]> jfr.im git - yt-dlp.git/blob - yt_dlp/YoutubeDL.py
[cleanup] Remove unused code paths (#2173)
[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 filter_dict,
69 float_or_none,
70 format_bytes,
71 format_field,
72 format_decimal_suffix,
73 formatSeconds,
74 GeoRestrictedError,
75 get_domain,
76 has_certifi,
77 HEADRequest,
78 InAdvancePagedList,
79 int_or_none,
80 iri_to_uri,
81 ISO3166Utils,
82 join_nonempty,
83 LazyList,
84 LINK_TEMPLATES,
85 locked_file,
86 make_dir,
87 make_HTTPS_handler,
88 MaxDownloadsReached,
89 merge_headers,
90 network_exceptions,
91 NO_DEFAULT,
92 number_of_digits,
93 orderedSet,
94 OUTTMPL_TYPES,
95 PagedList,
96 parse_filesize,
97 PerRequestProxyHandler,
98 platform_name,
99 Popen,
100 POSTPROCESS_WHEN,
101 PostProcessingError,
102 preferredencoding,
103 prepend_extension,
104 ReExtractInfo,
105 register_socks_protocols,
106 RejectedVideoReached,
107 remove_terminal_sequences,
108 render_table,
109 replace_extension,
110 SameFileError,
111 sanitize_filename,
112 sanitize_path,
113 sanitize_url,
114 sanitized_Request,
115 std_headers,
116 STR_FORMAT_RE_TMPL,
117 STR_FORMAT_TYPES,
118 str_or_none,
119 strftime_or_none,
120 subtitles_filename,
121 supports_terminal_sequences,
122 timetuple_from_msec,
123 to_high_limit_path,
124 traverse_obj,
125 try_get,
126 UnavailableVideoError,
127 url_basename,
128 variadic,
129 version_tuple,
130 write_json_file,
131 write_string,
132 YoutubeDLCookieProcessor,
133 YoutubeDLHandler,
134 YoutubeDLRedirectHandler,
135 )
136 from .cache import Cache
137 from .minicurses import format_text
138 from .extractor import (
139 gen_extractor_classes,
140 get_info_extractor,
141 _LAZY_LOADER,
142 _PLUGIN_CLASSES as plugin_extractors
143 )
144 from .extractor.openload import PhantomJSwrapper
145 from .downloader import (
146 FFmpegFD,
147 get_suitable_downloader,
148 shorten_protocol_name
149 )
150 from .downloader.rtmp import rtmpdump_version
151 from .postprocessor import (
152 get_postprocessor,
153 EmbedThumbnailPP,
154 FFmpegFixupDuplicateMoovPP,
155 FFmpegFixupDurationPP,
156 FFmpegFixupM3u8PP,
157 FFmpegFixupM4aPP,
158 FFmpegFixupStretchedPP,
159 FFmpegFixupTimestampPP,
160 FFmpegMergerPP,
161 FFmpegPostProcessor,
162 MoveFilesAfterDownloadPP,
163 _PLUGIN_CLASSES as plugin_postprocessors
164 )
165 from .update import detect_variant
166 from .version import __version__, RELEASE_GIT_HEAD
167
168 if compat_os_name == 'nt':
169 import ctypes
170
171
172 class YoutubeDL(object):
173 """YoutubeDL class.
174
175 YoutubeDL objects are the ones responsible of downloading the
176 actual video file and writing it to disk if the user has requested
177 it, among some other tasks. In most cases there should be one per
178 program. As, given a video URL, the downloader doesn't know how to
179 extract all the needed information, task that InfoExtractors do, it
180 has to pass the URL to one of them.
181
182 For this, YoutubeDL objects have a method that allows
183 InfoExtractors to be registered in a given order. When it is passed
184 a URL, the YoutubeDL object handles it to the first InfoExtractor it
185 finds that reports being able to handle it. The InfoExtractor extracts
186 all the information about the video or videos the URL refers to, and
187 YoutubeDL process the extracted information, possibly using a File
188 Downloader to download the video.
189
190 YoutubeDL objects accept a lot of parameters. In order not to saturate
191 the object constructor with arguments, it receives a dictionary of
192 options instead. These options are available through the params
193 attribute for the InfoExtractors to use. The YoutubeDL also
194 registers itself as the downloader in charge for the InfoExtractors
195 that are added to it, so this is a "mutual registration".
196
197 Available options:
198
199 username: Username for authentication purposes.
200 password: Password for authentication purposes.
201 videopassword: Password for accessing a video.
202 ap_mso: Adobe Pass multiple-system operator identifier.
203 ap_username: Multiple-system operator account username.
204 ap_password: Multiple-system operator account password.
205 usenetrc: Use netrc for authentication instead.
206 verbose: Print additional info to stdout.
207 quiet: Do not print messages to stdout.
208 no_warnings: Do not print out anything for warnings.
209 forceprint: A dict with keys WHEN mapped to a list of templates to
210 print to stdout. The allowed keys are video or any of the
211 items in utils.POSTPROCESS_WHEN.
212 For compatibility, a single list is also accepted
213 print_to_file: A dict with keys WHEN (same as forceprint) mapped to
214 a list of tuples with (template, filename)
215 forceurl: Force printing final URL. (Deprecated)
216 forcetitle: Force printing title. (Deprecated)
217 forceid: Force printing ID. (Deprecated)
218 forcethumbnail: Force printing thumbnail URL. (Deprecated)
219 forcedescription: Force printing description. (Deprecated)
220 forcefilename: Force printing final filename. (Deprecated)
221 forceduration: Force printing duration. (Deprecated)
222 forcejson: Force printing info_dict as JSON.
223 dump_single_json: Force printing the info_dict of the whole playlist
224 (or video) as a single JSON line.
225 force_write_download_archive: Force writing download archive regardless
226 of 'skip_download' or 'simulate'.
227 simulate: Do not download the video files. If unset (or None),
228 simulate only if listsubtitles, listformats or list_thumbnails is used
229 format: Video format code. see "FORMAT SELECTION" for more details.
230 You can also pass a function. The function takes 'ctx' as
231 argument and returns the formats to download.
232 See "build_format_selector" for an implementation
233 allow_unplayable_formats: Allow unplayable formats to be extracted and downloaded.
234 ignore_no_formats_error: Ignore "No video formats" error. Usefull for
235 extracting metadata even if the video is not actually
236 available for download (experimental)
237 format_sort: A list of fields by which to sort the video formats.
238 See "Sorting Formats" for more details.
239 format_sort_force: Force the given format_sort. see "Sorting Formats"
240 for more details.
241 prefer_free_formats: Whether to prefer video formats with free containers
242 over non-free ones of same quality.
243 allow_multiple_video_streams: Allow multiple video streams to be merged
244 into a single file
245 allow_multiple_audio_streams: Allow multiple audio streams to be merged
246 into a single file
247 check_formats Whether to test if the formats are downloadable.
248 Can be True (check all), False (check none),
249 'selected' (check selected formats),
250 or None (check only if requested by extractor)
251 paths: Dictionary of output paths. The allowed keys are 'home'
252 'temp' and the keys of OUTTMPL_TYPES (in utils.py)
253 outtmpl: Dictionary of templates for output names. Allowed keys
254 are 'default' and the keys of OUTTMPL_TYPES (in utils.py).
255 For compatibility with youtube-dl, a single string can also be used
256 outtmpl_na_placeholder: Placeholder for unavailable meta fields.
257 restrictfilenames: Do not allow "&" and spaces in file names
258 trim_file_name: Limit length of filename (extension excluded)
259 windowsfilenames: Force the filenames to be windows compatible
260 ignoreerrors: Do not stop on download/postprocessing errors.
261 Can be 'only_download' to ignore only download errors.
262 Default is 'only_download' for CLI, but False for API
263 skip_playlist_after_errors: Number of allowed failures until the rest of
264 the playlist is skipped
265 force_generic_extractor: Force downloader to use the generic extractor
266 overwrites: Overwrite all video and metadata files if True,
267 overwrite only non-video files if None
268 and don't overwrite any file if False
269 For compatibility with youtube-dl,
270 "nooverwrites" may also be used instead
271 playliststart: Playlist item to start at.
272 playlistend: Playlist item to end at.
273 playlist_items: Specific indices of playlist to download.
274 playlistreverse: Download playlist items in reverse order.
275 playlistrandom: Download playlist items in random order.
276 matchtitle: Download only matching titles.
277 rejecttitle: Reject downloads for matching titles.
278 logger: Log messages to a logging.Logger instance.
279 logtostderr: Log messages to stderr instead of stdout.
280 consoletitle: Display progress in console window's titlebar.
281 writedescription: Write the video description to a .description file
282 writeinfojson: Write the video description to a .info.json file
283 clean_infojson: Remove private fields from the infojson
284 getcomments: Extract video comments. This will not be written to disk
285 unless writeinfojson is also given
286 writeannotations: Write the video annotations to a .annotations.xml file
287 writethumbnail: Write the thumbnail image to a file
288 allow_playlist_files: Whether to write playlists' description, infojson etc
289 also to disk when using the 'write*' options
290 write_all_thumbnails: Write all thumbnail formats to files
291 writelink: Write an internet shortcut file, depending on the
292 current platform (.url/.webloc/.desktop)
293 writeurllink: Write a Windows internet shortcut file (.url)
294 writewebloclink: Write a macOS internet shortcut file (.webloc)
295 writedesktoplink: Write a Linux internet shortcut file (.desktop)
296 writesubtitles: Write the video subtitles to a file
297 writeautomaticsub: Write the automatically generated subtitles to a file
298 allsubtitles: Deprecated - Use subtitleslangs = ['all']
299 Downloads all the subtitles of the video
300 (requires writesubtitles or writeautomaticsub)
301 listsubtitles: Lists all available subtitles for the video
302 subtitlesformat: The format code for subtitles
303 subtitleslangs: List of languages of the subtitles to download (can be regex).
304 The list may contain "all" to refer to all the available
305 subtitles. The language can be prefixed with a "-" to
306 exclude it from the requested languages. Eg: ['all', '-live_chat']
307 keepvideo: Keep the video file after post-processing
308 daterange: A DateRange object, download only if the upload_date is in the range.
309 skip_download: Skip the actual download of the video file
310 cachedir: Location of the cache files in the filesystem.
311 False to disable filesystem cache.
312 noplaylist: Download single video instead of a playlist if in doubt.
313 age_limit: An integer representing the user's age in years.
314 Unsuitable videos for the given age are skipped.
315 min_views: An integer representing the minimum view count the video
316 must have in order to not be skipped.
317 Videos without view count information are always
318 downloaded. None for no limit.
319 max_views: An integer representing the maximum view count.
320 Videos that are more popular than that are not
321 downloaded.
322 Videos without view count information are always
323 downloaded. None for no limit.
324 download_archive: File name of a file where all downloads are recorded.
325 Videos already present in the file are not downloaded
326 again.
327 break_on_existing: Stop the download process after attempting to download a
328 file that is in the archive.
329 break_on_reject: Stop the download process when encountering a video that
330 has been filtered out.
331 break_per_url: Whether break_on_reject and break_on_existing
332 should act on each input URL as opposed to for the entire queue
333 cookiefile: File name where cookies should be read from and dumped to
334 cookiesfrombrowser: A tuple containing the name of the browser, the profile
335 name/pathfrom where cookies are loaded, and the name of the
336 keyring. Eg: ('chrome', ) or ('vivaldi', 'default', 'BASICTEXT')
337 legacyserverconnect: Explicitly allow HTTPS connection to servers that do not
338 support RFC 5746 secure renegotiation
339 nocheckcertificate: Do not verify SSL certificates
340 prefer_insecure: Use HTTP instead of HTTPS to retrieve information.
341 At the moment, this is only supported by YouTube.
342 http_headers: A dictionary of custom headers to be used for all requests
343 proxy: URL of the proxy server to use
344 geo_verification_proxy: URL of the proxy to use for IP address verification
345 on geo-restricted sites.
346 socket_timeout: Time to wait for unresponsive hosts, in seconds
347 bidi_workaround: Work around buggy terminals without bidirectional text
348 support, using fridibi
349 debug_printtraffic:Print out sent and received HTTP traffic
350 include_ads: Download ads as well (deprecated)
351 default_search: Prepend this string if an input url is not valid.
352 'auto' for elaborate guessing
353 encoding: Use this encoding instead of the system-specified.
354 extract_flat: Do not resolve URLs, return the immediate result.
355 Pass in 'in_playlist' to only show this behavior for
356 playlist items.
357 wait_for_video: If given, wait for scheduled streams to become available.
358 The value should be a tuple containing the range
359 (min_secs, max_secs) to wait between retries
360 postprocessors: A list of dictionaries, each with an entry
361 * key: The name of the postprocessor. See
362 yt_dlp/postprocessor/__init__.py for a list.
363 * when: When to run the postprocessor. Allowed values are
364 the entries of utils.POSTPROCESS_WHEN
365 Assumed to be 'post_process' if not given
366 post_hooks: Deprecated - Register a custom postprocessor instead
367 A list of functions that get called as the final step
368 for each video file, after all postprocessors have been
369 called. The filename will be passed as the only argument.
370 progress_hooks: A list of functions that get called on download
371 progress, with a dictionary with the entries
372 * status: One of "downloading", "error", or "finished".
373 Check this first and ignore unknown values.
374 * info_dict: The extracted info_dict
375
376 If status is one of "downloading", or "finished", the
377 following properties may also be present:
378 * filename: The final filename (always present)
379 * tmpfilename: The filename we're currently writing to
380 * downloaded_bytes: Bytes on disk
381 * total_bytes: Size of the whole file, None if unknown
382 * total_bytes_estimate: Guess of the eventual file size,
383 None if unavailable.
384 * elapsed: The number of seconds since download started.
385 * eta: The estimated time in seconds, None if unknown
386 * speed: The download speed in bytes/second, None if
387 unknown
388 * fragment_index: The counter of the currently
389 downloaded video fragment.
390 * fragment_count: The number of fragments (= individual
391 files that will be merged)
392
393 Progress hooks are guaranteed to be called at least once
394 (with status "finished") if the download is successful.
395 postprocessor_hooks: A list of functions that get called on postprocessing
396 progress, with a dictionary with the entries
397 * status: One of "started", "processing", or "finished".
398 Check this first and ignore unknown values.
399 * postprocessor: Name of the postprocessor
400 * info_dict: The extracted info_dict
401
402 Progress hooks are guaranteed to be called at least twice
403 (with status "started" and "finished") if the processing is successful.
404 merge_output_format: Extension to use when merging formats.
405 final_ext: Expected final extension; used to detect when the file was
406 already downloaded and converted
407 fixup: Automatically correct known faults of the file.
408 One of:
409 - "never": do nothing
410 - "warn": only emit a warning
411 - "detect_or_warn": check whether we can do anything
412 about it, warn otherwise (default)
413 source_address: Client-side IP address to bind to.
414 call_home: Boolean, true iff we are allowed to contact the
415 yt-dlp servers for debugging. (BROKEN)
416 sleep_interval_requests: Number of seconds to sleep between requests
417 during extraction
418 sleep_interval: Number of seconds to sleep before each download when
419 used alone or a lower bound of a range for randomized
420 sleep before each download (minimum possible number
421 of seconds to sleep) when used along with
422 max_sleep_interval.
423 max_sleep_interval:Upper bound of a range for randomized sleep before each
424 download (maximum possible number of seconds to sleep).
425 Must only be used along with sleep_interval.
426 Actual sleep time will be a random float from range
427 [sleep_interval; max_sleep_interval].
428 sleep_interval_subtitles: Number of seconds to sleep before each subtitle download
429 listformats: Print an overview of available video formats and exit.
430 list_thumbnails: Print a table of all thumbnails and exit.
431 match_filter: A function that gets called with the info_dict of
432 every video.
433 If it returns a message, the video is ignored.
434 If it returns None, the video is downloaded.
435 match_filter_func in utils.py is one example for this.
436 no_color: Do not emit color codes in output.
437 geo_bypass: Bypass geographic restriction via faking X-Forwarded-For
438 HTTP header
439 geo_bypass_country:
440 Two-letter ISO 3166-2 country code that will be used for
441 explicit geographic restriction bypassing via faking
442 X-Forwarded-For HTTP header
443 geo_bypass_ip_block:
444 IP range in CIDR notation that will be used similarly to
445 geo_bypass_country
446
447 The following options determine which downloader is picked:
448 external_downloader: A dictionary of protocol keys and the executable of the
449 external downloader to use for it. The allowed protocols
450 are default|http|ftp|m3u8|dash|rtsp|rtmp|mms.
451 Set the value to 'native' to use the native downloader
452 hls_prefer_native: Deprecated - Use external_downloader = {'m3u8': 'native'}
453 or {'m3u8': 'ffmpeg'} instead.
454 Use the native HLS downloader instead of ffmpeg/avconv
455 if True, otherwise use ffmpeg/avconv if False, otherwise
456 use downloader suggested by extractor if None.
457 compat_opts: Compatibility options. See "Differences in default behavior".
458 The following options do not work when used through the API:
459 filename, abort-on-error, multistreams, no-live-chat, format-sort
460 no-clean-infojson, no-playlist-metafiles, no-keep-subs, no-attach-info-json.
461 Refer __init__.py for their implementation
462 progress_template: Dictionary of templates for progress outputs.
463 Allowed keys are 'download', 'postprocess',
464 'download-title' (console title) and 'postprocess-title'.
465 The template is mapped on a dictionary with keys 'progress' and 'info'
466
467 The following parameters are not used by YoutubeDL itself, they are used by
468 the downloader (see yt_dlp/downloader/common.py):
469 nopart, updatetime, buffersize, ratelimit, throttledratelimit, min_filesize,
470 max_filesize, test, noresizebuffer, retries, file_access_retries, fragment_retries,
471 continuedl, noprogress, xattr_set_filesize, hls_use_mpegts, http_chunk_size,
472 external_downloader_args, concurrent_fragment_downloads.
473
474 The following options are used by the post processors:
475 prefer_ffmpeg: If False, use avconv instead of ffmpeg if both are available,
476 otherwise prefer ffmpeg. (avconv support is deprecated)
477 ffmpeg_location: Location of the ffmpeg/avconv binary; either the path
478 to the binary or its containing directory.
479 postprocessor_args: A dictionary of postprocessor/executable keys (in lower case)
480 and a list of additional command-line arguments for the
481 postprocessor/executable. The dict can also have "PP+EXE" keys
482 which are used when the given exe is used by the given PP.
483 Use 'default' as the name for arguments to passed to all PP
484 For compatibility with youtube-dl, a single list of args
485 can also be used
486
487 The following options are used by the extractors:
488 extractor_retries: Number of times to retry for known errors
489 dynamic_mpd: Whether to process dynamic DASH manifests (default: True)
490 hls_split_discontinuity: Split HLS playlists to different formats at
491 discontinuities such as ad breaks (default: False)
492 extractor_args: A dictionary of arguments to be passed to the extractors.
493 See "EXTRACTOR ARGUMENTS" for details.
494 Eg: {'youtube': {'skip': ['dash', 'hls']}}
495 mark_watched: Mark videos watched (even with --simulate). Only for YouTube
496 youtube_include_dash_manifest: Deprecated - Use extractor_args instead.
497 If True (default), DASH manifests and related
498 data will be downloaded and processed by extractor.
499 You can reduce network I/O by disabling it if you don't
500 care about DASH. (only for youtube)
501 youtube_include_hls_manifest: Deprecated - Use extractor_args instead.
502 If True (default), HLS manifests and related
503 data will be downloaded and processed by extractor.
504 You can reduce network I/O by disabling it if you don't
505 care about HLS. (only for youtube)
506 """
507
508 _NUMERIC_FIELDS = set((
509 'width', 'height', 'tbr', 'abr', 'asr', 'vbr', 'fps', 'filesize', 'filesize_approx',
510 'timestamp', 'release_timestamp',
511 'duration', 'view_count', 'like_count', 'dislike_count', 'repost_count',
512 'average_rating', 'comment_count', 'age_limit',
513 'start_time', 'end_time',
514 'chapter_number', 'season_number', 'episode_number',
515 'track_number', 'disc_number', 'release_year',
516 ))
517
518 _format_fields = {
519 # NB: Keep in sync with the docstring of extractor/common.py
520 'url', 'manifest_url', 'manifest_stream_number', 'ext', 'format', 'format_id', 'format_note',
521 'width', 'height', 'resolution', 'dynamic_range', 'tbr', 'abr', 'acodec', 'asr',
522 'vbr', 'fps', 'vcodec', 'container', 'filesize', 'filesize_approx',
523 'player_url', 'protocol', 'fragment_base_url', 'fragments', 'is_from_start',
524 'preference', 'language', 'language_preference', 'quality', 'source_preference',
525 'http_headers', 'stretched_ratio', 'no_resume', 'has_drm', 'downloader_options',
526 'page_url', 'app', 'play_path', 'tc_url', 'flash_version', 'rtmp_live', 'rtmp_conn', 'rtmp_protocol', 'rtmp_real_time'
527 }
528 _format_selection_exts = {
529 'audio': {'m4a', 'mp3', 'ogg', 'aac'},
530 'video': {'mp4', 'flv', 'webm', '3gp'},
531 'storyboards': {'mhtml'},
532 }
533
534 def __init__(self, params=None, auto_init=True):
535 """Create a FileDownloader object with the given options.
