(ctx['formats'])
909d24dd 1860 matches = list(filter(filter_f, formats)) if filter_f is not None else formats
e8e73840 1861 if format_fallback and ctx['incomplete_formats'] and not matches:
909d24dd 1862 # for extractors with incomplete formats (audio only (soundcloud)
1863 # or video only (imgur)) best/worst will fallback to
1864 # best/worst {video,audio}-only format
e8e73840 1865 matches = formats
981052c9 1866 matches = LazyList(_check_formats(matches[::-1 if format_reverse else 1]))
1867 try:
e8e73840 1868 yield matches[format_idx - 1]
981052c9 1869 except IndexError:
1870 return
083c9df9 1871
67134eab 1872 filters = [self._build_format_filter(f) for f in selector.filters]
083c9df9 1873
317f7ab6
S
1874 def final_selector(ctx):
1875 ctx_copy = copy.deepcopy(ctx)
67134eab 1876 for _filter in filters:
317f7ab6
S
1877 ctx_copy['formats'] = list(filter(_filter, ctx_copy['formats']))
1878 return selector_function(ctx_copy)
67134eab 1879 return final_selector
083c9df9 1880
67134eab 1881 stream = io.BytesIO(format_spec.encode('utf-8'))
0130afb7 1882 try:
232541df 1883 tokens = list(_remove_unused_ops(compat_tokenize_tokenize(stream.readline)))
0130afb7
JMF
1884 except tokenize.TokenError:
1885 raise syntax_error('Missing closing/opening brackets or parenthesis', (0, len(format_spec)))
1886
1887 class TokenIterator(object):
1888 def __init__(self, tokens):
1889 self.tokens = tokens
1890 self.counter = 0
1891
1892 def __iter__(self):
1893 return self
1894
1895 def __next__(self):
1896 if self.counter >= len(self.tokens):
1897 raise StopIteration()
1898 value = self.tokens[self.counter]
1899 self.counter += 1
1900 return value
1901
1902 next = __next__
1903
1904 def restore_last_token(self):
1905 self.counter -= 1
1906
1907 parsed_selector = _parse_format_selection(iter(TokenIterator(tokens)))
67134eab 1908 return _build_selector_function(parsed_selector)
a9c58ad9 1909
e5660ee6
JMF
1910 def _calc_headers(self, info_dict):
1911 res = std_headers.copy()
1912
1913 add_headers = info_dict.get('http_headers')
1914 if add_headers:
1915 res.update(add_headers)
1916
1917 cookies = self._calc_cookies(info_dict)
1918 if cookies:
1919 res['Cookie'] = cookies
1920
0016b84e
S
1921 if 'X-Forwarded-For' not in res:
1922 x_forwarded_for_ip = info_dict.get('__x_forwarded_for_ip')
1923 if x_forwarded_for_ip:
1924 res['X-Forwarded-For'] = x_forwarded_for_ip
1925
e5660ee6
JMF
1926 return res
1927
1928 def _calc_cookies(self, info_dict):
5c2266df 1929 pr = sanitized_Request(info_dict['url'])
e5660ee6 1930 self.cookiejar.add_cookie_header(pr)
662435f7 1931 return pr.get_header('Cookie')
e5660ee6 1932
b0249bca 1933 def _sanitize_thumbnails(self, info_dict):
bc516a3f 1934 thumbnails = info_dict.get('thumbnails')
1935 if thumbnails is None:
1936 thumbnail = info_dict.get('thumbnail')
1937 if thumbnail:
1938 info_dict['thumbnails'] = thumbnails = [{'url': thumbnail}]
1939 if thumbnails:
1940 thumbnails.sort(key=lambda t: (
1941 t.get('preference') if t.get('preference') is not None else -1,
1942 t.get('width') if t.get('width') is not None else -1,
1943 t.get('height') if t.get('height') is not None else -1,
1944 t.get('id') if t.get('id') is not None else '',
1945 t.get('url')))
b0249bca 1946
1947 def test_thumbnail(t):
1948 self.to_screen('[info] Testing thumbnail %s' % t['id'])
1949 try:
1950 self.urlopen(HEADRequest(t['url']))
1951 except network_exceptions as err:
1952 self.to_screen('[info] Unable to connect to thumbnail %s URL "%s" - %s. Skipping...' % (
1953 t['id'], t['url'], error_to_compat_str(err)))
1954 return False
1955 return True
1956
bc516a3f 1957 for i, t in enumerate(thumbnails):
bc516a3f 1958 if t.get('id') is None:
1959 t['id'] = '%d' % i
b0249bca 1960 if t.get('width') and t.get('height'):
1961 t['resolution'] = '%dx%d' % (t['width'], t['height'])
1962 t['url'] = sanitize_url(t['url'])
1963 if self.params.get('check_formats'):
981052c9 1964 info_dict['thumbnails'] = LazyList(filter(test_thumbnail, thumbnails[::-1])).reverse()
bc516a3f 1965
dd82ffea
JMF
1966 def process_video_result(self, info_dict, download=True):
1967 assert info_dict.get('_type', 'video') == 'video'
1968
bec1fad2
PH
1969 if 'id' not in info_dict:
1970 raise ExtractorError('Missing "id" field in extractor result')
1971 if 'title' not in info_dict:
1972 raise ExtractorError('Missing "title" field in extractor result')
1973
c9969434
S
1974 def report_force_conversion(field, field_not, conversion):
1975 self.report_warning(
1976 '"%s" field is not %s - forcing %s conversion, there is an error in extractor'
1977 % (field, field_not, conversion))
1978
1979 def sanitize_string_field(info, string_field):
1980 field = info.get(string_field)
1981 if field is None or isinstance(field, compat_str):
1982 return
1983 report_force_conversion(string_field, 'a string', 'string')
1984 info[string_field] = compat_str(field)
1985
1986 def sanitize_numeric_fields(info):
1987 for numeric_field in self._NUMERIC_FIELDS:
1988 field = info.get(numeric_field)
1989 if field is None or isinstance(field, compat_numeric_types):
1990 continue
1991 report_force_conversion(numeric_field, 'numeric', 'int')
1992 info[numeric_field] = int_or_none(field)
1993
1994 sanitize_string_field(info_dict, 'id')
1995 sanitize_numeric_fields(info_dict)
be6217b2 1996
dd82ffea
JMF
1997 if 'playlist' not in info_dict:
1998 # It isn't part of a playlist
1999 info_dict['playlist'] = None
2000 info_dict['playlist_index'] = None
2001
bc516a3f 2002 self._sanitize_thumbnails(info_dict)
d5519808 2003
536a55da 2004 thumbnail = info_dict.get('thumbnail')
bc516a3f 2005 thumbnails = info_dict.get('thumbnails')
536a55da
S
2006 if thumbnail:
2007 info_dict['thumbnail'] = sanitize_url(thumbnail)
2008 elif thumbnails:
d5519808
PH
2009 info_dict['thumbnail'] = thumbnails[-1]['url']
2010
c9ae7b95 2011 if 'display_id' not in info_dict and 'id' in info_dict:
0afef30b
PH
2012 info_dict['display_id'] = info_dict['id']
2013
10db0d2f 2014 for ts_key, date_key in (
2015 ('timestamp', 'upload_date'),
2016 ('release_timestamp', 'release_date'),
2017 ):
2018 if info_dict.get(date_key) is None and info_dict.get(ts_key) is not None:
2019 # Working around out-of-range timestamp values (e.g. negative ones on Windows,
2020 # see http://bugs.python.org/issue1646728)
2021 try:
2022 upload_date = datetime.datetime.utcfromtimestamp(info_dict[ts_key])
2023 info_dict[date_key] = upload_date.strftime('%Y%m%d')
2024 except (ValueError, OverflowError, OSError):
2025 pass
9d2ecdbc 2026
33d2fc2f
S
2027 # Auto generate title fields corresponding to the *_number fields when missing
2028 # in order to always have clean titles. This is very common for TV series.
2029 for field in ('chapter', 'season', 'episode'):
2030 if info_dict.get('%s_number' % field) is not None and not info_dict.get(field):
2031 info_dict[field] = '%s %d' % (field.capitalize(), info_dict['%s_number' % field])
2032
05108a49
S
2033 for cc_kind in ('subtitles', 'automatic_captions'):
2034 cc = info_dict.get(cc_kind)
2035 if cc:
2036 for _, subtitle in cc.items():
2037 for subtitle_format in subtitle:
2038 if subtitle_format.get('url'):
2039 subtitle_format['url'] = sanitize_url(subtitle_format['url'])
2040 if subtitle_format.get('ext') is None:
2041 subtitle_format['ext'] = determine_ext(subtitle_format['url']).lower()
2042
2043 automatic_captions = info_dict.get('automatic_captions')
4bba3716 2044 subtitles = info_dict.get('subtitles')
4bba3716 2045
360e1ca5 2046 info_dict['requested_subtitles'] = self.process_subtitles(
05108a49 2047 info_dict['id'], subtitles, automatic_captions)
a504ced0 2048
dd82ffea
JMF
2049 # We now pick which formats have to be downloaded
2050 if info_dict.get('formats') is None:
2051 # There's only one format available
2052 formats = [info_dict]
2053 else:
2054 formats = info_dict['formats']
2055
db95dc13 2056 if not formats:
b7da73eb 2057 if not self.params.get('ignore_no_formats_error'):
2058 raise ExtractorError('No video formats found!')
