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