536 @param auto_init Whether to load the default extractors and print header (if verbose).
537 Set to 'no_verbose_header' to not print the header
538 """
539 if params is None:
540 params = {}
541 self.params = params
542 self._ies = {}
543 self._ies_instances = {}
544 self._pps = {k: [] for k in POSTPROCESS_WHEN}
545 self._printed_messages = set()
546 self._first_webpage_request = True
547 self._post_hooks = []
548 self._progress_hooks = []
549 self._postprocessor_hooks = []
550 self._download_retcode = 0
551 self._num_downloads = 0
552 self._num_videos = 0
553 self._playlist_level = 0
554 self._playlist_urls = set()
555 self.cache = Cache(self)
556
557 windows_enable_vt_mode()
558 self._out_files = {
559 'error': sys.stderr,
560 'print': sys.stderr if self.params.get('logtostderr') else sys.stdout,
561 'console': None if compat_os_name == 'nt' else next(
562 filter(supports_terminal_sequences, (sys.stderr, sys.stdout)), None)
563 }
564 self._out_files['screen'] = sys.stderr if self.params.get('quiet') else self._out_files['print']
565 self._allow_colors = {
566 type_: not self.params.get('no_color') and supports_terminal_sequences(self._out_files[type_])
567 for type_ in ('screen', 'error')
568 }
569
570 if sys.version_info < (3, 6):
571 self.report_warning(
572 'Python version %d.%d is not supported! Please update to Python 3.6 or above' % sys.version_info[:2])
573
574 if self.params.get('allow_unplayable_formats'):
575 self.report_warning(
576 f'You have asked for {self._format_err("UNPLAYABLE", self.Styles.EMPHASIS)} formats to be listed/downloaded. '
577 'This is a developer option intended for debugging. \n'
578 ' If you experience any issues while using this option, '
579 f'{self._format_err("DO NOT", self.Styles.ERROR)} open a bug report')
580
581 def check_deprecated(param, option, suggestion):
582 if self.params.get(param) is not None:
583 self.report_warning('%s is deprecated. Use %s instead' % (option, suggestion))
584 return True
585 return False
586
587 if check_deprecated('cn_verification_proxy', '--cn-verification-proxy', '--geo-verification-proxy'):
588 if self.params.get('geo_verification_proxy') is None:
589 self.params['geo_verification_proxy'] = self.params['cn_verification_proxy']
590
591 check_deprecated('autonumber', '--auto-number', '-o "%(autonumber)s-%(title)s.%(ext)s"')
592 check_deprecated('usetitle', '--title', '-o "%(title)s-%(id)s.%(ext)s"')
593 check_deprecated('useid', '--id', '-o "%(id)s.%(ext)s"')
594
595 for msg in self.params.get('_warnings', []):
596 self.report_warning(msg)
597 for msg in self.params.get('_deprecation_warnings', []):
598 self.deprecation_warning(msg)
599
600 if 'list-formats' in self.params.get('compat_opts', []):
601 self.params['listformats_table'] = False
602
603 if 'overwrites' not in self.params and self.params.get('nooverwrites') is not None:
604 # nooverwrites was unnecessarily changed to overwrites
605 # in 0c3d0f51778b153f65c21906031c2e091fcfb641
606 # This ensures compatibility with both keys
607 self.params['overwrites'] = not self.params['nooverwrites']
608 elif self.params.get('overwrites') is None:
609 self.params.pop('overwrites', None)
610 else:
611 self.params['nooverwrites'] = not self.params['overwrites']
612
613 self.params.setdefault('forceprint', {})
614 self.params.setdefault('print_to_file', {})
615
616 # Compatibility with older syntax
617 if not isinstance(params['forceprint'], dict):
618 self.params['forceprint'] = {'video': params['forceprint']}
619
620 if self.params.get('bidi_workaround', False):
621 try:
622 import pty
623 master, slave = pty.openpty()
624 width = compat_get_terminal_size().columns
625 if width is None:
626 width_args = []
627 else:
628 width_args = ['-w', str(width)]
629 sp_kwargs = dict(
630 stdin=subprocess.PIPE,
631 stdout=slave,
632 stderr=self._out_files['error'])
633 try:
634 self._output_process = Popen(['bidiv'] + width_args, **sp_kwargs)
635 except OSError:
636 self._output_process = Popen(['fribidi', '-c', 'UTF-8'] + width_args, **sp_kwargs)
637 self._output_channel = os.fdopen(master, 'rb')
638 except OSError as ose:
639 if ose.errno == errno.ENOENT:
640 self.report_warning(
641 'Could not find fribidi executable, ignoring --bidi-workaround. '
642 'Make sure that fribidi is an executable file in one of the directories in your $PATH.')
643 else:
644 raise
645
646 if auto_init:
647 if auto_init != 'no_verbose_header':
648 self.print_debug_header()
649 self.add_default_info_extractors()
650
651 if (sys.platform != 'win32'
652 and sys.getfilesystemencoding() in ['ascii', 'ANSI_X3.4-1968']
653 and not self.params.get('restrictfilenames', False)):
654 # Unicode filesystem API will throw errors (#1474, #13027)
655 self.report_warning(
656 'Assuming --restrict-filenames since file system encoding '
657 'cannot encode all characters. '
658 'Set the LC_ALL environment variable to fix this.')
659 self.params['restrictfilenames'] = True
660
661 self.outtmpl_dict = self.parse_outtmpl()
662
663 # Creating format selector here allows us to catch syntax errors before the extraction
664 self.format_selector = (
665 self.params.get('format') if self.params.get('format') in (None, '-')
666 else self.params['format'] if callable(self.params['format'])
667 else self.build_format_selector(self.params['format']))
668
669 # Set http_headers defaults according to std_headers
670 self.params['http_headers'] = merge_headers(std_headers, self.params.get('http_headers', {}))
671
672 hooks = {
673 'post_hooks': self.add_post_hook,
674 'progress_hooks': self.add_progress_hook,
675 'postprocessor_hooks': self.add_postprocessor_hook,
676 }
677 for opt, fn in hooks.items():
678 for ph in self.params.get(opt, []):
679 fn(ph)
680
681 for pp_def_raw in self.params.get('postprocessors', []):
682 pp_def = dict(pp_def_raw)
683 when = pp_def.pop('when', 'post_process')
684 self.add_post_processor(
685 get_postprocessor(pp_def.pop('key'))(self, **compat_kwargs(pp_def)),
686 when=when)
687
688 self._setup_opener()
689 register_socks_protocols()
690
691 def preload_download_archive(fn):
692 """Preload the archive, if any is specified"""
693 if fn is None:
694 return False
695 self.write_debug(f'Loading archive file {fn!r}')
696 try:
697 with locked_file(fn, 'r', encoding='utf-8') as archive_file:
698 for line in archive_file:
699 self.archive.add(line.strip())
700 except IOError as ioe:
701 if ioe.errno != errno.ENOENT:
702 raise
703 return False
704 return True
705
706 self.archive = set()
707 preload_download_archive(self.params.get('download_archive'))
708
709 def warn_if_short_id(self, argv):
710 # short YouTube ID starting with dash?
711 idxs = [
712 i for i, a in enumerate(argv)
713 if re.match(r'^-[0-9A-Za-z_-]{10}$', a)]
714 if idxs:
715 correct_argv = (
716 ['yt-dlp']
717 + [a for i, a in enumerate(argv) if i not in idxs]
718 + ['--'] + [argv[i] for i in idxs]
719 )
720 self.report_warning(
721 'Long argument string detected. '
722 'Use -- to separate parameters and URLs, like this:\n%s' %
723 args_to_str(correct_argv))
724
725 def add_info_extractor(self, ie):
726 """Add an InfoExtractor object to the end of the list."""
727 ie_key = ie.ie_key()
728 self._ies[ie_key] = ie
729 if not isinstance(ie, type):
730 self._ies_instances[ie_key] = ie
731 ie.set_downloader(self)
732
733 def _get_info_extractor_class(self, ie_key):
734 ie = self._ies.get(ie_key)
735 if ie is None:
736 ie = get_info_extractor(ie_key)
737 self.add_info_extractor(ie)
738 return ie
739
740 def get_info_extractor(self, ie_key):
741 """
742 Get an instance of an IE with name ie_key, it will try to get one from
743 the _ies list, if there's no instance it will create a new one and add
744 it to the extractor list.
745 """
746 ie = self._ies_instances.get(ie_key)
747 if ie is None:
748 ie = get_info_extractor(ie_key)()
749 self.add_info_extractor(ie)
750 return ie
751
752 def add_default_info_extractors(self):
753 """
754 Add the InfoExtractors returned by gen_extractors to the end of the list
755 """
756 for ie in gen_extractor_classes():
757 self.add_info_extractor(ie)
758
759 def add_post_processor(self, pp, when='post_process'):
760 """Add a PostProcessor object to the end of the chain."""
761 self._pps[when].append(pp)
762 pp.set_downloader(self)
763
764 def add_post_hook(self, ph):
765 """Add the post hook"""
766 self._post_hooks.append(ph)
767
768 def add_progress_hook(self, ph):
769 """Add the download progress hook"""
770 self._progress_hooks.append(ph)
771
772 def add_postprocessor_hook(self, ph):
773 """Add the postprocessing progress hook"""
774 self._postprocessor_hooks.append(ph)
775 for pps in self._pps.values():
776 for pp in pps:
777 pp.add_progress_hook(ph)
778
779 def _bidi_workaround(self, message):
780 if not hasattr(self, '_output_channel'):
781 return message
782
783 assert hasattr(self, '_output_process')
784 assert isinstance(message, compat_str)
785 line_count = message.count('\n') + 1
786 self._output_process.stdin.write((message + '\n').encode('utf-8'))
787 self._output_process.stdin.flush()
788 res = ''.join(self._output_channel.readline().decode('utf-8')
789 for _ in range(line_count))
790 return res[:-len('\n')]
791
792 def _write_string(self, message, out=None, only_once=False):
793 if only_once:
794 if message in self._printed_messages:
795 return
796 self._printed_messages.add(message)
797 write_string(message, out=out, encoding=self.params.get('encoding'))
798
799 def to_stdout(self, message, skip_eol=False, quiet=None):
800 """Print message to stdout"""
801 if quiet is not None:
802 self.deprecation_warning('"YoutubeDL.to_stdout" no longer accepts the argument quiet. Use "YoutubeDL.to_screen" instead')
803 self._write_string(
804 '%s%s' % (self._bidi_workaround(message), ('' if skip_eol else '\n')),
805 self._out_files['print'])
806
807 def to_screen(self, message, skip_eol=False, quiet=None):
808 """Print message to screen if not in quiet mode"""
809 if self.params.get('logger'):
810 self.params['logger'].debug(message)
811 return
812 if (self.params.get('quiet') if quiet is None else quiet) and not self.params.get('verbose'):
813 return
814 self._write_string(
815 '%s%s' % (self._bidi_workaround(message), ('' if skip_eol else '\n')),
816 self._out_files['screen'])
817
818 def to_stderr(self, message, only_once=False):
819 """Print message to stderr"""
820 assert isinstance(message, compat_str)
821 if self.params.get('logger'):
822 self.params['logger'].error(message)
823 else:
824 self._write_string('%s\n' % self._bidi_workaround(message), self._out_files['error'], only_once=only_once)
825
826 def _send_console_code(self, code):
827 if compat_os_name == 'nt' or not self._out_files['console']:
828 return
829 self._write_string(code, self._out_files['console'])
830
831 def to_console_title(self, message):
832 if not self.params.get('consoletitle', False):
833 return
834 message = remove_terminal_sequences(message)
835 if compat_os_name == 'nt':
836 if ctypes.windll.kernel32.GetConsoleWindow():
837 # c_wchar_p() might not be necessary if `message` is
838 # already of type unicode()
839 ctypes.windll.kernel32.SetConsoleTitleW(ctypes.c_wchar_p(message))
840 else:
841 self._send_console_code(f'\033]0;{message}\007')
842
843 def save_console_title(self):
844 if not self.params.get('consoletitle') or self.params.get('simulate'):
845 return
846 self._send_console_code('\033[22;0t') # Save the title on stack
847
848 def restore_console_title(self):
849 if not self.params.get('consoletitle') or self.params.get('simulate'):
850 return
851 self._send_console_code('\033[23;0t') # Restore the title from stack
852
853 def __enter__(self):
854 self.save_console_title()
855 return self
856
857 def __exit__(self, *args):
858 self.restore_console_title()
859
860 if self.params.get('cookiefile') is not None:
861 self.cookiejar.save(ignore_discard=True, ignore_expires=True)
862
863 def trouble(self, message=None, tb=None, is_error=True):
864 """Determine action to take when a download problem appears.
865
866 Depending on if the downloader has been configured to ignore
867 download errors or not, this method may throw an exception or
868 not when errors are found, after printing the message.
869
870 @param tb If given, is additional traceback information
871 @param is_error Whether to raise error according to ignorerrors
872 """
873 if message is not None:
874 self.to_stderr(message)
875 if self.params.get('verbose'):
876 if tb is None:
877 if sys.exc_info()[0]: # if .trouble has been called from an except block
878 tb = ''
879 if hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
880 tb += ''.join(traceback.format_exception(*sys.exc_info()[1].exc_info))
881 tb += encode_compat_str(traceback.format_exc())
882 else:
883 tb_data = traceback.format_list(traceback.extract_stack())
884 tb = ''.join(tb_data)
885 if tb:
886 self.to_stderr(tb)
887 if not is_error:
888 return
889 if not self.params.get('ignoreerrors'):
890 if sys.exc_info()[0] and hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
891 exc_info = sys.exc_info()[1].exc_info
892 else:
893 exc_info = sys.exc_info()
894 raise DownloadError(message, exc_info)
895 self._download_retcode = 1
896
897 class Styles(Enum):
898 HEADERS = 'yellow'
899 EMPHASIS = 'light blue'
900 ID = 'green'
901 DELIM = 'blue'
902 ERROR = 'red'
903 WARNING = 'yellow'
904 SUPPRESS = 'light black'
905
906 def _format_text(self, handle, allow_colors, text, f, fallback=None, *, test_encoding=False):
907 if test_encoding:
908 original_text = text
909 # handle.encoding can be None. See https://github.com/yt-dlp/yt-dlp/issues/2711
910 encoding = self.params.get('encoding') or getattr(handle, 'encoding', None) or 'ascii'
911 text = text.encode(encoding, 'ignore').decode(encoding)
912 if fallback is not None and text != original_text:
913 text = fallback
914 if isinstance(f, self.Styles):
915 f = f.value
916 return format_text(text, f) if allow_colors else text if fallback is None else fallback
917
918 def _format_screen(self, *args, **kwargs):
919 return self._format_text(
920 self._out_files['screen'], self._allow_colors['screen'], *args, **kwargs)
921
922 def _format_err(self, *args, **kwargs):
923 return self._format_text(
924 self._out_files['error'], self._allow_colors['error'], *args, **kwargs)
925
926 def report_warning(self, message, only_once=False):
927 '''
928 Print the message to stderr, it will be prefixed with 'WARNING:'
929 If stderr is a tty file the 'WARNING:' will be colored
930 '''
931 if self.params.get('logger') is not None:
932 self.params['logger'].warning(message)
933 else:
934 if self.params.get('no_warnings'):
935 return
936 self.to_stderr(f'{self._format_err("WARNING:", self.Styles.WARNING)} {message}', only_once)
937
938 def deprecation_warning(self, message):
939 if self.params.get('logger') is not None:
940 self.params['logger'].warning(f'DeprecationWarning: {message}')
941 else:
942 self.to_stderr(f'{self._format_err("DeprecationWarning:", self.Styles.ERROR)} {message}', True)
943
944 def report_error(self, message, *args, **kwargs):
945 '''
946 Do the same as trouble, but prefixes the message with 'ERROR:', colored
947 in red if stderr is a tty file.
948 '''
949 self.trouble(f'{self._format_err("ERROR:", self.Styles.ERROR)} {message}', *args, **kwargs)
950
951 def write_debug(self, message, only_once=False):
952 '''Log debug message or Print message to stderr'''
953 if not self.params.get('verbose', False):
954 return
955 message = '[debug] %s' % message
956 if self.params.get('logger'):
957 self.params['logger'].debug(message)
958 else:
959 self.to_stderr(message, only_once)
960
961 def report_file_already_downloaded(self, file_name):
962 """Report file has already been fully downloaded."""
963 try:
964 self.to_screen('[download] %s has already been downloaded' % file_name)
965 except UnicodeEncodeError:
966 self.to_screen('[download] The file has already been downloaded')
967
968 def report_file_delete(self, file_name):
969 """Report that existing file will be deleted."""
970 try:
971 self.to_screen('Deleting existing file %s' % file_name)
972 except UnicodeEncodeError:
973 self.to_screen('Deleting existing file')
974
975 def raise_no_formats(self, info, forced=False, *, msg=None):
976 has_drm = info.get('__has_drm')
977 ignored, expected = self.params.get('ignore_no_formats_error'), bool(msg)
978 msg = msg or has_drm and 'This video is DRM protected' or 'No video formats found!'
979 if forced or not ignored:
980 raise ExtractorError(msg, video_id=info['id'], ie=info['extractor'],
981 expected=has_drm or ignored or expected)
982 else:
983 self.report_warning(msg)
984
985 def parse_outtmpl(self):
986 outtmpl_dict = self.params.get('outtmpl', {})
987 if not isinstance(outtmpl_dict, dict):
988 outtmpl_dict = {'default': outtmpl_dict}
989 # Remove spaces in the default template
990 if self.params.get('restrictfilenames'):
991 sanitize = lambda x: x.replace(' - ', ' ').replace(' ', '-')
992 else:
993 sanitize = lambda x: x
994 outtmpl_dict.update({
995 k: sanitize(v) for k, v in DEFAULT_OUTTMPL.items()
996 if outtmpl_dict.get(k) is None})
997 for key, val in outtmpl_dict.items():
998 if isinstance(val, bytes):
999 self.report_warning(
1000 'Parameter outtmpl is bytes, but should be a unicode string. '
1001 'Put from __future__ import unicode_literals at the top of your code file or consider switching to Python 3.x.')
1002 return outtmpl_dict
1003
1004 def get_output_path(self, dir_type='', filename=None):
1005 paths = self.params.get('paths', {})
1006 assert isinstance(paths, dict)
1007 path = os.path.join(
1008 expand_path(paths.get('home', '').strip()),
1009 expand_path(paths.get(dir_type, '').strip()) if dir_type else '',
1010 filename or '')
1011 return sanitize_path(path, force=self.params.get('windowsfilenames'))
1012
1013 @staticmethod
1014 def _outtmpl_expandpath(outtmpl):
1015 # expand_path translates '%%' into '%' and '$$' into '$'
1016 # correspondingly that is not what we want since we need to keep
1017 # '%%' intact for template dict substitution step. Working around
1018 # with boundary-alike separator hack.
1019 sep = ''.join([random.choice(ascii_letters) for _ in range(32)])
1020 outtmpl = outtmpl.replace('%%', '%{0}%'.format(sep)).replace('$$', '${0}$'.format(sep))
1021
1022 # outtmpl should be expand_path'ed before template dict substitution
1023 # because meta fields may contain env variables we don't want to
1024 # be expanded. For example, for outtmpl "%(title)s.%(ext)s" and
1025 # title "Hello $PATH", we don't want `$PATH` to be expanded.
1026 return expand_path(outtmpl).replace(sep, '')
1027
1028 @staticmethod
1029 def escape_outtmpl(outtmpl):
1030 ''' Escape any remaining strings like %s, %abc% etc. '''
1031 return re.sub(
1032 STR_FORMAT_RE_TMPL.format('', '(?![%(\0])'),
1033 lambda mobj: ('' if mobj.group('has_key') else '%') + mobj.group(0),
1034 outtmpl)
1035
1036 @classmethod
1037 def validate_outtmpl(cls, outtmpl):
1038 ''' @return None or Exception object '''
1039 outtmpl = re.sub(
1040 STR_FORMAT_RE_TMPL.format('[^)]*', '[ljqBUDS]'),
1041 lambda mobj: f'{mobj.group(0)[:-1]}s',
1042 cls._outtmpl_expandpath(outtmpl))
1043 try:
1044 cls.escape_outtmpl(outtmpl) % collections.defaultdict(int)
1045 return None
1046 except ValueError as err:
1047 return err
1048
1049 @staticmethod
1050 def _copy_infodict(info_dict):
1051 info_dict = dict(info_dict)
1052 info_dict.pop('__postprocessors', None)
1053 return info_dict
1054
1055 def prepare_outtmpl(self, outtmpl, info_dict, sanitize=False):
1056 """ Make the outtmpl and info_dict suitable for substitution: ydl.escape_outtmpl(outtmpl) % info_dict
1057 @param sanitize Whether to sanitize the output as a filename.
