]> jfr.im git - yt-dlp.git/blob - yt_dlp/YoutubeDL.py
[docs] Fix some mistakes and improve doc
[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 copy
9 import datetime
10 import errno
11 import fileinput
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
29 from string import ascii_letters
30 from zipimport import zipimporter
31
32 from .compat import (
33 compat_basestring,
34 compat_get_terminal_size,
35 compat_kwargs,
36 compat_numeric_types,
37 compat_os_name,
38 compat_shlex_quote,
39 compat_str,
40 compat_tokenize_tokenize,
41 compat_urllib_error,
42 compat_urllib_request,
43 compat_urllib_request_DataHandler,
44 )
45 from .cookies import load_cookies
46 from .utils import (
47 age_restricted,
48 args_to_str,
49 ContentTooShortError,
50 date_from_str,
51 DateRange,
52 DEFAULT_OUTTMPL,
53 determine_ext,
54 determine_protocol,
55 DOT_DESKTOP_LINK_TEMPLATE,
56 DOT_URL_LINK_TEMPLATE,
57 DOT_WEBLOC_LINK_TEMPLATE,
58 DownloadError,
59 encode_compat_str,
60 encodeFilename,
61 EntryNotInPlaylist,
62 error_to_compat_str,
63 ExistingVideoReached,
64 expand_path,
65 ExtractorError,
66 float_or_none,
67 format_bytes,
68 format_field,
69 STR_FORMAT_RE_TMPL,
70 STR_FORMAT_TYPES,
71 formatSeconds,
72 GeoRestrictedError,
73 HEADRequest,
74 int_or_none,
75 iri_to_uri,
76 ISO3166Utils,
77 LazyList,
78 locked_file,
79 make_dir,
80 make_HTTPS_handler,
81 MaxDownloadsReached,
82 network_exceptions,
83 orderedSet,
84 OUTTMPL_TYPES,
85 PagedList,
86 parse_filesize,
87 PerRequestProxyHandler,
88 platform_name,
89 PostProcessingError,
90 preferredencoding,
91 prepend_extension,
92 process_communicate_or_kill,
93 register_socks_protocols,
94 RejectedVideoReached,
95 render_table,
96 replace_extension,
97 SameFileError,
98 sanitize_filename,
99 sanitize_path,
100 sanitize_url,
101 sanitized_Request,
102 std_headers,
103 str_or_none,
104 strftime_or_none,
105 subtitles_filename,
106 ThrottledDownload,
107 to_high_limit_path,
108 traverse_obj,
109 try_get,
110 UnavailableVideoError,
111 url_basename,
112 variadic,
113 version_tuple,
114 write_json_file,
115 write_string,
116 YoutubeDLCookieProcessor,
117 YoutubeDLHandler,
118 YoutubeDLRedirectHandler,
119 )
120 from .cache import Cache
121 from .extractor import (
122 gen_extractor_classes,
123 get_info_extractor,
124 _LAZY_LOADER,
125 _PLUGIN_CLASSES
126 )
127 from .extractor.openload import PhantomJSwrapper
128 from .downloader import (
129 FFmpegFD,
130 get_suitable_downloader,
131 shorten_protocol_name
132 )
133 from .downloader.rtmp import rtmpdump_version
134 from .postprocessor import (
135 get_postprocessor,
136 FFmpegFixupDurationPP,
137 FFmpegFixupM3u8PP,
138 FFmpegFixupM4aPP,
139 FFmpegFixupStretchedPP,
140 FFmpegFixupTimestampPP,
141 FFmpegMergerPP,
142 FFmpegPostProcessor,
143 MoveFilesAfterDownloadPP,
144 )
145 from .version import __version__
146
147 if compat_os_name == 'nt':
148 import ctypes
149
150
151 class YoutubeDL(object):
152 """YoutubeDL class.
153
154 YoutubeDL objects are the ones responsible of downloading the
155 actual video file and writing it to disk if the user has requested
156 it, among some other tasks. In most cases there should be one per
157 program. As, given a video URL, the downloader doesn't know how to
158 extract all the needed information, task that InfoExtractors do, it
159 has to pass the URL to one of them.
160
161 For this, YoutubeDL objects have a method that allows
162 InfoExtractors to be registered in a given order. When it is passed
163 a URL, the YoutubeDL object handles it to the first InfoExtractor it
164 finds that reports being able to handle it. The InfoExtractor extracts
165 all the information about the video or videos the URL refers to, and
166 YoutubeDL process the extracted information, possibly using a File
167 Downloader to download the video.
168
169 YoutubeDL objects accept a lot of parameters. In order not to saturate
170 the object constructor with arguments, it receives a dictionary of
171 options instead. These options are available through the params
172 attribute for the InfoExtractors to use. The YoutubeDL also
173 registers itself as the downloader in charge for the InfoExtractors
174 that are added to it, so this is a "mutual registration".
175
176 Available options:
177
178 username: Username for authentication purposes.
179 password: Password for authentication purposes.
180 videopassword: Password for accessing a video.
181 ap_mso: Adobe Pass multiple-system operator identifier.
182 ap_username: Multiple-system operator account username.
183 ap_password: Multiple-system operator account password.
184 usenetrc: Use netrc for authentication instead.
185 verbose: Print additional info to stdout.
186 quiet: Do not print messages to stdout.
187 no_warnings: Do not print out anything for warnings.
188 forceprint: A list of templates to force print
189 forceurl: Force printing final URL. (Deprecated)
190 forcetitle: Force printing title. (Deprecated)
191 forceid: Force printing ID. (Deprecated)
192 forcethumbnail: Force printing thumbnail URL. (Deprecated)
193 forcedescription: Force printing description. (Deprecated)
194 forcefilename: Force printing final filename. (Deprecated)
195 forceduration: Force printing duration. (Deprecated)
196 forcejson: Force printing info_dict as JSON.
197 dump_single_json: Force printing the info_dict of the whole playlist
198 (or video) as a single JSON line.
199 force_write_download_archive: Force writing download archive regardless
200 of 'skip_download' or 'simulate'.
201 simulate: Do not download the video files. If unset (or None),
202 simulate only if listsubtitles, listformats or list_thumbnails is used
203 format: Video format code. see "FORMAT SELECTION" for more details.
204 allow_unplayable_formats: Allow unplayable formats to be extracted and downloaded.
205 ignore_no_formats_error: Ignore "No video formats" error. Usefull for
206 extracting metadata even if the video is not actually
207 available for download (experimental)
208 format_sort: How to sort the video formats. see "Sorting Formats"
209 for more details.
210 format_sort_force: Force the given format_sort. see "Sorting Formats"
211 for more details.
212 allow_multiple_video_streams: Allow multiple video streams to be merged
213 into a single file
214 allow_multiple_audio_streams: Allow multiple audio streams to be merged
215 into a single file
216 check_formats Whether to test if the formats are downloadable.
217 Can be True (check all), False (check none)
218 or None (check only if requested by extractor)
219 paths: Dictionary of output paths. The allowed keys are 'home'
220 'temp' and the keys of OUTTMPL_TYPES (in utils.py)
221 outtmpl: Dictionary of templates for output names. Allowed keys
222 are 'default' and the keys of OUTTMPL_TYPES (in utils.py).
223 For compatibility with youtube-dl, a single string can also be used
224 outtmpl_na_placeholder: Placeholder for unavailable meta fields.
225 restrictfilenames: Do not allow "&" and spaces in file names
226 trim_file_name: Limit length of filename (extension excluded)
227 windowsfilenames: Force the filenames to be windows compatible
228 ignoreerrors: Do not stop on download errors
229 (Default True when running yt-dlp,
230 but False when directly accessing YoutubeDL class)
231 skip_playlist_after_errors: Number of allowed failures until the rest of
232 the playlist is skipped
233 force_generic_extractor: Force downloader to use the generic extractor
234 overwrites: Overwrite all video and metadata files if True,
235 overwrite only non-video files if None
236 and don't overwrite any file if False
237 For compatibility with youtube-dl,
238 "nooverwrites" may also be used instead
239 playliststart: Playlist item to start at.
240 playlistend: Playlist item to end at.
241 playlist_items: Specific indices of playlist to download.
242 playlistreverse: Download playlist items in reverse order.
243 playlistrandom: Download playlist items in random order.
244 matchtitle: Download only matching titles.
245 rejecttitle: Reject downloads for matching titles.
246 logger: Log messages to a logging.Logger instance.
247 logtostderr: Log messages to stderr instead of stdout.
248 writedescription: Write the video description to a .description file
249 writeinfojson: Write the video description to a .info.json file
250 clean_infojson: Remove private fields from the infojson
251 getcomments: Extract video comments. This will not be written to disk
252 unless writeinfojson is also given
253 writeannotations: Write the video annotations to a .annotations.xml file
254 writethumbnail: Write the thumbnail image to a file
255 allow_playlist_files: Whether to write playlists' description, infojson etc
256 also to disk when using the 'write*' options
257 write_all_thumbnails: Write all thumbnail formats to files
258 writelink: Write an internet shortcut file, depending on the
259 current platform (.url/.webloc/.desktop)
260 writeurllink: Write a Windows internet shortcut file (.url)
261 writewebloclink: Write a macOS internet shortcut file (.webloc)
262 writedesktoplink: Write a Linux internet shortcut file (.desktop)
263 writesubtitles: Write the video subtitles to a file
264 writeautomaticsub: Write the automatically generated subtitles to a file
265 allsubtitles: Deprecated - Use subtitleslangs = ['all']
266 Downloads all the subtitles of the video
267 (requires writesubtitles or writeautomaticsub)
268 listsubtitles: Lists all available subtitles for the video
269 subtitlesformat: The format code for subtitles
270 subtitleslangs: List of languages of the subtitles to download (can be regex).
271 The list may contain "all" to refer to all the available
272 subtitles. The language can be prefixed with a "-" to
273 exclude it from the requested languages. Eg: ['all', '-live_chat']
274 keepvideo: Keep the video file after post-processing
275 daterange: A DateRange object, download only if the upload_date is in the range.
276 skip_download: Skip the actual download of the video file
277 cachedir: Location of the cache files in the filesystem.
278 False to disable filesystem cache.
279 noplaylist: Download single video instead of a playlist if in doubt.
280 age_limit: An integer representing the user's age in years.
281 Unsuitable videos for the given age are skipped.
282 min_views: An integer representing the minimum view count the video
283 must have in order to not be skipped.
284 Videos without view count information are always
285 downloaded. None for no limit.
286 max_views: An integer representing the maximum view count.
287 Videos that are more popular than that are not
288 downloaded.
289 Videos without view count information are always
290 downloaded. None for no limit.
291 download_archive: File name of a file where all downloads are recorded.
292 Videos already present in the file are not downloaded
293 again.
294 break_on_existing: Stop the download process after attempting to download a
295 file that is in the archive.
296 break_on_reject: Stop the download process when encountering a video that
297 has been filtered out.
298 cookiefile: File name where cookies should be read from and dumped to
299 cookiesfrombrowser: A tuple containing the name of the browser and the profile
300 name/path from where cookies are loaded.
301 Eg: ('chrome', ) or (vivaldi, 'default')
302 nocheckcertificate:Do not verify SSL certificates
303 prefer_insecure: Use HTTP instead of HTTPS to retrieve information.
304 At the moment, this is only supported by YouTube.
305 proxy: URL of the proxy server to use
306 geo_verification_proxy: URL of the proxy to use for IP address verification
307 on geo-restricted sites.
308 socket_timeout: Time to wait for unresponsive hosts, in seconds
309 bidi_workaround: Work around buggy terminals without bidirectional text
310 support, using fridibi
311 debug_printtraffic:Print out sent and received HTTP traffic
312 include_ads: Download ads as well
313 default_search: Prepend this string if an input url is not valid.
314 'auto' for elaborate guessing
315 encoding: Use this encoding instead of the system-specified.
316 extract_flat: Do not resolve URLs, return the immediate result.
317 Pass in 'in_playlist' to only show this behavior for
318 playlist items.
319 postprocessors: A list of dictionaries, each with an entry
320 * key: The name of the postprocessor. See
321 yt_dlp/postprocessor/__init__.py for a list.
322 * when: When to run the postprocessor. Can be one of
323 pre_process|before_dl|post_process|after_move.
324 Assumed to be 'post_process' if not given
325 post_hooks: A list of functions that get called as the final step
326 for each video file, after all postprocessors have been
327 called. The filename will be passed as the only argument.
328 progress_hooks: A list of functions that get called on download
329 progress, with a dictionary with the entries
330 * status: One of "downloading", "error", or "finished".
331 Check this first and ignore unknown values.
332 * info_dict: The extracted info_dict
333
334 If status is one of "downloading", or "finished", the
335 following properties may also be present:
336 * filename: The final filename (always present)
337 * tmpfilename: The filename we're currently writing to
338 * downloaded_bytes: Bytes on disk
339 * total_bytes: Size of the whole file, None if unknown
340 * total_bytes_estimate: Guess of the eventual file size,
341 None if unavailable.
342 * elapsed: The number of seconds since download started.
343 * eta: The estimated time in seconds, None if unknown
344 * speed: The download speed in bytes/second, None if
345 unknown
346 * fragment_index: The counter of the currently
347 downloaded video fragment.
348 * fragment_count: The number of fragments (= individual
349 files that will be merged)
350
351 Progress hooks are guaranteed to be called at least once
352 (with status "finished") if the download is successful.
353 merge_output_format: Extension to use when merging formats.
354 final_ext: Expected final extension; used to detect when the file was
355 already downloaded and converted. "merge_output_format" is
356 replaced by this extension when given
357 fixup: Automatically correct known faults of the file.
358 One of:
359 - "never": do nothing
360 - "warn": only emit a warning
361 - "detect_or_warn": check whether we can do anything
362 about it, warn otherwise (default)
363 source_address: Client-side IP address to bind to.
364 call_home: Boolean, true iff we are allowed to contact the
365 yt-dlp servers for debugging. (BROKEN)
366 sleep_interval_requests: Number of seconds to sleep between requests
367 during extraction
368 sleep_interval: Number of seconds to sleep before each download when
369 used alone or a lower bound of a range for randomized
370 sleep before each download (minimum possible number
371 of seconds to sleep) when used along with
372 max_sleep_interval.
373 max_sleep_interval:Upper bound of a range for randomized sleep before each
374 download (maximum possible number of seconds to sleep).
375 Must only be used along with sleep_interval.
376 Actual sleep time will be a random float from range
377 [sleep_interval; max_sleep_interval].
378 sleep_interval_subtitles: Number of seconds to sleep before each subtitle download
379 listformats: Print an overview of available video formats and exit.
380 list_thumbnails: Print a table of all thumbnails and exit.
381 match_filter: A function that gets called with the info_dict of
382 every video.
383 If it returns a message, the video is ignored.
384 If it returns None, the video is downloaded.
385 match_filter_func in utils.py is one example for this.
386 no_color: Do not emit color codes in output.
387 geo_bypass: Bypass geographic restriction via faking X-Forwarded-For
388 HTTP header
389 geo_bypass_country:
390 Two-letter ISO 3166-2 country code that will be used for
391 explicit geographic restriction bypassing via faking
392 X-Forwarded-For HTTP header
393 geo_bypass_ip_block:
394 IP range in CIDR notation that will be used similarly to
395 geo_bypass_country
396
397 The following options determine which downloader is picked:
398 external_downloader: A dictionary of protocol keys and the executable of the
399 external downloader to use for it. The allowed protocols
400 are default|http|ftp|m3u8|dash|rtsp|rtmp|mms.
401 Set the value to 'native' to use the native downloader
402 hls_prefer_native: Deprecated - Use external_downloader = {'m3u8': 'native'}
403 or {'m3u8': 'ffmpeg'} instead.
404 Use the native HLS downloader instead of ffmpeg/avconv
405 if True, otherwise use ffmpeg/avconv if False, otherwise
406 use downloader suggested by extractor if None.
407 compat_opts: Compatibility options. See "Differences in default behavior".
408 The following options do not work when used through the API:
409 filename, abort-on-error, multistreams, no-live-chat,
410 no-clean-infojson, no-playlist-metafiles, no-keep-subs.
411 Refer __init__.py for their implementation
412
413 The following parameters are not used by YoutubeDL itself, they are used by
414 the downloader (see yt_dlp/downloader/common.py):
415 nopart, updatetime, buffersize, ratelimit, throttledratelimit, min_filesize,
416 max_filesize, test, noresizebuffer, retries, continuedl, noprogress, consoletitle,
417 xattr_set_filesize, external_downloader_args, hls_use_mpegts, http_chunk_size.
418
419 The following options are used by the post processors:
420 prefer_ffmpeg: If False, use avconv instead of ffmpeg if both are available,
421 otherwise prefer ffmpeg. (avconv support is deprecated)
422 ffmpeg_location: Location of the ffmpeg/avconv binary; either the path
423 to the binary or its containing directory.
424 postprocessor_args: A dictionary of postprocessor/executable keys (in lower case)
425 and a list of additional command-line arguments for the
426 postprocessor/executable. The dict can also have "PP+EXE" keys
427 which are used when the given exe is used by the given PP.
428 Use 'default' as the name for arguments to passed to all PP
429 For compatibility with youtube-dl, a single list of args
430 can also be used
431
432 The following options are used by the extractors:
433 extractor_retries: Number of times to retry for known errors
434 dynamic_mpd: Whether to process dynamic DASH manifests (default: True)
435 hls_split_discontinuity: Split HLS playlists to different formats at
436 discontinuities such as ad breaks (default: False)
437 extractor_args: A dictionary of arguments to be passed to the extractors.