2059 else:
2060 self.report_warning('No video formats found!')
db95dc13 2061
73af5cc8
S
2062 def is_wellformed(f):
2063 url = f.get('url')
a5ac0c47 2064 if not url:
73af5cc8
S
2065 self.report_warning(
2066 '"url" field is missing or empty - skipping format, '
2067 'there is an error in extractor')
a5ac0c47
S
2068 return False
2069 if isinstance(url, bytes):
2070 sanitize_string_field(f, 'url')
2071 return True
73af5cc8
S
2072
2073 # Filter out malformed formats for better extraction robustness
2074 formats = list(filter(is_wellformed, formats))
2075
181c7053
S
2076 formats_dict = {}
2077
dd82ffea 2078 # We check that all the formats have the format and format_id fields
db95dc13 2079 for i, format in enumerate(formats):
c9969434
S
2080 sanitize_string_field(format, 'format_id')
2081 sanitize_numeric_fields(format)
dcf77cf1 2082 format['url'] = sanitize_url(format['url'])
e74e3b63 2083 if not format.get('format_id'):
8016c922 2084 format['format_id'] = compat_str(i)
e2effb08
S
2085 else:
2086 # Sanitize format_id from characters used in format selector expression
ec85ded8 2087 format['format_id'] = re.sub(r'[\s,/+\[\]()]', '_', format['format_id'])
181c7053
S
2088 format_id = format['format_id']
2089 if format_id not in formats_dict:
2090 formats_dict[format_id] = []
2091 formats_dict[format_id].append(format)
2092
2093 # Make sure all formats have unique format_id
2094 for format_id, ambiguous_formats in formats_dict.items():
2095 if len(ambiguous_formats) > 1:
2096 for i, format in enumerate(ambiguous_formats):
2097 format['format_id'] = '%s-%d' % (format_id, i)
2098
2099 for i, format in enumerate(formats):
8c51aa65 2100 if format.get('format') is None:
6febd1c1 2101 format['format'] = '{id} - {res}{note}'.format(
8c51aa65
JMF
2102 id=format['format_id'],
2103 res=self.format_resolution(format),
6febd1c1 2104 note=' ({0})'.format(format['format_note']) if format.get('format_note') is not None else '',
8c51aa65 2105 )
c1002e96 2106 # Automatically determine file extension if missing
5b1d8575 2107 if format.get('ext') is None:
cce929ea 2108 format['ext'] = determine_ext(format['url']).lower()
b5559424
S
2109 # Automatically determine protocol if missing (useful for format
2110 # selection purposes)
6f0be937 2111 if format.get('protocol') is None:
b5559424 2112 format['protocol'] = determine_protocol(format)
e5660ee6
JMF
2113 # Add HTTP headers, so that external programs can use them from the
2114 # json output
2115 full_format_info = info_dict.copy()
2116 full_format_info.update(format)
2117 format['http_headers'] = self._calc_headers(full_format_info)
0016b84e
S
2118 # Remove private housekeeping stuff
2119 if '__x_forwarded_for_ip' in info_dict:
2120 del info_dict['__x_forwarded_for_ip']
dd82ffea 2121
4bcc7bd1 2122 # TODO Central sorting goes here
99e206d5 2123
b7da73eb 2124 if formats and formats[0] is not info_dict:
b3d9ef88
JMF
2125 # only set the 'formats' fields if the original info_dict list them
2126 # otherwise we end up with a circular reference, the first (and unique)
f89197d7 2127 # element in the 'formats' field in info_dict is info_dict itself,
dfb1b146 2128 # which can't be exported to json
b3d9ef88 2129 info_dict['formats'] = formats
4ec82a72 2130
2131 info_dict, _ = self.pre_process(info_dict)
2132
169dbde9 2133 list_only = self.params.get('list_thumbnails') or self.params.get('listformats') or self.params.get('listsubtitles')
2134 if list_only:
2135 self.__forced_printings(info_dict, self.prepare_filename(info_dict), incomplete=True)
2136 if self.params.get('list_thumbnails'):
2137 self.list_thumbnails(info_dict)
2138 if self.params.get('listformats'):
2139 if not info_dict.get('formats'):
2140 raise ExtractorError('No video formats found', expected=True)
2141 self.list_formats(info_dict)
2142 if self.params.get('listsubtitles'):
2143 if 'automatic_captions' in info_dict:
2144 self.list_subtitles(
2145 info_dict['id'], automatic_captions, 'automatic captions')
2146 self.list_subtitles(info_dict['id'], subtitles, 'subtitles')
bfaae0a7 2147 return
2148
187986a8 2149 format_selector = self.format_selector
2150 if format_selector is None:
0017d9ad 2151 req_format = self._default_format_spec(info_dict, download=download)
0760b0a7 2152 self.write_debug('Default format spec: %s' % req_format)
187986a8 2153 format_selector = self.build_format_selector(req_format)
317f7ab6
S
2154
2155 # While in format selection we may need to have an access to the original
2156 # format set in order to calculate some metrics or do some processing.
2157 # For now we need to be able to guess whether original formats provided
2158 # by extractor are incomplete or not (i.e. whether extractor provides only
2159 # video-only or audio-only formats) for proper formats selection for
2160 # extractors with such incomplete formats (see
067aa17e 2161 # https://github.com/ytdl-org/youtube-dl/pull/5556).
317f7ab6
S
2162 # Since formats may be filtered during format selection and may not match
2163 # the original formats the results may be incorrect. Thus original formats
2164 # or pre-calculated metrics should be passed to format selection routines
2165 # as well.
2166 # We will pass a context object containing all necessary additional data
2167 # instead of just formats.
2168 # This fixes incorrect format selection issue (see
067aa17e 2169 # https://github.com/ytdl-org/youtube-dl/issues/10083).
2e221ca3 2170 incomplete_formats = (
317f7ab6 2171 # All formats are video-only or
3089bc74 2172 all(f.get('vcodec') != 'none' and f.get('acodec') == 'none' for f in formats)
317f7ab6 2173 # all formats are audio-only
3089bc74 2174 or all(f.get('vcodec') == 'none' and f.get('acodec') != 'none' for f in formats))
317f7ab6
S
2175
2176 ctx = {
2177 'formats': formats,
2178 'incomplete_formats': incomplete_formats,
2179 }
2180
2181 formats_to_download = list(format_selector(ctx))
dd82ffea 2182 if not formats_to_download:
b7da73eb 2183 if not self.params.get('ignore_no_formats_error'):
2184 raise ExtractorError('Requested format is not available', expected=True)
2185 else:
2186 self.report_warning('Requested format is not available')
4513a41a
A
2187 # Process what we can, even without any available formats.
