]> jfr.im git - yt-dlp.git/blob - yt_dlp/YoutubeDL.py
[cleanup] Misc fixes
[yt-dlp.git] / yt_dlp / YoutubeDL.py
1 #!/usr/bin/env python3
2 # coding: utf-8
3
4 from __future__ import absolute_import, unicode_literals
5
6 import collections
7 import contextlib
8 import datetime
9 import errno
10 import fileinput
11 import functools
12 import io
13 import itertools
14 import json
15 import locale
16 import operator
17 import os
18 import platform
19 import re
20 import shutil
21 import subprocess
22 import sys
23 import tempfile
24 import time
25 import tokenize
26 import traceback
27 import random
28 import unicodedata
29
30 from enum import Enum
31 from string import ascii_letters
32
33 from .compat import (
34 compat_basestring,
35 compat_brotli,
36 compat_get_terminal_size,
37 compat_kwargs,
38 compat_numeric_types,
39 compat_os_name,
40 compat_pycrypto_AES,
41 compat_shlex_quote,
42 compat_str,
43 compat_tokenize_tokenize,
44 compat_urllib_error,
45 compat_urllib_request,
46 compat_urllib_request_DataHandler,
47 windows_enable_vt_mode,
48 )
49 from .cookies import load_cookies
50 from .utils import (
51 age_restricted,
52 args_to_str,
53 ContentTooShortError,
54 date_from_str,
55 DateRange,
56 DEFAULT_OUTTMPL,
57 determine_ext,
58 determine_protocol,
59 DownloadCancelled,
60 DownloadError,
61 encode_compat_str,
62 encodeFilename,
63 EntryNotInPlaylist,
64 error_to_compat_str,
65 ExistingVideoReached,
66 expand_path,
67 ExtractorError,
68 filter_dict,
69 float_or_none,
70 format_bytes,
71 format_field,
72 format_decimal_suffix,
73 formatSeconds,
74 GeoRestrictedError,
75 get_domain,
76 has_certifi,
77 HEADRequest,
78 InAdvancePagedList,
79 int_or_none,
80 iri_to_uri,
81 ISO3166Utils,
82 join_nonempty,
83 LazyList,
84 LINK_TEMPLATES,
85 locked_file,
86 make_dir,
87 make_HTTPS_handler,
88 MaxDownloadsReached,
89 merge_headers,
90 network_exceptions,
91 NO_DEFAULT,
92 number_of_digits,
93 orderedSet,
94 OUTTMPL_TYPES,
95 PagedList,
96 parse_filesize,
97 PerRequestProxyHandler,
98 platform_name,
99 Popen,
100 POSTPROCESS_WHEN,
101 PostProcessingError,
102 preferredencoding,
103 prepend_extension,
104 ReExtractInfo,
105 register_socks_protocols,
106 RejectedVideoReached,
107 remove_terminal_sequences,
108 render_table,
109 replace_extension,
110 SameFileError,
111 sanitize_filename,
112 sanitize_path,
113 sanitize_url,
114 sanitized_Request,
115 std_headers,
116 STR_FORMAT_RE_TMPL,
117 STR_FORMAT_TYPES,
118 str_or_none,
119 strftime_or_none,
120 subtitles_filename,
121 supports_terminal_sequences,
122 timetuple_from_msec,
123 to_high_limit_path,
124 traverse_obj,
125 try_get,
126 UnavailableVideoError,
127 url_basename,
128 variadic,
129 version_tuple,
130 write_json_file,
131 write_string,
132 YoutubeDLCookieProcessor,
133 YoutubeDLHandler,
134 YoutubeDLRedirectHandler,
135 )
136 from .cache import Cache
137 from .minicurses import format_text
138 from .extractor import (
139 gen_extractor_classes,
140 get_info_extractor,
141 _LAZY_LOADER,
142 _PLUGIN_CLASSES as plugin_extractors
143 )
144 from .extractor.openload import PhantomJSwrapper
145 from .downloader import (
146 FFmpegFD,
147 get_suitable_downloader,
148 shorten_protocol_name
149 )
150 from .downloader.rtmp import rtmpdump_version
151 from .postprocessor import (
152 get_postprocessor,
153 EmbedThumbnailPP,
154 FFmpegFixupDuplicateMoovPP,
155 FFmpegFixupDurationPP,
156 FFmpegFixupM3u8PP,
157 FFmpegFixupM4aPP,
158 FFmpegFixupStretchedPP,
159 FFmpegFixupTimestampPP,
160 FFmpegMergerPP,
161 FFmpegPostProcessor,
162 MoveFilesAfterDownloadPP,
163 _PLUGIN_CLASSES as plugin_postprocessors
164 )
165 from .update import detect_variant
166 from .version import __version__, RELEASE_GIT_HEAD
167
168 if compat_os_name == 'nt':
169 import ctypes
170
171
172 class YoutubeDL(object):
173 """YoutubeDL class.
174
175 YoutubeDL objects are the ones responsible of downloading the
176 actual video file and writing it to disk if the user has requested
177 it, among some other tasks. In most cases there should be one per
178 program. As, given a video URL, the downloader doesn't know how to
179 extract all the needed information, task that InfoExtractors do, it
180 has to pass the URL to one of them.
181
182 For this, YoutubeDL objects have a method that allows
183 InfoExtractors to be registered in a given order. When it is passed
184 a URL, the YoutubeDL object handles it to the first InfoExtractor it
185 finds that reports being able to handle it. The InfoExtractor extracts
186 all the information about the video or videos the URL refers to, and
187 YoutubeDL process the extracted information, possibly using a File
188 Downloader to download the video.
189
190 YoutubeDL objects accept a lot of parameters. In order not to saturate
191 the object constructor with arguments, it receives a dictionary of
192 options instead. These options are available through the params
193 attribute for the InfoExtractors to use. The YoutubeDL also
194 registers itself as the downloader in charge for the InfoExtractors
195 that are added to it, so this is a "mutual registration".
196
197 Available options:
198
199 username: Username for authentication purposes.
200 password: Password for authentication purposes.
201 videopassword: Password for accessing a video.
202 ap_mso: Adobe Pass multiple-system operator identifier.
203 ap_username: Multiple-system operator account username.
204 ap_password: Multiple-system operator account password.
205 usenetrc: Use netrc for authentication instead.
206 verbose: Print additional info to stdout.
207 quiet: Do not print messages to stdout.
208 no_warnings: Do not print out anything for warnings.
209 forceprint: A dict with keys WHEN mapped to a list of templates to
210 print to stdout. The allowed keys are video or any of the
211 items in utils.POSTPROCESS_WHEN.
212 For compatibility, a single list is also accepted
213 print_to_file: A dict with keys WHEN (same as forceprint) mapped to
214 a list of tuples with (template, filename)
215 forceurl: Force printing final URL. (Deprecated)
216 forcetitle: Force printing title. (Deprecated)
217 forceid: Force printing ID. (Deprecated)
218 forcethumbnail: Force printing thumbnail URL. (Deprecated)
219 forcedescription: Force printing description. (Deprecated)
220 forcefilename: Force printing final filename. (Deprecated)
221 forceduration: Force printing duration. (Deprecated)
222 forcejson: Force printing info_dict as JSON.
223 dump_single_json: Force printing the info_dict of the whole playlist
224 (or video) as a single JSON line.
225 force_write_download_archive: Force writing download archive regardless
226 of 'skip_download' or 'simulate'.
227 simulate: Do not download the video files. If unset (or None),
228 simulate only if listsubtitles, listformats or list_thumbnails is used
229 format: Video format code. see "FORMAT SELECTION" for more details.
230 You can also pass a function. The function takes 'ctx' as
231 argument and returns the formats to download.
232 See "build_format_selector" for an implementation
233 allow_unplayable_formats: Allow unplayable formats to be extracted and downloaded.
234 ignore_no_formats_error: Ignore "No video formats" error. Usefull for
235 extracting metadata even if the video is not actually
236 available for download (experimental)
237 format_sort: A list of fields by which to sort the video formats.
238 See "Sorting Formats" for more details.
239 format_sort_force: Force the given format_sort. see "Sorting Formats"
240 for more details.
241 prefer_free_formats: Whether to prefer video formats with free containers
242 over non-free ones of same quality.
243 allow_multiple_video_streams: Allow multiple video streams to be merged
244 into a single file
245 allow_multiple_audio_streams: Allow multiple audio streams to be merged
246 into a single file
247 check_formats Whether to test if the formats are downloadable.
248 Can be True (check all), False (check none),
249 'selected' (check selected formats),
250 or None (check only if requested by extractor)
251 paths: Dictionary of output paths. The allowed keys are 'home'
252 'temp' and the keys of OUTTMPL_TYPES (in utils.py)
253 outtmpl: Dictionary of templates for output names. Allowed keys
254 are 'default' and the keys of OUTTMPL_TYPES (in utils.py).
255 For compatibility with youtube-dl, a single string can also be used
256 outtmpl_na_placeholder: Placeholder for unavailable meta fields.
257 restrictfilenames: Do not allow "&" and spaces in file names
258 trim_file_name: Limit length of filename (extension excluded)
259 windowsfilenames: Force the filenames to be windows compatible
260 ignoreerrors: Do not stop on download/postprocessing errors.
261 Can be 'only_download' to ignore only download errors.
262 Default is 'only_download' for CLI, but False for API
263 skip_playlist_after_errors: Number of allowed failures until the rest of
264 the playlist is skipped
265 force_generic_extractor: Force downloader to use the generic extractor
266 overwrites: Overwrite all video and metadata files if True,
267 overwrite only non-video files if None
268 and don't overwrite any file if False
269 For compatibility with youtube-dl,
270 "nooverwrites" may also be used instead
271 playliststart: Playlist item to start at.
272 playlistend: Playlist item to end at.
273 playlist_items: Specific indices of playlist to download.
274 playlistreverse: Download playlist items in reverse order.
275 playlistrandom: Download playlist items in random order.
276 matchtitle: Download only matching titles.
277 rejecttitle: Reject downloads for matching titles.
278 logger: Log messages to a logging.Logger instance.
279 logtostderr: Log messages to stderr instead of stdout.
280 consoletitle: Display progress in console window's titlebar.
281 writedescription: Write the video description to a .description file
282 writeinfojson: Write the video description to a .info.json file
283 clean_infojson: Remove private fields from the infojson
284 getcomments: Extract video comments. This will not be written to disk
285 unless writeinfojson is also given
286 writeannotations: Write the video annotations to a .annotations.xml file
287 writethumbnail: Write the thumbnail image to a file
288 allow_playlist_files: Whether to write playlists' description, infojson etc
289 also to disk when using the 'write*' options
290 write_all_thumbnails: Write all thumbnail formats to files
291 writelink: Write an internet shortcut file, depending on the
292 current platform (.url/.webloc/.desktop)
293 writeurllink: Write a Windows internet shortcut file (.url)
294 writewebloclink: Write a macOS internet shortcut file (.webloc)
295 writedesktoplink: Write a Linux internet shortcut file (.desktop)
296 writesubtitles: Write the video subtitles to a file
297 writeautomaticsub: Write the automatically generated subtitles to a file
298 allsubtitles: Deprecated - Use subtitleslangs = ['all']
299 Downloads all the subtitles of the video
300 (requires writesubtitles or writeautomaticsub)
301 listsubtitles: Lists all available subtitles for the video
302 subtitlesformat: The format code for subtitles
303 subtitleslangs: List of languages of the subtitles to download (can be regex).
304 The list may contain "all" to refer to all the available
305 subtitles. The language can be prefixed with a "-" to
306 exclude it from the requested languages. Eg: ['all', '-live_chat']
307 keepvideo: Keep the video file after post-processing
308 daterange: A DateRange object, download only if the upload_date is in the range.
309 skip_download: Skip the actual download of the video file
310 cachedir: Location of the cache files in the filesystem.
311 False to disable filesystem cache.
312 noplaylist: Download single video instead of a playlist if in doubt.
313 age_limit: An integer representing the user's age in years.
314 Unsuitable videos for the given age are skipped.
315 min_views: An integer representing the minimum view count the video
316 must have in order to not be skipped.
317 Videos without view count information are always
318 downloaded. None for no limit.
319 max_views: An integer representing the maximum view count.
320 Videos that are more popular than that are not
321 downloaded.
322 Videos without view count information are always
323 downloaded. None for no limit.
324 download_archive: File name of a file where all downloads are recorded.
325 Videos already present in the file are not downloaded
326 again.
327 break_on_existing: Stop the download process after attempting to download a
328 file that is in the archive.
329 break_on_reject: Stop the download process when encountering a video that
330 has been filtered out.
331 break_per_url: Whether break_on_reject and break_on_existing
332 should act on each input URL as opposed to for the entire queue
333 cookiefile: File name where cookies should be read from and dumped to
334 cookiesfrombrowser: A tuple containing the name of the browser, the profile
335 name/pathfrom where cookies are loaded, and the name of the
336 keyring. Eg: ('chrome', ) or ('vivaldi', 'default', 'BASICTEXT')
337 legacyserverconnect: Explicitly allow HTTPS connection to servers that do not
338 support RFC 5746 secure renegotiation
339 nocheckcertificate: Do not verify SSL certificates
340 prefer_insecure: Use HTTP instead of HTTPS to retrieve information.
341 At the moment, this is only supported by YouTube.
342 http_headers: A dictionary of custom headers to be used for all requests
343 proxy: URL of the proxy server to use
344 geo_verification_proxy: URL of the proxy to use for IP address verification
345 on geo-restricted sites.
346 socket_timeout: Time to wait for unresponsive hosts, in seconds
347 bidi_workaround: Work around buggy terminals without bidirectional text
348 support, using fridibi
349 debug_printtraffic:Print out sent and received HTTP traffic
350 include_ads: Download ads as well (deprecated)
351 default_search: Prepend this string if an input url is not valid.
352 'auto' for elaborate guessing
353 encoding: Use this encoding instead of the system-specified.
354 extract_flat: Do not resolve URLs, return the immediate result.
355 Pass in 'in_playlist' to only show this behavior for
356 playlist items.
357 wait_for_video: If given, wait for scheduled streams to become available.
358 The value should be a tuple containing the range
359 (min_secs, max_secs) to wait between retries
360 postprocessors: A list of dictionaries, each with an entry
361 * key: The name of the postprocessor. See
362 yt_dlp/postprocessor/__init__.py for a list.
363 * when: When to run the postprocessor. Allowed values are
364 the entries of utils.POSTPROCESS_WHEN
365 Assumed to be 'post_process' if not given
366 post_hooks: Deprecated - Register a custom postprocessor instead
367 A list of functions that get called as the final step
368 for each video file, after all postprocessors have been
369 called. The filename will be passed as the only argument.
370 progress_hooks: A list of functions that get called on download
371 progress, with a dictionary with the entries
372 * status: One of "downloading", "error", or "finished".
373 Check this first and ignore unknown values.
374 * info_dict: The extracted info_dict
375
376 If status is one of "downloading", or "finished", the
377 following properties may also be present:
378 * filename: The final filename (always present)
379 * tmpfilename: The filename we're currently writing to
380 * downloaded_bytes: Bytes on disk
381 * total_bytes: Size of the whole file, None if unknown
382 * total_bytes_estimate: Guess of the eventual file size,
383 None if unavailable.
384 * elapsed: The number of seconds since download started.
385 * eta: The estimated time in seconds, None if unknown
386 * speed: The download speed in bytes/second, None if
387 unknown
388 * fragment_index: The counter of the currently
389 downloaded video fragment.
390 * fragment_count: The number of fragments (= individual
391 files that will be merged)
392
393 Progress hooks are guaranteed to be called at least once
394 (with status "finished") if the download is successful.
395 postprocessor_hooks: A list of functions that get called on postprocessing
396 progress, with a dictionary with the entries
397 * status: One of "started", "processing", or "finished".
398 Check this first and ignore unknown values.
399 * postprocessor: Name of the postprocessor
400 * info_dict: The extracted info_dict
401
402 Progress hooks are guaranteed to be called at least twice
403 (with status "started" and "finished") if the processing is successful.
404 merge_output_format: Extension to use when merging formats.
405 final_ext: Expected final extension; used to detect when the file was
406 already downloaded and converted
407 fixup: Automatically correct known faults of the file.
408 One of:
409 - "never": do nothing
410 - "warn": only emit a warning
411 - "detect_or_warn": check whether we can do anything
412 about it, warn otherwise (default)
413 source_address: Client-side IP address to bind to.
414 call_home: Boolean, true iff we are allowed to contact the
415 yt-dlp servers for debugging. (BROKEN)
416 sleep_interval_requests: Number of seconds to sleep between requests
417 during extraction
418 sleep_interval: Number of seconds to sleep before each download when
419 used alone or a lower bound of a range for randomized
420 sleep before each download (minimum possible number
421 of seconds to sleep) when used along with
422 max_sleep_interval.
423 max_sleep_interval:Upper bound of a range for randomized sleep before each
424 download (maximum possible number of seconds to sleep).
425 Must only be used along with sleep_interval.
426 Actual sleep time will be a random float from range
427 [sleep_interval; max_sleep_interval].
428 sleep_interval_subtitles: Number of seconds to sleep before each subtitle download
429 listformats: Print an overview of available video formats and exit.
430 list_thumbnails: Print a table of all thumbnails and exit.
431 match_filter: A function that gets called with the info_dict of
432 every video.
433 If it returns a message, the video is ignored.
434 If it returns None, the video is downloaded.
435 match_filter_func in utils.py is one example for this.
436 no_color: Do not emit color codes in output.
437 geo_bypass: Bypass geographic restriction via faking X-Forwarded-For
438 HTTP header
439 geo_bypass_country:
440 Two-letter ISO 3166-2 country code that will be used for
441 explicit geographic restriction bypassing via faking
442 X-Forwarded-For HTTP header
443 geo_bypass_ip_block:
444 IP range in CIDR notation that will be used similarly to
445 geo_bypass_country
446
447 The following options determine which downloader is picked:
448 external_downloader: A dictionary of protocol keys and the executable of the
449 external downloader to use for it. The allowed protocols
450 are default|http|ftp|m3u8|dash|rtsp|rtmp|mms.
451 Set the value to 'native' to use the native downloader
452 hls_prefer_native: Deprecated - Use external_downloader = {'m3u8': 'native'}
453 or {'m3u8': 'ffmpeg'} instead.
454 Use the native HLS downloader instead of ffmpeg/avconv
455 if True, otherwise use ffmpeg/avconv if False, otherwise
456 use downloader suggested by extractor if None.
457 compat_opts: Compatibility options. See "Differences in default behavior".
458 The following options do not work when used through the API:
459 filename, abort-on-error, multistreams, no-live-chat, format-sort
460 no-clean-infojson, no-playlist-metafiles, no-keep-subs, no-attach-info-json.
461 Refer __init__.py for their implementation
462 progress_template: Dictionary of templates for progress outputs.
463 Allowed keys are 'download', 'postprocess',
464 'download-title' (console title) and 'postprocess-title'.
465 The template is mapped on a dictionary with keys 'progress' and 'info'
466
467 The following parameters are not used by YoutubeDL itself, they are used by
468 the downloader (see yt_dlp/downloader/common.py):
469 nopart, updatetime, buffersize, ratelimit, throttledratelimit, min_filesize,
470 max_filesize, test, noresizebuffer, retries, file_access_retries, fragment_retries,
471 continuedl, noprogress, xattr_set_filesize, hls_use_mpegts, http_chunk_size,
472 external_downloader_args, concurrent_fragment_downloads.
473
474 The following options are used by the post processors:
475 prefer_ffmpeg: If False, use avconv instead of ffmpeg if both are available,
476 otherwise prefer ffmpeg. (avconv support is deprecated)
477 ffmpeg_location: Location of the ffmpeg/avconv binary; either the path
478 to the binary or its containing directory.
479 postprocessor_args: A dictionary of postprocessor/executable keys (in lower case)
480 and a list of additional command-line arguments for the
481 postprocessor/executable. The dict can also have "PP+EXE" keys
482 which are used when the given exe is used by the given PP.
483 Use 'default' as the name for arguments to passed to all PP
484 For compatibility with youtube-dl, a single list of args
485 can also be used
486
487 The following options are used by the extractors:
488 extractor_retries: Number of times to retry for known errors
489 dynamic_mpd: Whether to process dynamic DASH manifests (default: True)
490 hls_split_discontinuity: Split HLS playlists to different formats at
491 discontinuities such as ad breaks (default: False)
492 extractor_args: A dictionary of arguments to be passed to the extractors.
493 See "EXTRACTOR ARGUMENTS" for details.
494 Eg: {'youtube': {'skip': ['dash', 'hls']}}
495 mark_watched: Mark videos watched (even with --simulate). Only for YouTube
496 youtube_include_dash_manifest: Deprecated - Use extractor_args instead.
497 If True (default), DASH manifests and related
498 data will be downloaded and processed by extractor.
499 You can reduce network I/O by disabling it if you don't
500 care about DASH. (only for youtube)
501 youtube_include_hls_manifest: Deprecated - Use extractor_args instead.
502 If True (default), HLS manifests and related
503 data will be downloaded and processed by extractor.