438 See "EXTRACTOR ARGUMENTS" for details.
439 Eg: {'youtube': {'skip': ['dash', 'hls']}}
440 youtube_include_dash_manifest: Deprecated - Use extractor_args instead.
441 If True (default), DASH manifests and related
442 data will be downloaded and processed by extractor.
443 You can reduce network I/O by disabling it if you don't
444 care about DASH. (only for youtube)
445 youtube_include_hls_manifest: Deprecated - Use extractor_args instead.
446 If True (default), HLS manifests and related
447 data will be downloaded and processed by extractor.
448 You can reduce network I/O by disabling it if you don't
449 care about HLS. (only for youtube)
450 """
451
452 _NUMERIC_FIELDS = set((
453 'width', 'height', 'tbr', 'abr', 'asr', 'vbr', 'fps', 'filesize', 'filesize_approx',
454 'timestamp', 'upload_year', 'upload_month', 'upload_day',
455 'duration', 'view_count', 'like_count', 'dislike_count', 'repost_count',
456 'average_rating', 'comment_count', 'age_limit',
457 'start_time', 'end_time',
458 'chapter_number', 'season_number', 'episode_number',
459 'track_number', 'disc_number', 'release_year',
460 'playlist_index',
461 ))
462
463 params = None
464 _ies = []
465 _pps = {'pre_process': [], 'before_dl': [], 'after_move': [], 'post_process': []}
466 _printed_messages = set()
467 _first_webpage_request = True
468 _download_retcode = None
469 _num_downloads = None
470 _playlist_level = 0
471 _playlist_urls = set()
472 _screen_file = None
473
474 def __init__(self, params=None, auto_init=True):
475 """Create a FileDownloader object with the given options."""
476 if params is None:
477 params = {}
478 self._ies = []
479 self._ies_instances = {}
480 self._pps = {'pre_process': [], 'before_dl': [], 'after_move': [], 'post_process': []}
481 self._printed_messages = set()
482 self._first_webpage_request = True
483 self._post_hooks = []
484 self._progress_hooks = []
485 self._download_retcode = 0
486 self._num_downloads = 0
487 self._screen_file = [sys.stdout, sys.stderr][params.get('logtostderr', False)]
488 self._err_file = sys.stderr
489 self.params = {
490 # Default parameters
491 'nocheckcertificate': False,
492 }
493 self.params.update(params)
494 self.cache = Cache(self)
495
496 if sys.version_info < (3, 6):
497 self.report_warning(
498 'Python version %d.%d is not supported! Please update to Python 3.6 or above' % sys.version_info[:2])
499
500 def check_deprecated(param, option, suggestion):
501 if self.params.get(param) is not None:
502 self.report_warning('%s is deprecated. Use %s instead' % (option, suggestion))
503 return True
504 return False
505
506 if check_deprecated('cn_verification_proxy', '--cn-verification-proxy', '--geo-verification-proxy'):
507 if self.params.get('geo_verification_proxy') is None:
508 self.params['geo_verification_proxy'] = self.params['cn_verification_proxy']
509
510 check_deprecated('autonumber', '--auto-number', '-o "%(autonumber)s-%(title)s.%(ext)s"')
511 check_deprecated('usetitle', '--title', '-o "%(title)s-%(id)s.%(ext)s"')
512 check_deprecated('useid', '--id', '-o "%(id)s.%(ext)s"')
513
514 for msg in self.params.get('warnings', []):
515 self.report_warning(msg)
516
517 if self.params.get('final_ext'):
518 if self.params.get('merge_output_format'):
519 self.report_warning('--merge-output-format will be ignored since --remux-video or --recode-video is given')
520 self.params['merge_output_format'] = self.params['final_ext']
521
522 if self.params.get('overwrites') is None:
523 self.params.pop('overwrites', None)
524 elif self.params.get('nooverwrites') is not None:
525 # nooverwrites was unnecessarily changed to overwrites
526 # in 0c3d0f51778b153f65c21906031c2e091fcfb641
527 # This ensures compatibility with both keys
528 self.params['overwrites'] = not self.params['nooverwrites']
529 else:
530 self.params['nooverwrites'] = not self.params['overwrites']
531
532 if params.get('bidi_workaround', False):
533 try:
534 import pty
535 master, slave = pty.openpty()
536 width = compat_get_terminal_size().columns
537 if width is None:
538 width_args = []
539 else:
540 width_args = ['-w', str(width)]
541 sp_kwargs = dict(
542 stdin=subprocess.PIPE,
543 stdout=slave,
544 stderr=self._err_file)
545 try:
546 self._output_process = subprocess.Popen(
547 ['bidiv'] + width_args, **sp_kwargs
548 )
549 except OSError:
550 self._output_process = subprocess.Popen(
551 ['fribidi', '-c', 'UTF-8'] + width_args, **sp_kwargs)
552 self._output_channel = os.fdopen(master, 'rb')
553 except OSError as ose:
554 if ose.errno == errno.ENOENT:
555 self.report_warning('Could not find fribidi executable, ignoring --bidi-workaround . Make sure that fribidi is an executable file in one of the directories in your $PATH.')
556 else:
557 raise
558
559 if (sys.platform != 'win32'
560 and sys.getfilesystemencoding() in ['ascii', 'ANSI_X3.4-1968']
561 and not params.get('restrictfilenames', False)):
562 # Unicode filesystem API will throw errors (#1474, #13027)
563 self.report_warning(
564 'Assuming --restrict-filenames since file system encoding '
565 'cannot encode all characters. '
566 'Set the LC_ALL environment variable to fix this.')
567 self.params['restrictfilenames'] = True
568
569 self.outtmpl_dict = self.parse_outtmpl()
570
571 # Creating format selector here allows us to catch syntax errors before the extraction
572 self.format_selector = (
573 None if self.params.get('format') is None
574 else self.build_format_selector(self.params['format']))
575
576 self._setup_opener()
577
578 """Preload the archive, if any is specified"""
579 def preload_download_archive(fn):
580 if fn is None:
581 return False
582 self.write_debug('Loading archive file %r\n' % fn)
583 try:
584 with locked_file(fn, 'r', encoding='utf-8') as archive_file:
585 for line in archive_file:
586 self.archive.add(line.strip())
587 except IOError as ioe:
588 if ioe.errno != errno.ENOENT:
589 raise
590 return False
591 return True
592
593 self.archive = set()
594 preload_download_archive(self.params.get('download_archive'))
595
596 if auto_init:
597 self.print_debug_header()
598 self.add_default_info_extractors()
599
600 for pp_def_raw in self.params.get('postprocessors', []):
601 pp_def = dict(pp_def_raw)
602 when = pp_def.pop('when', 'post_process')
603 pp_class = get_postprocessor(pp_def.pop('key'))
604 pp = pp_class(self, **compat_kwargs(pp_def))
605 self.add_post_processor(pp, when=when)
606
607 for ph in self.params.get('post_hooks', []):
608 self.add_post_hook(ph)
609
610 for ph in self.params.get('progress_hooks', []):
611 self.add_progress_hook(ph)
612
613 register_socks_protocols()
614
615 def warn_if_short_id(self, argv):
616 # short YouTube ID starting with dash?
617 idxs = [
618 i for i, a in enumerate(argv)
619 if re.match(r'^-[0-9A-Za-z_-]{10}$', a)]
620 if idxs:
621 correct_argv = (
622 ['yt-dlp']
623 + [a for i, a in enumerate(argv) if i not in idxs]
624 + ['--'] + [argv[i] for i in idxs]
625 )
626 self.report_warning(
627 'Long argument string detected. '
628 'Use -- to separate parameters and URLs, like this:\n%s\n' %
629 args_to_str(correct_argv))
630
631 def add_info_extractor(self, ie):
632 """Add an InfoExtractor object to the end of the list."""
633 self._ies.append(ie)
634 if not isinstance(ie, type):
635 self._ies_instances[ie.ie_key()] = ie
636 ie.set_downloader(self)
637
638 def get_info_extractor(self, ie_key):
639 """
640 Get an instance of an IE with name ie_key, it will try to get one from
641 the _ies list, if there's no instance it will create a new one and add
642 it to the extractor list.
643 """
644 ie = self._ies_instances.get(ie_key)
645 if ie is None:
646 ie = get_info_extractor(ie_key)()
647 self.add_info_extractor(ie)
648 return ie
649
650 def add_default_info_extractors(self):
651 """
652 Add the InfoExtractors returned by gen_extractors to the end of the list
653 """
654 for ie in gen_extractor_classes():
655 self.add_info_extractor(ie)
656
657 def add_post_processor(self, pp, when='post_process'):
658 """Add a PostProcessor object to the end of the chain."""
659 self._pps[when].append(pp)
660 pp.set_downloader(self)
661
662 def add_post_hook(self, ph):
663 """Add the post hook"""
664 self._post_hooks.append(ph)
665
666 def add_progress_hook(self, ph):
667 """Add the progress hook (currently only for the file downloader)"""
668 self._progress_hooks.append(ph)
669
670 def _bidi_workaround(self, message):
671 if not hasattr(self, '_output_channel'):
672 return message
673
674 assert hasattr(self, '_output_process')
675 assert isinstance(message, compat_str)
676 line_count = message.count('\n') + 1
677 self._output_process.stdin.write((message + '\n').encode('utf-8'))
678 self._output_process.stdin.flush()
679 res = ''.join(self._output_channel.readline().decode('utf-8')
680 for _ in range(line_count))
681 return res[:-len('\n')]
682
683 def _write_string(self, message, out=None, only_once=False):
684 if only_once:
685 if message in self._printed_messages:
686 return
687 self._printed_messages.add(message)
688 write_string(message, out=out, encoding=self.params.get('encoding'))
689
690 def to_stdout(self, message, skip_eol=False, quiet=False):
691 """Print message to stdout"""
692 if self.params.get('logger'):
693 self.params['logger'].debug(message)
694 elif not quiet or self.params.get('verbose'):
695 self._write_string(
696 '%s%s' % (self._bidi_workaround(message), ('' if skip_eol else '\n')),
697 self._err_file if quiet else self._screen_file)
698
699 def to_stderr(self, message, only_once=False):
700 """Print message to stderr"""
701 assert isinstance(message, compat_str)
702 if self.params.get('logger'):
703 self.params['logger'].error(message)
704 else:
705 self._write_string('%s\n' % self._bidi_workaround(message), self._err_file, only_once=only_once)
706
707 def to_console_title(self, message):
708 if not self.params.get('consoletitle', False):
709 return
710 if compat_os_name == 'nt':
711 if ctypes.windll.kernel32.GetConsoleWindow():
712 # c_wchar_p() might not be necessary if `message` is
713 # already of type unicode()
714 ctypes.windll.kernel32.SetConsoleTitleW(ctypes.c_wchar_p(message))
715 elif 'TERM' in os.environ:
716 self._write_string('\033]0;%s\007' % message, self._screen_file)
717
718 def save_console_title(self):
719 if not self.params.get('consoletitle', False):
720 return
721 if self.params.get('simulate'):
722 return
723 if compat_os_name != 'nt' and 'TERM' in os.environ:
724 # Save the title on stack
725 self._write_string('\033[22;0t', self._screen_file)
726
727 def restore_console_title(self):
728 if not self.params.get('consoletitle', False):
729 return
730 if self.params.get('simulate'):
731 return
732 if compat_os_name != 'nt' and 'TERM' in os.environ:
733 # Restore the title from stack
734 self._write_string('\033[23;0t', self._screen_file)
735
736 def __enter__(self):
737 self.save_console_title()
738 return self
739
740 def __exit__(self, *args):
741 self.restore_console_title()
742
743 if self.params.get('cookiefile') is not None:
744 self.cookiejar.save(ignore_discard=True, ignore_expires=True)
745
746 def trouble(self, message=None, tb=None):
747 """Determine action to take when a download problem appears.
748
749 Depending on if the downloader has been configured to ignore
750 download errors or not, this method may throw an exception or
751 not when errors are found, after printing the message.
752
753 tb, if given, is additional traceback information.
754 """
755 if message is not None:
756 self.to_stderr(message)
757 if self.params.get('verbose'):
758 if tb is None:
759 if sys.exc_info()[0]: # if .trouble has been called from an except block
760 tb = ''
761 if hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
762 tb += ''.join(traceback.format_exception(*sys.exc_info()[1].exc_info))
763 tb += encode_compat_str(traceback.format_exc())
764 else:
765 tb_data = traceback.format_list(traceback.extract_stack())
766 tb = ''.join(tb_data)
767 if tb:
768 self.to_stderr(tb)
769 if not self.params.get('ignoreerrors', False):
770 if sys.exc_info()[0] and hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
771 exc_info = sys.exc_info()[1].exc_info
772 else:
773 exc_info = sys.exc_info()
774 raise DownloadError(message, exc_info)
775 self._download_retcode = 1
776
777 def to_screen(self, message, skip_eol=False):
778 """Print message to stdout if not in quiet mode"""
779 self.to_stdout(
780 message, skip_eol, quiet=self.params.get('quiet', False))
781
782 def report_warning(self, message, only_once=False):
783 '''
784 Print the message to stderr, it will be prefixed with 'WARNING:'
785 If stderr is a tty file the 'WARNING:' will be colored
786 '''
787 if self.params.get('logger') is not None:
788 self.params['logger'].warning(message)
789 else:
790 if self.params.get('no_warnings'):
791 return
792 if not self.params.get('no_color') and self._err_file.isatty() and compat_os_name != 'nt':
793 _msg_header = '\033[0;33mWARNING:\033[0m'
794 else:
795 _msg_header = 'WARNING:'
796 warning_message = '%s %s' % (_msg_header, message)
797 self.to_stderr(warning_message, only_once)
798
799 def report_error(self, message, tb=None):
800 '''
801 Do the same as trouble, but prefixes the message with 'ERROR:', colored
802 in red if stderr is a tty file.
803 '''
804 if not self.params.get('no_color') and self._err_file.isatty() and compat_os_name != 'nt':
805 _msg_header = '\033[0;31mERROR:\033[0m'
806 else:
807 _msg_header = 'ERROR:'
808 error_message = '%s %s' % (_msg_header, message)
809 self.trouble(error_message, tb)
810
811 def write_debug(self, message, only_once=False):
812 '''Log debug message or Print message to stderr'''
813 if not self.params.get('verbose', False):
814 return
815 message = '[debug] %s' % message
816 if self.params.get('logger'):
817 self.params['logger'].debug(message)
818 else:
819 self.to_stderr(message, only_once)
820
821 def report_file_already_downloaded(self, file_name):
822 """Report file has already been fully downloaded."""
823 try:
824 self.to_screen('[download] %s has already been downloaded' % file_name)
825 except UnicodeEncodeError:
826 self.to_screen('[download] The file has already been downloaded')
827
828 def report_file_delete(self, file_name):
829 """Report that existing file will be deleted."""
830 try:
831 self.to_screen('Deleting existing file %s' % file_name)
832 except UnicodeEncodeError:
833 self.to_screen('Deleting existing file')
834
835 def parse_outtmpl(self):
836 outtmpl_dict = self.params.get('outtmpl', {})
837 if not isinstance(outtmpl_dict, dict):
838 outtmpl_dict = {'default': outtmpl_dict}
839 outtmpl_dict.update({
840 k: v for k, v in DEFAULT_OUTTMPL.items()
841 if not outtmpl_dict.get(k)})
842 for key, val in outtmpl_dict.items():
843 if isinstance(val, bytes):
844 self.report_warning(
845 'Parameter outtmpl is bytes, but should be a unicode string. '
846 'Put from __future__ import unicode_literals at the top of your code file or consider switching to Python 3.x.')
847 return outtmpl_dict
848
849 def get_output_path(self, dir_type='', filename=None):
850 paths = self.params.get('paths', {})
851 assert isinstance(paths, dict)
852 path = os.path.join(
853 expand_path(paths.get('home', '').strip()),
854 expand_path(paths.get(dir_type, '').strip()) if dir_type else '',
855 filename or '')
856
857 # Temporary fix for #4787
858 # 'Treat' all problem characters by passing filename through preferredencoding
859 # to workaround encoding issues with subprocess on python2 @ Windows
860 if sys.version_info < (3, 0) and sys.platform == 'win32':
861 path = encodeFilename(path, True).decode(preferredencoding())
862 return sanitize_path(path, force=self.params.get('windowsfilenames'))
863
864 @staticmethod
865 def _outtmpl_expandpath(outtmpl):
866 # expand_path translates '%%' into '%' and '$$' into '$'
867 # correspondingly that is not what we want since we need to keep
868 # '%%' intact for template dict substitution step. Working around
869 # with boundary-alike separator hack.
870 sep = ''.join([random.choice(ascii_letters) for _ in range(32)])
871 outtmpl = outtmpl.replace('%%', '%{0}%'.format(sep)).replace('$$', '${0}$'.format(sep))
872
873 # outtmpl should be expand_path'ed before template dict substitution
874 # because meta fields may contain env variables we don't want to
875 # be expanded. For example, for outtmpl "%(title)s.%(ext)s" and
876 # title "Hello $PATH", we don't want `$PATH` to be expanded.
