]> jfr.im git - yt-dlp.git/blob - youtube_dlc/YoutubeDL.py
Release 2021.02.19
[yt-dlp.git] / youtube_dlc / YoutubeDL.py
1 #!/usr/bin/env python
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 socket
23 import sys
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_cookiejar,
35 compat_get_terminal_size,
36 compat_http_client,
37 compat_kwargs,
38 compat_numeric_types,
39 compat_os_name,
40 compat_str,
41 compat_tokenize_tokenize,
42 compat_urllib_error,
43 compat_urllib_request,
44 compat_urllib_request_DataHandler,
45 )
46 from .utils import (
47 age_restricted,
48 args_to_str,
49 ContentTooShortError,
50 date_from_str,
51 DateRange,
52 DEFAULT_OUTTMPL,
53 OUTTMPL_TYPES,
54 determine_ext,
55 determine_protocol,
56 DOT_DESKTOP_LINK_TEMPLATE,
57 DOT_URL_LINK_TEMPLATE,
58 DOT_WEBLOC_LINK_TEMPLATE,
59 DownloadError,
60 encode_compat_str,
61 encodeFilename,
62 error_to_compat_str,
63 ExistingVideoReached,
64 expand_path,
65 ExtractorError,
66 float_or_none,
67 format_bytes,
68 format_field,
69 formatSeconds,
70 GeoRestrictedError,
71 int_or_none,
72 iri_to_uri,
73 ISO3166Utils,
74 locked_file,
75 make_dir,
76 make_HTTPS_handler,
77 MaxDownloadsReached,
78 orderedSet,
79 PagedList,
80 parse_filesize,
81 PerRequestProxyHandler,
82 platform_name,
83 PostProcessingError,
84 preferredencoding,
85 prepend_extension,
86 register_socks_protocols,
87 render_table,
88 replace_extension,
89 RejectedVideoReached,
90 SameFileError,
91 sanitize_filename,
92 sanitize_path,
93 sanitize_url,
94 sanitized_Request,
95 std_headers,
96 str_or_none,
97 strftime_or_none,
98 subtitles_filename,
99 to_high_limit_path,
100 UnavailableVideoError,
101 url_basename,
102 version_tuple,
103 write_json_file,
104 write_string,
105 YoutubeDLCookieJar,
106 YoutubeDLCookieProcessor,
107 YoutubeDLHandler,
108 YoutubeDLRedirectHandler,
109 process_communicate_or_kill,
110 )
111 from .cache import Cache
112 from .extractor import get_info_extractor, gen_extractor_classes, _LAZY_LOADER, _PLUGIN_CLASSES
113 from .extractor.openload import PhantomJSwrapper
114 from .downloader import get_suitable_downloader
115 from .downloader.rtmp import rtmpdump_version
116 from .postprocessor import (
117 FFmpegFixupM3u8PP,
118 FFmpegFixupM4aPP,
119 FFmpegFixupStretchedPP,
120 FFmpegMergerPP,
121 FFmpegPostProcessor,
122 # FFmpegSubtitlesConvertorPP,
123 get_postprocessor,
124 MoveFilesAfterDownloadPP,
125 )
126 from .version import __version__
127
128 if compat_os_name == 'nt':
129 import ctypes
130
131
132 class YoutubeDL(object):
133 """YoutubeDL class.
134
135 YoutubeDL objects are the ones responsible of downloading the
136 actual video file and writing it to disk if the user has requested
137 it, among some other tasks. In most cases there should be one per
138 program. As, given a video URL, the downloader doesn't know how to
139 extract all the needed information, task that InfoExtractors do, it
140 has to pass the URL to one of them.