504 You can reduce network I/O by disabling it if you don't
505 care about HLS. (only for youtube)
506 """
507
508 _NUMERIC_FIELDS = set((
509 'width', 'height', 'tbr', 'abr', 'asr', 'vbr', 'fps', 'filesize', 'filesize_approx',
510 'timestamp', 'release_timestamp',
511 'duration', 'view_count', 'like_count', 'dislike_count', 'repost_count',
512 'average_rating', 'comment_count', 'age_limit',
513 'start_time', 'end_time',
514 'chapter_number', 'season_number', 'episode_number',
515 'track_number', 'disc_number', 'release_year',
516 ))
517
518 _format_fields = {
519 # NB: Keep in sync with the docstring of extractor/common.py
520 'url', 'manifest_url', 'manifest_stream_number', 'ext', 'format', 'format_id', 'format_note',
521 'width', 'height', 'resolution', 'dynamic_range', 'tbr', 'abr', 'acodec', 'asr',
522 'vbr', 'fps', 'vcodec', 'container', 'filesize', 'filesize_approx',
523 'player_url', 'protocol', 'fragment_base_url', 'fragments', 'is_from_start',
524 'preference', 'language', 'language_preference', 'quality', 'source_preference',
525 'http_headers', 'stretched_ratio', 'no_resume', 'has_drm', 'downloader_options',
526 'page_url', 'app', 'play_path', 'tc_url', 'flash_version', 'rtmp_live', 'rtmp_conn', 'rtmp_protocol', 'rtmp_real_time'
527 }
528 _format_selection_exts = {
529 'audio': {'m4a', 'mp3', 'ogg', 'aac'},
530 'video': {'mp4', 'flv', 'webm', '3gp'},
531 'storyboards': {'mhtml'},
532 }
533
534 def __init__(self, params=None, auto_init=True):
535 """Create a FileDownloader object with the given options.
536 @param auto_init Whether to load the default extractors and print header (if verbose).
537 Set to 'no_verbose_header' to not print the header
538 """
539 if params is None:
540 params = {}
541 self.params = params
542 self._ies = {}
543 self._ies_instances = {}
544 self._pps = {k: [] for k in POSTPROCESS_WHEN}
545 self._printed_messages = set()
546 self._first_webpage_request = True
547 self._post_hooks = []
548 self._progress_hooks = []
549 self._postprocessor_hooks = []
550 self._download_retcode = 0
551 self._num_downloads = 0
552 self._num_videos = 0
553 self._playlist_level = 0
554 self._playlist_urls = set()
555 self.cache = Cache(self)
556
557 windows_enable_vt_mode()
558 self._out_files = {
559 'error': sys.stderr,
560 'print': sys.stderr if self.params.get('logtostderr') else sys.stdout,
561 'console': None if compat_os_name == 'nt' else next(
562 filter(supports_terminal_sequences, (sys.stderr, sys.stdout)), None)
563 }
564 self._out_files['screen'] = sys.stderr if self.params.get('quiet') else self._out_files['print']
565 self._allow_colors = {
566 type_: not self.params.get('no_color') and supports_terminal_sequences(self._out_files[type_])
567 for type_ in ('screen', 'error')
568 }
569
570 if sys.version_info < (3, 6):
571 self.report_warning(
572 'Python version %d.%d is not supported! Please update to Python 3.6 or above' % sys.version_info[:2])
573
574 if self.params.get('allow_unplayable_formats'):
575 self.report_warning(
576 f'You have asked for {self._format_err("UNPLAYABLE", self.Styles.EMPHASIS)} formats to be listed/downloaded. '
577 'This is a developer option intended for debugging. \n'
578 ' If you experience any issues while using this option, '
579 f'{self._format_err("DO NOT", self.Styles.ERROR)} open a bug report')
580
581 def check_deprecated(param, option, suggestion):
582 if self.params.get(param) is not None:
583 self.report_warning('%s is deprecated. Use %s instead' % (option, suggestion))
584 return True
585 return False
586
587 if check_deprecated('cn_verification_proxy', '--cn-verification-proxy', '--geo-verification-proxy'):
588 if self.params.get('geo_verification_proxy') is None:
589 self.params['geo_verification_proxy'] = self.params['cn_verification_proxy']
590
591 check_deprecated('autonumber', '--auto-number', '-o "%(autonumber)s-%(title)s.%(ext)s"')
592 check_deprecated('usetitle', '--title', '-o "%(title)s-%(id)s.%(ext)s"')
593 check_deprecated('useid', '--id', '-o "%(id)s.%(ext)s"')
594
595 for msg in self.params.get('_warnings', []):
596 self.report_warning(msg)
597 for msg in self.params.get('_deprecation_warnings', []):
598 self.deprecation_warning(msg)
599
600 if 'list-formats' in self.params.get('compat_opts', []):
601 self.params['listformats_table'] = False
602
603 if 'overwrites' not in self.params and self.params.get('nooverwrites') is not None:
604 # nooverwrites was unnecessarily changed to overwrites
605 # in 0c3d0f51778b153f65c21906031c2e091fcfb641
606 # This ensures compatibility with both keys
607 self.params['overwrites'] = not self.params['nooverwrites']
608 elif self.params.get('overwrites') is None:
609 self.params.pop('overwrites', None)
610 else:
611 self.params['nooverwrites'] = not self.params['overwrites']
612
613 self.params.setdefault('forceprint', {})
614 self.params.setdefault('print_to_file', {})
615
616 # Compatibility with older syntax
617 if not isinstance(params['forceprint'], dict):
618 self.params['forceprint'] = {'video': params['forceprint']}
619
620 if self.params.get('bidi_workaround', False):
621 try:
622 import pty
623 master, slave = pty.openpty()
624 width = compat_get_terminal_size().columns
625 if width is None:
626 width_args = []
627 else:
628 width_args = ['-w', str(width)]
629 sp_kwargs = dict(
630 stdin=subprocess.PIPE,
631 stdout=slave,
632 stderr=self._out_files['error'])
633 try:
634 self._output_process = Popen(['bidiv'] + width_args, **sp_kwargs)
635 except OSError:
636 self._output_process = Popen(['fribidi', '-c', 'UTF-8'] + width_args, **sp_kwargs)
637 self._output_channel = os.fdopen(master, 'rb')
638 except OSError as ose:
639 if ose.errno == errno.ENOENT:
640 self.report_warning(
641 'Could not find fribidi executable, ignoring --bidi-workaround. '
642 'Make sure that fribidi is an executable file in one of the directories in your $PATH.')
643 else:
644 raise
645
646 if (sys.platform != 'win32'
647 and sys.getfilesystemencoding() in ['ascii', 'ANSI_X3.4-1968']
648 and not self.params.get('restrictfilenames', False)):
649 # Unicode filesystem API will throw errors (#1474, #13027)
650 self.report_warning(
651 'Assuming --restrict-filenames since file system encoding '
652 'cannot encode all characters. '
653 'Set the LC_ALL environment variable to fix this.')
654 self.params['restrictfilenames'] = True
655
656 self.outtmpl_dict = self.parse_outtmpl()
657
658 # Creating format selector here allows us to catch syntax errors before the extraction
659 self.format_selector = (
660 self.params.get('format') if self.params.get('format') in (None, '-')
661 else self.params['format'] if callable(self.params['format'])
662 else self.build_format_selector(self.params['format']))
663
664 # Set http_headers defaults according to std_headers
665 self.params['http_headers'] = merge_headers(std_headers, self.params.get('http_headers', {}))
666
667 self._setup_opener()
668
669 if auto_init:
670 if auto_init != 'no_verbose_header':
671 self.print_debug_header()
672 self.add_default_info_extractors()
673
674 hooks = {
675 'post_hooks': self.add_post_hook,
676 'progress_hooks': self.add_progress_hook,
677 'postprocessor_hooks': self.add_postprocessor_hook,
678 }
679 for opt, fn in hooks.items():
680 for ph in self.params.get(opt, []):
681 fn(ph)
682
683 for pp_def_raw in self.params.get('postprocessors', []):
684 pp_def = dict(pp_def_raw)
685 when = pp_def.pop('when', 'post_process')
686 self.add_post_processor(
687 get_postprocessor(pp_def.pop('key'))(self, **compat_kwargs(pp_def)),
688 when=when)
689
690 register_socks_protocols()
691
692 def preload_download_archive(fn):
693 """Preload the archive, if any is specified"""
694 if fn is None:
695 return False
696 self.write_debug(f'Loading archive file {fn!r}')
697 try:
698 with locked_file(fn, 'r', encoding='utf-8') as archive_file:
699 for line in archive_file:
700 self.archive.add(line.strip())
701 except IOError as ioe:
702 if ioe.errno != errno.ENOENT:
703 raise
704 return False
705 return True
706
707 self.archive = set()
708 preload_download_archive(self.params.get('download_archive'))
709
710 def warn_if_short_id(self, argv):
711 # short YouTube ID starting with dash?
712 idxs = [
713 i for i, a in enumerate(argv)
714 if re.match(r'^-[0-9A-Za-z_-]{10}$', a)]
715 if idxs:
716 correct_argv = (
717 ['yt-dlp']
718 + [a for i, a in enumerate(argv) if i not in idxs]
719 + ['--'] + [argv[i] for i in idxs]
720 )
721 self.report_warning(
722 'Long argument string detected. '
723 'Use -- to separate parameters and URLs, like this:\n%s' %
724 args_to_str(correct_argv))
725
726 def add_info_extractor(self, ie):
727 """Add an InfoExtractor object to the end of the list."""
728 ie_key = ie.ie_key()
729 self._ies[ie_key] = ie
730 if not isinstance(ie, type):
731 self._ies_instances[ie_key] = ie
732 ie.set_downloader(self)
733
734 def _get_info_extractor_class(self, ie_key):
735 ie = self._ies.get(ie_key)
736 if ie is None:
737 ie = get_info_extractor(ie_key)
738 self.add_info_extractor(ie)
739 return ie
740
741 def get_info_extractor(self, ie_key):
742 """
743 Get an instance of an IE with name ie_key, it will try to get one from
744 the _ies list, if there's no instance it will create a new one and add
745 it to the extractor list.
746 """
747 ie = self._ies_instances.get(ie_key)
748 if ie is None:
749 ie = get_info_extractor(ie_key)()
750 self.add_info_extractor(ie)
751 return ie
752
753 def add_default_info_extractors(self):
754 """
755 Add the InfoExtractors returned by gen_extractors to the end of the list
756 """
757 for ie in gen_extractor_classes():
758 self.add_info_extractor(ie)
759
760 def add_post_processor(self, pp, when='post_process'):
761 """Add a PostProcessor object to the end of the chain."""
762 self._pps[when].append(pp)
763 pp.set_downloader(self)
764
765 def add_post_hook(self, ph):
766 """Add the post hook"""
767 self._post_hooks.append(ph)
768
769 def add_progress_hook(self, ph):
770 """Add the download progress hook"""
771 self._progress_hooks.append(ph)
772
773 def add_postprocessor_hook(self, ph):
774 """Add the postprocessing progress hook"""
775 self._postprocessor_hooks.append(ph)
776 for pps in self._pps.values():
777 for pp in pps:
778 pp.add_progress_hook(ph)
779
780 def _bidi_workaround(self, message):
781 if not hasattr(self, '_output_channel'):
782 return message
783
784 assert hasattr(self, '_output_process')
785 assert isinstance(message, compat_str)
786 line_count = message.count('\n') + 1
787 self._output_process.stdin.write((message + '\n').encode('utf-8'))
788 self._output_process.stdin.flush()
789 res = ''.join(self._output_channel.readline().decode('utf-8')
790 for _ in range(line_count))
791 return res[:-len('\n')]
792
793 def _write_string(self, message, out=None, only_once=False):
794 if only_once:
795 if message in self._printed_messages:
796 return
797 self._printed_messages.add(message)
798 write_string(message, out=out, encoding=self.params.get('encoding'))
799
800 def to_stdout(self, message, skip_eol=False, quiet=None):
801 """Print message to stdout"""
802 if quiet is not None:
803 self.deprecation_warning('"YoutubeDL.to_stdout" no longer accepts the argument quiet. Use "YoutubeDL.to_screen" instead')
804 self._write_string(
805 '%s%s' % (self._bidi_workaround(message), ('' if skip_eol else '\n')),
806 self._out_files['print'])
807
808 def to_screen(self, message, skip_eol=False, quiet=None):
809 """Print message to screen if not in quiet mode"""
810 if self.params.get('logger'):
811 self.params['logger'].debug(message)
812 return
813 if (self.params.get('quiet') if quiet is None else quiet) and not self.params.get('verbose'):
814 return
815 self._write_string(
816 '%s%s' % (self._bidi_workaround(message), ('' if skip_eol else '\n')),
817 self._out_files['screen'])
818
819 def to_stderr(self, message, only_once=False):
820 """Print message to stderr"""
821 assert isinstance(message, compat_str)
822 if self.params.get('logger'):
823 self.params['logger'].error(message)
824 else:
825 self._write_string('%s\n' % self._bidi_workaround(message), self._out_files['error'], only_once=only_once)
826
827 def _send_console_code(self, code):
828 if compat_os_name == 'nt' or not self._out_files['console']:
829 return
830 self._write_string(code, self._out_files['console'])
831
832 def to_console_title(self, message):
833 if not self.params.get('consoletitle', False):
834 return
835 message = remove_terminal_sequences(message)
836 if compat_os_name == 'nt':
837 if ctypes.windll.kernel32.GetConsoleWindow():
838 # c_wchar_p() might not be necessary if `message` is
839 # already of type unicode()
840 ctypes.windll.kernel32.SetConsoleTitleW(ctypes.c_wchar_p(message))
841 else:
842 self._send_console_code(f'\033]0;{message}\007')
843
844 def save_console_title(self):
845 if not self.params.get('consoletitle') or self.params.get('simulate'):
846 return
847 self._send_console_code('\033[22;0t') # Save the title on stack
848
849 def restore_console_title(self):
850 if not self.params.get('consoletitle') or self.params.get('simulate'):
851 return
852 self._send_console_code('\033[23;0t') # Restore the title from stack
853
854 def __enter__(self):
855 self.save_console_title()
856 return self
857
858 def __exit__(self, *args):
859 self.restore_console_title()
860
861 if self.params.get('cookiefile') is not None:
862 self.cookiejar.save(ignore_discard=True, ignore_expires=True)
863
864 def trouble(self, message=None, tb=None, is_error=True):
865 """Determine action to take when a download problem appears.
866
867 Depending on if the downloader has been configured to ignore
868 download errors or not, this method may throw an exception or
869 not when errors are found, after printing the message.
870
871 @param tb If given, is additional traceback information
872 @param is_error Whether to raise error according to ignorerrors
873 """
874 if message is not None:
875 self.to_stderr(message)
876 if self.params.get('verbose'):
877 if tb is None:
878 if sys.exc_info()[0]: # if .trouble has been called from an except block
879 tb = ''
880 if hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
881 tb += ''.join(traceback.format_exception(*sys.exc_info()[1].exc_info))
882 tb += encode_compat_str(traceback.format_exc())
883 else:
884 tb_data = traceback.format_list(traceback.extract_stack())
885 tb = ''.join(tb_data)
886 if tb:
887 self.to_stderr(tb)
888 if not is_error:
889 return
890 if not self.params.get('ignoreerrors'):
891 if sys.exc_info()[0] and hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
892 exc_info = sys.exc_info()[1].exc_info
893 else:
894 exc_info = sys.exc_info()
895 raise DownloadError(message, exc_info)
896 self._download_retcode = 1
897
898 class Styles(Enum):
899 HEADERS = 'yellow'
900 EMPHASIS = 'light blue'
901 ID = 'green'
902 DELIM = 'blue'
903 ERROR = 'red'
904 WARNING = 'yellow'
905 SUPPRESS = 'light black'
906
907 def _format_text(self, handle, allow_colors, text, f, fallback=None, *, test_encoding=False):
908 if test_encoding:
909 original_text = text
910 # handle.encoding can be None. See https://github.com/yt-dlp/yt-dlp/issues/2711
911 encoding = self.params.get('encoding') or getattr(handle, 'encoding', None) or 'ascii'
912 text = text.encode(encoding, 'ignore').decode(encoding)
913 if fallback is not None and text != original_text:
914 text = fallback
915 if isinstance(f, self.Styles):
916 f = f.value
917 return format_text(text, f) if allow_colors else text if fallback is None else fallback
918
919 def _format_screen(self, *args, **kwargs):
920 return self._format_text(
921 self._out_files['screen'], self._allow_colors['screen'], *args, **kwargs)
922
923 def _format_err(self, *args, **kwargs):
924 return self._format_text(
925 self._out_files['error'], self._allow_colors['error'], *args, **kwargs)
926
927 def report_warning(self, message, only_once=False):
928 '''
929 Print the message to stderr, it will be prefixed with 'WARNING:'
930 If stderr is a tty file the 'WARNING:' will be colored
931 '''
932 if self.params.get('logger') is not None:
933 self.params['logger'].warning(message)
934 else:
935 if self.params.get('no_warnings'):
936 return
937 self.to_stderr(f'{self._format_err("WARNING:", self.Styles.WARNING)} {message}', only_once)
938
939 def deprecation_warning(self, message):
940 if self.params.get('logger') is not None:
941 self.params['logger'].warning(f'DeprecationWarning: {message}')
942 else:
943 self.to_stderr(f'{self._format_err("DeprecationWarning:", self.Styles.ERROR)} {message}', True)
944
945 def report_error(self, message, *args, **kwargs):
946 '''
947 Do the same as trouble, but prefixes the message with 'ERROR:', colored
948 in red if stderr is a tty file.
949 '''
950 self.trouble(f'{self._format_err("ERROR:", self.Styles.ERROR)} {message}', *args, **kwargs)
951
952 def write_debug(self, message, only_once=False):
953 '''Log debug message or Print message to stderr'''
954 if not self.params.get('verbose', False):
955 return
956 message = '[debug] %s' % message
957 if self.params.get('logger'):
958 self.params['logger'].debug(message)
959 else:
960 self.to_stderr(message, only_once)
961
962 def report_file_already_downloaded(self, file_name):
963 """Report file has already been fully downloaded."""
964 try:
965 self.to_screen('[download] %s has already been downloaded' % file_name)
966 except UnicodeEncodeError:
967 self.to_screen('[download] The file has already been downloaded')
968
969 def report_file_delete(self, file_name):
970 """Report that existing file will be deleted."""
971 try:
972 self.to_screen('Deleting existing file %s' % file_name)
973 except UnicodeEncodeError:
974 self.to_screen('Deleting existing file')
975
976 def raise_no_formats(self, info, forced=False, *, msg=None):
977 has_drm = info.get('__has_drm')
978 ignored, expected = self.params.get('ignore_no_formats_error'), bool(msg)
979 msg = msg or has_drm and 'This video is DRM protected' or 'No video formats found!'
980 if forced or not ignored:
981 raise ExtractorError(msg, video_id=info['id'], ie=info['extractor'],
982 expected=has_drm or ignored or expected)
983 else:
984 self.report_warning(msg)
985
986 def parse_outtmpl(self):
987 outtmpl_dict = self.params.get('outtmpl', {})
988 if not isinstance(outtmpl_dict, dict):
989 outtmpl_dict = {'default': outtmpl_dict}
990 # Remove spaces in the default template
991 if self.params.get('restrictfilenames'):
992 sanitize = lambda x: x.replace(' - ', ' ').replace(' ', '-')
993 else:
994 sanitize = lambda x: x
995 outtmpl_dict.update({
996 k: sanitize(v) for k, v in DEFAULT_OUTTMPL.items()
997 if outtmpl_dict.get(k) is None})
998 for key, val in outtmpl_dict.items():
999 if isinstance(val, bytes):
1000 self.report_warning(
1001 'Parameter outtmpl is bytes, but should be a unicode string. '
1002 'Put from __future__ import unicode_literals at the top of your code file or consider switching to Python 3.x.')
1003 return outtmpl_dict
1004
1005 def get_output_path(self, dir_type='', filename=None):
1006 paths = self.params.get('paths', {})
1007 assert isinstance(paths, dict)
1008 path = os.path.join(
1009 expand_path(paths.get('home', '').strip()),
1010 expand_path(paths.get(dir_type, '').strip()) if dir_type else '',
1011 filename or '')
1012
1013 # Temporary fix for #4787
1014 # 'Treat' all problem characters by passing filename through preferredencoding
1015 # to workaround encoding issues with subprocess on python2 @ Windows
1016 if sys.version_info < (3, 0) and sys.platform == 'win32':
1017 path = encodeFilename(path, True).decode(preferredencoding())
1018 return sanitize_path(path, force=self.params.get('windowsfilenames'))
1019
1020 @staticmethod
1021 def _outtmpl_expandpath(outtmpl):
1022 # expand_path translates '%%' into '%' and '$$' into '$'
1023 # correspondingly that is not what we want since we need to keep
1024 # '%%' intact for template dict substitution step. Working around
1025 # with boundary-alike separator hack.
