]> jfr.im git - yt-dlp.git/blob - yt_dlp/YoutubeDL.py
17aa6348749240847666412539041f5079ba4e08
[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 if self.params.get('allow_unplayable_formats'):
501 self.report_warning(
502 'You have asked for unplayable formats to be listed/downloaded. '
503 'This is a developer option intended for debugging. '
504 'If you experience any issues while using this option, DO NOT open a bug report')
505
506 def check_deprecated(param, option, suggestion):
507 if self.params.get(param) is not None:
508 self.report_warning('%s is deprecated. Use %s instead' % (option, suggestion))
509 return True
510 return False
511
512 if check_deprecated('cn_verification_proxy', '--cn-verification-proxy', '--geo-verification-proxy'):
513 if self.params.get('geo_verification_proxy') is None:
514 self.params['geo_verification_proxy'] = self.params['cn_verification_proxy']
515
516 check_deprecated('autonumber', '--auto-number', '-o "%(autonumber)s-%(title)s.%(ext)s"')
517 check_deprecated('usetitle', '--title', '-o "%(title)s-%(id)s.%(ext)s"')
518 check_deprecated('useid', '--id', '-o "%(id)s.%(ext)s"')
519
520 for msg in self.params.get('warnings', []):
521 self.report_warning(msg)
522
523 if self.params.get('overwrites') is None:
524 self.params.pop('overwrites', None)
525 elif self.params.get('nooverwrites') is not None:
526 # nooverwrites was unnecessarily changed to overwrites
527 # in 0c3d0f51778b153f65c21906031c2e091fcfb641
528 # This ensures compatibility with both keys
529 self.params['overwrites'] = not self.params['nooverwrites']
530 else:
531 self.params['nooverwrites'] = not self.params['overwrites']
532
533 if params.get('bidi_workaround', False):
534 try:
535 import pty
536 master, slave = pty.openpty()
537 width = compat_get_terminal_size().columns
538 if width is None:
539 width_args = []
540 else:
541 width_args = ['-w', str(width)]
542 sp_kwargs = dict(
543 stdin=subprocess.PIPE,
544 stdout=slave,
545 stderr=self._err_file)
546 try:
547 self._output_process = subprocess.Popen(
548 ['bidiv'] + width_args, **sp_kwargs
549 )
550 except OSError:
551 self._output_process = subprocess.Popen(
552 ['fribidi', '-c', 'UTF-8'] + width_args, **sp_kwargs)
553 self._output_channel = os.fdopen(master, 'rb')
554 except OSError as ose:
555 if ose.errno == errno.ENOENT:
556 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.')
557 else:
558 raise
559
560 if (sys.platform != 'win32'
561 and sys.getfilesystemencoding() in ['ascii', 'ANSI_X3.4-1968']
562 and not params.get('restrictfilenames', False)):
563 # Unicode filesystem API will throw errors (#1474, #13027)
564 self.report_warning(
565 'Assuming --restrict-filenames since file system encoding '
566 'cannot encode all characters. '
567 'Set the LC_ALL environment variable to fix this.')
568 self.params['restrictfilenames'] = True
569
570 self.outtmpl_dict = self.parse_outtmpl()
571
572 # Creating format selector here allows us to catch syntax errors before the extraction
573 self.format_selector = (
574 None if self.params.get('format') is None
575 else self.build_format_selector(self.params['format']))
576
577 self._setup_opener()
578
579 """Preload the archive, if any is specified"""
580 def preload_download_archive(fn):
581 if fn is None:
582 return False
583 self.write_debug('Loading archive file %r\n' % fn)
584 try:
585 with locked_file(fn, 'r', encoding='utf-8') as archive_file:
586 for line in archive_file:
587 self.archive.add(line.strip())
588 except IOError as ioe:
589 if ioe.errno != errno.ENOENT:
590 raise
591 return False
592 return True
593
594 self.archive = set()
595 preload_download_archive(self.params.get('download_archive'))
596
597 if auto_init:
598 self.print_debug_header()
599 self.add_default_info_extractors()
600
601 for pp_def_raw in self.params.get('postprocessors', []):
602 pp_def = dict(pp_def_raw)
603 when = pp_def.pop('when', 'post_process')
604 pp_class = get_postprocessor(pp_def.pop('key'))
605 pp = pp_class(self, **compat_kwargs(pp_def))
606 self.add_post_processor(pp, when=when)
607
608 for ph in self.params.get('post_hooks', []):
609 self.add_post_hook(ph)
610
611 for ph in self.params.get('progress_hooks', []):
612 self.add_progress_hook(ph)
613
614 register_socks_protocols()
615
616 def warn_if_short_id(self, argv):
617 # short YouTube ID starting with dash?
618 idxs = [
619 i for i, a in enumerate(argv)
620 if re.match(r'^-[0-9A-Za-z_-]{10}$', a)]
621 if idxs:
622 correct_argv = (
623 ['yt-dlp']
624 + [a for i, a in enumerate(argv) if i not in idxs]
625 + ['--'] + [argv[i] for i in idxs]
626 )
627 self.report_warning(
628 'Long argument string detected. '
629 'Use -- to separate parameters and URLs, like this:\n%s\n' %
630 args_to_str(correct_argv))
631
632 def add_info_extractor(self, ie):
633 """Add an InfoExtractor object to the end of the list."""
634 self._ies.append(ie)
635 if not isinstance(ie, type):
636 self._ies_instances[ie.ie_key()] = ie
637 ie.set_downloader(self)
638
639 def get_info_extractor(self, ie_key):
640 """
641 Get an instance of an IE with name ie_key, it will try to get one from
642 the _ies list, if there's no instance it will create a new one and add
643 it to the extractor list.
644 """
645 ie = self._ies_instances.get(ie_key)
646 if ie is None:
647 ie = get_info_extractor(ie_key)()
648 self.add_info_extractor(ie)
649 return ie
650
651 def add_default_info_extractors(self):
652 """
653 Add the InfoExtractors returned by gen_extractors to the end of the list
654 """
655 for ie in gen_extractor_classes():
656 self.add_info_extractor(ie)
657
658 def add_post_processor(self, pp, when='post_process'):
659 """Add a PostProcessor object to the end of the chain."""
660 self._pps[when].append(pp)
661 pp.set_downloader(self)
662
663 def add_post_hook(self, ph):
664 """Add the post hook"""
665 self._post_hooks.append(ph)
666
667 def add_progress_hook(self, ph):
668 """Add the progress hook (currently only for the file downloader)"""
669 self._progress_hooks.append(ph)
670
671 def _bidi_workaround(self, message):
672 if not hasattr(self, '_output_channel'):
673 return message
674
675 assert hasattr(self, '_output_process')
676 assert isinstance(message, compat_str)
677 line_count = message.count('\n') + 1
678 self._output_process.stdin.write((message + '\n').encode('utf-8'))
679 self._output_process.stdin.flush()
680 res = ''.join(self._output_channel.readline().decode('utf-8')
681 for _ in range(line_count))
682 return res[:-len('\n')]
683
684 def _write_string(self, message, out=None, only_once=False):
685 if only_once:
686 if message in self._printed_messages:
687 return
688 self._printed_messages.add(message)
689 write_string(message, out=out, encoding=self.params.get('encoding'))
690
691 def to_stdout(self, message, skip_eol=False, quiet=False):
692 """Print message to stdout"""
693 if self.params.get('logger'):
694 self.params['logger'].debug(message)
695 elif not quiet or self.params.get('verbose'):
696 self._write_string(
697 '%s%s' % (self._bidi_workaround(message), ('' if skip_eol else '\n')),
698 self._err_file if quiet else self._screen_file)
699
700 def to_stderr(self, message, only_once=False):
701 """Print message to stderr"""
702 assert isinstance(message, compat_str)
703 if self.params.get('logger'):
704 self.params['logger'].error(message)
705 else:
706 self._write_string('%s\n' % self._bidi_workaround(message), self._err_file, only_once=only_once)
707
708 def to_console_title(self, message):
709 if not self.params.get('consoletitle', False):
710 return
711 if compat_os_name == 'nt':
712 if ctypes.windll.kernel32.GetConsoleWindow():
713 # c_wchar_p() might not be necessary if `message` is
714 # already of type unicode()
715 ctypes.windll.kernel32.SetConsoleTitleW(ctypes.c_wchar_p(message))
716 elif 'TERM' in os.environ:
717 self._write_string('\033]0;%s\007' % message, self._screen_file)
718
719 def save_console_title(self):
720 if not self.params.get('consoletitle', False):
721 return
722 if self.params.get('simulate'):
723 return
724 if compat_os_name != 'nt' and 'TERM' in os.environ:
725 # Save the title on stack
726 self._write_string('\033[22;0t', self._screen_file)
727
728 def restore_console_title(self):
729 if not self.params.get('consoletitle', False):
730 return
731 if self.params.get('simulate'):
732 return
733 if compat_os_name != 'nt' and 'TERM' in os.environ:
734 # Restore the title from stack
735 self._write_string('\033[23;0t', self._screen_file)
736
737 def __enter__(self):
738 self.save_console_title()
739 return self
740
741 def __exit__(self, *args):
742 self.restore_console_title()
743
744 if self.params.get('cookiefile') is not None:
745 self.cookiejar.save(ignore_discard=True, ignore_expires=True)
746
747 def trouble(self, message=None, tb=None):
748 """Determine action to take when a download problem appears.
749
750 Depending on if the downloader has been configured to ignore
751 download errors or not, this method may throw an exception or
752 not when errors are found, after printing the message.
753
754 tb, if given, is additional traceback information.
755 """
756 if message is not None:
757 self.to_stderr(message)
758 if self.params.get('verbose'):
759 if tb is None:
760 if sys.exc_info()[0]: # if .trouble has been called from an except block
761 tb = ''
762 if hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
763 tb += ''.join(traceback.format_exception(*sys.exc_info()[1].exc_info))
764 tb += encode_compat_str(traceback.format_exc())
765 else:
766 tb_data = traceback.format_list(traceback.extract_stack())
767 tb = ''.join(tb_data)
768 if tb:
769 self.to_stderr(tb)
770 if not self.params.get('ignoreerrors', False):
771 if sys.exc_info()[0] and hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
772 exc_info = sys.exc_info()[1].exc_info
773 else:
774 exc_info = sys.exc_info()
775 raise DownloadError(message, exc_info)
776 self._download_retcode = 1
777
778 def to_screen(self, message, skip_eol=False):
779 """Print message to stdout if not in quiet mode"""
780 self.to_stdout(
781 message, skip_eol, quiet=self.params.get('quiet', False))
782
783 def report_warning(self, message, only_once=False):
784 '''
785 Print the message to stderr, it will be prefixed with 'WARNING:'
786 If stderr is a tty file the 'WARNING:' will be colored
787 '''
788 if self.params.get('logger') is not None:
789 self.params['logger'].warning(message)
790 else:
791 if self.params.get('no_warnings'):
792 return
793 if not self.params.get('no_color') and self._err_file.isatty() and compat_os_name != 'nt':
794 _msg_header = '\033[0;33mWARNING:\033[0m'
795 else:
796 _msg_header = 'WARNING:'
797 warning_message = '%s %s' % (_msg_header, message)
798 self.to_stderr(warning_message, only_once)
799
800 def report_error(self, message, tb=None):
801 '''
802 Do the same as trouble, but prefixes the message with 'ERROR:', colored
803 in red if stderr is a tty file.
804 '''
805 if not self.params.get('no_color') and self._err_file.isatty() and compat_os_name != 'nt':
806 _msg_header = '\033[0;31mERROR:\033[0m'
807 else:
808 _msg_header = 'ERROR:'
809 error_message = '%s %s' % (_msg_header, message)
810 self.trouble(error_message, tb)
811
812 def write_debug(self, message, only_once=False):
813 '''Log debug message or Print message to stderr'''
814 if not self.params.get('verbose', False):
815 return
816 message = '[debug] %s' % message
817 if self.params.get('logger'):
818 self.params['logger'].debug(message)
819 else:
820 self.to_stderr(message, only_once)
821
822 def report_file_already_downloaded(self, file_name):
823 """Report file has already been fully downloaded."""
824 try:
825 self.to_screen('[download] %s has already been downloaded' % file_name)
826 except UnicodeEncodeError:
827 self.to_screen('[download] The file has already been downloaded')
828
829 def report_file_delete(self, file_name):
830 """Report that existing file will be deleted."""
831 try:
832 self.to_screen('Deleting existing file %s' % file_name)
833 except UnicodeEncodeError:
834 self.to_screen('Deleting existing file')
835
836 def raise_no_formats(self, info, forced=False):
837 has_drm = info.get('__has_drm')
838 msg = 'This video is DRM protected' if has_drm else 'No video formats found!'
839 expected = self.params.get('ignore_no_formats_error')
840 if forced or not expected:
841 raise ExtractorError(msg, video_id=info['id'], ie=info['extractor'],
842 expected=has_drm or expected)
843 else:
844 self.report_warning(msg)
845
846 def parse_outtmpl(self):
847 outtmpl_dict = self.params.get('outtmpl', {})
848 if not isinstance(outtmpl_dict, dict):
849 outtmpl_dict = {'default': outtmpl_dict}
850 outtmpl_dict.update({
851 k: v for k, v in DEFAULT_OUTTMPL.items()
852 if not outtmpl_dict.get(k)})
853 for key, val in outtmpl_dict.items():
854 if isinstance(val, bytes):
855 self.report_warning(
856 'Parameter outtmpl is bytes, but should be a unicode string. '
857 'Put from __future__ import unicode_literals at the top of your code file or consider switching to Python 3.x.')
858 return outtmpl_dict
859
860 def get_output_path(self, dir_type='', filename=None):
861 paths = self.params.get('paths', {})
862 assert isinstance(paths, dict)
863 path = os.path.join(
864 expand_path(paths.get('home', '').strip()),
865 expand_path(paths.get(dir_type, '').strip()) if dir_type else '',
866 filename or '')
867
868 # Temporary fix for #4787
869 # 'Treat' all problem characters by passing filename through preferredencoding
870 # to workaround encoding issues with subprocess on python2 @ Windows
871 if sys.version_info < (3, 0) and sys.platform == 'win32':
872 path = encodeFilename(path, True).decode(preferredencoding())
873 return sanitize_path(path, force=self.params.get('windowsfilenames'))
874
875 @staticmethod
876 def _outtmpl_expandpath(outtmpl):
877 # expand_path translates '%%' into '%' and '$$' into '$'
878 # correspondingly that is not what we want since we need to keep
879 # '%%' intact for template dict substitution step. Working around
880 # with boundary-alike separator hack.
881 sep = ''.join([random.choice(ascii_letters) for _ in range(32)])
882 outtmpl = outtmpl.replace('%%', '%{0}%'.format(sep)).replace('$$', '${0}$'.format(sep))
883
884 # outtmpl should be expand_path'ed before template dict substitution
885 # because meta fields may contain env variables we don't want to
886 # be expanded. For example, for outtmpl "%(title)s.%(ext)s" and
887 # title "Hello $PATH", we don't want `$PATH` to be expanded.