141
142 For this, YoutubeDL objects have a method that allows
143 InfoExtractors to be registered in a given order. When it is passed
144 a URL, the YoutubeDL object handles it to the first InfoExtractor it
145 finds that reports being able to handle it. The InfoExtractor extracts
146 all the information about the video or videos the URL refers to, and
147 YoutubeDL process the extracted information, possibly using a File
148 Downloader to download the video.
149
150 YoutubeDL objects accept a lot of parameters. In order not to saturate
151 the object constructor with arguments, it receives a dictionary of
152 options instead. These options are available through the params
153 attribute for the InfoExtractors to use. The YoutubeDL also
154 registers itself as the downloader in charge for the InfoExtractors
155 that are added to it, so this is a "mutual registration".
156
157 Available options:
158
159 username: Username for authentication purposes.
160 password: Password for authentication purposes.
161 videopassword: Password for accessing a video.
162 ap_mso: Adobe Pass multiple-system operator identifier.
163 ap_username: Multiple-system operator account username.
164 ap_password: Multiple-system operator account password.
165 usenetrc: Use netrc for authentication instead.
166 verbose: Print additional info to stdout.
167 quiet: Do not print messages to stdout.
168 no_warnings: Do not print out anything for warnings.
169 forceurl: Force printing final URL.
170 forcetitle: Force printing title.
171 forceid: Force printing ID.
172 forcethumbnail: Force printing thumbnail URL.
173 forcedescription: Force printing description.
174 forcefilename: Force printing final filename.
175 forceduration: Force printing duration.
176 forcejson: Force printing info_dict as JSON.
177 dump_single_json: Force printing the info_dict of the whole playlist
178 (or video) as a single JSON line.
179 force_write_download_archive: Force writing download archive regardless
180 of 'skip_download' or 'simulate'.
181 simulate: Do not download the video files.
182 format: Video format code. see "FORMAT SELECTION" for more details.
183 allow_unplayable_formats: Allow unplayable formats to be extracted and downloaded.
184 format_sort: How to sort the video formats. see "Sorting Formats"
185 for more details.
186 format_sort_force: Force the given format_sort. see "Sorting Formats"
187 for more details.
188 allow_multiple_video_streams: Allow multiple video streams to be merged
189 into a single file
190 allow_multiple_audio_streams: Allow multiple audio streams to be merged
191 into a single file
192 paths: Dictionary of output paths. The allowed keys are 'home'
193 'temp' and the keys of OUTTMPL_TYPES (in utils.py)
194 outtmpl: Dictionary of templates for output names. Allowed keys
195 are 'default' and the keys of OUTTMPL_TYPES (in utils.py).
196 A string a also accepted for backward compatibility
197 outtmpl_na_placeholder: Placeholder for unavailable meta fields.
198 restrictfilenames: Do not allow "&" and spaces in file names
199 trim_file_name: Limit length of filename (extension excluded)
200 windowsfilenames: Force the filenames to be windows compatible
201 ignoreerrors: Do not stop on download errors
202 (Default True when running youtube-dlc,
203 but False when directly accessing YoutubeDL class)
204 force_generic_extractor: Force downloader to use the generic extractor
205 overwrites: Overwrite all video and metadata files if True,
206 overwrite only non-video files if None
207 and don't overwrite any file if False
208 playliststart: Playlist item to start at.
209 playlistend: Playlist item to end at.
210 playlist_items: Specific indices of playlist to download.
211 playlistreverse: Download playlist items in reverse order.
212 playlistrandom: Download playlist items in random order.
213 matchtitle: Download only matching titles.
214 rejecttitle: Reject downloads for matching titles.
215 logger: Log messages to a logging.Logger instance.
216 logtostderr: Log messages to stderr instead of stdout.