1026 sep = ''.join([random.choice(ascii_letters) for _ in range(32)])
1027 outtmpl = outtmpl.replace('%%', '%{0}%'.format(sep)).replace('$$', '${0}$'.format(sep))
1028
1029 # outtmpl should be expand_path'ed before template dict substitution
1030 # because meta fields may contain env variables we don't want to
1031 # be expanded. For example, for outtmpl "%(title)s.%(ext)s" and
1032 # title "Hello $PATH", we don't want `$PATH` to be expanded.
1033 return expand_path(outtmpl).replace(sep, '')
1034
1035 @staticmethod
1036 def escape_outtmpl(outtmpl):
1037 ''' Escape any remaining strings like %s, %abc% etc. '''
1038 return re.sub(
1039 STR_FORMAT_RE_TMPL.format('', '(?![%(\0])'),
1040 lambda mobj: ('' if mobj.group('has_key') else '%') + mobj.group(0),
1041 outtmpl)
1042
1043 @classmethod
1044 def validate_outtmpl(cls, outtmpl):
1045 ''' @return None or Exception object '''
1046 outtmpl = re.sub(
1047 STR_FORMAT_RE_TMPL.format('[^)]*', '[ljqBUDS]'),
1048 lambda mobj: f'{mobj.group(0)[:-1]}s',
1049 cls._outtmpl_expandpath(outtmpl))
1050 try:
1051 cls.escape_outtmpl(outtmpl) % collections.defaultdict(int)
1052 return None
1053 except ValueError as err:
1054 return err
1055
1056 @staticmethod
1057 def _copy_infodict(info_dict):
1058 info_dict = dict(info_dict)
1059 info_dict.pop('__postprocessors', None)
1060 return info_dict
1061
1062 def prepare_outtmpl(self, outtmpl, info_dict, sanitize=False):
1063 """ Make the outtmpl and info_dict suitable for substitution: ydl.escape_outtmpl(outtmpl) % info_dict
1064 @param sanitize Whether to sanitize the output as a filename.
1065 For backward compatibility, a function can also be passed
1066 """
1067
1068 info_dict.setdefault('epoch', int(time.time())) # keep epoch consistent once set
1069
1070 info_dict = self._copy_infodict(info_dict)
1071 info_dict['duration_string'] = ( # %(duration>%H-%M-%S)s is wrong if duration > 24hrs
1072 formatSeconds(info_dict['duration'], '-' if sanitize else ':')
1073 if info_dict.get('duration', None) is not None
1074 else None)
1075 info_dict['autonumber'] = self.params.get('autonumber_start', 1) - 1 + self._num_downloads
1076 info_dict['video_autonumber'] = self._num_videos
1077 if info_dict.get('resolution') is None:
1078 info_dict['resolution'] = self.format_resolution(info_dict, default=None)
1079
1080 # For fields playlist_index, playlist_autonumber and autonumber convert all occurrences
1081 # of %(field)s to %(field)0Nd for backward compatibility
1082 field_size_compat_map = {
1083 'playlist_index': number_of_digits(info_dict.get('_last_playlist_index') or 0),
1084 'playlist_autonumber': number_of_digits(info_dict.get('n_entries') or 0),
1085 'autonumber': self.params.get('autonumber_size') or 5,
1086 }
1087
1088 TMPL_DICT = {}
1089 EXTERNAL_FORMAT_RE = re.compile(STR_FORMAT_RE_TMPL.format('[^)]*', f'[{STR_FORMAT_TYPES}ljqBUDS]'))
1090 MATH_FUNCTIONS = {
1091 '+': float.__add__,
1092 '-': float.__sub__,
1093 }
1094 # Field is of the form key1.key2...
1095 # where keys (except first) can be string, int or slice
1096 FIELD_RE = r'\w*(?:\.(?:\w+|{num}|{num}?(?::{num}?){{1,2}}))*'.format(num=r'(?:-?\d+)')
1097 MATH_FIELD_RE = r'''(?:{field}|{num})'''.format(field=FIELD_RE, num=r'-?\d+(?:.\d+)?')
1098 MATH_OPERATORS_RE = r'(?:%s)' % '|'.join(map(re.escape, MATH_FUNCTIONS.keys()))
1099 INTERNAL_FORMAT_RE = re.compile(r'''(?x)
1100 (?P<negate>-)?
1101 (?P<fields>{field})
1102 (?P<maths>(?:{math_op}{math_field})*)
1103 (?:>(?P<strf_format>.+?))?
1104 (?P<remaining>
1105 (?P<alternate>(?<!\\),[^|&)]+)?
1106 (?:&(?P<replacement>.*?))?
1107 (?:\|(?P<default>.*?))?
1108 )$'''.format(field=FIELD_RE, math_op=MATH_OPERATORS_RE, math_field=MATH_FIELD_RE))
1109
1110 def _traverse_infodict(k):
1111 k = k.split('.')
1112 if k[0] == '':
1113 k.pop(0)
1114 return traverse_obj(info_dict, k, is_user_input=True, traverse_string=True)
1115
1116 def get_value(mdict):
1117 # Object traversal
1118 value = _traverse_infodict(mdict['fields'])
1119 # Negative
1120 if mdict['negate']:
1121 value = float_or_none(value)
1122 if value is not None:
1123 value *= -1
1124 # Do maths
1125 offset_key = mdict['maths']
1126 if offset_key:
1127 value = float_or_none(value)
1128 operator = None
1129 while offset_key:
1130 item = re.match(
1131 MATH_FIELD_RE if operator else MATH_OPERATORS_RE,
1132 offset_key).group(0)
1133 offset_key = offset_key[len(item):]
1134 if operator is None:
1135 operator = MATH_FUNCTIONS[item]
1136 continue
1137 item, multiplier = (item[1:], -1) if item[0] == '-' else (item, 1)
1138 offset = float_or_none(item)
1139 if offset is None:
1140 offset = float_or_none(_traverse_infodict(item))
1141 try:
1142 value = operator(value, multiplier * offset)
1143 except (TypeError, ZeroDivisionError):
1144 return None
1145 operator = None
1146 # Datetime formatting
1147 if mdict['strf_format']:
1148 value = strftime_or_none(value, mdict['strf_format'].replace('\\,', ','))
1149
1150 return value
1151
1152 na = self.params.get('outtmpl_na_placeholder', 'NA')
1153
1154 def filename_sanitizer(key, value, restricted=self.params.get('restrictfilenames')):
1155 return sanitize_filename(str(value), restricted=restricted, is_id=(
1156 bool(re.search(r'(^|[_.])id(\.|$)', key))
1157 if 'filename-sanitization' in self.params.get('compat_opts', [])
1158 else NO_DEFAULT))
1159
1160 sanitizer = sanitize if callable(sanitize) else filename_sanitizer
1161 sanitize = bool(sanitize)
1162
1163 def _dumpjson_default(obj):
1164 if isinstance(obj, (set, LazyList)):
1165 return list(obj)
1166 return repr(obj)
1167
1168 def create_key(outer_mobj):
1169 if not outer_mobj.group('has_key'):
1170 return outer_mobj.group(0)
1171 key = outer_mobj.group('key')
1172 mobj = re.match(INTERNAL_FORMAT_RE, key)
1173 initial_field = mobj.group('fields') if mobj else ''
1174 value, replacement, default = None, None, na
1175 while mobj:
1176 mobj = mobj.groupdict()
1177 default = mobj['default'] if mobj['default'] is not None else default
1178 value = get_value(mobj)
1179 replacement = mobj['replacement']
1180 if value is None and mobj['alternate']:
1181 mobj = re.match(INTERNAL_FORMAT_RE, mobj['remaining'][1:])
1182 else:
1183 break
1184
1185 fmt = outer_mobj.group('format')
1186 if fmt == 's' and value is not None and key in field_size_compat_map.keys():
1187 fmt = '0{:d}d'.format(field_size_compat_map[key])
1188
1189 value = default if value is None else value if replacement is None else replacement
1190
1191 flags = outer_mobj.group('conversion') or ''
1192 str_fmt = f'{fmt[:-1]}s'
1193 if fmt[-1] == 'l': # list
1194 delim = '\n' if '#' in flags else ', '
1195 value, fmt = delim.join(map(str, variadic(value, allowed_types=(str, bytes)))), str_fmt
1196 elif fmt[-1] == 'j': # json
1197 value, fmt = json.dumps(value, default=_dumpjson_default, indent=4 if '#' in flags else None), str_fmt
1198 elif fmt[-1] == 'q': # quoted
1199 value = map(str, variadic(value) if '#' in flags else [value])
1200 value, fmt = ' '.join(map(compat_shlex_quote, value)), str_fmt
1201 elif fmt[-1] == 'B': # bytes
1202 value = f'%{str_fmt}'.encode('utf-8') % str(value).encode('utf-8')
1203 value, fmt = value.decode('utf-8', 'ignore'), 's'
1204 elif fmt[-1] == 'U': # unicode normalized
1205 value, fmt = unicodedata.normalize(
1206 # "+" = compatibility equivalence, "#" = NFD
1207 'NF%s%s' % ('K' if '+' in flags else '', 'D' if '#' in flags else 'C'),
1208 value), str_fmt
1209 elif fmt[-1] == 'D': # decimal suffix
1210 num_fmt, fmt = fmt[:-1].replace('#', ''), 's'
1211 value = format_decimal_suffix(value, f'%{num_fmt}f%s' if num_fmt else '%d%s',
1212 factor=1024 if '#' in flags else 1000)
1213 elif fmt[-1] == 'S': # filename sanitization
1214 value, fmt = filename_sanitizer(initial_field, value, restricted='#' in flags), str_fmt
1215 elif fmt[-1] == 'c':
1216 if value:
1217 value = str(value)[0]
1218 else:
1219 fmt = str_fmt
1220 elif fmt[-1] not in 'rs': # numeric
1221 value = float_or_none(value)
1222 if value is None:
1223 value, fmt = default, 's'
1224
1225 if sanitize:
1226 if fmt[-1] == 'r':
1227 # If value is an object, sanitize might convert it to a string
1228 # So we convert it to repr first
1229 value, fmt = repr(value), str_fmt
1230 if fmt[-1] in 'csr':
1231 value = sanitizer(initial_field, value)
1232
1233 key = '%s\0%s' % (key.replace('%', '%\0'), outer_mobj.group('format'))
1234 TMPL_DICT[key] = value
1235 return '{prefix}%({key}){fmt}'.format(key=key, fmt=fmt, prefix=outer_mobj.group('prefix'))
1236
1237 return EXTERNAL_FORMAT_RE.sub(create_key, outtmpl), TMPL_DICT
1238
1239 def evaluate_outtmpl(self, outtmpl, info_dict, *args, **kwargs):
1240 outtmpl, info_dict = self.prepare_outtmpl(outtmpl, info_dict, *args, **kwargs)
1241 return self.escape_outtmpl(outtmpl) % info_dict
1242
1243 def _prepare_filename(self, info_dict, *, outtmpl=None, tmpl_type=None):
1244 assert None in (outtmpl, tmpl_type), 'outtmpl and tmpl_type are mutually exclusive'
1245 if outtmpl is None:
1246 outtmpl = self.outtmpl_dict.get(tmpl_type or 'default', self.outtmpl_dict['default'])
1247 try:
1248 outtmpl = self._outtmpl_expandpath(outtmpl)
1249 filename = self.evaluate_outtmpl(outtmpl, info_dict, True)
1250 if not filename:
1251 return None
1252
1253 if tmpl_type in ('', 'temp'):
1254 final_ext, ext = self.params.get('final_ext'), info_dict.get('ext')
1255 if final_ext and ext and final_ext != ext and filename.endswith(f'.{final_ext}'):
1256 filename = replace_extension(filename, ext, final_ext)
1257 elif tmpl_type:
1258 force_ext = OUTTMPL_TYPES[tmpl_type]
1259 if force_ext:
1260 filename = replace_extension(filename, force_ext, info_dict.get('ext'))
1261
1262 # https://github.com/blackjack4494/youtube-dlc/issues/85
1263 trim_file_name = self.params.get('trim_file_name', False)
1264 if trim_file_name:
1265 no_ext, *ext = filename.rsplit('.', 2)
1266 filename = join_nonempty(no_ext[:trim_file_name], *ext, delim='.')
1267
1268 return filename
1269 except ValueError as err:
1270 self.report_error('Error in output template: ' + str(err) + ' (encoding: ' + repr(preferredencoding()) + ')')
1271 return None
1272
1273 def prepare_filename(self, info_dict, dir_type='', *, outtmpl=None, warn=False):
1274 """Generate the output filename"""
1275 if outtmpl:
1276 assert not dir_type, 'outtmpl and dir_type are mutually exclusive'
1277 dir_type = None
1278 filename = self._prepare_filename(info_dict, tmpl_type=dir_type, outtmpl=outtmpl)
1279 if not filename and dir_type not in ('', 'temp'):
1280 return ''
1281
1282 if warn:
1283 if not self.params.get('paths'):
1284 pass
1285 elif filename == '-':
1286 self.report_warning('--paths is ignored when an outputting to stdout', only_once=True)
1287 elif os.path.isabs(filename):
1288 self.report_warning('--paths is ignored since an absolute path is given in output template', only_once=True)
1289 if filename == '-' or not filename:
1290 return filename
1291
1292 return self.get_output_path(dir_type, filename)
1293
1294 def _match_entry(self, info_dict, incomplete=False, silent=False):
1295 """ Returns None if the file should be downloaded """
1296
1297 video_title = info_dict.get('title', info_dict.get('id', 'video'))
1298
1299 def check_filter():
1300 if 'title' in info_dict:
1301 # This can happen when we're just evaluating the playlist
1302 title = info_dict['title']
1303 matchtitle = self.params.get('matchtitle', False)
1304 if matchtitle:
1305 if not re.search(matchtitle, title, re.IGNORECASE):
1306 return '"' + title + '" title did not match pattern "' + matchtitle + '"'
1307 rejecttitle = self.params.get('rejecttitle', False)
1308 if rejecttitle:
1309 if re.search(rejecttitle, title, re.IGNORECASE):
1310 return '"' + title + '" title matched reject pattern "' + rejecttitle + '"'
1311 date = info_dict.get('upload_date')
1312 if date is not None:
1313 dateRange = self.params.get('daterange', DateRange())
1314 if date not in dateRange:
1315 return '%s upload date is not in range %s' % (date_from_str(date).isoformat(), dateRange)
1316 view_count = info_dict.get('view_count')
1317 if view_count is not None:
1318 min_views = self.params.get('min_views')
1319 if min_views is not None and view_count < min_views:
1320 return 'Skipping %s, because it has not reached minimum view count (%d/%d)' % (video_title, view_count, min_views)
1321 max_views = self.params.get('max_views')
1322 if max_views is not None and view_count > max_views:
1323 return 'Skipping %s, because it has exceeded the maximum view count (%d/%d)' % (video_title, view_count, max_views)
1324 if age_restricted(info_dict.get('age_limit'), self.params.get('age_limit')):
1325 return 'Skipping "%s" because it is age restricted' % video_title
1326
1327 match_filter = self.params.get('match_filter')
1328 if match_filter is not None:
1329 try:
1330 ret = match_filter(info_dict, incomplete=incomplete)
1331 except TypeError:
1332 # For backward compatibility
1333 ret = None if incomplete else match_filter(info_dict)
1334 if ret is not None:
1335 return ret
1336 return None
1337
1338 if self.in_download_archive(info_dict):
1339 reason = '%s has already been recorded in the archive' % video_title
1340 break_opt, break_err = 'break_on_existing', ExistingVideoReached
1341 else:
1342 reason = check_filter()
1343 break_opt, break_err = 'break_on_reject', RejectedVideoReached
1344 if reason is not None:
1345 if not silent:
1346 self.to_screen('[download] ' + reason)
1347 if self.params.get(break_opt, False):
1348 raise break_err()
1349 return reason
1350
1351 @staticmethod
1352 def add_extra_info(info_dict, extra_info):
1353 '''Set the keys from extra_info in info dict if they are missing'''
1354 for key, value in extra_info.items():
1355 info_dict.setdefault(key, value)
1356
1357 def extract_info(self, url, download=True, ie_key=None, extra_info=None,
1358 process=True, force_generic_extractor=False):
1359 """
1360 Return a list with a dictionary for each video extracted.
1361
1362 Arguments:
1363 url -- URL to extract
1364
1365 Keyword arguments:
1366 download -- whether to download videos during extraction
1367 ie_key -- extractor key hint
1368 extra_info -- dictionary containing the extra values to add to each result
1369 process -- whether to resolve all unresolved references (URLs, playlist items),
1370 must be True for download to work.
1371 force_generic_extractor -- force using the generic extractor
1372 """
1373
1374 if extra_info is None:
1375 extra_info = {}
1376
1377 if not ie_key and force_generic_extractor:
1378 ie_key = 'Generic'
1379
1380 if ie_key:
1381 ies = {ie_key: self._get_info_extractor_class(ie_key)}
1382 else:
1383 ies = self._ies
1384
1385 for ie_key, ie in ies.items():
1386 if not ie.suitable(url):
1387 continue
1388
1389 if not ie.working():
1390 self.report_warning('The program functionality for this site has been marked as broken, '
1391 'and will probably not work.')
1392
1393 temp_id = ie.get_temp_id(url)
1394 if temp_id is not None and self.in_download_archive({'id': temp_id, 'ie_key': ie_key}):
1395 self.to_screen(f'[{ie_key}] {temp_id}: has already been recorded in the archive')
1396 if self.params.get('break_on_existing', False):
1397 raise ExistingVideoReached()
1398 break
1399 return self.__extract_info(url, self.get_info_extractor(ie_key), download, extra_info, process)
1400 else:
1401 self.report_error('no suitable InfoExtractor for URL %s' % url)
1402
1403 def __handle_extraction_exceptions(func):
1404 @functools.wraps(func)
1405 def wrapper(self, *args, **kwargs):
1406 while True:
1407 try:
1408 return func(self, *args, **kwargs)
1409 except (DownloadCancelled, LazyList.IndexError, PagedList.IndexError):
1410 raise
1411 except ReExtractInfo as e:
1412 if e.expected:
1413 self.to_screen(f'{e}; Re-extracting data')
1414 else:
1415 self.to_stderr('\r')
1416 self.report_warning(f'{e}; Re-extracting data')
1417 continue
1418 except GeoRestrictedError as e:
1419 msg = e.msg
1420 if e.countries:
1421 msg += '\nThis video is available in %s.' % ', '.join(
1422 map(ISO3166Utils.short2full, e.countries))
1423 msg += '\nYou might want to use a VPN or a proxy server (with --proxy) to workaround.'
1424 self.report_error(msg)
1425 except ExtractorError as e: # An error we somewhat expected
1426 self.report_error(str(e), e.format_traceback())
1427 except Exception as e:
1428 if self.params.get('ignoreerrors'):
1429 self.report_error(str(e), tb=encode_compat_str(traceback.format_exc()))
1430 else:
1431 raise
1432 break
1433 return wrapper
1434
1435 def _wait_for_video(self, ie_result):
1436 if (not self.params.get('wait_for_video')
1437 or ie_result.get('_type', 'video') != 'video'
1438 or ie_result.get('formats') or ie_result.get('url')):
1439 return
1440
1441 format_dur = lambda dur: '%02d:%02d:%02d' % timetuple_from_msec(dur * 1000)[:-1]
1442 last_msg = ''
1443
1444 def progress(msg):
1445 nonlocal last_msg
1446 self.to_screen(msg + ' ' * (len(last_msg) - len(msg)) + '\r', skip_eol=True)
1447 last_msg = msg
1448
1449 min_wait, max_wait = self.params.get('wait_for_video')
1450 diff = try_get(ie_result, lambda x: x['release_timestamp'] - time.time())
1451 if diff is None and ie_result.get('live_status') == 'is_upcoming':
1452 diff = round(random.uniform(min_wait, max_wait) if (max_wait and min_wait) else (max_wait or min_wait), 0)
1453 self.report_warning('Release time of video is not known')
1454 elif (diff or 0) <= 0:
1455 self.report_warning('Video should already be available according to extracted info')
1456 diff = min(max(diff or 0, min_wait or 0), max_wait or float('inf'))
1457 self.to_screen(f'[wait] Waiting for {format_dur(diff)} - Press Ctrl+C to try now')
1458
1459 wait_till = time.time() + diff
1460 try:
1461 while True:
1462 diff = wait_till - time.time()
1463 if diff <= 0:
1464 progress('')
1465 raise ReExtractInfo('[wait] Wait period ended', expected=True)
1466 progress(f'[wait] Remaining time until next attempt: {self._format_screen(format_dur(diff), self.Styles.EMPHASIS)}')
1467 time.sleep(1)
1468 except KeyboardInterrupt:
1469 progress('')
1470 raise ReExtractInfo('[wait] Interrupted by user', expected=True)
1471 except BaseException as e:
1472 if not isinstance(e, ReExtractInfo):
1473 self.to_screen('')
1474 raise
1475
1476 @__handle_extraction_exceptions
1477 def __extract_info(self, url, ie, download, extra_info, process):
1478 ie_result = ie.extract(url)
1479 if ie_result is None: # Finished already (backwards compatibility; listformats and friends should be moved here)
1480 return
1481 if isinstance(ie_result, list):
1482 # Backwards compatibility: old IE result format
1483 ie_result = {
1484 '_type': 'compat_list',
1485 'entries': ie_result,
1486 }
1487 if extra_info.get('original_url'):
1488 ie_result.setdefault('original_url', extra_info['original_url'])
1489 self.add_default_extra_info(ie_result, ie, url)
1490 if process:
1491 self._wait_for_video(ie_result)
1492 return self.process_ie_result(ie_result, download, extra_info)
1493 else:
1494 return ie_result
1495
1496 def add_default_extra_info(self, ie_result, ie, url):
1497 if url is not None:
1498 self.add_extra_info(ie_result, {
1499 'webpage_url': url,
1500 'original_url': url,
1501 })
1502 webpage_url = ie_result.get('webpage_url')
1503 if webpage_url:
1504 self.add_extra_info(ie_result, {
1505 'webpage_url_basename': url_basename(webpage_url),
1506 'webpage_url_domain': get_domain(webpage_url),
1507 })
1508 if ie is not None:
1509 self.add_extra_info(ie_result, {
1510 'extractor': ie.IE_NAME,
1511 'extractor_key': ie.ie_key(),
1512 })
1513
1514 def process_ie_result(self, ie_result, download=True, extra_info=None):
1515 """
1516 Take the result of the ie(may be modified) and resolve all unresolved
1517 references (URLs, playlist items).