2188 self.process_info(dict(info_dict))
b7da73eb 2189 elif download:
2190 self.to_screen(
07cce701 2191 '[info] %s: Downloading %d format(s): %s' % (
2192 info_dict['id'], len(formats_to_download),
2193 ", ".join([f['format_id'] for f in formats_to_download])))
b7da73eb 2194 for fmt in formats_to_download:
dd82ffea 2195 new_info = dict(info_dict)
4ec82a72 2196 # Save a reference to the original info_dict so that it can be modified in process_info if needed
2197 new_info['__original_infodict'] = info_dict
b7da73eb 2198 new_info.update(fmt)
dd82ffea
JMF
2199 self.process_info(new_info)
2200 # We update the info dict with the best quality format (backwards compatibility)
b7da73eb 2201 if formats_to_download:
2202 info_dict.update(formats_to_download[-1])
dd82ffea
JMF
2203 return info_dict
2204
98c70d6f 2205 def process_subtitles(self, video_id, normal_subtitles, automatic_captions):
a504ced0 2206 """Select the requested subtitles and their format"""
98c70d6f
JMF
2207 available_subs = {}
2208 if normal_subtitles and self.params.get('writesubtitles'):
2209 available_subs.update(normal_subtitles)
2210 if automatic_captions and self.params.get('writeautomaticsub'):
2211 for lang, cap_info in automatic_captions.items():
360e1ca5
JMF
2212 if lang not in available_subs:
2213 available_subs[lang] = cap_info
2214
4d171848
JMF
2215 if (not self.params.get('writesubtitles') and not
2216 self.params.get('writeautomaticsub') or not
2217 available_subs):
2218 return None
a504ced0 2219
c32b0aab 2220 all_sub_langs = available_subs.keys()
a504ced0 2221 if self.params.get('allsubtitles', False):
c32b0aab 2222 requested_langs = all_sub_langs
2223 elif self.params.get('subtitleslangs', False):
2224 requested_langs = set()
2225 for lang in self.params.get('subtitleslangs'):
2226 if lang == 'all':
2227 requested_langs.update(all_sub_langs)
2228 continue
2229 discard = lang[0] == '-'
2230 if discard:
2231 lang = lang[1:]
2232 current_langs = filter(re.compile(lang + '$').match, all_sub_langs)
2233 if discard:
2234 for lang in current_langs:
2235 requested_langs.discard(lang)
2236 else:
2237 requested_langs.update(current_langs)
2238 elif 'en' in available_subs:
2239 requested_langs = ['en']
a504ced0 2240 else:
c32b0aab 2241 requested_langs = [list(all_sub_langs)[0]]
e632bce2 2242 self.write_debug('Downloading subtitles: %s' % ', '.join(requested_langs))
a504ced0
JMF
2243
2244 formats_query = self.params.get('subtitlesformat', 'best')
2245 formats_preference = formats_query.split('/') if formats_query else []
2246 subs = {}
2247 for lang in requested_langs:
2248 formats = available_subs.get(lang)
2249 if formats is None:
2250 self.report_warning('%s subtitles not available for %s' % (lang, video_id))
2251 continue
a504ced0
JMF
2252 for ext in formats_preference:
2253 if ext == 'best':
2254 f = formats[-1]
2255 break
2256 matches = list(filter(lambda f: f['ext'] == ext, formats))
2257 if matches:
2258 f = matches[-1]
2259 break
2260 else:
2261 f = formats[-1]
2262 self.report_warning(
2263 'No subtitle format found matching "%s" for language %s, '
2264 'using %s' % (formats_query, lang, f['ext']))
2265 subs[lang] = f
2266 return subs
2267
d06daf23 2268 def __forced_printings(self, info_dict, filename, incomplete):
53c18592 2269 def print_mandatory(field, actual_field=None):
2270 if actual_field is None:
2271 actual_field = field
d06daf23 2272 if (self.params.get('force%s' % field, False)
53c18592 2273 and (not incomplete or info_dict.get(actual_field) is not None)):
2274 self.to_stdout(info_dict[actual_field])
d06daf23
S
2275
2276 def print_optional(field):
2277 if (self.params.get('force%s' % field, False)
2278 and info_dict.get(field) is not None):
2279 self.to_stdout(info_dict[field])
2280
53c18592 2281 info_dict = info_dict.copy()
2282 if filename is not None:
2283 info_dict['filename'] = filename
2284 if info_dict.get('requested_formats') is not None:
2285 # For RTMP URLs, also include the playpath
2286 info_dict['urls'] = '\n'.join(f['url'] + f.get('play_path', '') for f in info_dict['requested_formats'])
2287 elif 'url' in info_dict:
2288 info_dict['urls'] = info_dict['url'] + info_dict.get('play_path', '')
2289
2290 for tmpl in self.params.get('forceprint', []):
2291 if re.match(r'\w+$', tmpl):
2292 tmpl = '%({})s'.format(tmpl)
2293 tmpl, info_copy = self.prepare_outtmpl(tmpl, info_dict)
2294 self.to_stdout(tmpl % info_copy)
2295
d06daf23
S
2296 print_mandatory('title')
2297 print_mandatory('id')
53c18592 2298 print_mandatory('url', 'urls')
d06daf23
S
2299 print_optional('thumbnail')
2300 print_optional('description')
53c18592 2301 print_optional('filename')
d06daf23
S
2302 if self.params.get('forceduration', False) and info_dict.get('duration') is not None:
2303 self.to_stdout(formatSeconds(info_dict['duration']))
2304 print_mandatory('format')
53c18592 2305
d06daf23 2306 if self.params.get('forcejson', False):
277d6ff5 2307 self.post_extract(info_dict)
75d43ca0 2308 self.to_stdout(json.dumps(info_dict, default=repr))
d06daf23 2309
e8e73840 2310 def dl(self, name, info, subtitle=False, test=False):
2311
2312 if test:
2313 verbose = self.params.get('verbose')
2314 params = {
2315 'test': True,
2316 'quiet': not verbose,
2317 'verbose': verbose,
2318 'noprogress': not verbose,
2319 'nopart': True,
2320 'skip_unavailable_fragments': False,
2321 'keep_fragments': False,
2322 'overwrites': True,
2323 '_no_ytdl_file': True,
2324 }
2325 else:
2326 params = self.params
2327 fd = get_suitable_downloader(info, params)(self, params)
2328 if not test:
2329 for ph in self._progress_hooks:
2330 fd.add_progress_hook(ph)
18e674b4 2331 urls = '", "'.join([f['url'] for f in info.get('requested_formats', [])] or [info['url']])
2332 self.write_debug('Invoking downloader on "%s"' % urls)
e8e73840 2333 new_info = dict(info)
2334 if new_info.get('http_headers') is None:
2335 new_info['http_headers'] = self._calc_headers(new_info)
2336 return fd.download(name, new_info, subtitle)
2337
8222d8de
JMF
2338 def process_info(self, info_dict):
2339 """Process a single resolved IE result."""
2340
2341 assert info_dict.get('_type', 'video') == 'video'
fd288278 2342
0202b52a 2343 info_dict.setdefault('__postprocessors', [])
2344
fd288278
PH
2345 max_downloads = self.params.get('max_downloads')
2346 if max_downloads is not None:
2347 if self._num_downloads >= int(max_downloads):
2348 raise MaxDownloadsReached()
8222d8de 2349
d06daf23 2350 # TODO: backward compatibility, to be removed
8222d8de 2351 info_dict['fulltitle'] = info_dict['title']
8222d8de 2352
4513a41a 2353 if 'format' not in info_dict and 'ext' in info_dict:
8222d8de
JMF
2354 info_dict['format'] = info_dict['ext']
2355
c77495e3 2356 if self._match_entry(info_dict) is not None:
8222d8de
JMF
2357 return
2358
277d6ff5 2359 self.post_extract(info_dict)
fd288278 2360 self._num_downloads += 1
8222d8de 2361
dcf64d43 2362 # info_dict['_filename'] needs to be set for backward compatibility
de6000d9 2363 info_dict['_filename'] = full_filename = self.prepare_filename(info_dict, warn=True)
2364 temp_filename = self.prepare_filename(info_dict, 'temp')
0202b52a 2365 files_to_move = {}
8222d8de
JMF
2366
2367 # Forced printings
4513a41a 2368 self.__forced_printings(info_dict, full_filename, incomplete=('format' not in info_dict))
8222d8de 2369
8222d8de 2370 if self.params.get('simulate', False):
2d30509f 2371 if self.params.get('force_write_download_archive', False):
2372 self.record_download_archive(info_dict)
2373
2374 # Do nothing else if in simulate mode
8222d8de
JMF
2375 return
2376
de6000d9 2377 if full_filename is None:
8222d8de
JMF
2378 return
2379
e92caff5 2380 if not self._ensure_dir_exists(encodeFilename(full_filename)):
0202b52a 2381 return
e92caff5 2382 if not self._ensure_dir_exists(encodeFilename(temp_filename)):
8222d8de
JMF
2383 return
2384
2385 if self.params.get('writedescription', False):
de6000d9 2386 descfn = self.prepare_filename(info_dict, 'description')
e92caff5 2387 if not self._ensure_dir_exists(encodeFilename(descfn)):
0202b52a 2388 return
0c3d0f51 2389 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(descfn)):
6febd1c1 2390 self.to_screen('[info] Video description is already present')
f00fd51d
JMF
2391 elif info_dict.get('description') is None:
2392 self.report_warning('There\'s no description to write.')
7b6fefc9
PH
2393 else:
2394 try:
6febd1c1 2395 self.to_screen('[info] Writing video description to: ' + descfn)
7b6fefc9
PH
2396 with io.open(encodeFilename(descfn), 'w', encoding='utf-8') as descfile:
2397 descfile.write(info_dict['description'])
7b6fefc9 2398 except (OSError, IOError):
6febd1c1 2399 self.report_error('Cannot write description file ' + descfn)
7b6fefc9 2400 return
8222d8de 2401
1fb07d10 2402 if self.params.get('writeannotations', False):
de6000d9 2403 annofn = self.prepare_filename(info_dict, 'annotation')
e92caff5 2404 if not self._ensure_dir_exists(encodeFilename(annofn)):
0202b52a 2405 return
0c3d0f51 2406 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(annofn)):
6febd1c1 2407 self.to_screen('[info] Video annotations are already present')
ffddb112
RA
2408 elif not info_dict.get('annotations'):
2409 self.report_warning('There are no annotations to write.')
7b6fefc9
PH
2410 else:
2411 try:
6febd1c1 2412 self.to_screen('[info] Writing video annotations to: ' + annofn)
7b6fefc9
PH
2413 with io.open(encodeFilename(annofn), 'w', encoding='utf-8') as annofile:
2414 annofile.write(info_dict['annotations'])
2415 except (KeyError, TypeError):
6febd1c1 2416 self.report_warning('There are no annotations to write.')
