]> jfr.im git - yt-dlp.git/blob - yt_dlp/YoutubeDL.py
Use certificates from `certifi` if installed (#3115)
[yt-dlp.git] / yt_dlp / YoutubeDL.py
1 #!/usr/bin/env python3
2 # coding: utf-8
3
4 from __future__ import absolute_import, unicode_literals
5
6 import collections
7 import contextlib
8 import datetime
9 import errno
10 import fileinput
11 import functools
12 import io
13 import itertools
14 import json
15 import locale
16 import operator
17 import os
18 import platform
19 import re
20 import shutil
21 import subprocess
22 import sys
23 import tempfile
24 import time
25 import tokenize
26 import traceback
27 import random
28 import unicodedata
29
30 from enum import Enum
31 from string import ascii_letters
32
33 from .compat import (
34 compat_basestring,
35 compat_brotli,
36 compat_get_terminal_size,
37 compat_kwargs,
38 compat_numeric_types,
39 compat_os_name,
40 compat_pycrypto_AES,
41 compat_shlex_quote,
42 compat_str,
43 compat_tokenize_tokenize,
44 compat_urllib_error,
45 compat_urllib_request,
46 compat_urllib_request_DataHandler,
47 windows_enable_vt_mode,
48 )
49 from .cookies import load_cookies
50 from .utils import (
51 age_restricted,
52 args_to_str,
53 ContentTooShortError,
54 date_from_str,
55 DateRange,
56 DEFAULT_OUTTMPL,
57 determine_ext,
58 determine_protocol,
59 DownloadCancelled,
60 DownloadError,
61 encode_compat_str,
62 encodeFilename,
63 EntryNotInPlaylist,
64 error_to_compat_str,
65 ExistingVideoReached,
66 expand_path,
67 ExtractorError,
68 float_or_none,
69 format_bytes,
70 format_field,
71 format_decimal_suffix,
72 formatSeconds,
73 GeoRestrictedError,
74 get_domain,
75 has_certifi,
76 HEADRequest,
77 InAdvancePagedList,
78 int_or_none,
79 iri_to_uri,
80 ISO3166Utils,
81 join_nonempty,
82 LazyList,
83 LINK_TEMPLATES,
84 locked_file,
85 make_dir,
86 make_HTTPS_handler,
87 MaxDownloadsReached,
88 merge_headers,
89 network_exceptions,
90 number_of_digits,
91 orderedSet,
92 OUTTMPL_TYPES,
93 PagedList,
94 parse_filesize,
95 PerRequestProxyHandler,
96 platform_name,
97 Popen,
98 POSTPROCESS_WHEN,
99 PostProcessingError,
100 preferredencoding,
101 prepend_extension,
102 ReExtractInfo,
103 register_socks_protocols,
104 RejectedVideoReached,
105 remove_terminal_sequences,
106 render_table,
107 replace_extension,
108 SameFileError,
109 sanitize_filename,
110 sanitize_path,
111 sanitize_url,
112 sanitized_Request,
113 std_headers,
114 STR_FORMAT_RE_TMPL,
115 STR_FORMAT_TYPES,
116 str_or_none,
117 strftime_or_none,
118 subtitles_filename,
119 supports_terminal_sequences,
120 timetuple_from_msec,
121 to_high_limit_path,
122 traverse_obj,
123 try_get,
124 UnavailableVideoError,
125 url_basename,
126 variadic,
127 version_tuple,
128 write_json_file,
129 write_string,
130 YoutubeDLCookieProcessor,
131 YoutubeDLHandler,
132 YoutubeDLRedirectHandler,
133 )
134 from .cache import Cache
135 from .minicurses import format_text
136 from .extractor import (
137 gen_extractor_classes,
138 get_info_extractor,
139 _LAZY_LOADER,
140 _PLUGIN_CLASSES as plugin_extractors
141 )
142 from .extractor.openload import PhantomJSwrapper
143 from .downloader import (
144 FFmpegFD,
145 get_suitable_downloader,
146 shorten_protocol_name
147 )
148 from .downloader.rtmp import rtmpdump_version
149 from .postprocessor import (
150 get_postprocessor,
151 EmbedThumbnailPP,
152 FFmpegFixupDuplicateMoovPP,
153 FFmpegFixupDurationPP,
154 FFmpegFixupM3u8PP,
155 FFmpegFixupM4aPP,
156 FFmpegFixupStretchedPP,
157 FFmpegFixupTimestampPP,
158 FFmpegMergerPP,
159 FFmpegPostProcessor,
160 MoveFilesAfterDownloadPP,
161 _PLUGIN_CLASSES as plugin_postprocessors
162 )
163 from .update import detect_variant
164 from .version import __version__, RELEASE_GIT_HEAD
165
166 if compat_os_name == 'nt':
167 import ctypes
168
169
170 class YoutubeDL(object):
171 """YoutubeDL class.
172
173 YoutubeDL objects are the ones responsible of downloading the
174 actual video file and writing it to disk if the user has requested
175 it, among some other tasks. In most cases there should be one per
176 program. As, given a video URL, the downloader doesn't know how to
177 extract all the needed information, task that InfoExtractors do, it
178 has to pass the URL to one of them.
179
180 For this, YoutubeDL objects have a method that allows
181 InfoExtractors to be registered in a given order. When it is passed
182 a URL, the YoutubeDL object handles it to the first InfoExtractor it
183 finds that reports being able to handle it. The InfoExtractor extracts
184 all the information about the video or videos the URL refers to, and
185 YoutubeDL process the extracted information, possibly using a File
186 Downloader to download the video.
187
188 YoutubeDL objects accept a lot of parameters. In order not to saturate
189 the object constructor with arguments, it receives a dictionary of
190 options instead. These options are available through the params
191 attribute for the InfoExtractors to use. The YoutubeDL also
192 registers itself as the downloader in charge for the InfoExtractors
193 that are added to it, so this is a "mutual registration".
194
195 Available options:
196
197 username: Username for authentication purposes.
198 password: Password for authentication purposes.
199 videopassword: Password for accessing a video.
200 ap_mso: Adobe Pass multiple-system operator identifier.
201 ap_username: Multiple-system operator account username.
202 ap_password: Multiple-system operator account password.
203 usenetrc: Use netrc for authentication instead.
204 verbose: Print additional info to stdout.
205 quiet: Do not print messages to stdout.
206 no_warnings: Do not print out anything for warnings.
207 forceprint: A dict with keys WHEN mapped to a list of templates to
208 print to stdout. The allowed keys are video or any of the
209 items in utils.POSTPROCESS_WHEN.
210 For compatibility, a single list is also accepted
211 print_to_file: A dict with keys WHEN (same as forceprint) mapped to
212 a list of tuples with (template, filename)
213 forceurl: Force printing final URL. (Deprecated)
214 forcetitle: Force printing title. (Deprecated)
215 forceid: Force printing ID. (Deprecated)
216 forcethumbnail: Force printing thumbnail URL. (Deprecated)
217 forcedescription: Force printing description. (Deprecated)
218 forcefilename: Force printing final filename. (Deprecated)
219 forceduration: Force printing duration. (Deprecated)
220 forcejson: Force printing info_dict as JSON.
221 dump_single_json: Force printing the info_dict of the whole playlist
222 (or video) as a single JSON line.
223 force_write_download_archive: Force writing download archive regardless
224 of 'skip_download' or 'simulate'.
225 simulate: Do not download the video files. If unset (or None),
226 simulate only if listsubtitles, listformats or list_thumbnails is used
227 format: Video format code. see "FORMAT SELECTION" for more details.
228 You can also pass a function. The function takes 'ctx' as
229 argument and returns the formats to download.
230 See "build_format_selector" for an implementation
231 allow_unplayable_formats: Allow unplayable formats to be extracted and downloaded.
232 ignore_no_formats_error: Ignore "No video formats" error. Usefull for
233 extracting metadata even if the video is not actually
234 available for download (experimental)
235 format_sort: A list of fields by which to sort the video formats.
236 See "Sorting Formats" for more details.
237 format_sort_force: Force the given format_sort. see "Sorting Formats"
238 for more details.
239 prefer_free_formats: Whether to prefer video formats with free containers
240 over non-free ones of same quality.
241 allow_multiple_video_streams: Allow multiple video streams to be merged
242 into a single file
243 allow_multiple_audio_streams: Allow multiple audio streams to be merged
244 into a single file
245 check_formats Whether to test if the formats are downloadable.
246 Can be True (check all), False (check none),
247 'selected' (check selected formats),
248 or None (check only if requested by extractor)
249 paths: Dictionary of output paths. The allowed keys are 'home'
250 'temp' and the keys of OUTTMPL_TYPES (in utils.py)
251 outtmpl: Dictionary of templates for output names. Allowed keys
252 are 'default' and the keys of OUTTMPL_TYPES (in utils.py).
253 For compatibility with youtube-dl, a single string can also be used
254 outtmpl_na_placeholder: Placeholder for unavailable meta fields.
255 restrictfilenames: Do not allow "&" and spaces in file names
256 trim_file_name: Limit length of filename (extension excluded)
257 windowsfilenames: Force the filenames to be windows compatible
258 ignoreerrors: Do not stop on download/postprocessing errors.
259 Can be 'only_download' to ignore only download errors.
260 Default is 'only_download' for CLI, but False for API
261 skip_playlist_after_errors: Number of allowed failures until the rest of
262 the playlist is skipped
263 force_generic_extractor: Force downloader to use the generic extractor
264 overwrites: Overwrite all video and metadata files if True,
265 overwrite only non-video files if None
266 and don't overwrite any file if False
267 For compatibility with youtube-dl,
268 "nooverwrites" may also be used instead
269 playliststart: Playlist item to start at.
270 playlistend: Playlist item to end at.
271 playlist_items: Specific indices of playlist to download.
272 playlistreverse: Download playlist items in reverse order.
273 playlistrandom: Download playlist items in random order.
274 matchtitle: Download only matching titles.
275 rejecttitle: Reject downloads for matching titles.
276 logger: Log messages to a logging.Logger instance.
277 logtostderr: Log messages to stderr instead of stdout.
278 consoletitle: Display progress in console window's titlebar.
279 writedescription: Write the video description to a .description file
280 writeinfojson: Write the video description to a .info.json file
281 clean_infojson: Remove private fields from the infojson
282 getcomments: Extract video comments. This will not be written to disk
283 unless writeinfojson is also given
284 writeannotations: Write the video annotations to a .annotations.xml file
285 writethumbnail: Write the thumbnail image to a file
286 allow_playlist_files: Whether to write playlists' description, infojson etc
287 also to disk when using the 'write*' options
288 write_all_thumbnails: Write all thumbnail formats to files
289 writelink: Write an internet shortcut file, depending on the
290 current platform (.url/.webloc/.desktop)
291 writeurllink: Write a Windows internet shortcut file (.url)
292 writewebloclink: Write a macOS internet shortcut file (.webloc)
293 writedesktoplink: Write a Linux internet shortcut file (.desktop)
294 writesubtitles: Write the video subtitles to a file
295 writeautomaticsub: Write the automatically generated subtitles to a file
296 allsubtitles: Deprecated - Use subtitleslangs = ['all']
297 Downloads all the subtitles of the video
298 (requires writesubtitles or writeautomaticsub)
299 listsubtitles: Lists all available subtitles for the video
300 subtitlesformat: The format code for subtitles
301 subtitleslangs: List of languages of the subtitles to download (can be regex).
302 The list may contain "all" to refer to all the available
303 subtitles. The language can be prefixed with a "-" to
304 exclude it from the requested languages. Eg: ['all', '-live_chat']
305 keepvideo: Keep the video file after post-processing
306 daterange: A DateRange object, download only if the upload_date is in the range.
307 skip_download: Skip the actual download of the video file
308 cachedir: Location of the cache files in the filesystem.
309 False to disable filesystem cache.
310 noplaylist: Download single video instead of a playlist if in doubt.
311 age_limit: An integer representing the user's age in years.
312 Unsuitable videos for the given age are skipped.
313 min_views: An integer representing the minimum view count the video
314 must have in order to not be skipped.
315 Videos without view count information are always
316 downloaded. None for no limit.
317 max_views: An integer representing the maximum view count.
318 Videos that are more popular than that are not
319 downloaded.
320 Videos without view count information are always
321 downloaded. None for no limit.
322 download_archive: File name of a file where all downloads are recorded.
323 Videos already present in the file are not downloaded
324 again.
325 break_on_existing: Stop the download process after attempting to download a
326 file that is in the archive.
327 break_on_reject: Stop the download process when encountering a video that
328 has been filtered out.
329 break_per_url: Whether break_on_reject and break_on_existing
330 should act on each input URL as opposed to for the entire queue
331 cookiefile: File name where cookies should be read from and dumped to
332 cookiesfrombrowser: A tuple containing the name of the browser, the profile
333 name/pathfrom where cookies are loaded, and the name of the
334 keyring. Eg: ('chrome', ) or ('vivaldi', 'default', 'BASICTEXT')
335 legacyserverconnect: Explicitly allow HTTPS connection to servers that do not
336 support RFC 5746 secure renegotiation
337 nocheckcertificate: Do not verify SSL certificates
338 prefer_insecure: Use HTTP instead of HTTPS to retrieve information.
339 At the moment, this is only supported by YouTube.
340 http_headers: A dictionary of custom headers to be used for all requests
341 proxy: URL of the proxy server to use
342 geo_verification_proxy: URL of the proxy to use for IP address verification
343 on geo-restricted sites.
344 socket_timeout: Time to wait for unresponsive hosts, in seconds
345 bidi_workaround: Work around buggy terminals without bidirectional text
346 support, using fridibi
347 debug_printtraffic:Print out sent and received HTTP traffic
348 include_ads: Download ads as well (deprecated)
349 default_search: Prepend this string if an input url is not valid.
350 'auto' for elaborate guessing
351 encoding: Use this encoding instead of the system-specified.
352 extract_flat: Do not resolve URLs, return the immediate result.
353 Pass in 'in_playlist' to only show this behavior for
354 playlist items.
355 wait_for_video: If given, wait for scheduled streams to become available.
356 The value should be a tuple containing the range
357 (min_secs, max_secs) to wait between retries
358 postprocessors: A list of dictionaries, each with an entry
359 * key: The name of the postprocessor. See
360 yt_dlp/postprocessor/__init__.py for a list.
361 * when: When to run the postprocessor. Allowed values are
362 the entries of utils.POSTPROCESS_WHEN
363 Assumed to be 'post_process' if not given
364 post_hooks: Deprecated - Register a custom postprocessor instead
365 A list of functions that get called as the final step
366 for each video file, after all postprocessors have been
367 called. The filename will be passed as the only argument.
368 progress_hooks: A list of functions that get called on download
369 progress, with a dictionary with the entries
370 * status: One of "downloading", "error", or "finished".
371 Check this first and ignore unknown values.
372 * info_dict: The extracted info_dict
373
374 If status is one of "downloading", or "finished", the
375 following properties may also be present:
376 * filename: The final filename (always present)
377 * tmpfilename: The filename we're currently writing to
378 * downloaded_bytes: Bytes on disk
379 * total_bytes: Size of the whole file, None if unknown
380 * total_bytes_estimate: Guess of the eventual file size,
381 None if unavailable.
382 * elapsed: The number of seconds since download started.
383 * eta: The estimated time in seconds, None if unknown
384 * speed: The download speed in bytes/second, None if
385 unknown
386 * fragment_index: The counter of the currently
387 downloaded video fragment.
388 * fragment_count: The number of fragments (= individual
389 files that will be merged)
390
391 Progress hooks are guaranteed to be called at least once
392 (with status "finished") if the download is successful.
393 postprocessor_hooks: A list of functions that get called on postprocessing
394 progress, with a dictionary with the entries
395 * status: One of "started", "processing", or "finished".
396 Check this first and ignore unknown values.
397 * postprocessor: Name of the postprocessor
398 * info_dict: The extracted info_dict
399
400 Progress hooks are guaranteed to be called at least twice
401 (with status "started" and "finished") if the processing is successful.
402 merge_output_format: Extension to use when merging formats.
403 final_ext: Expected final extension; used to detect when the file was
404 already downloaded and converted
405 fixup: Automatically correct known faults of the file.
406 One of:
407 - "never": do nothing
408 - "warn": only emit a warning
409 - "detect_or_warn": check whether we can do anything
410 about it, warn otherwise (default)
411 source_address: Client-side IP address to bind to.
412 call_home: Boolean, true iff we are allowed to contact the
413 yt-dlp servers for debugging. (BROKEN)
414 sleep_interval_requests: Number of seconds to sleep between requests
415 during extraction
416 sleep_interval: Number of seconds to sleep before each download when
417 used alone or a lower bound of a range for randomized
418 sleep before each download (minimum possible number
419 of seconds to sleep) when used along with
420 max_sleep_interval.
421 max_sleep_interval:Upper bound of a range for randomized sleep before each
422 download (maximum possible number of seconds to sleep).
423 Must only be used along with sleep_interval.
424 Actual sleep time will be a random float from range
425 [sleep_interval; max_sleep_interval].
426 sleep_interval_subtitles: Number of seconds to sleep before each subtitle download
427 listformats: Print an overview of available video formats and exit.
428 list_thumbnails: Print a table of all thumbnails and exit.
429 match_filter: A function that gets called with the info_dict of
430 every video.
431 If it returns a message, the video is ignored.
432 If it returns None, the video is downloaded.
433 match_filter_func in utils.py is one example for this.
434 no_color: Do not emit color codes in output.
435 geo_bypass: Bypass geographic restriction via faking X-Forwarded-For
436 HTTP header
437 geo_bypass_country:
438 Two-letter ISO 3166-2 country code that will be used for
439 explicit geographic restriction bypassing via faking
440 X-Forwarded-For HTTP header
441 geo_bypass_ip_block:
442 IP range in CIDR notation that will be used similarly to
443 geo_bypass_country
444
445 The following options determine which downloader is picked:
446 external_downloader: A dictionary of protocol keys and the executable of the
447 external downloader to use for it. The allowed protocols
448 are default|http|ftp|m3u8|dash|rtsp|rtmp|mms.
449 Set the value to 'native' to use the native downloader
450 hls_prefer_native: Deprecated - Use external_downloader = {'m3u8': 'native'}
451 or {'m3u8': 'ffmpeg'} instead.
452 Use the native HLS downloader instead of ffmpeg/avconv
453 if True, otherwise use ffmpeg/avconv if False, otherwise
454 use downloader suggested by extractor if None.
455 compat_opts: Compatibility options. See "Differences in default behavior".
456 The following options do not work when used through the API:
457 filename, abort-on-error, multistreams, no-live-chat, format-sort
458 no-clean-infojson, no-playlist-metafiles, no-keep-subs, no-attach-info-json.
459 Refer __init__.py for their implementation
460 progress_template: Dictionary of templates for progress outputs.
461 Allowed keys are 'download', 'postprocess',
462 'download-title' (console title) and 'postprocess-title'.
463 The template is mapped on a dictionary with keys 'progress' and 'info'
464
465 The following parameters are not used by YoutubeDL itself, they are used by
466 the downloader (see yt_dlp/downloader/common.py):
467 nopart, updatetime, buffersize, ratelimit, throttledratelimit, min_filesize,
468 max_filesize, test, noresizebuffer, retries, file_access_retries, fragment_retries,
469 continuedl, noprogress, xattr_set_filesize, hls_use_mpegts, http_chunk_size,
470 external_downloader_args, concurrent_fragment_downloads.
471
472 The following options are used by the post processors:
473 prefer_ffmpeg: If False, use avconv instead of ffmpeg if both are available,
474 otherwise prefer ffmpeg. (avconv support is deprecated)
475 ffmpeg_location: Location of the ffmpeg/avconv binary; either the path
476 to the binary or its containing directory.
477 postprocessor_args: A dictionary of postprocessor/executable keys (in lower case)
478 and a list of additional command-line arguments for the
479 postprocessor/executable. The dict can also have "PP+EXE" keys
480 which are used when the given exe is used by the given PP.
481 Use 'default' as the name for arguments to passed to all PP
482 For compatibility with youtube-dl, a single list of args
483 can also be used
484
485 The following options are used by the extractors:
486 extractor_retries: Number of times to retry for known errors
487 dynamic_mpd: Whether to process dynamic DASH manifests (default: True)
488 hls_split_discontinuity: Split HLS playlists to different formats at
489 discontinuities such as ad breaks (default: False)
490 extractor_args: A dictionary of arguments to be passed to the extractors.
491 See "EXTRACTOR ARGUMENTS" for details.
492 Eg: {'youtube': {'skip': ['dash', 'hls']}}
493 mark_watched: Mark videos watched (even with --simulate). Only for YouTube
494 youtube_include_dash_manifest: Deprecated - Use extractor_args instead.
495 If True (default), DASH manifests and related
496 data will be downloaded and processed by extractor.
497 You can reduce network I/O by disabling it if you don't
498 care about DASH. (only for youtube)
499 youtube_include_hls_manifest: Deprecated - Use extractor_args instead.