1058 For backward compatibility, a function can also be passed
1059 """
1060
1061 info_dict.setdefault('epoch', int(time.time())) # keep epoch consistent once set
1062
1063 info_dict = self._copy_infodict(info_dict)
1064 info_dict['duration_string'] = ( # %(duration>%H-%M-%S)s is wrong if duration > 24hrs
1065 formatSeconds(info_dict['duration'], '-' if sanitize else ':')
1066 if info_dict.get('duration', None) is not None
1067 else None)
1068 info_dict['autonumber'] = self.params.get('autonumber_start', 1) - 1 + self._num_downloads
1069 info_dict['video_autonumber'] = self._num_videos
1070 if info_dict.get('resolution') is None:
1071 info_dict['resolution'] = self.format_resolution(info_dict, default=None)
1072
1073 # For fields playlist_index, playlist_autonumber and autonumber convert all occurrences
1074 # of %(field)s to %(field)0Nd for backward compatibility
1075 field_size_compat_map = {
1076 'playlist_index': number_of_digits(info_dict.get('_last_playlist_index') or 0),
1077 'playlist_autonumber': number_of_digits(info_dict.get('n_entries') or 0),
1078 'autonumber': self.params.get('autonumber_size') or 5,
1079 }
1080
1081 TMPL_DICT = {}
1082 EXTERNAL_FORMAT_RE = re.compile(STR_FORMAT_RE_TMPL.format('[^)]*', f'[{STR_FORMAT_TYPES}ljqBUDS]'))
1083 MATH_FUNCTIONS = {
1084 '+': float.__add__,
1085 '-': float.__sub__,
1086 }
1087 # Field is of the form key1.key2...
1088 # where keys (except first) can be string, int or slice
1089 FIELD_RE = r'\w*(?:\.(?:\w+|{num}|{num}?(?::{num}?){{1,2}}))*'.format(num=r'(?:-?\d+)')
1090 MATH_FIELD_RE = r'''(?:{field}|{num})'''.format(field=FIELD_RE, num=r'-?\d+(?:.\d+)?')
1091 MATH_OPERATORS_RE = r'(?:%s)' % '|'.join(map(re.escape, MATH_FUNCTIONS.keys()))
1092 INTERNAL_FORMAT_RE = re.compile(r'''(?x)
1093 (?P<negate>-)?
1094 (?P<fields>{field})
1095 (?P<maths>(?:{math_op}{math_field})*)
1096 (?:>(?P<strf_format>.+?))?
1097 (?P<remaining>
1098 (?P<alternate>(?<!\\),[^|&)]+)?
1099 (?:&(?P<replacement>.*?))?
1100 (?:\|(?P<default>.*?))?
1101 )$'''.format(field=FIELD_RE, math_op=MATH_OPERATORS_RE, math_field=MATH_FIELD_RE))
1102
1103 def _traverse_infodict(k):
1104 k = k.split('.')
1105 if k[0] == '':
1106 k.pop(0)
1107 return traverse_obj(info_dict, k, is_user_input=True, traverse_string=True)
1108
1109 def get_value(mdict):
1110 # Object traversal
1111 value = _traverse_infodict(mdict['fields'])
1112 # Negative
1113 if mdict['negate']:
1114 value = float_or_none(value)
1115 if value is not None:
1116 value *= -1
1117 # Do maths
1118 offset_key = mdict['maths']
1119 if offset_key:
1120 value = float_or_none(value)
1121 operator = None
1122 while offset_key:
1123 item = re.match(
1124 MATH_FIELD_RE if operator else MATH_OPERATORS_RE,
1125 offset_key).group(0)
1126 offset_key = offset_key[len(item):]
1127 if operator is None:
1128 operator = MATH_FUNCTIONS[item]
1129 continue
1130 item, multiplier = (item[1:], -1) if item[0] == '-' else (item, 1)
1131 offset = float_or_none(item)
1132 if offset is None:
1133 offset = float_or_none(_traverse_infodict(item))
1134 try:
1135 value = operator(value, multiplier * offset)
1136 except (TypeError, ZeroDivisionError):
1137 return None
1138 operator = None
1139 # Datetime formatting
1140 if mdict['strf_format']:
1141 value = strftime_or_none(value, mdict['strf_format'].replace('\\,', ','))
1142
1143 return value
1144
1145 na = self.params.get('outtmpl_na_placeholder', 'NA')
1146
1147 def filename_sanitizer(key, value, restricted=self.params.get('restrictfilenames')):
1148 return sanitize_filename(str(value), restricted=restricted, is_id=(
1149 bool(re.search(r'(^|[_.])id(\.|$)', key))
1150 if 'filename-sanitization' in self.params.get('compat_opts', [])
1151 else NO_DEFAULT))
1152
1153 sanitizer = sanitize if callable(sanitize) else filename_sanitizer
1154 sanitize = bool(sanitize)
1155
1156 def _dumpjson_default(obj):
1157 if isinstance(obj, (set, LazyList)):
1158 return list(obj)
1159 return repr(obj)
1160
1161 def create_key(outer_mobj):
1162 if not outer_mobj.group('has_key'):
1163 return outer_mobj.group(0)
1164 key = outer_mobj.group('key')
1165 mobj = re.match(INTERNAL_FORMAT_RE, key)
1166 initial_field = mobj.group('fields') if mobj else ''
1167 value, replacement, default = None, None, na
1168 while mobj:
1169 mobj = mobj.groupdict()
1170 default = mobj['default'] if mobj['default'] is not None else default
1171 value = get_value(mobj)
1172 replacement = mobj['replacement']
1173 if value is None and mobj['alternate']:
1174 mobj = re.match(INTERNAL_FORMAT_RE, mobj['remaining'][1:])
1175 else:
1176 break
1177
1178 fmt = outer_mobj.group('format')
1179 if fmt == 's' and value is not None and key in field_size_compat_map.keys():
1180 fmt = '0{:d}d'.format(field_size_compat_map[key])
1181
1182 value = default if value is None else value if replacement is None else replacement
1183
1184 flags = outer_mobj.group('conversion') or ''
1185 str_fmt = f'{fmt[:-1]}s'
1186 if fmt[-1] == 'l': # list
1187 delim = '\n' if '#' in flags else ', '
1188 value, fmt = delim.join(map(str, variadic(value, allowed_types=(str, bytes)))), str_fmt
1189 elif fmt[-1] == 'j': # json
1190 value, fmt = json.dumps(value, default=_dumpjson_default, indent=4 if '#' in flags else None), str_fmt
1191 elif fmt[-1] == 'q': # quoted
1192 value = map(str, variadic(value) if '#' in flags else [value])
1193 value, fmt = ' '.join(map(compat_shlex_quote, value)), str_fmt
1194 elif fmt[-1] == 'B': # bytes
1195 value = f'%{str_fmt}'.encode('utf-8') % str(value).encode('utf-8')
1196 value, fmt = value.decode('utf-8', 'ignore'), 's'
1197 elif fmt[-1] == 'U': # unicode normalized
1198 value, fmt = unicodedata.normalize(
1199 # "+" = compatibility equivalence, "#" = NFD
1200 'NF%s%s' % ('K' if '+' in flags else '', 'D' if '#' in flags else 'C'),
1201 value), str_fmt
1202 elif fmt[-1] == 'D': # decimal suffix
1203 num_fmt, fmt = fmt[:-1].replace('#', ''), 's'
1204 value = format_decimal_suffix(value, f'%{num_fmt}f%s' if num_fmt else '%d%s',
1205 factor=1024 if '#' in flags else 1000)
1206 elif fmt[-1] == 'S': # filename sanitization
1207 value, fmt = filename_sanitizer(initial_field, value, restricted='#' in flags), str_fmt
1208 elif fmt[-1] == 'c':
1209 if value:
1210 value = str(value)[0]
1211 else:
1212 fmt = str_fmt
1213 elif fmt[-1] not in 'rs': # numeric
1214 value = float_or_none(value)
1215 if value is None:
1216 value, fmt = default, 's'
1217
1218 if sanitize:
1219 if fmt[-1] == 'r':
1220 # If value is an object, sanitize might convert it to a string
1221 # So we convert it to repr first
1222 value, fmt = repr(value), str_fmt
1223 if fmt[-1] in 'csr':
1224 value = sanitizer(initial_field, value)
1225
1226 key = '%s\0%s' % (key.replace('%', '%\0'), outer_mobj.group('format'))
1227 TMPL_DICT[key] = value
1228 return '{prefix}%({key}){fmt}'.format(key=key, fmt=fmt, prefix=outer_mobj.group('prefix'))
1229
1230 return EXTERNAL_FORMAT_RE.sub(create_key, outtmpl), TMPL_DICT
1231
1232 def evaluate_outtmpl(self, outtmpl, info_dict, *args, **kwargs):
1233 outtmpl, info_dict = self.prepare_outtmpl(outtmpl, info_dict, *args, **kwargs)
1234 return self.escape_outtmpl(outtmpl) % info_dict
1235
1236 def _prepare_filename(self, info_dict, *, outtmpl=None, tmpl_type=None):
1237 assert None in (outtmpl, tmpl_type), 'outtmpl and tmpl_type are mutually exclusive'
1238 if outtmpl is None:
1239 outtmpl = self.outtmpl_dict.get(tmpl_type or 'default', self.outtmpl_dict['default'])
1240 try:
1241 outtmpl = self._outtmpl_expandpath(outtmpl)
1242 filename = self.evaluate_outtmpl(outtmpl, info_dict, True)
1243 if not filename:
1244 return None
1245
1246 if tmpl_type in ('', 'temp'):
1247 final_ext, ext = self.params.get('final_ext'), info_dict.get('ext')
1248 if final_ext and ext and final_ext != ext and filename.endswith(f'.{final_ext}'):
1249 filename = replace_extension(filename, ext, final_ext)
1250 elif tmpl_type:
1251 force_ext = OUTTMPL_TYPES[tmpl_type]
1252 if force_ext:
1253 filename = replace_extension(filename, force_ext, info_dict.get('ext'))
1254
1255 # https://github.com/blackjack4494/youtube-dlc/issues/85
1256 trim_file_name = self.params.get('trim_file_name', False)
1257 if trim_file_name:
1258 no_ext, *ext = filename.rsplit('.', 2)
1259 filename = join_nonempty(no_ext[:trim_file_name], *ext, delim='.')
1260
1261 return filename
1262 except ValueError as err:
1263 self.report_error('Error in output template: ' + str(err) + ' (encoding: ' + repr(preferredencoding()) + ')')
1264 return None
1265
1266 def prepare_filename(self, info_dict, dir_type='', *, outtmpl=None, warn=False):
1267 """Generate the output filename"""
1268 if outtmpl:
1269 assert not dir_type, 'outtmpl and dir_type are mutually exclusive'
1270 dir_type = None
1271 filename = self._prepare_filename(info_dict, tmpl_type=dir_type, outtmpl=outtmpl)
1272 if not filename and dir_type not in ('', 'temp'):
1273 return ''
1274
1275 if warn:
1276 if not self.params.get('paths'):
1277 pass
1278 elif filename == '-':
1279 self.report_warning('--paths is ignored when an outputting to stdout', only_once=True)
1280 elif os.path.isabs(filename):
1281 self.report_warning('--paths is ignored since an absolute path is given in output template', only_once=True)
1282 if filename == '-' or not filename:
1283 return filename
1284
1285 return self.get_output_path(dir_type, filename)
1286
1287 def _match_entry(self, info_dict, incomplete=False, silent=False):
1288 """ Returns None if the file should be downloaded """
1289
1290 video_title = info_dict.get('title', info_dict.get('id', 'video'))
1291
1292 def check_filter():
1293 if 'title' in info_dict:
1294 # This can happen when we're just evaluating the playlist
1295 title = info_dict['title']
1296 matchtitle = self.params.get('matchtitle', False)
1297 if matchtitle:
1298 if not re.search(matchtitle, title, re.IGNORECASE):
1299 return '"' + title + '" title did not match pattern "' + matchtitle + '"'
1300 rejecttitle = self.params.get('rejecttitle', False)
1301 if rejecttitle:
1302 if re.search(rejecttitle, title, re.IGNORECASE):
1303 return '"' + title + '" title matched reject pattern "' + rejecttitle + '"'
1304 date = info_dict.get('upload_date')
1305 if date is not None:
1306 dateRange = self.params.get('daterange', DateRange())
1307 if date not in dateRange:
1308 return '%s upload date is not in range %s' % (date_from_str(date).isoformat(), dateRange)
1309 view_count = info_dict.get('view_count')
1310 if view_count is not None:
1311 min_views = self.params.get('min_views')
1312 if min_views is not None and view_count < min_views:
1313 return 'Skipping %s, because it has not reached minimum view count (%d/%d)' % (video_title, view_count, min_views)
1314 max_views = self.params.get('max_views')
1315 if max_views is not None and view_count > max_views:
1316 return 'Skipping %s, because it has exceeded the maximum view count (%d/%d)' % (video_title, view_count, max_views)
1317 if age_restricted(info_dict.get('age_limit'), self.params.get('age_limit')):
1318 return 'Skipping "%s" because it is age restricted' % video_title
1319
1320 match_filter = self.params.get('match_filter')
1321 if match_filter is not None:
1322 try:
1323 ret = match_filter(info_dict, incomplete=incomplete)
1324 except TypeError:
1325 # For backward compatibility
1326 ret = None if incomplete else match_filter(info_dict)
1327 if ret is not None:
1328 return ret
1329 return None
1330
1331 if self.in_download_archive(info_dict):
1332 reason = '%s has already been recorded in the archive' % video_title
1333 break_opt, break_err = 'break_on_existing', ExistingVideoReached
1334 else:
1335 reason = check_filter()
1336 break_opt, break_err = 'break_on_reject', RejectedVideoReached
1337 if reason is not None:
1338 if not silent:
1339 self.to_screen('[download] ' + reason)
1340 if self.params.get(break_opt, False):
1341 raise break_err()
1342 return reason
1343
1344 @staticmethod
1345 def add_extra_info(info_dict, extra_info):
1346 '''Set the keys from extra_info in info dict if they are missing'''
1347 for key, value in extra_info.items():
1348 info_dict.setdefault(key, value)
1349
1350 def extract_info(self, url, download=True, ie_key=None, extra_info=None,
1351 process=True, force_generic_extractor=False):
1352 """
1353 Return a list with a dictionary for each video extracted.
1354
1355 Arguments:
1356 url -- URL to extract
1357
1358 Keyword arguments:
1359 download -- whether to download videos during extraction
1360 ie_key -- extractor key hint
1361 extra_info -- dictionary containing the extra values to add to each result
1362 process -- whether to resolve all unresolved references (URLs, playlist items),
1363 must be True for download to work.
1364 force_generic_extractor -- force using the generic extractor
1365 """
1366
1367 if extra_info is None:
1368 extra_info = {}
1369
1370 if not ie_key and force_generic_extractor:
1371 ie_key = 'Generic'
1372
1373 if ie_key:
1374 ies = {ie_key: self._get_info_extractor_class(ie_key)}
1375 else:
1376 ies = self._ies
1377
1378 for ie_key, ie in ies.items():
1379 if not ie.suitable(url):
1380 continue
1381
1382 if not ie.working():
1383 self.report_warning('The program functionality for this site has been marked as broken, '
1384 'and will probably not work.')
1385
1386 temp_id = ie.get_temp_id(url)
1387 if temp_id is not None and self.in_download_archive({'id': temp_id, 'ie_key': ie_key}):
1388 self.to_screen(f'[{ie_key}] {temp_id}: has already been recorded in the archive')
1389 if self.params.get('break_on_existing', False):
1390 raise ExistingVideoReached()
1391 break
1392 return self.__extract_info(url, self.get_info_extractor(ie_key), download, extra_info, process)
1393 else:
1394 self.report_error('no suitable InfoExtractor for URL %s' % url)
1395
1396 def __handle_extraction_exceptions(func):
1397 @functools.wraps(func)
1398 def wrapper(self, *args, **kwargs):
1399 while True:
1400 try:
1401 return func(self, *args, **kwargs)
1402 except (DownloadCancelled, LazyList.IndexError, PagedList.IndexError):
1403 raise
1404 except ReExtractInfo as e:
1405 if e.expected:
1406 self.to_screen(f'{e}; Re-extracting data')
1407 else:
1408 self.to_stderr('\r')
1409 self.report_warning(f'{e}; Re-extracting data')
1410 continue
1411 except GeoRestrictedError as e:
1412 msg = e.msg
1413 if e.countries:
1414 msg += '\nThis video is available in %s.' % ', '.join(
1415 map(ISO3166Utils.short2full, e.countries))
1416 msg += '\nYou might want to use a VPN or a proxy server (with --proxy) to workaround.'
1417 self.report_error(msg)
1418 except ExtractorError as e: # An error we somewhat expected
1419 self.report_error(str(e), e.format_traceback())
1420 except Exception as e:
1421 if self.params.get('ignoreerrors'):
1422 self.report_error(str(e), tb=encode_compat_str(traceback.format_exc()))
1423 else:
1424 raise
1425 break
1426 return wrapper
1427
1428 def _wait_for_video(self, ie_result):
1429 if (not self.params.get('wait_for_video')
1430 or ie_result.get('_type', 'video') != 'video'
1431 or ie_result.get('formats') or ie_result.get('url')):
1432 return
1433
1434 format_dur = lambda dur: '%02d:%02d:%02d' % timetuple_from_msec(dur * 1000)[:-1]
1435 last_msg = ''
1436
1437 def progress(msg):
1438 nonlocal last_msg
1439 self.to_screen(msg + ' ' * (len(last_msg) - len(msg)) + '\r', skip_eol=True)
1440 last_msg = msg
1441
1442 min_wait, max_wait = self.params.get('wait_for_video')
1443 diff = try_get(ie_result, lambda x: x['release_timestamp'] - time.time())
1444 if diff is None and ie_result.get('live_status') == 'is_upcoming':
1445 diff = round(random.uniform(min_wait, max_wait) if (max_wait and min_wait) else (max_wait or min_wait), 0)
1446 self.report_warning('Release time of video is not known')
1447 elif (diff or 0) <= 0:
1448 self.report_warning('Video should already be available according to extracted info')
1449 diff = min(max(diff or 0, min_wait or 0), max_wait or float('inf'))
1450 self.to_screen(f'[wait] Waiting for {format_dur(diff)} - Press Ctrl+C to try now')
1451
1452 wait_till = time.time() + diff
1453 try:
1454 while True:
1455 diff = wait_till - time.time()
1456 if diff <= 0:
1457 progress('')
1458 raise ReExtractInfo('[wait] Wait period ended', expected=True)
1459 progress(f'[wait] Remaining time until next attempt: {self._format_screen(format_dur(diff), self.Styles.EMPHASIS)}')
1460 time.sleep(1)
1461 except KeyboardInterrupt:
1462 progress('')
1463 raise ReExtractInfo('[wait] Interrupted by user', expected=True)
1464 except BaseException as e:
1465 if not isinstance(e, ReExtractInfo):
1466 self.to_screen('')
1467 raise
1468
1469 @__handle_extraction_exceptions
1470 def __extract_info(self, url, ie, download, extra_info, process):
1471 ie_result = ie.extract(url)
1472 if ie_result is None: # Finished already (backwards compatibility; listformats and friends should be moved here)
1473 return
1474 if isinstance(ie_result, list):
1475 # Backwards compatibility: old IE result format
1476 ie_result = {
1477 '_type': 'compat_list',
1478 'entries': ie_result,
1479 }
1480 if extra_info.get('original_url'):
1481 ie_result.setdefault('original_url', extra_info['original_url'])
1482 self.add_default_extra_info(ie_result, ie, url)
1483 if process:
1484 self._wait_for_video(ie_result)
1485 return self.process_ie_result(ie_result, download, extra_info)
1486 else:
1487 return ie_result
1488
1489 def add_default_extra_info(self, ie_result, ie, url):
1490 if url is not None:
1491 self.add_extra_info(ie_result, {
1492 'webpage_url': url,
1493 'original_url': url,
1494 })
1495 webpage_url = ie_result.get('webpage_url')
1496 if webpage_url:
1497 self.add_extra_info(ie_result, {
1498 'webpage_url_basename': url_basename(webpage_url),
1499 'webpage_url_domain': get_domain(webpage_url),
1500 })
1501 if ie is not None:
1502 self.add_extra_info(ie_result, {
1503 'extractor': ie.IE_NAME,
1504 'extractor_key': ie.ie_key(),
1505 })
1506
1507 def process_ie_result(self, ie_result, download=True, extra_info=None):
1508 """
1509 Take the result of the ie(may be modified) and resolve all unresolved
1510 references (URLs, playlist items).