7b6fefc9 2417 except (OSError, IOError):
6febd1c1 2418 self.report_error('Cannot write annotations file: ' + annofn)
7b6fefc9 2419 return
1fb07d10 2420
c4a91be7 2421 subtitles_are_requested = any([self.params.get('writesubtitles', False),
0b7f3118 2422 self.params.get('writeautomaticsub')])
c4a91be7 2423
c84dd8a9 2424 if subtitles_are_requested and info_dict.get('requested_subtitles'):
8222d8de
JMF
2425 # subtitles download errors are already managed as troubles in relevant IE
2426 # that way it will silently go on when used with unsupporting IE
c84dd8a9 2427 subtitles = info_dict['requested_subtitles']
fa57af1e 2428 # ie = self.get_info_extractor(info_dict['extractor_key'])
a504ced0
JMF
2429 for sub_lang, sub_info in subtitles.items():
2430 sub_format = sub_info['ext']
56d868db 2431 sub_filename = subtitles_filename(temp_filename, sub_lang, sub_format, info_dict.get('ext'))
2432 sub_filename_final = subtitles_filename(
2433 self.prepare_filename(info_dict, 'subtitle'), sub_lang, sub_format, info_dict.get('ext'))
0c3d0f51 2434 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(sub_filename)):
5ff1bc0c 2435 self.to_screen('[info] Video subtitle %s.%s is already present' % (sub_lang, sub_format))
dcf64d43 2436 sub_info['filepath'] = sub_filename
0202b52a 2437 files_to_move[sub_filename] = sub_filename_final
a504ced0 2438 else:
0c9df79e 2439 self.to_screen('[info] Writing video subtitles to: ' + sub_filename)
5ff1bc0c
RA
2440 if sub_info.get('data') is not None:
2441 try:
2442 # Use newline='' to prevent conversion of newline characters
067aa17e 2443 # See https://github.com/ytdl-org/youtube-dl/issues/10268
5ff1bc0c
RA
2444 with io.open(encodeFilename(sub_filename), 'w', encoding='utf-8', newline='') as subfile:
2445 subfile.write(sub_info['data'])
dcf64d43 2446 sub_info['filepath'] = sub_filename
0202b52a 2447 files_to_move[sub_filename] = sub_filename_final
5ff1bc0c
RA
2448 except (OSError, IOError):
2449 self.report_error('Cannot write subtitles file ' + sub_filename)
2450 return
7b6fefc9 2451 else:
5ff1bc0c 2452 try:
e8e73840 2453 self.dl(sub_filename, sub_info.copy(), subtitle=True)
dcf64d43 2454 sub_info['filepath'] = sub_filename
0202b52a 2455 files_to_move[sub_filename] = sub_filename_final
fe346461 2456 except (ExtractorError, IOError, OSError, ValueError) + network_exceptions as err:
5ff1bc0c
RA
2457 self.report_warning('Unable to download subtitle for "%s": %s' %
2458 (sub_lang, error_to_compat_str(err)))
2459 continue
8222d8de 2460
8222d8de 2461 if self.params.get('writeinfojson', False):
de6000d9 2462 infofn = self.prepare_filename(info_dict, 'infojson')
e92caff5 2463 if not self._ensure_dir_exists(encodeFilename(infofn)):
0202b52a 2464 return
0c3d0f51 2465 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(infofn)):
66c935fb 2466 self.to_screen('[info] Video metadata is already present')
7b6fefc9 2467 else:
66c935fb 2468 self.to_screen('[info] Writing video metadata as JSON to: ' + infofn)
7b6fefc9 2469 try:
75d43ca0 2470 write_json_file(self.filter_requested_info(info_dict, self.params.get('clean_infojson', True)), infofn)
7b6fefc9 2471 except (OSError, IOError):
66c935fb 2472 self.report_error('Cannot write video metadata to JSON file ' + infofn)
7b6fefc9 2473 return
de6000d9 2474 info_dict['__infojson_filename'] = infofn
8222d8de 2475
56d868db 2476 for thumb_ext in self._write_thumbnails(info_dict, temp_filename):
2477 thumb_filename_temp = replace_extension(temp_filename, thumb_ext, info_dict.get('ext'))
2478 thumb_filename = replace_extension(
2479 self.prepare_filename(info_dict, 'thumbnail'), thumb_ext, info_dict.get('ext'))
dcf64d43 2480 files_to_move[thumb_filename_temp] = thumb_filename
8222d8de 2481
732044af 2482 # Write internet shortcut files
2483 url_link = webloc_link = desktop_link = False
2484 if self.params.get('writelink', False):
2485 if sys.platform == "darwin": # macOS.
2486 webloc_link = True
2487 elif sys.platform.startswith("linux"):
2488 desktop_link = True
2489 else: # if sys.platform in ['win32', 'cygwin']:
2490 url_link = True
2491 if self.params.get('writeurllink', False):
2492 url_link = True
2493 if self.params.get('writewebloclink', False):
2494 webloc_link = True
2495 if self.params.get('writedesktoplink', False):
2496 desktop_link = True
2497
2498 if url_link or webloc_link or desktop_link:
2499 if 'webpage_url' not in info_dict:
2500 self.report_error('Cannot write internet shortcut file because the "webpage_url" field is missing in the media information')
2501 return
2502 ascii_url = iri_to_uri(info_dict['webpage_url'])
2503
2504 def _write_link_file(extension, template, newline, embed_filename):
0202b52a 2505 linkfn = replace_extension(full_filename, extension, info_dict.get('ext'))
10e3742e 2506 if self.params.get('overwrites', True) and os.path.exists(encodeFilename(linkfn)):
732044af 2507 self.to_screen('[info] Internet shortcut is already present')
2508 else:
2509 try:
2510 self.to_screen('[info] Writing internet shortcut to: ' + linkfn)
2511 with io.open(encodeFilename(to_high_limit_path(linkfn)), 'w', encoding='utf-8', newline=newline) as linkfile:
2512 template_vars = {'url': ascii_url}
2513 if embed_filename:
2514 template_vars['filename'] = linkfn[:-(len(extension) + 1)]
2515 linkfile.write(template % template_vars)
2516 except (OSError, IOError):
2517 self.report_error('Cannot write internet shortcut ' + linkfn)
2518 return False
2519 return True
2520
2521 if url_link:
2522 if not _write_link_file('url', DOT_URL_LINK_TEMPLATE, '\r\n', embed_filename=False):
2523 return
2524 if webloc_link:
2525 if not _write_link_file('webloc', DOT_WEBLOC_LINK_TEMPLATE, '\n', embed_filename=False):
2526 return
2527 if desktop_link:
2528 if not _write_link_file('desktop', DOT_DESKTOP_LINK_TEMPLATE, '\n', embed_filename=True):
2529 return
2530
56d868db 2531 try:
2532 info_dict, files_to_move = self.pre_process(info_dict, 'before_dl', files_to_move)
2533 except PostProcessingError as err:
2534 self.report_error('Preprocessing: %s' % str(err))
2535 return
2536
732044af 2537 must_record_download_archive = False
56d868db 2538 if self.params.get('skip_download', False):
2539 info_dict['filepath'] = temp_filename
2540 info_dict['__finaldir'] = os.path.dirname(os.path.abspath(encodeFilename(full_filename)))
2541 info_dict['__files_to_move'] = files_to_move
2542 info_dict = self.run_pp(MoveFilesAfterDownloadPP(self, False), info_dict)
2543 else:
2544 # Download
4340deca 2545 try:
0202b52a 2546
6b591b29 2547 def existing_file(*filepaths):
2548 ext = info_dict.get('ext')
2549 final_ext = self.params.get('final_ext', ext)
2550 existing_files = []
2551 for file in orderedSet(filepaths):
2552 if final_ext != ext:
2553 converted = replace_extension(file, final_ext, ext)
2554 if os.path.exists(encodeFilename(converted)):
2555 existing_files.append(converted)
2556 if os.path.exists(encodeFilename(file)):
2557 existing_files.append(file)
2558
2559 if not existing_files or self.params.get('overwrites', False):
2560 for file in orderedSet(existing_files):
2561 self.report_file_delete(file)
2562 os.remove(encodeFilename(file))
2563 return None
2564
2565 self.report_file_already_downloaded(existing_files[0])
2566 info_dict['ext'] = os.path.splitext(existing_files[0])[1][1:]
2567 return existing_files[0]
0202b52a 2568
2569 success = True
4340deca 2570 if info_dict.get('requested_formats') is not None:
81cd954a
S
2571
2572 def compatible_formats(formats):
d03cfdce 2573 # TODO: some formats actually allow this (mkv, webm, ogg, mp4), but not all of them.