888 return expand_path(outtmpl).replace(sep, '')
889
890 @staticmethod
891 def escape_outtmpl(outtmpl):
892 ''' Escape any remaining strings like %s, %abc% etc. '''
893 return re.sub(
894 STR_FORMAT_RE_TMPL.format('', '(?![%(\0])'),
895 lambda mobj: ('' if mobj.group('has_key') else '%') + mobj.group(0),
896 outtmpl)
897
898 @classmethod
899 def validate_outtmpl(cls, outtmpl):
900 ''' @return None or Exception object '''
901 outtmpl = re.sub(
902 STR_FORMAT_RE_TMPL.format('[^)]*', '[ljq]'),
903 lambda mobj: f'{mobj.group(0)[:-1]}s',
904 cls._outtmpl_expandpath(outtmpl))
905 try:
906 cls.escape_outtmpl(outtmpl) % collections.defaultdict(int)
907 return None
908 except ValueError as err:
909 return err
910
911 def prepare_outtmpl(self, outtmpl, info_dict, sanitize=None):
912 """ Make the template and info_dict suitable for substitution : ydl.outtmpl_escape(outtmpl) % info_dict """
913 info_dict.setdefault('epoch', int(time.time())) # keep epoch consistent once set
914
915 info_dict = dict(info_dict) # Do not sanitize so as not to consume LazyList
916 for key in ('__original_infodict', '__postprocessors'):
917 info_dict.pop(key, None)
918 info_dict['duration_string'] = ( # %(duration>%H-%M-%S)s is wrong if duration > 24hrs
919 formatSeconds(info_dict['duration'], '-' if sanitize else ':')
920 if info_dict.get('duration', None) is not None
921 else None)
922 info_dict['autonumber'] = self.params.get('autonumber_start', 1) - 1 + self._num_downloads
923 if info_dict.get('resolution') is None:
924 info_dict['resolution'] = self.format_resolution(info_dict, default=None)
925
926 # For fields playlist_index and autonumber convert all occurrences
927 # of %(field)s to %(field)0Nd for backward compatibility
928 field_size_compat_map = {
929 'playlist_index': len(str(info_dict.get('_last_playlist_index') or '')),
930 'autonumber': self.params.get('autonumber_size') or 5,
931 }
932
933 TMPL_DICT = {}
934 EXTERNAL_FORMAT_RE = re.compile(STR_FORMAT_RE_TMPL.format('[^)]*', f'[{STR_FORMAT_TYPES}ljq]'))
935 MATH_FUNCTIONS = {
936 '+': float.__add__,
937 '-': float.__sub__,
938 }
939 # Field is of the form key1.key2...
940 # where keys (except first) can be string, int or slice
941 FIELD_RE = r'\w*(?:\.(?:\w+|{num}|{num}?(?::{num}?){{1,2}}))*'.format(num=r'(?:-?\d+)')
942 MATH_FIELD_RE = r'''{field}|{num}'''.format(field=FIELD_RE, num=r'-?\d+(?:.\d+)?')
943 MATH_OPERATORS_RE = r'(?:%s)' % '|'.join(map(re.escape, MATH_FUNCTIONS.keys()))
944 INTERNAL_FORMAT_RE = re.compile(r'''(?x)
945 (?P<negate>-)?
946 (?P<fields>{field})
947 (?P<maths>(?:{math_op}{math_field})*)
948 (?:>(?P<strf_format>.+?))?
949 (?:\|(?P<default>.*?))?
950 $'''.format(field=FIELD_RE, math_op=MATH_OPERATORS_RE, math_field=MATH_FIELD_RE))
951
952 def _traverse_infodict(k):
953 k = k.split('.')
954 if k[0] == '':
955 k.pop(0)
956 return traverse_obj(info_dict, k, is_user_input=True, traverse_string=True)
957
958 def get_value(mdict):
959 # Object traversal
960 value = _traverse_infodict(mdict['fields'])
961 # Negative
962 if mdict['negate']:
963 value = float_or_none(value)
964 if value is not None:
965 value *= -1
966 # Do maths
967 offset_key = mdict['maths']
968 if offset_key:
969 value = float_or_none(value)
970 operator = None
971 while offset_key:
972 item = re.match(
973 MATH_FIELD_RE if operator else MATH_OPERATORS_RE,
974 offset_key).group(0)
975 offset_key = offset_key[len(item):]
976 if operator is None:
977 operator = MATH_FUNCTIONS[item]
978 continue
979 item, multiplier = (item[1:], -1) if item[0] == '-' else (item, 1)
980 offset = float_or_none(item)
981 if offset is None:
982 offset = float_or_none(_traverse_infodict(item))
983 try:
984 value = operator(value, multiplier * offset)
985 except (TypeError, ZeroDivisionError):
986 return None
987 operator = None
988 # Datetime formatting
989 if mdict['strf_format']:
990 value = strftime_or_none(value, mdict['strf_format'])
991
992 return value
993
994 na = self.params.get('outtmpl_na_placeholder', 'NA')
995
996 def _dumpjson_default(obj):
997 if isinstance(obj, (set, LazyList)):
998 return list(obj)
999 raise TypeError(f'Object of type {type(obj).__name__} is not JSON serializable')
1000
1001 def create_key(outer_mobj):
1002 if not outer_mobj.group('has_key'):
1003 return f'%{outer_mobj.group(0)}'
1004 key = outer_mobj.group('key')
1005 mobj = re.match(INTERNAL_FORMAT_RE, key)
1006 if mobj is None:
1007 value, default, mobj = None, na, {'fields': ''}
1008 else:
1009 mobj = mobj.groupdict()
1010 default = mobj['default'] if mobj['default'] is not None else na
1011 value = get_value(mobj)
1012
1013 fmt = outer_mobj.group('format')
1014 if fmt == 's' and value is not None and key in field_size_compat_map.keys():
1015 fmt = '0{:d}d'.format(field_size_compat_map[key])
1016
1017 value = default if value is None else value
1018
1019 str_fmt = f'{fmt[:-1]}s'
1020 if fmt[-1] == 'l':
1021 value, fmt = ', '.join(variadic(value)), str_fmt
1022 elif fmt[-1] == 'j':
1023 value, fmt = json.dumps(value, default=_dumpjson_default), str_fmt
1024 elif fmt[-1] == 'q':
1025 value, fmt = compat_shlex_quote(str(value)), str_fmt
1026 elif fmt[-1] == 'c':
1027 value = str(value)
1028 if value is None:
1029 value, fmt = default, 's'
1030 else:
1031 value = value[0]
1032 elif fmt[-1] not in 'rs': # numeric
1033 value = float_or_none(value)
1034 if value is None:
1035 value, fmt = default, 's'
1036
1037 if sanitize:
1038 if fmt[-1] == 'r':
1039 # If value is an object, sanitize might convert it to a string
1040 # So we convert it to repr first
1041 value, fmt = repr(value), str_fmt
1042 if fmt[-1] in 'csr':
1043 value = sanitize(mobj['fields'].split('.')[-1], value)
1044
1045 key = '%s\0%s' % (key.replace('%', '%\0'), outer_mobj.group('format'))
1046 TMPL_DICT[key] = value
1047 return '{prefix}%({key}){fmt}'.format(key=key, fmt=fmt, prefix=outer_mobj.group('prefix'))
1048
1049 return EXTERNAL_FORMAT_RE.sub(create_key, outtmpl), TMPL_DICT
1050
1051 def _prepare_filename(self, info_dict, tmpl_type='default'):
1052 try:
1053 sanitize = lambda k, v: sanitize_filename(
1054 compat_str(v),
1055 restricted=self.params.get('restrictfilenames'),
1056 is_id=(k == 'id' or k.endswith('_id')))
1057 outtmpl = self.outtmpl_dict.get(tmpl_type, self.outtmpl_dict['default'])
1058 outtmpl, template_dict = self.prepare_outtmpl(outtmpl, info_dict, sanitize)
1059 outtmpl = self.escape_outtmpl(self._outtmpl_expandpath(outtmpl))
1060 filename = outtmpl % template_dict
1061
1062 force_ext = OUTTMPL_TYPES.get(tmpl_type)
1063 if force_ext is not None:
1064 filename = replace_extension(filename, force_ext, info_dict.get('ext'))
1065
1066 # https://github.com/blackjack4494/youtube-dlc/issues/85
1067 trim_file_name = self.params.get('trim_file_name', False)
1068 if trim_file_name:
1069 fn_groups = filename.rsplit('.')
1070 ext = fn_groups[-1]
1071 sub_ext = ''
1072 if len(fn_groups) > 2:
1073 sub_ext = fn_groups[-2]
1074 filename = '.'.join(filter(None, [fn_groups[0][:trim_file_name], sub_ext, ext]))
1075
1076 return filename
1077 except ValueError as err:
1078 self.report_error('Error in output template: ' + str(err) + ' (encoding: ' + repr(preferredencoding()) + ')')
1079 return None
1080
1081 def prepare_filename(self, info_dict, dir_type='', warn=False):
1082 """Generate the output filename."""
1083
1084 filename = self._prepare_filename(info_dict, dir_type or 'default')
1085
1086 if warn:
1087 if not self.params.get('paths'):
1088 pass
1089 elif filename == '-':
1090 self.report_warning('--paths is ignored when an outputting to stdout', only_once=True)
1091 elif os.path.isabs(filename):
1092 self.report_warning('--paths is ignored since an absolute path is given in output template', only_once=True)
1093 if filename == '-' or not filename:
1094 return filename
1095
1096 return self.get_output_path(dir_type, filename)
1097
1098 def _match_entry(self, info_dict, incomplete=False, silent=False):
1099 """ Returns None if the file should be downloaded """
1100
1101 video_title = info_dict.get('title', info_dict.get('id', 'video'))
1102
1103 def check_filter():
1104 if 'title' in info_dict:
1105 # This can happen when we're just evaluating the playlist
1106 title = info_dict['title']
1107 matchtitle = self.params.get('matchtitle', False)
1108 if matchtitle:
1109 if not re.search(matchtitle, title, re.IGNORECASE):
1110 return '"' + title + '" title did not match pattern "' + matchtitle + '"'
1111 rejecttitle = self.params.get('rejecttitle', False)
1112 if rejecttitle:
1113 if re.search(rejecttitle, title, re.IGNORECASE):
1114 return '"' + title + '" title matched reject pattern "' + rejecttitle + '"'
1115 date = info_dict.get('upload_date')
1116 if date is not None:
1117 dateRange = self.params.get('daterange', DateRange())
1118 if date not in dateRange:
1119 return '%s upload date is not in range %s' % (date_from_str(date).isoformat(), dateRange)
1120 view_count = info_dict.get('view_count')
1121 if view_count is not None:
1122 min_views = self.params.get('min_views')
1123 if min_views is not None and view_count < min_views:
1124 return 'Skipping %s, because it has not reached minimum view count (%d/%d)' % (video_title, view_count, min_views)
1125 max_views = self.params.get('max_views')
1126 if max_views is not None and view_count > max_views:
1127 return 'Skipping %s, because it has exceeded the maximum view count (%d/%d)' % (video_title, view_count, max_views)
1128 if age_restricted(info_dict.get('age_limit'), self.params.get('age_limit')):
1129 return 'Skipping "%s" because it is age restricted' % video_title
1130
1131 match_filter = self.params.get('match_filter')
1132 if match_filter is not None:
1133 try:
1134 ret = match_filter(info_dict, incomplete=incomplete)
1135 except TypeError:
1136 # For backward compatibility
1137 ret = None if incomplete else match_filter(info_dict)
1138 if ret is not None:
1139 return ret
1140 return None
1141
1142 if self.in_download_archive(info_dict):
1143 reason = '%s has already been recorded in the archive' % video_title
1144 break_opt, break_err = 'break_on_existing', ExistingVideoReached
1145 else:
1146 reason = check_filter()
1147 break_opt, break_err = 'break_on_reject', RejectedVideoReached
1148 if reason is not None:
1149 if not silent:
1150 self.to_screen('[download] ' + reason)
1151 if self.params.get(break_opt, False):
1152 raise break_err()
1153 return reason
1154
1155 @staticmethod
1156 def add_extra_info(info_dict, extra_info):
1157 '''Set the keys from extra_info in info dict if they are missing'''
1158 for key, value in extra_info.items():
1159 info_dict.setdefault(key, value)
1160
1161 def extract_info(self, url, download=True, ie_key=None, extra_info={},
1162 process=True, force_generic_extractor=False):
1163 """
1164 Return a list with a dictionary for each video extracted.
1165
1166 Arguments:
1167 url -- URL to extract
1168
1169 Keyword arguments:
1170 download -- whether to download videos during extraction
1171 ie_key -- extractor key hint
1172 extra_info -- dictionary containing the extra values to add to each result
1173 process -- whether to resolve all unresolved references (URLs, playlist items),
1174 must be True for download to work.
1175 force_generic_extractor -- force using the generic extractor
1176 """
1177
1178 if not ie_key and force_generic_extractor:
1179 ie_key = 'Generic'
1180
1181 if ie_key:
1182 ies = [self.get_info_extractor(ie_key)]
1183 else:
1184 ies = self._ies
1185
1186 for ie in ies:
1187 if not ie.suitable(url):
1188 continue
1189
1190 ie_key = ie.ie_key()
1191 ie = self.get_info_extractor(ie_key)
1192 if not ie.working():
1193 self.report_warning('The program functionality for this site has been marked as broken, '
1194 'and will probably not work.')
1195
1196 temp_id = ie.get_temp_id(url)
1197 if temp_id is not None and self.in_download_archive({'id': temp_id, 'ie_key': ie_key}):
1198 self.to_screen("[%s] %s: has already been recorded in archive" % (
1199 ie_key, temp_id))
1200 break
1201 return self.__extract_info(url, ie, download, extra_info, process)
1202 else:
1203 self.report_error('no suitable InfoExtractor for URL %s' % url)
1204
1205 def __handle_extraction_exceptions(func, handle_all_errors=True):
1206 def wrapper(self, *args, **kwargs):
1207 try:
1208 return func(self, *args, **kwargs)
1209 except GeoRestrictedError as e:
1210 msg = e.msg
1211 if e.countries:
1212 msg += '\nThis video is available in %s.' % ', '.join(
1213 map(ISO3166Utils.short2full, e.countries))
1214 msg += '\nYou might want to use a VPN or a proxy server (with --proxy) to workaround.'
1215 self.report_error(msg)
1216 except ExtractorError as e: # An error we somewhat expected
1217 self.report_error(compat_str(e), e.format_traceback())
1218 except ThrottledDownload:
1219 self.to_stderr('\r')
1220 self.report_warning('The download speed is below throttle limit. Re-extracting data')
1221 return wrapper(self, *args, **kwargs)
1222 except (MaxDownloadsReached, ExistingVideoReached, RejectedVideoReached):
1223 raise
1224 except Exception as e:
1225 if handle_all_errors and self.params.get('ignoreerrors', False):
1226 self.report_error(error_to_compat_str(e), tb=encode_compat_str(traceback.format_exc()))
1227 else:
1228 raise
1229 return wrapper
1230
1231 @__handle_extraction_exceptions
1232 def __extract_info(self, url, ie, download, extra_info, process):
1233 ie_result = ie.extract(url)
1234 if ie_result is None: # Finished already (backwards compatibility; listformats and friends should be moved here)
1235 return
1236 if isinstance(ie_result, list):
1237 # Backwards compatibility: old IE result format
1238 ie_result = {
1239 '_type': 'compat_list',
1240 'entries': ie_result,
1241 }
1242 if extra_info.get('original_url'):
1243 ie_result.setdefault('original_url', extra_info['original_url'])
1244 self.add_default_extra_info(ie_result, ie, url)
1245 if process:
1246 return self.process_ie_result(ie_result, download, extra_info)
1247 else:
1248 return ie_result
1249
1250 def add_default_extra_info(self, ie_result, ie, url):
1251 if url is not None:
1252 self.add_extra_info(ie_result, {
1253 'webpage_url': url,
1254 'original_url': url,
1255 'webpage_url_basename': url_basename(url),
1256 })
1257 if ie is not None:
1258 self.add_extra_info(ie_result, {
1259 'extractor': ie.IE_NAME,
1260 'extractor_key': ie.ie_key(),
1261 })
1262
1263 def process_ie_result(self, ie_result, download=True, extra_info=None):
1264 """
1265 Take the result of the ie(may be modified) and resolve all unresolved
1266 references (URLs, playlist items).