500 If True (default), HLS manifests and related
501 data will be downloaded and processed by extractor.
502 You can reduce network I/O by disabling it if you don't
503 care about HLS. (only for youtube)
504 """
505
506 _NUMERIC_FIELDS = set((
507 'width', 'height', 'tbr', 'abr', 'asr', 'vbr', 'fps', 'filesize', 'filesize_approx',
508 'timestamp', 'release_timestamp',
509 'duration', 'view_count', 'like_count', 'dislike_count', 'repost_count',
510 'average_rating', 'comment_count', 'age_limit',
511 'start_time', 'end_time',
512 'chapter_number', 'season_number', 'episode_number',
513 'track_number', 'disc_number', 'release_year',
514 ))
515
516 _format_selection_exts = {
517 'audio': {'m4a', 'mp3', 'ogg', 'aac'},
518 'video': {'mp4', 'flv', 'webm', '3gp'},
519 'storyboards': {'mhtml'},
520 }
521
522 def __init__(self, params=None, auto_init=True):
523 """Create a FileDownloader object with the given options.
524 @param auto_init Whether to load the default extractors and print header (if verbose).
525 Set to 'no_verbose_header' to not print the header
526 """
527 if params is None:
528 params = {}
529 self.params = params
530 self._ies = {}
531 self._ies_instances = {}
532 self._pps = {k: [] for k in POSTPROCESS_WHEN}
533 self._printed_messages = set()
534 self._first_webpage_request = True
535 self._post_hooks = []
536 self._progress_hooks = []
537 self._postprocessor_hooks = []
538 self._download_retcode = 0
539 self._num_downloads = 0
540 self._num_videos = 0
541 self._playlist_level = 0
542 self._playlist_urls = set()
543 self.cache = Cache(self)
544
545 windows_enable_vt_mode()
546 self._out_files = {
547 'error': sys.stderr,
548 'print': sys.stderr if self.params.get('logtostderr') else sys.stdout,
549 'console': None if compat_os_name == 'nt' else next(
550 filter(supports_terminal_sequences, (sys.stderr, sys.stdout)), None)
551 }
552 self._out_files['screen'] = sys.stderr if self.params.get('quiet') else self._out_files['print']
553 self._allow_colors = {
554 type_: not self.params.get('no_color') and supports_terminal_sequences(self._out_files[type_])
555 for type_ in ('screen', 'error')
556 }
557
558 if sys.version_info < (3, 6):
559 self.report_warning(
560 'Python version %d.%d is not supported! Please update to Python 3.6 or above' % sys.version_info[:2])
561
562 if self.params.get('allow_unplayable_formats'):
563 self.report_warning(
564 f'You have asked for {self._format_err("UNPLAYABLE", self.Styles.EMPHASIS)} formats to be listed/downloaded. '
565 'This is a developer option intended for debugging. \n'
566 ' If you experience any issues while using this option, '
567 f'{self._format_err("DO NOT", self.Styles.ERROR)} open a bug report')
568
569 def check_deprecated(param, option, suggestion):
570 if self.params.get(param) is not None:
571 self.report_warning('%s is deprecated. Use %s instead' % (option, suggestion))
572 return True
573 return False
574
575 if check_deprecated('cn_verification_proxy', '--cn-verification-proxy', '--geo-verification-proxy'):
576 if self.params.get('geo_verification_proxy') is None:
577 self.params['geo_verification_proxy'] = self.params['cn_verification_proxy']
578
579 check_deprecated('autonumber', '--auto-number', '-o "%(autonumber)s-%(title)s.%(ext)s"')
580 check_deprecated('usetitle', '--title', '-o "%(title)s-%(id)s.%(ext)s"')
581 check_deprecated('useid', '--id', '-o "%(id)s.%(ext)s"')
582
583 for msg in self.params.get('_warnings', []):
584 self.report_warning(msg)
585 for msg in self.params.get('_deprecation_warnings', []):
586 self.deprecation_warning(msg)
587
588 if 'list-formats' in self.params.get('compat_opts', []):
589 self.params['listformats_table'] = False
590
591 if 'overwrites' not in self.params and self.params.get('nooverwrites') is not None:
592 # nooverwrites was unnecessarily changed to overwrites
593 # in 0c3d0f51778b153f65c21906031c2e091fcfb641
594 # This ensures compatibility with both keys
595 self.params['overwrites'] = not self.params['nooverwrites']
596 elif self.params.get('overwrites') is None:
597 self.params.pop('overwrites', None)
598 else:
599 self.params['nooverwrites'] = not self.params['overwrites']
600
601 self.params.setdefault('forceprint', {})
602 self.params.setdefault('print_to_file', {})
603
604 # Compatibility with older syntax
605 if not isinstance(params['forceprint'], dict):
606 self.params['forceprint'] = {'video': params['forceprint']}
607
608 if self.params.get('bidi_workaround', False):
609 try:
610 import pty
611 master, slave = pty.openpty()
612 width = compat_get_terminal_size().columns
613 if width is None:
614 width_args = []
615 else:
616 width_args = ['-w', str(width)]
617 sp_kwargs = dict(
618 stdin=subprocess.PIPE,
619 stdout=slave,
620 stderr=self._out_files['error'])
621 try:
622 self._output_process = Popen(['bidiv'] + width_args, **sp_kwargs)
623 except OSError:
624 self._output_process = Popen(['fribidi', '-c', 'UTF-8'] + width_args, **sp_kwargs)
625 self._output_channel = os.fdopen(master, 'rb')
626 except OSError as ose:
627 if ose.errno == errno.ENOENT:
628 self.report_warning(
629 'Could not find fribidi executable, ignoring --bidi-workaround. '
630 'Make sure that fribidi is an executable file in one of the directories in your $PATH.')
631 else:
632 raise
633
634 if (sys.platform != 'win32'
635 and sys.getfilesystemencoding() in ['ascii', 'ANSI_X3.4-1968']
636 and not self.params.get('restrictfilenames', False)):
637 # Unicode filesystem API will throw errors (#1474, #13027)
638 self.report_warning(
639 'Assuming --restrict-filenames since file system encoding '
640 'cannot encode all characters. '
641 'Set the LC_ALL environment variable to fix this.')
642 self.params['restrictfilenames'] = True
643
644 self.outtmpl_dict = self.parse_outtmpl()
645
646 # Creating format selector here allows us to catch syntax errors before the extraction
647 self.format_selector = (
648 self.params.get('format') if self.params.get('format') in (None, '-')
649 else self.params['format'] if callable(self.params['format'])
650 else self.build_format_selector(self.params['format']))
651
652 # Set http_headers defaults according to std_headers
653 self.params['http_headers'] = merge_headers(std_headers, self.params.get('http_headers', {}))
654
655 self._setup_opener()
656
657 if auto_init:
658 if auto_init != 'no_verbose_header':
659 self.print_debug_header()
660 self.add_default_info_extractors()
661
662 hooks = {
663 'post_hooks': self.add_post_hook,
664 'progress_hooks': self.add_progress_hook,
665 'postprocessor_hooks': self.add_postprocessor_hook,
666 }
667 for opt, fn in hooks.items():
668 for ph in self.params.get(opt, []):
669 fn(ph)
670
671 for pp_def_raw in self.params.get('postprocessors', []):
672 pp_def = dict(pp_def_raw)
673 when = pp_def.pop('when', 'post_process')
674 self.add_post_processor(
675 get_postprocessor(pp_def.pop('key'))(self, **compat_kwargs(pp_def)),
676 when=when)
677
678 register_socks_protocols()
679
680 def preload_download_archive(fn):
681 """Preload the archive, if any is specified"""
682 if fn is None:
683 return False
684 self.write_debug(f'Loading archive file {fn!r}')
685 try:
686 with locked_file(fn, 'r', encoding='utf-8') as archive_file:
687 for line in archive_file:
688 self.archive.add(line.strip())
689 except IOError as ioe:
690 if ioe.errno != errno.ENOENT:
691 raise
692 return False
693 return True
694
695 self.archive = set()
696 preload_download_archive(self.params.get('download_archive'))
697
698 def warn_if_short_id(self, argv):
699 # short YouTube ID starting with dash?
700 idxs = [
701 i for i, a in enumerate(argv)
702 if re.match(r'^-[0-9A-Za-z_-]{10}$', a)]
703 if idxs:
704 correct_argv = (
705 ['yt-dlp']
706 + [a for i, a in enumerate(argv) if i not in idxs]
707 + ['--'] + [argv[i] for i in idxs]
708 )
709 self.report_warning(
710 'Long argument string detected. '
711 'Use -- to separate parameters and URLs, like this:\n%s' %
712 args_to_str(correct_argv))
713
714 def add_info_extractor(self, ie):
715 """Add an InfoExtractor object to the end of the list."""
716 ie_key = ie.ie_key()
717 self._ies[ie_key] = ie
718 if not isinstance(ie, type):
719 self._ies_instances[ie_key] = ie
720 ie.set_downloader(self)
721
722 def _get_info_extractor_class(self, ie_key):
723 ie = self._ies.get(ie_key)
724 if ie is None:
725 ie = get_info_extractor(ie_key)
726 self.add_info_extractor(ie)
727 return ie
728
729 def get_info_extractor(self, ie_key):
730 """
731 Get an instance of an IE with name ie_key, it will try to get one from
732 the _ies list, if there's no instance it will create a new one and add
733 it to the extractor list.
734 """
735 ie = self._ies_instances.get(ie_key)
736 if ie is None:
737 ie = get_info_extractor(ie_key)()
738 self.add_info_extractor(ie)
739 return ie
740
741 def add_default_info_extractors(self):
742 """
743 Add the InfoExtractors returned by gen_extractors to the end of the list
744 """
745 for ie in gen_extractor_classes():
746 self.add_info_extractor(ie)
747
748 def add_post_processor(self, pp, when='post_process'):
749 """Add a PostProcessor object to the end of the chain."""
750 self._pps[when].append(pp)
751 pp.set_downloader(self)
752
753 def add_post_hook(self, ph):
754 """Add the post hook"""
755 self._post_hooks.append(ph)
756
757 def add_progress_hook(self, ph):
758 """Add the download progress hook"""
759 self._progress_hooks.append(ph)
760
761 def add_postprocessor_hook(self, ph):
762 """Add the postprocessing progress hook"""
763 self._postprocessor_hooks.append(ph)
764 for pps in self._pps.values():
765 for pp in pps:
766 pp.add_progress_hook(ph)
767
768 def _bidi_workaround(self, message):
769 if not hasattr(self, '_output_channel'):
770 return message
771
772 assert hasattr(self, '_output_process')
773 assert isinstance(message, compat_str)
774 line_count = message.count('\n') + 1
775 self._output_process.stdin.write((message + '\n').encode('utf-8'))
776 self._output_process.stdin.flush()
777 res = ''.join(self._output_channel.readline().decode('utf-8')
778 for _ in range(line_count))
779 return res[:-len('\n')]
780
781 def _write_string(self, message, out=None, only_once=False):
782 if only_once:
783 if message in self._printed_messages:
784 return
785 self._printed_messages.add(message)
786 write_string(message, out=out, encoding=self.params.get('encoding'))
787
788 def to_stdout(self, message, skip_eol=False, quiet=None):
789 """Print message to stdout"""
790 if quiet is not None:
791 self.deprecation_warning('"ydl.to_stdout" no longer accepts the argument quiet. Use "ydl.to_screen" instead')
792 self._write_string(
793 '%s%s' % (self._bidi_workaround(message), ('' if skip_eol else '\n')),
794 self._out_files['print'])
795
796 def to_screen(self, message, skip_eol=False, quiet=None):
797 """Print message to screen if not in quiet mode"""
798 if self.params.get('logger'):
799 self.params['logger'].debug(message)
800 return
801 if (self.params.get('quiet') if quiet is None else quiet) and not self.params.get('verbose'):
802 return
803 self._write_string(
804 '%s%s' % (self._bidi_workaround(message), ('' if skip_eol else '\n')),
805 self._out_files['screen'])
806
807 def to_stderr(self, message, only_once=False):
808 """Print message to stderr"""
809 assert isinstance(message, compat_str)
810 if self.params.get('logger'):
811 self.params['logger'].error(message)
812 else:
813 self._write_string('%s\n' % self._bidi_workaround(message), self._out_files['error'], only_once=only_once)
814
815 def _send_console_code(self, code):
816 if compat_os_name == 'nt' or not self._out_files['console']:
817 return
818 self._write_string(code, self._out_files['console'])
819
820 def to_console_title(self, message):
821 if not self.params.get('consoletitle', False):
822 return
823 message = remove_terminal_sequences(message)
824 if compat_os_name == 'nt':
825 if ctypes.windll.kernel32.GetConsoleWindow():
826 # c_wchar_p() might not be necessary if `message` is
827 # already of type unicode()
828 ctypes.windll.kernel32.SetConsoleTitleW(ctypes.c_wchar_p(message))
829 else:
830 self._send_console_code(f'\033]0;{message}\007')
831
832 def save_console_title(self):
833 if not self.params.get('consoletitle') or self.params.get('simulate'):
834 return
835 self._send_console_code('\033[22;0t') # Save the title on stack
836
837 def restore_console_title(self):
838 if not self.params.get('consoletitle') or self.params.get('simulate'):
839 return
840 self._send_console_code('\033[23;0t') # Restore the title from stack
841
842 def __enter__(self):
843 self.save_console_title()
844 return self
845
846 def __exit__(self, *args):
847 self.restore_console_title()
848
849 if self.params.get('cookiefile') is not None:
850 self.cookiejar.save(ignore_discard=True, ignore_expires=True)
851
852 def trouble(self, message=None, tb=None, is_error=True):
853 """Determine action to take when a download problem appears.
854
855 Depending on if the downloader has been configured to ignore
856 download errors or not, this method may throw an exception or
857 not when errors are found, after printing the message.
858
859 @param tb If given, is additional traceback information
860 @param is_error Whether to raise error according to ignorerrors
861 """
862 if message is not None:
863 self.to_stderr(message)
864 if self.params.get('verbose'):
865 if tb is None:
866 if sys.exc_info()[0]: # if .trouble has been called from an except block
867 tb = ''
868 if hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
869 tb += ''.join(traceback.format_exception(*sys.exc_info()[1].exc_info))
870 tb += encode_compat_str(traceback.format_exc())
871 else:
872 tb_data = traceback.format_list(traceback.extract_stack())
873 tb = ''.join(tb_data)
874 if tb:
875 self.to_stderr(tb)
876 if not is_error:
877 return
878 if not self.params.get('ignoreerrors'):
879 if sys.exc_info()[0] and hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
880 exc_info = sys.exc_info()[1].exc_info
881 else:
882 exc_info = sys.exc_info()
883 raise DownloadError(message, exc_info)
884 self._download_retcode = 1
885
886 class Styles(Enum):
887 HEADERS = 'yellow'
888 EMPHASIS = 'light blue'
889 ID = 'green'
890 DELIM = 'blue'
891 ERROR = 'red'
892 WARNING = 'yellow'
893 SUPPRESS = 'light black'
894
895 def _format_text(self, handle, allow_colors, text, f, fallback=None, *, test_encoding=False):
896 if test_encoding:
897 original_text = text
898 # handle.encoding can be None. See https://github.com/yt-dlp/yt-dlp/issues/2711
899 encoding = self.params.get('encoding') or getattr(handle, 'encoding', None) or 'ascii'
900 text = text.encode(encoding, 'ignore').decode(encoding)
901 if fallback is not None and text != original_text:
902 text = fallback
903 if isinstance(f, self.Styles):
904 f = f.value
905 return format_text(text, f) if allow_colors else text if fallback is None else fallback
906
907 def _format_screen(self, *args, **kwargs):
908 return self._format_text(
909 self._out_files['screen'], self._allow_colors['screen'], *args, **kwargs)
910
911 def _format_err(self, *args, **kwargs):
912 return self._format_text(
913 self._out_files['error'], self._allow_colors['error'], *args, **kwargs)
914
915 def report_warning(self, message, only_once=False):
916 '''
917 Print the message to stderr, it will be prefixed with 'WARNING:'
918 If stderr is a tty file the 'WARNING:' will be colored
919 '''
920 if self.params.get('logger') is not None:
921 self.params['logger'].warning(message)
922 else:
923 if self.params.get('no_warnings'):
924 return
925 self.to_stderr(f'{self._format_err("WARNING:", self.Styles.WARNING)} {message}', only_once)
926
927 def deprecation_warning(self, message):
928 if self.params.get('logger') is not None:
929 self.params['logger'].warning('DeprecationWarning: {message}')
930 else:
931 self.to_stderr(f'{self._format_err("DeprecationWarning:", self.Styles.ERROR)} {message}', True)
932
933 def report_error(self, message, *args, **kwargs):
934 '''
935 Do the same as trouble, but prefixes the message with 'ERROR:', colored
936 in red if stderr is a tty file.
937 '''
938 self.trouble(f'{self._format_err("ERROR:", self.Styles.ERROR)} {message}', *args, **kwargs)
939
940 def write_debug(self, message, only_once=False):
941 '''Log debug message or Print message to stderr'''
942 if not self.params.get('verbose', False):
943 return
944 message = '[debug] %s' % message
945 if self.params.get('logger'):
946 self.params['logger'].debug(message)
947 else:
948 self.to_stderr(message, only_once)
949
950 def report_file_already_downloaded(self, file_name):
951 """Report file has already been fully downloaded."""
952 try:
953 self.to_screen('[download] %s has already been downloaded' % file_name)
954 except UnicodeEncodeError:
955 self.to_screen('[download] The file has already been downloaded')
956
957 def report_file_delete(self, file_name):
958 """Report that existing file will be deleted."""
959 try:
960 self.to_screen('Deleting existing file %s' % file_name)
961 except UnicodeEncodeError:
962 self.to_screen('Deleting existing file')
963
964 def raise_no_formats(self, info, forced=False, *, msg=None):
965 has_drm = info.get('__has_drm')
966 ignored, expected = self.params.get('ignore_no_formats_error'), bool(msg)
967 msg = msg or has_drm and 'This video is DRM protected' or 'No video formats found!'
968 if forced or not ignored:
969 raise ExtractorError(msg, video_id=info['id'], ie=info['extractor'],
970 expected=has_drm or ignored or expected)
971 else:
972 self.report_warning(msg)
973
974 def parse_outtmpl(self):
975 outtmpl_dict = self.params.get('outtmpl', {})
976 if not isinstance(outtmpl_dict, dict):
977 outtmpl_dict = {'default': outtmpl_dict}
978 # Remove spaces in the default template
979 if self.params.get('restrictfilenames'):
980 sanitize = lambda x: x.replace(' - ', ' ').replace(' ', '-')
981 else:
982 sanitize = lambda x: x
983 outtmpl_dict.update({
984 k: sanitize(v) for k, v in DEFAULT_OUTTMPL.items()
985 if outtmpl_dict.get(k) is None})
986 for key, val in outtmpl_dict.items():
987 if isinstance(val, bytes):
988 self.report_warning(
989 'Parameter outtmpl is bytes, but should be a unicode string. '
990 'Put from __future__ import unicode_literals at the top of your code file or consider switching to Python 3.x.')
991 return outtmpl_dict
992
993 def get_output_path(self, dir_type='', filename=None):
994 paths = self.params.get('paths', {})
995 assert isinstance(paths, dict)
996 path = os.path.join(
997 expand_path(paths.get('home', '').strip()),
998 expand_path(paths.get(dir_type, '').strip()) if dir_type else '',
999 filename or '')
1000
1001 # Temporary fix for #4787
1002 # 'Treat' all problem characters by passing filename through preferredencoding
1003 # to workaround encoding issues with subprocess on python2 @ Windows
1004 if sys.version_info < (3, 0) and sys.platform == 'win32':
1005 path = encodeFilename(path, True).decode(preferredencoding())
1006 return sanitize_path(path, force=self.params.get('windowsfilenames'))
1007
1008 @staticmethod
1009 def _outtmpl_expandpath(outtmpl):
1010 # expand_path translates '%%' into '%' and '$$' into '$'
1011 # correspondingly that is not what we want since we need to keep
1012 # '%%' intact for template dict substitution step. Working around
1013 # with boundary-alike separator hack.
1014 sep = ''.join([random.choice(ascii_letters) for _ in range(32)])
1015 outtmpl = outtmpl.replace('%%', '%{0}%'.format(sep)).replace('$$', '${0}$'.format(sep))
1016
1017 # outtmpl should be expand_path'ed before template dict substitution
1018 # because meta fields may contain env variables we don't want to
1019 # be expanded. For example, for outtmpl "%(title)s.%(ext)s" and
1020 # title "Hello $PATH", we don't want `$PATH` to be expanded.