877 return expand_path(outtmpl).replace(sep, '')
878
879 @staticmethod
880 def escape_outtmpl(outtmpl):
881 ''' Escape any remaining strings like %s, %abc% etc. '''
882 return re.sub(
883 STR_FORMAT_RE_TMPL.format('', '(?![%(\0])'),
884 lambda mobj: ('' if mobj.group('has_key') else '%') + mobj.group(0),
885 outtmpl)
886
887 @classmethod
888 def validate_outtmpl(cls, outtmpl):
889 ''' @return None or Exception object '''
890 outtmpl = re.sub(
891 STR_FORMAT_RE_TMPL.format('[^)]*', '[ljq]'),
892 lambda mobj: f'{mobj.group(0)[:-1]}s',
893 cls._outtmpl_expandpath(outtmpl))
894 try:
895 cls.escape_outtmpl(outtmpl) % collections.defaultdict(int)
896 return None
897 except ValueError as err:
898 return err
899
900 def prepare_outtmpl(self, outtmpl, info_dict, sanitize=None):
901 """ Make the template and info_dict suitable for substitution : ydl.outtmpl_escape(outtmpl) % info_dict """
902 info_dict.setdefault('epoch', int(time.time())) # keep epoch consistent once set
903
904 info_dict = dict(info_dict) # Do not sanitize so as not to consume LazyList
905 for key in ('__original_infodict', '__postprocessors'):
906 info_dict.pop(key, None)
907 info_dict['duration_string'] = ( # %(duration>%H-%M-%S)s is wrong if duration > 24hrs
908 formatSeconds(info_dict['duration'], '-' if sanitize else ':')
909 if info_dict.get('duration', None) is not None
910 else None)
911 info_dict['autonumber'] = self.params.get('autonumber_start', 1) - 1 + self._num_downloads
912 if info_dict.get('resolution') is None:
913 info_dict['resolution'] = self.format_resolution(info_dict, default=None)
914
915 # For fields playlist_index and autonumber convert all occurrences
916 # of %(field)s to %(field)0Nd for backward compatibility
917 field_size_compat_map = {
918 'playlist_index': len(str(info_dict.get('_last_playlist_index') or '')),
919 'autonumber': self.params.get('autonumber_size') or 5,
920 }
921
922 TMPL_DICT = {}
923 EXTERNAL_FORMAT_RE = re.compile(STR_FORMAT_RE_TMPL.format('[^)]*', f'[{STR_FORMAT_TYPES}ljq]'))
924 MATH_FUNCTIONS = {
925 '+': float.__add__,
926 '-': float.__sub__,
927 }
928 # Field is of the form key1.key2...
929 # where keys (except first) can be string, int or slice
930 FIELD_RE = r'\w*(?:\.(?:\w+|{num}|{num}?(?::{num}?){{1,2}}))*'.format(num=r'(?:-?\d+)')
931 MATH_FIELD_RE = r'''{field}|{num}'''.format(field=FIELD_RE, num=r'-?\d+(?:.\d+)?')
932 MATH_OPERATORS_RE = r'(?:%s)' % '|'.join(map(re.escape, MATH_FUNCTIONS.keys()))
933 INTERNAL_FORMAT_RE = re.compile(r'''(?x)
934 (?P<negate>-)?
935 (?P<fields>{field})
936 (?P<maths>(?:{math_op}{math_field})*)
937 (?:>(?P<strf_format>.+?))?
938 (?:\|(?P<default>.*?))?
939 $'''.format(field=FIELD_RE, math_op=MATH_OPERATORS_RE, math_field=MATH_FIELD_RE))
940
941 def _traverse_infodict(k):
942 k = k.split('.')
943 if k[0] == '':
944 k.pop(0)
945 return traverse_obj(info_dict, k, is_user_input=True, traverse_string=True)
946
947 def get_value(mdict):
948 # Object traversal
949 value = _traverse_infodict(mdict['fields'])
950 # Negative
951 if mdict['negate']:
952 value = float_or_none(value)
953 if value is not None:
954 value *= -1
955 # Do maths
956 offset_key = mdict['maths']
957 if offset_key:
958 value = float_or_none(value)
959 operator = None
960 while offset_key:
961 item = re.match(
962 MATH_FIELD_RE if operator else MATH_OPERATORS_RE,
963 offset_key).group(0)
964 offset_key = offset_key[len(item):]
965 if operator is None:
966 operator = MATH_FUNCTIONS[item]
967 continue
968 item, multiplier = (item[1:], -1) if item[0] == '-' else (item, 1)
969 offset = float_or_none(item)
970 if offset is None:
971 offset = float_or_none(_traverse_infodict(item))
972 try:
973 value = operator(value, multiplier * offset)
974 except (TypeError, ZeroDivisionError):
975 return None
976 operator = None
977 # Datetime formatting
978 if mdict['strf_format']:
979 value = strftime_or_none(value, mdict['strf_format'])
980
981 return value
982
983 na = self.params.get('outtmpl_na_placeholder', 'NA')
984
985 def _dumpjson_default(obj):
986 if isinstance(obj, (set, LazyList)):
987 return list(obj)
988 raise TypeError(f'Object of type {type(obj).__name__} is not JSON serializable')
989
990 def create_key(outer_mobj):
991 if not outer_mobj.group('has_key'):
992 return f'%{outer_mobj.group(0)}'
993 key = outer_mobj.group('key')
994 mobj = re.match(INTERNAL_FORMAT_RE, key)
995 if mobj is None:
996 value, default, mobj = None, na, {'fields': ''}
997 else:
998 mobj = mobj.groupdict()
999 default = mobj['default'] if mobj['default'] is not None else na
1000 value = get_value(mobj)
1001
1002 fmt = outer_mobj.group('format')
1003 if fmt == 's' and value is not None and key in field_size_compat_map.keys():
1004 fmt = '0{:d}d'.format(field_size_compat_map[key])
1005
1006 value = default if value is None else value
1007
1008 str_fmt = f'{fmt[:-1]}s'
1009 if fmt[-1] == 'l':
1010 value, fmt = ', '.join(variadic(value)), str_fmt
1011 elif fmt[-1] == 'j':
1012 value, fmt = json.dumps(value, default=_dumpjson_default), str_fmt
1013 elif fmt[-1] == 'q':
1014 value, fmt = compat_shlex_quote(str(value)), str_fmt
1015 elif fmt[-1] == 'c':
1016 value = str(value)
1017 if value is None:
1018 value, fmt = default, 's'
1019 else:
1020 value = value[0]
1021 elif fmt[-1] not in 'rs': # numeric
1022 value = float_or_none(value)
1023 if value is None:
1024 value, fmt = default, 's'
1025
1026 if sanitize:
1027 if fmt[-1] == 'r':
1028 # If value is an object, sanitize might convert it to a string
1029 # So we convert it to repr first
1030 value, fmt = repr(value), str_fmt
1031 if fmt[-1] in 'csr':
1032 value = sanitize(mobj['fields'].split('.')[-1], value)
1033
1034 key = '%s\0%s' % (key.replace('%', '%\0'), outer_mobj.group('format'))
1035 TMPL_DICT[key] = value
1036 return '{prefix}%({key}){fmt}'.format(key=key, fmt=fmt, prefix=outer_mobj.group('prefix'))
1037
1038 return EXTERNAL_FORMAT_RE.sub(create_key, outtmpl), TMPL_DICT
1039
1040 def _prepare_filename(self, info_dict, tmpl_type='default'):
1041 try:
1042 sanitize = lambda k, v: sanitize_filename(
1043 compat_str(v),
1044 restricted=self.params.get('restrictfilenames'),
1045 is_id=(k == 'id' or k.endswith('_id')))
1046 outtmpl = self.outtmpl_dict.get(tmpl_type, self.outtmpl_dict['default'])
1047 outtmpl, template_dict = self.prepare_outtmpl(outtmpl, info_dict, sanitize)
1048 outtmpl = self.escape_outtmpl(self._outtmpl_expandpath(outtmpl))
1049 filename = outtmpl % template_dict
1050
1051 force_ext = OUTTMPL_TYPES.get(tmpl_type)
1052 if force_ext is not None:
1053 filename = replace_extension(filename, force_ext, info_dict.get('ext'))
1054
1055 # https://github.com/blackjack4494/youtube-dlc/issues/85
1056 trim_file_name = self.params.get('trim_file_name', False)
1057 if trim_file_name:
1058 fn_groups = filename.rsplit('.')
1059 ext = fn_groups[-1]
1060 sub_ext = ''
1061 if len(fn_groups) > 2:
1062 sub_ext = fn_groups[-2]
1063 filename = '.'.join(filter(None, [fn_groups[0][:trim_file_name], sub_ext, ext]))
1064
1065 return filename
1066 except ValueError as err:
1067 self.report_error('Error in output template: ' + str(err) + ' (encoding: ' + repr(preferredencoding()) + ')')
1068 return None
1069
1070 def prepare_filename(self, info_dict, dir_type='', warn=False):
1071 """Generate the output filename."""
1072
1073 filename = self._prepare_filename(info_dict, dir_type or 'default')
1074
1075 if warn:
1076 if not self.params.get('paths'):
1077 pass
1078 elif filename == '-':
1079 self.report_warning('--paths is ignored when an outputting to stdout', only_once=True)
1080 elif os.path.isabs(filename):
1081 self.report_warning('--paths is ignored since an absolute path is given in output template', only_once=True)
1082 if filename == '-' or not filename:
1083 return filename
1084
1085 return self.get_output_path(dir_type, filename)
1086
1087 def _match_entry(self, info_dict, incomplete=False, silent=False):
1088 """ Returns None if the file should be downloaded """
1089
1090 video_title = info_dict.get('title', info_dict.get('id', 'video'))
1091
1092 def check_filter():
1093 if 'title' in info_dict:
1094 # This can happen when we're just evaluating the playlist
1095 title = info_dict['title']
1096 matchtitle = self.params.get('matchtitle', False)
1097 if matchtitle:
1098 if not re.search(matchtitle, title, re.IGNORECASE):
1099 return '"' + title + '" title did not match pattern "' + matchtitle + '"'
1100 rejecttitle = self.params.get('rejecttitle', False)
1101 if rejecttitle:
1102 if re.search(rejecttitle, title, re.IGNORECASE):
1103 return '"' + title + '" title matched reject pattern "' + rejecttitle + '"'
1104 date = info_dict.get('upload_date')
1105 if date is not None:
1106 dateRange = self.params.get('daterange', DateRange())
1107 if date not in dateRange:
1108 return '%s upload date is not in range %s' % (date_from_str(date).isoformat(), dateRange)
1109 view_count = info_dict.get('view_count')
1110 if view_count is not None:
1111 min_views = self.params.get('min_views')
1112 if min_views is not None and view_count < min_views:
1113 return 'Skipping %s, because it has not reached minimum view count (%d/%d)' % (video_title, view_count, min_views)
1114 max_views = self.params.get('max_views')
1115 if max_views is not None and view_count > max_views:
1116 return 'Skipping %s, because it has exceeded the maximum view count (%d/%d)' % (video_title, view_count, max_views)
1117 if age_restricted(info_dict.get('age_limit'), self.params.get('age_limit')):
1118 return 'Skipping "%s" because it is age restricted' % video_title
1119
1120 if not incomplete:
1121 match_filter = self.params.get('match_filter')
1122 if match_filter is not None:
1123 ret = match_filter(info_dict)
1124 if ret is not None:
1125 return ret
1126 return None
1127
1128 if self.in_download_archive(info_dict):
1129 reason = '%s has already been recorded in the archive' % video_title
1130 break_opt, break_err = 'break_on_existing', ExistingVideoReached
1131 else:
1132 reason = check_filter()
1133 break_opt, break_err = 'break_on_reject', RejectedVideoReached
1134 if reason is not None:
1135 if not silent:
1136 self.to_screen('[download] ' + reason)
1137 if self.params.get(break_opt, False):
1138 raise break_err()
1139 return reason
1140
1141 @staticmethod
1142 def add_extra_info(info_dict, extra_info):
1143 '''Set the keys from extra_info in info dict if they are missing'''
1144 for key, value in extra_info.items():
1145 info_dict.setdefault(key, value)
1146
1147 def extract_info(self, url, download=True, ie_key=None, extra_info={},
1148 process=True, force_generic_extractor=False):
1149 """
1150 Return a list with a dictionary for each video extracted.
1151
1152 Arguments:
1153 url -- URL to extract
1154
1155 Keyword arguments:
1156 download -- whether to download videos during extraction
1157 ie_key -- extractor key hint
1158 extra_info -- dictionary containing the extra values to add to each result
1159 process -- whether to resolve all unresolved references (URLs, playlist items),
1160 must be True for download to work.
1161 force_generic_extractor -- force using the generic extractor
1162 """
1163
1164 if not ie_key and force_generic_extractor:
1165 ie_key = 'Generic'
1166
1167 if ie_key:
1168 ies = [self.get_info_extractor(ie_key)]
1169 else:
1170 ies = self._ies
1171
1172 for ie in ies:
1173 if not ie.suitable(url):
1174 continue
1175
1176 ie_key = ie.ie_key()
1177 ie = self.get_info_extractor(ie_key)
1178 if not ie.working():
1179 self.report_warning('The program functionality for this site has been marked as broken, '
1180 'and will probably not work.')
1181
1182 try:
1183 temp_id = str_or_none(
1184 ie.extract_id(url) if callable(getattr(ie, 'extract_id', None))
1185 else ie._match_id(url))
1186 except (AssertionError, IndexError, AttributeError):
1187 temp_id = None
1188 if temp_id is not None and self.in_download_archive({'id': temp_id, 'ie_key': ie_key}):
1189 self.to_screen("[%s] %s: has already been recorded in archive" % (
1190 ie_key, temp_id))
1191 break
1192 return self.__extract_info(url, ie, download, extra_info, process)
1193 else:
1194 self.report_error('no suitable InfoExtractor for URL %s' % url)
1195
1196 def __handle_extraction_exceptions(func, handle_all_errors=True):
1197 def wrapper(self, *args, **kwargs):
1198 try:
1199 return func(self, *args, **kwargs)
1200 except GeoRestrictedError as e:
1201 msg = e.msg
1202 if e.countries:
1203 msg += '\nThis video is available in %s.' % ', '.join(
1204 map(ISO3166Utils.short2full, e.countries))
1205 msg += '\nYou might want to use a VPN or a proxy server (with --proxy) to workaround.'
1206 self.report_error(msg)
1207 except ExtractorError as e: # An error we somewhat expected
1208 self.report_error(compat_str(e), e.format_traceback())
1209 except ThrottledDownload:
1210 self.to_stderr('\r')
1211 self.report_warning('The download speed is below throttle limit. Re-extracting data')
1212 return wrapper(self, *args, **kwargs)
1213 except (MaxDownloadsReached, ExistingVideoReached, RejectedVideoReached):
1214 raise
1215 except Exception as e:
1216 if handle_all_errors and self.params.get('ignoreerrors', False):
1217 self.report_error(error_to_compat_str(e), tb=encode_compat_str(traceback.format_exc()))
1218 else:
1219 raise
1220 return wrapper
1221
1222 @__handle_extraction_exceptions
1223 def __extract_info(self, url, ie, download, extra_info, process):
1224 ie_result = ie.extract(url)
1225 if ie_result is None: # Finished already (backwards compatibility; listformats and friends should be moved here)
1226 return
1227 if isinstance(ie_result, list):
1228 # Backwards compatibility: old IE result format
1229 ie_result = {
1230 '_type': 'compat_list',
1231 'entries': ie_result,
1232 }
1233 if extra_info.get('original_url'):
1234 ie_result.setdefault('original_url', extra_info['original_url'])
1235 self.add_default_extra_info(ie_result, ie, url)
1236 if process:
1237 return self.process_ie_result(ie_result, download, extra_info)
1238 else:
1239 return ie_result
1240
1241 def add_default_extra_info(self, ie_result, ie, url):
1242 if url is not None:
1243 self.add_extra_info(ie_result, {
1244 'webpage_url': url,
1245 'original_url': url,
1246 'webpage_url_basename': url_basename(url),
1247 })
1248 if ie is not None:
1249 self.add_extra_info(ie_result, {
1250 'extractor': ie.IE_NAME,
1251 'extractor_key': ie.ie_key(),
1252 })
1253
1254 def process_ie_result(self, ie_result, download=True, extra_info={}):
1255 """
1256 Take the result of the ie(may be modified) and resolve all unresolved
1257 references (URLs, playlist items).
1258
1259 It will also download the videos if 'download'.
1260 Returns the resolved ie_result.
1261 """
1262 result_type = ie_result.get('_type', 'video')
1263
1264 if result_type in ('url', 'url_transparent'):
1265 ie_result['url'] = sanitize_url(ie_result['url'])
1266 if ie_result.get('original_url'):
1267 extra_info.setdefault('original_url', ie_result['original_url'])
1268
1269 extract_flat = self.params.get('extract_flat', False)
1270 if ((extract_flat == 'in_playlist' and 'playlist' in extra_info)
1271 or extract_flat is True):
1272 info_copy = ie_result.copy()
1273 self.add_extra_info(info_copy, extra_info)
1274 ie = try_get(ie_result.get('ie_key'), self.get_info_extractor)
1275 self.add_default_extra_info(info_copy, ie, ie_result['url'])
1276 self.__forced_printings(info_copy, self.prepare_filename(info_copy), incomplete=True)
1277 return ie_result
1278
1279 if result_type == 'video':
1280 self.add_extra_info(ie_result, extra_info)
1281 ie_result = self.process_video_result(ie_result, download=download)
1282 additional_urls = (ie_result or {}).get('additional_urls')
1283 if additional_urls:
1284 # TODO: Improve MetadataFromFieldPP to allow setting a list
1285 if isinstance(additional_urls, compat_str):
1286 additional_urls = [additional_urls]
1287 self.to_screen(
1288 '[info] %s: %d additional URL(s) requested' % (ie_result['id'], len(additional_urls)))
1289 self.write_debug('Additional URLs: "%s"' % '", "'.join(additional_urls))
1290 ie_result['additional_entries'] = [
1291 self.extract_info(
1292 url, download, extra_info,
1293 force_generic_extractor=self.params.get('force_generic_extractor'))
1294 for url in additional_urls
1295 ]
1296 return ie_result
1297 elif result_type == 'url':
1298 # We have to add extra_info to the results because it may be
1299 # contained in a playlist
1300 return self.extract_info(
1301 ie_result['url'], download,
1302 ie_key=ie_result.get('ie_key'),
1303 extra_info=extra_info)
1304 elif result_type == 'url_transparent':
1305 # Use the information from the embedding page
1306 info = self.extract_info(
1307 ie_result['url'], ie_key=ie_result.get('ie_key'),
1308 extra_info=extra_info, download=False, process=False)
1309
1310 # extract_info may return None when ignoreerrors is enabled and
1311 # extraction failed with an error, don't crash and return early
1312 # in this case
1313 if not info:
1314 return info
1315
1316 force_properties = dict(
1317 (k, v) for k, v in ie_result.items() if v is not None)
1318 for f in ('_type', 'url', 'id', 'extractor', 'extractor_key', 'ie_key'):
1319 if f in force_properties:
1320 del force_properties[f]
1321 new_result = info.copy()
1322 new_result.update(force_properties)
1323
1324 # Extracted info may not be a video result (i.e.