1267
1268 It will also download the videos if 'download'.
1269 Returns the resolved ie_result.
1270 """
1271 if extra_info is None:
1272 extra_info = {}
1273 result_type = ie_result.get('_type', 'video')
1274
1275 if result_type in ('url', 'url_transparent'):
1276 ie_result['url'] = sanitize_url(ie_result['url'])
1277 if ie_result.get('original_url'):
1278 extra_info.setdefault('original_url', ie_result['original_url'])
1279
1280 extract_flat = self.params.get('extract_flat', False)
1281 if ((extract_flat == 'in_playlist' and 'playlist' in extra_info)
1282 or extract_flat is True):
1283 info_copy = ie_result.copy()
1284 self.add_extra_info(info_copy, extra_info)
1285 ie = try_get(ie_result.get('ie_key'), self.get_info_extractor)
1286 self.add_default_extra_info(info_copy, ie, ie_result['url'])
1287 self.__forced_printings(info_copy, self.prepare_filename(info_copy), incomplete=True)
1288 return ie_result
1289
1290 if result_type == 'video':
1291 self.add_extra_info(ie_result, extra_info)
1292 ie_result = self.process_video_result(ie_result, download=download)
1293 additional_urls = (ie_result or {}).get('additional_urls')
1294 if additional_urls:
1295 # TODO: Improve MetadataParserPP to allow setting a list
1296 if isinstance(additional_urls, compat_str):
1297 additional_urls = [additional_urls]
1298 self.to_screen(
1299 '[info] %s: %d additional URL(s) requested' % (ie_result['id'], len(additional_urls)))
1300 self.write_debug('Additional URLs: "%s"' % '", "'.join(additional_urls))
1301 ie_result['additional_entries'] = [
1302 self.extract_info(
1303 url, download, extra_info,
1304 force_generic_extractor=self.params.get('force_generic_extractor'))
1305 for url in additional_urls
1306 ]
1307 return ie_result
1308 elif result_type == 'url':
1309 # We have to add extra_info to the results because it may be
1310 # contained in a playlist
1311 return self.extract_info(
1312 ie_result['url'], download,
1313 ie_key=ie_result.get('ie_key'),
1314 extra_info=extra_info)
1315 elif result_type == 'url_transparent':
1316 # Use the information from the embedding page
1317 info = self.extract_info(
1318 ie_result['url'], ie_key=ie_result.get('ie_key'),
1319 extra_info=extra_info, download=False, process=False)
1320
1321 # extract_info may return None when ignoreerrors is enabled and
1322 # extraction failed with an error, don't crash and return early
1323 # in this case
1324 if not info:
1325 return info
1326
1327 force_properties = dict(
1328 (k, v) for k, v in ie_result.items() if v is not None)
1329 for f in ('_type', 'url', 'id', 'extractor', 'extractor_key', 'ie_key'):
1330 if f in force_properties:
1331 del force_properties[f]
1332 new_result = info.copy()
1333 new_result.update(force_properties)
1334
1335 # Extracted info may not be a video result (i.e.
1336 # info.get('_type', 'video') != video) but rather an url or
1337 # url_transparent. In such cases outer metadata (from ie_result)
1338 # should be propagated to inner one (info). For this to happen
1339 # _type of info should be overridden with url_transparent. This
1340 # fixes issue from https://github.com/ytdl-org/youtube-dl/pull/11163.
1341 if new_result.get('_type') == 'url':
1342 new_result['_type'] = 'url_transparent'
1343
1344 return self.process_ie_result(
1345 new_result, download=download, extra_info=extra_info)
1346 elif result_type in ('playlist', 'multi_video'):
1347 # Protect from infinite recursion due to recursively nested playlists
1348 # (see https://github.com/ytdl-org/youtube-dl/issues/27833)
1349 webpage_url = ie_result['webpage_url']
1350 if webpage_url in self._playlist_urls:
1351 self.to_screen(
1352 '[download] Skipping already downloaded playlist: %s'
1353 % ie_result.get('title') or ie_result.get('id'))
1354 return
1355
1356 self._playlist_level += 1
1357 self._playlist_urls.add(webpage_url)
1358 self._sanitize_thumbnails(ie_result)
1359 try:
1360 return self.__process_playlist(ie_result, download)
1361 finally:
1362 self._playlist_level -= 1
1363 if not self._playlist_level:
1364 self._playlist_urls.clear()
1365 elif result_type == 'compat_list':
1366 self.report_warning(
1367 'Extractor %s returned a compat_list result. '
1368 'It needs to be updated.' % ie_result.get('extractor'))
1369
1370 def _fixup(r):
1371 self.add_extra_info(r, {
1372 'extractor': ie_result['extractor'],
1373 'webpage_url': ie_result['webpage_url'],
1374 'webpage_url_basename': url_basename(ie_result['webpage_url']),
1375 'extractor_key': ie_result['extractor_key'],
1376 })
1377 return r
1378 ie_result['entries'] = [
1379 self.process_ie_result(_fixup(r), download, extra_info)
1380 for r in ie_result['entries']
1381 ]
1382 return ie_result
1383 else:
1384 raise Exception('Invalid result type: %s' % result_type)
1385
1386 def _ensure_dir_exists(self, path):
1387 return make_dir(path, self.report_error)
1388
1389 def __process_playlist(self, ie_result, download):
1390 # We process each entry in the playlist
1391 playlist = ie_result.get('title') or ie_result.get('id')
1392 self.to_screen('[download] Downloading playlist: %s' % playlist)
1393
1394 if 'entries' not in ie_result:
1395 raise EntryNotInPlaylist()
1396 incomplete_entries = bool(ie_result.get('requested_entries'))
1397 if incomplete_entries:
1398 def fill_missing_entries(entries, indexes):
1399 ret = [None] * max(*indexes)
1400 for i, entry in zip(indexes, entries):
1401 ret[i - 1] = entry
1402 return ret
1403 ie_result['entries'] = fill_missing_entries(ie_result['entries'], ie_result['requested_entries'])
1404
1405 playlist_results = []
1406
1407 playliststart = self.params.get('playliststart', 1)
1408 playlistend = self.params.get('playlistend')
1409 # For backwards compatibility, interpret -1 as whole list
1410 if playlistend == -1:
1411 playlistend = None
1412
1413 playlistitems_str = self.params.get('playlist_items')
1414 playlistitems = None
1415 if playlistitems_str is not None:
1416 def iter_playlistitems(format):
1417 for string_segment in format.split(','):
1418 if '-' in string_segment:
1419 start, end = string_segment.split('-')
1420 for item in range(int(start), int(end) + 1):
1421 yield int(item)
1422 else:
1423 yield int(string_segment)
1424 playlistitems = orderedSet(iter_playlistitems(playlistitems_str))
1425
1426 ie_entries = ie_result['entries']
1427 msg = (
1428 'Downloading %d videos' if not isinstance(ie_entries, list)
1429 else 'Collected %d videos; downloading %%d of them' % len(ie_entries))
1430 if not isinstance(ie_entries, (list, PagedList)):
1431 ie_entries = LazyList(ie_entries)
1432
1433 def get_entry(i):
1434 return YoutubeDL.__handle_extraction_exceptions(
1435 lambda self, i: ie_entries[i - 1],
1436 False
1437 )(self, i)
1438
1439 entries = []
1440 for i in playlistitems or itertools.count(playliststart):
1441 if playlistitems is None and playlistend is not None and playlistend < i:
1442 break
1443 entry = None
1444 try:
1445 entry = get_entry(i)
1446 if entry is None:
1447 raise EntryNotInPlaylist()
1448 except (IndexError, EntryNotInPlaylist):
1449 if incomplete_entries:
1450 raise EntryNotInPlaylist()
1451 elif not playlistitems:
1452 break
1453 entries.append(entry)
1454 try:
1455 if entry is not None:
1456 self._match_entry(entry, incomplete=True, silent=True)
1457 except (ExistingVideoReached, RejectedVideoReached):
1458 break
1459 ie_result['entries'] = entries
1460
1461 # Save playlist_index before re-ordering
1462 entries = [
1463 ((playlistitems[i - 1] if playlistitems else i + playliststart - 1), entry)
1464 for i, entry in enumerate(entries, 1)
1465 if entry is not None]
1466 n_entries = len(entries)
1467
1468 if not playlistitems and (playliststart or playlistend):
1469 playlistitems = list(range(playliststart, playliststart + n_entries))
1470 ie_result['requested_entries'] = playlistitems
1471
1472 if self.params.get('allow_playlist_files', True):
1473 ie_copy = {
1474 'playlist': playlist,
1475 'playlist_id': ie_result.get('id'),
1476 'playlist_title': ie_result.get('title'),
1477 'playlist_uploader': ie_result.get('uploader'),
1478 'playlist_uploader_id': ie_result.get('uploader_id'),
1479 'playlist_index': 0,
1480 }
1481 ie_copy.update(dict(ie_result))
1482
1483 if self.params.get('writeinfojson', False):
1484 infofn = self.prepare_filename(ie_copy, 'pl_infojson')
1485 if not self._ensure_dir_exists(encodeFilename(infofn)):
1486 return
1487 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(infofn)):
1488 self.to_screen('[info] Playlist metadata is already present')
1489 else:
1490 self.to_screen('[info] Writing playlist metadata as JSON to: ' + infofn)
1491 try:
1492 write_json_file(self.sanitize_info(ie_result, self.params.get('clean_infojson', True)), infofn)
1493 except (OSError, IOError):
1494 self.report_error('Cannot write playlist metadata to JSON file ' + infofn)
1495
1496 # TODO: This should be passed to ThumbnailsConvertor if necessary
1497 self._write_thumbnails(ie_copy, self.prepare_filename(ie_copy, 'pl_thumbnail'))
1498
1499 if self.params.get('writedescription', False):
1500 descfn = self.prepare_filename(ie_copy, 'pl_description')
1501 if not self._ensure_dir_exists(encodeFilename(descfn)):
1502 return
1503 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(descfn)):
1504 self.to_screen('[info] Playlist description is already present')
1505 elif ie_result.get('description') is None:
1506 self.report_warning('There\'s no playlist description to write.')
1507 else:
1508 try:
1509 self.to_screen('[info] Writing playlist description to: ' + descfn)
1510 with io.open(encodeFilename(descfn), 'w', encoding='utf-8') as descfile:
1511 descfile.write(ie_result['description'])
1512 except (OSError, IOError):
1513 self.report_error('Cannot write playlist description file ' + descfn)
1514 return
1515
1516 if self.params.get('playlistreverse', False):
1517 entries = entries[::-1]
1518 if self.params.get('playlistrandom', False):
1519 random.shuffle(entries)
1520
1521 x_forwarded_for = ie_result.get('__x_forwarded_for_ip')
1522
1523 self.to_screen('[%s] playlist %s: %s' % (ie_result['extractor'], playlist, msg % n_entries))
1524 failures = 0
1525 max_failures = self.params.get('skip_playlist_after_errors') or float('inf')
1526 for i, entry_tuple in enumerate(entries, 1):
1527 playlist_index, entry = entry_tuple
1528 if 'playlist-index' in self.params.get('compat_options', []):
1529 playlist_index = playlistitems[i - 1] if playlistitems else i
1530 self.to_screen('[download] Downloading video %s of %s' % (i, n_entries))
1531 # This __x_forwarded_for_ip thing is a bit ugly but requires
1532 # minimal changes
1533 if x_forwarded_for:
1534 entry['__x_forwarded_for_ip'] = x_forwarded_for
1535 extra = {
1536 'n_entries': n_entries,
1537 '_last_playlist_index': max(playlistitems) if playlistitems else (playlistend or n_entries),
1538 'playlist_index': playlist_index,
1539 'playlist_autonumber': i,
1540 'playlist': playlist,
1541 'playlist_id': ie_result.get('id'),
1542 'playlist_title': ie_result.get('title'),
1543 'playlist_uploader': ie_result.get('uploader'),
1544 'playlist_uploader_id': ie_result.get('uploader_id'),
1545 'extractor': ie_result['extractor'],
1546 'webpage_url': ie_result['webpage_url'],
1547 'webpage_url_basename': url_basename(ie_result['webpage_url']),
1548 'extractor_key': ie_result['extractor_key'],
1549 }
1550
1551 if self._match_entry(entry, incomplete=True) is not None:
1552 continue
1553
1554 entry_result = self.__process_iterable_entry(entry, download, extra)
1555 if not entry_result:
1556 failures += 1
1557 if failures >= max_failures:
1558 self.report_error(
1559 'Skipping the remaining entries in playlist "%s" since %d items failed extraction' % (playlist, failures))
1560 break
1561 # TODO: skip failed (empty) entries?