2574 video_formats = [format for format in formats if format.get('vcodec') != 'none']
2575 audio_formats = [format for format in formats if format.get('acodec') != 'none']
2576 if len(video_formats) > 2 or len(audio_formats) > 2:
2577 return False
2578
81cd954a 2579 # Check extension
d03cfdce 2580 exts = set(format.get('ext') for format in formats)
2581 COMPATIBLE_EXTS = (
2582 set(('mp3', 'mp4', 'm4a', 'm4p', 'm4b', 'm4r', 'm4v', 'ismv', 'isma')),
2583 set(('webm',)),
2584 )
2585 for ext_sets in COMPATIBLE_EXTS:
2586 if ext_sets.issuperset(exts):
2587 return True
81cd954a
S
2588 # TODO: Check acodec/vcodec
2589 return False
2590
2591 requested_formats = info_dict['requested_formats']
0202b52a 2592 old_ext = info_dict['ext']
4d971a16 2593 if self.params.get('merge_output_format') is None:
2594 if not compatible_formats(requested_formats):
2595 info_dict['ext'] = 'mkv'
2596 self.report_warning(
2597 'Requested formats are incompatible for merge and will be merged into mkv.')
2598 if (info_dict['ext'] == 'webm'
2599 and self.params.get('writethumbnail', False)
2600 and info_dict.get('thumbnails')):
2601 info_dict['ext'] = 'mkv'
2602 self.report_warning(
2603 'webm doesn\'t support embedding a thumbnail, mkv will be used.')
0202b52a 2604
2605 def correct_ext(filename):
2606 filename_real_ext = os.path.splitext(filename)[1][1:]
2607 filename_wo_ext = (
2608 os.path.splitext(filename)[0]
2609 if filename_real_ext == old_ext
2610 else filename)
2611 return '%s.%s' % (filename_wo_ext, info_dict['ext'])
2612
38c6902b 2613 # Ensure filename always has a correct extension for successful merge
0202b52a 2614 full_filename = correct_ext(full_filename)
2615 temp_filename = correct_ext(temp_filename)
2616 dl_filename = existing_file(full_filename, temp_filename)
1ea24129 2617 info_dict['__real_download'] = False
18e674b4 2618
2619 _protocols = set(determine_protocol(f) for f in requested_formats)
2620 if len(_protocols) == 1:
2621 info_dict['protocol'] = _protocols.pop()
2622 directly_mergable = (
2623 'no-direct-merge' not in self.params.get('compat_opts', [])
2624 and info_dict.get('protocol') is not None # All requested formats have same protocol
2625 and not self.params.get('allow_unplayable_formats')
2626 and get_suitable_downloader(info_dict, self.params).__name__ == 'FFmpegFD')
2627 if directly_mergable:
2628 info_dict['url'] = requested_formats[0]['url']
2629 # Treat it as a single download
2630 dl_filename = existing_file(full_filename, temp_filename)
2631 if dl_filename is None:
2632 success, real_download = self.dl(temp_filename, info_dict)
2633 info_dict['__real_download'] = real_download
2634 else:
2635 downloaded = []
2636 merger = FFmpegMergerPP(self)
2637 if self.params.get('allow_unplayable_formats'):
2638 self.report_warning(
2639 'You have requested merging of multiple formats '
2640 'while also allowing unplayable formats to be downloaded. '
2641 'The formats won\'t be merged to prevent data corruption.')
2642 elif not merger.available:
2643 self.report_warning(
2644 'You have requested merging of multiple formats but ffmpeg is not installed. '
2645 'The formats won\'t be merged.')
2646
2647 if dl_filename is None:
2648 for f in requested_formats:
2649 new_info = dict(info_dict)
2650 del new_info['requested_formats']
2651 new_info.update(f)
2652 fname = prepend_extension(
2653 self.prepare_filename(new_info, 'temp'),
2654 'f%s' % f['format_id'], new_info['ext'])
2655 if not self._ensure_dir_exists(fname):
2656 return
2657 downloaded.append(fname)
2658 partial_success, real_download = self.dl(fname, new_info)
2659 info_dict['__real_download'] = info_dict['__real_download'] or real_download
2660 success = success and partial_success
2661 if merger.available and not self.params.get('allow_unplayable_formats'):
2662 info_dict['__postprocessors'].append(merger)
2663 info_dict['__files_to_merge'] = downloaded
2664 # Even if there were no downloads, it is being merged only now
2665 info_dict['__real_download'] = True
2666 else:
2667 for file in downloaded:
2668 files_to_move[file] = None
4340deca
P
2669 else:
2670 # Just a single file
0202b52a 2671 dl_filename = existing_file(full_filename, temp_filename)
2672 if dl_filename is None:
e8e73840 2673 success, real_download = self.dl(temp_filename, info_dict)
0202b52a 2674 info_dict['__real_download'] = real_download
2675
0202b52a 2676 dl_filename = dl_filename or temp_filename
c571435f 2677 info_dict['__finaldir'] = os.path.dirname(os.path.abspath(encodeFilename(full_filename)))
0202b52a 2678
3158150c 2679 except network_exceptions as err:
7960b056 2680 self.report_error('unable to download video data: %s' % error_to_compat_str(err))
4340deca
P
2681 return
2682 except (OSError, IOError) as err:
2683 raise UnavailableVideoError(err)
2684 except (ContentTooShortError, ) as err:
2685 self.report_error('content too short (expected %s bytes and served %s)' % (err.expected, err.downloaded))
2686 return
8222d8de 2687
de6000d9 2688 if success and full_filename != '-':
f17f8651 2689
fd7cfb64 2690 def fixup():
2691 do_fixup = True
2692 fixup_policy = self.params.get('fixup')
2693 vid = info_dict['id']
2694
2695 if fixup_policy in ('ignore', 'never'):
2696 return
2697 elif fixup_policy == 'warn':
2698 do_fixup = False
f89b3e2d 2699 elif fixup_policy != 'force':
2700 assert fixup_policy in ('detect_or_warn', None)
2701 if not info_dict.get('__real_download'):
2702 do_fixup = False
fd7cfb64 2703
2704 def ffmpeg_fixup(cndn, msg, cls):
2705 if not cndn:
2706 return
2707 if not do_fixup:
2708 self.report_warning(f'{vid}: {msg}')
2709 return
2710 pp = cls(self)
2711 if pp.available:
2712 info_dict['__postprocessors'].append(pp)
2713 else:
2714 self.report_warning(f'{vid}: {msg}. Install ffmpeg to fix this automatically')
2715
2716 stretched_ratio = info_dict.get('stretched_ratio')
2717 ffmpeg_fixup(
2718 stretched_ratio not in (1, None),
2719 f'Non-uniform pixel ratio {stretched_ratio}',
2720 FFmpegFixupStretchedPP)
2721
2722 ffmpeg_fixup(
2723 (info_dict.get('requested_formats') is None
2724 and info_dict.get('container') == 'm4a_dash'
2725 and info_dict.get('ext') == 'm4a'),
2726 'writing DASH m4a. Only some players support this container',
2727 FFmpegFixupM4aPP)
2728
2729 downloader = (get_suitable_downloader(info_dict, self.params).__name__
2730 if 'protocol' in info_dict else None)
2731 ffmpeg_fixup(downloader == 'HlsFD', 'malformed AAC bitstream detected', FFmpegFixupM3u8PP)
e36d50c5 2732 ffmpeg_fixup(downloader == 'WebSocketFragmentFD', 'malformed timestamps detected', FFmpegFixupTimestampPP)
2733 ffmpeg_fixup(downloader == 'WebSocketFragmentFD', 'malformed duration detected', FFmpegFixupDurationPP)
fd7cfb64 2734
2735 fixup()
8222d8de 2736 try:
23c1a667 2737 info_dict = self.post_process(dl_filename, info_dict, files_to_move)
af819c21 2738 except PostProcessingError as err:
2739 self.report_error('Postprocessing: %s' % str(err))
8222d8de 2740 return
ab8e5e51
AM
2741 try:
2742 for ph in self._post_hooks:
23c1a667 2743 ph(info_dict['filepath'])
ab8e5e51
AM
2744 except Exception as err:
2745 self.report_error('post hooks: %s' % str(err))
2746 return
2d30509f 2747 must_record_download_archive = True
2748
2749 if must_record_download_archive or self.params.get('force_write_download_archive', False):
2750 self.record_download_archive(info_dict)
c3e6ffba 2751 max_downloads = self.params.get('max_downloads')
2752 if max_downloads is not None and self._num_downloads >= int(max_downloads):
2753 raise MaxDownloadsReached()
8222d8de
JMF
2754
2755 def download(self, url_list):
2756 """Download a given list of URLs."""