1325 # info.get('_type', 'video') != video) but rather an url or
1326 # url_transparent. In such cases outer metadata (from ie_result)
1327 # should be propagated to inner one (info). For this to happen
1328 # _type of info should be overridden with url_transparent. This
1329 # fixes issue from https://github.com/ytdl-org/youtube-dl/pull/11163.
1330 if new_result.get('_type') == 'url':
1331 new_result['_type'] = 'url_transparent'
1332
1333 return self.process_ie_result(
1334 new_result, download=download, extra_info=extra_info)
1335 elif result_type in ('playlist', 'multi_video'):
1336 # Protect from infinite recursion due to recursively nested playlists
1337 # (see https://github.com/ytdl-org/youtube-dl/issues/27833)
1338 webpage_url = ie_result['webpage_url']
1339 if webpage_url in self._playlist_urls:
1340 self.to_screen(
1341 '[download] Skipping already downloaded playlist: %s'
1342 % ie_result.get('title') or ie_result.get('id'))
1343 return
1344
1345 self._playlist_level += 1
1346 self._playlist_urls.add(webpage_url)
1347 self._sanitize_thumbnails(ie_result)
1348 try:
1349 return self.__process_playlist(ie_result, download)
1350 finally:
1351 self._playlist_level -= 1
1352 if not self._playlist_level:
1353 self._playlist_urls.clear()
1354 elif result_type == 'compat_list':
1355 self.report_warning(
1356 'Extractor %s returned a compat_list result. '
1357 'It needs to be updated.' % ie_result.get('extractor'))
1358
1359 def _fixup(r):
1360 self.add_extra_info(r, {
1361 'extractor': ie_result['extractor'],
1362 'webpage_url': ie_result['webpage_url'],
1363 'webpage_url_basename': url_basename(ie_result['webpage_url']),
1364 'extractor_key': ie_result['extractor_key'],
1365 })
1366 return r
1367 ie_result['entries'] = [
1368 self.process_ie_result(_fixup(r), download, extra_info)
1369 for r in ie_result['entries']
1370 ]
1371 return ie_result
1372 else:
1373 raise Exception('Invalid result type: %s' % result_type)
1374
1375 def _ensure_dir_exists(self, path):
1376 return make_dir(path, self.report_error)
1377
1378 def __process_playlist(self, ie_result, download):
1379 # We process each entry in the playlist
1380 playlist = ie_result.get('title') or ie_result.get('id')
1381 self.to_screen('[download] Downloading playlist: %s' % playlist)
1382
1383 if 'entries' not in ie_result:
1384 raise EntryNotInPlaylist()
1385 incomplete_entries = bool(ie_result.get('requested_entries'))
1386 if incomplete_entries:
1387 def fill_missing_entries(entries, indexes):
1388 ret = [None] * max(*indexes)
1389 for i, entry in zip(indexes, entries):
1390 ret[i - 1] = entry
1391 return ret
1392 ie_result['entries'] = fill_missing_entries(ie_result['entries'], ie_result['requested_entries'])
1393
1394 playlist_results = []
1395
1396 playliststart = self.params.get('playliststart', 1)
1397 playlistend = self.params.get('playlistend')
1398 # For backwards compatibility, interpret -1 as whole list
1399 if playlistend == -1:
1400 playlistend = None
1401
1402 playlistitems_str = self.params.get('playlist_items')
1403 playlistitems = None
1404 if playlistitems_str is not None:
1405 def iter_playlistitems(format):
1406 for string_segment in format.split(','):
1407 if '-' in string_segment:
1408 start, end = string_segment.split('-')
1409 for item in range(int(start), int(end) + 1):
1410 yield int(item)
1411 else:
1412 yield int(string_segment)
1413 playlistitems = orderedSet(iter_playlistitems(playlistitems_str))
1414
1415 ie_entries = ie_result['entries']
1416 msg = (
1417 'Downloading %d videos' if not isinstance(ie_entries, list)
1418 else 'Collected %d videos; downloading %%d of them' % len(ie_entries))
1419 if not isinstance(ie_entries, (list, PagedList)):
1420 ie_entries = LazyList(ie_entries)
1421
1422 def get_entry(i):
1423 return YoutubeDL.__handle_extraction_exceptions(
1424 lambda self, i: ie_entries[i - 1],
1425 False
1426 )(self, i)
1427
1428 entries = []
1429 for i in playlistitems or itertools.count(playliststart):
1430 if playlistitems is None and playlistend is not None and playlistend < i:
1431 break
1432 entry = None
1433 try:
1434 entry = get_entry(i)
1435 if entry is None:
1436 raise EntryNotInPlaylist()
1437 except (IndexError, EntryNotInPlaylist):
1438 if incomplete_entries:
1439 raise EntryNotInPlaylist()
1440 elif not playlistitems:
1441 break
1442 entries.append(entry)
1443 try:
1444 if entry is not None:
1445 self._match_entry(entry, incomplete=True, silent=True)
1446 except (ExistingVideoReached, RejectedVideoReached):
1447 break
1448 ie_result['entries'] = entries
1449
1450 # Save playlist_index before re-ordering
1451 entries = [
1452 ((playlistitems[i - 1] if playlistitems else i), entry)
1453 for i, entry in enumerate(entries, 1)
1454 if entry is not None]
1455 n_entries = len(entries)
1456
1457 if not playlistitems and (playliststart or playlistend):
1458 playlistitems = list(range(playliststart, playliststart + n_entries))
1459 ie_result['requested_entries'] = playlistitems
1460
1461 if self.params.get('allow_playlist_files', True):
1462 ie_copy = {
1463 'playlist': playlist,
1464 'playlist_id': ie_result.get('id'),
1465 'playlist_title': ie_result.get('title'),
1466 'playlist_uploader': ie_result.get('uploader'),
1467 'playlist_uploader_id': ie_result.get('uploader_id'),
1468 'playlist_index': 0,
1469 }
1470 ie_copy.update(dict(ie_result))
1471
1472 if self.params.get('writeinfojson', False):
1473 infofn = self.prepare_filename(ie_copy, 'pl_infojson')
1474 if not self._ensure_dir_exists(encodeFilename(infofn)):
1475 return
1476 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(infofn)):
1477 self.to_screen('[info] Playlist metadata is already present')
1478 else:
1479 self.to_screen('[info] Writing playlist metadata as JSON to: ' + infofn)
1480 try:
1481 write_json_file(self.sanitize_info(ie_result, self.params.get('clean_infojson', True)), infofn)
1482 except (OSError, IOError):
1483 self.report_error('Cannot write playlist metadata to JSON file ' + infofn)
1484
1485 # TODO: This should be passed to ThumbnailsConvertor if necessary
1486 self._write_thumbnails(ie_copy, self.prepare_filename(ie_copy, 'pl_thumbnail'))
1487
1488 if self.params.get('writedescription', False):
1489 descfn = self.prepare_filename(ie_copy, 'pl_description')
1490 if not self._ensure_dir_exists(encodeFilename(descfn)):
1491 return
1492 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(descfn)):
1493 self.to_screen('[info] Playlist description is already present')
1494 elif ie_result.get('description') is None:
1495 self.report_warning('There\'s no playlist description to write.')
1496 else:
1497 try:
1498 self.to_screen('[info] Writing playlist description to: ' + descfn)
1499 with io.open(encodeFilename(descfn), 'w', encoding='utf-8') as descfile:
1500 descfile.write(ie_result['description'])
1501 except (OSError, IOError):
1502 self.report_error('Cannot write playlist description file ' + descfn)
1503 return
1504
1505 if self.params.get('playlistreverse', False):
1506 entries = entries[::-1]
1507 if self.params.get('playlistrandom', False):
1508 random.shuffle(entries)
1509
1510 x_forwarded_for = ie_result.get('__x_forwarded_for_ip')
1511
1512 self.to_screen('[%s] playlist %s: %s' % (ie_result['extractor'], playlist, msg % n_entries))
1513 failures = 0
1514 max_failures = self.params.get('skip_playlist_after_errors') or float('inf')
1515 for i, entry_tuple in enumerate(entries, 1):
1516 playlist_index, entry = entry_tuple
1517 if 'playlist_index' in self.params.get('compat_options', []):
1518 playlist_index = playlistitems[i - 1] if playlistitems else i
1519 self.to_screen('[download] Downloading video %s of %s' % (i, n_entries))
1520 # This __x_forwarded_for_ip thing is a bit ugly but requires
1521 # minimal changes
1522 if x_forwarded_for:
1523 entry['__x_forwarded_for_ip'] = x_forwarded_for
1524 extra = {
1525 'n_entries': n_entries,
1526 '_last_playlist_index': max(playlistitems) if playlistitems else (playlistend or n_entries),
1527 'playlist_index': playlist_index,
1528 'playlist_autonumber': i,
1529 'playlist': playlist,
1530 'playlist_id': ie_result.get('id'),
1531 'playlist_title': ie_result.get('title'),
1532 'playlist_uploader': ie_result.get('uploader'),
1533 'playlist_uploader_id': ie_result.get('uploader_id'),
1534 'extractor': ie_result['extractor'],
1535 'webpage_url': ie_result['webpage_url'],
1536 'webpage_url_basename': url_basename(ie_result['webpage_url']),
1537 'extractor_key': ie_result['extractor_key'],
1538 }
1539
1540 if self._match_entry(entry, incomplete=True) is not None:
1541 continue
1542
1543 entry_result = self.__process_iterable_entry(entry, download, extra)
1544 if not entry_result:
1545 failures += 1
1546 if failures >= max_failures:
1547 self.report_error(
1548 'Skipping the remaining entries in playlist "%s" since %d items failed extraction' % (playlist, failures))
1549 break
1550 # TODO: skip failed (empty) entries?
1551 playlist_results.append(entry_result)
1552 ie_result['entries'] = playlist_results
1553 self.to_screen('[download] Finished downloading playlist: %s' % playlist)
1554 return ie_result
1555
1556 @__handle_extraction_exceptions
1557 def __process_iterable_entry(self, entry, download, extra_info):
1558 return self.process_ie_result(
1559 entry, download=download, extra_info=extra_info)
1560
1561 def _build_format_filter(self, filter_spec):
1562 " Returns a function to filter the formats according to the filter_spec "
1563
1564 OPERATORS = {
1565 '<': operator.lt,
1566 '<=': operator.le,
1567 '>': operator.gt,
1568 '>=': operator.ge,
1569 '=': operator.eq,
1570 '!=': operator.ne,
1571 }
1572 operator_rex = re.compile(r'''(?x)\s*
1573 (?P<key>width|height|tbr|abr|vbr|asr|filesize|filesize_approx|fps)\s*
1574 (?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
1575 (?P<value>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?)\s*
1576 ''' % '|'.join(map(re.escape, OPERATORS.keys())))
1577 m = operator_rex.fullmatch(filter_spec)
1578 if m:
1579 try:
1580 comparison_value = int(m.group('value'))
1581 except ValueError:
1582 comparison_value = parse_filesize(m.group('value'))
1583 if comparison_value is None:
1584 comparison_value = parse_filesize(m.group('value') + 'B')
1585 if comparison_value is None:
1586 raise ValueError(
1587 'Invalid value %r in format specification %r' % (
1588 m.group('value'), filter_spec))
1589 op = OPERATORS[m.group('op')]
1590
1591 if not m:
1592 STR_OPERATORS = {
1593 '=': operator.eq,
1594 '^=': lambda attr, value: attr.startswith(value),
1595 '$=': lambda attr, value: attr.endswith(value),
1596 '*=': lambda attr, value: value in attr,
1597 }
1598 str_operator_rex = re.compile(r'''(?x)\s*
1599 (?P<key>[a-zA-Z0-9._-]+)\s*
1600 (?P<negation>!\s*)?(?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
1601 (?P<value>[a-zA-Z0-9._-]+)\s*
1602 ''' % '|'.join(map(re.escape, STR_OPERATORS.keys())))
1603 m = str_operator_rex.fullmatch(filter_spec)
1604 if m:
1605 comparison_value = m.group('value')
1606 str_op = STR_OPERATORS[m.group('op')]
1607 if m.group('negation'):
1608 op = lambda attr, value: not str_op(attr, value)
1609 else:
1610 op = str_op
1611
1612 if not m:
1613 raise SyntaxError('Invalid filter specification %r' % filter_spec)
1614
1615 def _filter(f):
1616 actual_value = f.get(m.group('key'))
1617 if actual_value is None:
1618 return m.group('none_inclusive')
1619 return op(actual_value, comparison_value)
1620 return _filter
1621
1622 def _default_format_spec(self, info_dict, download=True):
1623
1624 def can_merge():
1625 merger = FFmpegMergerPP(self)
1626 return merger.available and merger.can_merge()
1627
1628 prefer_best = (
1629 not self.params.get('simulate')
1630 and download
1631 and (
1632 not can_merge()
1633 or info_dict.get('is_live', False)
1634 or self.outtmpl_dict['default'] == '-'))
1635 compat = (
1636 prefer_best
1637 or self.params.get('allow_multiple_audio_streams', False)
1638 or 'format-spec' in self.params.get('compat_opts', []))
1639
1640 return (
1641 'best/bestvideo+bestaudio' if prefer_best
1642 else 'bestvideo*+bestaudio/best' if not compat
1643 else 'bestvideo+bestaudio/best')
1644
1645 def build_format_selector(self, format_spec):
1646 def syntax_error(note, start):
1647 message = (
1648 'Invalid format specification: '
1649 '{0}\n\t{1}\n\t{2}^'.format(note, format_spec, ' ' * start[1]))
1650 return SyntaxError(message)
1651
1652 PICKFIRST = 'PICKFIRST'
1653 MERGE = 'MERGE'
1654 SINGLE = 'SINGLE'
1655 GROUP = 'GROUP'
1656 FormatSelector = collections.namedtuple('FormatSelector', ['type', 'selector', 'filters'])
1657
1658 allow_multiple_streams = {'audio': self.params.get('allow_multiple_audio_streams', False),
1659 'video': self.params.get('allow_multiple_video_streams', False)}
1660
1661 check_formats = self.params.get('check_formats')
1662
1663 def _parse_filter(tokens):
1664 filter_parts = []
1665 for type, string, start, _, _ in tokens:
1666 if type == tokenize.OP and string == ']':
1667 return ''.join(filter_parts)
1668 else:
1669 filter_parts.append(string)
1670
1671 def _remove_unused_ops(tokens):
1672 # Remove operators that we don't use and join them with the surrounding strings
1673 # for example: 'mp4' '-' 'baseline' '-' '16x9' is converted to 'mp4-baseline-16x9'
1674 ALLOWED_OPS = ('/', '+', ',', '(', ')')
1675 last_string, last_start, last_end, last_line = None, None, None, None
1676 for type, string, start, end, line in tokens:
1677 if type == tokenize.OP and string == '[':
1678 if last_string:
1679 yield tokenize.NAME, last_string, last_start, last_end, last_line
1680 last_string = None
1681 yield type, string, start, end, line
1682 # everything inside brackets will be handled by _parse_filter
1683 for type, string, start, end, line in tokens:
1684 yield type, string, start, end, line
1685 if type == tokenize.OP and string == ']':
1686 break
1687 elif type == tokenize.OP and string in ALLOWED_OPS:
1688 if last_string:
1689 yield tokenize.NAME, last_string, last_start, last_end, last_line
1690 last_string = None
1691 yield type, string, start, end, line
1692 elif type in [tokenize.NAME, tokenize.NUMBER, tokenize.OP]:
1693 if not last_string:
1694 last_string = string
1695 last_start = start
1696 last_end = end
1697 else:
1698 last_string += string
1699 if last_string:
1700 yield tokenize.NAME, last_string, last_start, last_end, last_line
1701
1702 def _parse_format_selection(tokens, inside_merge=False, inside_choice=False, inside_group=False):
1703 selectors = []
1704 current_selector = None
1705 for type, string, start, _, _ in tokens:
1706 # ENCODING is only defined in python 3.x
1707 if type == getattr(tokenize, 'ENCODING', None):
1708 continue
1709 elif type in [tokenize.NAME, tokenize.NUMBER]:
1710 current_selector = FormatSelector(SINGLE, string, [])
1711 elif type == tokenize.OP:
1712 if string == ')':
1713 if not inside_group:
1714 # ')' will be handled by the parentheses group
1715 tokens.restore_last_token()
1716 break
1717 elif inside_merge and string in ['/', ',']:
1718 tokens.restore_last_token()
1719 break
1720 elif inside_choice and string == ',':
1721 tokens.restore_last_token()
1722 break
1723 elif string == ',':
1724 if not current_selector:
1725 raise syntax_error('"," must follow a format selector', start)
1726 selectors.append(current_selector)
1727 current_selector = None
1728 elif string == '/':
1729 if not current_selector:
1730 raise syntax_error('"/" must follow a format selector', start)
1731 first_choice = current_selector