1562 playlist_results.append(entry_result)
1563 ie_result['entries'] = playlist_results
1564 self.to_screen('[download] Finished downloading playlist: %s' % playlist)
1565 return ie_result
1566
1567 @__handle_extraction_exceptions
1568 def __process_iterable_entry(self, entry, download, extra_info):
1569 return self.process_ie_result(
1570 entry, download=download, extra_info=extra_info)
1571
1572 def _build_format_filter(self, filter_spec):
1573 " Returns a function to filter the formats according to the filter_spec "
1574
1575 OPERATORS = {
1576 '<': operator.lt,
1577 '<=': operator.le,
1578 '>': operator.gt,
1579 '>=': operator.ge,
1580 '=': operator.eq,
1581 '!=': operator.ne,
1582 }
1583 operator_rex = re.compile(r'''(?x)\s*
1584 (?P<key>width|height|tbr|abr|vbr|asr|filesize|filesize_approx|fps)\s*
1585 (?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
1586 (?P<value>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?)\s*
1587 ''' % '|'.join(map(re.escape, OPERATORS.keys())))
1588 m = operator_rex.fullmatch(filter_spec)
1589 if m:
1590 try:
1591 comparison_value = int(m.group('value'))
1592 except ValueError:
1593 comparison_value = parse_filesize(m.group('value'))
1594 if comparison_value is None:
1595 comparison_value = parse_filesize(m.group('value') + 'B')
1596 if comparison_value is None:
1597 raise ValueError(
1598 'Invalid value %r in format specification %r' % (
1599 m.group('value'), filter_spec))
1600 op = OPERATORS[m.group('op')]
1601
1602 if not m:
1603 STR_OPERATORS = {
1604 '=': operator.eq,
1605 '^=': lambda attr, value: attr.startswith(value),
1606 '$=': lambda attr, value: attr.endswith(value),
1607 '*=': lambda attr, value: value in attr,
1608 }
1609 str_operator_rex = re.compile(r'''(?x)\s*
1610 (?P<key>[a-zA-Z0-9._-]+)\s*
1611 (?P<negation>!\s*)?(?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
1612 (?P<value>[a-zA-Z0-9._-]+)\s*
1613 ''' % '|'.join(map(re.escape, STR_OPERATORS.keys())))
1614 m = str_operator_rex.fullmatch(filter_spec)
1615 if m:
1616 comparison_value = m.group('value')
1617 str_op = STR_OPERATORS[m.group('op')]
1618 if m.group('negation'):
1619 op = lambda attr, value: not str_op(attr, value)
1620 else:
1621 op = str_op
1622
1623 if not m:
1624 raise SyntaxError('Invalid filter specification %r' % filter_spec)
1625
1626 def _filter(f):
1627 actual_value = f.get(m.group('key'))
1628 if actual_value is None:
1629 return m.group('none_inclusive')
1630 return op(actual_value, comparison_value)
1631 return _filter
1632
1633 def _default_format_spec(self, info_dict, download=True):
1634
1635 def can_merge():
1636 merger = FFmpegMergerPP(self)
1637 return merger.available and merger.can_merge()
1638
1639 prefer_best = (
1640 not self.params.get('simulate')
1641 and download
1642 and (
1643 not can_merge()
1644 or info_dict.get('is_live', False)
1645 or self.outtmpl_dict['default'] == '-'))
1646 compat = (
1647 prefer_best
1648 or self.params.get('allow_multiple_audio_streams', False)
1649 or 'format-spec' in self.params.get('compat_opts', []))
1650
1651 return (
1652 'best/bestvideo+bestaudio' if prefer_best
1653 else 'bestvideo*+bestaudio/best' if not compat
1654 else 'bestvideo+bestaudio/best')
1655
1656 def build_format_selector(self, format_spec):
1657 def syntax_error(note, start):
1658 message = (
1659 'Invalid format specification: '
1660 '{0}\n\t{1}\n\t{2}^'.format(note, format_spec, ' ' * start[1]))
1661 return SyntaxError(message)
1662
1663 PICKFIRST = 'PICKFIRST'
1664 MERGE = 'MERGE'
1665 SINGLE = 'SINGLE'
1666 GROUP = 'GROUP'
1667 FormatSelector = collections.namedtuple('FormatSelector', ['type', 'selector', 'filters'])
1668
1669 allow_multiple_streams = {'audio': self.params.get('allow_multiple_audio_streams', False),
1670 'video': self.params.get('allow_multiple_video_streams', False)}
1671
1672 check_formats = self.params.get('check_formats')
1673
1674 def _parse_filter(tokens):
1675 filter_parts = []
1676 for type, string, start, _, _ in tokens:
1677 if type == tokenize.OP and string == ']':
1678 return ''.join(filter_parts)
1679 else:
1680 filter_parts.append(string)
1681
1682 def _remove_unused_ops(tokens):
1683 # Remove operators that we don't use and join them with the surrounding strings
1684 # for example: 'mp4' '-' 'baseline' '-' '16x9' is converted to 'mp4-baseline-16x9'
1685 ALLOWED_OPS = ('/', '+', ',', '(', ')')
1686 last_string, last_start, last_end, last_line = None, None, None, None
1687 for type, string, start, end, line in tokens:
1688 if type == tokenize.OP and string == '[':
1689 if last_string:
1690 yield tokenize.NAME, last_string, last_start, last_end, last_line
1691 last_string = None
1692 yield type, string, start, end, line
1693 # everything inside brackets will be handled by _parse_filter
1694 for type, string, start, end, line in tokens:
1695 yield type, string, start, end, line
1696 if type == tokenize.OP and string == ']':
1697 break
1698 elif type == tokenize.OP and string in ALLOWED_OPS:
1699 if last_string:
1700 yield tokenize.NAME, last_string, last_start, last_end, last_line
1701 last_string = None
1702 yield type, string, start, end, line
1703 elif type in [tokenize.NAME, tokenize.NUMBER, tokenize.OP]:
1704 if not last_string:
1705 last_string = string
1706 last_start = start
1707 last_end = end
1708 else:
1709 last_string += string
1710 if last_string:
1711 yield tokenize.NAME, last_string, last_start, last_end, last_line
1712
1713 def _parse_format_selection(tokens, inside_merge=False, inside_choice=False, inside_group=False):
1714 selectors = []
1715 current_selector = None
1716 for type, string, start, _, _ in tokens:
1717 # ENCODING is only defined in python 3.x
1718 if type == getattr(tokenize, 'ENCODING', None):
1719 continue
1720 elif type in [tokenize.NAME, tokenize.NUMBER]:
1721 current_selector = FormatSelector(SINGLE, string, [])
1722 elif type == tokenize.OP:
1723 if string == ')':
1724 if not inside_group:
1725 # ')' will be handled by the parentheses group
1726 tokens.restore_last_token()
1727 break
1728 elif inside_merge and string in ['/', ',']:
1729 tokens.restore_last_token()
1730 break
1731 elif inside_choice and string == ',':
1732 tokens.restore_last_token()
1733 break
1734 elif string == ',':
1735 if not current_selector:
1736 raise syntax_error('"," must follow a format selector', start)
1737 selectors.append(current_selector)
1738 current_selector = None
1739 elif string == '/':
1740 if not current_selector:
1741 raise syntax_error('"/" must follow a format selector', start)
1742 first_choice = current_selector
1743 second_choice = _parse_format_selection(tokens, inside_choice=True)
1744 current_selector = FormatSelector(PICKFIRST, (first_choice, second_choice), [])
1745 elif string == '[':
1746 if not current_selector:
1747 current_selector = FormatSelector(SINGLE, 'best', [])
1748 format_filter = _parse_filter(tokens)
1749 current_selector.filters.append(format_filter)
1750 elif string == '(':
1751 if current_selector:
1752 raise syntax_error('Unexpected "("', start)
1753 group = _parse_format_selection(tokens, inside_group=True)
1754 current_selector = FormatSelector(GROUP, group, [])
1755 elif string == '+':
1756 if not current_selector:
1757 raise syntax_error('Unexpected "+"', start)
1758 selector_1 = current_selector
1759 selector_2 = _parse_format_selection(tokens, inside_merge=True)
1760 if not selector_2:
1761 raise syntax_error('Expected a selector', start)
1762 current_selector = FormatSelector(MERGE, (selector_1, selector_2), [])
1763 else:
1764 raise syntax_error('Operator not recognized: "{0}"'.format(string), start)
1765 elif type == tokenize.ENDMARKER:
1766 break
1767 if current_selector:
1768 selectors.append(current_selector)
1769 return selectors
1770
1771 def _merge(formats_pair):
1772 format_1, format_2 = formats_pair
1773
1774 formats_info = []
1775 formats_info.extend(format_1.get('requested_formats', (format_1,)))
1776 formats_info.extend(format_2.get('requested_formats', (format_2,)))
1777
1778 if not allow_multiple_streams['video'] or not allow_multiple_streams['audio']:
1779 get_no_more = {'video': False, 'audio': False}
1780 for (i, fmt_info) in enumerate(formats_info):
1781 if fmt_info.get('acodec') == fmt_info.get('vcodec') == 'none':
1782 formats_info.pop(i)
1783 continue
1784 for aud_vid in ['audio', 'video']:
1785 if not allow_multiple_streams[aud_vid] and fmt_info.get(aud_vid[0] + 'codec') != 'none':
1786 if get_no_more[aud_vid]:
1787 formats_info.pop(i)
1788 break
1789 get_no_more[aud_vid] = True
1790
1791 if len(formats_info) == 1:
1792 return formats_info[0]
1793
1794 video_fmts = [fmt_info for fmt_info in formats_info if fmt_info.get('vcodec') != 'none']
1795 audio_fmts = [fmt_info for fmt_info in formats_info if fmt_info.get('acodec') != 'none']
1796
1797 the_only_video = video_fmts[0] if len(video_fmts) == 1 else None
1798 the_only_audio = audio_fmts[0] if len(audio_fmts) == 1 else None
1799
1800 output_ext = self.params.get('merge_output_format')
1801 if not output_ext:
1802 if the_only_video:
1803 output_ext = the_only_video['ext']
1804 elif the_only_audio and not video_fmts:
1805 output_ext = the_only_audio['ext']
1806 else:
1807 output_ext = 'mkv'
1808
1809 new_dict = {
1810 'requested_formats': formats_info,
1811 'format': '+'.join(fmt_info.get('format') for fmt_info in formats_info),
1812 'format_id': '+'.join(fmt_info.get('format_id') for fmt_info in formats_info),
1813 'ext': output_ext,
1814 }
1815
1816 if the_only_video:
1817 new_dict.update({
1818 'width': the_only_video.get('width'),
1819 'height': the_only_video.get('height'),
1820 'resolution': the_only_video.get('resolution') or self.format_resolution(the_only_video),
1821 'fps': the_only_video.get('fps'),
1822 'vcodec': the_only_video.get('vcodec'),
1823 'vbr': the_only_video.get('vbr'),
1824 'stretched_ratio': the_only_video.get('stretched_ratio'),
1825 })
1826
1827 if the_only_audio:
1828 new_dict.update({
1829 'acodec': the_only_audio.get('acodec'),
1830 'abr': the_only_audio.get('abr'),
1831 })
1832
1833 return new_dict
1834
1835 def _check_formats(formats):
1836 if not check_formats:
1837 yield from formats
1838 return
1839 for f in formats:
1840 self.to_screen('[info] Testing format %s' % f['format_id'])
1841 temp_file = tempfile.NamedTemporaryFile(
1842 suffix='.tmp', delete=False,
1843 dir=self.get_output_path('temp') or None)
1844 temp_file.close()
1845 try:
1846 success, _ = self.dl(temp_file.name, f, test=True)
1847 except (DownloadError, IOError, OSError, ValueError) + network_exceptions:
1848 success = False
1849 finally:
1850 if os.path.exists(temp_file.name):
1851 try:
1852 os.remove(temp_file.name)
1853 except OSError:
1854 self.report_warning('Unable to delete temporary file "%s"' % temp_file.name)
1855 if success:
1856 yield f
1857 else:
1858 self.to_screen('[info] Unable to download format %s. Skipping...' % f['format_id'])
1859
1860 def _build_selector_function(selector):
1861 if isinstance(selector, list): # ,
1862 fs = [_build_selector_function(s) for s in selector]
1863
1864 def selector_function(ctx):
1865 for f in fs:
1866 yield from f(ctx)
1867 return selector_function
1868
1869 elif selector.type == GROUP: # ()
1870 selector_function = _build_selector_function(selector.selector)
1871
1872 elif selector.type == PICKFIRST: # /
1873 fs = [_build_selector_function(s) for s in selector.selector]
1874
1875 def selector_function(ctx):
1876 for f in fs:
1877 picked_formats = list(f(ctx))
1878 if picked_formats:
1879 return picked_formats
1880 return []
1881
1882 elif selector.type == MERGE: # +
1883 selector_1, selector_2 = map(_build_selector_function, selector.selector)
1884
1885 def selector_function(ctx):
1886 for pair in itertools.product(
1887 selector_1(copy.deepcopy(ctx)), selector_2(copy.deepcopy(ctx))):
1888 yield _merge(pair)
1889
1890 elif selector.type == SINGLE: # atom
1891 format_spec = selector.selector or 'best'
1892
1893 # TODO: Add allvideo, allaudio etc by generalizing the code with best/worst selector
1894 if format_spec == 'all':
1895 def selector_function(ctx):
1896 yield from _check_formats(ctx['formats'])
1897 elif format_spec == 'mergeall':
1898 def selector_function(ctx):
1899 formats = list(_check_formats(ctx['formats']))
1900 if not formats:
1901 return
1902 merged_format = formats[-1]
1903 for f in formats[-2::-1]:
1904 merged_format = _merge((merged_format, f))
1905 yield merged_format
1906
1907 else:
1908 format_fallback, format_reverse, format_idx = False, True, 1
1909 mobj = re.match(
1910 r'(?P<bw>best|worst|b|w)(?P<type>video|audio|v|a)?(?P<mod>\*)?(?:\.(?P<n>[1-9]\d*))?$',
1911 format_spec)
1912 if mobj is not None:
1913 format_idx = int_or_none(mobj.group('n'), default=1)
1914 format_reverse = mobj.group('bw')[0] == 'b'
1915 format_type = (mobj.group('type') or [None])[0]
1916 not_format_type = {'v': 'a', 'a': 'v'}.get(format_type)
1917 format_modified = mobj.group('mod') is not None
1918
1919 format_fallback = not format_type and not format_modified # for b, w
1920 _filter_f = (
1921 (lambda f: f.get('%scodec' % format_type) != 'none')
1922 if format_type and format_modified # bv*, ba*, wv*, wa*
1923 else (lambda f: f.get('%scodec' % not_format_type) == 'none')
1924 if format_type # bv, ba, wv, wa
1925 else (lambda f: f.get('vcodec') != 'none' and f.get('acodec') != 'none')
1926 if not format_modified # b, w
1927 else lambda f: True) # b*, w*
1928 filter_f = lambda f: _filter_f(f) and (
1929 f.get('vcodec') != 'none' or f.get('acodec') != 'none')
1930 else:
1931 filter_f = ((lambda f: f.get('ext') == format_spec)
1932 if format_spec in ['mp4', 'flv', 'webm', '3gp', 'm4a', 'mp3', 'ogg', 'aac', 'wav'] # extension
1933 else (lambda f: f.get('format_id') == format_spec)) # id
1934
1935 def selector_function(ctx):
1936 formats = list(ctx['formats'])
1937 matches = list(filter(filter_f, formats)) if filter_f is not None else formats
1938 if format_fallback and ctx['incomplete_formats'] and not matches:
1939 # for extractors with incomplete formats (audio only (soundcloud)
1940 # or video only (imgur)) best/worst will fallback to
1941 # best/worst {video,audio}-only format
1942 matches = formats
1943 matches = LazyList(_check_formats(matches[::-1 if format_reverse else 1]))
1944 try:
1945 yield matches[format_idx - 1]
1946 except IndexError:
1947 return
1948
1949 filters = [self._build_format_filter(f) for f in selector.filters]
1950
1951 def final_selector(ctx):
1952 ctx_copy = copy.deepcopy(ctx)
1953 for _filter in filters:
1954 ctx_copy['formats'] = list(filter(_filter, ctx_copy['formats']))
1955 return selector_function(ctx_copy)
1956 return final_selector
1957
1958 stream = io.BytesIO(format_spec.encode('utf-8'))
1959 try:
1960 tokens = list(_remove_unused_ops(compat_tokenize_tokenize(stream.readline)))
1961 except tokenize.TokenError:
1962 raise syntax_error('Missing closing/opening brackets or parenthesis', (0, len(format_spec)))
1963
1964 class TokenIterator(object):
1965 def __init__(self, tokens):
1966 self.tokens = tokens
1967 self.counter = 0
1968
1969 def __iter__(self):
1970 return self
1971
1972 def __next__(self):
1973 if self.counter >= len(self.tokens):
1974 raise StopIteration()
1975 value = self.tokens[self.counter]
1976 self.counter += 1
1977 return value
1978
1979 next = __next__
1980
1981 def restore_last_token(self):
1982 self.counter -= 1
1983
1984 parsed_selector = _parse_format_selection(iter(TokenIterator(tokens)))
1985 return _build_selector_function(parsed_selector)
1986
1987 def _calc_headers(self, info_dict):
1988 res = std_headers.copy()
1989
1990 add_headers = info_dict.get('http_headers')
1991 if add_headers:
1992 res.update(add_headers)
1993
1994 cookies = self._calc_cookies(info_dict)
1995 if cookies:
1996 res['Cookie'] = cookies
1997
1998 if 'X-Forwarded-For' not in res:
1999 x_forwarded_for_ip = info_dict.get('__x_forwarded_for_ip')
2000 if x_forwarded_for_ip:
2001 res['X-Forwarded-For'] = x_forwarded_for_ip
2002
2003 return res
2004
2005 def _calc_cookies(self, info_dict):
2006 pr = sanitized_Request(info_dict['url'])
2007 self.cookiejar.add_cookie_header(pr)
2008 return pr.get_header('Cookie')
2009
2010 def _sanitize_thumbnails(self, info_dict):
2011 thumbnails = info_dict.get('thumbnails')
2012 if thumbnails is None:
2013 thumbnail = info_dict.get('thumbnail')
2014 if thumbnail:
2015 info_dict['thumbnails'] = thumbnails = [{'url': thumbnail}]
2016 if thumbnails:
2017 thumbnails.sort(key=lambda t: (
2018 t.get('preference') if t.get('preference') is not None else -1,
2019 t.get('width') if t.get('width') is not None else -1,
2020 t.get('height') if t.get('height') is not None else -1,
2021 t.get('id') if t.get('id') is not None else '',
2022 t.get('url')))
2023
2024 def thumbnail_tester():
2025 if self.params.get('check_formats'):
2026 test_all = True
2027 to_screen = lambda msg: self.to_screen(f'[info] {msg}')
2028 else:
2029 test_all = False
2030 to_screen = self.write_debug
2031
2032 def test_thumbnail(t):
2033 if not test_all and not t.get('_test_url'):
2034 return True
2035 to_screen('Testing thumbnail %s' % t['id'])
2036 try:
2037 self.urlopen(HEADRequest(t['url']))
2038 except network_exceptions as err:
2039 to_screen('Unable to connect to thumbnail %s URL "%s" - %s. Skipping...' % (
2040 t['id'], t['url'], error_to_compat_str(err)))
2041 return False
2042 return True
2043
2044 return test_thumbnail
2045
2046 for i, t in enumerate(thumbnails):
2047 if t.get('id') is None:
2048 t['id'] = '%d' % i
2049 if t.get('width') and t.get('height'):
2050 t['resolution'] = '%dx%d' % (t['width'], t['height'])
2051 t['url'] = sanitize_url(t['url'])
2052
2053 if self.params.get('check_formats') is not False:
2054 info_dict['thumbnails'] = LazyList(filter(thumbnail_tester(), thumbnails[::-1])).reverse()
2055 else:
2056 info_dict['thumbnails'] = thumbnails
2057
2058 def process_video_result(self, info_dict, download=True):
2059 assert info_dict.get('_type', 'video') == 'video'
2060
2061 if 'id' not in info_dict:
2062 raise ExtractorError('Missing "id" field in extractor result')
2063 if 'title' not in info_dict:
2064 raise ExtractorError('Missing "title" field in extractor result',
2065 video_id=info_dict['id'], ie=info_dict['extractor'])
2066
2067 def report_force_conversion(field, field_not, conversion):
2068 self.report_warning(
2069 '"%s" field is not %s - forcing %s conversion, there is an error in extractor'
2070 % (field, field_not, conversion))
2071
2072 def sanitize_string_field(info, string_field):
2073 field = info.get(string_field)
2074 if field is None or isinstance(field, compat_str):
2075 return
2076 report_force_conversion(string_field, 'a string', 'string')
2077 info[string_field] = compat_str(field)
2078
2079 def sanitize_numeric_fields(info):
2080 for numeric_field in self._NUMERIC_FIELDS:
2081 field = info.get(numeric_field)
2082 if field is None or isinstance(field, compat_numeric_types):
2083 continue
2084 report_force_conversion(numeric_field, 'numeric', 'int')
2085 info[numeric_field] = int_or_none(field)
2086
2087 sanitize_string_field(info_dict, 'id')
2088 sanitize_numeric_fields(info_dict)
2089
2090 if 'playlist' not in info_dict:
2091 # It isn't part of a playlist
2092 info_dict['playlist'] = None
2093 info_dict['playlist_index'] = None
2094
2095 self._sanitize_thumbnails(info_dict)
2096
2097 thumbnail = info_dict.get('thumbnail')
2098 thumbnails = info_dict.get('thumbnails')
2099 if thumbnail:
2100 info_dict['thumbnail'] = sanitize_url(thumbnail)
2101 elif thumbnails:
2102 info_dict['thumbnail'] = thumbnails[-1]['url']
2103
2104 if info_dict.get('display_id') is None and 'id' in info_dict:
2105 info_dict['display_id'] = info_dict['id']
2106
2107 for ts_key, date_key in (
2108 ('timestamp', 'upload_date'),
2109 ('release_timestamp', 'release_date'),
2110 ):
2111 if info_dict.get(date_key) is None and info_dict.get(ts_key) is not None:
2112 # Working around out-of-range timestamp values (e.g. negative ones on Windows,
2113 # see http://bugs.python.org/issue1646728)
2114 try:
2115 upload_date = datetime.datetime.utcfromtimestamp(info_dict[ts_key])
2116 info_dict[date_key] = upload_date.strftime('%Y%m%d')
2117 except (ValueError, OverflowError, OSError):
2118 pass
2119
2120 live_keys = ('is_live', 'was_live')
2121 live_status = info_dict.get('live_status')
2122 if live_status is None:
2123 for key in live_keys:
2124 if info_dict.get(key) is False:
2125 continue
2126 if info_dict.get(key):
2127 live_status = key
2128 break
2129 if all(info_dict.get(key) is False for key in live_keys):
2130 live_status = 'not_live'
2131 if live_status:
2132 info_dict['live_status'] = live_status
2133 for key in live_keys:
2134 if info_dict.get(key) is None:
2135 info_dict[key] = (live_status == key)
2136
2137 # Auto generate title fields corresponding to the *_number fields when missing
2138 # in order to always have clean titles. This is very common for TV series.
2139 for field in ('chapter', 'season', 'episode'):
2140 if info_dict.get('%s_number' % field) is not None and not info_dict.get(field):
2141 info_dict[field] = '%s %d' % (field.capitalize(), info_dict['%s_number' % field])
2142
2143 for cc_kind in ('subtitles', 'automatic_captions'):
2144 cc = info_dict.get(cc_kind)
2145 if cc:
2146 for _, subtitle in cc.items():
2147 for subtitle_format in subtitle:
2148 if subtitle_format.get('url'):
2149 subtitle_format['url'] = sanitize_url(subtitle_format['url'])
2150 if subtitle_format.get('ext') is None:
2151 subtitle_format['ext'] = determine_ext(subtitle_format['url']).lower()
2152
2153 automatic_captions = info_dict.get('automatic_captions')
2154 subtitles = info_dict.get('subtitles')
2155
2156 info_dict['requested_subtitles'] = self.process_subtitles(
2157 info_dict['id'], subtitles, automatic_captions)
2158
2159 # We now pick which formats have to be downloaded
2160 if info_dict.get('formats') is None:
2161 # There's only one format available
2162 formats = [info_dict]
2163 else:
2164 formats = info_dict['formats']
2165
2166 if not self.params.get('allow_unplayable_formats'):
2167 formats = [f for f in formats if not f.get('has_drm')]
2168 info_dict['__has_drm'] = len(info_dict.get('formats') or ['']) > len(formats)
2169
2170 if not formats:
2171 self.raise_no_formats(info_dict)
2172
2173 def is_wellformed(f):
2174 url = f.get('url')
2175 if not url:
2176 self.report_warning(
2177 '"url" field is missing or empty - skipping format, '
2178 'there is an error in extractor')
2179 return False
2180 if isinstance(url, bytes):
2181 sanitize_string_field(f, 'url')
2182 return True
2183
2184 # Filter out malformed formats for better extraction robustness
2185 formats = list(filter(is_wellformed, formats))
2186
2187 formats_dict = {}
2188
2189 # We check that all the formats have the format and format_id fields
2190 for i, format in enumerate(formats):
2191 sanitize_string_field(format, 'format_id')
2192 sanitize_numeric_fields(format)
2193 format['url'] = sanitize_url(format['url'])
2194 if not format.get('format_id'):
2195 format['format_id'] = compat_str(i)
2196 else:
2197 # Sanitize format_id from characters used in format selector expression
2198 format['format_id'] = re.sub(r'[\s,/+\[\]()]', '_', format['format_id'])
2199 format_id = format['format_id']
2200 if format_id not in formats_dict:
2201 formats_dict[format_id] = []
2202 formats_dict[format_id].append(format)
2203
2204 # Make sure all formats have unique format_id
2205 for format_id, ambiguous_formats in formats_dict.items():
2206 if len(ambiguous_formats) > 1:
2207 for i, format in enumerate(ambiguous_formats):
2208 format['format_id'] = '%s-%d' % (format_id, i)
2209
2210 for i, format in enumerate(formats):
2211 if format.get('format') is None:
2212 format['format'] = '{id} - {res}{note}'.format(
2213 id=format['format_id'],
2214 res=self.format_resolution(format),
2215 note=format_field(format, 'format_note', ' (%s)'),
2216 )
2217 # Automatically determine file extension if missing
2218 if format.get('ext') is None:
2219 format['ext'] = determine_ext(format['url']).lower()
2220 # Automatically determine protocol if missing (useful for format
2221 # selection purposes)
2222 if format.get('protocol') is None:
2223 format['protocol'] = determine_protocol(format)
2224 # Add HTTP headers, so that external programs can use them from the
2225 # json output
2226 full_format_info = info_dict.copy()
2227 full_format_info.update(format)
2228 format['http_headers'] = self._calc_headers(full_format_info)
2229 # Remove private housekeeping stuff
2230 if '__x_forwarded_for_ip' in info_dict:
2231 del info_dict['__x_forwarded_for_ip']
2232
2233 # TODO Central sorting goes here
2234
2235 if not formats or formats[0] is not info_dict:
2236 # only set the 'formats' fields if the original info_dict list them
2237 # otherwise we end up with a circular reference, the first (and unique)
2238 # element in the 'formats' field in info_dict is info_dict itself,
2239 # which can't be exported to json
2240 info_dict['formats'] = formats
2241
2242 info_dict, _ = self.pre_process(info_dict)
2243
2244 if self.params.get('list_thumbnails'):
2245 self.list_thumbnails(info_dict)
2246 if self.params.get('listformats'):
2247 if not info_dict.get('formats') and not info_dict.get('url'):
2248 self.to_screen('%s has no formats' % info_dict['id'])
2249 else:
2250 self.list_formats(info_dict)
2251 if self.params.get('listsubtitles'):
2252 if 'automatic_captions' in info_dict:
2253 self.list_subtitles(
2254 info_dict['id'], automatic_captions, 'automatic captions')
2255 self.list_subtitles(info_dict['id'], subtitles, 'subtitles')
2256 list_only = self.params.get('simulate') is None and (
2257 self.params.get('list_thumbnails') or self.params.get('listformats') or self.params.get('listsubtitles'))
2258 if list_only:
2259 # Without this printing, -F --print-json will not work
2260 self.__forced_printings(info_dict, self.prepare_filename(info_dict), incomplete=True)
2261 return
2262
2263 format_selector = self.format_selector
2264 if format_selector is None:
2265 req_format = self._default_format_spec(info_dict, download=download)
2266 self.write_debug('Default format spec: %s' % req_format)
2267 format_selector = self.build_format_selector(req_format)
2268
2269 # While in format selection we may need to have an access to the original
2270 # format set in order to calculate some metrics or do some processing.
2271 # For now we need to be able to guess whether original formats provided
2272 # by extractor are incomplete or not (i.e. whether extractor provides only
2273 # video-only or audio-only formats) for proper formats selection for
2274 # extractors with such incomplete formats (see
2275 # https://github.com/ytdl-org/youtube-dl/pull/5556).
2276 # Since formats may be filtered during format selection and may not match
2277 # the original formats the results may be incorrect. Thus original formats
2278 # or pre-calculated metrics should be passed to format selection routines
2279 # as well.
2280 # We will pass a context object containing all necessary additional data
2281 # instead of just formats.
2282 # This fixes incorrect format selection issue (see
2283 # https://github.com/ytdl-org/youtube-dl/issues/10083).
2284 incomplete_formats = (
2285 # All formats are video-only or
2286 all(f.get('vcodec') != 'none' and f.get('acodec') == 'none' for f in formats)
2287 # all formats are audio-only
2288 or all(f.get('vcodec') == 'none' and f.get('acodec') != 'none' for f in formats))
2289
2290 ctx = {
2291 'formats': formats,
2292 'incomplete_formats': incomplete_formats,
2293 }
2294
2295 formats_to_download = list(format_selector(ctx))
2296 if not formats_to_download:
2297 if not self.params.get('ignore_no_formats_error'):
2298 raise ExtractorError('Requested format is not available', expected=True,
2299 video_id=info_dict['id'], ie=info_dict['extractor'])
2300 else:
2301 self.report_warning('Requested format is not available')
2302 # Process what we can, even without any available formats.