217 writedescription: Write the video description to a .description file
218 writeinfojson: Write the video description to a .info.json file
219 writecomments: Extract video comments. This will not be written to disk
220 unless writeinfojson is also given
221 writeannotations: Write the video annotations to a .annotations.xml file
222 writethumbnail: Write the thumbnail image to a file
223 allow_playlist_files: Whether to write playlists' description, infojson etc
224 also to disk when using the 'write*' options
225 write_all_thumbnails: Write all thumbnail formats to files
226 writelink: Write an internet shortcut file, depending on the
227 current platform (.url/.webloc/.desktop)
228 writeurllink: Write a Windows internet shortcut file (.url)
229 writewebloclink: Write a macOS internet shortcut file (.webloc)
230 writedesktoplink: Write a Linux internet shortcut file (.desktop)
231 writesubtitles: Write the video subtitles to a file
232 writeautomaticsub: Write the automatically generated subtitles to a file
233 allsubtitles: Downloads all the subtitles of the video
234 (requires writesubtitles or writeautomaticsub)
235 listsubtitles: Lists all available subtitles for the video
236 subtitlesformat: The format code for subtitles
237 subtitleslangs: List of languages of the subtitles to download
238 keepvideo: Keep the video file after post-processing
239 daterange: A DateRange object, download only if the upload_date is in the range.
240 skip_download: Skip the actual download of the video file
241 cachedir: Location of the cache files in the filesystem.
242 False to disable filesystem cache.
243 noplaylist: Download single video instead of a playlist if in doubt.
244 age_limit: An integer representing the user's age in years.
245 Unsuitable videos for the given age are skipped.
246 min_views: An integer representing the minimum view count the video
247 must have in order to not be skipped.
248 Videos without view count information are always
249 downloaded. None for no limit.
250 max_views: An integer representing the maximum view count.
251 Videos that are more popular than that are not
252 downloaded.
253 Videos without view count information are always
254 downloaded. None for no limit.
255 download_archive: File name of a file where all downloads are recorded.
256 Videos already present in the file are not downloaded
257 again.
258 break_on_existing: Stop the download process after attempting to download a
259 file that is in the archive.
260 break_on_reject: Stop the download process when encountering a video that
261 has been filtered out.
262 cookiefile: File name where cookies should be read from and dumped to
263 nocheckcertificate:Do not verify SSL certificates
264 prefer_insecure: Use HTTP instead of HTTPS to retrieve information.
265 At the moment, this is only supported by YouTube.
266 proxy: URL of the proxy server to use
267 geo_verification_proxy: URL of the proxy to use for IP address verification
268 on geo-restricted sites.
269 socket_timeout: Time to wait for unresponsive hosts, in seconds
270 bidi_workaround: Work around buggy terminals without bidirectional text
271 support, using fridibi
272 debug_printtraffic:Print out sent and received HTTP traffic
273 include_ads: Download ads as well
274 default_search: Prepend this string if an input url is not valid.
275 'auto' for elaborate guessing
276 encoding: Use this encoding instead of the system-specified.
277 extract_flat: Do not resolve URLs, return the immediate result.
278 Pass in 'in_playlist' to only show this behavior for
279 playlist items.
280 postprocessors: A list of dictionaries, each with an entry
281 * key: The name of the postprocessor. See
282 youtube_dlc/postprocessor/__init__.py for a list.
283 * _after_move: Optional. If True, run this post_processor
284 after 'MoveFilesAfterDownload'
285 as well as any further keyword arguments for the
286 postprocessor.
287 post_hooks: A list of functions that get called as the final step
288 for each video file, after all postprocessors have been
289 called. The filename will be passed as the only argument.
290 progress_hooks: A list of functions that get called on download
291 progress, with a dictionary with the entries
292 * status: One of "downloading", "error", or "finished".
293 Check this first and ignore unknown values.
294
295 If status is one of "downloading", or "finished", the
296 following properties may also be present:
297 * filename: The final filename (always present)
298 * tmpfilename: The filename we're currently writing to
299 * downloaded_bytes: Bytes on disk
300 * total_bytes: Size of the whole file, None if unknown
301 * total_bytes_estimate: Guess of the eventual file size,
302 None if unavailable.
303 * elapsed: The number of seconds since download started.