1021 return expand_path(outtmpl).replace(sep, '')
1022
1023 @staticmethod
1024 def escape_outtmpl(outtmpl):
1025 ''' Escape any remaining strings like %s, %abc% etc. '''
1026 return re.sub(
1027 STR_FORMAT_RE_TMPL.format('', '(?![%(\0])'),
1028 lambda mobj: ('' if mobj.group('has_key') else '%') + mobj.group(0),
1029 outtmpl)
1030
1031 @classmethod
1032 def validate_outtmpl(cls, outtmpl):
1033 ''' @return None or Exception object '''
1034 outtmpl = re.sub(
1035 STR_FORMAT_RE_TMPL.format('[^)]*', '[ljqBUDS]'),
1036 lambda mobj: f'{mobj.group(0)[:-1]}s',
1037 cls._outtmpl_expandpath(outtmpl))
1038 try:
1039 cls.escape_outtmpl(outtmpl) % collections.defaultdict(int)
1040 return None
1041 except ValueError as err:
1042 return err
1043
1044 @staticmethod
1045 def _copy_infodict(info_dict):
1046 info_dict = dict(info_dict)
1047 info_dict.pop('__postprocessors', None)
1048 return info_dict
1049
1050 def prepare_outtmpl(self, outtmpl, info_dict, sanitize=False):
1051 """ Make the outtmpl and info_dict suitable for substitution: ydl.escape_outtmpl(outtmpl) % info_dict
1052 @param sanitize Whether to sanitize the output as a filename.
1053 For backward compatibility, a function can also be passed
1054 """
1055
1056 info_dict.setdefault('epoch', int(time.time())) # keep epoch consistent once set
1057
1058 info_dict = self._copy_infodict(info_dict)
1059 info_dict['duration_string'] = ( # %(duration>%H-%M-%S)s is wrong if duration > 24hrs
1060 formatSeconds(info_dict['duration'], '-' if sanitize else ':')
1061 if info_dict.get('duration', None) is not None
1062 else None)
1063 info_dict['autonumber'] = self.params.get('autonumber_start', 1) - 1 + self._num_downloads
1064 info_dict['video_autonumber'] = self._num_videos
1065 if info_dict.get('resolution') is None:
1066 info_dict['resolution'] = self.format_resolution(info_dict, default=None)
1067
1068 # For fields playlist_index, playlist_autonumber and autonumber convert all occurrences
1069 # of %(field)s to %(field)0Nd for backward compatibility
1070 field_size_compat_map = {
1071 'playlist_index': number_of_digits(info_dict.get('_last_playlist_index') or 0),
1072 'playlist_autonumber': number_of_digits(info_dict.get('n_entries') or 0),
1073 'autonumber': self.params.get('autonumber_size') or 5,
1074 }
1075
1076 TMPL_DICT = {}
1077 EXTERNAL_FORMAT_RE = re.compile(STR_FORMAT_RE_TMPL.format('[^)]*', f'[{STR_FORMAT_TYPES}ljqBUDS]'))
1078 MATH_FUNCTIONS = {
1079 '+': float.__add__,
1080 '-': float.__sub__,
1081 }
1082 # Field is of the form key1.key2...
1083 # where keys (except first) can be string, int or slice
1084 FIELD_RE = r'\w*(?:\.(?:\w+|{num}|{num}?(?::{num}?){{1,2}}))*'.format(num=r'(?:-?\d+)')
1085 MATH_FIELD_RE = r'''(?:{field}|{num})'''.format(field=FIELD_RE, num=r'-?\d+(?:.\d+)?')
1086 MATH_OPERATORS_RE = r'(?:%s)' % '|'.join(map(re.escape, MATH_FUNCTIONS.keys()))
1087 INTERNAL_FORMAT_RE = re.compile(r'''(?x)
1088 (?P<negate>-)?
1089 (?P<fields>{field})
1090 (?P<maths>(?:{math_op}{math_field})*)
1091 (?:>(?P<strf_format>.+?))?
1092 (?P<alternate>(?<!\\),[^|&)]+)?
1093 (?:&(?P<replacement>.*?))?
1094 (?:\|(?P<default>.*?))?
1095 $'''.format(field=FIELD_RE, math_op=MATH_OPERATORS_RE, math_field=MATH_FIELD_RE))
1096
1097 def _traverse_infodict(k):
1098 k = k.split('.')
1099 if k[0] == '':
1100 k.pop(0)
1101 return traverse_obj(info_dict, k, is_user_input=True, traverse_string=True)
1102
1103 def get_value(mdict):
1104 # Object traversal
1105 value = _traverse_infodict(mdict['fields'])
1106 # Negative
1107 if mdict['negate']:
1108 value = float_or_none(value)
1109 if value is not None:
1110 value *= -1
1111 # Do maths
1112 offset_key = mdict['maths']
1113 if offset_key:
1114 value = float_or_none(value)
1115 operator = None
1116 while offset_key:
1117 item = re.match(
1118 MATH_FIELD_RE if operator else MATH_OPERATORS_RE,
1119 offset_key).group(0)
1120 offset_key = offset_key[len(item):]
1121 if operator is None:
1122 operator = MATH_FUNCTIONS[item]
1123 continue
1124 item, multiplier = (item[1:], -1) if item[0] == '-' else (item, 1)
1125 offset = float_or_none(item)
1126 if offset is None:
1127 offset = float_or_none(_traverse_infodict(item))
1128 try:
1129 value = operator(value, multiplier * offset)
1130 except (TypeError, ZeroDivisionError):
1131 return None
1132 operator = None
1133 # Datetime formatting
1134 if mdict['strf_format']:
1135 value = strftime_or_none(value, mdict['strf_format'].replace('\\,', ','))
1136
1137 return value
1138
1139 na = self.params.get('outtmpl_na_placeholder', 'NA')
1140
1141 def filename_sanitizer(key, value, restricted=self.params.get('restrictfilenames')):
1142 return sanitize_filename(str(value), restricted=restricted,
1143 is_id=re.search(r'(^|[_.])id(\.|$)', key))
1144
1145 sanitizer = sanitize if callable(sanitize) else filename_sanitizer
1146 sanitize = bool(sanitize)
1147
1148 def _dumpjson_default(obj):
1149 if isinstance(obj, (set, LazyList)):
1150 return list(obj)
1151 return repr(obj)
1152
1153 def create_key(outer_mobj):
1154 if not outer_mobj.group('has_key'):
1155 return outer_mobj.group(0)
1156 key = outer_mobj.group('key')
1157 mobj = re.match(INTERNAL_FORMAT_RE, key)
1158 initial_field = mobj.group('fields') if mobj else ''
1159 value, replacement, default = None, None, na
1160 while mobj:
1161 mobj = mobj.groupdict()
1162 default = mobj['default'] if mobj['default'] is not None else default
1163 value = get_value(mobj)
1164 replacement = mobj['replacement']
1165 if value is None and mobj['alternate']:
1166 mobj = re.match(INTERNAL_FORMAT_RE, mobj['alternate'][1:])
1167 else:
1168 break
1169
1170 fmt = outer_mobj.group('format')
1171 if fmt == 's' and value is not None and key in field_size_compat_map.keys():
1172 fmt = '0{:d}d'.format(field_size_compat_map[key])
1173
1174 value = default if value is None else value if replacement is None else replacement
1175
1176 flags = outer_mobj.group('conversion') or ''
1177 str_fmt = f'{fmt[:-1]}s'
1178 if fmt[-1] == 'l': # list
1179 delim = '\n' if '#' in flags else ', '
1180 value, fmt = delim.join(map(str, variadic(value, allowed_types=(str, bytes)))), str_fmt
1181 elif fmt[-1] == 'j': # json
1182 value, fmt = json.dumps(value, default=_dumpjson_default, indent=4 if '#' in flags else None), str_fmt
1183 elif fmt[-1] == 'q': # quoted
1184 value = map(str, variadic(value) if '#' in flags else [value])
1185 value, fmt = ' '.join(map(compat_shlex_quote, value)), str_fmt
1186 elif fmt[-1] == 'B': # bytes
1187 value = f'%{str_fmt}'.encode('utf-8') % str(value).encode('utf-8')
1188 value, fmt = value.decode('utf-8', 'ignore'), 's'
1189 elif fmt[-1] == 'U': # unicode normalized
1190 value, fmt = unicodedata.normalize(
1191 # "+" = compatibility equivalence, "#" = NFD
1192 'NF%s%s' % ('K' if '+' in flags else '', 'D' if '#' in flags else 'C'),
1193 value), str_fmt
1194 elif fmt[-1] == 'D': # decimal suffix
1195 num_fmt, fmt = fmt[:-1].replace('#', ''), 's'
1196 value = format_decimal_suffix(value, f'%{num_fmt}f%s' if num_fmt else '%d%s',
1197 factor=1024 if '#' in flags else 1000)
1198 elif fmt[-1] == 'S': # filename sanitization
1199 value, fmt = filename_sanitizer(initial_field, value, restricted='#' in flags), str_fmt
1200 elif fmt[-1] == 'c':
1201 if value:
1202 value = str(value)[0]
1203 else:
1204 fmt = str_fmt
1205 elif fmt[-1] not in 'rs': # numeric
1206 value = float_or_none(value)
1207 if value is None:
1208 value, fmt = default, 's'
1209
1210 if sanitize:
1211 if fmt[-1] == 'r':
1212 # If value is an object, sanitize might convert it to a string
1213 # So we convert it to repr first
1214 value, fmt = repr(value), str_fmt
1215 if fmt[-1] in 'csr':
1216 value = sanitizer(initial_field, value)
1217
1218 key = '%s\0%s' % (key.replace('%', '%\0'), outer_mobj.group('format'))
1219 TMPL_DICT[key] = value
1220 return '{prefix}%({key}){fmt}'.format(key=key, fmt=fmt, prefix=outer_mobj.group('prefix'))
1221
1222 return EXTERNAL_FORMAT_RE.sub(create_key, outtmpl), TMPL_DICT
1223
1224 def evaluate_outtmpl(self, outtmpl, info_dict, *args, **kwargs):
1225 outtmpl, info_dict = self.prepare_outtmpl(outtmpl, info_dict, *args, **kwargs)
1226 return self.escape_outtmpl(outtmpl) % info_dict
1227
1228 def _prepare_filename(self, info_dict, tmpl_type='default'):
1229 try:
1230 outtmpl = self._outtmpl_expandpath(self.outtmpl_dict.get(tmpl_type, self.outtmpl_dict['default']))
1231 filename = self.evaluate_outtmpl(outtmpl, info_dict, True)
1232 if not filename:
1233 return None
1234
1235 if tmpl_type in ('default', 'temp'):
1236 final_ext, ext = self.params.get('final_ext'), info_dict.get('ext')
1237 if final_ext and ext and final_ext != ext and filename.endswith(f'.{final_ext}'):
1238 filename = replace_extension(filename, ext, final_ext)
1239 else:
1240 force_ext = OUTTMPL_TYPES[tmpl_type]
1241 if force_ext:
1242 filename = replace_extension(filename, force_ext, info_dict.get('ext'))
1243
1244 # https://github.com/blackjack4494/youtube-dlc/issues/85
1245 trim_file_name = self.params.get('trim_file_name', False)
1246 if trim_file_name:
1247 no_ext, *ext = filename.rsplit('.', 2)
1248 filename = join_nonempty(no_ext[:trim_file_name], *ext, delim='.')
1249
1250 return filename
1251 except ValueError as err:
1252 self.report_error('Error in output template: ' + str(err) + ' (encoding: ' + repr(preferredencoding()) + ')')
1253 return None
1254
1255 def prepare_filename(self, info_dict, dir_type='', warn=False):
1256 """Generate the output filename."""
1257
1258 filename = self._prepare_filename(info_dict, dir_type or 'default')
1259 if not filename and dir_type not in ('', 'temp'):
1260 return ''
1261
1262 if warn:
1263 if not self.params.get('paths'):
1264 pass
1265 elif filename == '-':
1266 self.report_warning('--paths is ignored when an outputting to stdout', only_once=True)
1267 elif os.path.isabs(filename):
1268 self.report_warning('--paths is ignored since an absolute path is given in output template', only_once=True)
1269 if filename == '-' or not filename:
1270 return filename
1271
1272 return self.get_output_path(dir_type, filename)
1273
1274 def _match_entry(self, info_dict, incomplete=False, silent=False):
1275 """ Returns None if the file should be downloaded """
1276
1277 video_title = info_dict.get('title', info_dict.get('id', 'video'))
1278
1279 def check_filter():
1280 if 'title' in info_dict:
1281 # This can happen when we're just evaluating the playlist
1282 title = info_dict['title']
1283 matchtitle = self.params.get('matchtitle', False)
1284 if matchtitle:
1285 if not re.search(matchtitle, title, re.IGNORECASE):
1286 return '"' + title + '" title did not match pattern "' + matchtitle + '"'
1287 rejecttitle = self.params.get('rejecttitle', False)
1288 if rejecttitle:
1289 if re.search(rejecttitle, title, re.IGNORECASE):
1290 return '"' + title + '" title matched reject pattern "' + rejecttitle + '"'
1291 date = info_dict.get('upload_date')
1292 if date is not None:
1293 dateRange = self.params.get('daterange', DateRange())
1294 if date not in dateRange:
1295 return '%s upload date is not in range %s' % (date_from_str(date).isoformat(), dateRange)
1296 view_count = info_dict.get('view_count')
1297 if view_count is not None:
1298 min_views = self.params.get('min_views')
1299 if min_views is not None and view_count < min_views:
1300 return 'Skipping %s, because it has not reached minimum view count (%d/%d)' % (video_title, view_count, min_views)
1301 max_views = self.params.get('max_views')
1302 if max_views is not None and view_count > max_views:
1303 return 'Skipping %s, because it has exceeded the maximum view count (%d/%d)' % (video_title, view_count, max_views)
1304 if age_restricted(info_dict.get('age_limit'), self.params.get('age_limit')):
1305 return 'Skipping "%s" because it is age restricted' % video_title
1306
1307 match_filter = self.params.get('match_filter')
1308 if match_filter is not None:
1309 try:
1310 ret = match_filter(info_dict, incomplete=incomplete)
1311 except TypeError:
1312 # For backward compatibility
1313 ret = None if incomplete else match_filter(info_dict)
1314 if ret is not None:
1315 return ret
1316 return None
1317
1318 if self.in_download_archive(info_dict):
1319 reason = '%s has already been recorded in the archive' % video_title
1320 break_opt, break_err = 'break_on_existing', ExistingVideoReached
1321 else:
1322 reason = check_filter()
1323 break_opt, break_err = 'break_on_reject', RejectedVideoReached
1324 if reason is not None:
1325 if not silent:
1326 self.to_screen('[download] ' + reason)
1327 if self.params.get(break_opt, False):
1328 raise break_err()
1329 return reason
1330
1331 @staticmethod
1332 def add_extra_info(info_dict, extra_info):
1333 '''Set the keys from extra_info in info dict if they are missing'''
1334 for key, value in extra_info.items():
1335 info_dict.setdefault(key, value)
1336
1337 def extract_info(self, url, download=True, ie_key=None, extra_info=None,
1338 process=True, force_generic_extractor=False):
1339 """
1340 Return a list with a dictionary for each video extracted.
1341
1342 Arguments:
1343 url -- URL to extract
1344
1345 Keyword arguments:
1346 download -- whether to download videos during extraction
1347 ie_key -- extractor key hint
1348 extra_info -- dictionary containing the extra values to add to each result
1349 process -- whether to resolve all unresolved references (URLs, playlist items),
1350 must be True for download to work.
1351 force_generic_extractor -- force using the generic extractor
1352 """
1353
1354 if extra_info is None:
1355 extra_info = {}
1356
1357 if not ie_key and force_generic_extractor:
1358 ie_key = 'Generic'
1359
1360 if ie_key:
1361 ies = {ie_key: self._get_info_extractor_class(ie_key)}
1362 else:
1363 ies = self._ies
1364
1365 for ie_key, ie in ies.items():
1366 if not ie.suitable(url):
1367 continue
1368
1369 if not ie.working():
1370 self.report_warning('The program functionality for this site has been marked as broken, '
1371 'and will probably not work.')
1372
1373 temp_id = ie.get_temp_id(url)
1374 if temp_id is not None and self.in_download_archive({'id': temp_id, 'ie_key': ie_key}):
1375 self.to_screen(f'[{ie_key}] {temp_id}: has already been recorded in the archive')
1376 if self.params.get('break_on_existing', False):
1377 raise ExistingVideoReached()
1378 break
1379 return self.__extract_info(url, self.get_info_extractor(ie_key), download, extra_info, process)
1380 else:
1381 self.report_error('no suitable InfoExtractor for URL %s' % url)
1382
1383 def __handle_extraction_exceptions(func):
1384 @functools.wraps(func)
1385 def wrapper(self, *args, **kwargs):
1386 while True:
1387 try:
1388 return func(self, *args, **kwargs)
1389 except (DownloadCancelled, LazyList.IndexError, PagedList.IndexError):
1390 raise
1391 except ReExtractInfo as e:
1392 if e.expected:
1393 self.to_screen(f'{e}; Re-extracting data')
1394 else:
1395 self.to_stderr('\r')
1396 self.report_warning(f'{e}; Re-extracting data')
1397 continue
1398 except GeoRestrictedError as e:
1399 msg = e.msg
1400 if e.countries:
1401 msg += '\nThis video is available in %s.' % ', '.join(
1402 map(ISO3166Utils.short2full, e.countries))
1403 msg += '\nYou might want to use a VPN or a proxy server (with --proxy) to workaround.'
1404 self.report_error(msg)
1405 except ExtractorError as e: # An error we somewhat expected
1406 self.report_error(str(e), e.format_traceback())
1407 except Exception as e:
1408 if self.params.get('ignoreerrors'):
1409 self.report_error(str(e), tb=encode_compat_str(traceback.format_exc()))
1410 else:
1411 raise
1412 break
1413 return wrapper
1414
1415 def _wait_for_video(self, ie_result):
1416 if (not self.params.get('wait_for_video')
1417 or ie_result.get('_type', 'video') != 'video'
1418 or ie_result.get('formats') or ie_result.get('url')):
1419 return
1420
1421 format_dur = lambda dur: '%02d:%02d:%02d' % timetuple_from_msec(dur * 1000)[:-1]
1422 last_msg = ''
1423
1424 def progress(msg):
1425 nonlocal last_msg
1426 self.to_screen(msg + ' ' * (len(last_msg) - len(msg)) + '\r', skip_eol=True)
1427 last_msg = msg
1428
1429 min_wait, max_wait = self.params.get('wait_for_video')
1430 diff = try_get(ie_result, lambda x: x['release_timestamp'] - time.time())
1431 if diff is None and ie_result.get('live_status') == 'is_upcoming':
1432 diff = round(random.uniform(min_wait, max_wait) if (max_wait and min_wait) else (max_wait or min_wait), 0)
1433 self.report_warning('Release time of video is not known')
1434 elif (diff or 0) <= 0:
1435 self.report_warning('Video should already be available according to extracted info')
1436 diff = min(max(diff or 0, min_wait or 0), max_wait or float('inf'))
1437 self.to_screen(f'[wait] Waiting for {format_dur(diff)} - Press Ctrl+C to try now')
1438
1439 wait_till = time.time() + diff
1440 try:
1441 while True:
1442 diff = wait_till - time.time()
1443 if diff <= 0:
1444 progress('')
1445 raise ReExtractInfo('[wait] Wait period ended', expected=True)
1446 progress(f'[wait] Remaining time until next attempt: {self._format_screen(format_dur(diff), self.Styles.EMPHASIS)}')
1447 time.sleep(1)
1448 except KeyboardInterrupt:
1449 progress('')
1450 raise ReExtractInfo('[wait] Interrupted by user', expected=True)
1451 except BaseException as e:
1452 if not isinstance(e, ReExtractInfo):
1453 self.to_screen('')
1454 raise
1455
1456 @__handle_extraction_exceptions
1457 def __extract_info(self, url, ie, download, extra_info, process):
1458 ie_result = ie.extract(url)
1459 if ie_result is None: # Finished already (backwards compatibility; listformats and friends should be moved here)
1460 return
1461 if isinstance(ie_result, list):
1462 # Backwards compatibility: old IE result format
1463 ie_result = {
1464 '_type': 'compat_list',
1465 'entries': ie_result,
1466 }
1467 if extra_info.get('original_url'):
1468 ie_result.setdefault('original_url', extra_info['original_url'])
1469 self.add_default_extra_info(ie_result, ie, url)
1470 if process:
1471 self._wait_for_video(ie_result)
1472 return self.process_ie_result(ie_result, download, extra_info)
1473 else:
1474 return ie_result
1475
1476 def add_default_extra_info(self, ie_result, ie, url):
1477 if url is not None:
1478 self.add_extra_info(ie_result, {
1479 'webpage_url': url,
1480 'original_url': url,
1481 })
1482 webpage_url = ie_result.get('webpage_url')
1483 if webpage_url:
1484 self.add_extra_info(ie_result, {
1485 'webpage_url_basename': url_basename(webpage_url),
1486 'webpage_url_domain': get_domain(webpage_url),
1487 })
1488 if ie is not None:
1489 self.add_extra_info(ie_result, {
1490 'extractor': ie.IE_NAME,
1491 'extractor_key': ie.ie_key(),
1492 })
1493
1494 def process_ie_result(self, ie_result, download=True, extra_info=None):
1495 """
1496 Take the result of the ie(may be modified) and resolve all unresolved
1497 references (URLs, playlist items).