1511
1512 It will also download the videos if 'download'.
1513 Returns the resolved ie_result.
1514 """
1515 if extra_info is None:
1516 extra_info = {}
1517 result_type = ie_result.get('_type', 'video')
1518
1519 if result_type in ('url', 'url_transparent'):
1520 ie_result['url'] = sanitize_url(ie_result['url'])
1521 if ie_result.get('original_url'):
1522 extra_info.setdefault('original_url', ie_result['original_url'])
1523
1524 extract_flat = self.params.get('extract_flat', False)
1525 if ((extract_flat == 'in_playlist' and 'playlist' in extra_info)
1526 or extract_flat is True):
1527 info_copy = ie_result.copy()
1528 ie = try_get(ie_result.get('ie_key'), self.get_info_extractor)
1529 if ie and not ie_result.get('id'):
1530 info_copy['id'] = ie.get_temp_id(ie_result['url'])
1531 self.add_default_extra_info(info_copy, ie, ie_result['url'])
1532 self.add_extra_info(info_copy, extra_info)
1533 info_copy, _ = self.pre_process(info_copy)
1534 self.__forced_printings(info_copy, self.prepare_filename(info_copy), incomplete=True)
1535 if self.params.get('force_write_download_archive', False):
1536 self.record_download_archive(info_copy)
1537 return ie_result
1538
1539 if result_type == 'video':
1540 self.add_extra_info(ie_result, extra_info)
1541 ie_result = self.process_video_result(ie_result, download=download)
1542 additional_urls = (ie_result or {}).get('additional_urls')
1543 if additional_urls:
1544 # TODO: Improve MetadataParserPP to allow setting a list
1545 if isinstance(additional_urls, compat_str):
1546 additional_urls = [additional_urls]
1547 self.to_screen(
1548 '[info] %s: %d additional URL(s) requested' % (ie_result['id'], len(additional_urls)))
1549 self.write_debug('Additional URLs: "%s"' % '", "'.join(additional_urls))
1550 ie_result['additional_entries'] = [
1551 self.extract_info(
1552 url, download, extra_info=extra_info,
1553 force_generic_extractor=self.params.get('force_generic_extractor'))
1554 for url in additional_urls
1555 ]
1556 return ie_result
1557 elif result_type == 'url':
1558 # We have to add extra_info to the results because it may be
1559 # contained in a playlist
1560 return self.extract_info(
1561 ie_result['url'], download,
1562 ie_key=ie_result.get('ie_key'),
1563 extra_info=extra_info)
1564 elif result_type == 'url_transparent':
1565 # Use the information from the embedding page
1566 info = self.extract_info(
1567 ie_result['url'], ie_key=ie_result.get('ie_key'),
1568 extra_info=extra_info, download=False, process=False)
1569
1570 # extract_info may return None when ignoreerrors is enabled and
1571 # extraction failed with an error, don't crash and return early
1572 # in this case
1573 if not info:
1574 return info
1575
1576 new_result = info.copy()
1577 new_result.update(filter_dict(ie_result, lambda k, v: (
1578 v is not None and k not in {'_type', 'url', 'id', 'extractor', 'extractor_key', 'ie_key'})))
1579
1580 # Extracted info may not be a video result (i.e.
1581 # info.get('_type', 'video') != video) but rather an url or
1582 # url_transparent. In such cases outer metadata (from ie_result)
1583 # should be propagated to inner one (info). For this to happen
1584 # _type of info should be overridden with url_transparent. This
1585 # fixes issue from https://github.com/ytdl-org/youtube-dl/pull/11163.
1586 if new_result.get('_type') == 'url':
1587 new_result['_type'] = 'url_transparent'
1588
1589 return self.process_ie_result(
1590 new_result, download=download, extra_info=extra_info)
1591 elif result_type in ('playlist', 'multi_video'):
1592 # Protect from infinite recursion due to recursively nested playlists
1593 # (see https://github.com/ytdl-org/youtube-dl/issues/27833)
1594 webpage_url = ie_result['webpage_url']
1595 if webpage_url in self._playlist_urls:
1596 self.to_screen(
1597 '[download] Skipping already downloaded playlist: %s'
1598 % ie_result.get('title') or ie_result.get('id'))
1599 return
1600
1601 self._playlist_level += 1
1602 self._playlist_urls.add(webpage_url)
1603 self._fill_common_fields(ie_result, False)
1604 self._sanitize_thumbnails(ie_result)
1605 try:
1606 return self.__process_playlist(ie_result, download)
1607 finally:
1608 self._playlist_level -= 1
1609 if not self._playlist_level:
1610 self._playlist_urls.clear()
1611 elif result_type == 'compat_list':
1612 self.report_warning(
1613 'Extractor %s returned a compat_list result. '
1614 'It needs to be updated.' % ie_result.get('extractor'))
1615
1616 def _fixup(r):
1617 self.add_extra_info(r, {
1618 'extractor': ie_result['extractor'],
1619 'webpage_url': ie_result['webpage_url'],
1620 'webpage_url_basename': url_basename(ie_result['webpage_url']),
1621 'webpage_url_domain': get_domain(ie_result['webpage_url']),
1622 'extractor_key': ie_result['extractor_key'],
1623 })
1624 return r
1625 ie_result['entries'] = [
1626 self.process_ie_result(_fixup(r), download, extra_info)
1627 for r in ie_result['entries']
1628 ]
1629 return ie_result
1630 else:
1631 raise Exception('Invalid result type: %s' % result_type)
1632
1633 def _ensure_dir_exists(self, path):
1634 return make_dir(path, self.report_error)
1635
1636 @staticmethod
1637 def _playlist_infodict(ie_result, **kwargs):
1638 return {
1639 **ie_result,
1640 'playlist': ie_result.get('title') or ie_result.get('id'),
1641 'playlist_id': ie_result.get('id'),
1642 'playlist_title': ie_result.get('title'),
1643 'playlist_uploader': ie_result.get('uploader'),
1644 'playlist_uploader_id': ie_result.get('uploader_id'),
1645 'playlist_index': 0,
1646 **kwargs,
1647 }
1648
1649 def __process_playlist(self, ie_result, download):
1650 # We process each entry in the playlist
1651 playlist = ie_result.get('title') or ie_result.get('id')
1652 self.to_screen('[download] Downloading playlist: %s' % playlist)
1653
1654 if 'entries' not in ie_result:
1655 raise EntryNotInPlaylist('There are no entries')
1656
1657 MissingEntry = object()
1658 incomplete_entries = bool(ie_result.get('requested_entries'))
1659 if incomplete_entries:
1660 def fill_missing_entries(entries, indices):
1661 ret = [MissingEntry] * max(indices)
1662 for i, entry in zip(indices, entries):
1663 ret[i - 1] = entry
1664 return ret
1665 ie_result['entries'] = fill_missing_entries(ie_result['entries'], ie_result['requested_entries'])
1666
1667 playlist_results = []
1668
1669 playliststart = self.params.get('playliststart', 1)
1670 playlistend = self.params.get('playlistend')
1671 # For backwards compatibility, interpret -1 as whole list
1672 if playlistend == -1:
1673 playlistend = None
1674
1675 playlistitems_str = self.params.get('playlist_items')
1676 playlistitems = None
1677 if playlistitems_str is not None:
1678 def iter_playlistitems(format):
1679 for string_segment in format.split(','):
1680 if '-' in string_segment:
1681 start, end = string_segment.split('-')
1682 for item in range(int(start), int(end) + 1):
1683 yield int(item)
1684 else:
1685 yield int(string_segment)
1686 playlistitems = orderedSet(iter_playlistitems(playlistitems_str))
1687
1688 ie_entries = ie_result['entries']
1689 if isinstance(ie_entries, list):
1690 playlist_count = len(ie_entries)
1691 msg = f'Collected {playlist_count} videos; downloading %d of them'
1692 ie_result['playlist_count'] = ie_result.get('playlist_count') or playlist_count
1693
1694 def get_entry(i):
1695 return ie_entries[i - 1]
1696 else:
1697 msg = 'Downloading %d videos'
1698 if not isinstance(ie_entries, (PagedList, LazyList)):
1699 ie_entries = LazyList(ie_entries)
1700 elif isinstance(ie_entries, InAdvancePagedList):
1701 if ie_entries._pagesize == 1:
1702 playlist_count = ie_entries._pagecount
1703
1704 def get_entry(i):
1705 return YoutubeDL.__handle_extraction_exceptions(
1706 lambda self, i: ie_entries[i - 1]
1707 )(self, i)
1708
1709 entries, broken = [], False
1710 items = playlistitems if playlistitems is not None else itertools.count(playliststart)
1711 for i in items:
1712 if i == 0:
1713 continue
1714 if playlistitems is None and playlistend is not None and playlistend < i:
1715 break
1716 entry = None
1717 try:
1718 entry = get_entry(i)
1719 if entry is MissingEntry:
1720 raise EntryNotInPlaylist()
1721 except (IndexError, EntryNotInPlaylist):
1722 if incomplete_entries:
1723 raise EntryNotInPlaylist(f'Entry {i} cannot be found')
1724 elif not playlistitems:
1725 break
1726 entries.append(entry)
1727 try:
1728 if entry is not None:
1729 self._match_entry(entry, incomplete=True, silent=True)
1730 except (ExistingVideoReached, RejectedVideoReached):
1731 broken = True
1732 break
1733 ie_result['entries'] = entries
1734
1735 # Save playlist_index before re-ordering
1736 entries = [
1737 ((playlistitems[i - 1] if playlistitems else i + playliststart - 1), entry)
1738 for i, entry in enumerate(entries, 1)
1739 if entry is not None]
1740 n_entries = len(entries)
1741
1742 if not (ie_result.get('playlist_count') or broken or playlistitems or playlistend):
1743 ie_result['playlist_count'] = n_entries
1744
1745 if not playlistitems and (playliststart != 1 or playlistend):
1746 playlistitems = list(range(playliststart, playliststart + n_entries))
1747 ie_result['requested_entries'] = playlistitems
1748
1749 _infojson_written = False
1750 write_playlist_files = self.params.get('allow_playlist_files', True)
1751 if write_playlist_files and self.params.get('list_thumbnails'):
1752 self.list_thumbnails(ie_result)
1753 if write_playlist_files and not self.params.get('simulate'):
1754 ie_copy = self._playlist_infodict(ie_result, n_entries=n_entries)
1755 _infojson_written = self._write_info_json(
1756 'playlist', ie_result, self.prepare_filename(ie_copy, 'pl_infojson'))
1757 if _infojson_written is None:
1758 return
1759 if self._write_description('playlist', ie_result,
1760 self.prepare_filename(ie_copy, 'pl_description')) is None:
1761 return
1762 # TODO: This should be passed to ThumbnailsConvertor if necessary
1763 self._write_thumbnails('playlist', ie_copy, self.prepare_filename(ie_copy, 'pl_thumbnail'))
1764
1765 if self.params.get('playlistreverse', False):
1766 entries = entries[::-1]
1767 if self.params.get('playlistrandom', False):
1768 random.shuffle(entries)
1769
1770 x_forwarded_for = ie_result.get('__x_forwarded_for_ip')
1771
1772 self.to_screen('[%s] playlist %s: %s' % (ie_result['extractor'], playlist, msg % n_entries))
1773 failures = 0
1774 max_failures = self.params.get('skip_playlist_after_errors') or float('inf')
1775 for i, entry_tuple in enumerate(entries, 1):
1776 playlist_index, entry = entry_tuple
1777 if 'playlist-index' in self.params.get('compat_opts', []):
1778 playlist_index = playlistitems[i - 1] if playlistitems else i + playliststart - 1
1779 self.to_screen('[download] Downloading video %s of %s' % (i, n_entries))
1780 # This __x_forwarded_for_ip thing is a bit ugly but requires
1781 # minimal changes
1782 if x_forwarded_for:
1783 entry['__x_forwarded_for_ip'] = x_forwarded_for
1784 extra = {
1785 'n_entries': n_entries,
1786 '_last_playlist_index': max(playlistitems) if playlistitems else (playlistend or n_entries),
1787 'playlist_count': ie_result.get('playlist_count'),
1788 'playlist_index': playlist_index,
1789 'playlist_autonumber': i,
1790 'playlist': playlist,
1791 'playlist_id': ie_result.get('id'),
1792 'playlist_title': ie_result.get('title'),
1793 'playlist_uploader': ie_result.get('uploader'),
1794 'playlist_uploader_id': ie_result.get('uploader_id'),
1795 'extractor': ie_result['extractor'],
1796 'webpage_url': ie_result['webpage_url'],
1797 'webpage_url_basename': url_basename(ie_result['webpage_url']),
1798 'webpage_url_domain': get_domain(ie_result['webpage_url']),
1799 'extractor_key': ie_result['extractor_key'],
1800 }
1801
1802 if self._match_entry(entry, incomplete=True) is not None:
1803 continue
1804
1805 entry_result = self.__process_iterable_entry(entry, download, extra)
1806 if not entry_result:
1807 failures += 1
1808 if failures >= max_failures:
1809 self.report_error(
1810 'Skipping the remaining entries in playlist "%s" since %d items failed extraction' % (playlist, failures))
1811 break
1812 playlist_results.append(entry_result)
1813 ie_result['entries'] = playlist_results
1814
1815 # Write the updated info to json
1816 if _infojson_written is True and self._write_info_json(
1817 'updated playlist', ie_result,
1818 self.prepare_filename(ie_copy, 'pl_infojson'), overwrite=True) is None:
1819 return
1820
1821 ie_result = self.run_all_pps('playlist', ie_result)
1822 self.to_screen(f'[download] Finished downloading playlist: {playlist}')
1823 return ie_result
1824
1825 @__handle_extraction_exceptions
1826 def __process_iterable_entry(self, entry, download, extra_info):
1827 return self.process_ie_result(
1828 entry, download=download, extra_info=extra_info)
1829
1830 def _build_format_filter(self, filter_spec):
1831 " Returns a function to filter the formats according to the filter_spec "
1832
1833 OPERATORS = {
1834 '<': operator.lt,
1835 '<=': operator.le,
1836 '>': operator.gt,
1837 '>=': operator.ge,
1838 '=': operator.eq,
1839 '!=': operator.ne,
1840 }
1841 operator_rex = re.compile(r'''(?x)\s*
1842 (?P<key>width|height|tbr|abr|vbr|asr|filesize|filesize_approx|fps)\s*
1843 (?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
1844 (?P<value>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?)\s*
1845 ''' % '|'.join(map(re.escape, OPERATORS.keys())))
1846 m = operator_rex.fullmatch(filter_spec)
1847 if m:
1848 try:
1849 comparison_value = int(m.group('value'))
1850 except ValueError:
1851 comparison_value = parse_filesize(m.group('value'))
1852 if comparison_value is None:
1853 comparison_value = parse_filesize(m.group('value') + 'B')
1854 if comparison_value is None:
1855 raise ValueError(
1856 'Invalid value %r in format specification %r' % (
1857 m.group('value'), filter_spec))
1858 op = OPERATORS[m.group('op')]
1859
1860 if not m:
1861 STR_OPERATORS = {
1862 '=': operator.eq,
1863 '^=': lambda attr, value: attr.startswith(value),
1864 '$=': lambda attr, value: attr.endswith(value),
1865 '*=': lambda attr, value: value in attr,
1866 '~=': lambda attr, value: value.search(attr) is not None
1867 }
1868 str_operator_rex = re.compile(r'''(?x)\s*
1869 (?P<key>[a-zA-Z0-9._-]+)\s*
1870 (?P<negation>!\s*)?(?P<op>%s)\s*(?P<none_inclusive>\?\s*)?
1871 (?P<quote>["'])?