1518
1519 It will also download the videos if 'download'.
1520 Returns the resolved ie_result.
1521 """
1522 if extra_info is None:
1523 extra_info = {}
1524 result_type = ie_result.get('_type', 'video')
1525
1526 if result_type in ('url', 'url_transparent'):
1527 ie_result['url'] = sanitize_url(ie_result['url'])
1528 if ie_result.get('original_url'):
1529 extra_info.setdefault('original_url', ie_result['original_url'])
1530
1531 extract_flat = self.params.get('extract_flat', False)
1532 if ((extract_flat == 'in_playlist' and 'playlist' in extra_info)
1533 or extract_flat is True):
1534 info_copy = ie_result.copy()
1535 ie = try_get(ie_result.get('ie_key'), self.get_info_extractor)
1536 if ie and not ie_result.get('id'):
1537 info_copy['id'] = ie.get_temp_id(ie_result['url'])
1538 self.add_default_extra_info(info_copy, ie, ie_result['url'])
1539 self.add_extra_info(info_copy, extra_info)
1540 info_copy, _ = self.pre_process(info_copy)
1541 self.__forced_printings(info_copy, self.prepare_filename(info_copy), incomplete=True)
1542 if self.params.get('force_write_download_archive', False):
1543 self.record_download_archive(info_copy)
1544 return ie_result
1545
1546 if result_type == 'video':
1547 self.add_extra_info(ie_result, extra_info)
1548 ie_result = self.process_video_result(ie_result, download=download)
1549 additional_urls = (ie_result or {}).get('additional_urls')
1550 if additional_urls:
1551 # TODO: Improve MetadataParserPP to allow setting a list
1552 if isinstance(additional_urls, compat_str):
1553 additional_urls = [additional_urls]
1554 self.to_screen(
1555 '[info] %s: %d additional URL(s) requested' % (ie_result['id'], len(additional_urls)))
1556 self.write_debug('Additional URLs: "%s"' % '", "'.join(additional_urls))
1557 ie_result['additional_entries'] = [
1558 self.extract_info(
1559 url, download, extra_info=extra_info,
1560 force_generic_extractor=self.params.get('force_generic_extractor'))
1561 for url in additional_urls
1562 ]
1563 return ie_result
1564 elif result_type == 'url':
1565 # We have to add extra_info to the results because it may be
1566 # contained in a playlist
1567 return self.extract_info(
1568 ie_result['url'], download,
1569 ie_key=ie_result.get('ie_key'),
1570 extra_info=extra_info)
1571 elif result_type == 'url_transparent':
1572 # Use the information from the embedding page
1573 info = self.extract_info(
1574 ie_result['url'], ie_key=ie_result.get('ie_key'),
1575 extra_info=extra_info, download=False, process=False)
1576
1577 # extract_info may return None when ignoreerrors is enabled and
1578 # extraction failed with an error, don't crash and return early
1579 # in this case
1580 if not info:
1581 return info
1582
1583 new_result = info.copy()
1584 new_result.update(filter_dict(ie_result, lambda k, v: (
1585 v is not None and k not in {'_type', 'url', 'id', 'extractor', 'extractor_key', 'ie_key'})))
1586
1587 # Extracted info may not be a video result (i.e.
1588 # info.get('_type', 'video') != video) but rather an url or
1589 # url_transparent. In such cases outer metadata (from ie_result)
1590 # should be propagated to inner one (info). For this to happen
1591 # _type of info should be overridden with url_transparent. This
1592 # fixes issue from https://github.com/ytdl-org/youtube-dl/pull/11163.
1593 if new_result.get('_type') == 'url':
1594 new_result['_type'] = 'url_transparent'
1595
1596 return self.process_ie_result(
1597 new_result, download=download, extra_info=extra_info)
1598 elif result_type in ('playlist', 'multi_video'):
1599 # Protect from infinite recursion due to recursively nested playlists
1600 # (see https://github.com/ytdl-org/youtube-dl/issues/27833)
1601 webpage_url = ie_result['webpage_url']
1602 if webpage_url in self._playlist_urls:
1603 self.to_screen(
1604 '[download] Skipping already downloaded playlist: %s'
1605 % ie_result.get('title') or ie_result.get('id'))
1606 return
1607
1608 self._playlist_level += 1
1609 self._playlist_urls.add(webpage_url)
1610 self._fill_common_fields(ie_result, False)
1611 self._sanitize_thumbnails(ie_result)
1612 try:
1613 return self.__process_playlist(ie_result, download)
1614 finally:
1615 self._playlist_level -= 1
1616 if not self._playlist_level:
1617 self._playlist_urls.clear()
1618 elif result_type == 'compat_list':
1619 self.report_warning(
1620 'Extractor %s returned a compat_list result. '
1621 'It needs to be updated.' % ie_result.get('extractor'))
1622
1623 def _fixup(r):
1624 self.add_extra_info(r, {
1625 'extractor': ie_result['extractor'],
1626 'webpage_url': ie_result['webpage_url'],
1627 'webpage_url_basename': url_basename(ie_result['webpage_url']),
1628 'webpage_url_domain': get_domain(ie_result['webpage_url']),
1629 'extractor_key': ie_result['extractor_key'],
1630 })
1631 return r
1632 ie_result['entries'] = [
1633 self.process_ie_result(_fixup(r), download, extra_info)
1634 for r in ie_result['entries']
1635 ]
1636 return ie_result
1637 else:
1638 raise Exception('Invalid result type: %s' % result_type)
1639
1640 def _ensure_dir_exists(self, path):
1641 return make_dir(path, self.report_error)
1642
1643 @staticmethod
1644 def _playlist_infodict(ie_result, **kwargs):
1645 return {
1646 **ie_result,
1647 'playlist': ie_result.get('title') or ie_result.get('id'),
1648 'playlist_id': ie_result.get('id'),
1649 'playlist_title': ie_result.get('title'),
1650 'playlist_uploader': ie_result.get('uploader'),
1651 'playlist_uploader_id': ie_result.get('uploader_id'),
1652 'playlist_index': 0,
1653 **kwargs,
1654 }
1655
1656 def __process_playlist(self, ie_result, download):
1657 # We process each entry in the playlist
1658 playlist = ie_result.get('title') or ie_result.get('id')
1659 self.to_screen('[download] Downloading playlist: %s' % playlist)
1660
1661 if 'entries' not in ie_result:
1662 raise EntryNotInPlaylist('There are no entries')
1663
1664 MissingEntry = object()
1665 incomplete_entries = bool(ie_result.get('requested_entries'))
1666 if incomplete_entries:
1667 def fill_missing_entries(entries, indices):
1668 ret = [MissingEntry] * max(indices)
1669 for i, entry in zip(indices, entries):
1670 ret[i - 1] = entry
1671 return ret
1672 ie_result['entries'] = fill_missing_entries(ie_result['entries'], ie_result['requested_entries'])
1673
1674 playlist_results = []
1675
1676 playliststart = self.params.get('playliststart', 1)
1677 playlistend = self.params.get('playlistend')
1678 # For backwards compatibility, interpret -1 as whole list
1679 if playlistend == -1:
1680 playlistend = None
1681
1682 playlistitems_str = self.params.get('playlist_items')
1683 playlistitems = None
1684 if playlistitems_str is not None:
1685 def iter_playlistitems(format):
1686 for string_segment in format.split(','):
1687 if '-' in string_segment:
1688 start, end = string_segment.split('-')
1689 for item in range(int(start), int(end) + 1):
1690 yield int(item)
1691 else:
1692 yield int(string_segment)
1693 playlistitems = orderedSet(iter_playlistitems(playlistitems_str))
1694
1695 ie_entries = ie_result['entries']
1696 if isinstance(ie_entries, list):
1697 playlist_count = len(ie_entries)
1698 msg = f'Collected {playlist_count} videos; downloading %d of them'
1699 ie_result['playlist_count'] = ie_result.get('playlist_count') or playlist_count
1700
1701 def get_entry(i):
1702 return ie_entries[i - 1]
1703 else:
1704 msg = 'Downloading %d videos'
1705 if not isinstance(ie_entries, (PagedList, LazyList)):
1706 ie_entries = LazyList(ie_entries)
1707 elif isinstance(ie_entries, InAdvancePagedList):
1708 if ie_entries._pagesize == 1:
1709 playlist_count = ie_entries._pagecount
1710
1711 def get_entry(i):
1712 return YoutubeDL.__handle_extraction_exceptions(
1713 lambda self, i: ie_entries[i - 1]
1714 )(self, i)
1715
1716 entries, broken = [], False
1717 items = playlistitems if playlistitems is not None else itertools.count(playliststart)
1718 for i in items:
1719 if i == 0:
1720 continue
1721 if playlistitems is None and playlistend is not None and playlistend < i:
1722 break
1723 entry = None
1724 try:
1725 entry = get_entry(i)
1726 if entry is MissingEntry:
1727 raise EntryNotInPlaylist()
1728 except (IndexError, EntryNotInPlaylist):
1729 if incomplete_entries:
1730 raise EntryNotInPlaylist(f'Entry {i} cannot be found')
1731 elif not playlistitems:
1732 break
1733 entries.append(entry)
1734 try:
1735 if entry is not None:
1736 self._match_entry(entry, incomplete=True, silent=True)
1737 except (ExistingVideoReached, RejectedVideoReached):
1738 broken = True
1739 break
1740 ie_result['entries'] = entries
1741
1742 # Save playlist_index before re-ordering
1743 entries = [
1744 ((playlistitems[i - 1] if playlistitems else i + playliststart - 1), entry)
1745 for i, entry in enumerate(entries, 1)
1746 if entry is not None]
1747 n_entries = len(entries)
1748
1749 if not (ie_result.get('playlist_count') or broken or playlistitems or playlistend):
1750 ie_result['playlist_count'] = n_entries
1751
1752 if not playlistitems and (playliststart != 1 or playlistend):
1753 playlistitems = list(range(playliststart, playliststart + n_entries))
1754 ie_result['requested_entries'] = playlistitems
1755
1756 _infojson_written = False
1757 write_playlist_files = self.params.get('allow_playlist_files', True)
1758 if write_playlist_files and self.params.get('list_thumbnails'):
1759 self.list_thumbnails(ie_result)
1760 if write_playlist_files and not self.params.get('simulate'):
1761 ie_copy = self._playlist_infodict(ie_result, n_entries=n_entries)
1762 _infojson_written = self._write_info_json(
1763 'playlist', ie_result, self.prepare_filename(ie_copy, 'pl_infojson'))
1764 if _infojson_written is None:
1765 return
1766 if self._write_description('playlist', ie_result,
1767 self.prepare_filename(ie_copy, 'pl_description')) is None:
1768 return
1769 # TODO: This should be passed to ThumbnailsConvertor if necessary
1770 self._write_thumbnails('playlist', ie_copy, self.prepare_filename(ie_copy, 'pl_thumbnail'))
1771
1772 if self.params.get('playlistreverse', False):
1773 entries = entries[::-1]
1774 if self.params.get('playlistrandom', False):
1775 random.shuffle(entries)
1776
1777 x_forwarded_for = ie_result.get('__x_forwarded_for_ip')
1778
1779 self.to_screen('[%s] playlist %s: %s' % (ie_result['extractor'], playlist, msg % n_entries))
1780 failures = 0
1781 max_failures = self.params.get('skip_playlist_after_errors') or float('inf')
1782 for i, entry_tuple in enumerate(entries, 1):
1783 playlist_index, entry = entry_tuple
1784 if 'playlist-index' in self.params.get('compat_opts', []):
1785 playlist_index = playlistitems[i - 1] if playlistitems else i + playliststart - 1
1786 self.to_screen('[download] Downloading video %s of %s' % (i, n_entries))
1787 # This __x_forwarded_for_ip thing is a bit ugly but requires
1788 # minimal changes
1789 if x_forwarded_for:
1790 entry['__x_forwarded_for_ip'] = x_forwarded_for
1791 extra = {
1792 'n_entries': n_entries,
1793 '_last_playlist_index': max(playlistitems) if playlistitems else (playlistend or n_entries),
1794 'playlist_count': ie_result.get('playlist_count'),
1795 'playlist_index': playlist_index,
1796 'playlist_autonumber': i,
1797 'playlist': playlist,
1798 'playlist_id': ie_result.get('id'),
1799 'playlist_title': ie_result.get('title'),
1800 'playlist_uploader': ie_result.get('uploader'),
1801 'playlist_uploader_id': ie_result.get('uploader_id'),
1802 'extractor': ie_result['extractor'],
1803 'webpage_url': ie_result['webpage_url'],
1804 'webpage_url_basename': url_basename(ie_result['webpage_url']),
1805 'webpage_url_domain': get_domain(ie_result['webpage_url']),
1806 'extractor_key': ie_result['extractor_key'],
1807 }
1808
1809 if self._match_entry(entry, incomplete=True) is not None:
1810 continue
1811
1812 entry_result = self.__process_iterable_entry(entry, download, extra)
1813 if not entry_result:
1814 failures += 1
1815 if failures >= max_failures:
1816 self.report_error(
1817 'Skipping the remaining entries in playlist "%s" since %d items failed extraction' % (playlist, failures))
1818 break
1819 playlist_results.append(entry_result)
1820 ie_result['entries'] = playlist_results
1821
1822 # Write the updated info to json
1823 if _infojson_written is True and self._write_info_json(
1824 'updated playlist', ie_result,
1825 self.prepare_filename(ie_copy, 'pl_infojson'), overwrite=True) is None:
1826 return
1827
1828 ie_result = self.run_all_pps('playlist', ie_result)
1829 self.to_screen(f'[download] Finished downloading playlist: {playlist}')
1830 return ie_result
1831
1832 @__handle_extraction_exceptions
1833 def __process_iterable_entry(self, entry, download, extra_info):
1834 return self.process_ie_result(
1835 entry, download=download, extra_info=extra_info)
1836
1837 def _build_format_filter(self, filter_spec):
1838 " Returns a function to filter the formats according to the filter_spec "
1839
1840 OPERATORS = {
1841 '<': operator.lt,
1842 '<=': operator.le,
1843 '>': operator.gt,
1844 '>=': operator.ge,
1845 '=': operator.eq,
1846 '!=': operator.ne,
1847 }
1848 operator_rex = re.compile(r'''(?x)\s*
1849 (?P<key>width|height|tbr|abr|vbr|asr|filesize|filesize_approx|fps)\s*
1850 (?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
1851 (?P<value>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?)\s*
1852 ''' % '|'.join(map(re.escape, OPERATORS.keys())))
1853 m = operator_rex.fullmatch(filter_spec)
1854 if m:
1855 try:
1856 comparison_value = int(m.group('value'))
1857 except ValueError:
1858 comparison_value = parse_filesize(m.group('value'))
1859 if comparison_value is None:
1860 comparison_value = parse_filesize(m.group('value') + 'B')
1861 if comparison_value is None:
1862 raise ValueError(
1863 'Invalid value %r in format specification %r' % (
1864 m.group('value'), filter_spec))
1865 op = OPERATORS[m.group('op')]
1866
1867 if not m:
1868 STR_OPERATORS = {
1869 '=': operator.eq,
1870 '^=': lambda attr, value: attr.startswith(value),
1871 '$=': lambda attr, value: attr.endswith(value),
1872 '*=': lambda attr, value: value in attr,
1873 '~=': lambda attr, value: value.search(attr) is not None
1874 }
1875 str_operator_rex = re.compile(r'''(?x)\s*
1876 (?P<key>[a-zA-Z0-9._-]+)\s*
1877 (?P<negation>!\s*)?(?P<op>%s)\s*(?P<none_inclusive>\?\s*)?
1878 (?P<quote>["'])?