1498
1499 It will also download the videos if 'download'.
1500 Returns the resolved ie_result.
1501 """
1502 if extra_info is None:
1503 extra_info = {}
1504 result_type = ie_result.get('_type', 'video')
1505
1506 if result_type in ('url', 'url_transparent'):
1507 ie_result['url'] = sanitize_url(ie_result['url'])
1508 if ie_result.get('original_url'):
1509 extra_info.setdefault('original_url', ie_result['original_url'])
1510
1511 extract_flat = self.params.get('extract_flat', False)
1512 if ((extract_flat == 'in_playlist' and 'playlist' in extra_info)
1513 or extract_flat is True):
1514 info_copy = ie_result.copy()
1515 ie = try_get(ie_result.get('ie_key'), self.get_info_extractor)
1516 if ie and not ie_result.get('id'):
1517 info_copy['id'] = ie.get_temp_id(ie_result['url'])
1518 self.add_default_extra_info(info_copy, ie, ie_result['url'])
1519 self.add_extra_info(info_copy, extra_info)
1520 info_copy, _ = self.pre_process(info_copy)
1521 self.__forced_printings(info_copy, self.prepare_filename(info_copy), incomplete=True)
1522 if self.params.get('force_write_download_archive', False):
1523 self.record_download_archive(info_copy)
1524 return ie_result
1525
1526 if result_type == 'video':
1527 self.add_extra_info(ie_result, extra_info)
1528 ie_result = self.process_video_result(ie_result, download=download)
1529 additional_urls = (ie_result or {}).get('additional_urls')
1530 if additional_urls:
1531 # TODO: Improve MetadataParserPP to allow setting a list
1532 if isinstance(additional_urls, compat_str):
1533 additional_urls = [additional_urls]
1534 self.to_screen(
1535 '[info] %s: %d additional URL(s) requested' % (ie_result['id'], len(additional_urls)))
1536 self.write_debug('Additional URLs: "%s"' % '", "'.join(additional_urls))
1537 ie_result['additional_entries'] = [
1538 self.extract_info(
1539 url, download, extra_info=extra_info,
1540 force_generic_extractor=self.params.get('force_generic_extractor'))
1541 for url in additional_urls
1542 ]
1543 return ie_result
1544 elif result_type == 'url':
1545 # We have to add extra_info to the results because it may be
1546 # contained in a playlist
1547 return self.extract_info(
1548 ie_result['url'], download,
1549 ie_key=ie_result.get('ie_key'),
1550 extra_info=extra_info)
1551 elif result_type == 'url_transparent':
1552 # Use the information from the embedding page
1553 info = self.extract_info(
1554 ie_result['url'], ie_key=ie_result.get('ie_key'),
1555 extra_info=extra_info, download=False, process=False)
1556
1557 # extract_info may return None when ignoreerrors is enabled and
1558 # extraction failed with an error, don't crash and return early
1559 # in this case
1560 if not info:
1561 return info
1562
1563 force_properties = dict(
1564 (k, v) for k, v in ie_result.items() if v is not None)
1565 for f in ('_type', 'url', 'id', 'extractor', 'extractor_key', 'ie_key'):
1566 if f in force_properties:
1567 del force_properties[f]
1568 new_result = info.copy()
1569 new_result.update(force_properties)
1570
1571 # Extracted info may not be a video result (i.e.
1572 # info.get('_type', 'video') != video) but rather an url or
1573 # url_transparent. In such cases outer metadata (from ie_result)
1574 # should be propagated to inner one (info). For this to happen
1575 # _type of info should be overridden with url_transparent. This
1576 # fixes issue from https://github.com/ytdl-org/youtube-dl/pull/11163.
1577 if new_result.get('_type') == 'url':
1578 new_result['_type'] = 'url_transparent'
1579
1580 return self.process_ie_result(
1581 new_result, download=download, extra_info=extra_info)
1582 elif result_type in ('playlist', 'multi_video'):
1583 # Protect from infinite recursion due to recursively nested playlists
1584 # (see https://github.com/ytdl-org/youtube-dl/issues/27833)
1585 webpage_url = ie_result['webpage_url']
1586 if webpage_url in self._playlist_urls:
1587 self.to_screen(
1588 '[download] Skipping already downloaded playlist: %s'
1589 % ie_result.get('title') or ie_result.get('id'))
1590 return
1591
1592 self._playlist_level += 1
1593 self._playlist_urls.add(webpage_url)
1594 self._fill_common_fields(ie_result, False)
1595 self._sanitize_thumbnails(ie_result)
1596 try:
1597 return self.__process_playlist(ie_result, download)
1598 finally:
1599 self._playlist_level -= 1
1600 if not self._playlist_level:
1601 self._playlist_urls.clear()
1602 elif result_type == 'compat_list':
1603 self.report_warning(
1604 'Extractor %s returned a compat_list result. '
1605 'It needs to be updated.' % ie_result.get('extractor'))
1606
1607 def _fixup(r):
1608 self.add_extra_info(r, {
1609 'extractor': ie_result['extractor'],
1610 'webpage_url': ie_result['webpage_url'],
1611 'webpage_url_basename': url_basename(ie_result['webpage_url']),
1612 'webpage_url_domain': get_domain(ie_result['webpage_url']),
1613 'extractor_key': ie_result['extractor_key'],
1614 })
1615 return r
1616 ie_result['entries'] = [
1617 self.process_ie_result(_fixup(r), download, extra_info)
1618 for r in ie_result['entries']
1619 ]
1620 return ie_result
1621 else:
1622 raise Exception('Invalid result type: %s' % result_type)
1623
1624 def _ensure_dir_exists(self, path):
1625 return make_dir(path, self.report_error)
1626
1627 @staticmethod
1628 def _playlist_infodict(ie_result, **kwargs):
1629 return {
1630 **ie_result,
1631 'playlist': ie_result.get('title') or ie_result.get('id'),
1632 'playlist_id': ie_result.get('id'),
1633 'playlist_title': ie_result.get('title'),
1634 'playlist_uploader': ie_result.get('uploader'),
1635 'playlist_uploader_id': ie_result.get('uploader_id'),
1636 'playlist_index': 0,
1637 **kwargs,
1638 }
1639
1640 def __process_playlist(self, ie_result, download):
1641 # We process each entry in the playlist
1642 playlist = ie_result.get('title') or ie_result.get('id')
1643 self.to_screen('[download] Downloading playlist: %s' % playlist)
1644
1645 if 'entries' not in ie_result:
1646 raise EntryNotInPlaylist('There are no entries')
1647
1648 MissingEntry = object()
1649 incomplete_entries = bool(ie_result.get('requested_entries'))
1650 if incomplete_entries:
1651 def fill_missing_entries(entries, indices):
1652 ret = [MissingEntry] * max(indices)
1653 for i, entry in zip(indices, entries):
1654 ret[i - 1] = entry
1655 return ret
1656 ie_result['entries'] = fill_missing_entries(ie_result['entries'], ie_result['requested_entries'])
1657
1658 playlist_results = []
1659
1660 playliststart = self.params.get('playliststart', 1)
1661 playlistend = self.params.get('playlistend')
1662 # For backwards compatibility, interpret -1 as whole list
1663 if playlistend == -1:
1664 playlistend = None
1665
1666 playlistitems_str = self.params.get('playlist_items')
1667 playlistitems = None
1668 if playlistitems_str is not None:
1669 def iter_playlistitems(format):
1670 for string_segment in format.split(','):
1671 if '-' in string_segment:
1672 start, end = string_segment.split('-')
1673 for item in range(int(start), int(end) + 1):
1674 yield int(item)
1675 else:
1676 yield int(string_segment)
1677 playlistitems = orderedSet(iter_playlistitems(playlistitems_str))
1678
1679 ie_entries = ie_result['entries']
1680 if isinstance(ie_entries, list):
1681 playlist_count = len(ie_entries)
1682 msg = f'Collected {playlist_count} videos; downloading %d of them'
1683 ie_result['playlist_count'] = ie_result.get('playlist_count') or playlist_count
1684
1685 def get_entry(i):
1686 return ie_entries[i - 1]
1687 else:
1688 msg = 'Downloading %d videos'
1689 if not isinstance(ie_entries, (PagedList, LazyList)):
1690 ie_entries = LazyList(ie_entries)
1691 elif isinstance(ie_entries, InAdvancePagedList):
1692 if ie_entries._pagesize == 1:
1693 playlist_count = ie_entries._pagecount
1694
1695 def get_entry(i):
1696 return YoutubeDL.__handle_extraction_exceptions(
1697 lambda self, i: ie_entries[i - 1]
1698 )(self, i)
1699
1700 entries, broken = [], False
1701 items = playlistitems if playlistitems is not None else itertools.count(playliststart)
1702 for i in items:
1703 if i == 0:
1704 continue
1705 if playlistitems is None and playlistend is not None and playlistend < i:
1706 break
1707 entry = None
1708 try:
1709 entry = get_entry(i)
1710 if entry is MissingEntry:
1711 raise EntryNotInPlaylist()
1712 except (IndexError, EntryNotInPlaylist):
1713 if incomplete_entries:
1714 raise EntryNotInPlaylist(f'Entry {i} cannot be found')
1715 elif not playlistitems:
1716 break
1717 entries.append(entry)
1718 try:
1719 if entry is not None:
1720 self._match_entry(entry, incomplete=True, silent=True)
1721 except (ExistingVideoReached, RejectedVideoReached):
1722 broken = True
1723 break
1724 ie_result['entries'] = entries
1725
1726 # Save playlist_index before re-ordering
1727 entries = [
1728 ((playlistitems[i - 1] if playlistitems else i + playliststart - 1), entry)
1729 for i, entry in enumerate(entries, 1)
1730 if entry is not None]
1731 n_entries = len(entries)
1732
1733 if not (ie_result.get('playlist_count') or broken or playlistitems or playlistend):
1734 ie_result['playlist_count'] = n_entries
1735
1736 if not playlistitems and (playliststart != 1 or playlistend):
1737 playlistitems = list(range(playliststart, playliststart + n_entries))
1738 ie_result['requested_entries'] = playlistitems
1739
1740 _infojson_written = False
1741 write_playlist_files = self.params.get('allow_playlist_files', True)
1742 if write_playlist_files and self.params.get('list_thumbnails'):
1743 self.list_thumbnails(ie_result)
1744 if write_playlist_files and not self.params.get('simulate'):
1745 ie_copy = self._playlist_infodict(ie_result, n_entries=n_entries)
1746 _infojson_written = self._write_info_json(
1747 'playlist', ie_result, self.prepare_filename(ie_copy, 'pl_infojson'))
1748 if _infojson_written is None:
1749 return
1750 if self._write_description('playlist', ie_result,
1751 self.prepare_filename(ie_copy, 'pl_description')) is None:
1752 return
1753 # TODO: This should be passed to ThumbnailsConvertor if necessary
1754 self._write_thumbnails('playlist', ie_copy, self.prepare_filename(ie_copy, 'pl_thumbnail'))
1755
1756 if self.params.get('playlistreverse', False):
1757 entries = entries[::-1]
1758 if self.params.get('playlistrandom', False):
1759 random.shuffle(entries)
1760
1761 x_forwarded_for = ie_result.get('__x_forwarded_for_ip')
1762
1763 self.to_screen('[%s] playlist %s: %s' % (ie_result['extractor'], playlist, msg % n_entries))
1764 failures = 0
1765 max_failures = self.params.get('skip_playlist_after_errors') or float('inf')
1766 for i, entry_tuple in enumerate(entries, 1):
1767 playlist_index, entry = entry_tuple
1768 if 'playlist-index' in self.params.get('compat_opts', []):
1769 playlist_index = playlistitems[i - 1] if playlistitems else i + playliststart - 1
1770 self.to_screen('[download] Downloading video %s of %s' % (i, n_entries))
1771 # This __x_forwarded_for_ip thing is a bit ugly but requires
1772 # minimal changes
1773 if x_forwarded_for:
1774 entry['__x_forwarded_for_ip'] = x_forwarded_for
1775 extra = {
1776 'n_entries': n_entries,
1777 '_last_playlist_index': max(playlistitems) if playlistitems else (playlistend or n_entries),
1778 'playlist_count': ie_result.get('playlist_count'),
1779 'playlist_index': playlist_index,
1780 'playlist_autonumber': i,
1781 'playlist': playlist,
1782 'playlist_id': ie_result.get('id'),
1783 'playlist_title': ie_result.get('title'),
1784 'playlist_uploader': ie_result.get('uploader'),
1785 'playlist_uploader_id': ie_result.get('uploader_id'),
1786 'extractor': ie_result['extractor'],
1787 'webpage_url': ie_result['webpage_url'],
1788 'webpage_url_basename': url_basename(ie_result['webpage_url']),
1789 'webpage_url_domain': get_domain(ie_result['webpage_url']),
1790 'extractor_key': ie_result['extractor_key'],
1791 }
1792
1793 if self._match_entry(entry, incomplete=True) is not None:
1794 continue
1795
1796 entry_result = self.__process_iterable_entry(entry, download, extra)
1797 if not entry_result:
1798 failures += 1
1799 if failures >= max_failures:
1800 self.report_error(
1801 'Skipping the remaining entries in playlist "%s" since %d items failed extraction' % (playlist, failures))
1802 break
1803 playlist_results.append(entry_result)
1804 ie_result['entries'] = playlist_results
1805
1806 # Write the updated info to json
1807 if _infojson_written and self._write_info_json(
1808 'updated playlist', ie_result,
1809 self.prepare_filename(ie_copy, 'pl_infojson'), overwrite=True) is None:
1810 return
1811
1812 ie_result = self.run_all_pps('playlist', ie_result)
1813 self.to_screen(f'[download] Finished downloading playlist: {playlist}')
1814 return ie_result
1815
1816 @__handle_extraction_exceptions
1817 def __process_iterable_entry(self, entry, download, extra_info):
1818 return self.process_ie_result(
1819 entry, download=download, extra_info=extra_info)
1820
1821 def _build_format_filter(self, filter_spec):
1822 " Returns a function to filter the formats according to the filter_spec "
1823
1824 OPERATORS = {
1825 '<': operator.lt,
1826 '<=': operator.le,
1827 '>': operator.gt,
1828 '>=': operator.ge,
1829 '=': operator.eq,
1830 '!=': operator.ne,
1831 }
1832 operator_rex = re.compile(r'''(?x)\s*
1833 (?P<key>width|height|tbr|abr|vbr|asr|filesize|filesize_approx|fps)\s*
1834 (?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
1835 (?P<value>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?)\s*
1836 ''' % '|'.join(map(re.escape, OPERATORS.keys())))
1837 m = operator_rex.fullmatch(filter_spec)
1838 if m:
1839 try:
1840 comparison_value = int(m.group('value'))
1841 except ValueError:
1842 comparison_value = parse_filesize(m.group('value'))
1843 if comparison_value is None:
1844 comparison_value = parse_filesize(m.group('value') + 'B')
1845 if comparison_value is None:
1846 raise ValueError(
1847 'Invalid value %r in format specification %r' % (
1848 m.group('value'), filter_spec))
1849 op = OPERATORS[m.group('op')]
1850
1851 if not m:
1852 STR_OPERATORS = {
1853 '=': operator.eq,
1854 '^=': lambda attr, value: attr.startswith(value),
1855 '$=': lambda attr, value: attr.endswith(value),
1856 '*=': lambda attr, value: value in attr,
1857 '~=': lambda attr, value: value.search(attr) is not None
1858 }
1859 str_operator_rex = re.compile(r'''(?x)\s*
1860 (?P<key>[a-zA-Z0-9._-]+)\s*
1861 (?P<negation>!\s*)?(?P<op>%s)\s*(?P<none_inclusive>\?\s*)?
1862 (?P<quote>["'])?