1872 (?P<value>(?(quote)(?:(?!(?P=quote))[^\\]|\\.)+|[\w.-]+))
1873 (?(quote)(?P=quote))\s*
1874 ''' % '|'.join(map(re.escape, STR_OPERATORS.keys())))
1875 m = str_operator_rex.fullmatch(filter_spec)
1876 if m:
1877 if m.group('op') == '~=':
1878 comparison_value = re.compile(m.group('value'))
1879 else:
1880 comparison_value = re.sub(r'''\\([\\"'])''', r'\1', m.group('value'))
1881 str_op = STR_OPERATORS[m.group('op')]
1882 if m.group('negation'):
1883 op = lambda attr, value: not str_op(attr, value)
1884 else:
1885 op = str_op
1886
1887 if not m:
1888 raise SyntaxError('Invalid filter specification %r' % filter_spec)
1889
1890 def _filter(f):
1891 actual_value = f.get(m.group('key'))
1892 if actual_value is None:
1893 return m.group('none_inclusive')
1894 return op(actual_value, comparison_value)
1895 return _filter
1896
1897 def _check_formats(self, formats):
1898 for f in formats:
1899 self.to_screen('[info] Testing format %s' % f['format_id'])
1900 path = self.get_output_path('temp')
1901 if not self._ensure_dir_exists(f'{path}/'):
1902 continue
1903 temp_file = tempfile.NamedTemporaryFile(suffix='.tmp', delete=False, dir=path or None)
1904 temp_file.close()
1905 try:
1906 success, _ = self.dl(temp_file.name, f, test=True)
1907 except (DownloadError, IOError, OSError, ValueError) + network_exceptions:
1908 success = False
1909 finally:
1910 if os.path.exists(temp_file.name):
1911 try:
1912 os.remove(temp_file.name)
1913 except OSError:
1914 self.report_warning('Unable to delete temporary file "%s"' % temp_file.name)
1915 if success:
1916 yield f
1917 else:
1918 self.to_screen('[info] Unable to download format %s. Skipping...' % f['format_id'])
1919
1920 def _default_format_spec(self, info_dict, download=True):
1921
1922 def can_merge():
1923 merger = FFmpegMergerPP(self)
1924 return merger.available and merger.can_merge()
1925
1926 prefer_best = (
1927 not self.params.get('simulate')
1928 and download
1929 and (
1930 not can_merge()
1931 or info_dict.get('is_live', False)
1932 or self.outtmpl_dict['default'] == '-'))
1933 compat = (
1934 prefer_best
1935 or self.params.get('allow_multiple_audio_streams', False)
1936 or 'format-spec' in self.params.get('compat_opts', []))
1937
1938 return (
1939 'best/bestvideo+bestaudio' if prefer_best
1940 else 'bestvideo*+bestaudio/best' if not compat
1941 else 'bestvideo+bestaudio/best')
1942
1943 def build_format_selector(self, format_spec):
1944 def syntax_error(note, start):
1945 message = (
1946 'Invalid format specification: '
1947 '{0}\n\t{1}\n\t{2}^'.format(note, format_spec, ' ' * start[1]))
1948 return SyntaxError(message)
1949
1950 PICKFIRST = 'PICKFIRST'
1951 MERGE = 'MERGE'
1952 SINGLE = 'SINGLE'
1953 GROUP = 'GROUP'
1954 FormatSelector = collections.namedtuple('FormatSelector', ['type', 'selector', 'filters'])
1955
1956 allow_multiple_streams = {'audio': self.params.get('allow_multiple_audio_streams', False),
1957 'video': self.params.get('allow_multiple_video_streams', False)}
1958
1959 check_formats = self.params.get('check_formats') == 'selected'
1960
1961 def _parse_filter(tokens):
1962 filter_parts = []
1963 for type, string, start, _, _ in tokens:
1964 if type == tokenize.OP and string == ']':
1965 return ''.join(filter_parts)
1966 else:
1967 filter_parts.append(string)
1968
1969 def _remove_unused_ops(tokens):
1970 # Remove operators that we don't use and join them with the surrounding strings
1971 # for example: 'mp4' '-' 'baseline' '-' '16x9' is converted to 'mp4-baseline-16x9'
1972 ALLOWED_OPS = ('/', '+', ',', '(', ')')
1973 last_string, last_start, last_end, last_line = None, None, None, None
1974 for type, string, start, end, line in tokens:
1975 if type == tokenize.OP and string == '[':
1976 if last_string:
1977 yield tokenize.NAME, last_string, last_start, last_end, last_line
1978 last_string = None
1979 yield type, string, start, end, line
1980 # everything inside brackets will be handled by _parse_filter
1981 for type, string, start, end, line in tokens:
1982 yield type, string, start, end, line
1983 if type == tokenize.OP and string == ']':
1984 break
1985 elif type == tokenize.OP and string in ALLOWED_OPS:
1986 if last_string:
1987 yield tokenize.NAME, last_string, last_start, last_end, last_line
1988 last_string = None
1989 yield type, string, start, end, line
1990 elif type in [tokenize.NAME, tokenize.NUMBER, tokenize.OP]:
1991 if not last_string:
1992 last_string = string
1993 last_start = start
1994 last_end = end
1995 else:
1996 last_string += string
1997 if last_string:
1998 yield tokenize.NAME, last_string, last_start, last_end, last_line
1999
2000 def _parse_format_selection(tokens, inside_merge=False, inside_choice=False, inside_group=False):
2001 selectors = []
2002 current_selector = None
2003 for type, string, start, _, _ in tokens:
2004 # ENCODING is only defined in python 3.x
2005 if type == getattr(tokenize, 'ENCODING', None):
2006 continue
2007 elif type in [tokenize.NAME, tokenize.NUMBER]:
2008 current_selector = FormatSelector(SINGLE, string, [])
2009 elif type == tokenize.OP:
2010 if string == ')':
2011 if not inside_group:
2012 # ')' will be handled by the parentheses group
2013 tokens.restore_last_token()
2014 break
2015 elif inside_merge and string in ['/', ',']:
2016 tokens.restore_last_token()
2017 break
2018 elif inside_choice and string == ',':
2019 tokens.restore_last_token()
2020 break
2021 elif string == ',':
2022 if not current_selector:
2023 raise syntax_error('"," must follow a format selector', start)
2024 selectors.append(current_selector)
2025 current_selector = None
2026 elif string == '/':
2027 if not current_selector:
2028 raise syntax_error('"/" must follow a format selector', start)
2029 first_choice = current_selector
2030 second_choice = _parse_format_selection(tokens, inside_choice=True)
2031 current_selector = FormatSelector(PICKFIRST, (first_choice, second_choice), [])
2032 elif string == '[':
2033 if not current_selector:
2034 current_selector = FormatSelector(SINGLE, 'best', [])
2035 format_filter = _parse_filter(tokens)
2036 current_selector.filters.append(format_filter)
2037 elif string == '(':
2038 if current_selector:
2039 raise syntax_error('Unexpected "("', start)
2040 group = _parse_format_selection(tokens, inside_group=True)
2041 current_selector = FormatSelector(GROUP, group, [])
2042 elif string == '+':
2043 if not current_selector:
2044 raise syntax_error('Unexpected "+"', start)
2045 selector_1 = current_selector
2046 selector_2 = _parse_format_selection(tokens, inside_merge=True)
2047 if not selector_2:
2048 raise syntax_error('Expected a selector', start)
2049 current_selector = FormatSelector(MERGE, (selector_1, selector_2), [])
2050 else:
2051 raise syntax_error('Operator not recognized: "{0}"'.format(string), start)
2052 elif type == tokenize.ENDMARKER:
2053 break
2054 if current_selector:
2055 selectors.append(current_selector)
2056 return selectors
2057
2058 def _merge(formats_pair):
2059 format_1, format_2 = formats_pair
2060
2061 formats_info = []
2062 formats_info.extend(format_1.get('requested_formats', (format_1,)))
2063 formats_info.extend(format_2.get('requested_formats', (format_2,)))
2064
2065 if not allow_multiple_streams['video'] or not allow_multiple_streams['audio']:
2066 get_no_more = {'video': False, 'audio': False}
2067 for (i, fmt_info) in enumerate(formats_info):
2068 if fmt_info.get('acodec') == fmt_info.get('vcodec') == 'none':
2069 formats_info.pop(i)
2070 continue
2071 for aud_vid in ['audio', 'video']:
2072 if not allow_multiple_streams[aud_vid] and fmt_info.get(aud_vid[0] + 'codec') != 'none':
2073 if get_no_more[aud_vid]:
2074 formats_info.pop(i)
2075 break
2076 get_no_more[aud_vid] = True
2077
2078 if len(formats_info) == 1:
2079 return formats_info[0]
2080
2081 video_fmts = [fmt_info for fmt_info in formats_info if fmt_info.get('vcodec') != 'none']
2082 audio_fmts = [fmt_info for fmt_info in formats_info if fmt_info.get('acodec') != 'none']
2083
2084 the_only_video = video_fmts[0] if len(video_fmts) == 1 else None
2085 the_only_audio = audio_fmts[0] if len(audio_fmts) == 1 else None
2086
2087 output_ext = self.params.get('merge_output_format')
2088 if not output_ext:
2089 if the_only_video:
2090 output_ext = the_only_video['ext']
2091 elif the_only_audio and not video_fmts:
2092 output_ext = the_only_audio['ext']
2093 else:
2094 output_ext = 'mkv'
2095
2096 filtered = lambda *keys: filter(None, (traverse_obj(fmt, *keys) for fmt in formats_info))
2097
2098 new_dict = {
2099 'requested_formats': formats_info,
2100 'format': '+'.join(filtered('format')),
2101 'format_id': '+'.join(filtered('format_id')),
2102 'ext': output_ext,
2103 'protocol': '+'.join(map(determine_protocol, formats_info)),
2104 'language': '+'.join(orderedSet(filtered('language'))) or None,
2105 'format_note': '+'.join(orderedSet(filtered('format_note'))) or None,
2106 'filesize_approx': sum(filtered('filesize', 'filesize_approx')) or None,
2107 'tbr': sum(filtered('tbr', 'vbr', 'abr')),
2108 }
2109
2110 if the_only_video:
2111 new_dict.update({
2112 'width': the_only_video.get('width'),
2113 'height': the_only_video.get('height'),
2114 'resolution': the_only_video.get('resolution') or self.format_resolution(the_only_video),
2115 'fps': the_only_video.get('fps'),
2116 'dynamic_range': the_only_video.get('dynamic_range'),
2117 'vcodec': the_only_video.get('vcodec'),
2118 'vbr': the_only_video.get('vbr'),
2119 'stretched_ratio': the_only_video.get('stretched_ratio'),
2120 })
2121
2122 if the_only_audio:
2123 new_dict.update({
2124 'acodec': the_only_audio.get('acodec'),
2125 'abr': the_only_audio.get('abr'),
2126 'asr': the_only_audio.get('asr'),
2127 })
2128
2129 return new_dict
2130
2131 def _check_formats(formats):
2132 if not check_formats:
2133 yield from formats
2134 return
2135 yield from self._check_formats(formats)
2136
2137 def _build_selector_function(selector):
2138 if isinstance(selector, list): # ,
2139 fs = [_build_selector_function(s) for s in selector]
2140
2141 def selector_function(ctx):
2142 for f in fs:
2143 yield from f(ctx)
2144 return selector_function
2145
2146 elif selector.type == GROUP: # ()
2147 selector_function = _build_selector_function(selector.selector)
2148
2149 elif selector.type == PICKFIRST: # /
2150 fs = [_build_selector_function(s) for s in selector.selector]
2151
2152 def selector_function(ctx):
2153 for f in fs:
2154 picked_formats = list(f(ctx))
2155 if picked_formats:
2156 return picked_formats
2157 return []
2158
2159 elif selector.type == MERGE: # +
2160 selector_1, selector_2 = map(_build_selector_function, selector.selector)
2161
2162 def selector_function(ctx):
2163 for pair in itertools.product(selector_1(ctx), selector_2(ctx)):
2164 yield _merge(pair)
2165
2166 elif selector.type == SINGLE: # atom
2167 format_spec = selector.selector or 'best'
2168
2169 # TODO: Add allvideo, allaudio etc by generalizing the code with best/worst selector
2170 if format_spec == 'all':
2171 def selector_function(ctx):
2172 yield from _check_formats(ctx['formats'][::-1])
2173 elif format_spec == 'mergeall':
2174 def selector_function(ctx):
2175 formats = list(_check_formats(
2176 f for f in ctx['formats'] if f.get('vcodec') != 'none' or f.get('acodec') != 'none'))
2177 if not formats:
2178 return
2179 merged_format = formats[-1]
2180 for f in formats[-2::-1]:
2181 merged_format = _merge((merged_format, f))
2182 yield merged_format
2183
2184 else:
2185 format_fallback, seperate_fallback, format_reverse, format_idx = False, None, True, 1
2186 mobj = re.match(
2187 r'(?P<bw>best|worst|b|w)(?P<type>video|audio|v|a)?(?P<mod>\*)?(?:\.(?P<n>[1-9]\d*))?$',
2188 format_spec)
2189 if mobj is not None:
2190 format_idx = int_or_none(mobj.group('n'), default=1)
2191 format_reverse = mobj.group('bw')[0] == 'b'
2192 format_type = (mobj.group('type') or [None])[0]
2193 not_format_type = {'v': 'a', 'a': 'v'}.get(format_type)
2194 format_modified = mobj.group('mod') is not None
2195
2196 format_fallback = not format_type and not format_modified # for b, w
2197 _filter_f = (
2198 (lambda f: f.get('%scodec' % format_type) != 'none')
2199 if format_type and format_modified # bv*, ba*, wv*, wa*
2200 else (lambda f: f.get('%scodec' % not_format_type) == 'none')
2201 if format_type # bv, ba, wv, wa
2202 else (lambda f: f.get('vcodec') != 'none' and f.get('acodec') != 'none')
2203 if not format_modified # b, w
2204 else lambda f: True) # b*, w*
2205 filter_f = lambda f: _filter_f(f) and (
2206 f.get('vcodec') != 'none' or f.get('acodec') != 'none')
2207 else:
2208 if format_spec in self._format_selection_exts['audio']:
2209 filter_f = lambda f: f.get('ext') == format_spec and f.get('acodec') != 'none'
2210 elif format_spec in self._format_selection_exts['video']:
2211 filter_f = lambda f: f.get('ext') == format_spec and f.get('acodec') != 'none' and f.get('vcodec') != 'none'
2212 seperate_fallback = lambda f: f.get('ext') == format_spec and f.get('vcodec') != 'none'
2213 elif format_spec in self._format_selection_exts['storyboards']:
2214 filter_f = lambda f: f.get('ext') == format_spec and f.get('acodec') == 'none' and f.get('vcodec') == 'none'
2215 else:
2216 filter_f = lambda f: f.get('format_id') == format_spec # id
2217
2218 def selector_function(ctx):
2219 formats = list(ctx['formats'])
2220 matches = list(filter(filter_f, formats)) if filter_f is not None else formats
2221 if not matches:
2222 if format_fallback and ctx['incomplete_formats']:
2223 # for extractors with incomplete formats (audio only (soundcloud)
2224 # or video only (imgur)) best/worst will fallback to
2225 # best/worst {video,audio}-only format
2226 matches = formats
2227 elif seperate_fallback and not ctx['has_merged_format']:
2228 # for compatibility with youtube-dl when there is no pre-merged format
2229 matches = list(filter(seperate_fallback, formats))
2230 matches = LazyList(_check_formats(matches[::-1 if format_reverse else 1]))
2231 try:
2232 yield matches[format_idx - 1]
2233 except LazyList.IndexError:
2234 return
2235
2236 filters = [self._build_format_filter(f) for f in selector.filters]
2237
2238 def final_selector(ctx):
2239 ctx_copy = dict(ctx)
2240 for _filter in filters:
2241 ctx_copy['formats'] = list(filter(_filter, ctx_copy['formats']))
2242 return selector_function(ctx_copy)
2243 return final_selector
2244
2245 stream = io.BytesIO(format_spec.encode('utf-8'))
2246 try:
2247 tokens = list(_remove_unused_ops(compat_tokenize_tokenize(stream.readline)))
2248 except tokenize.TokenError:
2249 raise syntax_error('Missing closing/opening brackets or parenthesis', (0, len(format_spec)))
2250
2251 class TokenIterator(object):
2252 def __init__(self, tokens):
2253 self.tokens = tokens
2254 self.counter = 0
2255
2256 def __iter__(self):
2257 return self
2258
2259 def __next__(self):
2260 if self.counter >= len(self.tokens):
2261 raise StopIteration()
2262 value = self.tokens[self.counter]
2263 self.counter += 1
2264 return value
2265
2266 next = __next__
2267
2268 def restore_last_token(self):
2269 self.counter -= 1
2270
2271 parsed_selector = _parse_format_selection(iter(TokenIterator(tokens)))
2272 return _build_selector_function(parsed_selector)
2273
2274 def _calc_headers(self, info_dict):
2275 res = merge_headers(self.params['http_headers'], info_dict.get('http_headers') or {})
2276
2277 cookies = self._calc_cookies(info_dict)
2278 if cookies:
2279 res['Cookie'] = cookies
2280
2281 if 'X-Forwarded-For' not in res:
2282 x_forwarded_for_ip = info_dict.get('__x_forwarded_for_ip')
2283 if x_forwarded_for_ip:
2284 res['X-Forwarded-For'] = x_forwarded_for_ip
2285
2286 return res
2287
2288 def _calc_cookies(self, info_dict):
2289 pr = sanitized_Request(info_dict['url'])
2290 self.cookiejar.add_cookie_header(pr)
2291 return pr.get_header('Cookie')
2292
2293 def _sort_thumbnails(self, thumbnails):
2294 thumbnails.sort(key=lambda t: (
2295 t.get('preference') if t.get('preference') is not None else -1,
2296 t.get('width') if t.get('width') is not None else -1,
2297 t.get('height') if t.get('height') is not None else -1,
2298 t.get('id') if t.get('id') is not None else '',
2299 t.get('url')))
2300
2301 def _sanitize_thumbnails(self, info_dict):
2302 thumbnails = info_dict.get('thumbnails')
2303 if thumbnails is None:
2304 thumbnail = info_dict.get('thumbnail')
2305 if thumbnail:
2306 info_dict['thumbnails'] = thumbnails = [{'url': thumbnail}]
2307 if not thumbnails:
2308 return
2309
2310 def check_thumbnails(thumbnails):
2311 for t in thumbnails:
2312 self.to_screen(f'[info] Testing thumbnail {t["id"]}')
2313 try:
2314 self.urlopen(HEADRequest(t['url']))
2315 except network_exceptions as err:
2316 self.to_screen(f'[info] Unable to connect to thumbnail {t["id"]} URL {t["url"]!r} - {err}. Skipping...')
2317 continue
2318 yield t
2319
2320 self._sort_thumbnails(thumbnails)
2321 for i, t in enumerate(thumbnails):
2322 if t.get('id') is None:
2323 t['id'] = '%d' % i
2324 if t.get('width') and t.get('height'):
2325 t['resolution'] = '%dx%d' % (t['width'], t['height'])
2326 t['url'] = sanitize_url(t['url'])
2327
2328 if self.params.get('check_formats') is True:
2329 info_dict['thumbnails'] = LazyList(check_thumbnails(thumbnails[::-1]), reverse=True)
2330 else:
2331 info_dict['thumbnails'] = thumbnails
2332
2333 def _fill_common_fields(self, info_dict, is_video=True):
2334 # TODO: move sanitization here
2335 if is_video:
2336 # playlists are allowed to lack "title"
2337 info_dict['fulltitle'] = info_dict.get('title')
2338 if 'title' not in info_dict:
2339 raise ExtractorError('Missing "title" field in extractor result',
2340 video_id=info_dict['id'], ie=info_dict['extractor'])
2341 elif not info_dict.get('title'):
2342 self.report_warning('Extractor failed to obtain "title". Creating a generic title instead')
2343 info_dict['title'] = f'{info_dict["extractor"]} video #{info_dict["id"]}'
2344
2345 if info_dict.get('duration') is not None:
2346 info_dict['duration_string'] = formatSeconds(info_dict['duration'])
2347
2348 for ts_key, date_key in (
2349 ('timestamp', 'upload_date'),
2350 ('release_timestamp', 'release_date'),
2351 ('modified_timestamp', 'modified_date'),
2352 ):
2353 if info_dict.get(date_key) is None and info_dict.get(ts_key) is not None:
2354 # Working around out-of-range timestamp values (e.g. negative ones on Windows,
2355 # see http://bugs.python.org/issue1646728)
2356 try:
2357 upload_date = datetime.datetime.utcfromtimestamp(info_dict[ts_key])
2358 info_dict[date_key] = upload_date.strftime('%Y%m%d')
2359 except (ValueError, OverflowError, OSError):
2360 pass
2361
2362 live_keys = ('is_live', 'was_live')
2363 live_status = info_dict.get('live_status')
2364 if live_status is None:
2365 for key in live_keys:
2366 if info_dict.get(key) is False:
2367 continue
2368 if info_dict.get(key):
2369 live_status = key
2370 break
2371 if all(info_dict.get(key) is False for key in live_keys):
2372 live_status = 'not_live'
2373 if live_status:
2374 info_dict['live_status'] = live_status
2375 for key in live_keys:
2376 if info_dict.get(key) is None:
2377 info_dict[key] = (live_status == key)
2378
2379 # Auto generate title fields corresponding to the *_number fields when missing
2380 # in order to always have clean titles. This is very common for TV series.