1732 second_choice = _parse_format_selection(tokens, inside_choice=True)
1733 current_selector = FormatSelector(PICKFIRST, (first_choice, second_choice), [])
1734 elif string == '[':
1735 if not current_selector:
1736 current_selector = FormatSelector(SINGLE, 'best', [])
1737 format_filter = _parse_filter(tokens)
1738 current_selector.filters.append(format_filter)
1739 elif string == '(':
1740 if current_selector:
1741 raise syntax_error('Unexpected "("', start)
1742 group = _parse_format_selection(tokens, inside_group=True)
1743 current_selector = FormatSelector(GROUP, group, [])
1744 elif string == '+':
1745 if not current_selector:
1746 raise syntax_error('Unexpected "+"', start)
1747 selector_1 = current_selector
1748 selector_2 = _parse_format_selection(tokens, inside_merge=True)
1749 if not selector_2:
1750 raise syntax_error('Expected a selector', start)
1751 current_selector = FormatSelector(MERGE, (selector_1, selector_2), [])
1752 else:
1753 raise syntax_error('Operator not recognized: "{0}"'.format(string), start)
1754 elif type == tokenize.ENDMARKER:
1755 break
1756 if current_selector:
1757 selectors.append(current_selector)
1758 return selectors
1759
1760 def _merge(formats_pair):
1761 format_1, format_2 = formats_pair
1762
1763 formats_info = []
1764 formats_info.extend(format_1.get('requested_formats', (format_1,)))
1765 formats_info.extend(format_2.get('requested_formats', (format_2,)))
1766
1767 if not allow_multiple_streams['video'] or not allow_multiple_streams['audio']:
1768 get_no_more = {'video': False, 'audio': False}
1769 for (i, fmt_info) in enumerate(formats_info):
1770 if fmt_info.get('acodec') == fmt_info.get('vcodec') == 'none':
1771 formats_info.pop(i)
1772 continue
1773 for aud_vid in ['audio', 'video']:
1774 if not allow_multiple_streams[aud_vid] and fmt_info.get(aud_vid[0] + 'codec') != 'none':
1775 if get_no_more[aud_vid]:
1776 formats_info.pop(i)
1777 break
1778 get_no_more[aud_vid] = True
1779
1780 if len(formats_info) == 1:
1781 return formats_info[0]
1782
1783 video_fmts = [fmt_info for fmt_info in formats_info if fmt_info.get('vcodec') != 'none']
1784 audio_fmts = [fmt_info for fmt_info in formats_info if fmt_info.get('acodec') != 'none']
1785
1786 the_only_video = video_fmts[0] if len(video_fmts) == 1 else None
1787 the_only_audio = audio_fmts[0] if len(audio_fmts) == 1 else None
1788
1789 output_ext = self.params.get('merge_output_format')
1790 if not output_ext:
1791 if the_only_video:
1792 output_ext = the_only_video['ext']
1793 elif the_only_audio and not video_fmts:
1794 output_ext = the_only_audio['ext']
1795 else:
1796 output_ext = 'mkv'
1797
1798 new_dict = {
1799 'requested_formats': formats_info,
1800 'format': '+'.join(fmt_info.get('format') for fmt_info in formats_info),
1801 'format_id': '+'.join(fmt_info.get('format_id') for fmt_info in formats_info),
1802 'ext': output_ext,
1803 }
1804
1805 if the_only_video:
1806 new_dict.update({
1807 'width': the_only_video.get('width'),
1808 'height': the_only_video.get('height'),
1809 'resolution': the_only_video.get('resolution') or self.format_resolution(the_only_video),
1810 'fps': the_only_video.get('fps'),
1811 'vcodec': the_only_video.get('vcodec'),
1812 'vbr': the_only_video.get('vbr'),
1813 'stretched_ratio': the_only_video.get('stretched_ratio'),
1814 })
1815
1816 if the_only_audio:
1817 new_dict.update({
1818 'acodec': the_only_audio.get('acodec'),
1819 'abr': the_only_audio.get('abr'),
1820 })
1821
1822 return new_dict
1823
1824 def _check_formats(formats):
1825 if not check_formats:
1826 yield from formats
1827 return
1828 for f in formats:
1829 self.to_screen('[info] Testing format %s' % f['format_id'])
1830 temp_file = tempfile.NamedTemporaryFile(
1831 suffix='.tmp', delete=False,
1832 dir=self.get_output_path('temp') or None)
1833 temp_file.close()
1834 try:
1835 success, _ = self.dl(temp_file.name, f, test=True)
1836 except (DownloadError, IOError, OSError, ValueError) + network_exceptions:
1837 success = False
1838 finally:
1839 if os.path.exists(temp_file.name):
1840 try:
1841 os.remove(temp_file.name)
1842 except OSError:
1843 self.report_warning('Unable to delete temporary file "%s"' % temp_file.name)
1844 if success:
1845 yield f
1846 else:
1847 self.to_screen('[info] Unable to download format %s. Skipping...' % f['format_id'])
1848
1849 def _build_selector_function(selector):
1850 if isinstance(selector, list): # ,
1851 fs = [_build_selector_function(s) for s in selector]
1852
1853 def selector_function(ctx):
1854 for f in fs:
1855 yield from f(ctx)
1856 return selector_function
1857
1858 elif selector.type == GROUP: # ()
1859 selector_function = _build_selector_function(selector.selector)
1860
1861 elif selector.type == PICKFIRST: # /
1862 fs = [_build_selector_function(s) for s in selector.selector]
1863
1864 def selector_function(ctx):
1865 for f in fs:
1866 picked_formats = list(f(ctx))
1867 if picked_formats:
1868 return picked_formats
1869 return []
1870
1871 elif selector.type == MERGE: # +
1872 selector_1, selector_2 = map(_build_selector_function, selector.selector)
1873
1874 def selector_function(ctx):
1875 for pair in itertools.product(
1876 selector_1(copy.deepcopy(ctx)), selector_2(copy.deepcopy(ctx))):
1877 yield _merge(pair)
1878
1879 elif selector.type == SINGLE: # atom
1880 format_spec = selector.selector or 'best'
1881
1882 # TODO: Add allvideo, allaudio etc by generalizing the code with best/worst selector
1883 if format_spec == 'all':
1884 def selector_function(ctx):
1885 yield from _check_formats(ctx['formats'])
1886 elif format_spec == 'mergeall':
1887 def selector_function(ctx):
1888 formats = list(_check_formats(ctx['formats']))
1889 if not formats:
1890 return
1891 merged_format = formats[-1]
1892 for f in formats[-2::-1]:
1893 merged_format = _merge((merged_format, f))
1894 yield merged_format
1895
1896 else:
1897 format_fallback, format_reverse, format_idx = False, True, 1
1898 mobj = re.match(
1899 r'(?P<bw>best|worst|b|w)(?P<type>video|audio|v|a)?(?P<mod>\*)?(?:\.(?P<n>[1-9]\d*))?$',
1900 format_spec)
1901 if mobj is not None:
1902 format_idx = int_or_none(mobj.group('n'), default=1)
1903 format_reverse = mobj.group('bw')[0] == 'b'
1904 format_type = (mobj.group('type') or [None])[0]
1905 not_format_type = {'v': 'a', 'a': 'v'}.get(format_type)
1906 format_modified = mobj.group('mod') is not None
1907
1908 format_fallback = not format_type and not format_modified # for b, w
1909 _filter_f = (
1910 (lambda f: f.get('%scodec' % format_type) != 'none')
1911 if format_type and format_modified # bv*, ba*, wv*, wa*
1912 else (lambda f: f.get('%scodec' % not_format_type) == 'none')
1913 if format_type # bv, ba, wv, wa
1914 else (lambda f: f.get('vcodec') != 'none' and f.get('acodec') != 'none')
1915 if not format_modified # b, w
1916 else lambda f: True) # b*, w*
1917 filter_f = lambda f: _filter_f(f) and (
1918 f.get('vcodec') != 'none' or f.get('acodec') != 'none')
1919 else:
1920 filter_f = ((lambda f: f.get('ext') == format_spec)
1921 if format_spec in ['mp4', 'flv', 'webm', '3gp', 'm4a', 'mp3', 'ogg', 'aac', 'wav'] # extension
1922 else (lambda f: f.get('format_id') == format_spec)) # id
1923
1924 def selector_function(ctx):
1925 formats = list(ctx['formats'])
1926 matches = list(filter(filter_f, formats)) if filter_f is not None else formats
1927 if format_fallback and ctx['incomplete_formats'] and not matches:
1928 # for extractors with incomplete formats (audio only (soundcloud)
1929 # or video only (imgur)) best/worst will fallback to
1930 # best/worst {video,audio}-only format
1931 matches = formats
1932 matches = LazyList(_check_formats(matches[::-1 if format_reverse else 1]))
1933 try:
1934 yield matches[format_idx - 1]
1935 except IndexError:
1936 return
1937
1938 filters = [self._build_format_filter(f) for f in selector.filters]
1939
1940 def final_selector(ctx):
1941 ctx_copy = copy.deepcopy(ctx)
1942 for _filter in filters:
1943 ctx_copy['formats'] = list(filter(_filter, ctx_copy['formats']))
1944 return selector_function(ctx_copy)
1945 return final_selector
1946
1947 stream = io.BytesIO(format_spec.encode('utf-8'))
1948 try:
1949 tokens = list(_remove_unused_ops(compat_tokenize_tokenize(stream.readline)))
1950 except tokenize.TokenError:
1951 raise syntax_error('Missing closing/opening brackets or parenthesis', (0, len(format_spec)))
1952
1953 class TokenIterator(object):
1954 def __init__(self, tokens):
1955 self.tokens = tokens
1956 self.counter = 0
1957
1958 def __iter__(self):
1959 return self
1960
1961 def __next__(self):
1962 if self.counter >= len(self.tokens):
1963 raise StopIteration()
1964 value = self.tokens[self.counter]
1965 self.counter += 1
1966 return value
1967
1968 next = __next__
1969
1970 def restore_last_token(self):
1971 self.counter -= 1
1972
1973 parsed_selector = _parse_format_selection(iter(TokenIterator(tokens)))
1974 return _build_selector_function(parsed_selector)
1975
1976 def _calc_headers(self, info_dict):
1977 res = std_headers.copy()
1978
1979 add_headers = info_dict.get('http_headers')
1980 if add_headers:
1981 res.update(add_headers)
1982
1983 cookies = self._calc_cookies(info_dict)
1984 if cookies:
1985 res['Cookie'] = cookies
1986
1987 if 'X-Forwarded-For' not in res:
1988 x_forwarded_for_ip = info_dict.get('__x_forwarded_for_ip')
1989 if x_forwarded_for_ip:
1990 res['X-Forwarded-For'] = x_forwarded_for_ip
1991
1992 return res
1993
1994 def _calc_cookies(self, info_dict):
1995 pr = sanitized_Request(info_dict['url'])
1996 self.cookiejar.add_cookie_header(pr)
1997 return pr.get_header('Cookie')
1998
1999 def _sanitize_thumbnails(self, info_dict):
2000 thumbnails = info_dict.get('thumbnails')
2001 if thumbnails is None:
2002 thumbnail = info_dict.get('thumbnail')
2003 if thumbnail:
2004 info_dict['thumbnails'] = thumbnails = [{'url': thumbnail}]
2005 if thumbnails:
2006 thumbnails.sort(key=lambda t: (
2007 t.get('preference') if t.get('preference') is not None else -1,
2008 t.get('width') if t.get('width') is not None else -1,
2009 t.get('height') if t.get('height') is not None else -1,
2010 t.get('id') if t.get('id') is not None else '',
2011 t.get('url')))
2012
2013 def thumbnail_tester():
2014 if self.params.get('check_formats'):
2015 test_all = True
2016 to_screen = lambda msg: self.to_screen(f'[info] {msg}')
2017 else:
2018 test_all = False
2019 to_screen = self.write_debug
2020
2021 def test_thumbnail(t):
2022 if not test_all and not t.get('_test_url'):
2023 return True
2024 to_screen('Testing thumbnail %s' % t['id'])
2025 try:
2026 self.urlopen(HEADRequest(t['url']))
2027 except network_exceptions as err:
2028 to_screen('Unable to connect to thumbnail %s URL "%s" - %s. Skipping...' % (
2029 t['id'], t['url'], error_to_compat_str(err)))
2030 return False
2031 return True
2032
2033 return test_thumbnail
2034
2035 for i, t in enumerate(thumbnails):
2036 if t.get('id') is None:
2037 t['id'] = '%d' % i
2038 if t.get('width') and t.get('height'):
2039 t['resolution'] = '%dx%d' % (t['width'], t['height'])
2040 t['url'] = sanitize_url(t['url'])
2041
2042 if self.params.get('check_formats') is not False:
2043 info_dict['thumbnails'] = LazyList(filter(thumbnail_tester(), thumbnails[::-1])).reverse()
2044 else:
2045 info_dict['thumbnails'] = thumbnails
2046
2047 def process_video_result(self, info_dict, download=True):
2048 assert info_dict.get('_type', 'video') == 'video'
2049
2050 if 'id' not in info_dict:
2051 raise ExtractorError('Missing "id" field in extractor result')
2052 if 'title' not in info_dict:
2053 raise ExtractorError('Missing "title" field in extractor result')
2054
2055 def report_force_conversion(field, field_not, conversion):
2056 self.report_warning(
2057 '"%s" field is not %s - forcing %s conversion, there is an error in extractor'
2058 % (field, field_not, conversion))
2059
2060 def sanitize_string_field(info, string_field):
2061 field = info.get(string_field)
2062 if field is None or isinstance(field, compat_str):
2063 return
2064 report_force_conversion(string_field, 'a string', 'string')
2065 info[string_field] = compat_str(field)
2066
2067 def sanitize_numeric_fields(info):
2068 for numeric_field in self._NUMERIC_FIELDS:
2069 field = info.get(numeric_field)
2070 if field is None or isinstance(field, compat_numeric_types):
2071 continue
2072 report_force_conversion(numeric_field, 'numeric', 'int')
2073 info[numeric_field] = int_or_none(field)
2074
2075 sanitize_string_field(info_dict, 'id')
2076 sanitize_numeric_fields(info_dict)
2077
2078 if 'playlist' not in info_dict:
2079 # It isn't part of a playlist
2080 info_dict['playlist'] = None
2081 info_dict['playlist_index'] = None
2082
2083 self._sanitize_thumbnails(info_dict)
2084
2085 thumbnail = info_dict.get('thumbnail')
2086 thumbnails = info_dict.get('thumbnails')
2087 if thumbnail:
2088 info_dict['thumbnail'] = sanitize_url(thumbnail)
2089 elif thumbnails:
2090 info_dict['thumbnail'] = thumbnails[-1]['url']
2091
2092 if info_dict.get('display_id') is None and 'id' in info_dict:
2093 info_dict['display_id'] = info_dict['id']
2094
2095 for ts_key, date_key in (
2096 ('timestamp', 'upload_date'),
2097 ('release_timestamp', 'release_date'),
2098 ):
2099 if info_dict.get(date_key) is None and info_dict.get(ts_key) is not None:
2100 # Working around out-of-range timestamp values (e.g. negative ones on Windows,
2101 # see http://bugs.python.org/issue1646728)
2102 try:
2103 upload_date = datetime.datetime.utcfromtimestamp(info_dict[ts_key])
2104 info_dict[date_key] = upload_date.strftime('%Y%m%d')
2105 except (ValueError, OverflowError, OSError):
2106 pass
2107
2108 live_keys = ('is_live', 'was_live')
2109 live_status = info_dict.get('live_status')
2110 if live_status is None:
2111 for key in live_keys:
2112 if info_dict.get(key) is False:
2113 continue
2114 if info_dict.get(key):
2115 live_status = key
2116 break
2117 if all(info_dict.get(key) is False for key in live_keys):
2118 live_status = 'not_live'
2119 if live_status:
2120 info_dict['live_status'] = live_status
2121 for key in live_keys:
2122 if info_dict.get(key) is None:
2123 info_dict[key] = (live_status == key)
2124
2125 # Auto generate title fields corresponding to the *_number fields when missing
2126 # in order to always have clean titles. This is very common for TV series.
2127 for field in ('chapter', 'season', 'episode'):
2128 if info_dict.get('%s_number' % field) is not None and not info_dict.get(field):
2129 info_dict[field] = '%s %d' % (field.capitalize(), info_dict['%s_number' % field])
2130
2131 for cc_kind in ('subtitles', 'automatic_captions'):
2132 cc = info_dict.get(cc_kind)
2133 if cc:
2134 for _, subtitle in cc.items():
2135 for subtitle_format in subtitle:
2136 if subtitle_format.get('url'):
2137 subtitle_format['url'] = sanitize_url(subtitle_format['url'])
2138 if subtitle_format.get('ext') is None:
2139 subtitle_format['ext'] = determine_ext(subtitle_format['url']).lower()
2140
2141 automatic_captions = info_dict.get('automatic_captions')
2142 subtitles = info_dict.get('subtitles')
2143
2144 info_dict['requested_subtitles'] = self.process_subtitles(
2145 info_dict['id'], subtitles, automatic_captions)
2146
2147 # We now pick which formats have to be downloaded
2148 if info_dict.get('formats') is None:
2149 # There's only one format available
2150 formats = [info_dict]
2151 else:
2152 formats = info_dict['formats']
2153
2154 if not formats:
2155 if not self.params.get('ignore_no_formats_error'):
2156 raise ExtractorError('No video formats found!')