1879 (?P<value>(?(quote)(?:(?!(?P=quote))[^\\]|\\.)+|[\w.-]+))
1880 (?(quote)(?P=quote))\s*
1881 ''' % '|'.join(map(re.escape, STR_OPERATORS.keys())))
1882 m = str_operator_rex.fullmatch(filter_spec)
1883 if m:
1884 if m.group('op') == '~=':
1885 comparison_value = re.compile(m.group('value'))
1886 else:
1887 comparison_value = re.sub(r'''\\([\\"'])''', r'\1', m.group('value'))
1888 str_op = STR_OPERATORS[m.group('op')]
1889 if m.group('negation'):
1890 op = lambda attr, value: not str_op(attr, value)
1891 else:
1892 op = str_op
1893
1894 if not m:
1895 raise SyntaxError('Invalid filter specification %r' % filter_spec)
1896
1897 def _filter(f):
1898 actual_value = f.get(m.group('key'))
1899 if actual_value is None:
1900 return m.group('none_inclusive')
1901 return op(actual_value, comparison_value)
1902 return _filter
1903
1904 def _check_formats(self, formats):
1905 for f in formats:
1906 self.to_screen('[info] Testing format %s' % f['format_id'])
1907 path = self.get_output_path('temp')
1908 if not self._ensure_dir_exists(f'{path}/'):
1909 continue
1910 temp_file = tempfile.NamedTemporaryFile(suffix='.tmp', delete=False, dir=path or None)
1911 temp_file.close()
1912 try:
1913 success, _ = self.dl(temp_file.name, f, test=True)
1914 except (DownloadError, IOError, OSError, ValueError) + network_exceptions:
1915 success = False
1916 finally:
1917 if os.path.exists(temp_file.name):
1918 try:
1919 os.remove(temp_file.name)
1920 except OSError:
1921 self.report_warning('Unable to delete temporary file "%s"' % temp_file.name)
1922 if success:
1923 yield f
1924 else:
1925 self.to_screen('[info] Unable to download format %s. Skipping...' % f['format_id'])
1926
1927 def _default_format_spec(self, info_dict, download=True):
1928
1929 def can_merge():
1930 merger = FFmpegMergerPP(self)
1931 return merger.available and merger.can_merge()
1932
1933 prefer_best = (
1934 not self.params.get('simulate')
1935 and download
1936 and (
1937 not can_merge()
1938 or info_dict.get('is_live', False)
1939 or self.outtmpl_dict['default'] == '-'))
1940 compat = (
1941 prefer_best
1942 or self.params.get('allow_multiple_audio_streams', False)
1943 or 'format-spec' in self.params.get('compat_opts', []))
1944
1945 return (
1946 'best/bestvideo+bestaudio' if prefer_best
1947 else 'bestvideo*+bestaudio/best' if not compat
1948 else 'bestvideo+bestaudio/best')
1949
1950 def build_format_selector(self, format_spec):
1951 def syntax_error(note, start):
1952 message = (
1953 'Invalid format specification: '
1954 '{0}\n\t{1}\n\t{2}^'.format(note, format_spec, ' ' * start[1]))
1955 return SyntaxError(message)
1956
1957 PICKFIRST = 'PICKFIRST'
1958 MERGE = 'MERGE'
1959 SINGLE = 'SINGLE'
1960 GROUP = 'GROUP'
1961 FormatSelector = collections.namedtuple('FormatSelector', ['type', 'selector', 'filters'])
1962
1963 allow_multiple_streams = {'audio': self.params.get('allow_multiple_audio_streams', False),
1964 'video': self.params.get('allow_multiple_video_streams', False)}
1965
1966 check_formats = self.params.get('check_formats') == 'selected'
1967
1968 def _parse_filter(tokens):
1969 filter_parts = []
1970 for type, string, start, _, _ in tokens:
1971 if type == tokenize.OP and string == ']':
1972 return ''.join(filter_parts)
1973 else:
1974 filter_parts.append(string)
1975
1976 def _remove_unused_ops(tokens):
1977 # Remove operators that we don't use and join them with the surrounding strings
1978 # for example: 'mp4' '-' 'baseline' '-' '16x9' is converted to 'mp4-baseline-16x9'
1979 ALLOWED_OPS = ('/', '+', ',', '(', ')')
1980 last_string, last_start, last_end, last_line = None, None, None, None
1981 for type, string, start, end, line in tokens:
1982 if type == tokenize.OP and string == '[':
1983 if last_string:
1984 yield tokenize.NAME, last_string, last_start, last_end, last_line
1985 last_string = None
1986 yield type, string, start, end, line
1987 # everything inside brackets will be handled by _parse_filter
1988 for type, string, start, end, line in tokens:
1989 yield type, string, start, end, line
1990 if type == tokenize.OP and string == ']':
1991 break
1992 elif type == tokenize.OP and string in ALLOWED_OPS:
1993 if last_string:
1994 yield tokenize.NAME, last_string, last_start, last_end, last_line
1995 last_string = None
1996 yield type, string, start, end, line
1997 elif type in [tokenize.NAME, tokenize.NUMBER, tokenize.OP]:
1998 if not last_string:
1999 last_string = string
2000 last_start = start
2001 last_end = end
2002 else:
2003 last_string += string
2004 if last_string:
2005 yield tokenize.NAME, last_string, last_start, last_end, last_line
2006
2007 def _parse_format_selection(tokens, inside_merge=False, inside_choice=False, inside_group=False):
2008 selectors = []
2009 current_selector = None
2010 for type, string, start, _, _ in tokens:
2011 # ENCODING is only defined in python 3.x
2012 if type == getattr(tokenize, 'ENCODING', None):
2013 continue
2014 elif type in [tokenize.NAME, tokenize.NUMBER]:
2015 current_selector = FormatSelector(SINGLE, string, [])
2016 elif type == tokenize.OP:
2017 if string == ')':
2018 if not inside_group:
2019 # ')' will be handled by the parentheses group
2020 tokens.restore_last_token()
2021 break
2022 elif inside_merge and string in ['/', ',']:
2023 tokens.restore_last_token()
2024 break
2025 elif inside_choice and string == ',':
2026 tokens.restore_last_token()
2027 break
2028 elif string == ',':
2029 if not current_selector:
2030 raise syntax_error('"," must follow a format selector', start)
2031 selectors.append(current_selector)
2032 current_selector = None
2033 elif string == '/':
2034 if not current_selector:
2035 raise syntax_error('"/" must follow a format selector', start)
2036 first_choice = current_selector
2037 second_choice = _parse_format_selection(tokens, inside_choice=True)
2038 current_selector = FormatSelector(PICKFIRST, (first_choice, second_choice), [])
2039 elif string == '[':
2040 if not current_selector:
2041 current_selector = FormatSelector(SINGLE, 'best', [])
2042 format_filter = _parse_filter(tokens)
2043 current_selector.filters.append(format_filter)
2044 elif string == '(':
2045 if current_selector:
2046 raise syntax_error('Unexpected "("', start)
2047 group = _parse_format_selection(tokens, inside_group=True)
2048 current_selector = FormatSelector(GROUP, group, [])
2049 elif string == '+':
2050 if not current_selector:
2051 raise syntax_error('Unexpected "+"', start)
2052 selector_1 = current_selector
2053 selector_2 = _parse_format_selection(tokens, inside_merge=True)
2054 if not selector_2:
2055 raise syntax_error('Expected a selector', start)
2056 current_selector = FormatSelector(MERGE, (selector_1, selector_2), [])
2057 else:
2058 raise syntax_error('Operator not recognized: "{0}"'.format(string), start)
2059 elif type == tokenize.ENDMARKER:
2060 break
2061 if current_selector:
2062 selectors.append(current_selector)
2063 return selectors
2064
2065 def _merge(formats_pair):
2066 format_1, format_2 = formats_pair
2067
2068 formats_info = []
2069 formats_info.extend(format_1.get('requested_formats', (format_1,)))
2070 formats_info.extend(format_2.get('requested_formats', (format_2,)))
2071
2072 if not allow_multiple_streams['video'] or not allow_multiple_streams['audio']:
2073 get_no_more = {'video': False, 'audio': False}
2074 for (i, fmt_info) in enumerate(formats_info):
2075 if fmt_info.get('acodec') == fmt_info.get('vcodec') == 'none':
2076 formats_info.pop(i)
2077 continue
2078 for aud_vid in ['audio', 'video']:
2079 if not allow_multiple_streams[aud_vid] and fmt_info.get(aud_vid[0] + 'codec') != 'none':
2080 if get_no_more[aud_vid]:
2081 formats_info.pop(i)
2082 break
2083 get_no_more[aud_vid] = True
2084
2085 if len(formats_info) == 1:
2086 return formats_info[0]
2087
2088 video_fmts = [fmt_info for fmt_info in formats_info if fmt_info.get('vcodec') != 'none']
2089 audio_fmts = [fmt_info for fmt_info in formats_info if fmt_info.get('acodec') != 'none']
2090
2091 the_only_video = video_fmts[0] if len(video_fmts) == 1 else None
2092 the_only_audio = audio_fmts[0] if len(audio_fmts) == 1 else None
2093
2094 output_ext = self.params.get('merge_output_format')
2095 if not output_ext:
2096 if the_only_video:
2097 output_ext = the_only_video['ext']
2098 elif the_only_audio and not video_fmts:
2099 output_ext = the_only_audio['ext']
2100 else:
2101 output_ext = 'mkv'
2102
2103 filtered = lambda *keys: filter(None, (traverse_obj(fmt, *keys) for fmt in formats_info))
2104
2105 new_dict = {
2106 'requested_formats': formats_info,
2107 'format': '+'.join(filtered('format')),
2108 'format_id': '+'.join(filtered('format_id')),
2109 'ext': output_ext,
2110 'protocol': '+'.join(map(determine_protocol, formats_info)),
2111 'language': '+'.join(orderedSet(filtered('language'))) or None,
2112 'format_note': '+'.join(orderedSet(filtered('format_note'))) or None,
2113 'filesize_approx': sum(filtered('filesize', 'filesize_approx')) or None,
2114 'tbr': sum(filtered('tbr', 'vbr', 'abr')),
2115 }
2116
2117 if the_only_video:
2118 new_dict.update({
2119 'width': the_only_video.get('width'),
2120 'height': the_only_video.get('height'),
2121 'resolution': the_only_video.get('resolution') or self.format_resolution(the_only_video),
2122 'fps': the_only_video.get('fps'),
2123 'dynamic_range': the_only_video.get('dynamic_range'),
2124 'vcodec': the_only_video.get('vcodec'),
2125 'vbr': the_only_video.get('vbr'),
2126 'stretched_ratio': the_only_video.get('stretched_ratio'),
2127 })
2128
2129 if the_only_audio:
2130 new_dict.update({
2131 'acodec': the_only_audio.get('acodec'),
2132 'abr': the_only_audio.get('abr'),
2133 'asr': the_only_audio.get('asr'),
2134 })
2135
2136 return new_dict
2137
2138 def _check_formats(formats):
2139 if not check_formats:
2140 yield from formats
2141 return
2142 yield from self._check_formats(formats)
2143
2144 def _build_selector_function(selector):
2145 if isinstance(selector, list): # ,
2146 fs = [_build_selector_function(s) for s in selector]
2147
2148 def selector_function(ctx):
2149 for f in fs:
2150 yield from f(ctx)
2151 return selector_function
2152
2153 elif selector.type == GROUP: # ()
2154 selector_function = _build_selector_function(selector.selector)
2155
2156 elif selector.type == PICKFIRST: # /
2157 fs = [_build_selector_function(s) for s in selector.selector]
2158
2159 def selector_function(ctx):
2160 for f in fs:
2161 picked_formats = list(f(ctx))
2162 if picked_formats:
2163 return picked_formats
2164 return []
2165
2166 elif selector.type == MERGE: # +
2167 selector_1, selector_2 = map(_build_selector_function, selector.selector)
2168
2169 def selector_function(ctx):
2170 for pair in itertools.product(selector_1(ctx), selector_2(ctx)):
2171 yield _merge(pair)
2172
2173 elif selector.type == SINGLE: # atom
2174 format_spec = selector.selector or 'best'
2175
2176 # TODO: Add allvideo, allaudio etc by generalizing the code with best/worst selector
2177 if format_spec == 'all':
2178 def selector_function(ctx):
2179 yield from _check_formats(ctx['formats'][::-1])
2180 elif format_spec == 'mergeall':
2181 def selector_function(ctx):
2182 formats = list(_check_formats(ctx['formats']))
2183 if not formats:
2184 return
2185 merged_format = formats[-1]
2186 for f in formats[-2::-1]:
2187 merged_format = _merge((merged_format, f))
2188 yield merged_format
2189
2190 else:
2191 format_fallback, seperate_fallback, format_reverse, format_idx = False, None, True, 1
2192 mobj = re.match(
2193 r'(?P<bw>best|worst|b|w)(?P<type>video|audio|v|a)?(?P<mod>\*)?(?:\.(?P<n>[1-9]\d*))?$',
2194 format_spec)
2195 if mobj is not None:
2196 format_idx = int_or_none(mobj.group('n'), default=1)
2197 format_reverse = mobj.group('bw')[0] == 'b'
2198 format_type = (mobj.group('type') or [None])[0]
2199 not_format_type = {'v': 'a', 'a': 'v'}.get(format_type)
2200 format_modified = mobj.group('mod') is not None
2201
2202 format_fallback = not format_type and not format_modified # for b, w
2203 _filter_f = (
2204 (lambda f: f.get('%scodec' % format_type) != 'none')
2205 if format_type and format_modified # bv*, ba*, wv*, wa*
2206 else (lambda f: f.get('%scodec' % not_format_type) == 'none')
2207 if format_type # bv, ba, wv, wa
2208 else (lambda f: f.get('vcodec') != 'none' and f.get('acodec') != 'none')
2209 if not format_modified # b, w
2210 else lambda f: True) # b*, w*
2211 filter_f = lambda f: _filter_f(f) and (
2212 f.get('vcodec') != 'none' or f.get('acodec') != 'none')
2213 else:
2214 if format_spec in self._format_selection_exts['audio']:
2215 filter_f = lambda f: f.get('ext') == format_spec and f.get('acodec') != 'none'
2216 elif format_spec in self._format_selection_exts['video']:
2217 filter_f = lambda f: f.get('ext') == format_spec and f.get('acodec') != 'none' and f.get('vcodec') != 'none'
2218 seperate_fallback = lambda f: f.get('ext') == format_spec and f.get('vcodec') != 'none'
2219 elif format_spec in self._format_selection_exts['storyboards']:
2220 filter_f = lambda f: f.get('ext') == format_spec and f.get('acodec') == 'none' and f.get('vcodec') == 'none'
2221 else:
2222 filter_f = lambda f: f.get('format_id') == format_spec # id
2223
2224 def selector_function(ctx):
2225 formats = list(ctx['formats'])
2226 matches = list(filter(filter_f, formats)) if filter_f is not None else formats
2227 if not matches:
2228 if format_fallback and ctx['incomplete_formats']:
2229 # for extractors with incomplete formats (audio only (soundcloud)
2230 # or video only (imgur)) best/worst will fallback to
2231 # best/worst {video,audio}-only format
2232 matches = formats
2233 elif seperate_fallback and not ctx['has_merged_format']:
2234 # for compatibility with youtube-dl when there is no pre-merged format
2235 matches = list(filter(seperate_fallback, formats))
2236 matches = LazyList(_check_formats(matches[::-1 if format_reverse else 1]))
2237 try:
2238 yield matches[format_idx - 1]
2239 except IndexError:
2240 return
2241
2242 filters = [self._build_format_filter(f) for f in selector.filters]
2243
2244 def final_selector(ctx):
2245 ctx_copy = dict(ctx)
2246 for _filter in filters:
2247 ctx_copy['formats'] = list(filter(_filter, ctx_copy['formats']))
2248 return selector_function(ctx_copy)
2249 return final_selector
2250
2251 stream = io.BytesIO(format_spec.encode('utf-8'))
2252 try:
2253 tokens = list(_remove_unused_ops(compat_tokenize_tokenize(stream.readline)))
2254 except tokenize.TokenError:
2255 raise syntax_error('Missing closing/opening brackets or parenthesis', (0, len(format_spec)))
2256
2257 class TokenIterator(object):
2258 def __init__(self, tokens):
2259 self.tokens = tokens
2260 self.counter = 0
2261
2262 def __iter__(self):
2263 return self
2264
2265 def __next__(self):
2266 if self.counter >= len(self.tokens):
2267 raise StopIteration()
2268 value = self.tokens[self.counter]
2269 self.counter += 1
2270 return value
2271
2272 next = __next__
2273
2274 def restore_last_token(self):
2275 self.counter -= 1
2276
2277 parsed_selector = _parse_format_selection(iter(TokenIterator(tokens)))
2278 return _build_selector_function(parsed_selector)
2279
2280 def _calc_headers(self, info_dict):
2281 res = merge_headers(self.params['http_headers'], info_dict.get('http_headers') or {})
2282
2283 cookies = self._calc_cookies(info_dict)
2284 if cookies:
2285 res['Cookie'] = cookies
2286
2287 if 'X-Forwarded-For' not in res:
2288 x_forwarded_for_ip = info_dict.get('__x_forwarded_for_ip')
2289 if x_forwarded_for_ip:
2290 res['X-Forwarded-For'] = x_forwarded_for_ip
2291
2292 return res
2293
2294 def _calc_cookies(self, info_dict):
2295 pr = sanitized_Request(info_dict['url'])
2296 self.cookiejar.add_cookie_header(pr)
2297 return pr.get_header('Cookie')
2298
2299 def _sort_thumbnails(self, thumbnails):
2300 thumbnails.sort(key=lambda t: (
2301 t.get('preference') if t.get('preference') is not None else -1,
2302 t.get('width') if t.get('width') is not None else -1,
2303 t.get('height') if t.get('height') is not None else -1,
2304 t.get('id') if t.get('id') is not None else '',
2305 t.get('url')))
2306
2307 def _sanitize_thumbnails(self, info_dict):
2308 thumbnails = info_dict.get('thumbnails')
2309 if thumbnails is None:
2310 thumbnail = info_dict.get('thumbnail')
2311 if thumbnail:
2312 info_dict['thumbnails'] = thumbnails = [{'url': thumbnail}]
2313 if not thumbnails:
2314 return
2315
2316 def check_thumbnails(thumbnails):
2317 for t in thumbnails:
2318 self.to_screen(f'[info] Testing thumbnail {t["id"]}')
2319 try:
2320 self.urlopen(HEADRequest(t['url']))
2321 except network_exceptions as err:
2322 self.to_screen(f'[info] Unable to connect to thumbnail {t["id"]} URL {t["url"]!r} - {err}. Skipping...')
2323 continue
2324 yield t
2325
2326 self._sort_thumbnails(thumbnails)
2327 for i, t in enumerate(thumbnails):
2328 if t.get('id') is None:
2329 t['id'] = '%d' % i
2330 if t.get('width') and t.get('height'):
2331 t['resolution'] = '%dx%d' % (t['width'], t['height'])
2332 t['url'] = sanitize_url(t['url'])
2333
2334 if self.params.get('check_formats') is True:
2335 info_dict['thumbnails'] = LazyList(check_thumbnails(thumbnails[::-1]), reverse=True)
2336 else:
2337 info_dict['thumbnails'] = thumbnails
2338
2339 def _fill_common_fields(self, info_dict, is_video=True):
2340 # TODO: move sanitization here
2341 if is_video:
2342 # playlists are allowed to lack "title"
2343 info_dict['fulltitle'] = info_dict.get('title')
2344 if 'title' not in info_dict:
2345 raise ExtractorError('Missing "title" field in extractor result',
2346 video_id=info_dict['id'], ie=info_dict['extractor'])
2347 elif not info_dict.get('title'):
2348 self.report_warning('Extractor failed to obtain "title". Creating a generic title instead')
2349 info_dict['title'] = f'{info_dict["extractor"]} video #{info_dict["id"]}'
2350
2351 if info_dict.get('duration') is not None:
2352 info_dict['duration_string'] = formatSeconds(info_dict['duration'])
2353
2354 for ts_key, date_key in (
2355 ('timestamp', 'upload_date'),
2356 ('release_timestamp', 'release_date'),
2357 ('modified_timestamp', 'modified_date'),
2358 ):
2359 if info_dict.get(date_key) is None and info_dict.get(ts_key) is not None:
2360 # Working around out-of-range timestamp values (e.g. negative ones on Windows,
2361 # see http://bugs.python.org/issue1646728)
2362 try:
2363 upload_date = datetime.datetime.utcfromtimestamp(info_dict[ts_key])
2364 info_dict[date_key] = upload_date.strftime('%Y%m%d')
2365 except (ValueError, OverflowError, OSError):
2366 pass
2367
2368 live_keys = ('is_live', 'was_live')
2369 live_status = info_dict.get('live_status')
2370 if live_status is None:
2371 for key in live_keys:
2372 if info_dict.get(key) is False:
2373 continue
2374 if info_dict.get(key):
2375 live_status = key
2376 break
2377 if all(info_dict.get(key) is False for key in live_keys):
2378 live_status = 'not_live'
2379 if live_status:
2380 info_dict['live_status'] = live_status
2381 for key in live_keys:
2382 if info_dict.get(key) is None:
2383 info_dict[key] = (live_status == key)
2384
2385 # Auto generate title fields corresponding to the *_number fields when missing
2386 # in order to always have clean titles. This is very common for TV series.