1863 (?P<value>(?(quote)(?:(?!(?P=quote))[^\\]|\\.)+|[\w.-]+))
1864 (?(quote)(?P=quote))\s*
1865 ''' % '|'.join(map(re.escape, STR_OPERATORS.keys())))
1866 m = str_operator_rex.fullmatch(filter_spec)
1867 if m:
1868 if m.group('op') == '~=':
1869 comparison_value = re.compile(m.group('value'))
1870 else:
1871 comparison_value = re.sub(r'''\\([\\"'])''', r'\1', m.group('value'))
1872 str_op = STR_OPERATORS[m.group('op')]
1873 if m.group('negation'):
1874 op = lambda attr, value: not str_op(attr, value)
1875 else:
1876 op = str_op
1877
1878 if not m:
1879 raise SyntaxError('Invalid filter specification %r' % filter_spec)
1880
1881 def _filter(f):
1882 actual_value = f.get(m.group('key'))
1883 if actual_value is None:
1884 return m.group('none_inclusive')
1885 return op(actual_value, comparison_value)
1886 return _filter
1887
1888 def _check_formats(self, formats):
1889 for f in formats:
1890 self.to_screen('[info] Testing format %s' % f['format_id'])
1891 path = self.get_output_path('temp')
1892 if not self._ensure_dir_exists(f'{path}/'):
1893 continue
1894 temp_file = tempfile.NamedTemporaryFile(suffix='.tmp', delete=False, dir=path or None)
1895 temp_file.close()
1896 try:
1897 success, _ = self.dl(temp_file.name, f, test=True)
1898 except (DownloadError, IOError, OSError, ValueError) + network_exceptions:
1899 success = False
1900 finally:
1901 if os.path.exists(temp_file.name):
1902 try:
1903 os.remove(temp_file.name)
1904 except OSError:
1905 self.report_warning('Unable to delete temporary file "%s"' % temp_file.name)
1906 if success:
1907 yield f
1908 else:
1909 self.to_screen('[info] Unable to download format %s. Skipping...' % f['format_id'])
1910
1911 def _default_format_spec(self, info_dict, download=True):
1912
1913 def can_merge():
1914 merger = FFmpegMergerPP(self)
1915 return merger.available and merger.can_merge()
1916
1917 prefer_best = (
1918 not self.params.get('simulate')
1919 and download
1920 and (
1921 not can_merge()
1922 or info_dict.get('is_live', False)
1923 or self.outtmpl_dict['default'] == '-'))
1924 compat = (
1925 prefer_best
1926 or self.params.get('allow_multiple_audio_streams', False)
1927 or 'format-spec' in self.params.get('compat_opts', []))
1928
1929 return (
1930 'best/bestvideo+bestaudio' if prefer_best
1931 else 'bestvideo*+bestaudio/best' if not compat
1932 else 'bestvideo+bestaudio/best')
1933
1934 def build_format_selector(self, format_spec):
1935 def syntax_error(note, start):
1936 message = (
1937 'Invalid format specification: '
1938 '{0}\n\t{1}\n\t{2}^'.format(note, format_spec, ' ' * start[1]))
1939 return SyntaxError(message)
1940
1941 PICKFIRST = 'PICKFIRST'
1942 MERGE = 'MERGE'
1943 SINGLE = 'SINGLE'
1944 GROUP = 'GROUP'
1945 FormatSelector = collections.namedtuple('FormatSelector', ['type', 'selector', 'filters'])
1946
1947 allow_multiple_streams = {'audio': self.params.get('allow_multiple_audio_streams', False),
1948 'video': self.params.get('allow_multiple_video_streams', False)}
1949
1950 check_formats = self.params.get('check_formats') == 'selected'
1951
1952 def _parse_filter(tokens):
1953 filter_parts = []
1954 for type, string, start, _, _ in tokens:
1955 if type == tokenize.OP and string == ']':
1956 return ''.join(filter_parts)
1957 else:
1958 filter_parts.append(string)
1959
1960 def _remove_unused_ops(tokens):
1961 # Remove operators that we don't use and join them with the surrounding strings
1962 # for example: 'mp4' '-' 'baseline' '-' '16x9' is converted to 'mp4-baseline-16x9'
1963 ALLOWED_OPS = ('/', '+', ',', '(', ')')
1964 last_string, last_start, last_end, last_line = None, None, None, None
1965 for type, string, start, end, line in tokens:
1966 if type == tokenize.OP and string == '[':
1967 if last_string:
1968 yield tokenize.NAME, last_string, last_start, last_end, last_line
1969 last_string = None
1970 yield type, string, start, end, line
1971 # everything inside brackets will be handled by _parse_filter
1972 for type, string, start, end, line in tokens:
1973 yield type, string, start, end, line
1974 if type == tokenize.OP and string == ']':
1975 break
1976 elif type == tokenize.OP and string in ALLOWED_OPS:
1977 if last_string:
1978 yield tokenize.NAME, last_string, last_start, last_end, last_line
1979 last_string = None
1980 yield type, string, start, end, line
1981 elif type in [tokenize.NAME, tokenize.NUMBER, tokenize.OP]:
1982 if not last_string:
1983 last_string = string
1984 last_start = start
1985 last_end = end
1986 else:
1987 last_string += string
1988 if last_string:
1989 yield tokenize.NAME, last_string, last_start, last_end, last_line
1990
1991 def _parse_format_selection(tokens, inside_merge=False, inside_choice=False, inside_group=False):
1992 selectors = []
1993 current_selector = None
1994 for type, string, start, _, _ in tokens:
1995 # ENCODING is only defined in python 3.x
1996 if type == getattr(tokenize, 'ENCODING', None):
1997 continue
1998 elif type in [tokenize.NAME, tokenize.NUMBER]:
1999 current_selector = FormatSelector(SINGLE, string, [])
2000 elif type == tokenize.OP:
2001 if string == ')':
2002 if not inside_group:
2003 # ')' will be handled by the parentheses group
2004 tokens.restore_last_token()
2005 break
2006 elif inside_merge and string in ['/', ',']:
2007 tokens.restore_last_token()
2008 break
2009 elif inside_choice and string == ',':
2010 tokens.restore_last_token()
2011 break
2012 elif string == ',':
2013 if not current_selector:
2014 raise syntax_error('"," must follow a format selector', start)
2015 selectors.append(current_selector)
2016 current_selector = None
2017 elif string == '/':
2018 if not current_selector:
2019 raise syntax_error('"/" must follow a format selector', start)
2020 first_choice = current_selector
2021 second_choice = _parse_format_selection(tokens, inside_choice=True)
2022 current_selector = FormatSelector(PICKFIRST, (first_choice, second_choice), [])
2023 elif string == '[':
2024 if not current_selector:
2025 current_selector = FormatSelector(SINGLE, 'best', [])
2026 format_filter = _parse_filter(tokens)
2027 current_selector.filters.append(format_filter)
2028 elif string == '(':
2029 if current_selector:
2030 raise syntax_error('Unexpected "("', start)
2031 group = _parse_format_selection(tokens, inside_group=True)
2032 current_selector = FormatSelector(GROUP, group, [])
2033 elif string == '+':
2034 if not current_selector:
2035 raise syntax_error('Unexpected "+"', start)
2036 selector_1 = current_selector
2037 selector_2 = _parse_format_selection(tokens, inside_merge=True)
2038 if not selector_2:
2039 raise syntax_error('Expected a selector', start)
2040 current_selector = FormatSelector(MERGE, (selector_1, selector_2), [])
2041 else:
2042 raise syntax_error('Operator not recognized: "{0}"'.format(string), start)
2043 elif type == tokenize.ENDMARKER:
2044 break
2045 if current_selector:
2046 selectors.append(current_selector)
2047 return selectors
2048
2049 def _merge(formats_pair):
2050 format_1, format_2 = formats_pair
2051
2052 formats_info = []
2053 formats_info.extend(format_1.get('requested_formats', (format_1,)))
2054 formats_info.extend(format_2.get('requested_formats', (format_2,)))
2055
2056 if not allow_multiple_streams['video'] or not allow_multiple_streams['audio']:
2057 get_no_more = {'video': False, 'audio': False}
2058 for (i, fmt_info) in enumerate(formats_info):
2059 if fmt_info.get('acodec') == fmt_info.get('vcodec') == 'none':
2060 formats_info.pop(i)
2061 continue
2062 for aud_vid in ['audio', 'video']:
2063 if not allow_multiple_streams[aud_vid] and fmt_info.get(aud_vid[0] + 'codec') != 'none':
2064 if get_no_more[aud_vid]:
2065 formats_info.pop(i)
2066 break
2067 get_no_more[aud_vid] = True
2068
2069 if len(formats_info) == 1:
2070 return formats_info[0]
2071
2072 video_fmts = [fmt_info for fmt_info in formats_info if fmt_info.get('vcodec') != 'none']
2073 audio_fmts = [fmt_info for fmt_info in formats_info if fmt_info.get('acodec') != 'none']
2074
2075 the_only_video = video_fmts[0] if len(video_fmts) == 1 else None
2076 the_only_audio = audio_fmts[0] if len(audio_fmts) == 1 else None
2077
2078 output_ext = self.params.get('merge_output_format')
2079 if not output_ext:
2080 if the_only_video:
2081 output_ext = the_only_video['ext']
2082 elif the_only_audio and not video_fmts:
2083 output_ext = the_only_audio['ext']
2084 else:
2085 output_ext = 'mkv'
2086
2087 filtered = lambda *keys: filter(None, (traverse_obj(fmt, *keys) for fmt in formats_info))
2088
2089 new_dict = {
2090 'requested_formats': formats_info,
2091 'format': '+'.join(filtered('format')),
2092 'format_id': '+'.join(filtered('format_id')),
2093 'ext': output_ext,
2094 'protocol': '+'.join(map(determine_protocol, formats_info)),
2095 'language': '+'.join(orderedSet(filtered('language'))) or None,
2096 'format_note': '+'.join(orderedSet(filtered('format_note'))) or None,
2097 'filesize_approx': sum(filtered('filesize', 'filesize_approx')) or None,
2098 'tbr': sum(filtered('tbr', 'vbr', 'abr')),
2099 }
2100
2101 if the_only_video:
2102 new_dict.update({
2103 'width': the_only_video.get('width'),
2104 'height': the_only_video.get('height'),
2105 'resolution': the_only_video.get('resolution') or self.format_resolution(the_only_video),
2106 'fps': the_only_video.get('fps'),
2107 'dynamic_range': the_only_video.get('dynamic_range'),
2108 'vcodec': the_only_video.get('vcodec'),
2109 'vbr': the_only_video.get('vbr'),
2110 'stretched_ratio': the_only_video.get('stretched_ratio'),
2111 })
2112
2113 if the_only_audio:
2114 new_dict.update({
2115 'acodec': the_only_audio.get('acodec'),
2116 'abr': the_only_audio.get('abr'),
2117 'asr': the_only_audio.get('asr'),
2118 })
2119
2120 return new_dict
2121
2122 def _check_formats(formats):
2123 if not check_formats:
2124 yield from formats
2125 return
2126 yield from self._check_formats(formats)
2127
2128 def _build_selector_function(selector):
2129 if isinstance(selector, list): # ,
2130 fs = [_build_selector_function(s) for s in selector]
2131
2132 def selector_function(ctx):
2133 for f in fs:
2134 yield from f(ctx)
2135 return selector_function
2136
2137 elif selector.type == GROUP: # ()
2138 selector_function = _build_selector_function(selector.selector)
2139
2140 elif selector.type == PICKFIRST: # /
2141 fs = [_build_selector_function(s) for s in selector.selector]
2142
2143 def selector_function(ctx):
2144 for f in fs:
2145 picked_formats = list(f(ctx))
2146 if picked_formats:
2147 return picked_formats
2148 return []
2149
2150 elif selector.type == MERGE: # +
2151 selector_1, selector_2 = map(_build_selector_function, selector.selector)
2152
2153 def selector_function(ctx):
2154 for pair in itertools.product(selector_1(ctx), selector_2(ctx)):
2155 yield _merge(pair)
2156
2157 elif selector.type == SINGLE: # atom
2158 format_spec = selector.selector or 'best'
2159
2160 # TODO: Add allvideo, allaudio etc by generalizing the code with best/worst selector
2161 if format_spec == 'all':
2162 def selector_function(ctx):
2163 yield from _check_formats(ctx['formats'][::-1])
2164 elif format_spec == 'mergeall':
2165 def selector_function(ctx):
2166 formats = list(_check_formats(ctx['formats']))
2167 if not formats:
2168 return
2169 merged_format = formats[-1]
2170 for f in formats[-2::-1]:
2171 merged_format = _merge((merged_format, f))
2172 yield merged_format
2173
2174 else:
2175 format_fallback, format_reverse, format_idx = False, True, 1
2176 mobj = re.match(
2177 r'(?P<bw>best|worst|b|w)(?P<type>video|audio|v|a)?(?P<mod>\*)?(?:\.(?P<n>[1-9]\d*))?$',
2178 format_spec)
2179 if mobj is not None:
2180 format_idx = int_or_none(mobj.group('n'), default=1)
2181 format_reverse = mobj.group('bw')[0] == 'b'
2182 format_type = (mobj.group('type') or [None])[0]
2183 not_format_type = {'v': 'a', 'a': 'v'}.get(format_type)
2184 format_modified = mobj.group('mod') is not None
2185
2186 format_fallback = not format_type and not format_modified # for b, w
2187 _filter_f = (
2188 (lambda f: f.get('%scodec' % format_type) != 'none')
2189 if format_type and format_modified # bv*, ba*, wv*, wa*
2190 else (lambda f: f.get('%scodec' % not_format_type) == 'none')
2191 if format_type # bv, ba, wv, wa
2192 else (lambda f: f.get('vcodec') != 'none' and f.get('acodec') != 'none')
2193 if not format_modified # b, w
2194 else lambda f: True) # b*, w*
2195 filter_f = lambda f: _filter_f(f) and (
2196 f.get('vcodec') != 'none' or f.get('acodec') != 'none')
2197 else:
2198 if format_spec in self._format_selection_exts['audio']:
2199 filter_f = lambda f: f.get('ext') == format_spec and f.get('acodec') != 'none'
2200 elif format_spec in self._format_selection_exts['video']:
2201 filter_f = lambda f: f.get('ext') == format_spec and f.get('acodec') != 'none' and f.get('vcodec') != 'none'
2202 elif format_spec in self._format_selection_exts['storyboards']:
2203 filter_f = lambda f: f.get('ext') == format_spec and f.get('acodec') == 'none' and f.get('vcodec') == 'none'
2204 else:
2205 filter_f = lambda f: f.get('format_id') == format_spec # id
2206
2207 def selector_function(ctx):
2208 formats = list(ctx['formats'])
2209 matches = list(filter(filter_f, formats)) if filter_f is not None else formats
2210 if format_fallback and ctx['incomplete_formats'] and not matches:
2211 # for extractors with incomplete formats (audio only (soundcloud)
2212 # or video only (imgur)) best/worst will fallback to
2213 # best/worst {video,audio}-only format
2214 matches = formats
2215 matches = LazyList(_check_formats(matches[::-1 if format_reverse else 1]))
2216 try:
2217 yield matches[format_idx - 1]
2218 except IndexError:
2219 return
2220
2221 filters = [self._build_format_filter(f) for f in selector.filters]
2222
2223 def final_selector(ctx):
2224 ctx_copy = dict(ctx)
2225 for _filter in filters:
2226 ctx_copy['formats'] = list(filter(_filter, ctx_copy['formats']))
2227 return selector_function(ctx_copy)
2228 return final_selector
2229
2230 stream = io.BytesIO(format_spec.encode('utf-8'))
2231 try:
2232 tokens = list(_remove_unused_ops(compat_tokenize_tokenize(stream.readline)))
2233 except tokenize.TokenError:
2234 raise syntax_error('Missing closing/opening brackets or parenthesis', (0, len(format_spec)))
2235
2236 class TokenIterator(object):
2237 def __init__(self, tokens):
2238 self.tokens = tokens
2239 self.counter = 0
2240
2241 def __iter__(self):
2242 return self
2243
2244 def __next__(self):
2245 if self.counter >= len(self.tokens):
2246 raise StopIteration()
2247 value = self.tokens[self.counter]
2248 self.counter += 1
2249 return value
2250
2251 next = __next__
2252
2253 def restore_last_token(self):
2254 self.counter -= 1
2255
2256 parsed_selector = _parse_format_selection(iter(TokenIterator(tokens)))
2257 return _build_selector_function(parsed_selector)
2258
2259 def _calc_headers(self, info_dict):
2260 res = merge_headers(self.params['http_headers'], info_dict.get('http_headers') or {})
2261
2262 cookies = self._calc_cookies(info_dict)
2263 if cookies:
2264 res['Cookie'] = cookies
2265
2266 if 'X-Forwarded-For' not in res:
2267 x_forwarded_for_ip = info_dict.get('__x_forwarded_for_ip')
2268 if x_forwarded_for_ip:
2269 res['X-Forwarded-For'] = x_forwarded_for_ip
2270
2271 return res
2272
2273 def _calc_cookies(self, info_dict):
2274 pr = sanitized_Request(info_dict['url'])
2275 self.cookiejar.add_cookie_header(pr)
2276 return pr.get_header('Cookie')
2277
2278 def _sort_thumbnails(self, thumbnails):
2279 thumbnails.sort(key=lambda t: (
2280 t.get('preference') if t.get('preference') is not None else -1,
2281 t.get('width') if t.get('width') is not None else -1,
2282 t.get('height') if t.get('height') is not None else -1,
2283 t.get('id') if t.get('id') is not None else '',
2284 t.get('url')))
2285
2286 def _sanitize_thumbnails(self, info_dict):
2287 thumbnails = info_dict.get('thumbnails')
2288 if thumbnails is None:
2289 thumbnail = info_dict.get('thumbnail')
2290 if thumbnail:
2291 info_dict['thumbnails'] = thumbnails = [{'url': thumbnail}]
2292 if not thumbnails:
2293 return
2294
2295 def check_thumbnails(thumbnails):
2296 for t in thumbnails:
2297 self.to_screen(f'[info] Testing thumbnail {t["id"]}')
2298 try:
2299 self.urlopen(HEADRequest(t['url']))
2300 except network_exceptions as err:
2301 self.to_screen(f'[info] Unable to connect to thumbnail {t["id"]} URL {t["url"]!r} - {err}. Skipping...')
2302 continue
2303 yield t
2304
2305 self._sort_thumbnails(thumbnails)
2306 for i, t in enumerate(thumbnails):
2307 if t.get('id') is None:
2308 t['id'] = '%d' % i
2309 if t.get('width') and t.get('height'):
2310 t['resolution'] = '%dx%d' % (t['width'], t['height'])
2311 t['url'] = sanitize_url(t['url'])
2312
2313 if self.params.get('check_formats') is True:
2314 info_dict['thumbnails'] = LazyList(check_thumbnails(thumbnails[::-1]), reverse=True)
2315 else:
2316 info_dict['thumbnails'] = thumbnails
2317
2318 def _fill_common_fields(self, info_dict, is_video=True):
2319 # TODO: move sanitization here
2320 if is_video:
2321 # playlists are allowed to lack "title"
2322 info_dict['fulltitle'] = info_dict.get('title')
2323 if 'title' not in info_dict:
2324 raise ExtractorError('Missing "title" field in extractor result',
2325 video_id=info_dict['id'], ie=info_dict['extractor'])
2326 elif not info_dict.get('title'):
2327 self.report_warning('Extractor failed to obtain "title". Creating a generic title instead')
2328 info_dict['title'] = f'{info_dict["extractor"]} video #{info_dict["id"]}'
2329
2330 if info_dict.get('duration') is not None:
2331 info_dict['duration_string'] = formatSeconds(info_dict['duration'])
2332
2333 for ts_key, date_key in (
2334 ('timestamp', 'upload_date'),
2335 ('release_timestamp', 'release_date'),
2336 ('modified_timestamp', 'modified_date'),
2337 ):
2338 if info_dict.get(date_key) is None and info_dict.get(ts_key) is not None:
2339 # Working around out-of-range timestamp values (e.g. negative ones on Windows,
2340 # see http://bugs.python.org/issue1646728)
2341 try:
2342 upload_date = datetime.datetime.utcfromtimestamp(info_dict[ts_key])
2343 info_dict[date_key] = upload_date.strftime('%Y%m%d')
2344 except (ValueError, OverflowError, OSError):
2345 pass
2346
2347 live_keys = ('is_live', 'was_live')
2348 live_status = info_dict.get('live_status')
2349 if live_status is None:
2350 for key in live_keys:
2351 if info_dict.get(key) is False:
2352 continue
2353 if info_dict.get(key):
2354 live_status = key
2355 break
2356 if all(info_dict.get(key) is False for key in live_keys):
2357 live_status = 'not_live'
2358 if live_status:
2359 info_dict['live_status'] = live_status
2360 for key in live_keys:
2361 if info_dict.get(key) is None:
2362 info_dict[key] = (live_status == key)
2363
2364 # Auto generate title fields corresponding to the *_number fields when missing
2365 # in order to always have clean titles. This is very common for TV series.