2157 else:
2158 self.report_warning('No video formats found!')
2159
2160 def is_wellformed(f):
2161 url = f.get('url')
2162 if not url:
2163 self.report_warning(
2164 '"url" field is missing or empty - skipping format, '
2165 'there is an error in extractor')
2166 return False
2167 if isinstance(url, bytes):
2168 sanitize_string_field(f, 'url')
2169 return True
2170
2171 # Filter out malformed formats for better extraction robustness
2172 formats = list(filter(is_wellformed, formats))
2173
2174 formats_dict = {}
2175
2176 # We check that all the formats have the format and format_id fields
2177 for i, format in enumerate(formats):
2178 sanitize_string_field(format, 'format_id')
2179 sanitize_numeric_fields(format)
2180 format['url'] = sanitize_url(format['url'])
2181 if not format.get('format_id'):
2182 format['format_id'] = compat_str(i)
2183 else:
2184 # Sanitize format_id from characters used in format selector expression
2185 format['format_id'] = re.sub(r'[\s,/+\[\]()]', '_', format['format_id'])
2186 format_id = format['format_id']
2187 if format_id not in formats_dict:
2188 formats_dict[format_id] = []
2189 formats_dict[format_id].append(format)
2190
2191 # Make sure all formats have unique format_id
2192 for format_id, ambiguous_formats in formats_dict.items():
2193 if len(ambiguous_formats) > 1:
2194 for i, format in enumerate(ambiguous_formats):
2195 format['format_id'] = '%s-%d' % (format_id, i)
2196
2197 for i, format in enumerate(formats):
2198 if format.get('format') is None:
2199 format['format'] = '{id} - {res}{note}'.format(
2200 id=format['format_id'],
2201 res=self.format_resolution(format),
2202 note=format_field(format, 'format_note', ' (%s)'),
2203 )
2204 # Automatically determine file extension if missing
2205 if format.get('ext') is None:
2206 format['ext'] = determine_ext(format['url']).lower()
2207 # Automatically determine protocol if missing (useful for format
2208 # selection purposes)
2209 if format.get('protocol') is None:
2210 format['protocol'] = determine_protocol(format)
2211 # Add HTTP headers, so that external programs can use them from the
2212 # json output
2213 full_format_info = info_dict.copy()
2214 full_format_info.update(format)
2215 format['http_headers'] = self._calc_headers(full_format_info)
2216 # Remove private housekeeping stuff
2217 if '__x_forwarded_for_ip' in info_dict:
2218 del info_dict['__x_forwarded_for_ip']
2219
2220 # TODO Central sorting goes here
2221
2222 if formats and formats[0] is not info_dict:
2223 # only set the 'formats' fields if the original info_dict list them
2224 # otherwise we end up with a circular reference, the first (and unique)
2225 # element in the 'formats' field in info_dict is info_dict itself,
2226 # which can't be exported to json
2227 info_dict['formats'] = formats
2228
2229 info_dict, _ = self.pre_process(info_dict)
2230
2231 if self.params.get('list_thumbnails'):
2232 self.list_thumbnails(info_dict)
2233 if self.params.get('listformats'):
2234 if not info_dict.get('formats'):
2235 raise ExtractorError('No video formats found', expected=True)
2236 self.list_formats(info_dict)
2237 if self.params.get('listsubtitles'):
2238 if 'automatic_captions' in info_dict:
2239 self.list_subtitles(
2240 info_dict['id'], automatic_captions, 'automatic captions')
2241 self.list_subtitles(info_dict['id'], subtitles, 'subtitles')
2242 list_only = self.params.get('simulate') is None and (
2243 self.params.get('list_thumbnails') or self.params.get('listformats') or self.params.get('listsubtitles'))
2244 if list_only:
2245 # Without this printing, -F --print-json will not work
2246 self.__forced_printings(info_dict, self.prepare_filename(info_dict), incomplete=True)
2247 return
2248
2249 format_selector = self.format_selector
2250 if format_selector is None:
2251 req_format = self._default_format_spec(info_dict, download=download)
2252 self.write_debug('Default format spec: %s' % req_format)
2253 format_selector = self.build_format_selector(req_format)
2254
2255 # While in format selection we may need to have an access to the original
2256 # format set in order to calculate some metrics or do some processing.
2257 # For now we need to be able to guess whether original formats provided
2258 # by extractor are incomplete or not (i.e. whether extractor provides only
2259 # video-only or audio-only formats) for proper formats selection for
2260 # extractors with such incomplete formats (see
2261 # https://github.com/ytdl-org/youtube-dl/pull/5556).
2262 # Since formats may be filtered during format selection and may not match
2263 # the original formats the results may be incorrect. Thus original formats
2264 # or pre-calculated metrics should be passed to format selection routines
2265 # as well.
2266 # We will pass a context object containing all necessary additional data
2267 # instead of just formats.
2268 # This fixes incorrect format selection issue (see
2269 # https://github.com/ytdl-org/youtube-dl/issues/10083).
2270 incomplete_formats = (
2271 # All formats are video-only or
2272 all(f.get('vcodec') != 'none' and f.get('acodec') == 'none' for f in formats)
2273 # all formats are audio-only
2274 or all(f.get('vcodec') == 'none' and f.get('acodec') != 'none' for f in formats))
2275
2276 ctx = {
2277 'formats': formats,
2278 'incomplete_formats': incomplete_formats,
2279 }
2280
2281 formats_to_download = list(format_selector(ctx))
2282 if not formats_to_download:
2283 if not self.params.get('ignore_no_formats_error'):
2284 raise ExtractorError('Requested format is not available', expected=True)
2285 else:
2286 self.report_warning('Requested format is not available')
2287 # Process what we can, even without any available formats.
2288 self.process_info(dict(info_dict))
2289 elif download:
2290 self.to_screen(
2291 '[info] %s: Downloading %d format(s): %s' % (
2292 info_dict['id'], len(formats_to_download),
2293 ", ".join([f['format_id'] for f in formats_to_download])))
2294 for fmt in formats_to_download:
2295 new_info = dict(info_dict)
2296 # Save a reference to the original info_dict so that it can be modified in process_info if needed
2297 new_info['__original_infodict'] = info_dict
2298 new_info.update(fmt)
2299 self.process_info(new_info)
2300 # We update the info dict with the best quality format (backwards compatibility)
2301 if formats_to_download:
2302 info_dict.update(formats_to_download[-1])
2303 return info_dict
2304
2305 def process_subtitles(self, video_id, normal_subtitles, automatic_captions):
2306 """Select the requested subtitles and their format"""
2307 available_subs = {}
2308 if normal_subtitles and self.params.get('writesubtitles'):
2309 available_subs.update(normal_subtitles)
2310 if automatic_captions and self.params.get('writeautomaticsub'):
2311 for lang, cap_info in automatic_captions.items():
2312 if lang not in available_subs:
2313 available_subs[lang] = cap_info
2314
2315 if (not self.params.get('writesubtitles') and not
2316 self.params.get('writeautomaticsub') or not
2317 available_subs):
2318 return None
2319
2320 all_sub_langs = available_subs.keys()
2321 if self.params.get('allsubtitles', False):
2322 requested_langs = all_sub_langs
2323 elif self.params.get('subtitleslangs', False):
2324 requested_langs = set()
2325 for lang in self.params.get('subtitleslangs'):
2326 if lang == 'all':
2327 requested_langs.update(all_sub_langs)
2328 continue
2329 discard = lang[0] == '-'
2330 if discard:
2331 lang = lang[1:]
2332 current_langs = filter(re.compile(lang + '$').match, all_sub_langs)
2333 if discard:
2334 for lang in current_langs:
2335 requested_langs.discard(lang)
2336 else:
2337 requested_langs.update(current_langs)
2338 elif 'en' in available_subs:
2339 requested_langs = ['en']
2340 else:
2341 requested_langs = [list(all_sub_langs)[0]]
2342 self.write_debug('Downloading subtitles: %s' % ', '.join(requested_langs))
2343
2344 formats_query = self.params.get('subtitlesformat', 'best')
2345 formats_preference = formats_query.split('/') if formats_query else []
2346 subs = {}
2347 for lang in requested_langs:
2348 formats = available_subs.get(lang)
2349 if formats is None:
2350 self.report_warning('%s subtitles not available for %s' % (lang, video_id))
2351 continue
2352 for ext in formats_preference:
2353 if ext == 'best':
2354 f = formats[-1]
2355 break
2356 matches = list(filter(lambda f: f['ext'] == ext, formats))
2357 if matches:
2358 f = matches[-1]
2359 break
2360 else:
2361 f = formats[-1]
2362 self.report_warning(
2363 'No subtitle format found matching "%s" for language %s, '
2364 'using %s' % (formats_query, lang, f['ext']))
2365 subs[lang] = f
2366 return subs
2367
2368 def __forced_printings(self, info_dict, filename, incomplete):
2369 def print_mandatory(field, actual_field=None):
2370 if actual_field is None:
2371 actual_field = field
2372 if (self.params.get('force%s' % field, False)
2373 and (not incomplete or info_dict.get(actual_field) is not None)):
2374 self.to_stdout(info_dict[actual_field])
2375
2376 def print_optional(field):
2377 if (self.params.get('force%s' % field, False)
2378 and info_dict.get(field) is not None):
2379 self.to_stdout(info_dict[field])
2380
2381 info_dict = info_dict.copy()
2382 if filename is not None:
2383 info_dict['filename'] = filename
2384 if info_dict.get('requested_formats') is not None:
2385 # For RTMP URLs, also include the playpath
2386 info_dict['urls'] = '\n'.join(f['url'] + f.get('play_path', '') for f in info_dict['requested_formats'])
2387 elif 'url' in info_dict:
2388 info_dict['urls'] = info_dict['url'] + info_dict.get('play_path', '')
2389
2390 if self.params.get('forceprint') or self.params.get('forcejson'):
2391 self.post_extract(info_dict)
2392 for tmpl in self.params.get('forceprint', []):
2393 if re.match(r'\w+$', tmpl):
2394 tmpl = '%({})s'.format(tmpl)
2395 tmpl, info_copy = self.prepare_outtmpl(tmpl, info_dict)
2396 self.to_stdout(self.escape_outtmpl(tmpl) % info_copy)
2397
2398 print_mandatory('title')
2399 print_mandatory('id')
2400 print_mandatory('url', 'urls')
2401 print_optional('thumbnail')
2402 print_optional('description')
2403 print_optional('filename')
2404 if self.params.get('forceduration') and info_dict.get('duration') is not None:
2405 self.to_stdout(formatSeconds(info_dict['duration']))
2406 print_mandatory('format')
2407
2408 if self.params.get('forcejson'):
2409 self.to_stdout(json.dumps(self.sanitize_info(info_dict)))
2410
2411 def dl(self, name, info, subtitle=False, test=False):
2412
2413 if test:
2414 verbose = self.params.get('verbose')
2415 params = {
2416 'test': True,
2417 'quiet': not verbose,
2418 'verbose': verbose,
2419 'noprogress': not verbose,
2420 'nopart': True,
2421 'skip_unavailable_fragments': False,
2422 'keep_fragments': False,
2423 'overwrites': True,
2424 '_no_ytdl_file': True,
2425 }
2426 else:
2427 params = self.params
2428 fd = get_suitable_downloader(info, params, to_stdout=(name == '-'))(self, params)
2429 if not test:
2430 for ph in self._progress_hooks:
2431 fd.add_progress_hook(ph)
2432 urls = '", "'.join([f['url'] for f in info.get('requested_formats', [])] or [info['url']])
2433 self.write_debug('Invoking downloader on "%s"' % urls)
2434 new_info = dict(info)
2435 if new_info.get('http_headers') is None:
2436 new_info['http_headers'] = self._calc_headers(new_info)
2437 return fd.download(name, new_info, subtitle)
2438
2439 def process_info(self, info_dict):
2440 """Process a single resolved IE result."""
2441
2442 assert info_dict.get('_type', 'video') == 'video'
2443
2444 max_downloads = self.params.get('max_downloads')
2445 if max_downloads is not None:
2446 if self._num_downloads >= int(max_downloads):
2447 raise MaxDownloadsReached()
2448
2449 # TODO: backward compatibility, to be removed
2450 info_dict['fulltitle'] = info_dict['title']
2451
2452 if 'format' not in info_dict and 'ext' in info_dict:
2453 info_dict['format'] = info_dict['ext']
2454
2455 if self._match_entry(info_dict) is not None:
2456 return
2457
2458 self.post_extract(info_dict)
2459 self._num_downloads += 1
2460
2461 # info_dict['_filename'] needs to be set for backward compatibility
2462 info_dict['_filename'] = full_filename = self.prepare_filename(info_dict, warn=True)
2463 temp_filename = self.prepare_filename(info_dict, 'temp')
2464 files_to_move = {}
2465
2466 # Forced printings
2467 self.__forced_printings(info_dict, full_filename, incomplete=('format' not in info_dict))
2468
2469 if self.params.get('simulate'):
2470 if self.params.get('force_write_download_archive', False):
2471 self.record_download_archive(info_dict)
2472
2473 # Do nothing else if in simulate mode
2474 return
2475
2476 if full_filename is None:
2477 return
2478
2479 if not self._ensure_dir_exists(encodeFilename(full_filename)):
2480 return
2481 if not self._ensure_dir_exists(encodeFilename(temp_filename)):
2482 return
2483
2484 if self.params.get('writedescription', False):
2485 descfn = self.prepare_filename(info_dict, 'description')
2486 if not self._ensure_dir_exists(encodeFilename(descfn)):
2487 return
2488 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(descfn)):
2489 self.to_screen('[info] Video description is already present')
2490 elif info_dict.get('description') is None:
2491 self.report_warning('There\'s no description to write.')
2492 else:
2493 try:
2494 self.to_screen('[info] Writing video description to: ' + descfn)
2495 with io.open(encodeFilename(descfn), 'w', encoding='utf-8') as descfile:
2496 descfile.write(info_dict['description'])
2497 except (OSError, IOError):
2498 self.report_error('Cannot write description file ' + descfn)
2499 return
2500
2501 if self.params.get('writeannotations', False):
2502 annofn = self.prepare_filename(info_dict, 'annotation')
2503 if not self._ensure_dir_exists(encodeFilename(annofn)):
2504 return
2505 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(annofn)):
2506 self.to_screen('[info] Video annotations are already present')
2507 elif not info_dict.get('annotations'):
2508 self.report_warning('There are no annotations to write.')
2509 else:
2510 try:
2511 self.to_screen('[info] Writing video annotations to: ' + annofn)
2512 with io.open(encodeFilename(annofn), 'w', encoding='utf-8') as annofile:
2513 annofile.write(info_dict['annotations'])
2514 except (KeyError, TypeError):
2515 self.report_warning('There are no annotations to write.')
2516 except (OSError, IOError):
2517 self.report_error('Cannot write annotations file: ' + annofn)
2518 return
2519
2520 subtitles_are_requested = any([self.params.get('writesubtitles', False),
2521 self.params.get('writeautomaticsub')])
2522
2523 if subtitles_are_requested and info_dict.get('requested_subtitles'):
2524 # subtitles download errors are already managed as troubles in relevant IE
2525 # that way it will silently go on when used with unsupporting IE
2526 subtitles = info_dict['requested_subtitles']
2527 # ie = self.get_info_extractor(info_dict['extractor_key'])
2528 for sub_lang, sub_info in subtitles.items():
2529 sub_format = sub_info['ext']
2530 sub_filename = subtitles_filename(temp_filename, sub_lang, sub_format, info_dict.get('ext'))
2531 sub_filename_final = subtitles_filename(
2532 self.prepare_filename(info_dict, 'subtitle'), sub_lang, sub_format, info_dict.get('ext'))
2533 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(sub_filename)):
2534 self.to_screen('[info] Video subtitle %s.%s is already present' % (sub_lang, sub_format))
2535 sub_info['filepath'] = sub_filename
2536 files_to_move[sub_filename] = sub_filename_final
2537 else:
2538 self.to_screen('[info] Writing video subtitles to: ' + sub_filename)
2539 if sub_info.get('data') is not None:
2540 try:
2541 # Use newline='' to prevent conversion of newline characters
2542 # See https://github.com/ytdl-org/youtube-dl/issues/10268
2543 with io.open(encodeFilename(sub_filename), 'w', encoding='utf-8', newline='') as subfile:
2544 subfile.write(sub_info['data'])
2545 sub_info['filepath'] = sub_filename
2546 files_to_move[sub_filename] = sub_filename_final
2547 except (OSError, IOError):
2548 self.report_error('Cannot write subtitles file ' + sub_filename)
2549 return
2550 else:
2551 try:
2552 self.dl(sub_filename, sub_info.copy(), subtitle=True)
2553 sub_info['filepath'] = sub_filename
2554 files_to_move[sub_filename] = sub_filename_final
2555 except (ExtractorError, IOError, OSError, ValueError) + network_exceptions as err:
2556 self.report_warning('Unable to download subtitle for "%s": %s' %
2557 (sub_lang, error_to_compat_str(err)))
2558 continue
2559
2560 if self.params.get('writeinfojson', False):
2561 infofn = self.prepare_filename(info_dict, 'infojson')
2562 if not self._ensure_dir_exists(encodeFilename(infofn)):
2563 return
2564 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(infofn)):
2565 self.to_screen('[info] Video metadata is already present')
2566 else:
2567 self.to_screen('[info] Writing video metadata as JSON to: ' + infofn)
2568 try:
2569 write_json_file(self.sanitize_info(info_dict, self.params.get('clean_infojson', True)), infofn)
2570 except (OSError, IOError):
2571 self.report_error('Cannot write video metadata to JSON file ' + infofn)
2572 return
2573 info_dict['__infojson_filename'] = infofn
2574
2575 for thumb_ext in self._write_thumbnails(info_dict, temp_filename):
2576 thumb_filename_temp = replace_extension(temp_filename, thumb_ext, info_dict.get('ext'))
2577 thumb_filename = replace_extension(
2578 self.prepare_filename(info_dict, 'thumbnail'), thumb_ext, info_dict.get('ext'))
2579 files_to_move[thumb_filename_temp] = thumb_filename
2580
2581 # Write internet shortcut files
2582 url_link = webloc_link = desktop_link = False
2583 if self.params.get('writelink', False):
2584 if sys.platform == "darwin": # macOS.