2381 for field in ('chapter', 'season', 'episode'):
2382 if info_dict.get('%s_number' % field) is not None and not info_dict.get(field):
2383 info_dict[field] = '%s %d' % (field.capitalize(), info_dict['%s_number' % field])
2384
2385 def process_video_result(self, info_dict, download=True):
2386 assert info_dict.get('_type', 'video') == 'video'
2387 self._num_videos += 1
2388
2389 if 'id' not in info_dict:
2390 raise ExtractorError('Missing "id" field in extractor result', ie=info_dict['extractor'])
2391 elif not info_dict.get('id'):
2392 raise ExtractorError('Extractor failed to obtain "id"', ie=info_dict['extractor'])
2393
2394 def report_force_conversion(field, field_not, conversion):
2395 self.report_warning(
2396 '"%s" field is not %s - forcing %s conversion, there is an error in extractor'
2397 % (field, field_not, conversion))
2398
2399 def sanitize_string_field(info, string_field):
2400 field = info.get(string_field)
2401 if field is None or isinstance(field, compat_str):
2402 return
2403 report_force_conversion(string_field, 'a string', 'string')
2404 info[string_field] = compat_str(field)
2405
2406 def sanitize_numeric_fields(info):
2407 for numeric_field in self._NUMERIC_FIELDS:
2408 field = info.get(numeric_field)
2409 if field is None or isinstance(field, compat_numeric_types):
2410 continue
2411 report_force_conversion(numeric_field, 'numeric', 'int')
2412 info[numeric_field] = int_or_none(field)
2413
2414 sanitize_string_field(info_dict, 'id')
2415 sanitize_numeric_fields(info_dict)
2416 if (info_dict.get('duration') or 0) <= 0 and info_dict.pop('duration', None):
2417 self.report_warning('"duration" field is negative, there is an error in extractor')
2418
2419 if 'playlist' not in info_dict:
2420 # It isn't part of a playlist
2421 info_dict['playlist'] = None
2422 info_dict['playlist_index'] = None
2423
2424 self._sanitize_thumbnails(info_dict)
2425
2426 thumbnail = info_dict.get('thumbnail')
2427 thumbnails = info_dict.get('thumbnails')
2428 if thumbnail:
2429 info_dict['thumbnail'] = sanitize_url(thumbnail)
2430 elif thumbnails:
2431 info_dict['thumbnail'] = thumbnails[-1]['url']
2432
2433 if info_dict.get('display_id') is None and 'id' in info_dict:
2434 info_dict['display_id'] = info_dict['id']
2435
2436 self._fill_common_fields(info_dict)
2437
2438 for cc_kind in ('subtitles', 'automatic_captions'):
2439 cc = info_dict.get(cc_kind)
2440 if cc:
2441 for _, subtitle in cc.items():
2442 for subtitle_format in subtitle:
2443 if subtitle_format.get('url'):
2444 subtitle_format['url'] = sanitize_url(subtitle_format['url'])
2445 if subtitle_format.get('ext') is None:
2446 subtitle_format['ext'] = determine_ext(subtitle_format['url']).lower()
2447
2448 automatic_captions = info_dict.get('automatic_captions')
2449 subtitles = info_dict.get('subtitles')
2450
2451 info_dict['requested_subtitles'] = self.process_subtitles(
2452 info_dict['id'], subtitles, automatic_captions)
2453
2454 if info_dict.get('formats') is None:
2455 # There's only one format available
2456 formats = [info_dict]
2457 else:
2458 formats = info_dict['formats']
2459
2460 info_dict['__has_drm'] = any(f.get('has_drm') for f in formats)
2461 if not self.params.get('allow_unplayable_formats'):
2462 formats = [f for f in formats if not f.get('has_drm')]
2463 if info_dict['__has_drm'] and all(
2464 f.get('acodec') == f.get('vcodec') == 'none' for f in formats):
2465 self.report_warning(
2466 'This video is DRM protected and only images are available for download. '
2467 'Use --list-formats to see them')
2468
2469 get_from_start = not info_dict.get('is_live') or bool(self.params.get('live_from_start'))
2470 if not get_from_start:
2471 info_dict['title'] += ' ' + datetime.datetime.now().strftime('%Y-%m-%d %H:%M')
2472 if info_dict.get('is_live') and formats:
2473 formats = [f for f in formats if bool(f.get('is_from_start')) == get_from_start]
2474 if get_from_start and not formats:
2475 self.raise_no_formats(info_dict, msg=(
2476 '--live-from-start is passed, but there are no formats that can be downloaded from the start. '
2477 'If you want to download from the current time, use --no-live-from-start'))
2478
2479 if not formats:
2480 self.raise_no_formats(info_dict)
2481
2482 def is_wellformed(f):
2483 url = f.get('url')
2484 if not url:
2485 self.report_warning(
2486 '"url" field is missing or empty - skipping format, '
2487 'there is an error in extractor')
2488 return False
2489 if isinstance(url, bytes):
2490 sanitize_string_field(f, 'url')
2491 return True
2492
2493 # Filter out malformed formats for better extraction robustness
2494 formats = list(filter(is_wellformed, formats))
2495
2496 formats_dict = {}
2497
2498 # We check that all the formats have the format and format_id fields
2499 for i, format in enumerate(formats):
2500 sanitize_string_field(format, 'format_id')
2501 sanitize_numeric_fields(format)
2502 format['url'] = sanitize_url(format['url'])
2503 if not format.get('format_id'):
2504 format['format_id'] = compat_str(i)
2505 else:
2506 # Sanitize format_id from characters used in format selector expression
2507 format['format_id'] = re.sub(r'[\s,/+\[\]()]', '_', format['format_id'])
2508 format_id = format['format_id']
2509 if format_id not in formats_dict:
2510 formats_dict[format_id] = []
2511 formats_dict[format_id].append(format)
2512
2513 # Make sure all formats have unique format_id
2514 common_exts = set(itertools.chain(*self._format_selection_exts.values()))
2515 for format_id, ambiguous_formats in formats_dict.items():
2516 ambigious_id = len(ambiguous_formats) > 1
2517 for i, format in enumerate(ambiguous_formats):
2518 if ambigious_id:
2519 format['format_id'] = '%s-%d' % (format_id, i)
2520 if format.get('ext') is None:
2521 format['ext'] = determine_ext(format['url']).lower()
2522 # Ensure there is no conflict between id and ext in format selection
2523 # See https://github.com/yt-dlp/yt-dlp/issues/1282
2524 if format['format_id'] != format['ext'] and format['format_id'] in common_exts:
2525 format['format_id'] = 'f%s' % format['format_id']
2526
2527 for i, format in enumerate(formats):
2528 if format.get('format') is None:
2529 format['format'] = '{id} - {res}{note}'.format(
2530 id=format['format_id'],
2531 res=self.format_resolution(format),
2532 note=format_field(format, 'format_note', ' (%s)'),
2533 )
2534 if format.get('protocol') is None:
2535 format['protocol'] = determine_protocol(format)
2536 if format.get('resolution') is None:
2537 format['resolution'] = self.format_resolution(format, default=None)
2538 if format.get('dynamic_range') is None and format.get('vcodec') != 'none':
2539 format['dynamic_range'] = 'SDR'
2540 if (info_dict.get('duration') and format.get('tbr')
2541 and not format.get('filesize') and not format.get('filesize_approx')):
2542 format['filesize_approx'] = info_dict['duration'] * format['tbr'] * (1024 / 8)
2543
2544 # Add HTTP headers, so that external programs can use them from the
2545 # json output
2546 full_format_info = info_dict.copy()
2547 full_format_info.update(format)
2548 format['http_headers'] = self._calc_headers(full_format_info)
2549 # Remove private housekeeping stuff
2550 if '__x_forwarded_for_ip' in info_dict:
2551 del info_dict['__x_forwarded_for_ip']
2552
2553 if self.params.get('check_formats') is True:
2554 formats = LazyList(self._check_formats(formats[::-1]), reverse=True)
2555
2556 if not formats or formats[0] is not info_dict:
2557 # only set the 'formats' fields if the original info_dict list them
2558 # otherwise we end up with a circular reference, the first (and unique)
2559 # element in the 'formats' field in info_dict is info_dict itself,
2560 # which can't be exported to json
2561 info_dict['formats'] = formats
2562
2563 info_dict, _ = self.pre_process(info_dict)
2564
2565 if self._match_entry(info_dict, incomplete=self._format_fields) is not None:
2566 return info_dict
2567
2568 self.post_extract(info_dict)
2569 info_dict, _ = self.pre_process(info_dict, 'after_filter')
2570
2571 # The pre-processors may have modified the formats
2572 formats = info_dict.get('formats', [info_dict])
2573
2574 list_only = self.params.get('simulate') is None and (
2575 self.params.get('list_thumbnails') or self.params.get('listformats') or self.params.get('listsubtitles'))
2576 interactive_format_selection = not list_only and self.format_selector == '-'
2577 if self.params.get('list_thumbnails'):
2578 self.list_thumbnails(info_dict)
2579 if self.params.get('listsubtitles'):
2580 if 'automatic_captions' in info_dict:
2581 self.list_subtitles(
2582 info_dict['id'], automatic_captions, 'automatic captions')
2583 self.list_subtitles(info_dict['id'], subtitles, 'subtitles')
2584 if self.params.get('listformats') or interactive_format_selection:
2585 self.list_formats(info_dict)
2586 if list_only:
2587 # Without this printing, -F --print-json will not work
2588 self.__forced_printings(info_dict, self.prepare_filename(info_dict), incomplete=True)
2589 return
2590
2591 format_selector = self.format_selector
2592 if format_selector is None:
2593 req_format = self._default_format_spec(info_dict, download=download)
2594 self.write_debug('Default format spec: %s' % req_format)
2595 format_selector = self.build_format_selector(req_format)
2596
2597 while True:
2598 if interactive_format_selection:
2599 req_format = input(
2600 self._format_screen('\nEnter format selector: ', self.Styles.EMPHASIS))
2601 try:
2602 format_selector = self.build_format_selector(req_format)
2603 except SyntaxError as err:
2604 self.report_error(err, tb=False, is_error=False)
2605 continue
2606
2607 formats_to_download = list(format_selector({
2608 'formats': formats,
2609 'has_merged_format': any('none' not in (f.get('acodec'), f.get('vcodec')) for f in formats),
2610 'incomplete_formats': (
2611 # All formats are video-only or
2612 all(f.get('vcodec') != 'none' and f.get('acodec') == 'none' for f in formats)
2613 # all formats are audio-only
2614 or all(f.get('vcodec') == 'none' and f.get('acodec') != 'none' for f in formats)),
2615 }))
2616 if interactive_format_selection and not formats_to_download:
2617 self.report_error('Requested format is not available', tb=False, is_error=False)
2618 continue
2619 break
2620
2621 if not formats_to_download:
2622 if not self.params.get('ignore_no_formats_error'):
2623 raise ExtractorError(
2624 'Requested format is not available. Use --list-formats for a list of available formats',
2625 expected=True, video_id=info_dict['id'], ie=info_dict['extractor'])
2626 self.report_warning('Requested format is not available')
2627 # Process what we can, even without any available formats.
2628 formats_to_download = [{}]
2629
2630 best_format = formats_to_download[-1]
2631 if download:
2632 if best_format:
2633 self.to_screen(
2634 f'[info] {info_dict["id"]}: Downloading {len(formats_to_download)} format(s): '
2635 + ', '.join([f['format_id'] for f in formats_to_download]))
2636 max_downloads_reached = False
2637 for i, fmt in enumerate(formats_to_download):
2638 formats_to_download[i] = new_info = self._copy_infodict(info_dict)
2639 new_info.update(fmt)
2640 try:
2641 self.process_info(new_info)
2642 except MaxDownloadsReached:
2643 max_downloads_reached = True
2644 # Remove copied info
2645 for key, val in tuple(new_info.items()):
2646 if info_dict.get(key) == val:
2647 new_info.pop(key)
2648 if max_downloads_reached:
2649 break
2650
2651 write_archive = set(f.get('__write_download_archive', False) for f in formats_to_download)
2652 assert write_archive.issubset({True, False, 'ignore'})
2653 if True in write_archive and False not in write_archive:
2654 self.record_download_archive(info_dict)
2655
2656 info_dict['requested_downloads'] = formats_to_download
2657 info_dict = self.run_all_pps('after_video', info_dict)
2658 if max_downloads_reached:
2659 raise MaxDownloadsReached()
2660
2661 # We update the info dict with the selected best quality format (backwards compatibility)
2662 info_dict.update(best_format)
2663 return info_dict
2664
2665 def process_subtitles(self, video_id, normal_subtitles, automatic_captions):
2666 """Select the requested subtitles and their format"""
2667 available_subs, normal_sub_langs = {}, []
2668 if normal_subtitles and self.params.get('writesubtitles'):
2669 available_subs.update(normal_subtitles)
2670 normal_sub_langs = tuple(normal_subtitles.keys())
2671 if automatic_captions and self.params.get('writeautomaticsub'):
2672 for lang, cap_info in automatic_captions.items():
2673 if lang not in available_subs:
2674 available_subs[lang] = cap_info
2675
2676 if (not self.params.get('writesubtitles') and not
2677 self.params.get('writeautomaticsub') or not
2678 available_subs):
2679 return None
2680
2681 all_sub_langs = tuple(available_subs.keys())
2682 if self.params.get('allsubtitles', False):
2683 requested_langs = all_sub_langs
2684 elif self.params.get('subtitleslangs', False):
2685 # A list is used so that the order of languages will be the same as
2686 # given in subtitleslangs. See https://github.com/yt-dlp/yt-dlp/issues/1041
2687 requested_langs = []
2688 for lang_re in self.params.get('subtitleslangs'):
2689 discard = lang_re[0] == '-'
2690 if discard:
2691 lang_re = lang_re[1:]
2692 if lang_re == 'all':
2693 if discard:
2694 requested_langs = []
2695 else:
2696 requested_langs.extend(all_sub_langs)
2697 continue
2698 current_langs = filter(re.compile(lang_re + '$').match, all_sub_langs)
2699 if discard:
2700 for lang in current_langs:
2701 while lang in requested_langs:
2702 requested_langs.remove(lang)
2703 else:
2704 requested_langs.extend(current_langs)
2705 requested_langs = orderedSet(requested_langs)
2706 elif normal_sub_langs:
2707 requested_langs = ['en'] if 'en' in normal_sub_langs else normal_sub_langs[:1]
2708 else:
2709 requested_langs = ['en'] if 'en' in all_sub_langs else all_sub_langs[:1]
2710 if requested_langs:
2711 self.write_debug('Downloading subtitles: %s' % ', '.join(requested_langs))
2712
2713 formats_query = self.params.get('subtitlesformat', 'best')
2714 formats_preference = formats_query.split('/') if formats_query else []
2715 subs = {}
2716 for lang in requested_langs:
2717 formats = available_subs.get(lang)
2718 if formats is None:
2719 self.report_warning('%s subtitles not available for %s' % (lang, video_id))
2720 continue
2721 for ext in formats_preference:
2722 if ext == 'best':
2723 f = formats[-1]
2724 break
2725 matches = list(filter(lambda f: f['ext'] == ext, formats))
2726 if matches:
2727 f = matches[-1]
2728 break
2729 else:
2730 f = formats[-1]
2731 self.report_warning(
2732 'No subtitle format found matching "%s" for language %s, '
2733 'using %s' % (formats_query, lang, f['ext']))
2734 subs[lang] = f
2735 return subs
2736
2737 def _forceprint(self, key, info_dict):
2738 if info_dict is None:
2739 return
2740 info_copy = info_dict.copy()
2741 info_copy['formats_table'] = self.render_formats_table(info_dict)
2742 info_copy['thumbnails_table'] = self.render_thumbnails_table(info_dict)
2743 info_copy['subtitles_table'] = self.render_subtitles_table(info_dict.get('id'), info_dict.get('subtitles'))
2744 info_copy['automatic_captions_table'] = self.render_subtitles_table(info_dict.get('id'), info_dict.get('automatic_captions'))
2745
2746 def format_tmpl(tmpl):
2747 mobj = re.match(r'\w+(=?)$', tmpl)
2748 if mobj and mobj.group(1):
2749 return f'{tmpl[:-1]} = %({tmpl[:-1]})r'
2750 elif mobj:
2751 return f'%({tmpl})s'
2752 return tmpl
2753
2754 for tmpl in self.params['forceprint'].get(key, []):
2755 self.to_stdout(self.evaluate_outtmpl(format_tmpl(tmpl), info_copy))
2756
2757 for tmpl, file_tmpl in self.params['print_to_file'].get(key, []):
2758 filename = self.prepare_filename(info_dict, outtmpl=file_tmpl)
2759 tmpl = format_tmpl(tmpl)
2760 self.to_screen(f'[info] Writing {tmpl!r} to: {filename}')
2761 if self._ensure_dir_exists(filename):
2762 with io.open(filename, 'a', encoding='utf-8') as f:
2763 f.write(self.evaluate_outtmpl(tmpl, info_copy) + '\n')
2764
2765 def __forced_printings(self, info_dict, filename, incomplete):
2766 def print_mandatory(field, actual_field=None):
2767 if actual_field is None:
2768 actual_field = field
2769 if (self.params.get('force%s' % field, False)
2770 and (not incomplete or info_dict.get(actual_field) is not None)):
2771 self.to_stdout(info_dict[actual_field])
2772
2773 def print_optional(field):
2774 if (self.params.get('force%s' % field, False)
2775 and info_dict.get(field) is not None):
2776 self.to_stdout(info_dict[field])
2777
2778 info_dict = info_dict.copy()
2779 if filename is not None:
2780 info_dict['filename'] = filename
2781 if info_dict.get('requested_formats') is not None:
2782 # For RTMP URLs, also include the playpath
2783 info_dict['urls'] = '\n'.join(f['url'] + f.get('play_path', '') for f in info_dict['requested_formats'])
2784 elif info_dict.get('url'):
2785 info_dict['urls'] = info_dict['url'] + info_dict.get('play_path', '')
2786
2787 if (self.params.get('forcejson')
2788 or self.params['forceprint'].get('video')
2789 or self.params['print_to_file'].get('video')):
2790 self.post_extract(info_dict)
2791 self._forceprint('video', info_dict)
2792
2793 print_mandatory('title')
2794 print_mandatory('id')
2795 print_mandatory('url', 'urls')
2796 print_optional('thumbnail')
2797 print_optional('description')
2798 print_optional('filename')
2799 if self.params.get('forceduration') and info_dict.get('duration') is not None:
2800 self.to_stdout(formatSeconds(info_dict['duration']))
2801 print_mandatory('format')
2802
2803 if self.params.get('forcejson'):
2804 self.to_stdout(json.dumps(self.sanitize_info(info_dict)))
2805
2806 def dl(self, name, info, subtitle=False, test=False):
2807 if not info.get('url'):
2808 self.raise_no_formats(info, True)
2809
2810 if test:
2811 verbose = self.params.get('verbose')
2812 params = {
2813 'test': True,
2814 'quiet': self.params.get('quiet') or not verbose,
2815 'verbose': verbose,
2816 'noprogress': not verbose,
2817 'nopart': True,
2818 'skip_unavailable_fragments': False,
2819 'keep_fragments': False,
2820 'overwrites': True,
2821 '_no_ytdl_file': True,
2822 }
2823 else:
2824 params = self.params
2825 fd = get_suitable_downloader(info, params, to_stdout=(name == '-'))(self, params)
2826 if not test:
2827 for ph in self._progress_hooks:
2828 fd.add_progress_hook(ph)
2829 urls = '", "'.join(
2830 (f['url'].split(',')[0] + ',<data>' if f['url'].startswith('data:') else f['url'])
2831 for f in info.get('requested_formats', []) or [info])
2832 self.write_debug('Invoking downloader on "%s"' % urls)
2833
2834 # Note: Ideally info should be a deep-copied so that hooks cannot modify it.
2835 # But it may contain objects that are not deep-copyable
2836 new_info = self._copy_infodict(info)
2837 if new_info.get('http_headers') is None:
2838 new_info['http_headers'] = self._calc_headers(new_info)
2839 return fd.download(name, new_info, subtitle)
2840
2841 def existing_file(self, filepaths, *, default_overwrite=True):
2842 existing_files = list(filter(os.path.exists, orderedSet(filepaths)))
2843 if existing_files and not self.params.get('overwrites', default_overwrite):
2844 return existing_files[0]
2845
2846 for file in existing_files:
2847 self.report_file_delete(file)
2848 os.remove(file)
2849 return None
2850
2851 def process_info(self, info_dict):
2852 """Process a single resolved IE result. (Modifies it in-place)"""
2853
2854 assert info_dict.get('_type', 'video') == 'video'
2855 original_infodict = info_dict
2856
2857 if 'format' not in info_dict and 'ext' in info_dict:
2858 info_dict['format'] = info_dict['ext']
2859
2860 # This is mostly just for backward compatibility of process_info
2861 # As a side-effect, this allows for format-specific filters
2862 if self._match_entry(info_dict) is not None:
2863 info_dict['__write_download_archive'] = 'ignore'
2864 return
2865
2866 # Does nothing under normal operation - for backward compatibility of process_info
2867 self.post_extract(info_dict)
2868 self._num_downloads += 1
2869
2870 # info_dict['_filename'] needs to be set for backward compatibility
2871 info_dict['_filename'] = full_filename = self.prepare_filename(info_dict, warn=True)
2872 temp_filename = self.prepare_filename(info_dict, 'temp')
2873 files_to_move = {}
2874
2875 # Forced printings
2876 self.__forced_printings(info_dict, full_filename, incomplete=('format' not in info_dict))
2877
2878 if self.params.get('simulate'):
2879 info_dict['__write_download_archive'] = self.params.get('force_write_download_archive')
2880 return
2881
2882 if full_filename is None:
2883 return
2884 if not self._ensure_dir_exists(encodeFilename(full_filename)):
2885 return
2886 if not self._ensure_dir_exists(encodeFilename(temp_filename)):
2887 return
2888
2889 if self._write_description('video', info_dict,
2890 self.prepare_filename(info_dict, 'description')) is None:
2891 return
2892
2893 sub_files = self._write_subtitles(info_dict, temp_filename)
2894 if sub_files is None:
2895 return
2896 files_to_move.update(dict(sub_files))
2897
2898 thumb_files = self._write_thumbnails(
2899 'video', info_dict, temp_filename, self.prepare_filename(info_dict, 'thumbnail'))
2900 if thumb_files is None:
2901 return
2902 files_to_move.update(dict(thumb_files))
2903
2904 infofn = self.prepare_filename(info_dict, 'infojson')
2905 _infojson_written = self._write_info_json('video', info_dict, infofn)
2906 if _infojson_written:
2907 info_dict['infojson_filename'] = infofn
2908 # For backward compatibility, even though it was a private field
2909 info_dict['__infojson_filename'] = infofn
2910 elif _infojson_written is None:
2911 return
2912
2913 # Note: Annotations are deprecated
2914 annofn = None
2915 if self.params.get('writeannotations', False):
2916 annofn = self.prepare_filename(info_dict, 'annotation')
2917 if annofn:
2918 if not self._ensure_dir_exists(encodeFilename(annofn)):
2919 return
2920 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(annofn)):
2921 self.to_screen('[info] Video annotations are already present')
2922 elif not info_dict.get('annotations'):
2923 self.report_warning('There are no annotations to write.')
2924 else:
2925 try:
2926 self.to_screen('[info] Writing video annotations to: ' + annofn)
2927 with io.open(encodeFilename(annofn), 'w', encoding='utf-8') as annofile:
2928 annofile.write(info_dict['annotations'])
2929 except (KeyError, TypeError):
2930 self.report_warning('There are no annotations to write.')
2931 except (OSError, IOError):
2932 self.report_error('Cannot write annotations file: ' + annofn)
2933 return
2934
2935 # Write internet shortcut files
2936 def _write_link_file(link_type):
2937 url = try_get(info_dict['webpage_url'], iri_to_uri)
2938 if not url:
2939 self.report_warning(
2940 f'Cannot write internet shortcut file because the actual URL of "{info_dict["webpage_url"]}" is unknown')
2941 return True
2942 linkfn = replace_extension(self.prepare_filename(info_dict, 'link'), link_type, info_dict.get('ext'))
2943 if not self._ensure_dir_exists(encodeFilename(linkfn)):
2944 return False
2945 if self.params.get('overwrites', True) and os.path.exists(encodeFilename(linkfn)):
2946 self.to_screen(f'[info] Internet shortcut (.{link_type}) is already present')
2947 return True
2948 try:
2949 self.to_screen(f'[info] Writing internet shortcut (.{link_type}) to: {linkfn}')
2950 with io.open(encodeFilename(to_high_limit_path(linkfn)), 'w', encoding='utf-8',
2951 newline='\r\n' if link_type == 'url' else '\n') as linkfile:
2952 template_vars = {'url': url}
2953 if link_type == 'desktop':
2954 template_vars['filename'] = linkfn[:-(len(link_type) + 1)]
2955 linkfile.write(LINK_TEMPLATES[link_type] % template_vars)
2956 except (OSError, IOError):
2957 self.report_error(f'Cannot write internet shortcut {linkfn}')
2958 return False
2959 return True
2960
2961 write_links = {
2962 'url': self.params.get('writeurllink'),
2963 'webloc': self.params.get('writewebloclink'),
2964 'desktop': self.params.get('writedesktoplink'),
2965 }
2966 if self.params.get('writelink'):
2967 link_type = ('webloc' if sys.platform == 'darwin'
2968 else 'desktop' if sys.platform.startswith('linux')
2969 else 'url')
2970 write_links[link_type] = True
2971
2972 if any(should_write and not _write_link_file(link_type)
2973 for link_type, should_write in write_links.items()):
2974 return
2975
2976 def replace_info_dict(new_info):
2977 nonlocal info_dict
2978 if new_info == info_dict:
2979 return
2980 info_dict.clear()
2981 info_dict.update(new_info)
2982
2983 try:
2984 new_info, files_to_move = self.pre_process(info_dict, 'before_dl', files_to_move)
2985 replace_info_dict(new_info)
2986 except PostProcessingError as err:
2987 self.report_error('Preprocessing: %s' % str(err))
2988 return
2989
2990 if self.params.get('skip_download'):
2991 info_dict['filepath'] = temp_filename
2992 info_dict['__finaldir'] = os.path.dirname(os.path.abspath(encodeFilename(full_filename)))
2993 info_dict['__files_to_move'] = files_to_move
2994 replace_info_dict(self.run_pp(MoveFilesAfterDownloadPP(self, False), info_dict))
2995 info_dict['__write_download_archive'] = self.params.get('force_write_download_archive')
2996 else:
2997 # Download
2998 info_dict.setdefault('__postprocessors', [])
2999 try:
3000
3001 def existing_video_file(*filepaths):
3002 ext = info_dict.get('ext')
3003 converted = lambda file: replace_extension(file, self.params.get('final_ext') or ext, ext)
3004 file = self.existing_file(itertools.chain(*zip(map(converted, filepaths), filepaths)),
3005 default_overwrite=False)
3006 if file:
3007 info_dict['ext'] = os.path.splitext(file)[1][1:]
3008 return file
3009
3010 success = True
3011 if info_dict.get('requested_formats') is not None:
3012
3013 def compatible_formats(formats):
3014 # TODO: some formats actually allow this (mkv, webm, ogg, mp4), but not all of them.