2387 for field in ('chapter', 'season', 'episode'):
2388 if info_dict.get('%s_number' % field) is not None and not info_dict.get(field):
2389 info_dict[field] = '%s %d' % (field.capitalize(), info_dict['%s_number' % field])
2390
2391 def process_video_result(self, info_dict, download=True):
2392 assert info_dict.get('_type', 'video') == 'video'
2393 self._num_videos += 1
2394
2395 if 'id' not in info_dict:
2396 raise ExtractorError('Missing "id" field in extractor result', ie=info_dict['extractor'])
2397 elif not info_dict.get('id'):
2398 raise ExtractorError('Extractor failed to obtain "id"', ie=info_dict['extractor'])
2399
2400 def report_force_conversion(field, field_not, conversion):
2401 self.report_warning(
2402 '"%s" field is not %s - forcing %s conversion, there is an error in extractor'
2403 % (field, field_not, conversion))
2404
2405 def sanitize_string_field(info, string_field):
2406 field = info.get(string_field)
2407 if field is None or isinstance(field, compat_str):
2408 return
2409 report_force_conversion(string_field, 'a string', 'string')
2410 info[string_field] = compat_str(field)
2411
2412 def sanitize_numeric_fields(info):
2413 for numeric_field in self._NUMERIC_FIELDS:
2414 field = info.get(numeric_field)
2415 if field is None or isinstance(field, compat_numeric_types):
2416 continue
2417 report_force_conversion(numeric_field, 'numeric', 'int')
2418 info[numeric_field] = int_or_none(field)
2419
2420 sanitize_string_field(info_dict, 'id')
2421 sanitize_numeric_fields(info_dict)
2422 if (info_dict.get('duration') or 0) <= 0 and info_dict.pop('duration', None):
2423 self.report_warning('"duration" field is negative, there is an error in extractor')
2424
2425 if 'playlist' not in info_dict:
2426 # It isn't part of a playlist
2427 info_dict['playlist'] = None
2428 info_dict['playlist_index'] = None
2429
2430 self._sanitize_thumbnails(info_dict)
2431
2432 thumbnail = info_dict.get('thumbnail')
2433 thumbnails = info_dict.get('thumbnails')
2434 if thumbnail:
2435 info_dict['thumbnail'] = sanitize_url(thumbnail)
2436 elif thumbnails:
2437 info_dict['thumbnail'] = thumbnails[-1]['url']
2438
2439 if info_dict.get('display_id') is None and 'id' in info_dict:
2440 info_dict['display_id'] = info_dict['id']
2441
2442 self._fill_common_fields(info_dict)
2443
2444 for cc_kind in ('subtitles', 'automatic_captions'):
2445 cc = info_dict.get(cc_kind)
2446 if cc:
2447 for _, subtitle in cc.items():
2448 for subtitle_format in subtitle:
2449 if subtitle_format.get('url'):
2450 subtitle_format['url'] = sanitize_url(subtitle_format['url'])
2451 if subtitle_format.get('ext') is None:
2452 subtitle_format['ext'] = determine_ext(subtitle_format['url']).lower()
2453
2454 automatic_captions = info_dict.get('automatic_captions')
2455 subtitles = info_dict.get('subtitles')
2456
2457 info_dict['requested_subtitles'] = self.process_subtitles(
2458 info_dict['id'], subtitles, automatic_captions)
2459
2460 if info_dict.get('formats') is None:
2461 # There's only one format available
2462 formats = [info_dict]
2463 else:
2464 formats = info_dict['formats']
2465
2466 info_dict['__has_drm'] = any(f.get('has_drm') for f in formats)
2467 if not self.params.get('allow_unplayable_formats'):
2468 formats = [f for f in formats if not f.get('has_drm')]
2469 if info_dict['__has_drm'] and all(
2470 f.get('acodec') == f.get('vcodec') == 'none' for f in formats):
2471 self.report_warning(
2472 'This video is DRM protected and only images are available for download. '
2473 'Use --list-formats to see them')
2474
2475 get_from_start = not info_dict.get('is_live') or bool(self.params.get('live_from_start'))
2476 if not get_from_start:
2477 info_dict['title'] += ' ' + datetime.datetime.now().strftime('%Y-%m-%d %H:%M')
2478 if info_dict.get('is_live') and formats:
2479 formats = [f for f in formats if bool(f.get('is_from_start')) == get_from_start]
2480 if get_from_start and not formats:
2481 self.raise_no_formats(info_dict, msg=(
2482 '--live-from-start is passed, but there are no formats that can be downloaded from the start. '
2483 'If you want to download from the current time, use --no-live-from-start'))
2484
2485 if not formats:
2486 self.raise_no_formats(info_dict)
2487
2488 def is_wellformed(f):
2489 url = f.get('url')
2490 if not url:
2491 self.report_warning(
2492 '"url" field is missing or empty - skipping format, '
2493 'there is an error in extractor')
2494 return False
2495 if isinstance(url, bytes):
2496 sanitize_string_field(f, 'url')
2497 return True
2498
2499 # Filter out malformed formats for better extraction robustness
2500 formats = list(filter(is_wellformed, formats))
2501
2502 formats_dict = {}
2503
2504 # We check that all the formats have the format and format_id fields
2505 for i, format in enumerate(formats):
2506 sanitize_string_field(format, 'format_id')
2507 sanitize_numeric_fields(format)
2508 format['url'] = sanitize_url(format['url'])
2509 if not format.get('format_id'):
2510 format['format_id'] = compat_str(i)
2511 else:
2512 # Sanitize format_id from characters used in format selector expression
2513 format['format_id'] = re.sub(r'[\s,/+\[\]()]', '_', format['format_id'])
2514 format_id = format['format_id']
2515 if format_id not in formats_dict:
2516 formats_dict[format_id] = []
2517 formats_dict[format_id].append(format)
2518
2519 # Make sure all formats have unique format_id
2520 common_exts = set(itertools.chain(*self._format_selection_exts.values()))
2521 for format_id, ambiguous_formats in formats_dict.items():
2522 ambigious_id = len(ambiguous_formats) > 1
2523 for i, format in enumerate(ambiguous_formats):
2524 if ambigious_id:
2525 format['format_id'] = '%s-%d' % (format_id, i)
2526 if format.get('ext') is None:
2527 format['ext'] = determine_ext(format['url']).lower()
2528 # Ensure there is no conflict between id and ext in format selection
2529 # See https://github.com/yt-dlp/yt-dlp/issues/1282
2530 if format['format_id'] != format['ext'] and format['format_id'] in common_exts:
2531 format['format_id'] = 'f%s' % format['format_id']
2532
2533 for i, format in enumerate(formats):
2534 if format.get('format') is None:
2535 format['format'] = '{id} - {res}{note}'.format(
2536 id=format['format_id'],
2537 res=self.format_resolution(format),
2538 note=format_field(format, 'format_note', ' (%s)'),
2539 )
2540 if format.get('protocol') is None:
2541 format['protocol'] = determine_protocol(format)
2542 if format.get('resolution') is None:
2543 format['resolution'] = self.format_resolution(format, default=None)
2544 if format.get('dynamic_range') is None and format.get('vcodec') != 'none':
2545 format['dynamic_range'] = 'SDR'
2546 if (info_dict.get('duration') and format.get('tbr')
2547 and not format.get('filesize') and not format.get('filesize_approx')):
2548 format['filesize_approx'] = info_dict['duration'] * format['tbr'] * (1024 / 8)
2549
2550 # Add HTTP headers, so that external programs can use them from the
2551 # json output
2552 full_format_info = info_dict.copy()
2553 full_format_info.update(format)
2554 format['http_headers'] = self._calc_headers(full_format_info)
2555 # Remove private housekeeping stuff
2556 if '__x_forwarded_for_ip' in info_dict:
2557 del info_dict['__x_forwarded_for_ip']
2558
2559 if self.params.get('check_formats') is True:
2560 formats = LazyList(self._check_formats(formats[::-1]), reverse=True)
2561
2562 if not formats or formats[0] is not info_dict:
2563 # only set the 'formats' fields if the original info_dict list them
2564 # otherwise we end up with a circular reference, the first (and unique)
2565 # element in the 'formats' field in info_dict is info_dict itself,
2566 # which can't be exported to json
2567 info_dict['formats'] = formats
2568
2569 info_dict, _ = self.pre_process(info_dict)
2570
2571 if self._match_entry(info_dict, incomplete=self._format_fields) is not None:
2572 return info_dict
2573
2574 self.post_extract(info_dict)
2575 info_dict, _ = self.pre_process(info_dict, 'after_filter')
2576
2577 # The pre-processors may have modified the formats
2578 formats = info_dict.get('formats', [info_dict])
2579
2580 list_only = self.params.get('simulate') is None and (
2581 self.params.get('list_thumbnails') or self.params.get('listformats') or self.params.get('listsubtitles'))
2582 interactive_format_selection = not list_only and self.format_selector == '-'
2583 if self.params.get('list_thumbnails'):
2584 self.list_thumbnails(info_dict)
2585 if self.params.get('listsubtitles'):
2586 if 'automatic_captions' in info_dict:
2587 self.list_subtitles(
2588 info_dict['id'], automatic_captions, 'automatic captions')
2589 self.list_subtitles(info_dict['id'], subtitles, 'subtitles')
2590 if self.params.get('listformats') or interactive_format_selection:
2591 self.list_formats(info_dict)
2592 if list_only:
2593 # Without this printing, -F --print-json will not work
2594 self.__forced_printings(info_dict, self.prepare_filename(info_dict), incomplete=True)
2595 return
2596
2597 format_selector = self.format_selector
2598 if format_selector is None:
2599 req_format = self._default_format_spec(info_dict, download=download)
2600 self.write_debug('Default format spec: %s' % req_format)
2601 format_selector = self.build_format_selector(req_format)
2602
2603 while True:
2604 if interactive_format_selection:
2605 req_format = input(
2606 self._format_screen('\nEnter format selector: ', self.Styles.EMPHASIS))
2607 try:
2608 format_selector = self.build_format_selector(req_format)
2609 except SyntaxError as err:
2610 self.report_error(err, tb=False, is_error=False)
2611 continue
2612
2613 formats_to_download = list(format_selector({
2614 'formats': formats,
2615 'has_merged_format': any('none' not in (f.get('acodec'), f.get('vcodec')) for f in formats),
2616 'incomplete_formats': (
2617 # All formats are video-only or
2618 all(f.get('vcodec') != 'none' and f.get('acodec') == 'none' for f in formats)
2619 # all formats are audio-only
2620 or all(f.get('vcodec') == 'none' and f.get('acodec') != 'none' for f in formats)),
2621 }))
2622 if interactive_format_selection and not formats_to_download:
2623 self.report_error('Requested format is not available', tb=False, is_error=False)
2624 continue
2625 break
2626
2627 if not formats_to_download:
2628 if not self.params.get('ignore_no_formats_error'):
2629 raise ExtractorError(
2630 'Requested format is not available. Use --list-formats for a list of available formats',
2631 expected=True, video_id=info_dict['id'], ie=info_dict['extractor'])
2632 self.report_warning('Requested format is not available')
2633 # Process what we can, even without any available formats.
2634 formats_to_download = [{}]
2635
2636 best_format = formats_to_download[-1]
2637 if download:
2638 if best_format:
2639 self.to_screen(
2640 f'[info] {info_dict["id"]}: Downloading {len(formats_to_download)} format(s): '
2641 + ', '.join([f['format_id'] for f in formats_to_download]))
2642 max_downloads_reached = False
2643 for i, fmt in enumerate(formats_to_download):
2644 formats_to_download[i] = new_info = self._copy_infodict(info_dict)
2645 new_info.update(fmt)
2646 try:
2647 self.process_info(new_info)
2648 except MaxDownloadsReached:
2649 max_downloads_reached = True
2650 # Remove copied info
2651 for key, val in tuple(new_info.items()):
2652 if info_dict.get(key) == val:
2653 new_info.pop(key)
2654 if max_downloads_reached:
2655 break
2656
2657 write_archive = set(f.get('__write_download_archive', False) for f in formats_to_download)
2658 assert write_archive.issubset({True, False, 'ignore'})
2659 if True in write_archive and False not in write_archive:
2660 self.record_download_archive(info_dict)
2661
2662 info_dict['requested_downloads'] = formats_to_download
2663 info_dict = self.run_all_pps('after_video', info_dict)
2664 if max_downloads_reached:
2665 raise MaxDownloadsReached()
2666
2667 # We update the info dict with the selected best quality format (backwards compatibility)
2668 info_dict.update(best_format)
2669 return info_dict
2670
2671 def process_subtitles(self, video_id, normal_subtitles, automatic_captions):
2672 """Select the requested subtitles and their format"""
2673 available_subs = {}
2674 if normal_subtitles and self.params.get('writesubtitles'):
2675 available_subs.update(normal_subtitles)
2676 if automatic_captions and self.params.get('writeautomaticsub'):
2677 for lang, cap_info in automatic_captions.items():
2678 if lang not in available_subs:
2679 available_subs[lang] = cap_info
2680
2681 if (not self.params.get('writesubtitles') and not
2682 self.params.get('writeautomaticsub') or not
2683 available_subs):
2684 return None
2685
2686 all_sub_langs = available_subs.keys()
2687 if self.params.get('allsubtitles', False):
2688 requested_langs = all_sub_langs
2689 elif self.params.get('subtitleslangs', False):
2690 # A list is used so that the order of languages will be the same as
2691 # given in subtitleslangs. See https://github.com/yt-dlp/yt-dlp/issues/1041
2692 requested_langs = []
2693 for lang_re in self.params.get('subtitleslangs'):
2694 discard = lang_re[0] == '-'
2695 if discard:
2696 lang_re = lang_re[1:]
2697 if lang_re == 'all':
2698 if discard:
2699 requested_langs = []
2700 else:
2701 requested_langs.extend(all_sub_langs)
2702 continue
2703 current_langs = filter(re.compile(lang_re + '$').match, all_sub_langs)
2704 if discard:
2705 for lang in current_langs:
2706 while lang in requested_langs:
2707 requested_langs.remove(lang)
2708 else:
2709 requested_langs.extend(current_langs)
2710 requested_langs = orderedSet(requested_langs)
2711 elif 'en' in available_subs:
2712 requested_langs = ['en']
2713 else:
2714 requested_langs = [list(all_sub_langs)[0]]
2715 if requested_langs:
2716 self.write_debug('Downloading subtitles: %s' % ', '.join(requested_langs))
2717
2718 formats_query = self.params.get('subtitlesformat', 'best')
2719 formats_preference = formats_query.split('/') if formats_query else []
2720 subs = {}
2721 for lang in requested_langs:
2722 formats = available_subs.get(lang)
2723 if formats is None:
2724 self.report_warning('%s subtitles not available for %s' % (lang, video_id))
2725 continue
2726 for ext in formats_preference:
2727 if ext == 'best':
2728 f = formats[-1]
2729 break
2730 matches = list(filter(lambda f: f['ext'] == ext, formats))
2731 if matches:
2732 f = matches[-1]
2733 break
2734 else:
2735 f = formats[-1]
2736 self.report_warning(
2737 'No subtitle format found matching "%s" for language %s, '
2738 'using %s' % (formats_query, lang, f['ext']))
2739 subs[lang] = f
2740 return subs
2741
2742 def _forceprint(self, key, info_dict):
2743 if info_dict is None:
2744 return
2745 info_copy = info_dict.copy()
2746 info_copy['formats_table'] = self.render_formats_table(info_dict)
2747 info_copy['thumbnails_table'] = self.render_thumbnails_table(info_dict)
2748 info_copy['subtitles_table'] = self.render_subtitles_table(info_dict.get('id'), info_dict.get('subtitles'))
2749 info_copy['automatic_captions_table'] = self.render_subtitles_table(info_dict.get('id'), info_dict.get('automatic_captions'))
2750
2751 def format_tmpl(tmpl):
2752 mobj = re.match(r'\w+(=?)$', tmpl)
2753 if mobj and mobj.group(1):
2754 return f'{tmpl[:-1]} = %({tmpl[:-1]})r'
2755 elif mobj:
2756 return f'%({tmpl})s'
2757 return tmpl
2758
2759 for tmpl in self.params['forceprint'].get(key, []):
2760 self.to_stdout(self.evaluate_outtmpl(format_tmpl(tmpl), info_copy))
2761
2762 for tmpl, file_tmpl in self.params['print_to_file'].get(key, []):
2763 filename = self.prepare_filename(info_dict, outtmpl=file_tmpl)
2764 tmpl = format_tmpl(tmpl)
2765 self.to_screen(f'[info] Writing {tmpl!r} to: {filename}')
2766 if self._ensure_dir_exists(filename):
2767 with io.open(filename, 'a', encoding='utf-8') as f:
2768 f.write(self.evaluate_outtmpl(tmpl, info_copy) + '\n')
2769
2770 def __forced_printings(self, info_dict, filename, incomplete):
2771 def print_mandatory(field, actual_field=None):
2772 if actual_field is None:
2773 actual_field = field
2774 if (self.params.get('force%s' % field, False)
2775 and (not incomplete or info_dict.get(actual_field) is not None)):
2776 self.to_stdout(info_dict[actual_field])
2777
2778 def print_optional(field):
2779 if (self.params.get('force%s' % field, False)
2780 and info_dict.get(field) is not None):
2781 self.to_stdout(info_dict[field])
2782
2783 info_dict = info_dict.copy()
2784 if filename is not None:
2785 info_dict['filename'] = filename
2786 if info_dict.get('requested_formats') is not None:
2787 # For RTMP URLs, also include the playpath
2788 info_dict['urls'] = '\n'.join(f['url'] + f.get('play_path', '') for f in info_dict['requested_formats'])
2789 elif info_dict.get('url'):
2790 info_dict['urls'] = info_dict['url'] + info_dict.get('play_path', '')
2791
2792 if (self.params.get('forcejson')
2793 or self.params['forceprint'].get('video')
2794 or self.params['print_to_file'].get('video')):
2795 self.post_extract(info_dict)
2796 self._forceprint('video', info_dict)
2797
2798 print_mandatory('title')
2799 print_mandatory('id')
2800 print_mandatory('url', 'urls')
2801 print_optional('thumbnail')
2802 print_optional('description')
2803 print_optional('filename')
2804 if self.params.get('forceduration') and info_dict.get('duration') is not None:
2805 self.to_stdout(formatSeconds(info_dict['duration']))
2806 print_mandatory('format')
2807
2808 if self.params.get('forcejson'):
2809 self.to_stdout(json.dumps(self.sanitize_info(info_dict)))
2810
2811 def dl(self, name, info, subtitle=False, test=False):
2812 if not info.get('url'):
2813 self.raise_no_formats(info, True)
2814
2815 if test:
2816 verbose = self.params.get('verbose')
2817 params = {
2818 'test': True,
2819 'quiet': self.params.get('quiet') or not verbose,
2820 'verbose': verbose,
2821 'noprogress': not verbose,
2822 'nopart': True,
2823 'skip_unavailable_fragments': False,
2824 'keep_fragments': False,
2825 'overwrites': True,
2826 '_no_ytdl_file': True,
2827 }
2828 else:
2829 params = self.params
2830 fd = get_suitable_downloader(info, params, to_stdout=(name == '-'))(self, params)
2831 if not test:
2832 for ph in self._progress_hooks:
2833 fd.add_progress_hook(ph)
2834 urls = '", "'.join(
2835 (f['url'].split(',')[0] + ',<data>' if f['url'].startswith('data:') else f['url'])
2836 for f in info.get('requested_formats', []) or [info])
2837 self.write_debug('Invoking downloader on "%s"' % urls)
2838
2839 # Note: Ideally info should be a deep-copied so that hooks cannot modify it.
2840 # But it may contain objects that are not deep-copyable
2841 new_info = self._copy_infodict(info)
2842 if new_info.get('http_headers') is None:
2843 new_info['http_headers'] = self._calc_headers(new_info)
2844 return fd.download(name, new_info, subtitle)
2845
2846 def existing_file(self, filepaths, *, default_overwrite=True):
2847 existing_files = list(filter(os.path.exists, orderedSet(filepaths)))
2848 if existing_files and not self.params.get('overwrites', default_overwrite):
2849 return existing_files[0]
2850
2851 for file in existing_files:
2852 self.report_file_delete(file)
2853 os.remove(file)
2854 return None
2855
2856 def process_info(self, info_dict):
2857 """Process a single resolved IE result. (Modifies it in-place)"""
2858
2859 assert info_dict.get('_type', 'video') == 'video'
2860 original_infodict = info_dict
2861
2862 if 'format' not in info_dict and 'ext' in info_dict:
2863 info_dict['format'] = info_dict['ext']
2864
2865 # This is mostly just for backward compatibility of process_info
2866 # As a side-effect, this allows for format-specific filters
2867 if self._match_entry(info_dict) is not None:
2868 info_dict['__write_download_archive'] = 'ignore'
2869 return
2870
2871 # Does nothing under normal operation - for backward compatibility of process_info
2872 self.post_extract(info_dict)
2873 self._num_downloads += 1
2874
2875 # info_dict['_filename'] needs to be set for backward compatibility
2876 info_dict['_filename'] = full_filename = self.prepare_filename(info_dict, warn=True)
2877 temp_filename = self.prepare_filename(info_dict, 'temp')
2878 files_to_move = {}
2879
2880 # Forced printings
2881 self.__forced_printings(info_dict, full_filename, incomplete=('format' not in info_dict))
2882
2883 if self.params.get('simulate'):
2884 info_dict['__write_download_archive'] = self.params.get('force_write_download_archive')
2885 return
2886
2887 if full_filename is None:
2888 return
2889 if not self._ensure_dir_exists(encodeFilename(full_filename)):
2890 return
2891 if not self._ensure_dir_exists(encodeFilename(temp_filename)):
2892 return
2893
2894 if self._write_description('video', info_dict,
2895 self.prepare_filename(info_dict, 'description')) is None:
2896 return
2897
2898 sub_files = self._write_subtitles(info_dict, temp_filename)
2899 if sub_files is None:
2900 return
2901 files_to_move.update(dict(sub_files))
2902
2903 thumb_files = self._write_thumbnails(
2904 'video', info_dict, temp_filename, self.prepare_filename(info_dict, 'thumbnail'))
2905 if thumb_files is None:
2906 return
2907 files_to_move.update(dict(thumb_files))
2908
2909 infofn = self.prepare_filename(info_dict, 'infojson')
2910 _infojson_written = self._write_info_json('video', info_dict, infofn)
2911 if _infojson_written:
2912 info_dict['infojson_filename'] = infofn
2913 # For backward compatibility, even though it was a private field
2914 info_dict['__infojson_filename'] = infofn
2915 elif _infojson_written is None:
2916 return
2917
2918 # Note: Annotations are deprecated
2919 annofn = None
2920 if self.params.get('writeannotations', False):
2921 annofn = self.prepare_filename(info_dict, 'annotation')
2922 if annofn:
2923 if not self._ensure_dir_exists(encodeFilename(annofn)):
2924 return
2925 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(annofn)):
2926 self.to_screen('[info] Video annotations are already present')
2927 elif not info_dict.get('annotations'):
2928 self.report_warning('There are no annotations to write.')
2929 else:
2930 try:
2931 self.to_screen('[info] Writing video annotations to: ' + annofn)
2932 with io.open(encodeFilename(annofn), 'w', encoding='utf-8') as annofile:
2933 annofile.write(info_dict['annotations'])
2934 except (KeyError, TypeError):
2935 self.report_warning('There are no annotations to write.')