de6000d9 2757 outtmpl = self.outtmpl_dict['default']
3089bc74
S
2758 if (len(url_list) > 1
2759 and outtmpl != '-'
2760 and '%' not in outtmpl
2761 and self.params.get('max_downloads') != 1):
acd69589 2762 raise SameFileError(outtmpl)
8222d8de
JMF
2763
2764 for url in url_list:
2765 try:
5f6a1245 2766 # It also downloads the videos
61aa5ba3
S
2767 res = self.extract_info(
2768 url, force_generic_extractor=self.params.get('force_generic_extractor', False))
8222d8de 2769 except UnavailableVideoError:
6febd1c1 2770 self.report_error('unable to download video')
8222d8de 2771 except MaxDownloadsReached:
8b0d7497 2772 self.to_screen('[info] Maximum number of downloaded files reached')
2773 raise
2774 except ExistingVideoReached:
d83cb531 2775 self.to_screen('[info] Encountered a file that is already in the archive, stopping due to --break-on-existing')
8b0d7497 2776 raise
2777 except RejectedVideoReached:
d83cb531 2778 self.to_screen('[info] Encountered a file that did not match filter, stopping due to --break-on-reject')
8222d8de 2779 raise
63e0be34
PH
2780 else:
2781 if self.params.get('dump_single_json', False):
277d6ff5 2782 self.post_extract(res)
75d43ca0 2783 self.to_stdout(json.dumps(res, default=repr))
8222d8de
JMF
2784
2785 return self._download_retcode
2786
1dcc4c0c 2787 def download_with_info_file(self, info_filename):
31bd3925
JMF
2788 with contextlib.closing(fileinput.FileInput(
2789 [info_filename], mode='r',
2790 openhook=fileinput.hook_encoded('utf-8'))) as f:
2791 # FileInput doesn't have a read method, we can't call json.load
498f5606 2792 info = self.filter_requested_info(json.loads('\n'.join(f)), self.params.get('clean_infojson', True))
d4943898
JMF
2793 try:
2794 self.process_ie_result(info, download=True)
d3f62c19 2795 except (DownloadError, EntryNotInPlaylist, ThrottledDownload):
d4943898
JMF
2796 webpage_url = info.get('webpage_url')
2797 if webpage_url is not None:
6febd1c1 2798 self.report_warning('The info failed to download, trying with "%s"' % webpage_url)
d4943898
JMF
2799 return self.download([webpage_url])
2800 else:
2801 raise
2802 return self._download_retcode
1dcc4c0c 2803
cb202fd2 2804 @staticmethod
75d43ca0 2805 def filter_requested_info(info_dict, actually_filter=True):
ae8f99e6 2806 remove_keys = ['__original_infodict'] # Always remove this since this may contain a copy of the entire dict
2807 keep_keys = ['_type'], # Always keep this to facilitate load-info-json
2808 if actually_filter:
bd99f6e6 2809 remove_keys += ('requested_formats', 'requested_subtitles', 'requested_entries', 'filepath', 'entries', 'original_url')
ae8f99e6 2810 empty_values = (None, {}, [], set(), tuple())
2811 reject = lambda k, v: k not in keep_keys and (
2812 k.startswith('_') or k in remove_keys or v in empty_values)
2813 else:
394dcd44 2814 info_dict['epoch'] = int(time.time())
ae8f99e6 2815 reject = lambda k, v: k in remove_keys
5226731e 2816 filter_fn = lambda obj: (
b0249bca 2817 list(map(filter_fn, obj)) if isinstance(obj, (LazyList, list, tuple, set))
a515a78d 2818 else obj if not isinstance(obj, dict)
ae8f99e6 2819 else dict((k, filter_fn(v)) for k, v in obj.items() if not reject(k, v)))
5226731e 2820 return filter_fn(info_dict)
cb202fd2 2821
dcf64d43 2822 def run_pp(self, pp, infodict):
5bfa4862 2823 files_to_delete = []
dcf64d43 2824 if '__files_to_move' not in infodict:
2825 infodict['__files_to_move'] = {}
af819c21 2826 files_to_delete, infodict = pp.run(infodict)
5bfa4862 2827 if not files_to_delete:
dcf64d43 2828 return infodict
5bfa4862 2829
2830 if self.params.get('keepvideo', False):
2831 for f in files_to_delete:
dcf64d43 2832 infodict['__files_to_move'].setdefault(f, '')
5bfa4862 2833 else:
2834 for old_filename in set(files_to_delete):
2835 self.to_screen('Deleting original file %s (pass -k to keep)' % old_filename)
2836 try:
2837 os.remove(encodeFilename(old_filename))
2838 except (IOError, OSError):
2839 self.report_warning('Unable to remove downloaded original file')
dcf64d43 2840 if old_filename in infodict['__files_to_move']:
2841 del infodict['__files_to_move'][old_filename]
2842 return infodict
5bfa4862 2843
277d6ff5 2844 @staticmethod
2845 def post_extract(info_dict):
2846 def actual_post_extract(info_dict):
2847 if info_dict.get('_type') in ('playlist', 'multi_video'):
2848 for video_dict in info_dict.get('entries', {}):
b050d210 2849 actual_post_extract(video_dict or {})
277d6ff5 2850 return
2851
07cce701 2852 post_extractor = info_dict.get('__post_extractor') or (lambda: {})
4ec82a72 2853 extra = post_extractor().items()
2854 info_dict.update(extra)
07cce701 2855 info_dict.pop('__post_extractor', None)
277d6ff5 2856
4ec82a72 2857 original_infodict = info_dict.get('__original_infodict') or {}
2858 original_infodict.update(extra)
2859 original_infodict.pop('__post_extractor', None)
2860
b050d210 2861 actual_post_extract(info_dict or {})
277d6ff5 2862
56d868db 2863 def pre_process(self, ie_info, key='pre_process', files_to_move=None):
5bfa4862 2864 info = dict(ie_info)
56d868db 2865 info['__files_to_move'] = files_to_move or {}
2866 for pp in self._pps[key]:
dcf64d43 2867 info = self.run_pp(pp, info)
56d868db 2868 return info, info.pop('__files_to_move', None)
5bfa4862 2869
dcf64d43 2870 def post_process(self, filename, ie_info, files_to_move=None):
8222d8de
JMF
2871 """Run all the postprocessors on the given file."""
2872 info = dict(ie_info)
2873 info['filepath'] = filename
dcf64d43 2874 info['__files_to_move'] = files_to_move or {}
0202b52a 2875
56d868db 2876 for pp in ie_info.get('__postprocessors', []) + self._pps['post_process']:
dcf64d43 2877 info = self.run_pp(pp, info)
2878 info = self.run_pp(MoveFilesAfterDownloadPP(self), info)
2879 del info['__files_to_move']
56d868db 2880 for pp in self._pps['after_move']:
dcf64d43 2881 info = self.run_pp(pp, info)
23c1a667 2882 return info
c1c9a79c 2883
5db07df6 2884 def _make_archive_id(self, info_dict):
e9fef7ee
S
2885 video_id = info_dict.get('id')
2886 if not video_id:
2887 return
5db07df6
PH
2888 # Future-proof against any change in case
2889 # and backwards compatibility with prior versions
e9fef7ee 2890 extractor = info_dict.get('extractor_key') or info_dict.get('ie_key') # key in a playlist
7012b23c 2891 if extractor is None:
1211bb6d
S
2892 url = str_or_none(info_dict.get('url'))
2893 if not url:
2894 return
e9fef7ee
S
2895 # Try to find matching extractor for the URL and take its ie_key
2896 for ie in self._ies:
1211bb6d 2897 if ie.suitable(url):
e9fef7ee
S
2898 extractor = ie.ie_key()
2899 break
2900 else:
2901 return
d0757229 2902 return '%s %s' % (extractor.lower(), video_id)
5db07df6
PH
2903
2904 def in_download_archive(self, info_dict):
2905 fn = self.params.get('download_archive')
2906 if fn is None:
2907 return False
2908
2909 vid_id = self._make_archive_id(info_dict)
e9fef7ee 2910 if not vid_id:
7012b23c 2911 return False # Incomplete video information
5db07df6 2912
a45e8619 2913 return vid_id in self.archive
c1c9a79c
PH
2914
2915 def record_download_archive(self, info_dict):
2916 fn = self.params.get('download_archive')
2917 if fn is None:
2918 return
5db07df6
PH
2919 vid_id = self._make_archive_id(info_dict)
2920 assert vid_id
c1c9a79c 2921 with locked_file(fn, 'a', encoding='utf-8') as archive_file:
6febd1c1 2922 archive_file.write(vid_id + '\n')
a45e8619 2923 self.archive.add(vid_id)
dd82ffea 2924
8c51aa65 2925 @staticmethod
8abeeb94 2926 def format_resolution(format, default='unknown'):
fb04e403 2927 if format.get('vcodec') == 'none':