3015 video_formats = [format for format in formats if format.get('vcodec') != 'none']
3016 audio_formats = [format for format in formats if format.get('acodec') != 'none']
3017 if len(video_formats) > 2 or len(audio_formats) > 2:
3018 return False
3019
3020 # Check extension
3021 exts = set(format.get('ext') for format in formats)
3022 COMPATIBLE_EXTS = (
3023 set(('mp3', 'mp4', 'm4a', 'm4p', 'm4b', 'm4r', 'm4v', 'ismv', 'isma')),
3024 set(('webm',)),
3025 )
3026 for ext_sets in COMPATIBLE_EXTS:
3027 if ext_sets.issuperset(exts):
3028 return True
3029 # TODO: Check acodec/vcodec
3030 return False
3031
3032 requested_formats = info_dict['requested_formats']
3033 old_ext = info_dict['ext']
3034 if self.params.get('merge_output_format') is None:
3035 if not compatible_formats(requested_formats):
3036 info_dict['ext'] = 'mkv'
3037 self.report_warning(
3038 'Requested formats are incompatible for merge and will be merged into mkv')
3039 if (info_dict['ext'] == 'webm'
3040 and info_dict.get('thumbnails')
3041 # check with type instead of pp_key, __name__, or isinstance
3042 # since we dont want any custom PPs to trigger this
3043 and any(type(pp) == EmbedThumbnailPP for pp in self._pps['post_process'])):
3044 info_dict['ext'] = 'mkv'
3045 self.report_warning(
3046 'webm doesn\'t support embedding a thumbnail, mkv will be used')
3047 new_ext = info_dict['ext']
3048
3049 def correct_ext(filename, ext=new_ext):
3050 if filename == '-':
3051 return filename
3052 filename_real_ext = os.path.splitext(filename)[1][1:]
3053 filename_wo_ext = (
3054 os.path.splitext(filename)[0]
3055 if filename_real_ext in (old_ext, new_ext)
3056 else filename)
3057 return '%s.%s' % (filename_wo_ext, ext)
3058
3059 # Ensure filename always has a correct extension for successful merge
3060 full_filename = correct_ext(full_filename)
3061 temp_filename = correct_ext(temp_filename)
3062 dl_filename = existing_video_file(full_filename, temp_filename)
3063 info_dict['__real_download'] = False
3064
3065 downloaded = []
3066 merger = FFmpegMergerPP(self)
3067
3068 fd = get_suitable_downloader(info_dict, self.params, to_stdout=temp_filename == '-')
3069 if dl_filename is not None:
3070 self.report_file_already_downloaded(dl_filename)
3071 elif fd:
3072 for f in requested_formats if fd != FFmpegFD else []:
3073 f['filepath'] = fname = prepend_extension(
3074 correct_ext(temp_filename, info_dict['ext']),
3075 'f%s' % f['format_id'], info_dict['ext'])
3076 downloaded.append(fname)
3077 info_dict['url'] = '\n'.join(f['url'] for f in requested_formats)
3078 success, real_download = self.dl(temp_filename, info_dict)
3079 info_dict['__real_download'] = real_download
3080 else:
3081 if self.params.get('allow_unplayable_formats'):
3082 self.report_warning(
3083 'You have requested merging of multiple formats '
3084 'while also allowing unplayable formats to be downloaded. '
3085 'The formats won\'t be merged to prevent data corruption.')
3086 elif not merger.available:
3087 msg = 'You have requested merging of multiple formats but ffmpeg is not installed'
3088 if not self.params.get('ignoreerrors'):
3089 self.report_error(f'{msg}. Aborting due to --abort-on-error')
3090 return
3091 self.report_warning(f'{msg}. The formats won\'t be merged')
3092
3093 if temp_filename == '-':
3094 reason = ('using a downloader other than ffmpeg' if FFmpegFD.can_merge_formats(info_dict, self.params)
3095 else 'but the formats are incompatible for simultaneous download' if merger.available
3096 else 'but ffmpeg is not installed')
3097 self.report_warning(
3098 f'You have requested downloading multiple formats to stdout {reason}. '
3099 'The formats will be streamed one after the other')
3100 fname = temp_filename
3101 for f in requested_formats:
3102 new_info = dict(info_dict)
3103 del new_info['requested_formats']
3104 new_info.update(f)
3105 if temp_filename != '-':
3106 fname = prepend_extension(
3107 correct_ext(temp_filename, new_info['ext']),
3108 'f%s' % f['format_id'], new_info['ext'])
3109 if not self._ensure_dir_exists(fname):
3110 return
3111 f['filepath'] = fname
3112 downloaded.append(fname)
3113 partial_success, real_download = self.dl(fname, new_info)
3114 info_dict['__real_download'] = info_dict['__real_download'] or real_download
3115 success = success and partial_success
3116
3117 if downloaded and merger.available and not self.params.get('allow_unplayable_formats'):
3118 info_dict['__postprocessors'].append(merger)
3119 info_dict['__files_to_merge'] = downloaded
3120 # Even if there were no downloads, it is being merged only now
3121 info_dict['__real_download'] = True
3122 else:
3123 for file in downloaded:
3124 files_to_move[file] = None
3125 else:
3126 # Just a single file
3127 dl_filename = existing_video_file(full_filename, temp_filename)
3128 if dl_filename is None or dl_filename == temp_filename:
3129 # dl_filename == temp_filename could mean that the file was partially downloaded with --no-part.
3130 # So we should try to resume the download
3131 success, real_download = self.dl(temp_filename, info_dict)
3132 info_dict['__real_download'] = real_download
3133 else:
3134 self.report_file_already_downloaded(dl_filename)
3135
3136 dl_filename = dl_filename or temp_filename
3137 info_dict['__finaldir'] = os.path.dirname(os.path.abspath(encodeFilename(full_filename)))
3138
3139 except network_exceptions as err:
3140 self.report_error('unable to download video data: %s' % error_to_compat_str(err))
3141 return
3142 except (OSError, IOError) as err:
3143 raise UnavailableVideoError(err)
3144 except (ContentTooShortError, ) as err:
3145 self.report_error('content too short (expected %s bytes and served %s)' % (err.expected, err.downloaded))
3146 return
3147
3148 if success and full_filename != '-':
3149
3150 def fixup():
3151 do_fixup = True
3152 fixup_policy = self.params.get('fixup')
3153 vid = info_dict['id']
3154
3155 if fixup_policy in ('ignore', 'never'):
3156 return
3157 elif fixup_policy == 'warn':
3158 do_fixup = False
3159 elif fixup_policy != 'force':
3160 assert fixup_policy in ('detect_or_warn', None)
3161 if not info_dict.get('__real_download'):
3162 do_fixup = False
3163
3164 def ffmpeg_fixup(cndn, msg, cls):
3165 if not cndn:
3166 return
3167 if not do_fixup:
3168 self.report_warning(f'{vid}: {msg}')
3169 return
3170 pp = cls(self)
3171 if pp.available:
3172 info_dict['__postprocessors'].append(pp)
3173 else:
3174 self.report_warning(f'{vid}: {msg}. Install ffmpeg to fix this automatically')
3175
3176 stretched_ratio = info_dict.get('stretched_ratio')
3177 ffmpeg_fixup(
3178 stretched_ratio not in (1, None),
3179 f'Non-uniform pixel ratio {stretched_ratio}',
3180 FFmpegFixupStretchedPP)
3181
3182 ffmpeg_fixup(
3183 (info_dict.get('requested_formats') is None
3184 and info_dict.get('container') == 'm4a_dash'
3185 and info_dict.get('ext') == 'm4a'),
3186 'writing DASH m4a. Only some players support this container',
3187 FFmpegFixupM4aPP)
3188
3189 downloader = get_suitable_downloader(info_dict, self.params) if 'protocol' in info_dict else None
3190 downloader = downloader.__name__ if downloader else None
3191
3192 if info_dict.get('requested_formats') is None: # Not necessary if doing merger
3193 ffmpeg_fixup(downloader == 'HlsFD',
3194 'Possible MPEG-TS in MP4 container or malformed AAC timestamps',
3195 FFmpegFixupM3u8PP)
3196 ffmpeg_fixup(info_dict.get('is_live') and downloader == 'DashSegmentsFD',
3197 'Possible duplicate MOOV atoms', FFmpegFixupDuplicateMoovPP)
3198
3199 ffmpeg_fixup(downloader == 'WebSocketFragmentFD', 'Malformed timestamps detected', FFmpegFixupTimestampPP)
3200 ffmpeg_fixup(downloader == 'WebSocketFragmentFD', 'Malformed duration detected', FFmpegFixupDurationPP)
3201
3202 fixup()
3203 try:
3204 replace_info_dict(self.post_process(dl_filename, info_dict, files_to_move))
3205 except PostProcessingError as err:
3206 self.report_error('Postprocessing: %s' % str(err))
3207 return
3208 try:
3209 for ph in self._post_hooks:
3210 ph(info_dict['filepath'])
3211 except Exception as err:
3212 self.report_error('post hooks: %s' % str(err))
3213 return
3214 info_dict['__write_download_archive'] = True
3215
3216 if self.params.get('force_write_download_archive'):
3217 info_dict['__write_download_archive'] = True
3218
3219 # Make sure the info_dict was modified in-place
3220 assert info_dict is original_infodict
3221
3222 max_downloads = self.params.get('max_downloads')
3223 if max_downloads is not None and self._num_downloads >= int(max_downloads):
3224 raise MaxDownloadsReached()
3225
3226 def __download_wrapper(self, func):
3227 @functools.wraps(func)
3228 def wrapper(*args, **kwargs):
3229 try:
3230 res = func(*args, **kwargs)
3231 except UnavailableVideoError as e:
3232 self.report_error(e)
3233 except MaxDownloadsReached as e:
3234 self.to_screen(f'[info] {e}')
3235 raise
3236 except DownloadCancelled as e:
3237 self.to_screen(f'[info] {e}')
3238 if not self.params.get('break_per_url'):
3239 raise
3240 else:
3241 if self.params.get('dump_single_json', False):
3242 self.post_extract(res)
3243 self.to_stdout(json.dumps(self.sanitize_info(res)))
3244 return wrapper
3245
3246 def download(self, url_list):
3247 """Download a given list of URLs."""
3248 url_list = variadic(url_list) # Passing a single URL is a common mistake
3249 outtmpl = self.outtmpl_dict['default']
3250 if (len(url_list) > 1
3251 and outtmpl != '-'
3252 and '%' not in outtmpl
3253 and self.params.get('max_downloads') != 1):
3254 raise SameFileError(outtmpl)
3255
3256 for url in url_list:
3257 self.__download_wrapper(self.extract_info)(
3258 url, force_generic_extractor=self.params.get('force_generic_extractor', False))
3259
3260 return self._download_retcode
3261
3262 def download_with_info_file(self, info_filename):
3263 with contextlib.closing(fileinput.FileInput(
3264 [info_filename], mode='r',
3265 openhook=fileinput.hook_encoded('utf-8'))) as f:
3266 # FileInput doesn't have a read method, we can't call json.load
3267 info = self.sanitize_info(json.loads('\n'.join(f)), self.params.get('clean_infojson', True))
3268 try:
3269 self.__download_wrapper(self.process_ie_result)(info, download=True)
3270 except (DownloadError, EntryNotInPlaylist, ReExtractInfo) as e:
3271 if not isinstance(e, EntryNotInPlaylist):
3272 self.to_stderr('\r')
3273 webpage_url = info.get('webpage_url')
3274 if webpage_url is not None:
3275 self.report_warning(f'The info failed to download: {e}; trying with URL {webpage_url}')
3276 return self.download([webpage_url])
3277 else:
3278 raise
3279 return self._download_retcode
3280
3281 @staticmethod
3282 def sanitize_info(info_dict, remove_private_keys=False):
3283 ''' Sanitize the infodict for converting to json '''
3284 if info_dict is None:
3285 return info_dict
3286 info_dict.setdefault('epoch', int(time.time()))
3287 info_dict.setdefault('_type', 'video')
3288
3289 if remove_private_keys:
3290 reject = lambda k, v: v is None or (k.startswith('_') and k != '_type') or k in {
3291 'requested_downloads', 'requested_formats', 'requested_subtitles', 'requested_entries',
3292 'entries', 'filepath', 'infojson_filename', 'original_url', 'playlist_autonumber',
3293 }
3294 else:
3295 reject = lambda k, v: False
3296
3297 def filter_fn(obj):
3298 if isinstance(obj, dict):
3299 return {k: filter_fn(v) for k, v in obj.items() if not reject(k, v)}
3300 elif isinstance(obj, (list, tuple, set, LazyList)):
3301 return list(map(filter_fn, obj))
3302 elif obj is None or isinstance(obj, (str, int, float, bool)):
3303 return obj
3304 else:
3305 return repr(obj)
3306
3307 return filter_fn(info_dict)
3308
3309 @staticmethod
3310 def filter_requested_info(info_dict, actually_filter=True):
3311 ''' Alias of sanitize_info for backward compatibility '''
3312 return YoutubeDL.sanitize_info(info_dict, actually_filter)
3313
3314 @staticmethod
3315 def post_extract(info_dict):
3316 def actual_post_extract(info_dict):
3317 if info_dict.get('_type') in ('playlist', 'multi_video'):
3318 for video_dict in info_dict.get('entries', {}):
3319 actual_post_extract(video_dict or {})
3320 return
3321
3322 post_extractor = info_dict.pop('__post_extractor', None) or (lambda: {})
3323 info_dict.update(post_extractor())
3324
3325 actual_post_extract(info_dict or {})
3326
3327 def run_pp(self, pp, infodict):
3328 files_to_delete = []
3329 if '__files_to_move' not in infodict:
3330 infodict['__files_to_move'] = {}
3331 try:
3332 files_to_delete, infodict = pp.run(infodict)
3333 except PostProcessingError as e:
3334 # Must be True and not 'only_download'
3335 if self.params.get('ignoreerrors') is True:
3336 self.report_error(e)
3337 return infodict
3338 raise
3339
3340 if not files_to_delete:
3341 return infodict
3342 if self.params.get('keepvideo', False):
3343 for f in files_to_delete:
3344 infodict['__files_to_move'].setdefault(f, '')
3345 else:
3346 for old_filename in set(files_to_delete):
3347 self.to_screen('Deleting original file %s (pass -k to keep)' % old_filename)
3348 try:
3349 os.remove(encodeFilename(old_filename))
3350 except (IOError, OSError):
3351 self.report_warning('Unable to remove downloaded original file')
3352 if old_filename in infodict['__files_to_move']:
3353 del infodict['__files_to_move'][old_filename]
3354 return infodict
3355
3356 def run_all_pps(self, key, info, *, additional_pps=None):
3357 self._forceprint(key, info)
3358 for pp in (additional_pps or []) + self._pps[key]:
3359 info = self.run_pp(pp, info)
3360 return info
3361
3362 def pre_process(self, ie_info, key='pre_process', files_to_move=None):
3363 info = dict(ie_info)
3364 info['__files_to_move'] = files_to_move or {}
3365 info = self.run_all_pps(key, info)
3366 return info, info.pop('__files_to_move', None)
3367
3368 def post_process(self, filename, info, files_to_move=None):
3369 """Run all the postprocessors on the given file."""