2303 self.process_info(dict(info_dict))
2304 elif download:
2305 self.to_screen(
2306 '[info] %s: Downloading %d format(s): %s' % (
2307 info_dict['id'], len(formats_to_download),
2308 ", ".join([f['format_id'] for f in formats_to_download])))
2309 for fmt in formats_to_download:
2310 new_info = dict(info_dict)
2311 # Save a reference to the original info_dict so that it can be modified in process_info if needed
2312 new_info['__original_infodict'] = info_dict
2313 new_info.update(fmt)
2314 self.process_info(new_info)
2315 # We update the info dict with the best quality format (backwards compatibility)
2316 if formats_to_download:
2317 info_dict.update(formats_to_download[-1])
2318 return info_dict
2319
2320 def process_subtitles(self, video_id, normal_subtitles, automatic_captions):
2321 """Select the requested subtitles and their format"""
2322 available_subs = {}
2323 if normal_subtitles and self.params.get('writesubtitles'):
2324 available_subs.update(normal_subtitles)
2325 if automatic_captions and self.params.get('writeautomaticsub'):
2326 for lang, cap_info in automatic_captions.items():
2327 if lang not in available_subs:
2328 available_subs[lang] = cap_info
2329
2330 if (not self.params.get('writesubtitles') and not
2331 self.params.get('writeautomaticsub') or not
2332 available_subs):
2333 return None
2334
2335 all_sub_langs = available_subs.keys()
2336 if self.params.get('allsubtitles', False):
2337 requested_langs = all_sub_langs
2338 elif self.params.get('subtitleslangs', False):
2339 requested_langs = set()
2340 for lang in self.params.get('subtitleslangs'):
2341 if lang == 'all':
2342 requested_langs.update(all_sub_langs)
2343 continue
2344 discard = lang[0] == '-'
2345 if discard:
2346 lang = lang[1:]
2347 current_langs = filter(re.compile(lang + '$').match, all_sub_langs)
2348 if discard:
2349 for lang in current_langs:
2350 requested_langs.discard(lang)
2351 else:
2352 requested_langs.update(current_langs)
2353 elif 'en' in available_subs:
2354 requested_langs = ['en']
2355 else:
2356 requested_langs = [list(all_sub_langs)[0]]
2357 if requested_langs:
2358 self.write_debug('Downloading subtitles: %s' % ', '.join(requested_langs))
2359
2360 formats_query = self.params.get('subtitlesformat', 'best')
2361 formats_preference = formats_query.split('/') if formats_query else []
2362 subs = {}
2363 for lang in requested_langs:
2364 formats = available_subs.get(lang)
2365 if formats is None:
2366 self.report_warning('%s subtitles not available for %s' % (lang, video_id))
2367 continue
2368 for ext in formats_preference:
2369 if ext == 'best':
2370 f = formats[-1]
2371 break
2372 matches = list(filter(lambda f: f['ext'] == ext, formats))
2373 if matches:
2374 f = matches[-1]
2375 break
2376 else:
2377 f = formats[-1]
2378 self.report_warning(
2379 'No subtitle format found matching "%s" for language %s, '
2380 'using %s' % (formats_query, lang, f['ext']))
2381 subs[lang] = f
2382 return subs
2383
2384 def __forced_printings(self, info_dict, filename, incomplete):
2385 def print_mandatory(field, actual_field=None):
2386 if actual_field is None:
2387 actual_field = field
2388 if (self.params.get('force%s' % field, False)
2389 and (not incomplete or info_dict.get(actual_field) is not None)):
2390 self.to_stdout(info_dict[actual_field])
2391
2392 def print_optional(field):
2393 if (self.params.get('force%s' % field, False)
2394 and info_dict.get(field) is not None):
2395 self.to_stdout(info_dict[field])
2396
2397 info_dict = info_dict.copy()
2398 if filename is not None:
2399 info_dict['filename'] = filename
2400 if info_dict.get('requested_formats') is not None:
2401 # For RTMP URLs, also include the playpath
2402 info_dict['urls'] = '\n'.join(f['url'] + f.get('play_path', '') for f in info_dict['requested_formats'])
2403 elif 'url' in info_dict:
2404 info_dict['urls'] = info_dict['url'] + info_dict.get('play_path', '')
2405
2406 if self.params.get('forceprint') or self.params.get('forcejson'):
2407 self.post_extract(info_dict)
2408 for tmpl in self.params.get('forceprint', []):
2409 if re.match(r'\w+$', tmpl):
2410 tmpl = '%({})s'.format(tmpl)
2411 tmpl, info_copy = self.prepare_outtmpl(tmpl, info_dict)
2412 self.to_stdout(self.escape_outtmpl(tmpl) % info_copy)
2413
2414 print_mandatory('title')
2415 print_mandatory('id')
2416 print_mandatory('url', 'urls')
2417 print_optional('thumbnail')
2418 print_optional('description')
2419 print_optional('filename')
2420 if self.params.get('forceduration') and info_dict.get('duration') is not None:
2421 self.to_stdout(formatSeconds(info_dict['duration']))
2422 print_mandatory('format')
2423
2424 if self.params.get('forcejson'):
2425 self.to_stdout(json.dumps(self.sanitize_info(info_dict)))
2426
2427 def dl(self, name, info, subtitle=False, test=False):
2428 if not info.get('url'):
2429 self.raise_no_formats(info, True)
2430
2431 if test:
2432 verbose = self.params.get('verbose')
2433 params = {
2434 'test': True,
2435 'quiet': not verbose,
2436 'verbose': verbose,
2437 'noprogress': not verbose,
2438 'nopart': True,
2439 'skip_unavailable_fragments': False,
2440 'keep_fragments': False,
2441 'overwrites': True,
2442 '_no_ytdl_file': True,
2443 }
2444 else:
2445 params = self.params
2446 fd = get_suitable_downloader(info, params, to_stdout=(name == '-'))(self, params)
2447 if not test:
2448 for ph in self._progress_hooks:
2449 fd.add_progress_hook(ph)
2450 urls = '", "'.join([f['url'] for f in info.get('requested_formats', [])] or [info['url']])
2451 self.write_debug('Invoking downloader on "%s"' % urls)
2452 new_info = dict(info)
2453 if new_info.get('http_headers') is None:
2454 new_info['http_headers'] = self._calc_headers(new_info)
2455 return fd.download(name, new_info, subtitle)
2456
2457 def process_info(self, info_dict):
2458 """Process a single resolved IE result."""
2459
2460 assert info_dict.get('_type', 'video') == 'video'
2461
2462 max_downloads = self.params.get('max_downloads')
2463 if max_downloads is not None:
2464 if self._num_downloads >= int(max_downloads):
2465 raise MaxDownloadsReached()
2466
2467 # TODO: backward compatibility, to be removed
2468 info_dict['fulltitle'] = info_dict['title']
2469
2470 if 'format' not in info_dict and 'ext' in info_dict:
2471 info_dict['format'] = info_dict['ext']
2472
2473 if self._match_entry(info_dict) is not None:
2474 return
2475
2476 self.post_extract(info_dict)
2477 self._num_downloads += 1
2478
2479 # info_dict['_filename'] needs to be set for backward compatibility
2480 info_dict['_filename'] = full_filename = self.prepare_filename(info_dict, warn=True)
2481 temp_filename = self.prepare_filename(info_dict, 'temp')
2482 files_to_move = {}
2483
2484 # Forced printings
2485 self.__forced_printings(info_dict, full_filename, incomplete=('format' not in info_dict))
2486
2487 if self.params.get('simulate'):
2488 if self.params.get('force_write_download_archive', False):
2489 self.record_download_archive(info_dict)
2490
2491 # Do nothing else if in simulate mode
2492 return
2493
2494 if full_filename is None:
2495 return
2496
2497 if not self._ensure_dir_exists(encodeFilename(full_filename)):
2498 return
2499 if not self._ensure_dir_exists(encodeFilename(temp_filename)):
2500 return
2501
2502 if self.params.get('writedescription', False):
2503 descfn = self.prepare_filename(info_dict, 'description')
2504 if not self._ensure_dir_exists(encodeFilename(descfn)):
2505 return
2506 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(descfn)):
2507 self.to_screen('[info] Video description is already present')
2508 elif info_dict.get('description') is None:
2509 self.report_warning('There\'s no description to write.')
2510 else:
2511 try:
2512 self.to_screen('[info] Writing video description to: ' + descfn)
2513 with io.open(encodeFilename(descfn), 'w', encoding='utf-8') as descfile:
2514 descfile.write(info_dict['description'])
2515 except (OSError, IOError):
2516 self.report_error('Cannot write description file ' + descfn)
2517 return
2518
2519 if self.params.get('writeannotations', False):
2520 annofn = self.prepare_filename(info_dict, 'annotation')
2521 if not self._ensure_dir_exists(encodeFilename(annofn)):
2522 return
2523 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(annofn)):
2524 self.to_screen('[info] Video annotations are already present')
2525 elif not info_dict.get('annotations'):
2526 self.report_warning('There are no annotations to write.')
2527 else:
2528 try:
2529 self.to_screen('[info] Writing video annotations to: ' + annofn)
2530 with io.open(encodeFilename(annofn), 'w', encoding='utf-8') as annofile:
2531 annofile.write(info_dict['annotations'])
2532 except (KeyError, TypeError):
2533 self.report_warning('There are no annotations to write.')
2534 except (OSError, IOError):
2535 self.report_error('Cannot write annotations file: ' + annofn)
2536 return
2537
2538 subtitles_are_requested = any([self.params.get('writesubtitles', False),
2539 self.params.get('writeautomaticsub')])
2540
2541 if subtitles_are_requested and info_dict.get('requested_subtitles'):
2542 # subtitles download errors are already managed as troubles in relevant IE
2543 # that way it will silently go on when used with unsupporting IE
2544 subtitles = info_dict['requested_subtitles']
2545 # ie = self.get_info_extractor(info_dict['extractor_key'])
2546 for sub_lang, sub_info in subtitles.items():
2547 sub_format = sub_info['ext']
2548 sub_filename = subtitles_filename(temp_filename, sub_lang, sub_format, info_dict.get('ext'))
2549 sub_filename_final = subtitles_filename(
2550 self.prepare_filename(info_dict, 'subtitle'), sub_lang, sub_format, info_dict.get('ext'))
2551 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(sub_filename)):
2552 self.to_screen('[info] Video subtitle %s.%s is already present' % (sub_lang, sub_format))
2553 sub_info['filepath'] = sub_filename
2554 files_to_move[sub_filename] = sub_filename_final
2555 else:
2556 self.to_screen('[info] Writing video subtitles to: ' + sub_filename)
2557 if sub_info.get('data') is not None:
2558 try:
2559 # Use newline='' to prevent conversion of newline characters
2560 # See https://github.com/ytdl-org/youtube-dl/issues/10268
2561 with io.open(encodeFilename(sub_filename), 'w', encoding='utf-8', newline='') as subfile:
2562 subfile.write(sub_info['data'])
2563 sub_info['filepath'] = sub_filename
2564 files_to_move[sub_filename] = sub_filename_final
2565 except (OSError, IOError):
2566 self.report_error('Cannot write subtitles file ' + sub_filename)
2567 return
2568 else:
2569 try:
2570 self.dl(sub_filename, sub_info.copy(), subtitle=True)
2571 sub_info['filepath'] = sub_filename
2572 files_to_move[sub_filename] = sub_filename_final
2573 except (ExtractorError, IOError, OSError, ValueError) + network_exceptions as err:
2574 self.report_warning('Unable to download subtitle for "%s": %s' %
2575 (sub_lang, error_to_compat_str(err)))
2576 continue
2577
2578 if self.params.get('writeinfojson', False):
2579 infofn = self.prepare_filename(info_dict, 'infojson')
2580 if not self._ensure_dir_exists(encodeFilename(infofn)):
2581 return
2582 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(infofn)):
2583 self.to_screen('[info] Video metadata is already present')
2584 else:
2585 self.to_screen('[info] Writing video metadata as JSON to: ' + infofn)
2586 try:
2587 write_json_file(self.sanitize_info(info_dict, self.params.get('clean_infojson', True)), infofn)
2588 except (OSError, IOError):
2589 self.report_error('Cannot write video metadata to JSON file ' + infofn)
2590 return
2591 info_dict['__infojson_filename'] = infofn
2592
2593 for thumb_ext in self._write_thumbnails(info_dict, temp_filename):
2594 thumb_filename_temp = replace_extension(temp_filename, thumb_ext, info_dict.get('ext'))
2595 thumb_filename = replace_extension(
2596 self.prepare_filename(info_dict, 'thumbnail'), thumb_ext, info_dict.get('ext'))
2597 files_to_move[thumb_filename_temp] = thumb_filename
2598
2599 # Write internet shortcut files
2600 url_link = webloc_link = desktop_link = False
2601 if self.params.get('writelink', False):
2602 if sys.platform == "darwin": # macOS.
2603 webloc_link = True
2604 elif sys.platform.startswith("linux"):
2605 desktop_link = True
2606 else: # if sys.platform in ['win32', 'cygwin']:
2607 url_link = True
2608 if self.params.get('writeurllink', False):
2609 url_link = True
2610 if self.params.get('writewebloclink', False):
2611 webloc_link = True
2612 if self.params.get('writedesktoplink', False):
2613 desktop_link = True
2614
2615 if url_link or webloc_link or desktop_link:
2616 if 'webpage_url' not in info_dict:
2617 self.report_error('Cannot write internet shortcut file because the "webpage_url" field is missing in the media information')
2618 return
2619 ascii_url = iri_to_uri(info_dict['webpage_url'])
2620
2621 def _write_link_file(extension, template, newline, embed_filename):
2622 linkfn = replace_extension(full_filename, extension, info_dict.get('ext'))
2623 if self.params.get('overwrites', True) and os.path.exists(encodeFilename(linkfn)):
2624 self.to_screen('[info] Internet shortcut is already present')
2625 else:
2626 try:
2627 self.to_screen('[info] Writing internet shortcut to: ' + linkfn)
2628 with io.open(encodeFilename(to_high_limit_path(linkfn)), 'w', encoding='utf-8', newline=newline) as linkfile:
2629 template_vars = {'url': ascii_url}
2630 if embed_filename:
2631 template_vars['filename'] = linkfn[:-(len(extension) + 1)]
2632 linkfile.write(template % template_vars)
2633 except (OSError, IOError):
2634 self.report_error('Cannot write internet shortcut ' + linkfn)
2635 return False
2636 return True
2637
2638 if url_link:
2639 if not _write_link_file('url', DOT_URL_LINK_TEMPLATE, '\r\n', embed_filename=False):
2640 return
2641 if webloc_link:
2642 if not _write_link_file('webloc', DOT_WEBLOC_LINK_TEMPLATE, '\n', embed_filename=False):
2643 return
2644 if desktop_link:
2645 if not _write_link_file('desktop', DOT_DESKTOP_LINK_TEMPLATE, '\n', embed_filename=True):
2646 return
2647
2648 try:
2649 info_dict, files_to_move = self.pre_process(info_dict, 'before_dl', files_to_move)
2650 except PostProcessingError as err:
2651 self.report_error('Preprocessing: %s' % str(err))
2652 return
2653
2654 must_record_download_archive = False
2655 if self.params.get('skip_download', False):
2656 info_dict['filepath'] = temp_filename
2657 info_dict['__finaldir'] = os.path.dirname(os.path.abspath(encodeFilename(full_filename)))
2658 info_dict['__files_to_move'] = files_to_move
2659 info_dict = self.run_pp(MoveFilesAfterDownloadPP(self, False), info_dict)
2660 else:
2661 # Download
2662 info_dict.setdefault('__postprocessors', [])
2663 try:
2664
2665 def existing_file(*filepaths):
2666 ext = info_dict.get('ext')
2667 final_ext = self.params.get('final_ext', ext)
2668 existing_files = []
2669 for file in orderedSet(filepaths):
2670 if final_ext != ext:
2671 converted = replace_extension(file, final_ext, ext)
2672 if os.path.exists(encodeFilename(converted)):
2673 existing_files.append(converted)
2674 if os.path.exists(encodeFilename(file)):
2675 existing_files.append(file)
2676
2677 if not existing_files or self.params.get('overwrites', False):
2678 for file in orderedSet(existing_files):
2679 self.report_file_delete(file)
2680 os.remove(encodeFilename(file))
2681 return None
2682
2683 info_dict['ext'] = os.path.splitext(existing_files[0])[1][1:]
2684 return existing_files[0]
2685
2686 success = True
2687 if info_dict.get('requested_formats') is not None:
2688
2689 def compatible_formats(formats):
2690 # TODO: some formats actually allow this (mkv, webm, ogg, mp4), but not all of them.
2691 video_formats = [format for format in formats if format.get('vcodec') != 'none']
2692 audio_formats = [format for format in formats if format.get('acodec') != 'none']
2693 if len(video_formats) > 2 or len(audio_formats) > 2:
2694 return False
2695
2696 # Check extension
2697 exts = set(format.get('ext') for format in formats)
2698 COMPATIBLE_EXTS = (
2699 set(('mp3', 'mp4', 'm4a', 'm4p', 'm4b', 'm4r', 'm4v', 'ismv', 'isma')),
2700 set(('webm',)),
2701 )
2702 for ext_sets in COMPATIBLE_EXTS:
2703 if ext_sets.issuperset(exts):
2704 return True
2705 # TODO: Check acodec/vcodec
2706 return False
2707
2708 requested_formats = info_dict['requested_formats']
2709 old_ext = info_dict['ext']
2710 if self.params.get('merge_output_format') is None and not compatible_formats(requested_formats):
2711 info_dict['ext'] = 'mkv'
2712 self.report_warning(
2713 'Requested formats are incompatible for merge and will be merged into mkv.')