304 * eta: The estimated time in seconds, None if unknown
305 * speed: The download speed in bytes/second, None if
306 unknown
307 * fragment_index: The counter of the currently
308 downloaded video fragment.
309 * fragment_count: The number of fragments (= individual
310 files that will be merged)
311
312 Progress hooks are guaranteed to be called at least once
313 (with status "finished") if the download is successful.
314 merge_output_format: Extension to use when merging formats.
315 final_ext: Expected final extension; used to detect when the file was
316 already downloaded and converted. "merge_output_format" is
317 replaced by this extension when given
318 fixup: Automatically correct known faults of the file.
319 One of:
320 - "never": do nothing
321 - "warn": only emit a warning
322 - "detect_or_warn": check whether we can do anything
323 about it, warn otherwise (default)
324 source_address: Client-side IP address to bind to.
325 call_home: Boolean, true iff we are allowed to contact the
326 youtube-dlc servers for debugging.
327 sleep_interval: Number of seconds to sleep before each download when
328 used alone or a lower bound of a range for randomized
329 sleep before each download (minimum possible number
330 of seconds to sleep) when used along with
331 max_sleep_interval.
332 max_sleep_interval:Upper bound of a range for randomized sleep before each
333 download (maximum possible number of seconds to sleep).
334 Must only be used along with sleep_interval.
335 Actual sleep time will be a random float from range
336 [sleep_interval; max_sleep_interval].
337 listformats: Print an overview of available video formats and exit.
338 list_thumbnails: Print a table of all thumbnails and exit.
339 match_filter: A function that gets called with the info_dict of
340 every video.
341 If it returns a message, the video is ignored.
342 If it returns None, the video is downloaded.
343 match_filter_func in utils.py is one example for this.
344 no_color: Do not emit color codes in output.
345 geo_bypass: Bypass geographic restriction via faking X-Forwarded-For
346 HTTP header
347 geo_bypass_country:
348 Two-letter ISO 3166-2 country code that will be used for
349 explicit geographic restriction bypassing via faking
350 X-Forwarded-For HTTP header
351 geo_bypass_ip_block:
352 IP range in CIDR notation that will be used similarly to
353 geo_bypass_country
354
355 The following options determine which downloader is picked:
356 external_downloader: Executable of the external downloader to call.
357 None or unset for standard (built-in) downloader.
358 hls_prefer_native: Use the native HLS downloader instead of ffmpeg/avconv
359 if True, otherwise use ffmpeg/avconv if False, otherwise
360 use downloader suggested by extractor if None.
361
362 The following parameters are not used by YoutubeDL itself, they are used by
363 the downloader (see youtube_dlc/downloader/common.py):
364 nopart, updatetime, buffersize, ratelimit, min_filesize, max_filesize, test,
365 noresizebuffer, retries, continuedl, noprogress, consoletitle,
366 xattr_set_filesize, external_downloader_args, hls_use_mpegts,
367 http_chunk_size.
368
369 The following options are used by the post processors:
370 prefer_ffmpeg: If False, use avconv instead of ffmpeg if both are available,
371 otherwise prefer ffmpeg. (avconv support is deprecated)
372 ffmpeg_location: Location of the ffmpeg/avconv binary; either the path
373 to the binary or its containing directory.
374 postprocessor_args: A dictionary of postprocessor/executable keys (in lower case)
375 and a list of additional command-line arguments for the
376 postprocessor/executable. The dict can also have "PP+EXE" keys
377 which are used when the given exe is used by the given PP.
378 Use 'default' as the name for arguments to passed to all PP
379 The following options are used by the Youtube extractor:
380 youtube_include_dash_manifest: If True (default), DASH manifests and related
381 data will be downloaded and processed by extractor.
382 You can reduce network I/O by disabling it if you don't
383 care about DASH.