2366 for field in ('chapter', 'season', 'episode'):
2367 if info_dict.get('%s_number' % field) is not None and not info_dict.get(field):
2368 info_dict[field] = '%s %d' % (field.capitalize(), info_dict['%s_number' % field])
2369
2370 def process_video_result(self, info_dict, download=True):
2371 assert info_dict.get('_type', 'video') == 'video'
2372 self._num_videos += 1
2373
2374 if 'id' not in info_dict:
2375 raise ExtractorError('Missing "id" field in extractor result', ie=info_dict['extractor'])
2376 elif not info_dict.get('id'):
2377 raise ExtractorError('Extractor failed to obtain "id"', ie=info_dict['extractor'])
2378
2379 def report_force_conversion(field, field_not, conversion):
2380 self.report_warning(
2381 '"%s" field is not %s - forcing %s conversion, there is an error in extractor'
2382 % (field, field_not, conversion))
2383
2384 def sanitize_string_field(info, string_field):
2385 field = info.get(string_field)
2386 if field is None or isinstance(field, compat_str):
2387 return
2388 report_force_conversion(string_field, 'a string', 'string')
2389 info[string_field] = compat_str(field)
2390
2391 def sanitize_numeric_fields(info):
2392 for numeric_field in self._NUMERIC_FIELDS:
2393 field = info.get(numeric_field)
2394 if field is None or isinstance(field, compat_numeric_types):
2395 continue
2396 report_force_conversion(numeric_field, 'numeric', 'int')
2397 info[numeric_field] = int_or_none(field)
2398
2399 sanitize_string_field(info_dict, 'id')
2400 sanitize_numeric_fields(info_dict)
2401 if (info_dict.get('duration') or 0) <= 0 and info_dict.pop('duration', None):
2402 self.report_warning('"duration" field is negative, there is an error in extractor')
2403
2404 if 'playlist' not in info_dict:
2405 # It isn't part of a playlist
2406 info_dict['playlist'] = None
2407 info_dict['playlist_index'] = None
2408
2409 self._sanitize_thumbnails(info_dict)
2410
2411 thumbnail = info_dict.get('thumbnail')
2412 thumbnails = info_dict.get('thumbnails')
2413 if thumbnail:
2414 info_dict['thumbnail'] = sanitize_url(thumbnail)
2415 elif thumbnails:
2416 info_dict['thumbnail'] = thumbnails[-1]['url']
2417
2418 if info_dict.get('display_id') is None and 'id' in info_dict:
2419 info_dict['display_id'] = info_dict['id']
2420
2421 self._fill_common_fields(info_dict)
2422
2423 for cc_kind in ('subtitles', 'automatic_captions'):
2424 cc = info_dict.get(cc_kind)
2425 if cc:
2426 for _, subtitle in cc.items():
2427 for subtitle_format in subtitle:
2428 if subtitle_format.get('url'):
2429 subtitle_format['url'] = sanitize_url(subtitle_format['url'])
2430 if subtitle_format.get('ext') is None:
2431 subtitle_format['ext'] = determine_ext(subtitle_format['url']).lower()
2432
2433 automatic_captions = info_dict.get('automatic_captions')
2434 subtitles = info_dict.get('subtitles')
2435
2436 info_dict['requested_subtitles'] = self.process_subtitles(
2437 info_dict['id'], subtitles, automatic_captions)
2438
2439 if info_dict.get('formats') is None:
2440 # There's only one format available
2441 formats = [info_dict]
2442 else:
2443 formats = info_dict['formats']
2444
2445 info_dict['__has_drm'] = any(f.get('has_drm') for f in formats)
2446 if not self.params.get('allow_unplayable_formats'):
2447 formats = [f for f in formats if not f.get('has_drm')]
2448
2449 get_from_start = not info_dict.get('is_live') or bool(self.params.get('live_from_start'))
2450 if not get_from_start:
2451 info_dict['title'] += ' ' + datetime.datetime.now().strftime('%Y-%m-%d %H:%M')
2452 if info_dict.get('is_live') and formats:
2453 formats = [f for f in formats if bool(f.get('is_from_start')) == get_from_start]
2454 if get_from_start and not formats:
2455 self.raise_no_formats(info_dict, msg='--live-from-start is passed, but there are no formats that can be downloaded from the start. '
2456 'If you want to download from the current time, pass --no-live-from-start')
2457
2458 if not formats:
2459 self.raise_no_formats(info_dict)
2460
2461 def is_wellformed(f):
2462 url = f.get('url')
2463 if not url:
2464 self.report_warning(
2465 '"url" field is missing or empty - skipping format, '
2466 'there is an error in extractor')
2467 return False
2468 if isinstance(url, bytes):
2469 sanitize_string_field(f, 'url')
2470 return True
2471
2472 # Filter out malformed formats for better extraction robustness
2473 formats = list(filter(is_wellformed, formats))
2474
2475 formats_dict = {}
2476
2477 # We check that all the formats have the format and format_id fields
2478 for i, format in enumerate(formats):
2479 sanitize_string_field(format, 'format_id')
2480 sanitize_numeric_fields(format)
2481 format['url'] = sanitize_url(format['url'])
2482 if not format.get('format_id'):
2483 format['format_id'] = compat_str(i)
2484 else:
2485 # Sanitize format_id from characters used in format selector expression
2486 format['format_id'] = re.sub(r'[\s,/+\[\]()]', '_', format['format_id'])
2487 format_id = format['format_id']
2488 if format_id not in formats_dict:
2489 formats_dict[format_id] = []
2490 formats_dict[format_id].append(format)
2491
2492 # Make sure all formats have unique format_id
2493 common_exts = set(itertools.chain(*self._format_selection_exts.values()))
2494 for format_id, ambiguous_formats in formats_dict.items():
2495 ambigious_id = len(ambiguous_formats) > 1
2496 for i, format in enumerate(ambiguous_formats):
2497 if ambigious_id:
2498 format['format_id'] = '%s-%d' % (format_id, i)
2499 if format.get('ext') is None:
2500 format['ext'] = determine_ext(format['url']).lower()
2501 # Ensure there is no conflict between id and ext in format selection
2502 # See https://github.com/yt-dlp/yt-dlp/issues/1282
2503 if format['format_id'] != format['ext'] and format['format_id'] in common_exts:
2504 format['format_id'] = 'f%s' % format['format_id']
2505
2506 for i, format in enumerate(formats):
2507 if format.get('format') is None:
2508 format['format'] = '{id} - {res}{note}'.format(
2509 id=format['format_id'],
2510 res=self.format_resolution(format),
2511 note=format_field(format, 'format_note', ' (%s)'),
2512 )
2513 if format.get('protocol') is None:
2514 format['protocol'] = determine_protocol(format)
2515 if format.get('resolution') is None:
2516 format['resolution'] = self.format_resolution(format, default=None)
2517 if format.get('dynamic_range') is None and format.get('vcodec') != 'none':
2518 format['dynamic_range'] = 'SDR'
2519 if (info_dict.get('duration') and format.get('tbr')
2520 and not format.get('filesize') and not format.get('filesize_approx')):
2521 format['filesize_approx'] = info_dict['duration'] * format['tbr'] * (1024 / 8)
2522
2523 # Add HTTP headers, so that external programs can use them from the
2524 # json output
2525 full_format_info = info_dict.copy()
2526 full_format_info.update(format)
2527 format['http_headers'] = self._calc_headers(full_format_info)
2528 # Remove private housekeeping stuff
2529 if '__x_forwarded_for_ip' in info_dict:
2530 del info_dict['__x_forwarded_for_ip']
2531
2532 if self.params.get('check_formats') is True:
2533 formats = LazyList(self._check_formats(formats[::-1]), reverse=True)
2534
2535 if not formats or formats[0] is not info_dict:
2536 # only set the 'formats' fields if the original info_dict list them
2537 # otherwise we end up with a circular reference, the first (and unique)
2538 # element in the 'formats' field in info_dict is info_dict itself,
2539 # which can't be exported to json
2540 info_dict['formats'] = formats
2541
2542 info_dict, _ = self.pre_process(info_dict)
2543
2544 if self._match_entry(info_dict) is not None:
2545 return info_dict
2546
2547 self.post_extract(info_dict)
2548 info_dict, _ = self.pre_process(info_dict, 'after_filter')
2549
2550 # The pre-processors may have modified the formats
2551 formats = info_dict.get('formats', [info_dict])
2552
2553 list_only = self.params.get('simulate') is None and (
2554 self.params.get('list_thumbnails') or self.params.get('listformats') or self.params.get('listsubtitles'))
2555 interactive_format_selection = not list_only and self.format_selector == '-'
2556 if self.params.get('list_thumbnails'):
2557 self.list_thumbnails(info_dict)
2558 if self.params.get('listsubtitles'):
2559 if 'automatic_captions' in info_dict:
2560 self.list_subtitles(
2561 info_dict['id'], automatic_captions, 'automatic captions')
2562 self.list_subtitles(info_dict['id'], subtitles, 'subtitles')
2563 if self.params.get('listformats') or interactive_format_selection:
2564 self.list_formats(info_dict)
2565 if list_only:
2566 # Without this printing, -F --print-json will not work
2567 self.__forced_printings(info_dict, self.prepare_filename(info_dict), incomplete=True)
2568 return
2569
2570 format_selector = self.format_selector
2571 if format_selector is None:
2572 req_format = self._default_format_spec(info_dict, download=download)
2573 self.write_debug('Default format spec: %s' % req_format)
2574 format_selector = self.build_format_selector(req_format)
2575
2576 while True:
2577 if interactive_format_selection:
2578 req_format = input(
2579 self._format_screen('\nEnter format selector: ', self.Styles.EMPHASIS))
2580 try:
2581 format_selector = self.build_format_selector(req_format)
2582 except SyntaxError as err:
2583 self.report_error(err, tb=False, is_error=False)
2584 continue
2585
2586 # While in format selection we may need to have an access to the original
2587 # format set in order to calculate some metrics or do some processing.
2588 # For now we need to be able to guess whether original formats provided
2589 # by extractor are incomplete or not (i.e. whether extractor provides only
2590 # video-only or audio-only formats) for proper formats selection for
2591 # extractors with such incomplete formats (see
2592 # https://github.com/ytdl-org/youtube-dl/pull/5556).
2593 # Since formats may be filtered during format selection and may not match
2594 # the original formats the results may be incorrect. Thus original formats
2595 # or pre-calculated metrics should be passed to format selection routines
2596 # as well.
2597 # We will pass a context object containing all necessary additional data
2598 # instead of just formats.
2599 # This fixes incorrect format selection issue (see
2600 # https://github.com/ytdl-org/youtube-dl/issues/10083).
2601 incomplete_formats = (
2602 # All formats are video-only or
2603 all(f.get('vcodec') != 'none' and f.get('acodec') == 'none' for f in formats)
2604 # all formats are audio-only
2605 or all(f.get('vcodec') == 'none' and f.get('acodec') != 'none' for f in formats))
2606
2607 ctx = {
2608 'formats': formats,
2609 'incomplete_formats': incomplete_formats,
2610 }
2611
2612 formats_to_download = list(format_selector(ctx))
2613 if interactive_format_selection and not formats_to_download:
2614 self.report_error('Requested format is not available', tb=False, is_error=False)
2615 continue
2616 break
2617
2618 if not formats_to_download:
2619 if not self.params.get('ignore_no_formats_error'):
2620 raise ExtractorError('Requested format is not available', expected=True,
2621 video_id=info_dict['id'], ie=info_dict['extractor'])
2622 self.report_warning('Requested format is not available')
2623 # Process what we can, even without any available formats.
2624 formats_to_download = [{}]
2625
2626 best_format = formats_to_download[-1]
2627 if download:
2628 if best_format:
2629 self.to_screen(
2630 f'[info] {info_dict["id"]}: Downloading {len(formats_to_download)} format(s): '
2631 + ', '.join([f['format_id'] for f in formats_to_download]))
2632 max_downloads_reached = False
2633 for i, fmt in enumerate(formats_to_download):
2634 formats_to_download[i] = new_info = self._copy_infodict(info_dict)
2635 new_info.update(fmt)
2636 try:
2637 self.process_info(new_info)
2638 except MaxDownloadsReached:
2639 max_downloads_reached = True
2640 # Remove copied info
2641 for key, val in tuple(new_info.items()):
2642 if info_dict.get(key) == val:
2643 new_info.pop(key)
2644 if max_downloads_reached:
2645 break
2646
2647 write_archive = set(f.get('__write_download_archive', False) for f in formats_to_download)
2648 assert write_archive.issubset({True, False, 'ignore'})
2649 if True in write_archive and False not in write_archive:
2650 self.record_download_archive(info_dict)
2651
2652 info_dict['requested_downloads'] = formats_to_download
2653 info_dict = self.run_all_pps('after_video', info_dict)
2654 if max_downloads_reached:
2655 raise MaxDownloadsReached()
2656
2657 # We update the info dict with the selected best quality format (backwards compatibility)
2658 info_dict.update(best_format)
2659 return info_dict
2660
2661 def process_subtitles(self, video_id, normal_subtitles, automatic_captions):
2662 """Select the requested subtitles and their format"""
2663 available_subs = {}
2664 if normal_subtitles and self.params.get('writesubtitles'):
2665 available_subs.update(normal_subtitles)
2666 if automatic_captions and self.params.get('writeautomaticsub'):
2667 for lang, cap_info in automatic_captions.items():
2668 if lang not in available_subs:
2669 available_subs[lang] = cap_info
2670
2671 if (not self.params.get('writesubtitles') and not
2672 self.params.get('writeautomaticsub') or not
2673 available_subs):
2674 return None
2675
2676 all_sub_langs = available_subs.keys()
2677 if self.params.get('allsubtitles', False):
2678 requested_langs = all_sub_langs
2679 elif self.params.get('subtitleslangs', False):
2680 # A list is used so that the order of languages will be the same as
2681 # given in subtitleslangs. See https://github.com/yt-dlp/yt-dlp/issues/1041
2682 requested_langs = []
2683 for lang_re in self.params.get('subtitleslangs'):
2684 discard = lang_re[0] == '-'
2685 if discard:
2686 lang_re = lang_re[1:]
2687 if lang_re == 'all':
2688 if discard:
2689 requested_langs = []
2690 else:
2691 requested_langs.extend(all_sub_langs)
2692 continue
2693 current_langs = filter(re.compile(lang_re + '$').match, all_sub_langs)
2694 if discard:
2695 for lang in current_langs:
2696 while lang in requested_langs:
2697 requested_langs.remove(lang)
2698 else:
2699 requested_langs.extend(current_langs)
2700 requested_langs = orderedSet(requested_langs)
2701 elif 'en' in available_subs:
2702 requested_langs = ['en']
2703 else:
2704 requested_langs = [list(all_sub_langs)[0]]
2705 if requested_langs:
2706 self.write_debug('Downloading subtitles: %s' % ', '.join(requested_langs))
2707
2708 formats_query = self.params.get('subtitlesformat', 'best')
2709 formats_preference = formats_query.split('/') if formats_query else []
2710 subs = {}
2711 for lang in requested_langs:
2712 formats = available_subs.get(lang)
2713 if formats is None:
2714 self.report_warning('%s subtitles not available for %s' % (lang, video_id))
2715 continue
2716 for ext in formats_preference:
2717 if ext == 'best':
2718 f = formats[-1]
2719 break
2720 matches = list(filter(lambda f: f['ext'] == ext, formats))
2721 if matches:
2722 f = matches[-1]
2723 break
2724 else:
2725 f = formats[-1]
2726 self.report_warning(
2727 'No subtitle format found matching "%s" for language %s, '
2728 'using %s' % (formats_query, lang, f['ext']))
2729 subs[lang] = f
2730 return subs
2731
2732 def _forceprint(self, key, info_dict):
2733 if info_dict is None:
2734 return
2735 info_copy = info_dict.copy()
2736 info_copy['formats_table'] = self.render_formats_table(info_dict)
2737 info_copy['thumbnails_table'] = self.render_thumbnails_table(info_dict)
2738 info_copy['subtitles_table'] = self.render_subtitles_table(info_dict.get('id'), info_dict.get('subtitles'))
2739 info_copy['automatic_captions_table'] = self.render_subtitles_table(info_dict.get('id'), info_dict.get('automatic_captions'))
2740
2741 def format_tmpl(tmpl):
2742 mobj = re.match(r'\w+(=?)$', tmpl)
2743 if mobj and mobj.group(1):
2744 return f'{tmpl[:-1]} = %({tmpl[:-1]})r'
2745 elif mobj:
2746 return f'%({tmpl})s'
2747 return tmpl
2748
2749 for tmpl in self.params['forceprint'].get(key, []):
2750 self.to_stdout(self.evaluate_outtmpl(format_tmpl(tmpl), info_copy))
2751
2752 for tmpl, file_tmpl in self.params['print_to_file'].get(key, []):
2753 filename = self.evaluate_outtmpl(file_tmpl, info_dict)
2754 tmpl = format_tmpl(tmpl)
2755 self.to_screen(f'[info] Writing {tmpl!r} to: {filename}')
2756 if self._ensure_dir_exists(filename):
2757 with io.open(filename, 'a', encoding='utf-8') as f:
2758 f.write(self.evaluate_outtmpl(tmpl, info_copy) + '\n')
2759
2760 def __forced_printings(self, info_dict, filename, incomplete):
2761 def print_mandatory(field, actual_field=None):
2762 if actual_field is None:
2763 actual_field = field
2764 if (self.params.get('force%s' % field, False)
2765 and (not incomplete or info_dict.get(actual_field) is not None)):
2766 self.to_stdout(info_dict[actual_field])
2767
2768 def print_optional(field):
2769 if (self.params.get('force%s' % field, False)
2770 and info_dict.get(field) is not None):
2771 self.to_stdout(info_dict[field])
2772
2773 info_dict = info_dict.copy()
2774 if filename is not None:
2775 info_dict['filename'] = filename
2776 if info_dict.get('requested_formats') is not None:
2777 # For RTMP URLs, also include the playpath
2778 info_dict['urls'] = '\n'.join(f['url'] + f.get('play_path', '') for f in info_dict['requested_formats'])
2779 elif info_dict.get('url'):
2780 info_dict['urls'] = info_dict['url'] + info_dict.get('play_path', '')
2781
2782 if (self.params.get('forcejson')
2783 or self.params['forceprint'].get('video')
2784 or self.params['print_to_file'].get('video')):
2785 self.post_extract(info_dict)
2786 self._forceprint('video', info_dict)
2787
2788 print_mandatory('title')
2789 print_mandatory('id')
2790 print_mandatory('url', 'urls')
2791 print_optional('thumbnail')
2792 print_optional('description')
2793 print_optional('filename')
2794 if self.params.get('forceduration') and info_dict.get('duration') is not None:
2795 self.to_stdout(formatSeconds(info_dict['duration']))
2796 print_mandatory('format')
2797
2798 if self.params.get('forcejson'):
2799 self.to_stdout(json.dumps(self.sanitize_info(info_dict)))
2800
2801 def dl(self, name, info, subtitle=False, test=False):
2802 if not info.get('url'):
2803 self.raise_no_formats(info, True)
2804
2805 if test:
2806 verbose = self.params.get('verbose')
2807 params = {
2808 'test': True,
2809 'quiet': self.params.get('quiet') or not verbose,
2810 'verbose': verbose,
2811 'noprogress': not verbose,
2812 'nopart': True,
2813 'skip_unavailable_fragments': False,
2814 'keep_fragments': False,
2815 'overwrites': True,
2816 '_no_ytdl_file': True,
2817 }
2818 else:
2819 params = self.params
2820 fd = get_suitable_downloader(info, params, to_stdout=(name == '-'))(self, params)
2821 if not test:
2822 for ph in self._progress_hooks:
2823 fd.add_progress_hook(ph)
2824 urls = '", "'.join(
2825 (f['url'].split(',')[0] + ',<data>' if f['url'].startswith('data:') else f['url'])
2826 for f in info.get('requested_formats', []) or [info])
2827 self.write_debug('Invoking downloader on "%s"' % urls)
2828
2829 # Note: Ideally info should be a deep-copied so that hooks cannot modify it.
2830 # But it may contain objects that are not deep-copyable
2831 new_info = self._copy_infodict(info)
2832 if new_info.get('http_headers') is None:
2833 new_info['http_headers'] = self._calc_headers(new_info)
2834 return fd.download(name, new_info, subtitle)
2835
2836 def existing_file(self, filepaths, *, default_overwrite=True):
2837 existing_files = list(filter(os.path.exists, orderedSet(filepaths)))
2838 if existing_files and not self.params.get('overwrites', default_overwrite):
2839 return existing_files[0]
2840
2841 for file in existing_files:
2842 self.report_file_delete(file)
2843 os.remove(file)
2844 return None
2845
2846 def process_info(self, info_dict):
2847 """Process a single resolved IE result. (Modifies it in-place)"""
2848
2849 assert info_dict.get('_type', 'video') == 'video'
2850 original_infodict = info_dict
2851
2852 if 'format' not in info_dict and 'ext' in info_dict:
2853 info_dict['format'] = info_dict['ext']
2854
2855 # This is mostly just for backward compatibility of process_info
2856 # As a side-effect, this allows for format-specific filters
2857 if self._match_entry(info_dict) is not None:
2858 info_dict['__write_download_archive'] = 'ignore'
2859 return
2860
2861 # Does nothing under normal operation - for backward compatibility of process_info
2862 self.post_extract(info_dict)
2863 self._num_downloads += 1
2864
2865 # info_dict['_filename'] needs to be set for backward compatibility
2866 info_dict['_filename'] = full_filename = self.prepare_filename(info_dict, warn=True)
2867 temp_filename = self.prepare_filename(info_dict, 'temp')
2868 files_to_move = {}
2869
2870 # Forced printings
2871 self.__forced_printings(info_dict, full_filename, incomplete=('format' not in info_dict))
2872
2873 if self.params.get('simulate'):
2874 info_dict['__write_download_archive'] = self.params.get('force_write_download_archive')
2875 return
2876
2877 if full_filename is None:
2878 return
2879 if not self._ensure_dir_exists(encodeFilename(full_filename)):
2880 return
2881 if not self._ensure_dir_exists(encodeFilename(temp_filename)):
2882 return
2883
2884 if self._write_description('video', info_dict,
2885 self.prepare_filename(info_dict, 'description')) is None:
2886 return
2887
2888 sub_files = self._write_subtitles(info_dict, temp_filename)
2889 if sub_files is None:
2890 return
2891 files_to_move.update(dict(sub_files))
2892
2893 thumb_files = self._write_thumbnails(
2894 'video', info_dict, temp_filename, self.prepare_filename(info_dict, 'thumbnail'))
2895 if thumb_files is None:
2896 return
2897 files_to_move.update(dict(thumb_files))
2898
2899 infofn = self.prepare_filename(info_dict, 'infojson')
2900 _infojson_written = self._write_info_json('video', info_dict, infofn)
2901 if _infojson_written:
2902 info_dict['infojson_filename'] = infofn
2903 # For backward compatibility, even though it was a private field
2904 info_dict['__infojson_filename'] = infofn
2905 elif _infojson_written is None:
2906 return
2907
2908 # Note: Annotations are deprecated
2909 annofn = None
2910 if self.params.get('writeannotations', False):
2911 annofn = self.prepare_filename(info_dict, 'annotation')
2912 if annofn:
2913 if not self._ensure_dir_exists(encodeFilename(annofn)):
2914 return
2915 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(annofn)):
2916 self.to_screen('[info] Video annotations are already present')
2917 elif not info_dict.get('annotations'):
2918 self.report_warning('There are no annotations to write.')
2919 else:
2920 try:
2921 self.to_screen('[info] Writing video annotations to: ' + annofn)
2922 with io.open(encodeFilename(annofn), 'w', encoding='utf-8') as annofile:
2923 annofile.write(info_dict['annotations'])
2924 except (KeyError, TypeError):
2925 self.report_warning('There are no annotations to write.')