8326b00a 2928 if format.get('acodec') == 'none':
2929 return 'images'
fb04e403 2930 return 'audio only'
f49d89ee
PH
2931 if format.get('resolution') is not None:
2932 return format['resolution']
35615307
DA
2933 if format.get('width') and format.get('height'):
2934 res = '%dx%d' % (format['width'], format['height'])
2935 elif format.get('height'):
2936 res = '%sp' % format['height']
2937 elif format.get('width'):
388ae76b 2938 res = '%dx?' % format['width']
8c51aa65 2939 else:
8abeeb94 2940 res = default
8c51aa65
JMF
2941 return res
2942
c57f7757
PH
2943 def _format_note(self, fdict):
2944 res = ''
2945 if fdict.get('ext') in ['f4f', 'f4m']:
2946 res += '(unsupported) '
32f90364
PH
2947 if fdict.get('language'):
2948 if res:
2949 res += ' '
9016d76f 2950 res += '[%s] ' % fdict['language']
c57f7757
PH
2951 if fdict.get('format_note') is not None:
2952 res += fdict['format_note'] + ' '
2953 if fdict.get('tbr') is not None:
2954 res += '%4dk ' % fdict['tbr']
2955 if fdict.get('container') is not None:
2956 if res:
2957 res += ', '
2958 res += '%s container' % fdict['container']
3089bc74
S
2959 if (fdict.get('vcodec') is not None
2960 and fdict.get('vcodec') != 'none'):
c57f7757
PH
2961 if res:
2962 res += ', '
2963 res += fdict['vcodec']
91c7271a 2964 if fdict.get('vbr') is not None:
c57f7757
PH
2965 res += '@'
2966 elif fdict.get('vbr') is not None and fdict.get('abr') is not None:
2967 res += 'video@'
2968 if fdict.get('vbr') is not None:
2969 res += '%4dk' % fdict['vbr']
fbb21cf5 2970 if fdict.get('fps') is not None:
5d583bdf
S
2971 if res:
2972 res += ', '
2973 res += '%sfps' % fdict['fps']
c57f7757
PH
2974 if fdict.get('acodec') is not None:
2975 if res:
2976 res += ', '
2977 if fdict['acodec'] == 'none':
2978 res += 'video only'
2979 else:
2980 res += '%-5s' % fdict['acodec']
2981 elif fdict.get('abr') is not None:
2982 if res:
2983 res += ', '
2984 res += 'audio'
2985 if fdict.get('abr') is not None:
2986 res += '@%3dk' % fdict['abr']
2987 if fdict.get('asr') is not None:
2988 res += ' (%5dHz)' % fdict['asr']
2989 if fdict.get('filesize') is not None:
2990 if res:
2991 res += ', '
2992 res += format_bytes(fdict['filesize'])
9732d77e
PH
2993 elif fdict.get('filesize_approx') is not None:
2994 if res:
2995 res += ', '
2996 res += '~' + format_bytes(fdict['filesize_approx'])
c57f7757 2997 return res
91c7271a 2998
76d321f6 2999 def _format_note_table(self, f):
3000 def join_fields(*vargs):
3001 return ', '.join((val for val in vargs if val != ''))
3002
3003 return join_fields(
3004 'UNSUPPORTED' if f.get('ext') in ('f4f', 'f4m') else '',
3005 format_field(f, 'language', '[%s]'),
3006 format_field(f, 'format_note'),
3007 format_field(f, 'container', ignore=(None, f.get('ext'))),
3008 format_field(f, 'asr', '%5dHz'))
3009
c57f7757 3010 def list_formats(self, info_dict):
94badb25 3011 formats = info_dict.get('formats', [info_dict])
53ed7066 3012 new_format = (
3013 'list-formats' not in self.params.get('compat_opts', [])
169dbde9 3014 and self.params.get('listformats_table', True) is not False)
76d321f6 3015 if new_format:
3016 table = [
3017 [
3018 format_field(f, 'format_id'),
3019 format_field(f, 'ext'),
3020 self.format_resolution(f),
3021 format_field(f, 'fps', '%d'),
3022 '|',
3023 format_field(f, 'filesize', ' %s', func=format_bytes) + format_field(f, 'filesize_approx', '~%s', func=format_bytes),
3024 format_field(f, 'tbr', '%4dk'),
52a8a1e1 3025 shorten_protocol_name(f.get('protocol', '').replace("native", "n")),
76d321f6 3026 '|',
3027 format_field(f, 'vcodec', default='unknown').replace('none', ''),
3028 format_field(f, 'vbr', '%4dk'),
3029 format_field(f, 'acodec', default='unknown').replace('none', ''),
3030 format_field(f, 'abr', '%3dk'),
3031 format_field(f, 'asr', '%5dHz'),
3032 self._format_note_table(f)]
3033 for f in formats
3034 if f.get('preference') is None or f['preference'] >= -1000]
3035 header_line = ['ID', 'EXT', 'RESOLUTION', 'FPS', '|', ' FILESIZE', ' TBR', 'PROTO',
3036 '|', 'VCODEC', ' VBR', 'ACODEC', ' ABR', ' ASR', 'NOTE']
3037 else:
3038 table = [
3039 [
3040 format_field(f, 'format_id'),
3041 format_field(f, 'ext'),
3042 self.format_resolution(f),
3043 self._format_note(f)]
3044 for f in formats
3045 if f.get('preference') is None or f['preference'] >= -1000]
3046 header_line = ['format code', 'extension', 'resolution', 'note']
57dd9a8f 3047
cfb56d1a 3048 self.to_screen(
169dbde9 3049 '[info] Available formats for %s:' % info_dict['id'])
3050 self.to_stdout(render_table(
bc97cdae 3051 header_line, table, delim=new_format, extraGap=(0 if new_format else 1), hideEmpty=new_format))
cfb56d1a
PH
3052
3053 def list_thumbnails(self, info_dict):
b0249bca 3054 thumbnails = list(info_dict.get('thumbnails'))
cfb56d1a 3055 if not thumbnails:
b7b72db9 3056 self.to_screen('[info] No thumbnails present for %s' % info_dict['id'])
3057 return
cfb56d1a
PH
3058
3059 self.to_screen(
3060 '[info] Thumbnails for %s:' % info_dict['id'])
169dbde9 3061 self.to_stdout(render_table(
cfb56d1a
PH
3062 ['ID', 'width', 'height', 'URL'],
3063 [[t['id'], t.get('width', 'unknown'), t.get('height', 'unknown'), t['url']] for t in thumbnails]))
dca08720 3064
360e1ca5 3065 def list_subtitles(self, video_id, subtitles, name='subtitles'):
a504ced0 3066 if not subtitles:
360e1ca5 3067 self.to_screen('%s has no %s' % (video_id, name))
a504ced0 3068 return
a504ced0 3069 self.to_screen(
edab9dbf 3070 'Available %s for %s:' % (name, video_id))
2412044c 3071
3072 def _row(lang, formats):
49c258e1 3073 exts, names = zip(*((f['ext'], f.get('name') or 'unknown') for f in reversed(formats)))
2412044c 3074 if len(set(names)) == 1:
7aee40c1 3075 names = [] if names[0] == 'unknown' else names[:1]
2412044c 3076 return [lang, ', '.join(names), ', '.join(exts)]
3077
169dbde9 3078 self.to_stdout(render_table(
2412044c 3079 ['Language', 'Name', 'Formats'],
3080 [_row(lang, formats) for lang, formats in subtitles.items()],
3081 hideEmpty=True))
a504ced0 3082
dca08720
PH
3083 def urlopen(self, req):
3084 """ Start an HTTP download """
82d8a8b6 3085 if isinstance(req, compat_basestring):
67dda517 3086 req = sanitized_Request(req)
19a41fc6 3087 return self._opener.open(req, timeout=self._socket_timeout)
dca08720
PH
3088
3089 def print_debug_header(self):
3090 if not self.params.get('verbose'):
3091 return
62fec3b2 3092
4192b51c 3093 if type('') is not compat_str:
067aa17e 3094 # Python 2.6 on SLES11 SP1 (https://github.com/ytdl-org/youtube-dl/issues/3326)
4192b51c
PH
3095 self.report_warning(
3096 'Your Python is broken! Update to a newer and supported version')
3097
c6afed48
PH
3098 stdout_encoding = getattr(
3099 sys.stdout, 'encoding', 'missing (%s)' % type(sys.stdout).__name__)
b0472057 3100 encoding_str = (
734f90bb
PH
3101 '[debug] Encodings: locale %s, fs %s, out %s, pref %s\n' % (
3102 locale.getpreferredencoding(),
3103 sys.getfilesystemencoding(),
c6afed48 3104 stdout_encoding,
b0472057 3105 self.get_encoding()))
4192b51c 3106 write_string(encoding_str, encoding=None)
734f90bb 3107
e5813e53 3108 source = (
3109 '(exe)' if hasattr(sys, 'frozen')
3110 else '(zip)' if isinstance(globals().get('__loader__'), zipimporter)
3111 else '(source)' if os.path.basename(sys.argv[0]) == '__main__.py'
3112 else '')
3113 self._write_string('[debug] yt-dlp version %s %s\n' % (__version__, source))
e0986e31 3114 if _LAZY_LOADER:
f74980cb 3115 self._write_string('[debug] Lazy loading extractors enabled\n')
3116 if _PLUGIN_CLASSES:
3117 self._write_string(
3118 '[debug] Plugin Extractors: %s\n' % [ie.ie_key() for ie in _PLUGIN_CLASSES])