3370 info['filepath'] = filename
3371 info['__files_to_move'] = files_to_move or {}
3372 info = self.run_all_pps('post_process', info, additional_pps=info.get('__postprocessors'))
3373 info = self.run_pp(MoveFilesAfterDownloadPP(self), info)
3374 del info['__files_to_move']
3375 return self.run_all_pps('after_move', info)
3376
3377 def _make_archive_id(self, info_dict):
3378 video_id = info_dict.get('id')
3379 if not video_id:
3380 return
3381 # Future-proof against any change in case
3382 # and backwards compatibility with prior versions
3383 extractor = info_dict.get('extractor_key') or info_dict.get('ie_key') # key in a playlist
3384 if extractor is None:
3385 url = str_or_none(info_dict.get('url'))
3386 if not url:
3387 return
3388 # Try to find matching extractor for the URL and take its ie_key
3389 for ie_key, ie in self._ies.items():
3390 if ie.suitable(url):
3391 extractor = ie_key
3392 break
3393 else:
3394 return
3395 return '%s %s' % (extractor.lower(), video_id)
3396
3397 def in_download_archive(self, info_dict):
3398 fn = self.params.get('download_archive')
3399 if fn is None:
3400 return False
3401
3402 vid_id = self._make_archive_id(info_dict)
3403 if not vid_id:
3404 return False # Incomplete video information
3405
3406 return vid_id in self.archive
3407
3408 def record_download_archive(self, info_dict):
3409 fn = self.params.get('download_archive')
3410 if fn is None:
3411 return
3412 vid_id = self._make_archive_id(info_dict)
3413 assert vid_id
3414 self.write_debug(f'Adding to archive: {vid_id}')
3415 with locked_file(fn, 'a', encoding='utf-8') as archive_file:
3416 archive_file.write(vid_id + '\n')
3417 self.archive.add(vid_id)
3418
3419 @staticmethod
3420 def format_resolution(format, default='unknown'):
3421 if format.get('vcodec') == 'none' and format.get('acodec') != 'none':
3422 return 'audio only'
3423 if format.get('resolution') is not None:
3424 return format['resolution']
3425 if format.get('width') and format.get('height'):
3426 return '%dx%d' % (format['width'], format['height'])
3427 elif format.get('height'):
3428 return '%sp' % format['height']
3429 elif format.get('width'):
3430 return '%dx?' % format['width']
3431 return default
3432
3433 def _list_format_headers(self, *headers):
3434 if self.params.get('listformats_table', True) is not False:
3435 return [self._format_screen(header, self.Styles.HEADERS) for header in headers]
3436 return headers
3437
3438 def _format_note(self, fdict):
3439 res = ''
3440 if fdict.get('ext') in ['f4f', 'f4m']:
3441 res += '(unsupported)'
3442 if fdict.get('language'):
3443 if res:
3444 res += ' '
3445 res += '[%s]' % fdict['language']
3446 if fdict.get('format_note') is not None:
3447 if res:
3448 res += ' '
3449 res += fdict['format_note']
3450 if fdict.get('tbr') is not None:
3451 if res:
3452 res += ', '
3453 res += '%4dk' % fdict['tbr']
3454 if fdict.get('container') is not None:
3455 if res:
3456 res += ', '
3457 res += '%s container' % fdict['container']
3458 if (fdict.get('vcodec') is not None
3459 and fdict.get('vcodec') != 'none'):
3460 if res:
3461 res += ', '
3462 res += fdict['vcodec']
3463 if fdict.get('vbr') is not None:
3464 res += '@'
3465 elif fdict.get('vbr') is not None and fdict.get('abr') is not None:
3466 res += 'video@'
3467 if fdict.get('vbr') is not None:
3468 res += '%4dk' % fdict['vbr']
3469 if fdict.get('fps') is not None:
3470 if res:
3471 res += ', '
3472 res += '%sfps' % fdict['fps']
3473 if fdict.get('acodec') is not None:
3474 if res:
3475 res += ', '
3476 if fdict['acodec'] == 'none':
3477 res += 'video only'
3478 else:
3479 res += '%-5s' % fdict['acodec']
3480 elif fdict.get('abr') is not None:
3481 if res:
3482 res += ', '
3483 res += 'audio'
3484 if fdict.get('abr') is not None:
3485 res += '@%3dk' % fdict['abr']
3486 if fdict.get('asr') is not None:
3487 res += ' (%5dHz)' % fdict['asr']
3488 if fdict.get('filesize') is not None:
3489 if res:
3490 res += ', '
3491 res += format_bytes(fdict['filesize'])
3492 elif fdict.get('filesize_approx') is not None:
3493 if res:
3494 res += ', '
3495 res += '~' + format_bytes(fdict['filesize_approx'])
3496 return res
3497
3498 def render_formats_table(self, info_dict):
3499 if not info_dict.get('formats') and not info_dict.get('url'):
3500 return None
3501
3502 formats = info_dict.get('formats', [info_dict])
3503 if not self.params.get('listformats_table', True) is not False:
3504 table = [
3505 [
3506 format_field(f, 'format_id'),
3507 format_field(f, 'ext'),
3508 self.format_resolution(f),
3509 self._format_note(f)
3510 ] for f in formats if f.get('preference') is None or f['preference'] >= -1000]
3511 return render_table(['format code', 'extension', 'resolution', 'note'], table, extra_gap=1)
3512
3513 delim = self._format_screen('\u2502', self.Styles.DELIM, '|', test_encoding=True)
3514 table = [
3515 [
3516 self._format_screen(format_field(f, 'format_id'), self.Styles.ID),
3517 format_field(f, 'ext'),
3518 format_field(f, func=self.format_resolution, ignore=('audio only', 'images')),
3519 format_field(f, 'fps', '\t%d'),
3520 format_field(f, 'dynamic_range', '%s', ignore=(None, 'SDR')).replace('HDR', ''),
3521 delim,
3522 format_field(f, 'filesize', ' \t%s', func=format_bytes) + format_field(f, 'filesize_approx', '~\t%s', func=format_bytes),
3523 format_field(f, 'tbr', '\t%dk'),
3524 shorten_protocol_name(f.get('protocol', '')),
3525 delim,
3526 format_field(f, 'vcodec', default='unknown').replace(
3527 'none', 'images' if f.get('acodec') == 'none'
3528 else self._format_screen('audio only', self.Styles.SUPPRESS)),
3529 format_field(f, 'vbr', '\t%dk'),
3530 format_field(f, 'acodec', default='unknown').replace(
3531 'none', '' if f.get('vcodec') == 'none'
3532 else self._format_screen('video only', self.Styles.SUPPRESS)),
3533 format_field(f, 'abr', '\t%dk'),
3534 format_field(f, 'asr', '\t%dHz'),
3535 join_nonempty(
3536 self._format_screen('UNSUPPORTED', 'light red') if f.get('ext') in ('f4f', 'f4m') else None,
3537 format_field(f, 'language', '[%s]'),
3538 join_nonempty(format_field(f, 'format_note'),
3539 format_field(f, 'container', ignore=(None, f.get('ext'))),
3540 delim=', '),
3541 delim=' '),
3542 ] for f in formats if f.get('preference') is None or f['preference'] >= -1000]
3543 header_line = self._list_format_headers(
3544 'ID', 'EXT', 'RESOLUTION', '\tFPS', 'HDR', delim, '\tFILESIZE', '\tTBR', 'PROTO',
3545 delim, 'VCODEC', '\tVBR', 'ACODEC', '\tABR', '\tASR', 'MORE INFO')
3546
3547 return render_table(
3548 header_line, table, hide_empty=True,
3549 delim=self._format_screen('\u2500', self.Styles.DELIM, '-', test_encoding=True))
3550
3551 def render_thumbnails_table(self, info_dict):
3552 thumbnails = list(info_dict.get('thumbnails') or [])
3553 if not thumbnails:
3554 return None
3555 return render_table(
3556 self._list_format_headers('ID', 'Width', 'Height', 'URL'),
3557 [[t.get('id'), t.get('width', 'unknown'), t.get('height', 'unknown'), t['url']] for t in thumbnails])
3558
3559 def render_subtitles_table(self, video_id, subtitles):
3560 def _row(lang, formats):
3561 exts, names = zip(*((f['ext'], f.get('name') or 'unknown') for f in reversed(formats)))
3562 if len(set(names)) == 1:
3563 names = [] if names[0] == 'unknown' else names[:1]
3564 return [lang, ', '.join(names), ', '.join(exts)]
3565
3566 if not subtitles:
3567 return None
3568 return render_table(
3569 self._list_format_headers('Language', 'Name', 'Formats'),
3570 [_row(lang, formats) for lang, formats in subtitles.items()],
3571 hide_empty=True)
3572
3573 def __list_table(self, video_id, name, func, *args):
3574 table = func(*args)
3575 if not table:
3576 self.to_screen(f'{video_id} has no {name}')
3577 return
3578 self.to_screen(f'[info] Available {name} for {video_id}:')
3579 self.to_stdout(table)
3580
3581 def list_formats(self, info_dict):
3582 self.__list_table(info_dict['id'], 'formats', self.render_formats_table, info_dict)
3583
3584 def list_thumbnails(self, info_dict):
3585 self.__list_table(info_dict['id'], 'thumbnails', self.render_thumbnails_table, info_dict)
3586
3587 def list_subtitles(self, video_id, subtitles, name='subtitles'):
3588 self.__list_table(video_id, name, self.render_subtitles_table, video_id, subtitles)
3589
3590 def urlopen(self, req):
3591 """ Start an HTTP download """
3592 if isinstance(req, compat_basestring):
3593 req = sanitized_Request(req)
3594 return self._opener.open(req, timeout=self._socket_timeout)
3595
3596 def print_debug_header(self):
3597 if not self.params.get('verbose'):
3598 return
3599
3600 def get_encoding(stream):
3601 ret = str(getattr(stream, 'encoding', 'missing (%s)' % type(stream).__name__))
3602 if not supports_terminal_sequences(stream):
3603 from .compat import WINDOWS_VT_MODE
3604 ret += ' (No VT)' if WINDOWS_VT_MODE is False else ' (No ANSI)'
3605 return ret
3606
3607 encoding_str = 'Encodings: locale %s, fs %s, out %s, err %s, pref %s' % (
3608 locale.getpreferredencoding(),
3609 sys.getfilesystemencoding(),
3610 get_encoding(self._out_files['screen']), get_encoding(self._out_files['error']),
3611 self.get_encoding())
3612
3613 logger = self.params.get('logger')
3614 if logger:
3615 write_debug = lambda msg: logger.debug(f'[debug] {msg}')
3616 write_debug(encoding_str)
3617 else:
3618 write_string(f'[debug] {encoding_str}\n', encoding=None)
3619 write_debug = lambda msg: self._write_string(f'[debug] {msg}\n')
3620
3621 source = detect_variant()
3622 write_debug(join_nonempty(
3623 'yt-dlp version', __version__,
3624 f'[{RELEASE_GIT_HEAD}]' if RELEASE_GIT_HEAD else '',
3625 '' if source == 'unknown' else f'({source})',
3626 delim=' '))
3627 if not _LAZY_LOADER:
3628 if os.environ.get('YTDLP_NO_LAZY_EXTRACTORS'):
3629 write_debug('Lazy loading extractors is forcibly disabled')
3630 else:
3631 write_debug('Lazy loading extractors is disabled')
3632 if plugin_extractors or plugin_postprocessors:
3633 write_debug('Plugins: %s' % [
3634 '%s%s' % (klass.__name__, '' if klass.__name__ == name else f' as {name}')
3635 for name, klass in itertools.chain(plugin_extractors.items(), plugin_postprocessors.items())])
3636 if self.params.get('compat_opts'):
3637 write_debug('Compatibility options: %s' % ', '.join(self.params.get('compat_opts')))
3638
3639 if source == 'source':
3640 try:
3641 sp = Popen(
3642 ['git', 'rev-parse', '--short', 'HEAD'],
3643 stdout=subprocess.PIPE, stderr=subprocess.PIPE,
3644 cwd=os.path.dirname(os.path.abspath(__file__)))
3645 out, err = sp.communicate_or_kill()
3646 out = out.decode().strip()
3647 if re.match('[0-9a-f]+', out):
3648 write_debug('Git HEAD: %s' % out)
3649 except Exception:
3650 try:
3651 sys.exc_clear()
3652 except Exception:
3653 pass
3654
3655 def python_implementation():
3656 impl_name = platform.python_implementation()
3657 if impl_name == 'PyPy' and hasattr(sys, 'pypy_version_info'):
3658 return impl_name + ' version %d.%d.%d' % sys.pypy_version_info[:3]
3659 return impl_name
3660
3661 write_debug('Python version %s (%s %s) - %s' % (
3662 platform.python_version(),
3663 python_implementation(),
3664 platform.architecture()[0],
3665 platform_name()))
3666
3667 exe_versions, ffmpeg_features = FFmpegPostProcessor.get_versions_and_features(self)
3668 ffmpeg_features = {key for key, val in ffmpeg_features.items() if val}
3669 if ffmpeg_features:
3670 exe_versions['ffmpeg'] += ' (%s)' % ','.join(ffmpeg_features)
3671
3672 exe_versions['rtmpdump'] = rtmpdump_version()
3673 exe_versions['phantomjs'] = PhantomJSwrapper._version()
3674 exe_str = ', '.join(
3675 f'{exe} {v}' for exe, v in sorted(exe_versions.items()) if v
3676 ) or 'none'
3677 write_debug('exe versions: %s' % exe_str)
3678
3679 from .downloader.websocket import has_websockets
3680 from .postprocessor.embedthumbnail import has_mutagen
3681 from .cookies import SQLITE_AVAILABLE, SECRETSTORAGE_AVAILABLE
3682
3683 lib_str = join_nonempty(
3684 compat_brotli and compat_brotli.__name__,
3685 has_certifi and 'certifi',
3686 compat_pycrypto_AES and compat_pycrypto_AES.__name__.split('.')[0],
3687 SECRETSTORAGE_AVAILABLE and 'secretstorage',
3688 has_mutagen and 'mutagen',
3689 SQLITE_AVAILABLE and 'sqlite',
3690 has_websockets and 'websockets',
3691 delim=', ') or 'none'
3692 write_debug('Optional libraries: %s' % lib_str)
3693
3694 self._setup_opener()
3695 proxy_map = {}
3696 for handler in self._opener.handlers:
3697 if hasattr(handler, 'proxies'):
3698 proxy_map.update(handler.proxies)
3699 write_debug(f'Proxy map: {proxy_map}')
3700
3701 # Not implemented
3702 if False and self.params.get('call_home'):
3703 ipaddr = self.urlopen('https://yt-dl.org/ip').read().decode('utf-8')
3704 write_debug('Public IP address: %s' % ipaddr)
3705 latest_version = self.urlopen(
3706 'https://yt-dl.org/latest/version').read().decode('utf-8')
3707 if version_tuple(latest_version) > version_tuple(__version__):
3708 self.report_warning(
3709 'You are using an outdated version (newest version: %s)! '
3710 'See https://yt-dl.org/update if you need help updating.' %
3711 latest_version)
3712
3713 def _setup_opener(self):
3714 if hasattr(self, '_opener'):
3715 return
3716 timeout_val = self.params.get('socket_timeout')
3717 self._socket_timeout = 20 if timeout_val is None else float(timeout_val)
3718
3719 opts_cookiesfrombrowser = self.params.get('cookiesfrombrowser')
3720 opts_cookiefile = self.params.get('cookiefile')
3721 opts_proxy = self.params.get('proxy')
3722
3723 self.cookiejar = load_cookies(opts_cookiefile, opts_cookiesfrombrowser, self)
3724
3725 cookie_processor = YoutubeDLCookieProcessor(self.cookiejar)
3726 if opts_proxy is not None:
3727 if opts_proxy == '':
3728 proxies = {}
3729 else:
3730 proxies = {'http': opts_proxy, 'https': opts_proxy}
3731 else:
3732 proxies = compat_urllib_request.getproxies()
3733 # Set HTTPS proxy to HTTP one if given (https://github.com/ytdl-org/youtube-dl/issues/805)
3734 if 'http' in proxies and 'https' not in proxies:
3735 proxies['https'] = proxies['http']
3736 proxy_handler = PerRequestProxyHandler(proxies)
3737
3738 debuglevel = 1 if self.params.get('debug_printtraffic') else 0
3739 https_handler = make_HTTPS_handler(self.params, debuglevel=debuglevel)
3740 ydlh = YoutubeDLHandler(self.params, debuglevel=debuglevel)
3741 redirect_handler = YoutubeDLRedirectHandler()
3742 data_handler = compat_urllib_request_DataHandler()
3743
3744 # When passing our own FileHandler instance, build_opener won't add the
3745 # default FileHandler and allows us to disable the file protocol, which
3746 # can be used for malicious purposes (see
3747 # https://github.com/ytdl-org/youtube-dl/issues/8227)
3748 file_handler = compat_urllib_request.FileHandler()
3749
3750 def file_open(*args, **kwargs):
3751 raise compat_urllib_error.URLError('file:// scheme is explicitly disabled in yt-dlp for security reasons')
3752 file_handler.file_open = file_open
3753
3754 opener = compat_urllib_request.build_opener(
3755 proxy_handler, https_handler, cookie_processor, ydlh, redirect_handler, data_handler, file_handler)
3756
3757 # Delete the default user-agent header, which would otherwise apply in
3758 # cases where our custom HTTP handler doesn't come into play
3759 # (See https://github.com/ytdl-org/youtube-dl/issues/1309 for details)
3760 opener.addheaders = []
3761 self._opener = opener
3762
3763 def encode(self, s):
3764 if isinstance(s, bytes):
3765 return s # Already encoded
3766
3767 try:
3768 return s.encode(self.get_encoding())
3769 except UnicodeEncodeError as err:
3770 err.reason = err.reason + '. Check your system encoding configuration or use the --encoding option.'
3771 raise
3772
3773 def get_encoding(self):
3774 encoding = self.params.get('encoding')
3775 if encoding is None:
3776 encoding = preferredencoding()
3777 return encoding
3778
3779 def _write_info_json(self, label, ie_result, infofn, overwrite=None):
3780 ''' Write infojson and returns True = written, 'exists' = Already exists, False = skip, None = error '''
3781 if overwrite is None:
3782 overwrite = self.params.get('overwrites', True)
3783 if not self.params.get('writeinfojson'):
3784 return False
3785 elif not infofn:
3786 self.write_debug(f'Skipping writing {label} infojson')
3787 return False
3788 elif not self._ensure_dir_exists(infofn):
3789 return None
3790 elif not overwrite and os.path.exists(infofn):
3791 self.to_screen(f'[info] {label.title()} metadata is already present')
3792 return 'exists'
3793
3794 self.to_screen(f'[info] Writing {label} metadata as JSON to: {infofn}')
3795 try:
3796 write_json_file(self.sanitize_info(ie_result, self.params.get('clean_infojson', True)), infofn)
3797 return True
3798 except (OSError, IOError):
3799 self.report_error(f'Cannot write {label} metadata to JSON file {infofn}')
3800 return None
3801
3802 def _write_description(self, label, ie_result, descfn):
3803 ''' Write description and returns True = written, False = skip, None = error '''
3804 if not self.params.get('writedescription'):
3805 return False
3806 elif not descfn:
3807 self.write_debug(f'Skipping writing {label} description')
3808 return False
3809 elif not self._ensure_dir_exists(descfn):
3810 return None
3811 elif not self.params.get('overwrites', True) and os.path.exists(descfn):
3812 self.to_screen(f'[info] {label.title()} description is already present')
3813 elif ie_result.get('description') is None:
3814 self.report_warning(f'There\'s no {label} description to write')
3815 return False
3816 else:
3817 try:
3818 self.to_screen(f'[info] Writing {label} description to: {descfn}')
3819 with io.open(encodeFilename(descfn), 'w', encoding='utf-8') as descfile:
3820 descfile.write(ie_result['description'])
3821 except (OSError, IOError):
3822 self.report_error(f'Cannot write {label} description file {descfn}')
3823 return None
3824 return True
3825
3826 def _write_subtitles(self, info_dict, filename):
3827 ''' Write subtitles to file and return list of (sub_filename, final_sub_filename); or None if error'''
3828 ret = []
3829 subtitles = info_dict.get('requested_subtitles')
3830 if not subtitles or not (self.params.get('writesubtitles') or self.params.get('writeautomaticsub')):
3831 # subtitles download errors are already managed as troubles in relevant IE
3832 # that way it will silently go on when used with unsupporting IE
3833 return ret
3834
3835 sub_filename_base = self.prepare_filename(info_dict, 'subtitle')
3836 if not sub_filename_base:
3837 self.to_screen('[info] Skipping writing video subtitles')
3838 return ret
3839 for sub_lang, sub_info in subtitles.items():
3840 sub_format = sub_info['ext']
3841 sub_filename = subtitles_filename(filename, sub_lang, sub_format, info_dict.get('ext'))
3842 sub_filename_final = subtitles_filename(sub_filename_base, sub_lang, sub_format, info_dict.get('ext'))
3843 existing_sub = self.existing_file((sub_filename_final, sub_filename))
3844 if existing_sub:
3845 self.to_screen(f'[info] Video subtitle {sub_lang}.{sub_format} is already present')
3846 sub_info['filepath'] = existing_sub
3847 ret.append((existing_sub, sub_filename_final))
3848 continue
3849
3850 self.to_screen(f'[info] Writing video subtitles to: {sub_filename}')
3851 if sub_info.get('data') is not None:
3852 try:
3853 # Use newline='' to prevent conversion of newline characters
3854 # See https://github.com/ytdl-org/youtube-dl/issues/10268
3855 with io.open(sub_filename, 'w', encoding='utf-8', newline='') as subfile:
3856 subfile.write(sub_info['data'])
3857 sub_info['filepath'] = sub_filename
3858 ret.append((sub_filename, sub_filename_final))
3859 continue
3860 except (OSError, IOError):
3861 self.report_error(f'Cannot write video subtitles file {sub_filename}')
3862 return None
3863
3864 try:
3865 sub_copy = sub_info.copy()
3866 sub_copy.setdefault('http_headers', info_dict.get('http_headers'))
3867 self.dl(sub_filename, sub_copy, subtitle=True)
3868 sub_info['filepath'] = sub_filename
3869 ret.append((sub_filename, sub_filename_final))
3870 except (DownloadError, ExtractorError, IOError, OSError, ValueError) + network_exceptions as err:
3871 msg = f'Unable to download video subtitles for {sub_lang!r}: {err}'
3872 if self.params.get('ignoreerrors') is not True: # False or 'only_download'
3873 if not self.params.get('ignoreerrors'):
3874 self.report_error(msg)
3875 raise DownloadError(msg)
3876 self.report_warning(msg)
3877 return ret
3878
3879 def _write_thumbnails(self, label, info_dict, filename, thumb_filename_base=None):
3880 ''' Write thumbnails to file and return list of (thumb_filename, final_thumb_filename) '''
3881 write_all = self.params.get('write_all_thumbnails', False)
3882 thumbnails, ret = [], []
3883 if write_all or self.params.get('writethumbnail', False):
3884 thumbnails = info_dict.get('thumbnails') or []
3885 multiple = write_all and len(thumbnails) > 1
3886
3887 if thumb_filename_base is None:
3888 thumb_filename_base = filename
3889 if thumbnails and not thumb_filename_base:
3890 self.write_debug(f'Skipping writing {label} thumbnail')
3891 return ret
3892
3893 for idx, t in list(enumerate(thumbnails))[::-1]:
3894 thumb_ext = (f'{t["id"]}.' if multiple else '') + determine_ext(t['url'], 'jpg')
3895 thumb_display_id = f'{label} thumbnail {t["id"]}'
3896 thumb_filename = replace_extension(filename, thumb_ext, info_dict.get('ext'))
3897 thumb_filename_final = replace_extension(thumb_filename_base, thumb_ext, info_dict.get('ext'))
3898
3899 existing_thumb = self.existing_file((thumb_filename_final, thumb_filename))
3900 if existing_thumb:
3901 self.to_screen('[info] %s is already present' % (
3902 thumb_display_id if multiple else f'{label} thumbnail').capitalize())
3903 t['filepath'] = existing_thumb
3904 ret.append((existing_thumb, thumb_filename_final))
3905 else:
3906 self.to_screen(f'[info] Downloading {thumb_display_id} ...')
3907 try:
3908 uf = self.urlopen(sanitized_Request(t['url'], headers=t.get('http_headers', {})))
3909 self.to_screen(f'[info] Writing {thumb_display_id} to: {thumb_filename}')
3910 with open(encodeFilename(thumb_filename), 'wb') as thumbf:
3911 shutil.copyfileobj(uf, thumbf)
3912 ret.append((thumb_filename, thumb_filename_final))
3913 t['filepath'] = thumb_filename
3914 except network_exceptions as err:
3915 thumbnails.pop(idx)
3916 self.report_warning(f'Unable to download {thumb_display_id}: {err}')
3917 if ret and not write_all:
3918 break
3919 return ret