2926 except (OSError, IOError):
2927 self.report_error('Cannot write annotations file: ' + annofn)
2928 return
2929
2930 # Write internet shortcut files
2931 def _write_link_file(link_type):
2932 url = try_get(info_dict['webpage_url'], iri_to_uri)
2933 if not url:
2934 self.report_warning(
2935 f'Cannot write internet shortcut file because the actual URL of "{info_dict["webpage_url"]}" is unknown')
2936 return True
2937 linkfn = replace_extension(self.prepare_filename(info_dict, 'link'), link_type, info_dict.get('ext'))
2938 if not self._ensure_dir_exists(encodeFilename(linkfn)):
2939 return False
2940 if self.params.get('overwrites', True) and os.path.exists(encodeFilename(linkfn)):
2941 self.to_screen(f'[info] Internet shortcut (.{link_type}) is already present')
2942 return True
2943 try:
2944 self.to_screen(f'[info] Writing internet shortcut (.{link_type}) to: {linkfn}')
2945 with io.open(encodeFilename(to_high_limit_path(linkfn)), 'w', encoding='utf-8',
2946 newline='\r\n' if link_type == 'url' else '\n') as linkfile:
2947 template_vars = {'url': url}
2948 if link_type == 'desktop':
2949 template_vars['filename'] = linkfn[:-(len(link_type) + 1)]
2950 linkfile.write(LINK_TEMPLATES[link_type] % template_vars)
2951 except (OSError, IOError):
2952 self.report_error(f'Cannot write internet shortcut {linkfn}')
2953 return False
2954 return True
2955
2956 write_links = {
2957 'url': self.params.get('writeurllink'),
2958 'webloc': self.params.get('writewebloclink'),
2959 'desktop': self.params.get('writedesktoplink'),
2960 }
2961 if self.params.get('writelink'):
2962 link_type = ('webloc' if sys.platform == 'darwin'
2963 else 'desktop' if sys.platform.startswith('linux')
2964 else 'url')
2965 write_links[link_type] = True
2966
2967 if any(should_write and not _write_link_file(link_type)
2968 for link_type, should_write in write_links.items()):
2969 return
2970
2971 def replace_info_dict(new_info):
2972 nonlocal info_dict
2973 if new_info == info_dict:
2974 return
2975 info_dict.clear()
2976 info_dict.update(new_info)
2977
2978 try:
2979 new_info, files_to_move = self.pre_process(info_dict, 'before_dl', files_to_move)
2980 replace_info_dict(new_info)
2981 except PostProcessingError as err:
2982 self.report_error('Preprocessing: %s' % str(err))
2983 return
2984
2985 if self.params.get('skip_download'):
2986 info_dict['filepath'] = temp_filename
2987 info_dict['__finaldir'] = os.path.dirname(os.path.abspath(encodeFilename(full_filename)))
2988 info_dict['__files_to_move'] = files_to_move
2989 replace_info_dict(self.run_pp(MoveFilesAfterDownloadPP(self, False), info_dict))
2990 info_dict['__write_download_archive'] = self.params.get('force_write_download_archive')
2991 else:
2992 # Download
2993 info_dict.setdefault('__postprocessors', [])
2994 try:
2995
2996 def existing_video_file(*filepaths):
2997 ext = info_dict.get('ext')
2998 converted = lambda file: replace_extension(file, self.params.get('final_ext') or ext, ext)
2999 file = self.existing_file(itertools.chain(*zip(map(converted, filepaths), filepaths)),
3000 default_overwrite=False)
3001 if file:
3002 info_dict['ext'] = os.path.splitext(file)[1][1:]
3003 return file
3004
3005 success = True
3006 if info_dict.get('requested_formats') is not None:
3007
3008 def compatible_formats(formats):
3009 # TODO: some formats actually allow this (mkv, webm, ogg, mp4), but not all of them.
3010 video_formats = [format for format in formats if format.get('vcodec') != 'none']
3011 audio_formats = [format for format in formats if format.get('acodec') != 'none']
3012 if len(video_formats) > 2 or len(audio_formats) > 2:
3013 return False
3014
3015 # Check extension
3016 exts = set(format.get('ext') for format in formats)
3017 COMPATIBLE_EXTS = (
3018 set(('mp3', 'mp4', 'm4a', 'm4p', 'm4b', 'm4r', 'm4v', 'ismv', 'isma')),
3019 set(('webm',)),
3020 )
3021 for ext_sets in COMPATIBLE_EXTS:
3022 if ext_sets.issuperset(exts):
3023 return True
3024 # TODO: Check acodec/vcodec
3025 return False
3026
3027 requested_formats = info_dict['requested_formats']
3028 old_ext = info_dict['ext']
3029 if self.params.get('merge_output_format') is None:
3030 if not compatible_formats(requested_formats):
3031 info_dict['ext'] = 'mkv'
3032 self.report_warning(
3033 'Requested formats are incompatible for merge and will be merged into mkv')
3034 if (info_dict['ext'] == 'webm'
3035 and info_dict.get('thumbnails')
3036 # check with type instead of pp_key, __name__, or isinstance
3037 # since we dont want any custom PPs to trigger this
3038 and any(type(pp) == EmbedThumbnailPP for pp in self._pps['post_process'])):
3039 info_dict['ext'] = 'mkv'
3040 self.report_warning(
3041 'webm doesn\'t support embedding a thumbnail, mkv will be used')
3042 new_ext = info_dict['ext']
3043
3044 def correct_ext(filename, ext=new_ext):
3045 if filename == '-':
3046 return filename
3047 filename_real_ext = os.path.splitext(filename)[1][1:]
3048 filename_wo_ext = (
3049 os.path.splitext(filename)[0]
3050 if filename_real_ext in (old_ext, new_ext)
3051 else filename)
3052 return '%s.%s' % (filename_wo_ext, ext)
3053
3054 # Ensure filename always has a correct extension for successful merge
3055 full_filename = correct_ext(full_filename)
3056 temp_filename = correct_ext(temp_filename)
3057 dl_filename = existing_video_file(full_filename, temp_filename)
3058 info_dict['__real_download'] = False
3059
3060 downloaded = []
3061 merger = FFmpegMergerPP(self)
3062
3063 fd = get_suitable_downloader(info_dict, self.params, to_stdout=temp_filename == '-')
3064 if dl_filename is not None:
3065 self.report_file_already_downloaded(dl_filename)
3066 elif fd:
3067 for f in requested_formats if fd != FFmpegFD else []:
3068 f['filepath'] = fname = prepend_extension(
3069 correct_ext(temp_filename, info_dict['ext']),
3070 'f%s' % f['format_id'], info_dict['ext'])
3071 downloaded.append(fname)
3072 info_dict['url'] = '\n'.join(f['url'] for f in requested_formats)
3073 success, real_download = self.dl(temp_filename, info_dict)
3074 info_dict['__real_download'] = real_download
3075 else:
3076 if self.params.get('allow_unplayable_formats'):
3077 self.report_warning(
3078 'You have requested merging of multiple formats '
3079 'while also allowing unplayable formats to be downloaded. '
3080 'The formats won\'t be merged to prevent data corruption.')
3081 elif not merger.available:
3082 msg = 'You have requested merging of multiple formats but ffmpeg is not installed'
3083 if not self.params.get('ignoreerrors'):
3084 self.report_error(f'{msg}. Aborting due to --abort-on-error')
3085 return
3086 self.report_warning(f'{msg}. The formats won\'t be merged')
3087
3088 if temp_filename == '-':
3089 reason = ('using a downloader other than ffmpeg' if FFmpegFD.can_merge_formats(info_dict, self.params)
3090 else 'but the formats are incompatible for simultaneous download' if merger.available
3091 else 'but ffmpeg is not installed')
3092 self.report_warning(
3093 f'You have requested downloading multiple formats to stdout {reason}. '
3094 'The formats will be streamed one after the other')
3095 fname = temp_filename
3096 for f in requested_formats:
3097 new_info = dict(info_dict)
3098 del new_info['requested_formats']
3099 new_info.update(f)
3100 if temp_filename != '-':
3101 fname = prepend_extension(
3102 correct_ext(temp_filename, new_info['ext']),
3103 'f%s' % f['format_id'], new_info['ext'])
3104 if not self._ensure_dir_exists(fname):
3105 return
3106 f['filepath'] = fname
3107 downloaded.append(fname)
3108 partial_success, real_download = self.dl(fname, new_info)
3109 info_dict['__real_download'] = info_dict['__real_download'] or real_download
3110 success = success and partial_success
3111
3112 if downloaded and merger.available and not self.params.get('allow_unplayable_formats'):
3113 info_dict['__postprocessors'].append(merger)
3114 info_dict['__files_to_merge'] = downloaded
3115 # Even if there were no downloads, it is being merged only now
3116 info_dict['__real_download'] = True
3117 else:
3118 for file in downloaded:
3119 files_to_move[file] = None
3120 else:
3121 # Just a single file
3122 dl_filename = existing_video_file(full_filename, temp_filename)
3123 if dl_filename is None or dl_filename == temp_filename:
3124 # dl_filename == temp_filename could mean that the file was partially downloaded with --no-part.
3125 # So we should try to resume the download
3126 success, real_download = self.dl(temp_filename, info_dict)
3127 info_dict['__real_download'] = real_download
3128 else:
3129 self.report_file_already_downloaded(dl_filename)
3130
3131 dl_filename = dl_filename or temp_filename
3132 info_dict['__finaldir'] = os.path.dirname(os.path.abspath(encodeFilename(full_filename)))
3133
3134 except network_exceptions as err:
3135 self.report_error('unable to download video data: %s' % error_to_compat_str(err))
3136 return
3137 except (OSError, IOError) as err:
3138 raise UnavailableVideoError(err)
3139 except (ContentTooShortError, ) as err:
3140 self.report_error('content too short (expected %s bytes and served %s)' % (err.expected, err.downloaded))
3141 return
3142
3143 if success and full_filename != '-':
3144
3145 def fixup():
3146 do_fixup = True
3147 fixup_policy = self.params.get('fixup')
3148 vid = info_dict['id']
3149
3150 if fixup_policy in ('ignore', 'never'):
3151 return
3152 elif fixup_policy == 'warn':
3153 do_fixup = False
3154 elif fixup_policy != 'force':
3155 assert fixup_policy in ('detect_or_warn', None)
3156 if not info_dict.get('__real_download'):
3157 do_fixup = False
3158
3159 def ffmpeg_fixup(cndn, msg, cls):
3160 if not cndn:
3161 return
3162 if not do_fixup:
3163 self.report_warning(f'{vid}: {msg}')
3164 return
3165 pp = cls(self)
3166 if pp.available:
3167 info_dict['__postprocessors'].append(pp)
3168 else:
3169 self.report_warning(f'{vid}: {msg}. Install ffmpeg to fix this automatically')
3170
3171 stretched_ratio = info_dict.get('stretched_ratio')
3172 ffmpeg_fixup(
3173 stretched_ratio not in (1, None),
3174 f'Non-uniform pixel ratio {stretched_ratio}',
3175 FFmpegFixupStretchedPP)
3176
3177 ffmpeg_fixup(
3178 (info_dict.get('requested_formats') is None
3179 and info_dict.get('container') == 'm4a_dash'
3180 and info_dict.get('ext') == 'm4a'),
3181 'writing DASH m4a. Only some players support this container',
3182 FFmpegFixupM4aPP)
3183
3184 downloader = get_suitable_downloader(info_dict, self.params) if 'protocol' in info_dict else None
3185 downloader = downloader.__name__ if downloader else None
3186
3187 if info_dict.get('requested_formats') is None: # Not necessary if doing merger
3188 ffmpeg_fixup(downloader == 'HlsFD',
3189 'Possible MPEG-TS in MP4 container or malformed AAC timestamps',
3190 FFmpegFixupM3u8PP)
3191 ffmpeg_fixup(info_dict.get('is_live') and downloader == 'DashSegmentsFD',
3192 'Possible duplicate MOOV atoms', FFmpegFixupDuplicateMoovPP)
3193
3194 ffmpeg_fixup(downloader == 'WebSocketFragmentFD', 'Malformed timestamps detected', FFmpegFixupTimestampPP)
3195 ffmpeg_fixup(downloader == 'WebSocketFragmentFD', 'Malformed duration detected', FFmpegFixupDurationPP)
3196
3197 fixup()
3198 try:
3199 replace_info_dict(self.post_process(dl_filename, info_dict, files_to_move))
3200 except PostProcessingError as err:
3201 self.report_error('Postprocessing: %s' % str(err))
3202 return
3203 try:
3204 for ph in self._post_hooks:
3205 ph(info_dict['filepath'])
3206 except Exception as err:
3207 self.report_error('post hooks: %s' % str(err))
3208 return
3209 info_dict['__write_download_archive'] = True
3210
3211 if self.params.get('force_write_download_archive'):
3212 info_dict['__write_download_archive'] = True
3213
3214 # Make sure the info_dict was modified in-place
3215 assert info_dict is original_infodict
3216
3217 max_downloads = self.params.get('max_downloads')
3218 if max_downloads is not None and self._num_downloads >= int(max_downloads):
3219 raise MaxDownloadsReached()
3220
3221 def __download_wrapper(self, func):
3222 @functools.wraps(func)
3223 def wrapper(*args, **kwargs):
3224 try:
3225 res = func(*args, **kwargs)
3226 except UnavailableVideoError as e:
3227 self.report_error(e)
3228 except MaxDownloadsReached as e:
3229 self.to_screen(f'[info] {e}')
3230 raise
3231 except DownloadCancelled as e:
3232 self.to_screen(f'[info] {e}')
3233 if not self.params.get('break_per_url'):
3234 raise
3235 else:
3236 if self.params.get('dump_single_json', False):
3237 self.post_extract(res)
3238 self.to_stdout(json.dumps(self.sanitize_info(res)))
3239 return wrapper
3240
3241 def download(self, url_list):
3242 """Download a given list of URLs."""
3243 url_list = variadic(url_list) # Passing a single URL is a common mistake
3244 outtmpl = self.outtmpl_dict['default']
3245 if (len(url_list) > 1
3246 and outtmpl != '-'
3247 and '%' not in outtmpl
3248 and self.params.get('max_downloads') != 1):
3249 raise SameFileError(outtmpl)
3250
3251 for url in url_list:
3252 self.__download_wrapper(self.extract_info)(
3253 url, force_generic_extractor=self.params.get('force_generic_extractor', False))
3254
3255 return self._download_retcode
3256
3257 def download_with_info_file(self, info_filename):
3258 with contextlib.closing(fileinput.FileInput(
3259 [info_filename], mode='r',
3260 openhook=fileinput.hook_encoded('utf-8'))) as f:
3261 # FileInput doesn't have a read method, we can't call json.load
3262 info = self.sanitize_info(json.loads('\n'.join(f)), self.params.get('clean_infojson', True))
3263 try:
3264 self.__download_wrapper(self.process_ie_result)(info, download=True)
3265 except (DownloadError, EntryNotInPlaylist, ReExtractInfo) as e:
3266 if not isinstance(e, EntryNotInPlaylist):
3267 self.to_stderr('\r')
3268 webpage_url = info.get('webpage_url')
3269 if webpage_url is not None:
3270 self.report_warning(f'The info failed to download: {e}; trying with URL {webpage_url}')
3271 return self.download([webpage_url])
3272 else:
3273 raise
3274 return self._download_retcode
3275
3276 @staticmethod
3277 def sanitize_info(info_dict, remove_private_keys=False):
3278 ''' Sanitize the infodict for converting to json '''
3279 if info_dict is None:
3280 return info_dict
3281 info_dict.setdefault('epoch', int(time.time()))
3282 info_dict.setdefault('_type', 'video')
3283
3284 if remove_private_keys:
3285 reject = lambda k, v: v is None or (k.startswith('_') and k != '_type') or k in {
3286 'requested_downloads', 'requested_formats', 'requested_subtitles', 'requested_entries',
3287 'entries', 'filepath', 'infojson_filename', 'original_url', 'playlist_autonumber',
3288 }
3289 else:
3290 reject = lambda k, v: False
3291
3292 def filter_fn(obj):
3293 if isinstance(obj, dict):
3294 return {k: filter_fn(v) for k, v in obj.items() if not reject(k, v)}
3295 elif isinstance(obj, (list, tuple, set, LazyList)):
3296 return list(map(filter_fn, obj))
3297 elif obj is None or isinstance(obj, (str, int, float, bool)):
3298 return obj
3299 else:
3300 return repr(obj)
3301
3302 return filter_fn(info_dict)
3303
3304 @staticmethod
3305 def filter_requested_info(info_dict, actually_filter=True):
3306 ''' Alias of sanitize_info for backward compatibility '''
3307 return YoutubeDL.sanitize_info(info_dict, actually_filter)
3308
3309 @staticmethod
3310 def post_extract(info_dict):
3311 def actual_post_extract(info_dict):
3312 if info_dict.get('_type') in ('playlist', 'multi_video'):
3313 for video_dict in info_dict.get('entries', {}):
3314 actual_post_extract(video_dict or {})
3315 return
3316
3317 post_extractor = info_dict.pop('__post_extractor', None) or (lambda: {})
3318 info_dict.update(post_extractor())
3319
3320 actual_post_extract(info_dict or {})
3321
3322 def run_pp(self, pp, infodict):
3323 files_to_delete = []
3324 if '__files_to_move' not in infodict:
3325 infodict['__files_to_move'] = {}
3326 try:
3327 files_to_delete, infodict = pp.run(infodict)
3328 except PostProcessingError as e:
3329 # Must be True and not 'only_download'
3330 if self.params.get('ignoreerrors') is True:
3331 self.report_error(e)
3332 return infodict
3333 raise
3334
3335 if not files_to_delete:
3336 return infodict
3337 if self.params.get('keepvideo', False):
3338 for f in files_to_delete:
3339 infodict['__files_to_move'].setdefault(f, '')
3340 else:
3341 for old_filename in set(files_to_delete):
3342 self.to_screen('Deleting original file %s (pass -k to keep)' % old_filename)
3343 try:
3344 os.remove(encodeFilename(old_filename))
3345 except (IOError, OSError):
3346 self.report_warning('Unable to remove downloaded original file')
3347 if old_filename in infodict['__files_to_move']:
3348 del infodict['__files_to_move'][old_filename]
3349 return infodict
3350
3351 def run_all_pps(self, key, info, *, additional_pps=None):
3352 self._forceprint(key, info)
3353 for pp in (additional_pps or []) + self._pps[key]:
3354 info = self.run_pp(pp, info)
3355 return info
3356
3357 def pre_process(self, ie_info, key='pre_process', files_to_move=None):
3358 info = dict(ie_info)
3359 info['__files_to_move'] = files_to_move or {}
3360 info = self.run_all_pps(key, info)
3361 return info, info.pop('__files_to_move', None)
3362
3363 def post_process(self, filename, info, files_to_move=None):
3364 """Run all the postprocessors on the given file."""