2714 new_ext = info_dict['ext']
2715
2716 def correct_ext(filename, ext=new_ext):
2717 if filename == '-':
2718 return filename
2719 filename_real_ext = os.path.splitext(filename)[1][1:]
2720 filename_wo_ext = (
2721 os.path.splitext(filename)[0]
2722 if filename_real_ext in (old_ext, new_ext)
2723 else filename)
2724 return '%s.%s' % (filename_wo_ext, ext)
2725
2726 # Ensure filename always has a correct extension for successful merge
2727 full_filename = correct_ext(full_filename)
2728 temp_filename = correct_ext(temp_filename)
2729 dl_filename = existing_file(full_filename, temp_filename)
2730 info_dict['__real_download'] = False
2731
2732 _protocols = set(determine_protocol(f) for f in requested_formats)
2733 if len(_protocols) == 1: # All requested formats have same protocol
2734 info_dict['protocol'] = _protocols.pop()
2735 directly_mergable = FFmpegFD.can_merge_formats(info_dict)
2736 if dl_filename is not None:
2737 self.report_file_already_downloaded(dl_filename)
2738 elif (directly_mergable and get_suitable_downloader(
2739 info_dict, self.params, to_stdout=(temp_filename == '-')) == FFmpegFD):
2740 info_dict['url'] = '\n'.join(f['url'] for f in requested_formats)
2741 success, real_download = self.dl(temp_filename, info_dict)
2742 info_dict['__real_download'] = real_download
2743 else:
2744 downloaded = []
2745 merger = FFmpegMergerPP(self)
2746 if self.params.get('allow_unplayable_formats'):
2747 self.report_warning(
2748 'You have requested merging of multiple formats '
2749 'while also allowing unplayable formats to be downloaded. '
2750 'The formats won\'t be merged to prevent data corruption.')
2751 elif not merger.available:
2752 self.report_warning(
2753 'You have requested merging of multiple formats but ffmpeg is not installed. '
2754 'The formats won\'t be merged.')
2755
2756 if temp_filename == '-':
2757 reason = ('using a downloader other than ffmpeg' if directly_mergable
2758 else 'but the formats are incompatible for simultaneous download' if merger.available
2759 else 'but ffmpeg is not installed')
2760 self.report_warning(
2761 f'You have requested downloading multiple formats to stdout {reason}. '
2762 'The formats will be streamed one after the other')
2763 fname = temp_filename
2764 for f in requested_formats:
2765 new_info = dict(info_dict)
2766 del new_info['requested_formats']
2767 new_info.update(f)
2768 if temp_filename != '-':
2769 fname = prepend_extension(
2770 correct_ext(temp_filename, new_info['ext']),
2771 'f%s' % f['format_id'], new_info['ext'])
2772 if not self._ensure_dir_exists(fname):
2773 return
2774 downloaded.append(fname)
2775 partial_success, real_download = self.dl(fname, new_info)
2776 info_dict['__real_download'] = info_dict['__real_download'] or real_download
2777 success = success and partial_success
2778 if merger.available and not self.params.get('allow_unplayable_formats'):
2779 info_dict['__postprocessors'].append(merger)
2780 info_dict['__files_to_merge'] = downloaded
2781 # Even if there were no downloads, it is being merged only now
2782 info_dict['__real_download'] = True
2783 else:
2784 for file in downloaded:
2785 files_to_move[file] = None
2786 else:
2787 # Just a single file
2788 dl_filename = existing_file(full_filename, temp_filename)
2789 if dl_filename is None or dl_filename == temp_filename:
2790 # dl_filename == temp_filename could mean that the file was partially downloaded with --no-part.
2791 # So we should try to resume the download
2792 success, real_download = self.dl(temp_filename, info_dict)
2793 info_dict['__real_download'] = real_download
2794 else:
2795 self.report_file_already_downloaded(dl_filename)
2796
2797 dl_filename = dl_filename or temp_filename
2798 info_dict['__finaldir'] = os.path.dirname(os.path.abspath(encodeFilename(full_filename)))
2799
2800 except network_exceptions as err:
2801 self.report_error('unable to download video data: %s' % error_to_compat_str(err))
2802 return
2803 except (OSError, IOError) as err:
2804 raise UnavailableVideoError(err)
2805 except (ContentTooShortError, ) as err:
2806 self.report_error('content too short (expected %s bytes and served %s)' % (err.expected, err.downloaded))
2807 return
2808
2809 if success and full_filename != '-':
2810
2811 def fixup():
2812 do_fixup = True
2813 fixup_policy = self.params.get('fixup')
2814 vid = info_dict['id']
2815
2816 if fixup_policy in ('ignore', 'never'):
2817 return
2818 elif fixup_policy == 'warn':
2819 do_fixup = False
2820 elif fixup_policy != 'force':
2821 assert fixup_policy in ('detect_or_warn', None)
2822 if not info_dict.get('__real_download'):
2823 do_fixup = False
2824
2825 def ffmpeg_fixup(cndn, msg, cls):
2826 if not cndn:
2827 return
2828 if not do_fixup:
2829 self.report_warning(f'{vid}: {msg}')
2830 return
2831 pp = cls(self)
2832 if pp.available:
2833 info_dict['__postprocessors'].append(pp)
2834 else:
2835 self.report_warning(f'{vid}: {msg}. Install ffmpeg to fix this automatically')
2836
2837 stretched_ratio = info_dict.get('stretched_ratio')
2838 ffmpeg_fixup(
2839 stretched_ratio not in (1, None),
2840 f'Non-uniform pixel ratio {stretched_ratio}',
2841 FFmpegFixupStretchedPP)
2842
2843 ffmpeg_fixup(
2844 (info_dict.get('requested_formats') is None
2845 and info_dict.get('container') == 'm4a_dash'
2846 and info_dict.get('ext') == 'm4a'),
2847 'writing DASH m4a. Only some players support this container',
2848 FFmpegFixupM4aPP)
2849
2850 downloader = (get_suitable_downloader(info_dict, self.params).__name__
2851 if 'protocol' in info_dict else None)
2852 ffmpeg_fixup(downloader == 'HlsFD', 'malformed AAC bitstream detected', FFmpegFixupM3u8PP)
2853 ffmpeg_fixup(downloader == 'WebSocketFragmentFD', 'malformed timestamps detected', FFmpegFixupTimestampPP)
2854 ffmpeg_fixup(downloader == 'WebSocketFragmentFD', 'malformed duration detected', FFmpegFixupDurationPP)
2855
2856 fixup()
2857 try:
2858 info_dict = self.post_process(dl_filename, info_dict, files_to_move)
2859 except PostProcessingError as err:
2860 self.report_error('Postprocessing: %s' % str(err))
2861 return
2862 try:
2863 for ph in self._post_hooks:
2864 ph(info_dict['filepath'])
2865 except Exception as err:
2866 self.report_error('post hooks: %s' % str(err))
2867 return
2868 must_record_download_archive = True
2869
2870 if must_record_download_archive or self.params.get('force_write_download_archive', False):
2871 self.record_download_archive(info_dict)
2872 max_downloads = self.params.get('max_downloads')
2873 if max_downloads is not None and self._num_downloads >= int(max_downloads):
2874 raise MaxDownloadsReached()
2875
2876 def download(self, url_list):
2877 """Download a given list of URLs."""
2878 outtmpl = self.outtmpl_dict['default']
2879 if (len(url_list) > 1
2880 and outtmpl != '-'
2881 and '%' not in outtmpl
2882 and self.params.get('max_downloads') != 1):
2883 raise SameFileError(outtmpl)
2884
2885 for url in url_list:
2886 try:
2887 # It also downloads the videos
2888 res = self.extract_info(
2889 url, force_generic_extractor=self.params.get('force_generic_extractor', False))
2890 except UnavailableVideoError:
2891 self.report_error('unable to download video')
2892 except MaxDownloadsReached:
2893 self.to_screen('[info] Maximum number of downloads reached')
2894 raise
2895 except ExistingVideoReached:
2896 self.to_screen('[info] Encountered a video that is already in the archive, stopping due to --break-on-existing')
2897 raise
2898 except RejectedVideoReached:
2899 self.to_screen('[info] Encountered a video that did not match filter, stopping due to --break-on-reject')
2900 raise
2901 else:
2902 if self.params.get('dump_single_json', False):
2903 self.post_extract(res)
2904 self.to_stdout(json.dumps(self.sanitize_info(res)))
2905
2906 return self._download_retcode
2907
2908 def download_with_info_file(self, info_filename):
2909 with contextlib.closing(fileinput.FileInput(
2910 [info_filename], mode='r',
2911 openhook=fileinput.hook_encoded('utf-8'))) as f:
2912 # FileInput doesn't have a read method, we can't call json.load
2913 info = self.sanitize_info(json.loads('\n'.join(f)), self.params.get('clean_infojson', True))
2914 try:
2915 self.process_ie_result(info, download=True)
2916 except (DownloadError, EntryNotInPlaylist, ThrottledDownload):
2917 webpage_url = info.get('webpage_url')
2918 if webpage_url is not None:
2919 self.report_warning('The info failed to download, trying with "%s"' % webpage_url)
2920 return self.download([webpage_url])
2921 else:
2922 raise
2923 return self._download_retcode
2924
2925 @staticmethod
2926 def sanitize_info(info_dict, remove_private_keys=False):
2927 ''' Sanitize the infodict for converting to json '''
2928 if info_dict is None:
2929 return info_dict
2930 info_dict.setdefault('epoch', int(time.time()))
2931 remove_keys = {'__original_infodict'} # Always remove this since this may contain a copy of the entire dict
2932 keep_keys = ['_type'], # Always keep this to facilitate load-info-json
2933 if remove_private_keys:
2934 remove_keys |= {
2935 'requested_formats', 'requested_subtitles', 'requested_entries',
2936 'filepath', 'entries', 'original_url', 'playlist_autonumber',
2937 }
2938 empty_values = (None, {}, [], set(), tuple())
2939 reject = lambda k, v: k not in keep_keys and (
2940 k.startswith('_') or k in remove_keys or v in empty_values)
2941 else:
2942 reject = lambda k, v: k in remove_keys
2943 filter_fn = lambda obj: (
2944 list(map(filter_fn, obj)) if isinstance(obj, (LazyList, list, tuple, set))
2945 else obj if not isinstance(obj, dict)
2946 else dict((k, filter_fn(v)) for k, v in obj.items() if not reject(k, v)))
2947 return filter_fn(info_dict)
2948
2949 @staticmethod
2950 def filter_requested_info(info_dict, actually_filter=True):
2951 ''' Alias of sanitize_info for backward compatibility '''
2952 return YoutubeDL.sanitize_info(info_dict, actually_filter)
2953
2954 def run_pp(self, pp, infodict):
2955 files_to_delete = []
2956 if '__files_to_move' not in infodict:
2957 infodict['__files_to_move'] = {}
2958 files_to_delete, infodict = pp.run(infodict)
2959 if not files_to_delete:
2960 return infodict
2961
2962 if self.params.get('keepvideo', False):
2963 for f in files_to_delete:
2964 infodict['__files_to_move'].setdefault(f, '')
2965 else:
2966 for old_filename in set(files_to_delete):
2967 self.to_screen('Deleting original file %s (pass -k to keep)' % old_filename)
2968 try:
2969 os.remove(encodeFilename(old_filename))
2970 except (IOError, OSError):
2971 self.report_warning('Unable to remove downloaded original file')
2972 if old_filename in infodict['__files_to_move']:
2973 del infodict['__files_to_move'][old_filename]
2974 return infodict
2975
2976 @staticmethod
2977 def post_extract(info_dict):
2978 def actual_post_extract(info_dict):
2979 if info_dict.get('_type') in ('playlist', 'multi_video'):
2980 for video_dict in info_dict.get('entries', {}):
2981 actual_post_extract(video_dict or {})
2982 return
2983
2984 post_extractor = info_dict.get('__post_extractor') or (lambda: {})
2985 extra = post_extractor().items()
2986 info_dict.update(extra)
2987 info_dict.pop('__post_extractor', None)
2988
2989 original_infodict = info_dict.get('__original_infodict') or {}
2990 original_infodict.update(extra)
2991 original_infodict.pop('__post_extractor', None)
2992
2993 actual_post_extract(info_dict or {})
2994
2995 def pre_process(self, ie_info, key='pre_process', files_to_move=None):
2996 info = dict(ie_info)
2997 info['__files_to_move'] = files_to_move or {}
2998 for pp in self._pps[key]:
2999 info = self.run_pp(pp, info)
3000 return info, info.pop('__files_to_move', None)
3001
3002 def post_process(self, filename, ie_info, files_to_move=None):
3003 """Run all the postprocessors on the given file."""