2585 webloc_link = True
2586 elif sys.platform.startswith("linux"):
2587 desktop_link = True
2588 else: # if sys.platform in ['win32', 'cygwin']:
2589 url_link = True
2590 if self.params.get('writeurllink', False):
2591 url_link = True
2592 if self.params.get('writewebloclink', False):
2593 webloc_link = True
2594 if self.params.get('writedesktoplink', False):
2595 desktop_link = True
2596
2597 if url_link or webloc_link or desktop_link:
2598 if 'webpage_url' not in info_dict:
2599 self.report_error('Cannot write internet shortcut file because the "webpage_url" field is missing in the media information')
2600 return
2601 ascii_url = iri_to_uri(info_dict['webpage_url'])
2602
2603 def _write_link_file(extension, template, newline, embed_filename):
2604 linkfn = replace_extension(full_filename, extension, info_dict.get('ext'))
2605 if self.params.get('overwrites', True) and os.path.exists(encodeFilename(linkfn)):
2606 self.to_screen('[info] Internet shortcut is already present')
2607 else:
2608 try:
2609 self.to_screen('[info] Writing internet shortcut to: ' + linkfn)
2610 with io.open(encodeFilename(to_high_limit_path(linkfn)), 'w', encoding='utf-8', newline=newline) as linkfile:
2611 template_vars = {'url': ascii_url}
2612 if embed_filename:
2613 template_vars['filename'] = linkfn[:-(len(extension) + 1)]
2614 linkfile.write(template % template_vars)
2615 except (OSError, IOError):
2616 self.report_error('Cannot write internet shortcut ' + linkfn)
2617 return False
2618 return True
2619
2620 if url_link:
2621 if not _write_link_file('url', DOT_URL_LINK_TEMPLATE, '\r\n', embed_filename=False):
2622 return
2623 if webloc_link:
2624 if not _write_link_file('webloc', DOT_WEBLOC_LINK_TEMPLATE, '\n', embed_filename=False):
2625 return
2626 if desktop_link:
2627 if not _write_link_file('desktop', DOT_DESKTOP_LINK_TEMPLATE, '\n', embed_filename=True):
2628 return
2629
2630 try:
2631 info_dict, files_to_move = self.pre_process(info_dict, 'before_dl', files_to_move)
2632 except PostProcessingError as err:
2633 self.report_error('Preprocessing: %s' % str(err))
2634 return
2635
2636 must_record_download_archive = False
2637 if self.params.get('skip_download', False):
2638 info_dict['filepath'] = temp_filename
2639 info_dict['__finaldir'] = os.path.dirname(os.path.abspath(encodeFilename(full_filename)))
2640 info_dict['__files_to_move'] = files_to_move
2641 info_dict = self.run_pp(MoveFilesAfterDownloadPP(self, False), info_dict)
2642 else:
2643 # Download
2644 info_dict.setdefault('__postprocessors', [])
2645 try:
2646
2647 def existing_file(*filepaths):
2648 ext = info_dict.get('ext')
2649 final_ext = self.params.get('final_ext', ext)
2650 existing_files = []
2651 for file in orderedSet(filepaths):
2652 if final_ext != ext:
2653 converted = replace_extension(file, final_ext, ext)
2654 if os.path.exists(encodeFilename(converted)):
2655 existing_files.append(converted)
2656 if os.path.exists(encodeFilename(file)):
2657 existing_files.append(file)
2658
2659 if not existing_files or self.params.get('overwrites', False):
2660 for file in orderedSet(existing_files):
2661 self.report_file_delete(file)
2662 os.remove(encodeFilename(file))
2663 return None
2664
2665 self.report_file_already_downloaded(existing_files[0])
2666 info_dict['ext'] = os.path.splitext(existing_files[0])[1][1:]
2667 return existing_files[0]
2668
2669 success = True
2670 if info_dict.get('requested_formats') is not None:
2671
2672 def compatible_formats(formats):
2673 # TODO: some formats actually allow this (mkv, webm, ogg, mp4), but not all of them.
2674 video_formats = [format for format in formats if format.get('vcodec') != 'none']
2675 audio_formats = [format for format in formats if format.get('acodec') != 'none']
2676 if len(video_formats) > 2 or len(audio_formats) > 2:
2677 return False
2678
2679 # Check extension
2680 exts = set(format.get('ext') for format in formats)
2681 COMPATIBLE_EXTS = (
2682 set(('mp3', 'mp4', 'm4a', 'm4p', 'm4b', 'm4r', 'm4v', 'ismv', 'isma')),
2683 set(('webm',)),
2684 )
2685 for ext_sets in COMPATIBLE_EXTS:
2686 if ext_sets.issuperset(exts):
2687 return True
2688 # TODO: Check acodec/vcodec
2689 return False
2690
2691 requested_formats = info_dict['requested_formats']
2692 old_ext = info_dict['ext']
2693 if self.params.get('merge_output_format') is None and not compatible_formats(requested_formats):
2694 info_dict['ext'] = 'mkv'
2695 self.report_warning(
2696 'Requested formats are incompatible for merge and will be merged into mkv.')
2697 new_ext = info_dict['ext']
2698
2699 def correct_ext(filename, ext=new_ext):
2700 if filename == '-':
2701 return filename
2702 filename_real_ext = os.path.splitext(filename)[1][1:]
2703 filename_wo_ext = (
2704 os.path.splitext(filename)[0]
2705 if filename_real_ext in (old_ext, new_ext)
2706 else filename)
2707 return '%s.%s' % (filename_wo_ext, ext)
2708
2709 # Ensure filename always has a correct extension for successful merge
2710 full_filename = correct_ext(full_filename)
2711 temp_filename = correct_ext(temp_filename)
2712 dl_filename = existing_file(full_filename, temp_filename)
2713 info_dict['__real_download'] = False
2714
2715 _protocols = set(determine_protocol(f) for f in requested_formats)
2716 if len(_protocols) == 1: # All requested formats have same protocol
2717 info_dict['protocol'] = _protocols.pop()
2718 directly_mergable = FFmpegFD.can_merge_formats(info_dict)
2719 if dl_filename is not None:
2720 pass
2721 elif (directly_mergable and get_suitable_downloader(
2722 info_dict, self.params, to_stdout=(temp_filename == '-')) == FFmpegFD):
2723 info_dict['url'] = '\n'.join(f['url'] for f in requested_formats)
2724 success, real_download = self.dl(temp_filename, info_dict)
2725 info_dict['__real_download'] = real_download
2726 else:
2727 downloaded = []
2728 merger = FFmpegMergerPP(self)
2729 if self.params.get('allow_unplayable_formats'):
2730 self.report_warning(
2731 'You have requested merging of multiple formats '
2732 'while also allowing unplayable formats to be downloaded. '
2733 'The formats won\'t be merged to prevent data corruption.')
2734 elif not merger.available:
2735 self.report_warning(
2736 'You have requested merging of multiple formats but ffmpeg is not installed. '
2737 'The formats won\'t be merged.')
2738
2739 if temp_filename == '-':
2740 reason = ('using a downloader other than ffmpeg' if directly_mergable
2741 else 'but the formats are incompatible for simultaneous download' if merger.available
2742 else 'but ffmpeg is not installed')
2743 self.report_warning(
2744 f'You have requested downloading multiple formats to stdout {reason}. '
2745 'The formats will be streamed one after the other')
2746 fname = temp_filename
2747 for f in requested_formats:
2748 new_info = dict(info_dict)
2749 del new_info['requested_formats']
2750 new_info.update(f)
2751 if temp_filename != '-':
2752 fname = prepend_extension(
2753 correct_ext(temp_filename, new_info['ext']),
2754 'f%s' % f['format_id'], new_info['ext'])
2755 if not self._ensure_dir_exists(fname):
2756 return
2757 downloaded.append(fname)
2758 partial_success, real_download = self.dl(fname, new_info)
2759 info_dict['__real_download'] = info_dict['__real_download'] or real_download
2760 success = success and partial_success
2761 if merger.available and not self.params.get('allow_unplayable_formats'):
2762 info_dict['__postprocessors'].append(merger)
2763 info_dict['__files_to_merge'] = downloaded
2764 # Even if there were no downloads, it is being merged only now
2765 info_dict['__real_download'] = True
2766 else:
2767 for file in downloaded:
2768 files_to_move[file] = None
2769 else:
2770 # Just a single file
2771 dl_filename = existing_file(full_filename, temp_filename)
2772 if dl_filename is None:
2773 success, real_download = self.dl(temp_filename, info_dict)
2774 info_dict['__real_download'] = real_download
2775
2776 dl_filename = dl_filename or temp_filename
2777 info_dict['__finaldir'] = os.path.dirname(os.path.abspath(encodeFilename(full_filename)))
2778
2779 except network_exceptions as err:
2780 self.report_error('unable to download video data: %s' % error_to_compat_str(err))
2781 return
2782 except (OSError, IOError) as err:
2783 raise UnavailableVideoError(err)
2784 except (ContentTooShortError, ) as err:
2785 self.report_error('content too short (expected %s bytes and served %s)' % (err.expected, err.downloaded))
2786 return
2787
2788 if success and full_filename != '-':
2789
2790 def fixup():
2791 do_fixup = True
2792 fixup_policy = self.params.get('fixup')
2793 vid = info_dict['id']
2794
2795 if fixup_policy in ('ignore', 'never'):
2796 return
2797 elif fixup_policy == 'warn':
2798 do_fixup = False
2799 elif fixup_policy != 'force':
2800 assert fixup_policy in ('detect_or_warn', None)
2801 if not info_dict.get('__real_download'):
2802 do_fixup = False
2803
2804 def ffmpeg_fixup(cndn, msg, cls):
2805 if not cndn:
2806 return
2807 if not do_fixup:
2808 self.report_warning(f'{vid}: {msg}')
2809 return
2810 pp = cls(self)
2811 if pp.available:
2812 info_dict['__postprocessors'].append(pp)
2813 else:
2814 self.report_warning(f'{vid}: {msg}. Install ffmpeg to fix this automatically')
2815
2816 stretched_ratio = info_dict.get('stretched_ratio')
2817 ffmpeg_fixup(
2818 stretched_ratio not in (1, None),
2819 f'Non-uniform pixel ratio {stretched_ratio}',
2820 FFmpegFixupStretchedPP)
2821
2822 ffmpeg_fixup(
2823 (info_dict.get('requested_formats') is None
2824 and info_dict.get('container') == 'm4a_dash'
2825 and info_dict.get('ext') == 'm4a'),
2826 'writing DASH m4a. Only some players support this container',
2827 FFmpegFixupM4aPP)
2828
2829 downloader = (get_suitable_downloader(info_dict, self.params).__name__
2830 if 'protocol' in info_dict else None)
2831 ffmpeg_fixup(downloader == 'HlsFD', 'malformed AAC bitstream detected', FFmpegFixupM3u8PP)
2832 ffmpeg_fixup(downloader == 'WebSocketFragmentFD', 'malformed timestamps detected', FFmpegFixupTimestampPP)
2833 ffmpeg_fixup(downloader == 'WebSocketFragmentFD', 'malformed duration detected', FFmpegFixupDurationPP)
2834
2835 fixup()
2836 try:
2837 info_dict = self.post_process(dl_filename, info_dict, files_to_move)
2838 except PostProcessingError as err:
2839 self.report_error('Postprocessing: %s' % str(err))
2840 return
2841 try:
2842 for ph in self._post_hooks:
2843 ph(info_dict['filepath'])
2844 except Exception as err:
2845 self.report_error('post hooks: %s' % str(err))
2846 return
2847 must_record_download_archive = True
2848
2849 if must_record_download_archive or self.params.get('force_write_download_archive', False):
2850 self.record_download_archive(info_dict)
2851 max_downloads = self.params.get('max_downloads')
2852 if max_downloads is not None and self._num_downloads >= int(max_downloads):
2853 raise MaxDownloadsReached()
2854
2855 def download(self, url_list):
2856 """Download a given list of URLs."""
2857 outtmpl = self.outtmpl_dict['default']
2858 if (len(url_list) > 1
2859 and outtmpl != '-'
2860 and '%' not in outtmpl
2861 and self.params.get('max_downloads') != 1):
2862 raise SameFileError(outtmpl)
2863
2864 for url in url_list:
2865 try:
2866 # It also downloads the videos
2867 res = self.extract_info(
2868 url, force_generic_extractor=self.params.get('force_generic_extractor', False))
2869 except UnavailableVideoError:
2870 self.report_error('unable to download video')
2871 except MaxDownloadsReached:
2872 self.to_screen('[info] Maximum number of downloaded files reached')
2873 raise
2874 except ExistingVideoReached:
2875 self.to_screen('[info] Encountered a file that is already in the archive, stopping due to --break-on-existing')
2876 raise
2877 except RejectedVideoReached:
2878 self.to_screen('[info] Encountered a file that did not match filter, stopping due to --break-on-reject')
2879 raise
2880 else:
2881 if self.params.get('dump_single_json', False):
2882 self.post_extract(res)
2883 self.to_stdout(json.dumps(self.sanitize_info(res)))
2884
2885 return self._download_retcode
2886
2887 def download_with_info_file(self, info_filename):
2888 with contextlib.closing(fileinput.FileInput(
2889 [info_filename], mode='r',
2890 openhook=fileinput.hook_encoded('utf-8'))) as f:
2891 # FileInput doesn't have a read method, we can't call json.load
2892 info = self.sanitize_info(json.loads('\n'.join(f)), self.params.get('clean_infojson', True))
2893 try:
2894 self.process_ie_result(info, download=True)
2895 except (DownloadError, EntryNotInPlaylist, ThrottledDownload):
2896 webpage_url = info.get('webpage_url')
2897 if webpage_url is not None:
2898 self.report_warning('The info failed to download, trying with "%s"' % webpage_url)
2899 return self.download([webpage_url])
2900 else:
2901 raise
2902 return self._download_retcode
2903
2904 @staticmethod
2905 def sanitize_info(info_dict, remove_private_keys=False):
2906 ''' Sanitize the infodict for converting to json '''
2907 info_dict.setdefault('epoch', int(time.time()))
2908 remove_keys = {'__original_infodict'} # Always remove this since this may contain a copy of the entire dict
2909 keep_keys = ['_type'], # Always keep this to facilitate load-info-json
2910 if remove_private_keys:
2911 remove_keys |= {
2912 'requested_formats', 'requested_subtitles', 'requested_entries',
2913 'filepath', 'entries', 'original_url', 'playlist_autonumber',
2914 }
2915 empty_values = (None, {}, [], set(), tuple())
2916 reject = lambda k, v: k not in keep_keys and (
2917 k.startswith('_') or k in remove_keys or v in empty_values)
2918 else:
2919 reject = lambda k, v: k in remove_keys
2920 filter_fn = lambda obj: (
2921 list(map(filter_fn, obj)) if isinstance(obj, (LazyList, list, tuple, set))
2922 else obj if not isinstance(obj, dict)
2923 else dict((k, filter_fn(v)) for k, v in obj.items() if not reject(k, v)))
2924 return filter_fn(info_dict)
2925
2926 @staticmethod
2927 def filter_requested_info(info_dict, actually_filter=True):
2928 ''' Alias of sanitize_info for backward compatibility '''
2929 return YoutubeDL.sanitize_info(info_dict, actually_filter)
2930
2931 def run_pp(self, pp, infodict):
2932 files_to_delete = []
2933 if '__files_to_move' not in infodict:
2934 infodict['__files_to_move'] = {}
2935 files_to_delete, infodict = pp.run(infodict)
2936 if not files_to_delete:
2937 return infodict
2938
2939 if self.params.get('keepvideo', False):
2940 for f in files_to_delete:
2941 infodict['__files_to_move'].setdefault(f, '')
2942 else:
2943 for old_filename in set(files_to_delete):
2944 self.to_screen('Deleting original file %s (pass -k to keep)' % old_filename)
2945 try:
2946 os.remove(encodeFilename(old_filename))
2947 except (IOError, OSError):
2948 self.report_warning('Unable to remove downloaded original file')
2949 if old_filename in infodict['__files_to_move']:
2950 del infodict['__files_to_move'][old_filename]
2951 return infodict
2952
2953 @staticmethod
2954 def post_extract(info_dict):
2955 def actual_post_extract(info_dict):
2956 if info_dict.get('_type') in ('playlist', 'multi_video'):
2957 for video_dict in info_dict.get('entries', {}):
2958 actual_post_extract(video_dict or {})
2959 return
2960
2961 post_extractor = info_dict.get('__post_extractor') or (lambda: {})
2962 extra = post_extractor().items()
2963 info_dict.update(extra)
2964 info_dict.pop('__post_extractor', None)
2965
2966 original_infodict = info_dict.get('__original_infodict') or {}
2967 original_infodict.update(extra)
2968 original_infodict.pop('__post_extractor', None)
2969
2970 actual_post_extract(info_dict or {})
2971
2972 def pre_process(self, ie_info, key='pre_process', files_to_move=None):
2973 info = dict(ie_info)
2974 info['__files_to_move'] = files_to_move or {}
2975 for pp in self._pps[key]:
2976 info = self.run_pp(pp, info)
2977 return info, info.pop('__files_to_move', None)
2978
2979 def post_process(self, filename, ie_info, files_to_move=None):
2980 """Run all the postprocessors on the given file."""