2936 except (OSError, IOError):
2937 self.report_error('Cannot write annotations file: ' + annofn)
2938 return
2939
2940 # Write internet shortcut files
2941 def _write_link_file(link_type):
2942 url = try_get(info_dict['webpage_url'], iri_to_uri)
2943 if not url:
2944 self.report_warning(
2945 f'Cannot write internet shortcut file because the actual URL of "{info_dict["webpage_url"]}" is unknown')
2946 return True
2947 linkfn = replace_extension(self.prepare_filename(info_dict, 'link'), link_type, info_dict.get('ext'))
2948 if not self._ensure_dir_exists(encodeFilename(linkfn)):
2949 return False
2950 if self.params.get('overwrites', True) and os.path.exists(encodeFilename(linkfn)):
2951 self.to_screen(f'[info] Internet shortcut (.{link_type}) is already present')
2952 return True
2953 try:
2954 self.to_screen(f'[info] Writing internet shortcut (.{link_type}) to: {linkfn}')
2955 with io.open(encodeFilename(to_high_limit_path(linkfn)), 'w', encoding='utf-8',
2956 newline='\r\n' if link_type == 'url' else '\n') as linkfile:
2957 template_vars = {'url': url}
2958 if link_type == 'desktop':
2959 template_vars['filename'] = linkfn[:-(len(link_type) + 1)]
2960 linkfile.write(LINK_TEMPLATES[link_type] % template_vars)
2961 except (OSError, IOError):
2962 self.report_error(f'Cannot write internet shortcut {linkfn}')
2963 return False
2964 return True
2965
2966 write_links = {
2967 'url': self.params.get('writeurllink'),
2968 'webloc': self.params.get('writewebloclink'),
2969 'desktop': self.params.get('writedesktoplink'),
2970 }
2971 if self.params.get('writelink'):
2972 link_type = ('webloc' if sys.platform == 'darwin'
2973 else 'desktop' if sys.platform.startswith('linux')
2974 else 'url')
2975 write_links[link_type] = True
2976
2977 if any(should_write and not _write_link_file(link_type)
2978 for link_type, should_write in write_links.items()):
2979 return
2980
2981 def replace_info_dict(new_info):
2982 nonlocal info_dict
2983 if new_info == info_dict:
2984 return
2985 info_dict.clear()
2986 info_dict.update(new_info)
2987
2988 try:
2989 new_info, files_to_move = self.pre_process(info_dict, 'before_dl', files_to_move)
2990 replace_info_dict(new_info)
2991 except PostProcessingError as err:
2992 self.report_error('Preprocessing: %s' % str(err))
2993 return
2994
2995 if self.params.get('skip_download'):
2996 info_dict['filepath'] = temp_filename
2997 info_dict['__finaldir'] = os.path.dirname(os.path.abspath(encodeFilename(full_filename)))
2998 info_dict['__files_to_move'] = files_to_move
2999 replace_info_dict(self.run_pp(MoveFilesAfterDownloadPP(self, False), info_dict))
3000 info_dict['__write_download_archive'] = self.params.get('force_write_download_archive')
3001 else:
3002 # Download
3003 info_dict.setdefault('__postprocessors', [])
3004 try:
3005
3006 def existing_video_file(*filepaths):
3007 ext = info_dict.get('ext')
3008 converted = lambda file: replace_extension(file, self.params.get('final_ext') or ext, ext)
3009 file = self.existing_file(itertools.chain(*zip(map(converted, filepaths), filepaths)),
3010 default_overwrite=False)
3011 if file:
3012 info_dict['ext'] = os.path.splitext(file)[1][1:]
3013 return file
3014
3015 success = True
3016 if info_dict.get('requested_formats') is not None:
3017
3018 def compatible_formats(formats):
3019 # TODO: some formats actually allow this (mkv, webm, ogg, mp4), but not all of them.
3020 video_formats = [format for format in formats if format.get('vcodec') != 'none']
3021 audio_formats = [format for format in formats if format.get('acodec') != 'none']
3022 if len(video_formats) > 2 or len(audio_formats) > 2:
3023 return False
3024
3025 # Check extension
3026 exts = set(format.get('ext') for format in formats)
3027 COMPATIBLE_EXTS = (
3028 set(('mp3', 'mp4', 'm4a', 'm4p', 'm4b', 'm4r', 'm4v', 'ismv', 'isma')),
3029 set(('webm',)),
3030 )
3031 for ext_sets in COMPATIBLE_EXTS:
3032 if ext_sets.issuperset(exts):
3033 return True
3034 # TODO: Check acodec/vcodec
3035 return False
3036
3037 requested_formats = info_dict['requested_formats']
3038 old_ext = info_dict['ext']
3039 if self.params.get('merge_output_format') is None:
3040 if not compatible_formats(requested_formats):
3041 info_dict['ext'] = 'mkv'
3042 self.report_warning(
3043 'Requested formats are incompatible for merge and will be merged into mkv')
3044 if (info_dict['ext'] == 'webm'
3045 and info_dict.get('thumbnails')
3046 # check with type instead of pp_key, __name__, or isinstance
3047 # since we dont want any custom PPs to trigger this
3048 and any(type(pp) == EmbedThumbnailPP for pp in self._pps['post_process'])):
3049 info_dict['ext'] = 'mkv'
3050 self.report_warning(
3051 'webm doesn\'t support embedding a thumbnail, mkv will be used')
3052 new_ext = info_dict['ext']
3053
3054 def correct_ext(filename, ext=new_ext):
3055 if filename == '-':
3056 return filename
3057 filename_real_ext = os.path.splitext(filename)[1][1:]
3058 filename_wo_ext = (
3059 os.path.splitext(filename)[0]
3060 if filename_real_ext in (old_ext, new_ext)
3061 else filename)
3062 return '%s.%s' % (filename_wo_ext, ext)
3063
3064 # Ensure filename always has a correct extension for successful merge
3065 full_filename = correct_ext(full_filename)
3066 temp_filename = correct_ext(temp_filename)
3067 dl_filename = existing_video_file(full_filename, temp_filename)
3068 info_dict['__real_download'] = False
3069
3070 downloaded = []
3071 merger = FFmpegMergerPP(self)
3072
3073 fd = get_suitable_downloader(info_dict, self.params, to_stdout=temp_filename == '-')
3074 if dl_filename is not None:
3075 self.report_file_already_downloaded(dl_filename)
3076 elif fd:
3077 for f in requested_formats if fd != FFmpegFD else []:
3078 f['filepath'] = fname = prepend_extension(
3079 correct_ext(temp_filename, info_dict['ext']),
3080 'f%s' % f['format_id'], info_dict['ext'])
3081 downloaded.append(fname)
3082 info_dict['url'] = '\n'.join(f['url'] for f in requested_formats)
3083 success, real_download = self.dl(temp_filename, info_dict)
3084 info_dict['__real_download'] = real_download
3085 else:
3086 if self.params.get('allow_unplayable_formats'):
3087 self.report_warning(
3088 'You have requested merging of multiple formats '
3089 'while also allowing unplayable formats to be downloaded. '
3090 'The formats won\'t be merged to prevent data corruption.')
3091 elif not merger.available:
3092 msg = 'You have requested merging of multiple formats but ffmpeg is not installed'
3093 if not self.params.get('ignoreerrors'):
3094 self.report_error(f'{msg}. Aborting due to --abort-on-error')
3095 return
3096 self.report_warning(f'{msg}. The formats won\'t be merged')
3097
3098 if temp_filename == '-':
3099 reason = ('using a downloader other than ffmpeg' if FFmpegFD.can_merge_formats(info_dict, self.params)
3100 else 'but the formats are incompatible for simultaneous download' if merger.available
3101 else 'but ffmpeg is not installed')
3102 self.report_warning(
3103 f'You have requested downloading multiple formats to stdout {reason}. '
3104 'The formats will be streamed one after the other')
3105 fname = temp_filename
3106 for f in requested_formats:
3107 new_info = dict(info_dict)
3108 del new_info['requested_formats']
3109 new_info.update(f)
3110 if temp_filename != '-':
3111 fname = prepend_extension(
3112 correct_ext(temp_filename, new_info['ext']),
3113 'f%s' % f['format_id'], new_info['ext'])
3114 if not self._ensure_dir_exists(fname):
3115 return
3116 f['filepath'] = fname
3117 downloaded.append(fname)
3118 partial_success, real_download = self.dl(fname, new_info)
3119 info_dict['__real_download'] = info_dict['__real_download'] or real_download
3120 success = success and partial_success
3121
3122 if downloaded and merger.available and not self.params.get('allow_unplayable_formats'):
3123 info_dict['__postprocessors'].append(merger)
3124 info_dict['__files_to_merge'] = downloaded
3125 # Even if there were no downloads, it is being merged only now
3126 info_dict['__real_download'] = True
3127 else:
3128 for file in downloaded:
3129 files_to_move[file] = None
3130 else:
3131 # Just a single file
3132 dl_filename = existing_video_file(full_filename, temp_filename)
3133 if dl_filename is None or dl_filename == temp_filename:
3134 # dl_filename == temp_filename could mean that the file was partially downloaded with --no-part.
3135 # So we should try to resume the download
3136 success, real_download = self.dl(temp_filename, info_dict)
3137 info_dict['__real_download'] = real_download
3138 else:
3139 self.report_file_already_downloaded(dl_filename)
3140
3141 dl_filename = dl_filename or temp_filename
3142 info_dict['__finaldir'] = os.path.dirname(os.path.abspath(encodeFilename(full_filename)))
3143
3144 except network_exceptions as err:
3145 self.report_error('unable to download video data: %s' % error_to_compat_str(err))
3146 return
3147 except (OSError, IOError) as err:
3148 raise UnavailableVideoError(err)
3149 except (ContentTooShortError, ) as err:
3150 self.report_error('content too short (expected %s bytes and served %s)' % (err.expected, err.downloaded))
3151 return
3152
3153 if success and full_filename != '-':
3154
3155 def fixup():
3156 do_fixup = True
3157 fixup_policy = self.params.get('fixup')
3158 vid = info_dict['id']
3159
3160 if fixup_policy in ('ignore', 'never'):
3161 return
3162 elif fixup_policy == 'warn':
3163 do_fixup = False
3164 elif fixup_policy != 'force':
3165 assert fixup_policy in ('detect_or_warn', None)
3166 if not info_dict.get('__real_download'):
3167 do_fixup = False
3168
3169 def ffmpeg_fixup(cndn, msg, cls):
3170 if not cndn:
3171 return
3172 if not do_fixup:
3173 self.report_warning(f'{vid}: {msg}')
3174 return
3175 pp = cls(self)
3176 if pp.available:
3177 info_dict['__postprocessors'].append(pp)
3178 else:
3179 self.report_warning(f'{vid}: {msg}. Install ffmpeg to fix this automatically')
3180
3181 stretched_ratio = info_dict.get('stretched_ratio')
3182 ffmpeg_fixup(
3183 stretched_ratio not in (1, None),
3184 f'Non-uniform pixel ratio {stretched_ratio}',
3185 FFmpegFixupStretchedPP)
3186
3187 ffmpeg_fixup(
3188 (info_dict.get('requested_formats') is None
3189 and info_dict.get('container') == 'm4a_dash'
3190 and info_dict.get('ext') == 'm4a'),
3191 'writing DASH m4a. Only some players support this container',
3192 FFmpegFixupM4aPP)
3193
3194 downloader = get_suitable_downloader(info_dict, self.params) if 'protocol' in info_dict else None
3195 downloader = downloader.__name__ if downloader else None
3196
3197 if info_dict.get('requested_formats') is None: # Not necessary if doing merger
3198 ffmpeg_fixup(downloader == 'HlsFD',
3199 'Possible MPEG-TS in MP4 container or malformed AAC timestamps',
3200 FFmpegFixupM3u8PP)
3201 ffmpeg_fixup(info_dict.get('is_live') and downloader == 'DashSegmentsFD',
3202 'Possible duplicate MOOV atoms', FFmpegFixupDuplicateMoovPP)
3203
3204 ffmpeg_fixup(downloader == 'WebSocketFragmentFD', 'Malformed timestamps detected', FFmpegFixupTimestampPP)
3205 ffmpeg_fixup(downloader == 'WebSocketFragmentFD', 'Malformed duration detected', FFmpegFixupDurationPP)
3206
3207 fixup()
3208 try:
3209 replace_info_dict(self.post_process(dl_filename, info_dict, files_to_move))
3210 except PostProcessingError as err:
3211 self.report_error('Postprocessing: %s' % str(err))
3212 return
3213 try:
3214 for ph in self._post_hooks:
3215 ph(info_dict['filepath'])
3216 except Exception as err:
3217 self.report_error('post hooks: %s' % str(err))
3218 return
3219 info_dict['__write_download_archive'] = True
3220
3221 if self.params.get('force_write_download_archive'):
3222 info_dict['__write_download_archive'] = True
3223
3224 # Make sure the info_dict was modified in-place
3225 assert info_dict is original_infodict
3226
3227 max_downloads = self.params.get('max_downloads')
3228 if max_downloads is not None and self._num_downloads >= int(max_downloads):
3229 raise MaxDownloadsReached()
3230
3231 def __download_wrapper(self, func):
3232 @functools.wraps(func)
3233 def wrapper(*args, **kwargs):
3234 try:
3235 res = func(*args, **kwargs)
3236 except UnavailableVideoError as e:
3237 self.report_error(e)
3238 except MaxDownloadsReached as e:
3239 self.to_screen(f'[info] {e}')
3240 raise
3241 except DownloadCancelled as e:
3242 self.to_screen(f'[info] {e}')
3243 if not self.params.get('break_per_url'):
3244 raise
3245 else:
3246 if self.params.get('dump_single_json', False):
3247 self.post_extract(res)
3248 self.to_stdout(json.dumps(self.sanitize_info(res)))
3249 return wrapper
3250
3251 def download(self, url_list):
3252 """Download a given list of URLs."""
3253 url_list = variadic(url_list) # Passing a single URL is a common mistake
3254 outtmpl = self.outtmpl_dict['default']
3255 if (len(url_list) > 1
3256 and outtmpl != '-'
3257 and '%' not in outtmpl
3258 and self.params.get('max_downloads') != 1):
3259 raise SameFileError(outtmpl)
3260
3261 for url in url_list:
3262 self.__download_wrapper(self.extract_info)(
3263 url, force_generic_extractor=self.params.get('force_generic_extractor', False))
3264
3265 return self._download_retcode
3266
3267 def download_with_info_file(self, info_filename):
3268 with contextlib.closing(fileinput.FileInput(
3269 [info_filename], mode='r',
3270 openhook=fileinput.hook_encoded('utf-8'))) as f:
3271 # FileInput doesn't have a read method, we can't call json.load
3272 info = self.sanitize_info(json.loads('\n'.join(f)), self.params.get('clean_infojson', True))
3273 try:
3274 self.__download_wrapper(self.process_ie_result)(info, download=True)
3275 except (DownloadError, EntryNotInPlaylist, ReExtractInfo) as e:
3276 if not isinstance(e, EntryNotInPlaylist):
3277 self.to_stderr('\r')
3278 webpage_url = info.get('webpage_url')
3279 if webpage_url is not None:
3280 self.report_warning(f'The info failed to download: {e}; trying with URL {webpage_url}')
3281 return self.download([webpage_url])
3282 else:
3283 raise
3284 return self._download_retcode
3285
3286 @staticmethod
3287 def sanitize_info(info_dict, remove_private_keys=False):
3288 ''' Sanitize the infodict for converting to json '''
3289 if info_dict is None:
3290 return info_dict
3291 info_dict.setdefault('epoch', int(time.time()))
3292 info_dict.setdefault('_type', 'video')
3293
3294 if remove_private_keys:
3295 reject = lambda k, v: v is None or (k.startswith('_') and k != '_type') or k in {
3296 'requested_downloads', 'requested_formats', 'requested_subtitles', 'requested_entries',
3297 'entries', 'filepath', 'infojson_filename', 'original_url', 'playlist_autonumber',
3298 }
3299 else:
3300 reject = lambda k, v: False
3301
3302 def filter_fn(obj):
3303 if isinstance(obj, dict):
3304 return {k: filter_fn(v) for k, v in obj.items() if not reject(k, v)}
3305 elif isinstance(obj, (list, tuple, set, LazyList)):
3306 return list(map(filter_fn, obj))
3307 elif obj is None or isinstance(obj, (str, int, float, bool)):
3308 return obj
3309 else:
3310 return repr(obj)
3311
3312 return filter_fn(info_dict)
3313
3314 @staticmethod
3315 def filter_requested_info(info_dict, actually_filter=True):
3316 ''' Alias of sanitize_info for backward compatibility '''
3317 return YoutubeDL.sanitize_info(info_dict, actually_filter)
3318
3319 @staticmethod
3320 def post_extract(info_dict):
3321 def actual_post_extract(info_dict):
3322 if info_dict.get('_type') in ('playlist', 'multi_video'):
3323 for video_dict in info_dict.get('entries', {}):
3324 actual_post_extract(video_dict or {})
3325 return
3326
3327 post_extractor = info_dict.pop('__post_extractor', None) or (lambda: {})
3328 info_dict.update(post_extractor())
3329
3330 actual_post_extract(info_dict or {})
3331
3332 def run_pp(self, pp, infodict):
3333 files_to_delete = []
3334 if '__files_to_move' not in infodict:
3335 infodict['__files_to_move'] = {}
3336 try:
3337 files_to_delete, infodict = pp.run(infodict)
3338 except PostProcessingError as e:
3339 # Must be True and not 'only_download'
3340 if self.params.get('ignoreerrors') is True:
3341 self.report_error(e)
3342 return infodict
3343 raise
3344
3345 if not files_to_delete:
3346 return infodict
3347 if self.params.get('keepvideo', False):
3348 for f in files_to_delete:
3349 infodict['__files_to_move'].setdefault(f, '')
3350 else:
3351 for old_filename in set(files_to_delete):
3352 self.to_screen('Deleting original file %s (pass -k to keep)' % old_filename)
3353 try:
3354 os.remove(encodeFilename(old_filename))
3355 except (IOError, OSError):
3356 self.report_warning('Unable to remove downloaded original file')
3357 if old_filename in infodict['__files_to_move']:
3358 del infodict['__files_to_move'][old_filename]
3359 return infodict
3360
3361 def run_all_pps(self, key, info, *, additional_pps=None):
3362 self._forceprint(key, info)
3363 for pp in (additional_pps or []) + self._pps[key]:
3364 info = self.run_pp(pp, info)
3365 return info
3366
3367 def pre_process(self, ie_info, key='pre_process', files_to_move=None):
3368 info = dict(ie_info)
3369 info['__files_to_move'] = files_to_move or {}
3370 info = self.run_all_pps(key, info)
3371 return info, info.pop('__files_to_move', None)
3372
3373 def post_process(self, filename, info, files_to_move=None):
3374 """Run all the postprocessors on the given file."""