384 """
385
386 _NUMERIC_FIELDS = set((
387 'width', 'height', 'tbr', 'abr', 'asr', 'vbr', 'fps', 'filesize', 'filesize_approx',
388 'timestamp', 'upload_year', 'upload_month', 'upload_day',
389 'duration', 'view_count', 'like_count', 'dislike_count', 'repost_count',
390 'average_rating', 'comment_count', 'age_limit',
391 'start_time', 'end_time',
392 'chapter_number', 'season_number', 'episode_number',
393 'track_number', 'disc_number', 'release_year',
394 'playlist_index',
395 ))
396
397 params = None
398 _ies = []
399 _pps = {'beforedl': [], 'aftermove': [], 'normal': []}
400 __prepare_filename_warned = False
401 _download_retcode = None
402 _num_downloads = None
403 _playlist_level = 0
404 _playlist_urls = set()
405 _screen_file = None
406
407 def __init__(self, params=None, auto_init=True):
408 """Create a FileDownloader object with the given options."""
409 if params is None:
410 params = {}
411 self._ies = []
412 self._ies_instances = {}
413 self._pps = {'beforedl': [], 'aftermove': [], 'normal': []}
414 self.__prepare_filename_warned = False
415 self._post_hooks = []
416 self._progress_hooks = []
417 self._download_retcode = 0
418 self._num_downloads = 0
419 self._screen_file = [sys.stdout, sys.stderr][params.get('logtostderr', False)]
420 self._err_file = sys.stderr
421 self.params = {
422 # Default parameters
423 'nocheckcertificate': False,
424 }
425 self.params.update(params)
426 self.cache = Cache(self)
427 self.archive = set()
428
429 """Preload the archive, if any is specified"""
430 def preload_download_archive(self):
431 fn = self.params.get('download_archive')
432 if fn is None:
433 return False
434 try:
435 with locked_file(fn, 'r', encoding='utf-8') as archive_file:
436 for line in archive_file:
437 self.archive.add(line.strip())
438 except IOError as ioe:
439 if ioe.errno != errno.ENOENT:
440 raise
441 return False
442 return True
443
444 def check_deprecated(param, option, suggestion):
445 if self.params.get(param) is not None:
446 self.report_warning(
447 '%s is deprecated. Use %s instead.' % (option, suggestion))
448 return True
449 return False
450
451 if self.params.get('verbose'):
452 self.to_stdout('[debug] Loading archive file %r' % self.params.get('download_archive'))
453
454 preload_download_archive(self)
455
456 if check_deprecated('cn_verification_proxy', '--cn-verification-proxy', '--geo-verification-proxy'):
457 if self.params.get('geo_verification_proxy') is None:
458 self.params['geo_verification_proxy'] = self.params['cn_verification_proxy']
459
460 if self.params.get('final_ext'):
461 if self.params.get('merge_output_format'):
462 self.report_warning('--merge-output-format will be ignored since --remux-video or --recode-video is given')
463 self.params['merge_output_format'] = self.params['final_ext']
464
465 if 'overwrites' in self.params and self.params['overwrites'] is None:
466 del self.params['overwrites']
467
468 check_deprecated('autonumber_size', '--autonumber-size', 'output template with %(autonumber)0Nd, where N in the number of digits')
469 check_deprecated('autonumber', '--auto-number', '-o "%(autonumber)s-%(title)s.%(ext)s"')
470 check_deprecated('usetitle', '--title', '-o "%(title)s-%(id)s.%(ext)s"')
471
472 if params.get('bidi_workaround', False):
473 try:
474 import pty
475 master, slave = pty.openpty()
476 width = compat_get_terminal_size().columns
477 if width is None:
478 width_args = []
479 else:
480 width_args = ['-w', str(width)]
481 sp_kwargs = dict(
482 stdin=subprocess.PIPE,
483 stdout=slave,
484 stderr=self._err_file)
485 try:
486 self._output_process = subprocess.Popen(
487 ['bidiv'] + width_args, **sp_kwargs
488 )
489 except OSError:
490 self._output_process = subprocess.Popen(
491 ['fribidi', '-c', 'UTF-8'] + width_args, **sp_kwargs)
492 self._output_channel = os.fdopen(master, 'rb')
493 except OSError as ose:
494 if ose.errno == errno.ENOENT:
495 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 $PATH.')