3004 info = dict(ie_info)
3005 info['filepath'] = filename
3006 info['__files_to_move'] = files_to_move or {}
3007
3008 for pp in ie_info.get('__postprocessors', []) + self._pps['post_process']:
3009 info = self.run_pp(pp, info)
3010 info = self.run_pp(MoveFilesAfterDownloadPP(self), info)
3011 del info['__files_to_move']
3012 for pp in self._pps['after_move']:
3013 info = self.run_pp(pp, info)
3014 return info
3015
3016 def _make_archive_id(self, info_dict):
3017 video_id = info_dict.get('id')
3018 if not video_id:
3019 return
3020 # Future-proof against any change in case
3021 # and backwards compatibility with prior versions
3022 extractor = info_dict.get('extractor_key') or info_dict.get('ie_key') # key in a playlist
3023 if extractor is None:
3024 url = str_or_none(info_dict.get('url'))
3025 if not url:
3026 return
3027 # Try to find matching extractor for the URL and take its ie_key
3028 for ie in self._ies:
3029 if ie.suitable(url):
3030 extractor = ie.ie_key()
3031 break
3032 else:
3033 return
3034 return '%s %s' % (extractor.lower(), video_id)
3035
3036 def in_download_archive(self, info_dict):
3037 fn = self.params.get('download_archive')
3038 if fn is None:
3039 return False
3040
3041 vid_id = self._make_archive_id(info_dict)
3042 if not vid_id:
3043 return False # Incomplete video information
3044
3045 return vid_id in self.archive
3046
3047 def record_download_archive(self, info_dict):
3048 fn = self.params.get('download_archive')
3049 if fn is None:
3050 return
3051 vid_id = self._make_archive_id(info_dict)
3052 assert vid_id
3053 with locked_file(fn, 'a', encoding='utf-8') as archive_file:
3054 archive_file.write(vid_id + '\n')
3055 self.archive.add(vid_id)
3056
3057 @staticmethod
3058 def format_resolution(format, default='unknown'):
3059 if format.get('vcodec') == 'none':
3060 if format.get('acodec') == 'none':
3061 return 'images'
3062 return 'audio only'
3063 if format.get('resolution') is not None:
3064 return format['resolution']
3065 if format.get('width') and format.get('height'):
3066 res = '%dx%d' % (format['width'], format['height'])
3067 elif format.get('height'):
3068 res = '%sp' % format['height']
3069 elif format.get('width'):
3070 res = '%dx?' % format['width']
3071 else:
3072 res = default
3073 return res
3074
3075 def _format_note(self, fdict):
3076 res = ''
3077 if fdict.get('ext') in ['f4f', 'f4m']:
3078 res += '(unsupported) '
3079 if fdict.get('language'):
3080 if res:
3081 res += ' '
3082 res += '[%s] ' % fdict['language']
3083 if fdict.get('format_note') is not None:
3084 res += fdict['format_note'] + ' '
3085 if fdict.get('tbr') is not None:
3086 res += '%4dk ' % fdict['tbr']
3087 if fdict.get('container') is not None:
3088 if res:
3089 res += ', '
3090 res += '%s container' % fdict['container']
3091 if (fdict.get('vcodec') is not None
3092 and fdict.get('vcodec') != 'none'):
3093 if res:
3094 res += ', '
3095 res += fdict['vcodec']
3096 if fdict.get('vbr') is not None:
3097 res += '@'
3098 elif fdict.get('vbr') is not None and fdict.get('abr') is not None:
3099 res += 'video@'
3100 if fdict.get('vbr') is not None:
3101 res += '%4dk' % fdict['vbr']
3102 if fdict.get('fps') is not None:
3103 if res:
3104 res += ', '
3105 res += '%sfps' % fdict['fps']
3106 if fdict.get('acodec') is not None:
3107 if res:
3108 res += ', '
3109 if fdict['acodec'] == 'none':
3110 res += 'video only'
3111 else:
3112 res += '%-5s' % fdict['acodec']
3113 elif fdict.get('abr') is not None:
3114 if res:
3115 res += ', '
3116 res += 'audio'
3117 if fdict.get('abr') is not None:
3118 res += '@%3dk' % fdict['abr']
3119 if fdict.get('asr') is not None:
3120 res += ' (%5dHz)' % fdict['asr']
3121 if fdict.get('filesize') is not None:
3122 if res:
3123 res += ', '
3124 res += format_bytes(fdict['filesize'])
3125 elif fdict.get('filesize_approx') is not None:
3126 if res:
3127 res += ', '
3128 res += '~' + format_bytes(fdict['filesize_approx'])
3129 return res
3130
3131 def list_formats(self, info_dict):
3132 formats = info_dict.get('formats', [info_dict])
3133 new_format = (
3134 'list-formats' not in self.params.get('compat_opts', [])
3135 and self.params.get('listformats_table', True) is not False)
3136 if new_format:
3137 table = [
3138 [
3139 format_field(f, 'format_id'),
3140 format_field(f, 'ext'),
3141 self.format_resolution(f),
3142 format_field(f, 'fps', '%d'),
3143 '|',
3144 format_field(f, 'filesize', ' %s', func=format_bytes) + format_field(f, 'filesize_approx', '~%s', func=format_bytes),
3145 format_field(f, 'tbr', '%4dk'),
3146 shorten_protocol_name(f.get('protocol', '').replace("native", "n")),
3147 '|',
3148 format_field(f, 'vcodec', default='unknown').replace('none', ''),
3149 format_field(f, 'vbr', '%4dk'),
3150 format_field(f, 'acodec', default='unknown').replace('none', ''),
3151 format_field(f, 'abr', '%3dk'),
3152 format_field(f, 'asr', '%5dHz'),
3153 ', '.join(filter(None, (
3154 'UNSUPPORTED' if f.get('ext') in ('f4f', 'f4m') else '',
3155 format_field(f, 'language', '[%s]'),
3156 format_field(f, 'format_note'),
3157 format_field(f, 'container', ignore=(None, f.get('ext'))),
3158 ))),
3159 ] for f in formats if f.get('preference') is None or f['preference'] >= -1000]
3160 header_line = ['ID', 'EXT', 'RESOLUTION', 'FPS', '|', ' FILESIZE', ' TBR', 'PROTO',
3161 '|', 'VCODEC', ' VBR', 'ACODEC', ' ABR', ' ASR', 'MORE INFO']
3162 else:
3163 table = [
3164 [
3165 format_field(f, 'format_id'),
3166 format_field(f, 'ext'),
3167 self.format_resolution(f),
3168 self._format_note(f)]
3169 for f in formats
3170 if f.get('preference') is None or f['preference'] >= -1000]
3171 header_line = ['format code', 'extension', 'resolution', 'note']
3172
3173 self.to_screen(
3174 '[info] Available formats for %s:' % info_dict['id'])
3175 self.to_stdout(render_table(
3176 header_line, table, delim=new_format, extraGap=(0 if new_format else 1), hideEmpty=new_format))
3177
3178 def list_thumbnails(self, info_dict):
3179 thumbnails = list(info_dict.get('thumbnails'))
3180 if not thumbnails:
3181 self.to_screen('[info] No thumbnails present for %s' % info_dict['id'])
3182 return
3183
3184 self.to_screen(
3185 '[info] Thumbnails for %s:' % info_dict['id'])
3186 self.to_stdout(render_table(
3187 ['ID', 'width', 'height', 'URL'],
3188 [[t['id'], t.get('width', 'unknown'), t.get('height', 'unknown'), t['url']] for t in thumbnails]))
3189
3190 def list_subtitles(self, video_id, subtitles, name='subtitles'):
3191 if not subtitles:
3192 self.to_screen('%s has no %s' % (video_id, name))
3193 return
3194 self.to_screen(
3195 'Available %s for %s:' % (name, video_id))
3196
3197 def _row(lang, formats):
3198 exts, names = zip(*((f['ext'], f.get('name') or 'unknown') for f in reversed(formats)))
3199 if len(set(names)) == 1:
3200 names = [] if names[0] == 'unknown' else names[:1]
3201 return [lang, ', '.join(names), ', '.join(exts)]
3202
3203 self.to_stdout(render_table(
3204 ['Language', 'Name', 'Formats'],
3205 [_row(lang, formats) for lang, formats in subtitles.items()],
3206 hideEmpty=True))
3207
3208 def urlopen(self, req):
3209 """ Start an HTTP download """
3210 if isinstance(req, compat_basestring):
3211 req = sanitized_Request(req)
3212 return self._opener.open(req, timeout=self._socket_timeout)
3213
3214 def print_debug_header(self):
3215 if not self.params.get('verbose'):
3216 return
3217
3218 stdout_encoding = getattr(
3219 sys.stdout, 'encoding', 'missing (%s)' % type(sys.stdout).__name__)
3220 encoding_str = (
3221 '[debug] Encodings: locale %s, fs %s, out %s, pref %s\n' % (
3222 locale.getpreferredencoding(),
3223 sys.getfilesystemencoding(),
3224 stdout_encoding,
3225 self.get_encoding()))
3226 write_string(encoding_str, encoding=None)
3227
3228 source = (
3229 '(exe)' if hasattr(sys, 'frozen')
3230 else '(zip)' if isinstance(globals().get('__loader__'), zipimporter)
3231 else '(source)' if os.path.basename(sys.argv[0]) == '__main__.py'
3232 else '')
3233 self._write_string('[debug] yt-dlp version %s %s\n' % (__version__, source))
3234 if _LAZY_LOADER:
3235 self._write_string('[debug] Lazy loading extractors enabled\n')
3236 if _PLUGIN_CLASSES:
3237 self._write_string(
3238 '[debug] Plugin Extractors: %s\n' % [ie.ie_key() for ie in _PLUGIN_CLASSES])
3239 if self.params.get('compat_opts'):
3240 self._write_string(
3241 '[debug] Compatibility options: %s\n' % ', '.join(self.params.get('compat_opts')))
3242 try:
3243 sp = subprocess.Popen(
3244 ['git', 'rev-parse', '--short', 'HEAD'],
3245 stdout=subprocess.PIPE, stderr=subprocess.PIPE,
3246 cwd=os.path.dirname(os.path.abspath(__file__)))
3247 out, err = process_communicate_or_kill(sp)
3248 out = out.decode().strip()
3249 if re.match('[0-9a-f]+', out):
3250 self._write_string('[debug] Git HEAD: %s\n' % out)
3251 except Exception:
3252 try:
3253 sys.exc_clear()
3254 except Exception:
3255 pass
3256
3257 def python_implementation():
3258 impl_name = platform.python_implementation()
3259 if impl_name == 'PyPy' and hasattr(sys, 'pypy_version_info'):
3260 return impl_name + ' version %d.%d.%d' % sys.pypy_version_info[:3]
3261 return impl_name
3262
3263 self._write_string('[debug] Python version %s (%s %s) - %s\n' % (
3264 platform.python_version(),
3265 python_implementation(),
3266 platform.architecture()[0],
3267 platform_name()))
3268
3269 exe_versions = FFmpegPostProcessor.get_versions(self)
3270 exe_versions['rtmpdump'] = rtmpdump_version()
3271 exe_versions['phantomjs'] = PhantomJSwrapper._version()
3272 exe_str = ', '.join(
3273 f'{exe} {v}' for exe, v in sorted(exe_versions.items()) if v
3274 ) or 'none'
3275 self._write_string('[debug] exe versions: %s\n' % exe_str)
3276
3277 from .downloader.fragment import can_decrypt_frag
3278 from .downloader.websocket import has_websockets
3279 from .postprocessor.embedthumbnail import has_mutagen
3280 from .cookies import SQLITE_AVAILABLE, KEYRING_AVAILABLE
3281
3282 lib_str = ', '.join(sorted(filter(None, (
3283 can_decrypt_frag and 'pycryptodome',
3284 has_websockets and 'websockets',
3285 has_mutagen and 'mutagen',
3286 SQLITE_AVAILABLE and 'sqlite',
3287 KEYRING_AVAILABLE and 'keyring',
3288 )))) or 'none'
3289 self._write_string('[debug] Optional libraries: %s\n' % lib_str)
3290
3291 proxy_map = {}
3292 for handler in self._opener.handlers:
3293 if hasattr(handler, 'proxies'):
3294 proxy_map.update(handler.proxies)
3295 self._write_string('[debug] Proxy map: ' + compat_str(proxy_map) + '\n')
3296
3297 if self.params.get('call_home', False):
3298 ipaddr = self.urlopen('https://yt-dl.org/ip').read().decode('utf-8')
3299 self._write_string('[debug] Public IP address: %s\n' % ipaddr)
3300 return
3301 latest_version = self.urlopen(
3302 'https://yt-dl.org/latest/version').read().decode('utf-8')
3303 if version_tuple(latest_version) > version_tuple(__version__):
3304 self.report_warning(
3305 'You are using an outdated version (newest version: %s)! '
3306 'See https://yt-dl.org/update if you need help updating.' %
3307 latest_version)
3308
3309 def _setup_opener(self):
3310 timeout_val = self.params.get('socket_timeout')
3311 self._socket_timeout = 600 if timeout_val is None else float(timeout_val)
3312
3313 opts_cookiesfrombrowser = self.params.get('cookiesfrombrowser')
3314 opts_cookiefile = self.params.get('cookiefile')
3315 opts_proxy = self.params.get('proxy')
3316
3317 self.cookiejar = load_cookies(opts_cookiefile, opts_cookiesfrombrowser, self)
3318
3319 cookie_processor = YoutubeDLCookieProcessor(self.cookiejar)
3320 if opts_proxy is not None:
3321 if opts_proxy == '':
3322 proxies = {}
3323 else:
3324 proxies = {'http': opts_proxy, 'https': opts_proxy}
3325 else:
3326 proxies = compat_urllib_request.getproxies()
3327 # Set HTTPS proxy to HTTP one if given (https://github.com/ytdl-org/youtube-dl/issues/805)
3328 if 'http' in proxies and 'https' not in proxies:
3329 proxies['https'] = proxies['http']
3330 proxy_handler = PerRequestProxyHandler(proxies)
3331
3332 debuglevel = 1 if self.params.get('debug_printtraffic') else 0
3333 https_handler = make_HTTPS_handler(self.params, debuglevel=debuglevel)
3334 ydlh = YoutubeDLHandler(self.params, debuglevel=debuglevel)
3335 redirect_handler = YoutubeDLRedirectHandler()
3336 data_handler = compat_urllib_request_DataHandler()
3337
3338 # When passing our own FileHandler instance, build_opener won't add the
3339 # default FileHandler and allows us to disable the file protocol, which
3340 # can be used for malicious purposes (see
3341 # https://github.com/ytdl-org/youtube-dl/issues/8227)
3342 file_handler = compat_urllib_request.FileHandler()
3343
3344 def file_open(*args, **kwargs):
3345 raise compat_urllib_error.URLError('file:// scheme is explicitly disabled in yt-dlp for security reasons')
3346 file_handler.file_open = file_open
3347
3348 opener = compat_urllib_request.build_opener(
3349 proxy_handler, https_handler, cookie_processor, ydlh, redirect_handler, data_handler, file_handler)
3350
3351 # Delete the default user-agent header, which would otherwise apply in
3352 # cases where our custom HTTP handler doesn't come into play
3353 # (See https://github.com/ytdl-org/youtube-dl/issues/1309 for details)
3354 opener.addheaders = []
3355 self._opener = opener
3356
3357 def encode(self, s):
3358 if isinstance(s, bytes):
3359 return s # Already encoded
3360
3361 try:
3362 return s.encode(self.get_encoding())
3363 except UnicodeEncodeError as err:
3364 err.reason = err.reason + '. Check your system encoding configuration or use the --encoding option.'
3365 raise
3366
3367 def get_encoding(self):
3368 encoding = self.params.get('encoding')
3369 if encoding is None:
3370 encoding = preferredencoding()
3371 return encoding
3372
3373 def _write_thumbnails(self, info_dict, filename): # return the extensions
3374 write_all = self.params.get('write_all_thumbnails', False)
3375 thumbnails = []
3376 if write_all or self.params.get('writethumbnail', False):
3377 thumbnails = info_dict.get('thumbnails') or []
3378 multiple = write_all and len(thumbnails) > 1
3379
3380 ret = []
3381 for t in thumbnails[::-1]:
3382 thumb_ext = determine_ext(t['url'], 'jpg')
3383 suffix = '%s.' % t['id'] if multiple else ''
3384 thumb_display_id = '%s ' % t['id'] if multiple else ''
3385 thumb_filename = replace_extension(filename, suffix + thumb_ext, info_dict.get('ext'))
3386
3387 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(thumb_filename)):
3388 ret.append(suffix + thumb_ext)
3389 t['filepath'] = thumb_filename
3390 self.to_screen('[%s] %s: Thumbnail %sis already present' %
3391 (info_dict['extractor'], info_dict['id'], thumb_display_id))
3392 else:
3393 self.to_screen('[%s] %s: Downloading thumbnail %s ...' %
3394 (info_dict['extractor'], info_dict['id'], thumb_display_id))
3395 try:
3396 uf = self.urlopen(t['url'])
3397 with open(encodeFilename(thumb_filename), 'wb') as thumbf:
3398 shutil.copyfileobj(uf, thumbf)
3399 ret.append(suffix + thumb_ext)
3400 self.to_screen('[%s] %s: Writing thumbnail %sto: %s' %
3401 (info_dict['extractor'], info_dict['id'], thumb_display_id, thumb_filename))
3402 t['filepath'] = thumb_filename
3403 except network_exceptions as err:
3404 self.report_warning('Unable to download thumbnail "%s": %s' %
3405 (t['url'], error_to_compat_str(err)))
3406 if ret and not write_all:
3407 break
3408 return ret