3375 info['filepath'] = filename
3376 info['__files_to_move'] = files_to_move or {}
3377 info = self.run_all_pps('post_process', info, additional_pps=info.get('__postprocessors'))
3378 info = self.run_pp(MoveFilesAfterDownloadPP(self), info)
3379 del info['__files_to_move']
3380 return self.run_all_pps('after_move', info)
3381
3382 def _make_archive_id(self, info_dict):
3383 video_id = info_dict.get('id')
3384 if not video_id:
3385 return
3386 # Future-proof against any change in case
3387 # and backwards compatibility with prior versions
3388 extractor = info_dict.get('extractor_key') or info_dict.get('ie_key') # key in a playlist
3389 if extractor is None:
3390 url = str_or_none(info_dict.get('url'))
3391 if not url:
3392 return
3393 # Try to find matching extractor for the URL and take its ie_key
3394 for ie_key, ie in self._ies.items():
3395 if ie.suitable(url):
3396 extractor = ie_key
3397 break
3398 else:
3399 return
3400 return '%s %s' % (extractor.lower(), video_id)
3401
3402 def in_download_archive(self, info_dict):
3403 fn = self.params.get('download_archive')
3404 if fn is None:
3405 return False
3406
3407 vid_id = self._make_archive_id(info_dict)
3408 if not vid_id:
3409 return False # Incomplete video information
3410
3411 return vid_id in self.archive
3412
3413 def record_download_archive(self, info_dict):
3414 fn = self.params.get('download_archive')
3415 if fn is None:
3416 return
3417 vid_id = self._make_archive_id(info_dict)
3418 assert vid_id
3419 self.write_debug(f'Adding to archive: {vid_id}')
3420 with locked_file(fn, 'a', encoding='utf-8') as archive_file:
3421 archive_file.write(vid_id + '\n')
3422 self.archive.add(vid_id)
3423
3424 @staticmethod
3425 def format_resolution(format, default='unknown'):
3426 if format.get('vcodec') == 'none' and format.get('acodec') != 'none':
3427 return 'audio only'
3428 if format.get('resolution') is not None:
3429 return format['resolution']
3430 if format.get('width') and format.get('height'):
3431 return '%dx%d' % (format['width'], format['height'])
3432 elif format.get('height'):
3433 return '%sp' % format['height']
3434 elif format.get('width'):
3435 return '%dx?' % format['width']
3436 return default
3437
3438 def _list_format_headers(self, *headers):
3439 if self.params.get('listformats_table', True) is not False:
3440 return [self._format_screen(header, self.Styles.HEADERS) for header in headers]
3441 return headers
3442
3443 def _format_note(self, fdict):
3444 res = ''
3445 if fdict.get('ext') in ['f4f', 'f4m']:
3446 res += '(unsupported)'
3447 if fdict.get('language'):
3448 if res:
3449 res += ' '
3450 res += '[%s]' % fdict['language']
3451 if fdict.get('format_note') is not None:
3452 if res:
3453 res += ' '
3454 res += fdict['format_note']
3455 if fdict.get('tbr') is not None:
3456 if res:
3457 res += ', '
3458 res += '%4dk' % fdict['tbr']
3459 if fdict.get('container') is not None:
3460 if res:
3461 res += ', '
3462 res += '%s container' % fdict['container']
3463 if (fdict.get('vcodec') is not None
3464 and fdict.get('vcodec') != 'none'):
3465 if res:
3466 res += ', '
3467 res += fdict['vcodec']
3468 if fdict.get('vbr') is not None:
3469 res += '@'
3470 elif fdict.get('vbr') is not None and fdict.get('abr') is not None:
3471 res += 'video@'
3472 if fdict.get('vbr') is not None:
3473 res += '%4dk' % fdict['vbr']
3474 if fdict.get('fps') is not None:
3475 if res:
3476 res += ', '
3477 res += '%sfps' % fdict['fps']
3478 if fdict.get('acodec') is not None:
3479 if res:
3480 res += ', '
3481 if fdict['acodec'] == 'none':
3482 res += 'video only'
3483 else:
3484 res += '%-5s' % fdict['acodec']
3485 elif fdict.get('abr') is not None:
3486 if res:
3487 res += ', '
3488 res += 'audio'
3489 if fdict.get('abr') is not None:
3490 res += '@%3dk' % fdict['abr']
3491 if fdict.get('asr') is not None:
3492 res += ' (%5dHz)' % fdict['asr']
3493 if fdict.get('filesize') is not None:
3494 if res:
3495 res += ', '
3496 res += format_bytes(fdict['filesize'])
3497 elif fdict.get('filesize_approx') is not None:
3498 if res:
3499 res += ', '
3500 res += '~' + format_bytes(fdict['filesize_approx'])
3501 return res
3502
3503 def render_formats_table(self, info_dict):
3504 if not info_dict.get('formats') and not info_dict.get('url'):
3505 return None
3506
3507 formats = info_dict.get('formats', [info_dict])
3508 if not self.params.get('listformats_table', True) is not False:
3509 table = [
3510 [
3511 format_field(f, 'format_id'),
3512 format_field(f, 'ext'),
3513 self.format_resolution(f),
3514 self._format_note(f)
3515 ] for f in formats if f.get('preference') is None or f['preference'] >= -1000]
3516 return render_table(['format code', 'extension', 'resolution', 'note'], table, extra_gap=1)
3517
3518 delim = self._format_screen('\u2502', self.Styles.DELIM, '|', test_encoding=True)
3519 table = [
3520 [
3521 self._format_screen(format_field(f, 'format_id'), self.Styles.ID),
3522 format_field(f, 'ext'),
3523 format_field(f, func=self.format_resolution, ignore=('audio only', 'images')),
3524 format_field(f, 'fps', '\t%d'),
3525 format_field(f, 'dynamic_range', '%s', ignore=(None, 'SDR')).replace('HDR', ''),
3526 delim,
3527 format_field(f, 'filesize', ' \t%s', func=format_bytes) + format_field(f, 'filesize_approx', '~\t%s', func=format_bytes),
3528 format_field(f, 'tbr', '\t%dk'),
3529 shorten_protocol_name(f.get('protocol', '')),
3530 delim,
3531 format_field(f, 'vcodec', default='unknown').replace(
3532 'none', 'images' if f.get('acodec') == 'none'
3533 else self._format_screen('audio only', self.Styles.SUPPRESS)),
3534 format_field(f, 'vbr', '\t%dk'),
3535 format_field(f, 'acodec', default='unknown').replace(
3536 'none', '' if f.get('vcodec') == 'none'
3537 else self._format_screen('video only', self.Styles.SUPPRESS)),
3538 format_field(f, 'abr', '\t%dk'),
3539 format_field(f, 'asr', '\t%dHz'),
3540 join_nonempty(
3541 self._format_screen('UNSUPPORTED', 'light red') if f.get('ext') in ('f4f', 'f4m') else None,
3542 format_field(f, 'language', '[%s]'),
3543 join_nonempty(format_field(f, 'format_note'),
3544 format_field(f, 'container', ignore=(None, f.get('ext'))),
3545 delim=', '),
3546 delim=' '),
3547 ] for f in formats if f.get('preference') is None or f['preference'] >= -1000]
3548 header_line = self._list_format_headers(
3549 'ID', 'EXT', 'RESOLUTION', '\tFPS', 'HDR', delim, '\tFILESIZE', '\tTBR', 'PROTO',
3550 delim, 'VCODEC', '\tVBR', 'ACODEC', '\tABR', '\tASR', 'MORE INFO')
3551
3552 return render_table(
3553 header_line, table, hide_empty=True,
3554 delim=self._format_screen('\u2500', self.Styles.DELIM, '-', test_encoding=True))
3555
3556 def render_thumbnails_table(self, info_dict):
3557 thumbnails = list(info_dict.get('thumbnails') or [])
3558 if not thumbnails:
3559 return None
3560 return render_table(
3561 self._list_format_headers('ID', 'Width', 'Height', 'URL'),
3562 [[t.get('id'), t.get('width', 'unknown'), t.get('height', 'unknown'), t['url']] for t in thumbnails])
3563
3564 def render_subtitles_table(self, video_id, subtitles):
3565 def _row(lang, formats):
3566 exts, names = zip(*((f['ext'], f.get('name') or 'unknown') for f in reversed(formats)))
3567 if len(set(names)) == 1:
3568 names = [] if names[0] == 'unknown' else names[:1]
3569 return [lang, ', '.join(names), ', '.join(exts)]
3570
3571 if not subtitles:
3572 return None
3573 return render_table(
3574 self._list_format_headers('Language', 'Name', 'Formats'),
3575 [_row(lang, formats) for lang, formats in subtitles.items()],
3576 hide_empty=True)
3577
3578 def __list_table(self, video_id, name, func, *args):
3579 table = func(*args)
3580 if not table:
3581 self.to_screen(f'{video_id} has no {name}')
3582 return
3583 self.to_screen(f'[info] Available {name} for {video_id}:')
3584 self.to_stdout(table)
3585
3586 def list_formats(self, info_dict):
3587 self.__list_table(info_dict['id'], 'formats', self.render_formats_table, info_dict)
3588
3589 def list_thumbnails(self, info_dict):
3590 self.__list_table(info_dict['id'], 'thumbnails', self.render_thumbnails_table, info_dict)
3591
3592 def list_subtitles(self, video_id, subtitles, name='subtitles'):
3593 self.__list_table(video_id, name, self.render_subtitles_table, video_id, subtitles)
3594
3595 def urlopen(self, req):
3596 """ Start an HTTP download """
3597 if isinstance(req, compat_basestring):
3598 req = sanitized_Request(req)
3599 return self._opener.open(req, timeout=self._socket_timeout)
3600
3601 def print_debug_header(self):
3602 if not self.params.get('verbose'):
3603 return
3604
3605 def get_encoding(stream):
3606 ret = str(getattr(stream, 'encoding', 'missing (%s)' % type(stream).__name__))
3607 if not supports_terminal_sequences(stream):
3608 from .compat import WINDOWS_VT_MODE
3609 ret += ' (No VT)' if WINDOWS_VT_MODE is False else ' (No ANSI)'
3610 return ret
3611
3612 encoding_str = 'Encodings: locale %s, fs %s, out %s, err %s, pref %s' % (
3613 locale.getpreferredencoding(),
3614 sys.getfilesystemencoding(),
3615 get_encoding(self._out_files['screen']), get_encoding(self._out_files['error']),
3616 self.get_encoding())
3617
3618 logger = self.params.get('logger')
3619 if logger:
3620 write_debug = lambda msg: logger.debug(f'[debug] {msg}')
3621 write_debug(encoding_str)
3622 else:
3623 write_string(f'[debug] {encoding_str}\n', encoding=None)
3624 write_debug = lambda msg: self._write_string(f'[debug] {msg}\n')
3625
3626 source = detect_variant()
3627 write_debug(join_nonempty(
3628 'yt-dlp version', __version__,
3629 f'[{RELEASE_GIT_HEAD}]' if RELEASE_GIT_HEAD else '',
3630 '' if source == 'unknown' else f'({source})',
3631 delim=' '))
3632 if not _LAZY_LOADER:
3633 if os.environ.get('YTDLP_NO_LAZY_EXTRACTORS'):
3634 write_debug('Lazy loading extractors is forcibly disabled')
3635 else:
3636 write_debug('Lazy loading extractors is disabled')
3637 if plugin_extractors or plugin_postprocessors:
3638 write_debug('Plugins: %s' % [
3639 '%s%s' % (klass.__name__, '' if klass.__name__ == name else f' as {name}')
3640 for name, klass in itertools.chain(plugin_extractors.items(), plugin_postprocessors.items())])
3641 if self.params.get('compat_opts'):
3642 write_debug('Compatibility options: %s' % ', '.join(self.params.get('compat_opts')))
3643
3644 if source == 'source':
3645 try:
3646 sp = Popen(
3647 ['git', 'rev-parse', '--short', 'HEAD'],
3648 stdout=subprocess.PIPE, stderr=subprocess.PIPE,
3649 cwd=os.path.dirname(os.path.abspath(__file__)))
3650 out, err = sp.communicate_or_kill()
3651 out = out.decode().strip()
3652 if re.match('[0-9a-f]+', out):
3653 write_debug('Git HEAD: %s' % out)
3654 except Exception:
3655 try:
3656 sys.exc_clear()
3657 except Exception:
3658 pass
3659
3660 def python_implementation():
3661 impl_name = platform.python_implementation()
3662 if impl_name == 'PyPy' and hasattr(sys, 'pypy_version_info'):
3663 return impl_name + ' version %d.%d.%d' % sys.pypy_version_info[:3]
3664 return impl_name
3665
3666 write_debug('Python version %s (%s %s) - %s' % (
3667 platform.python_version(),
3668 python_implementation(),
3669 platform.architecture()[0],
3670 platform_name()))
3671
3672 exe_versions, ffmpeg_features = FFmpegPostProcessor.get_versions_and_features(self)
3673 ffmpeg_features = {key for key, val in ffmpeg_features.items() if val}
3674 if ffmpeg_features:
3675 exe_versions['ffmpeg'] += ' (%s)' % ','.join(ffmpeg_features)
3676
3677 exe_versions['rtmpdump'] = rtmpdump_version()
3678 exe_versions['phantomjs'] = PhantomJSwrapper._version()
3679 exe_str = ', '.join(
3680 f'{exe} {v}' for exe, v in sorted(exe_versions.items()) if v
3681 ) or 'none'
3682 write_debug('exe versions: %s' % exe_str)
3683
3684 from .downloader.websocket import has_websockets
3685 from .postprocessor.embedthumbnail import has_mutagen
3686 from .cookies import SQLITE_AVAILABLE, SECRETSTORAGE_AVAILABLE
3687
3688 lib_str = join_nonempty(
3689 compat_brotli and compat_brotli.__name__,
3690 has_certifi and 'certifi',
3691 compat_pycrypto_AES and compat_pycrypto_AES.__name__.split('.')[0],
3692 SECRETSTORAGE_AVAILABLE and 'secretstorage',
3693 has_mutagen and 'mutagen',
3694 SQLITE_AVAILABLE and 'sqlite',
3695 has_websockets and 'websockets',
3696 delim=', ') or 'none'
3697 write_debug('Optional libraries: %s' % lib_str)
3698
3699 proxy_map = {}
3700 for handler in self._opener.handlers:
3701 if hasattr(handler, 'proxies'):
3702 proxy_map.update(handler.proxies)
3703 write_debug(f'Proxy map: {proxy_map}')
3704
3705 # Not implemented
3706 if False and self.params.get('call_home'):
3707 ipaddr = self.urlopen('https://yt-dl.org/ip').read().decode('utf-8')
3708 write_debug('Public IP address: %s' % ipaddr)
3709 latest_version = self.urlopen(
3710 'https://yt-dl.org/latest/version').read().decode('utf-8')
3711 if version_tuple(latest_version) > version_tuple(__version__):
3712 self.report_warning(
3713 'You are using an outdated version (newest version: %s)! '
3714 'See https://yt-dl.org/update if you need help updating.' %
3715 latest_version)
3716
3717 def _setup_opener(self):
3718 timeout_val = self.params.get('socket_timeout')
3719 self._socket_timeout = 20 if timeout_val is None else float(timeout_val)
3720
3721 opts_cookiesfrombrowser = self.params.get('cookiesfrombrowser')
3722 opts_cookiefile = self.params.get('cookiefile')
3723 opts_proxy = self.params.get('proxy')
3724
3725 self.cookiejar = load_cookies(opts_cookiefile, opts_cookiesfrombrowser, self)
3726
3727 cookie_processor = YoutubeDLCookieProcessor(self.cookiejar)
3728 if opts_proxy is not None:
3729 if opts_proxy == '':
3730 proxies = {}
3731 else:
3732 proxies = {'http': opts_proxy, 'https': opts_proxy}
3733 else:
3734 proxies = compat_urllib_request.getproxies()
3735 # Set HTTPS proxy to HTTP one if given (https://github.com/ytdl-org/youtube-dl/issues/805)
3736 if 'http' in proxies and 'https' not in proxies:
3737 proxies['https'] = proxies['http']
3738 proxy_handler = PerRequestProxyHandler(proxies)
3739
3740 debuglevel = 1 if self.params.get('debug_printtraffic') else 0
3741 https_handler = make_HTTPS_handler(self.params, debuglevel=debuglevel)
3742 ydlh = YoutubeDLHandler(self.params, debuglevel=debuglevel)
3743 redirect_handler = YoutubeDLRedirectHandler()
3744 data_handler = compat_urllib_request_DataHandler()
3745
3746 # When passing our own FileHandler instance, build_opener won't add the
3747 # default FileHandler and allows us to disable the file protocol, which
3748 # can be used for malicious purposes (see
3749 # https://github.com/ytdl-org/youtube-dl/issues/8227)
3750 file_handler = compat_urllib_request.FileHandler()
3751
3752 def file_open(*args, **kwargs):
3753 raise compat_urllib_error.URLError('file:// scheme is explicitly disabled in yt-dlp for security reasons')
3754 file_handler.file_open = file_open
3755
3756 opener = compat_urllib_request.build_opener(
3757 proxy_handler, https_handler, cookie_processor, ydlh, redirect_handler, data_handler, file_handler)
3758
3759 # Delete the default user-agent header, which would otherwise apply in
3760 # cases where our custom HTTP handler doesn't come into play
3761 # (See https://github.com/ytdl-org/youtube-dl/issues/1309 for details)
3762 opener.addheaders = []
3763 self._opener = opener
3764
3765 def encode(self, s):
3766 if isinstance(s, bytes):
3767 return s # Already encoded
3768
3769 try:
3770 return s.encode(self.get_encoding())
3771 except UnicodeEncodeError as err:
3772 err.reason = err.reason + '. Check your system encoding configuration or use the --encoding option.'
3773 raise
3774
3775 def get_encoding(self):
3776 encoding = self.params.get('encoding')
3777 if encoding is None:
3778 encoding = preferredencoding()
3779 return encoding
3780
3781 def _write_info_json(self, label, ie_result, infofn, overwrite=None):
3782 ''' Write infojson and returns True = written, 'exists' = Already exists, False = skip, None = error '''
3783 if overwrite is None:
3784 overwrite = self.params.get('overwrites', True)
3785 if not self.params.get('writeinfojson'):
3786 return False
3787 elif not infofn:
3788 self.write_debug(f'Skipping writing {label} infojson')
3789 return False
3790 elif not self._ensure_dir_exists(infofn):
3791 return None
3792 elif not overwrite and os.path.exists(infofn):
3793 self.to_screen(f'[info] {label.title()} metadata is already present')
3794 return 'exists'
3795
3796 self.to_screen(f'[info] Writing {label} metadata as JSON to: {infofn}')
3797 try:
3798 write_json_file(self.sanitize_info(ie_result, self.params.get('clean_infojson', True)), infofn)
3799 return True
3800 except (OSError, IOError):
3801 self.report_error(f'Cannot write {label} metadata to JSON file {infofn}')
3802 return None
3803
3804 def _write_description(self, label, ie_result, descfn):
3805 ''' Write description and returns True = written, False = skip, None = error '''
3806 if not self.params.get('writedescription'):
3807 return False
3808 elif not descfn:
3809 self.write_debug(f'Skipping writing {label} description')
3810 return False
3811 elif not self._ensure_dir_exists(descfn):
3812 return None
3813 elif not self.params.get('overwrites', True) and os.path.exists(descfn):
3814 self.to_screen(f'[info] {label.title()} description is already present')
3815 elif ie_result.get('description') is None:
3816 self.report_warning(f'There\'s no {label} description to write')
3817 return False
3818 else:
3819 try:
3820 self.to_screen(f'[info] Writing {label} description to: {descfn}')
3821 with io.open(encodeFilename(descfn), 'w', encoding='utf-8') as descfile:
3822 descfile.write(ie_result['description'])
3823 except (OSError, IOError):
3824 self.report_error(f'Cannot write {label} description file {descfn}')
3825 return None
3826 return True
3827
3828 def _write_subtitles(self, info_dict, filename):
3829 ''' Write subtitles to file and return list of (sub_filename, final_sub_filename); or None if error'''
3830 ret = []
3831 subtitles = info_dict.get('requested_subtitles')
3832 if not subtitles or not (self.params.get('writesubtitles') or self.params.get('writeautomaticsub')):
3833 # subtitles download errors are already managed as troubles in relevant IE
3834 # that way it will silently go on when used with unsupporting IE
3835 return ret
3836
3837 sub_filename_base = self.prepare_filename(info_dict, 'subtitle')
3838 if not sub_filename_base:
3839 self.to_screen('[info] Skipping writing video subtitles')
3840 return ret
3841 for sub_lang, sub_info in subtitles.items():
3842 sub_format = sub_info['ext']
3843 sub_filename = subtitles_filename(filename, sub_lang, sub_format, info_dict.get('ext'))
3844 sub_filename_final = subtitles_filename(sub_filename_base, sub_lang, sub_format, info_dict.get('ext'))
3845 existing_sub = self.existing_file((sub_filename_final, sub_filename))
3846 if existing_sub:
3847 self.to_screen(f'[info] Video subtitle {sub_lang}.{sub_format} is already present')
3848 sub_info['filepath'] = existing_sub
3849 ret.append((existing_sub, sub_filename_final))
3850 continue
3851
3852 self.to_screen(f'[info] Writing video subtitles to: {sub_filename}')
3853 if sub_info.get('data') is not None:
3854 try:
3855 # Use newline='' to prevent conversion of newline characters
3856 # See https://github.com/ytdl-org/youtube-dl/issues/10268
3857 with io.open(sub_filename, 'w', encoding='utf-8', newline='') as subfile:
3858 subfile.write(sub_info['data'])
3859 sub_info['filepath'] = sub_filename
3860 ret.append((sub_filename, sub_filename_final))
3861 continue
3862 except (OSError, IOError):
3863 self.report_error(f'Cannot write video subtitles file {sub_filename}')
3864 return None
3865
3866 try:
3867 sub_copy = sub_info.copy()
3868 sub_copy.setdefault('http_headers', info_dict.get('http_headers'))
3869 self.dl(sub_filename, sub_copy, subtitle=True)
3870 sub_info['filepath'] = sub_filename
3871 ret.append((sub_filename, sub_filename_final))
3872 except (DownloadError, ExtractorError, IOError, OSError, ValueError) + network_exceptions as err:
3873 msg = f'Unable to download video subtitles for {sub_lang!r}: {err}'
3874 if self.params.get('ignoreerrors') is not True: # False or 'only_download'
3875 if not self.params.get('ignoreerrors'):
3876 self.report_error(msg)
3877 raise DownloadError(msg)
3878 self.report_warning(msg)
3879 return ret
3880
3881 def _write_thumbnails(self, label, info_dict, filename, thumb_filename_base=None):
3882 ''' Write thumbnails to file and return list of (thumb_filename, final_thumb_filename) '''
3883 write_all = self.params.get('write_all_thumbnails', False)
3884 thumbnails, ret = [], []
3885 if write_all or self.params.get('writethumbnail', False):
3886 thumbnails = info_dict.get('thumbnails') or []
3887 multiple = write_all and len(thumbnails) > 1
3888
3889 if thumb_filename_base is None:
3890 thumb_filename_base = filename
3891 if thumbnails and not thumb_filename_base:
3892 self.write_debug(f'Skipping writing {label} thumbnail')
3893 return ret
3894
3895 for idx, t in list(enumerate(thumbnails))[::-1]:
3896 thumb_ext = (f'{t["id"]}.' if multiple else '') + determine_ext(t['url'], 'jpg')
3897 thumb_display_id = f'{label} thumbnail {t["id"]}'
3898 thumb_filename = replace_extension(filename, thumb_ext, info_dict.get('ext'))
3899 thumb_filename_final = replace_extension(thumb_filename_base, thumb_ext, info_dict.get('ext'))
3900
3901 existing_thumb = self.existing_file((thumb_filename_final, thumb_filename))
3902 if existing_thumb:
3903 self.to_screen('[info] %s is already present' % (
3904 thumb_display_id if multiple else f'{label} thumbnail').capitalize())
3905 t['filepath'] = existing_thumb
3906 ret.append((existing_thumb, thumb_filename_final))
3907 else:
3908 self.to_screen(f'[info] Downloading {thumb_display_id} ...')
3909 try:
3910 uf = self.urlopen(sanitized_Request(t['url'], headers=t.get('http_headers', {})))
3911 self.to_screen(f'[info] Writing {thumb_display_id} to: {thumb_filename}')
3912 with open(encodeFilename(thumb_filename), 'wb') as thumbf:
3913 shutil.copyfileobj(uf, thumbf)
3914 ret.append((thumb_filename, thumb_filename_final))
3915 t['filepath'] = thumb_filename
3916 except network_exceptions as err:
3917 thumbnails.pop(idx)
3918 self.report_warning(f'Unable to download {thumb_display_id}: {err}')
3919 if ret and not write_all:
3920 break
3921 return ret