53ed7066 3119 if self.params.get('compat_opts'):
3120 self._write_string(
3121 '[debug] Compatibility options: %s\n' % ', '.join(self.params.get('compat_opts')))
dca08720
PH
3122 try:
3123 sp = subprocess.Popen(
3124 ['git', 'rev-parse', '--short', 'HEAD'],
3125 stdout=subprocess.PIPE, stderr=subprocess.PIPE,
3126 cwd=os.path.dirname(os.path.abspath(__file__)))
f5b1bca9 3127 out, err = process_communicate_or_kill(sp)
dca08720
PH
3128 out = out.decode().strip()
3129 if re.match('[0-9a-f]+', out):
f74980cb 3130 self._write_string('[debug] Git HEAD: %s\n' % out)
70a1165b 3131 except Exception:
dca08720
PH
3132 try:
3133 sys.exc_clear()
70a1165b 3134 except Exception:
dca08720 3135 pass
b300cda4
S
3136
3137 def python_implementation():
3138 impl_name = platform.python_implementation()
3139 if impl_name == 'PyPy' and hasattr(sys, 'pypy_version_info'):
3140 return impl_name + ' version %d.%d.%d' % sys.pypy_version_info[:3]
3141 return impl_name
3142
e5813e53 3143 self._write_string('[debug] Python version %s (%s %s) - %s\n' % (
3144 platform.python_version(),
3145 python_implementation(),
3146 platform.architecture()[0],
b300cda4 3147 platform_name()))
d28b5171 3148
73fac4e9 3149 exe_versions = FFmpegPostProcessor.get_versions(self)
4c83c967 3150 exe_versions['rtmpdump'] = rtmpdump_version()
feee8d32 3151 exe_versions['phantomjs'] = PhantomJSwrapper._version()
d28b5171
PH
3152 exe_str = ', '.join(
3153 '%s %s' % (exe, v)
3154 for exe, v in sorted(exe_versions.items())
3155 if v
3156 )
3157 if not exe_str:
3158 exe_str = 'none'
3159 self._write_string('[debug] exe versions: %s\n' % exe_str)
dca08720
PH
3160
3161 proxy_map = {}
3162 for handler in self._opener.handlers:
3163 if hasattr(handler, 'proxies'):
3164 proxy_map.update(handler.proxies)
734f90bb 3165 self._write_string('[debug] Proxy map: ' + compat_str(proxy_map) + '\n')
dca08720 3166
58b1f00d
PH
3167 if self.params.get('call_home', False):
3168 ipaddr = self.urlopen('https://yt-dl.org/ip').read().decode('utf-8')
3169 self._write_string('[debug] Public IP address: %s\n' % ipaddr)
f5546c0b 3170 return
58b1f00d
PH
3171 latest_version = self.urlopen(
3172 'https://yt-dl.org/latest/version').read().decode('utf-8')
3173 if version_tuple(latest_version) > version_tuple(__version__):
3174 self.report_warning(
3175 'You are using an outdated version (newest version: %s)! '
3176 'See https://yt-dl.org/update if you need help updating.' %
3177 latest_version)
3178
e344693b 3179 def _setup_opener(self):
6ad14cab 3180 timeout_val = self.params.get('socket_timeout')
19a41fc6 3181 self._socket_timeout = 600 if timeout_val is None else float(timeout_val)
6ad14cab 3182
dca08720
PH
3183 opts_cookiefile = self.params.get('cookiefile')
3184 opts_proxy = self.params.get('proxy')
3185
3186 if opts_cookiefile is None:
3187 self.cookiejar = compat_cookiejar.CookieJar()
3188 else:
590bc6f6 3189 opts_cookiefile = expand_path(opts_cookiefile)
1bab3437 3190 self.cookiejar = YoutubeDLCookieJar(opts_cookiefile)
dca08720 3191 if os.access(opts_cookiefile, os.R_OK):
1d88b3e6 3192 self.cookiejar.load(ignore_discard=True, ignore_expires=True)
dca08720 3193
6a3f4c3f 3194 cookie_processor = YoutubeDLCookieProcessor(self.cookiejar)
dca08720
PH
3195 if opts_proxy is not None:
3196 if opts_proxy == '':
3197 proxies = {}
3198 else:
3199 proxies = {'http': opts_proxy, 'https': opts_proxy}
3200 else:
3201 proxies = compat_urllib_request.getproxies()
067aa17e 3202 # Set HTTPS proxy to HTTP one if given (https://github.com/ytdl-org/youtube-dl/issues/805)
dca08720
PH
3203 if 'http' in proxies and 'https' not in proxies:
3204 proxies['https'] = proxies['http']
91410c9b 3205 proxy_handler = PerRequestProxyHandler(proxies)
a0ddb8a2
PH
3206
3207 debuglevel = 1 if self.params.get('debug_printtraffic') else 0
be4a824d
PH
3208 https_handler = make_HTTPS_handler(self.params, debuglevel=debuglevel)
3209 ydlh = YoutubeDLHandler(self.params, debuglevel=debuglevel)
fca6dba8 3210 redirect_handler = YoutubeDLRedirectHandler()
8b172c2e 3211 data_handler = compat_urllib_request_DataHandler()
6240b0a2
JMF
3212
3213 # When passing our own FileHandler instance, build_opener won't add the
3214 # default FileHandler and allows us to disable the file protocol, which
3215 # can be used for malicious purposes (see
067aa17e 3216 # https://github.com/ytdl-org/youtube-dl/issues/8227)
6240b0a2
JMF
3217 file_handler = compat_urllib_request.FileHandler()
3218
3219 def file_open(*args, **kwargs):
7a5c1cfe 3220 raise compat_urllib_error.URLError('file:// scheme is explicitly disabled in yt-dlp for security reasons')
6240b0a2
JMF
3221 file_handler.file_open = file_open
3222
3223 opener = compat_urllib_request.build_opener(
fca6dba8 3224 proxy_handler, https_handler, cookie_processor, ydlh, redirect_handler, data_handler, file_handler)
2461f79d 3225
dca08720
PH
3226 # Delete the default user-agent header, which would otherwise apply in
3227 # cases where our custom HTTP handler doesn't come into play
067aa17e 3228 # (See https://github.com/ytdl-org/youtube-dl/issues/1309 for details)
dca08720
PH
3229 opener.addheaders = []
3230 self._opener = opener
62fec3b2
PH
3231
3232 def encode(self, s):
3233 if isinstance(s, bytes):
3234 return s # Already encoded
3235
3236 try:
3237 return s.encode(self.get_encoding())
3238 except UnicodeEncodeError as err:
3239 err.reason = err.reason + '. Check your system encoding configuration or use the --encoding option.'
3240 raise
3241
3242 def get_encoding(self):
3243 encoding = self.params.get('encoding')
3244 if encoding is None:
3245 encoding = preferredencoding()
3246 return encoding
ec82d85a 3247
de6000d9 3248 def _write_thumbnails(self, info_dict, filename): # return the extensions
6c4fd172 3249 write_all = self.params.get('write_all_thumbnails', False)
3250 thumbnails = []
3251 if write_all or self.params.get('writethumbnail', False):
0202b52a 3252 thumbnails = info_dict.get('thumbnails') or []
6c4fd172 3253 multiple = write_all and len(thumbnails) > 1
ec82d85a 3254
0202b52a 3255 ret = []
981052c9 3256 for t in thumbnails[::-1]:
ec82d85a 3257 thumb_ext = determine_ext(t['url'], 'jpg')
6c4fd172 3258 suffix = '%s.' % t['id'] if multiple else ''
3259 thumb_display_id = '%s ' % t['id'] if multiple else ''
885cc0b7 3260 thumb_filename = replace_extension(filename, suffix + thumb_ext, info_dict.get('ext'))
ec82d85a 3261
0c3d0f51 3262 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(thumb_filename)):
de6000d9 3263 ret.append(suffix + thumb_ext)
8ba87148 3264 t['filepath'] = thumb_filename
ec82d85a
PH
3265 self.to_screen('[%s] %s: Thumbnail %sis already present' %
3266 (info_dict['extractor'], info_dict['id'], thumb_display_id))
3267 else:
5ef7d9bd 3268 self.to_screen('[%s] %s: Downloading thumbnail %s ...' %
ec82d85a
PH
3269 (info_dict['extractor'], info_dict['id'], thumb_display_id))
3270 try:
3271 uf = self.urlopen(t['url'])
d3d89c32 3272 with open(encodeFilename(thumb_filename), 'wb') as thumbf:
ec82d85a 3273 shutil.copyfileobj(uf, thumbf)
de6000d9 3274 ret.append(suffix + thumb_ext)
ec82d85a
PH
3275 self.to_screen('[%s] %s: Writing thumbnail %sto: %s' %
3276 (info_dict['extractor'], info_dict['id'], thumb_display_id, thumb_filename))
885cc0b7 3277 t['filepath'] = thumb_filename
3158150c 3278 except network_exceptions as err:
ec82d85a 3279 self.report_warning('Unable to download thumbnail "%s": %s' %
9b9c5355 3280 (t['url'], error_to_compat_str(err)))
6c4fd172 3281 if ret and not write_all:
3282 break
0202b52a 3283 return ret