3365 info['filepath'] = filename
3366 info['__files_to_move'] = files_to_move or {}
3367 info = self.run_all_pps('post_process', info, additional_pps=info.get('__postprocessors'))
3368 info = self.run_pp(MoveFilesAfterDownloadPP(self), info)
3369 del info['__files_to_move']
3370 return self.run_all_pps('after_move', info)
3371
3372 def _make_archive_id(self, info_dict):
3373 video_id = info_dict.get('id')
3374 if not video_id:
3375 return
3376 # Future-proof against any change in case
3377 # and backwards compatibility with prior versions
3378 extractor = info_dict.get('extractor_key') or info_dict.get('ie_key') # key in a playlist
3379 if extractor is None:
3380 url = str_or_none(info_dict.get('url'))
3381 if not url:
3382 return
3383 # Try to find matching extractor for the URL and take its ie_key
3384 for ie_key, ie in self._ies.items():
3385 if ie.suitable(url):
3386 extractor = ie_key
3387 break
3388 else:
3389 return
3390 return '%s %s' % (extractor.lower(), video_id)
3391
3392 def in_download_archive(self, info_dict):
3393 fn = self.params.get('download_archive')
3394 if fn is None:
3395 return False
3396
3397 vid_id = self._make_archive_id(info_dict)
3398 if not vid_id:
3399 return False # Incomplete video information
3400
3401 return vid_id in self.archive
3402
3403 def record_download_archive(self, info_dict):
3404 fn = self.params.get('download_archive')
3405 if fn is None:
3406 return
3407 vid_id = self._make_archive_id(info_dict)
3408 assert vid_id
3409 self.write_debug(f'Adding to archive: {vid_id}')
3410 with locked_file(fn, 'a', encoding='utf-8') as archive_file:
3411 archive_file.write(vid_id + '\n')
3412 self.archive.add(vid_id)
3413
3414 @staticmethod
3415 def format_resolution(format, default='unknown'):
3416 if format.get('vcodec') == 'none' and format.get('acodec') != 'none':
3417 return 'audio only'
3418 if format.get('resolution') is not None:
3419 return format['resolution']
3420 if format.get('width') and format.get('height'):
3421 return '%dx%d' % (format['width'], format['height'])
3422 elif format.get('height'):
3423 return '%sp' % format['height']
3424 elif format.get('width'):
3425 return '%dx?' % format['width']
3426 return default
3427
3428 def _list_format_headers(self, *headers):
3429 if self.params.get('listformats_table', True) is not False:
3430 return [self._format_screen(header, self.Styles.HEADERS) for header in headers]
3431 return headers
3432
3433 def _format_note(self, fdict):
3434 res = ''
3435 if fdict.get('ext') in ['f4f', 'f4m']:
3436 res += '(unsupported)'
3437 if fdict.get('language'):
3438 if res:
3439 res += ' '
3440 res += '[%s]' % fdict['language']
3441 if fdict.get('format_note') is not None:
3442 if res:
3443 res += ' '
3444 res += fdict['format_note']
3445 if fdict.get('tbr') is not None:
3446 if res:
3447 res += ', '
3448 res += '%4dk' % fdict['tbr']
3449 if fdict.get('container') is not None:
3450 if res:
3451 res += ', '
3452 res += '%s container' % fdict['container']
3453 if (fdict.get('vcodec') is not None
3454 and fdict.get('vcodec') != 'none'):
3455 if res:
3456 res += ', '
3457 res += fdict['vcodec']
3458 if fdict.get('vbr') is not None:
3459 res += '@'
3460 elif fdict.get('vbr') is not None and fdict.get('abr') is not None:
3461 res += 'video@'
3462 if fdict.get('vbr') is not None:
3463 res += '%4dk' % fdict['vbr']
3464 if fdict.get('fps') is not None:
3465 if res:
3466 res += ', '
3467 res += '%sfps' % fdict['fps']
3468 if fdict.get('acodec') is not None:
3469 if res:
3470 res += ', '
3471 if fdict['acodec'] == 'none':
3472 res += 'video only'
3473 else:
3474 res += '%-5s' % fdict['acodec']
3475 elif fdict.get('abr') is not None:
3476 if res:
3477 res += ', '
3478 res += 'audio'
3479 if fdict.get('abr') is not None:
3480 res += '@%3dk' % fdict['abr']
3481 if fdict.get('asr') is not None:
3482 res += ' (%5dHz)' % fdict['asr']
3483 if fdict.get('filesize') is not None:
3484 if res:
3485 res += ', '
3486 res += format_bytes(fdict['filesize'])
3487 elif fdict.get('filesize_approx') is not None:
3488 if res:
3489 res += ', '
3490 res += '~' + format_bytes(fdict['filesize_approx'])
3491 return res
3492
3493 def render_formats_table(self, info_dict):
3494 if not info_dict.get('formats') and not info_dict.get('url'):
3495 return None
3496
3497 formats = info_dict.get('formats', [info_dict])
3498 if not self.params.get('listformats_table', True) is not False:
3499 table = [
3500 [
3501 format_field(f, 'format_id'),
3502 format_field(f, 'ext'),
3503 self.format_resolution(f),
3504 self._format_note(f)
3505 ] for f in formats if f.get('preference') is None or f['preference'] >= -1000]
3506 return render_table(['format code', 'extension', 'resolution', 'note'], table, extra_gap=1)
3507
3508 delim = self._format_screen('\u2502', self.Styles.DELIM, '|', test_encoding=True)
3509 table = [
3510 [
3511 self._format_screen(format_field(f, 'format_id'), self.Styles.ID),
3512 format_field(f, 'ext'),
3513 format_field(f, func=self.format_resolution, ignore=('audio only', 'images')),
3514 format_field(f, 'fps', '\t%d'),
3515 format_field(f, 'dynamic_range', '%s', ignore=(None, 'SDR')).replace('HDR', ''),
3516 delim,
3517 format_field(f, 'filesize', ' \t%s', func=format_bytes) + format_field(f, 'filesize_approx', '~\t%s', func=format_bytes),
3518 format_field(f, 'tbr', '\t%dk'),
3519 shorten_protocol_name(f.get('protocol', '')),
3520 delim,
3521 format_field(f, 'vcodec', default='unknown').replace(
3522 'none', 'images' if f.get('acodec') == 'none'
3523 else self._format_screen('audio only', self.Styles.SUPPRESS)),
3524 format_field(f, 'vbr', '\t%dk'),
3525 format_field(f, 'acodec', default='unknown').replace(
3526 'none', '' if f.get('vcodec') == 'none'
3527 else self._format_screen('video only', self.Styles.SUPPRESS)),
3528 format_field(f, 'abr', '\t%dk'),
3529 format_field(f, 'asr', '\t%dHz'),
3530 join_nonempty(
3531 self._format_screen('UNSUPPORTED', 'light red') if f.get('ext') in ('f4f', 'f4m') else None,
3532 format_field(f, 'language', '[%s]'),
3533 join_nonempty(format_field(f, 'format_note'),
3534 format_field(f, 'container', ignore=(None, f.get('ext'))),
3535 delim=', '),
3536 delim=' '),
3537 ] for f in formats if f.get('preference') is None or f['preference'] >= -1000]
3538 header_line = self._list_format_headers(
3539 'ID', 'EXT', 'RESOLUTION', '\tFPS', 'HDR', delim, '\tFILESIZE', '\tTBR', 'PROTO',
3540 delim, 'VCODEC', '\tVBR', 'ACODEC', '\tABR', '\tASR', 'MORE INFO')
3541
3542 return render_table(
3543 header_line, table, hide_empty=True,
3544 delim=self._format_screen('\u2500', self.Styles.DELIM, '-', test_encoding=True))
3545
3546 def render_thumbnails_table(self, info_dict):
3547 thumbnails = list(info_dict.get('thumbnails') or [])
3548 if not thumbnails:
3549 return None
3550 return render_table(
3551 self._list_format_headers('ID', 'Width', 'Height', 'URL'),
3552 [[t.get('id'), t.get('width', 'unknown'), t.get('height', 'unknown'), t['url']] for t in thumbnails])
3553
3554 def render_subtitles_table(self, video_id, subtitles):
3555 def _row(lang, formats):
3556 exts, names = zip(*((f['ext'], f.get('name') or 'unknown') for f in reversed(formats)))
3557 if len(set(names)) == 1:
3558 names = [] if names[0] == 'unknown' else names[:1]
3559 return [lang, ', '.join(names), ', '.join(exts)]
3560
3561 if not subtitles:
3562 return None
3563 return render_table(
3564 self._list_format_headers('Language', 'Name', 'Formats'),
3565 [_row(lang, formats) for lang, formats in subtitles.items()],
3566 hide_empty=True)
3567
3568 def __list_table(self, video_id, name, func, *args):
3569 table = func(*args)
3570 if not table:
3571 self.to_screen(f'{video_id} has no {name}')
3572 return
3573 self.to_screen(f'[info] Available {name} for {video_id}:')
3574 self.to_stdout(table)
3575
3576 def list_formats(self, info_dict):
3577 self.__list_table(info_dict['id'], 'formats', self.render_formats_table, info_dict)
3578
3579 def list_thumbnails(self, info_dict):
3580 self.__list_table(info_dict['id'], 'thumbnails', self.render_thumbnails_table, info_dict)
3581
3582 def list_subtitles(self, video_id, subtitles, name='subtitles'):
3583 self.__list_table(video_id, name, self.render_subtitles_table, video_id, subtitles)
3584
3585 def urlopen(self, req):
3586 """ Start an HTTP download """
3587 if isinstance(req, compat_basestring):
3588 req = sanitized_Request(req)
3589 return self._opener.open(req, timeout=self._socket_timeout)
3590
3591 def print_debug_header(self):
3592 if not self.params.get('verbose'):
3593 return
3594
3595 def get_encoding(stream):
3596 ret = str(getattr(stream, 'encoding', 'missing (%s)' % type(stream).__name__))
3597 if not supports_terminal_sequences(stream):
3598 from .compat import WINDOWS_VT_MODE
3599 ret += ' (No VT)' if WINDOWS_VT_MODE is False else ' (No ANSI)'
3600 return ret
3601
3602 encoding_str = 'Encodings: locale %s, fs %s, out %s, err %s, pref %s' % (
3603 locale.getpreferredencoding(),
3604 sys.getfilesystemencoding(),
3605 get_encoding(self._out_files['screen']), get_encoding(self._out_files['error']),
3606 self.get_encoding())
3607
3608 logger = self.params.get('logger')
3609 if logger:
3610 write_debug = lambda msg: logger.debug(f'[debug] {msg}')
3611 write_debug(encoding_str)
3612 else:
3613 write_string(f'[debug] {encoding_str}\n', encoding=None)
3614 write_debug = lambda msg: self._write_string(f'[debug] {msg}\n')
3615
3616 source = detect_variant()
3617 write_debug(join_nonempty(
3618 'yt-dlp version', __version__,
3619 f'[{RELEASE_GIT_HEAD}]' if RELEASE_GIT_HEAD else '',
3620 '' if source == 'unknown' else f'({source})',
3621 delim=' '))
3622 if not _LAZY_LOADER:
3623 if os.environ.get('YTDLP_NO_LAZY_EXTRACTORS'):
3624 write_debug('Lazy loading extractors is forcibly disabled')
3625 else:
3626 write_debug('Lazy loading extractors is disabled')
3627 if plugin_extractors or plugin_postprocessors:
3628 write_debug('Plugins: %s' % [
3629 '%s%s' % (klass.__name__, '' if klass.__name__ == name else f' as {name}')
3630 for name, klass in itertools.chain(plugin_extractors.items(), plugin_postprocessors.items())])
3631 if self.params.get('compat_opts'):
3632 write_debug('Compatibility options: %s' % ', '.join(self.params.get('compat_opts')))
3633
3634 if source == 'source':
3635 try:
3636 sp = Popen(
3637 ['git', 'rev-parse', '--short', 'HEAD'],
3638 stdout=subprocess.PIPE, stderr=subprocess.PIPE,
3639 cwd=os.path.dirname(os.path.abspath(__file__)))
3640 out, err = sp.communicate_or_kill()
3641 out = out.decode().strip()
3642 if re.match('[0-9a-f]+', out):
3643 write_debug('Git HEAD: %s' % out)
3644 except Exception:
3645 try:
3646 sys.exc_clear()
3647 except Exception:
3648 pass
3649
3650 def python_implementation():
3651 impl_name = platform.python_implementation()
3652 if impl_name == 'PyPy' and hasattr(sys, 'pypy_version_info'):
3653 return impl_name + ' version %d.%d.%d' % sys.pypy_version_info[:3]
3654 return impl_name
3655
3656 write_debug('Python version %s (%s %s) - %s' % (
3657 platform.python_version(),
3658 python_implementation(),
3659 platform.architecture()[0],
3660 platform_name()))
3661
3662 exe_versions, ffmpeg_features = FFmpegPostProcessor.get_versions_and_features(self)
3663 ffmpeg_features = {key for key, val in ffmpeg_features.items() if val}
3664 if ffmpeg_features:
3665 exe_versions['ffmpeg'] += ' (%s)' % ','.join(ffmpeg_features)
3666
3667 exe_versions['rtmpdump'] = rtmpdump_version()
3668 exe_versions['phantomjs'] = PhantomJSwrapper._version()
3669 exe_str = ', '.join(
3670 f'{exe} {v}' for exe, v in sorted(exe_versions.items()) if v
3671 ) or 'none'
3672 write_debug('exe versions: %s' % exe_str)
3673
3674 from .downloader.websocket import has_websockets
3675 from .postprocessor.embedthumbnail import has_mutagen
3676 from .cookies import SQLITE_AVAILABLE, SECRETSTORAGE_AVAILABLE
3677
3678 lib_str = join_nonempty(
3679 compat_brotli and compat_brotli.__name__,
3680 has_certifi and 'certifi',
3681 compat_pycrypto_AES and compat_pycrypto_AES.__name__.split('.')[0],
3682 SECRETSTORAGE_AVAILABLE and 'secretstorage',
3683 has_mutagen and 'mutagen',
3684 SQLITE_AVAILABLE and 'sqlite',
3685 has_websockets and 'websockets',
3686 delim=', ') or 'none'
3687 write_debug('Optional libraries: %s' % lib_str)
3688
3689 proxy_map = {}
3690 for handler in self._opener.handlers:
3691 if hasattr(handler, 'proxies'):
3692 proxy_map.update(handler.proxies)
3693 write_debug(f'Proxy map: {proxy_map}')
3694
3695 # Not implemented
3696 if False and self.params.get('call_home'):
3697 ipaddr = self.urlopen('https://yt-dl.org/ip').read().decode('utf-8')
3698 write_debug('Public IP address: %s' % ipaddr)
3699 latest_version = self.urlopen(
3700 'https://yt-dl.org/latest/version').read().decode('utf-8')
3701 if version_tuple(latest_version) > version_tuple(__version__):
3702 self.report_warning(
3703 'You are using an outdated version (newest version: %s)! '
3704 'See https://yt-dl.org/update if you need help updating.' %
3705 latest_version)
3706
3707 def _setup_opener(self):
3708 timeout_val = self.params.get('socket_timeout')
3709 self._socket_timeout = 20 if timeout_val is None else float(timeout_val)
3710
3711 opts_cookiesfrombrowser = self.params.get('cookiesfrombrowser')
3712 opts_cookiefile = self.params.get('cookiefile')
3713 opts_proxy = self.params.get('proxy')
3714
3715 self.cookiejar = load_cookies(opts_cookiefile, opts_cookiesfrombrowser, self)
3716
3717 cookie_processor = YoutubeDLCookieProcessor(self.cookiejar)
3718 if opts_proxy is not None:
3719 if opts_proxy == '':
3720 proxies = {}
3721 else:
3722 proxies = {'http': opts_proxy, 'https': opts_proxy}
3723 else:
3724 proxies = compat_urllib_request.getproxies()
3725 # Set HTTPS proxy to HTTP one if given (https://github.com/ytdl-org/youtube-dl/issues/805)
3726 if 'http' in proxies and 'https' not in proxies:
3727 proxies['https'] = proxies['http']
3728 proxy_handler = PerRequestProxyHandler(proxies)
3729
3730 debuglevel = 1 if self.params.get('debug_printtraffic') else 0
3731 https_handler = make_HTTPS_handler(self.params, debuglevel=debuglevel)
3732 ydlh = YoutubeDLHandler(self.params, debuglevel=debuglevel)
3733 redirect_handler = YoutubeDLRedirectHandler()
3734 data_handler = compat_urllib_request_DataHandler()
3735
3736 # When passing our own FileHandler instance, build_opener won't add the
3737 # default FileHandler and allows us to disable the file protocol, which
3738 # can be used for malicious purposes (see
3739 # https://github.com/ytdl-org/youtube-dl/issues/8227)
3740 file_handler = compat_urllib_request.FileHandler()
3741
3742 def file_open(*args, **kwargs):
3743 raise compat_urllib_error.URLError('file:// scheme is explicitly disabled in yt-dlp for security reasons')
3744 file_handler.file_open = file_open
3745
3746 opener = compat_urllib_request.build_opener(
3747 proxy_handler, https_handler, cookie_processor, ydlh, redirect_handler, data_handler, file_handler)
3748
3749 # Delete the default user-agent header, which would otherwise apply in
3750 # cases where our custom HTTP handler doesn't come into play
3751 # (See https://github.com/ytdl-org/youtube-dl/issues/1309 for details)
3752 opener.addheaders = []
3753 self._opener = opener
3754
3755 def encode(self, s):
3756 if isinstance(s, bytes):
3757 return s # Already encoded
3758
3759 try:
3760 return s.encode(self.get_encoding())
3761 except UnicodeEncodeError as err:
3762 err.reason = err.reason + '. Check your system encoding configuration or use the --encoding option.'
3763 raise
3764
3765 def get_encoding(self):
3766 encoding = self.params.get('encoding')
3767 if encoding is None:
3768 encoding = preferredencoding()
3769 return encoding
3770
3771 def _write_info_json(self, label, ie_result, infofn, overwrite=None):
3772 ''' Write infojson and returns True = written, False = skip, None = error '''
3773 if overwrite is None:
3774 overwrite = self.params.get('overwrites', True)
3775 if not self.params.get('writeinfojson'):
3776 return False
3777 elif not infofn:
3778 self.write_debug(f'Skipping writing {label} infojson')
3779 return False
3780 elif not self._ensure_dir_exists(infofn):
3781 return None
3782 elif not overwrite and os.path.exists(infofn):
3783 self.to_screen(f'[info] {label.title()} metadata is already present')
3784 else:
3785 self.to_screen(f'[info] Writing {label} metadata as JSON to: {infofn}')
3786 try:
3787 write_json_file(self.sanitize_info(ie_result, self.params.get('clean_infojson', True)), infofn)
3788 except (OSError, IOError):
3789 self.report_error(f'Cannot write {label} metadata to JSON file {infofn}')
3790 return None
3791 return True
3792
3793 def _write_description(self, label, ie_result, descfn):
3794 ''' Write description and returns True = written, False = skip, None = error '''
3795 if not self.params.get('writedescription'):
3796 return False
3797 elif not descfn:
3798 self.write_debug(f'Skipping writing {label} description')
3799 return False
3800 elif not self._ensure_dir_exists(descfn):
3801 return None
3802 elif not self.params.get('overwrites', True) and os.path.exists(descfn):
3803 self.to_screen(f'[info] {label.title()} description is already present')
3804 elif ie_result.get('description') is None:
3805 self.report_warning(f'There\'s no {label} description to write')
3806 return False
3807 else:
3808 try:
3809 self.to_screen(f'[info] Writing {label} description to: {descfn}')
3810 with io.open(encodeFilename(descfn), 'w', encoding='utf-8') as descfile:
3811 descfile.write(ie_result['description'])
3812 except (OSError, IOError):
3813 self.report_error(f'Cannot write {label} description file {descfn}')
3814 return None
3815 return True
3816
3817 def _write_subtitles(self, info_dict, filename):
3818 ''' Write subtitles to file and return list of (sub_filename, final_sub_filename); or None if error'''
3819 ret = []
3820 subtitles = info_dict.get('requested_subtitles')
3821 if not subtitles or not (self.params.get('writesubtitles') or self.params.get('writeautomaticsub')):
3822 # subtitles download errors are already managed as troubles in relevant IE
3823 # that way it will silently go on when used with unsupporting IE
3824 return ret
3825
3826 sub_filename_base = self.prepare_filename(info_dict, 'subtitle')
3827 if not sub_filename_base:
3828 self.to_screen('[info] Skipping writing video subtitles')
3829 return ret
3830 for sub_lang, sub_info in subtitles.items():
3831 sub_format = sub_info['ext']
3832 sub_filename = subtitles_filename(filename, sub_lang, sub_format, info_dict.get('ext'))
3833 sub_filename_final = subtitles_filename(sub_filename_base, sub_lang, sub_format, info_dict.get('ext'))
3834 existing_sub = self.existing_file((sub_filename_final, sub_filename))
3835 if existing_sub:
3836 self.to_screen(f'[info] Video subtitle {sub_lang}.{sub_format} is already present')
3837 sub_info['filepath'] = existing_sub
3838 ret.append((existing_sub, sub_filename_final))
3839 continue
3840
3841 self.to_screen(f'[info] Writing video subtitles to: {sub_filename}')
3842 if sub_info.get('data') is not None:
3843 try:
3844 # Use newline='' to prevent conversion of newline characters
3845 # See https://github.com/ytdl-org/youtube-dl/issues/10268
3846 with io.open(sub_filename, 'w', encoding='utf-8', newline='') as subfile:
3847 subfile.write(sub_info['data'])
3848 sub_info['filepath'] = sub_filename
3849 ret.append((sub_filename, sub_filename_final))
3850 continue
3851 except (OSError, IOError):
3852 self.report_error(f'Cannot write video subtitles file {sub_filename}')
3853 return None
3854
3855 try:
3856 sub_copy = sub_info.copy()
3857 sub_copy.setdefault('http_headers', info_dict.get('http_headers'))
3858 self.dl(sub_filename, sub_copy, subtitle=True)
3859 sub_info['filepath'] = sub_filename
3860 ret.append((sub_filename, sub_filename_final))
3861 except (DownloadError, ExtractorError, IOError, OSError, ValueError) + network_exceptions as err:
3862 if self.params.get('ignoreerrors') is not True: # False or 'only_download'
3863 raise DownloadError(f'Unable to download video subtitles for {sub_lang!r}: {err}', err)
3864 self.report_warning(f'Unable to download video subtitles for {sub_lang!r}: {err}')
3865 return ret
3866
3867 def _write_thumbnails(self, label, info_dict, filename, thumb_filename_base=None):
3868 ''' Write thumbnails to file and return list of (thumb_filename, final_thumb_filename) '''
3869 write_all = self.params.get('write_all_thumbnails', False)
3870 thumbnails, ret = [], []
3871 if write_all or self.params.get('writethumbnail', False):
3872 thumbnails = info_dict.get('thumbnails') or []
3873 multiple = write_all and len(thumbnails) > 1
3874
3875 if thumb_filename_base is None:
3876 thumb_filename_base = filename
3877 if thumbnails and not thumb_filename_base:
3878 self.write_debug(f'Skipping writing {label} thumbnail')
3879 return ret
3880
3881 for idx, t in list(enumerate(thumbnails))[::-1]:
3882 thumb_ext = (f'{t["id"]}.' if multiple else '') + determine_ext(t['url'], 'jpg')
3883 thumb_display_id = f'{label} thumbnail {t["id"]}'
3884 thumb_filename = replace_extension(filename, thumb_ext, info_dict.get('ext'))
3885 thumb_filename_final = replace_extension(thumb_filename_base, thumb_ext, info_dict.get('ext'))
3886
3887 existing_thumb = self.existing_file((thumb_filename_final, thumb_filename))
3888 if existing_thumb:
3889 self.to_screen('[info] %s is already present' % (
3890 thumb_display_id if multiple else f'{label} thumbnail').capitalize())
3891 t['filepath'] = existing_thumb
3892 ret.append((existing_thumb, thumb_filename_final))
3893 else:
3894 self.to_screen(f'[info] Downloading {thumb_display_id} ...')
3895 try:
3896 uf = self.urlopen(sanitized_Request(t['url'], headers=t.get('http_headers', {})))
3897 self.to_screen(f'[info] Writing {thumb_display_id} to: {thumb_filename}')
3898 with open(encodeFilename(thumb_filename), 'wb') as thumbf:
3899 shutil.copyfileobj(uf, thumbf)
3900 ret.append((thumb_filename, thumb_filename_final))
3901 t['filepath'] = thumb_filename
3902 except network_exceptions as err:
3903 thumbnails.pop(idx)
3904 self.report_warning(f'Unable to download {thumb_display_id}: {err}')
3905 if ret and not write_all:
3906 break
3907 return ret