2981 info = dict(ie_info)
2982 info['filepath'] = filename
2983 info['__files_to_move'] = files_to_move or {}
2984
2985 for pp in ie_info.get('__postprocessors', []) + self._pps['post_process']:
2986 info = self.run_pp(pp, info)
2987 info = self.run_pp(MoveFilesAfterDownloadPP(self), info)
2988 del info['__files_to_move']
2989 for pp in self._pps['after_move']:
2990 info = self.run_pp(pp, info)
2991 return info
2992
2993 def _make_archive_id(self, info_dict):
2994 video_id = info_dict.get('id')
2995 if not video_id:
2996 return
2997 # Future-proof against any change in case
2998 # and backwards compatibility with prior versions
2999 extractor = info_dict.get('extractor_key') or info_dict.get('ie_key') # key in a playlist
3000 if extractor is None:
3001 url = str_or_none(info_dict.get('url'))
3002 if not url:
3003 return
3004 # Try to find matching extractor for the URL and take its ie_key
3005 for ie in self._ies:
3006 if ie.suitable(url):
3007 extractor = ie.ie_key()
3008 break
3009 else:
3010 return
3011 return '%s %s' % (extractor.lower(), video_id)
3012
3013 def in_download_archive(self, info_dict):
3014 fn = self.params.get('download_archive')
3015 if fn is None:
3016 return False
3017
3018 vid_id = self._make_archive_id(info_dict)
3019 if not vid_id:
3020 return False # Incomplete video information
3021
3022 return vid_id in self.archive
3023
3024 def record_download_archive(self, info_dict):
3025 fn = self.params.get('download_archive')
3026 if fn is None:
3027 return
3028 vid_id = self._make_archive_id(info_dict)
3029 assert vid_id
3030 with locked_file(fn, 'a', encoding='utf-8') as archive_file:
3031 archive_file.write(vid_id + '\n')
3032 self.archive.add(vid_id)
3033
3034 @staticmethod
3035 def format_resolution(format, default='unknown'):
3036 if format.get('vcodec') == 'none':
3037 if format.get('acodec') == 'none':
3038 return 'images'
3039 return 'audio only'
3040 if format.get('resolution') is not None:
3041 return format['resolution']
3042 if format.get('width') and format.get('height'):
3043 res = '%dx%d' % (format['width'], format['height'])
3044 elif format.get('height'):
3045 res = '%sp' % format['height']
3046 elif format.get('width'):
3047 res = '%dx?' % format['width']
3048 else:
3049 res = default
3050 return res
3051
3052 def _format_note(self, fdict):
3053 res = ''
3054 if fdict.get('ext') in ['f4f', 'f4m']:
3055 res += '(unsupported) '
3056 if fdict.get('language'):
3057 if res:
3058 res += ' '
3059 res += '[%s] ' % fdict['language']
3060 if fdict.get('format_note') is not None:
3061 res += fdict['format_note'] + ' '
3062 if fdict.get('tbr') is not None:
3063 res += '%4dk ' % fdict['tbr']
3064 if fdict.get('container') is not None:
3065 if res:
3066 res += ', '
3067 res += '%s container' % fdict['container']
3068 if (fdict.get('vcodec') is not None
3069 and fdict.get('vcodec') != 'none'):
3070 if res:
3071 res += ', '
3072 res += fdict['vcodec']
3073 if fdict.get('vbr') is not None:
3074 res += '@'
3075 elif fdict.get('vbr') is not None and fdict.get('abr') is not None:
3076 res += 'video@'
3077 if fdict.get('vbr') is not None:
3078 res += '%4dk' % fdict['vbr']
3079 if fdict.get('fps') is not None:
3080 if res:
3081 res += ', '
3082 res += '%sfps' % fdict['fps']
3083 if fdict.get('acodec') is not None:
3084 if res:
3085 res += ', '
3086 if fdict['acodec'] == 'none':
3087 res += 'video only'
3088 else:
3089 res += '%-5s' % fdict['acodec']
3090 elif fdict.get('abr') is not None:
3091 if res:
3092 res += ', '
3093 res += 'audio'
3094 if fdict.get('abr') is not None:
3095 res += '@%3dk' % fdict['abr']
3096 if fdict.get('asr') is not None:
3097 res += ' (%5dHz)' % fdict['asr']
3098 if fdict.get('filesize') is not None:
3099 if res:
3100 res += ', '
3101 res += format_bytes(fdict['filesize'])
3102 elif fdict.get('filesize_approx') is not None:
3103 if res:
3104 res += ', '
3105 res += '~' + format_bytes(fdict['filesize_approx'])
3106 return res
3107
3108 def list_formats(self, info_dict):
3109 formats = info_dict.get('formats', [info_dict])
3110 new_format = (
3111 'list-formats' not in self.params.get('compat_opts', [])
3112 and self.params.get('listformats_table', True) is not False)
3113 if new_format:
3114 table = [
3115 [
3116 format_field(f, 'format_id'),
3117 format_field(f, 'ext'),
3118 self.format_resolution(f),
3119 format_field(f, 'fps', '%d'),
3120 '|',
3121 format_field(f, 'filesize', ' %s', func=format_bytes) + format_field(f, 'filesize_approx', '~%s', func=format_bytes),
3122 format_field(f, 'tbr', '%4dk'),
3123 shorten_protocol_name(f.get('protocol', '').replace("native", "n")),
3124 '|',
3125 format_field(f, 'vcodec', default='unknown').replace('none', ''),
3126 format_field(f, 'vbr', '%4dk'),
3127 format_field(f, 'acodec', default='unknown').replace('none', ''),
3128 format_field(f, 'abr', '%3dk'),
3129 format_field(f, 'asr', '%5dHz'),
3130 ', '.join(filter(None, (
3131 'UNSUPPORTED' if f.get('ext') in ('f4f', 'f4m') else '',
3132 format_field(f, 'language', '[%s]'),
3133 format_field(f, 'format_note'),
3134 format_field(f, 'container', ignore=(None, f.get('ext'))),
3135 ))),
3136 ] for f in formats if f.get('preference') is None or f['preference'] >= -1000]
3137 header_line = ['ID', 'EXT', 'RESOLUTION', 'FPS', '|', ' FILESIZE', ' TBR', 'PROTO',
3138 '|', 'VCODEC', ' VBR', 'ACODEC', ' ABR', ' ASR', 'MORE INFO']
3139 else:
3140 table = [
3141 [
3142 format_field(f, 'format_id'),
3143 format_field(f, 'ext'),
3144 self.format_resolution(f),
3145 self._format_note(f)]
3146 for f in formats
3147 if f.get('preference') is None or f['preference'] >= -1000]
3148 header_line = ['format code', 'extension', 'resolution', 'note']
3149
3150 self.to_screen(
3151 '[info] Available formats for %s:' % info_dict['id'])
3152 self.to_stdout(render_table(
3153 header_line, table, delim=new_format, extraGap=(0 if new_format else 1), hideEmpty=new_format))
3154
3155 def list_thumbnails(self, info_dict):
3156 thumbnails = list(info_dict.get('thumbnails'))
3157 if not thumbnails:
3158 self.to_screen('[info] No thumbnails present for %s' % info_dict['id'])
3159 return
3160
3161 self.to_screen(
3162 '[info] Thumbnails for %s:' % info_dict['id'])
3163 self.to_stdout(render_table(
3164 ['ID', 'width', 'height', 'URL'],
3165 [[t['id'], t.get('width', 'unknown'), t.get('height', 'unknown'), t['url']] for t in thumbnails]))
3166
3167 def list_subtitles(self, video_id, subtitles, name='subtitles'):
3168 if not subtitles:
3169 self.to_screen('%s has no %s' % (video_id, name))
3170 return
3171 self.to_screen(
3172 'Available %s for %s:' % (name, video_id))
3173
3174 def _row(lang, formats):
3175 exts, names = zip(*((f['ext'], f.get('name') or 'unknown') for f in reversed(formats)))
3176 if len(set(names)) == 1:
3177 names = [] if names[0] == 'unknown' else names[:1]
3178 return [lang, ', '.join(names), ', '.join(exts)]
3179
3180 self.to_stdout(render_table(
3181 ['Language', 'Name', 'Formats'],
3182 [_row(lang, formats) for lang, formats in subtitles.items()],
3183 hideEmpty=True))
3184
3185 def urlopen(self, req):
3186 """ Start an HTTP download """
3187 if isinstance(req, compat_basestring):
3188 req = sanitized_Request(req)
3189 return self._opener.open(req, timeout=self._socket_timeout)
3190
3191 def print_debug_header(self):
3192 if not self.params.get('verbose'):
3193 return
3194
3195 if type('') is not compat_str:
3196 # Python 2.6 on SLES11 SP1 (https://github.com/ytdl-org/youtube-dl/issues/3326)
3197 self.report_warning(
3198 'Your Python is broken! Update to a newer and supported version')
3199
3200 stdout_encoding = getattr(
3201 sys.stdout, 'encoding', 'missing (%s)' % type(sys.stdout).__name__)
3202 encoding_str = (
3203 '[debug] Encodings: locale %s, fs %s, out %s, pref %s\n' % (
3204 locale.getpreferredencoding(),
3205 sys.getfilesystemencoding(),
3206 stdout_encoding,
3207 self.get_encoding()))
3208 write_string(encoding_str, encoding=None)
3209
3210 source = (
3211 '(exe)' if hasattr(sys, 'frozen')
3212 else '(zip)' if isinstance(globals().get('__loader__'), zipimporter)
3213 else '(source)' if os.path.basename(sys.argv[0]) == '__main__.py'
3214 else '')
3215 self._write_string('[debug] yt-dlp version %s %s\n' % (__version__, source))
3216 if _LAZY_LOADER:
3217 self._write_string('[debug] Lazy loading extractors enabled\n')
3218 if _PLUGIN_CLASSES:
3219 self._write_string(
3220 '[debug] Plugin Extractors: %s\n' % [ie.ie_key() for ie in _PLUGIN_CLASSES])
3221 if self.params.get('compat_opts'):
3222 self._write_string(
3223 '[debug] Compatibility options: %s\n' % ', '.join(self.params.get('compat_opts')))
3224 try:
3225 sp = subprocess.Popen(
3226 ['git', 'rev-parse', '--short', 'HEAD'],
3227 stdout=subprocess.PIPE, stderr=subprocess.PIPE,
3228 cwd=os.path.dirname(os.path.abspath(__file__)))
3229 out, err = process_communicate_or_kill(sp)
3230 out = out.decode().strip()
3231 if re.match('[0-9a-f]+', out):
3232 self._write_string('[debug] Git HEAD: %s\n' % out)
3233 except Exception:
3234 try:
3235 sys.exc_clear()
3236 except Exception:
3237 pass
3238
3239 def python_implementation():
3240 impl_name = platform.python_implementation()
3241 if impl_name == 'PyPy' and hasattr(sys, 'pypy_version_info'):
3242 return impl_name + ' version %d.%d.%d' % sys.pypy_version_info[:3]
3243 return impl_name
3244
3245 self._write_string('[debug] Python version %s (%s %s) - %s\n' % (
3246 platform.python_version(),
3247 python_implementation(),
3248 platform.architecture()[0],
3249 platform_name()))
3250
3251 exe_versions = FFmpegPostProcessor.get_versions(self)
3252 exe_versions['rtmpdump'] = rtmpdump_version()
3253 exe_versions['phantomjs'] = PhantomJSwrapper._version()
3254 exe_str = ', '.join(
3255 '%s %s' % (exe, v)
3256 for exe, v in sorted(exe_versions.items())
3257 if v
3258 )
3259 if not exe_str:
3260 exe_str = 'none'
3261 self._write_string('[debug] exe versions: %s\n' % exe_str)
3262
3263 proxy_map = {}
3264 for handler in self._opener.handlers:
3265 if hasattr(handler, 'proxies'):
3266 proxy_map.update(handler.proxies)
3267 self._write_string('[debug] Proxy map: ' + compat_str(proxy_map) + '\n')
3268
3269 if self.params.get('call_home', False):
3270 ipaddr = self.urlopen('https://yt-dl.org/ip').read().decode('utf-8')
3271 self._write_string('[debug] Public IP address: %s\n' % ipaddr)
3272 return
3273 latest_version = self.urlopen(
3274 'https://yt-dl.org/latest/version').read().decode('utf-8')
3275 if version_tuple(latest_version) > version_tuple(__version__):
3276 self.report_warning(
3277 'You are using an outdated version (newest version: %s)! '
3278 'See https://yt-dl.org/update if you need help updating.' %
3279 latest_version)
3280
3281 def _setup_opener(self):
3282 timeout_val = self.params.get('socket_timeout')
3283 self._socket_timeout = 600 if timeout_val is None else float(timeout_val)
3284
3285 opts_cookiesfrombrowser = self.params.get('cookiesfrombrowser')
3286 opts_cookiefile = self.params.get('cookiefile')
3287 opts_proxy = self.params.get('proxy')
3288
3289 self.cookiejar = load_cookies(opts_cookiefile, opts_cookiesfrombrowser, self)
3290
3291 cookie_processor = YoutubeDLCookieProcessor(self.cookiejar)
3292 if opts_proxy is not None:
3293 if opts_proxy == '':
3294 proxies = {}
3295 else:
3296 proxies = {'http': opts_proxy, 'https': opts_proxy}
3297 else:
3298 proxies = compat_urllib_request.getproxies()
3299 # Set HTTPS proxy to HTTP one if given (https://github.com/ytdl-org/youtube-dl/issues/805)
3300 if 'http' in proxies and 'https' not in proxies:
3301 proxies['https'] = proxies['http']
3302 proxy_handler = PerRequestProxyHandler(proxies)
3303
3304 debuglevel = 1 if self.params.get('debug_printtraffic') else 0
3305 https_handler = make_HTTPS_handler(self.params, debuglevel=debuglevel)
3306 ydlh = YoutubeDLHandler(self.params, debuglevel=debuglevel)
3307 redirect_handler = YoutubeDLRedirectHandler()
3308 data_handler = compat_urllib_request_DataHandler()
3309
3310 # When passing our own FileHandler instance, build_opener won't add the
3311 # default FileHandler and allows us to disable the file protocol, which
3312 # can be used for malicious purposes (see
3313 # https://github.com/ytdl-org/youtube-dl/issues/8227)
3314 file_handler = compat_urllib_request.FileHandler()
3315
3316 def file_open(*args, **kwargs):
3317 raise compat_urllib_error.URLError('file:// scheme is explicitly disabled in yt-dlp for security reasons')
3318 file_handler.file_open = file_open
3319
3320 opener = compat_urllib_request.build_opener(
3321 proxy_handler, https_handler, cookie_processor, ydlh, redirect_handler, data_handler, file_handler)
3322
3323 # Delete the default user-agent header, which would otherwise apply in
3324 # cases where our custom HTTP handler doesn't come into play
3325 # (See https://github.com/ytdl-org/youtube-dl/issues/1309 for details)
3326 opener.addheaders = []
3327 self._opener = opener
3328
3329 def encode(self, s):
3330 if isinstance(s, bytes):
3331 return s # Already encoded
3332
3333 try:
3334 return s.encode(self.get_encoding())
3335 except UnicodeEncodeError as err:
3336 err.reason = err.reason + '. Check your system encoding configuration or use the --encoding option.'
3337 raise
3338
3339 def get_encoding(self):
3340 encoding = self.params.get('encoding')
3341 if encoding is None:
3342 encoding = preferredencoding()
3343 return encoding
3344
3345 def _write_thumbnails(self, info_dict, filename): # return the extensions
3346 write_all = self.params.get('write_all_thumbnails', False)
3347 thumbnails = []
3348 if write_all or self.params.get('writethumbnail', False):
3349 thumbnails = info_dict.get('thumbnails') or []
3350 multiple = write_all and len(thumbnails) > 1
3351
3352 ret = []
3353 for t in thumbnails[::-1]:
3354 thumb_ext = determine_ext(t['url'], 'jpg')
3355 suffix = '%s.' % t['id'] if multiple else ''
3356 thumb_display_id = '%s ' % t['id'] if multiple else ''
3357 thumb_filename = replace_extension(filename, suffix + thumb_ext, info_dict.get('ext'))
3358
3359 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(thumb_filename)):
3360 ret.append(suffix + thumb_ext)
3361 t['filepath'] = thumb_filename
3362 self.to_screen('[%s] %s: Thumbnail %sis already present' %
3363 (info_dict['extractor'], info_dict['id'], thumb_display_id))
3364 else:
3365 self.to_screen('[%s] %s: Downloading thumbnail %s ...' %
3366 (info_dict['extractor'], info_dict['id'], thumb_display_id))
3367 try:
3368 uf = self.urlopen(t['url'])
3369 with open(encodeFilename(thumb_filename), 'wb') as thumbf:
3370 shutil.copyfileobj(uf, thumbf)
3371 ret.append(suffix + thumb_ext)
3372 self.to_screen('[%s] %s: Writing thumbnail %sto: %s' %
3373 (info_dict['extractor'], info_dict['id'], thumb_display_id, thumb_filename))
3374 t['filepath'] = thumb_filename
3375 except network_exceptions as err:
3376 self.report_warning('Unable to download thumbnail "%s": %s' %
3377 (t['url'], error_to_compat_str(err)))
3378 if ret and not write_all:
3379 break
3380 return ret