496 else:
497 raise
498
499 if (sys.platform != 'win32'
500 and sys.getfilesystemencoding() in ['ascii', 'ANSI_X3.4-1968']
501 and not params.get('restrictfilenames', False)):
502 # Unicode filesystem API will throw errors (#1474, #13027)
503 self.report_warning(
504 'Assuming --restrict-filenames since file system encoding '
505 'cannot encode all characters. '
506 'Set the LC_ALL environment variable to fix this.')
507 self.params['restrictfilenames'] = True
508
509 self.outtmpl_dict = self.parse_outtmpl()
510
511 self._setup_opener()
512
513 if auto_init:
514 self.print_debug_header()
515 self.add_default_info_extractors()
516
517 for pp_def_raw in self.params.get('postprocessors', []):
518 pp_class = get_postprocessor(pp_def_raw['key'])
519 pp_def = dict(pp_def_raw)
520 del pp_def['key']
521 if 'when' in pp_def:
522 when = pp_def['when']
523 del pp_def['when']
524 else:
525 when = 'normal'
526 pp = pp_class(self, **compat_kwargs(pp_def))
527 self.add_post_processor(pp, when=when)
528
529 for ph in self.params.get('post_hooks', []):
530 self.add_post_hook(ph)
531
532 for ph in self.params.get('progress_hooks', []):
533 self.add_progress_hook(ph)
534
535 register_socks_protocols()
536
537 def warn_if_short_id(self, argv):
538 # short YouTube ID starting with dash?
539 idxs = [
540 i for i, a in enumerate(argv)
541 if re.match(r'^-[0-9A-Za-z_-]{10}$', a)]
542 if idxs:
543 correct_argv = (
544 ['youtube-dlc']
545 + [a for i, a in enumerate(argv) if i not in idxs]
546 + ['--'] + [argv[i] for i in idxs]
547 )
548 self.report_warning(
549 'Long argument string detected. '
550 'Use -- to separate parameters and URLs, like this:\n%s\n' %
551 args_to_str(correct_argv))
552
553 def add_info_extractor(self, ie):
554 """Add an InfoExtractor object to the end of the list."""
555 self._ies.append(ie)
556 if not isinstance(ie, type):
557 self._ies_instances[ie.ie_key()] = ie
558 ie.set_downloader(self)
559
560 def get_info_extractor(self, ie_key):
561 """
562 Get an instance of an IE with name ie_key, it will try to get one from
563 the _ies list, if there's no instance it will create a new one and add
564 it to the extractor list.
565 """
566 ie = self._ies_instances.get(ie_key)
567 if ie is None:
568 ie = get_info_extractor(ie_key)()
569 self.add_info_extractor(ie)
570 return ie
571
572 def add_default_info_extractors(self):
573 """
574 Add the InfoExtractors returned by gen_extractors to the end of the list
575 """
576 for ie in gen_extractor_classes():
577 self.add_info_extractor(ie)
578
579 def add_post_processor(self, pp, when='normal'):
580 """Add a PostProcessor object to the end of the chain."""
581 self._pps[when].append(pp)
582 pp.set_downloader(self)
583
584 def add_post_hook(self, ph):
585 """Add the post hook"""
586 self._post_hooks.append(ph)
587
588 def add_progress_hook(self, ph):
589 """Add the progress hook (currently only for the file downloader)"""
590 self._progress_hooks.append(ph)
591
592 def _bidi_workaround(self, message):
593 if not hasattr(self, '_output_channel'):
594 return message
595
596 assert hasattr(self, '_output_process')
597 assert isinstance(message, compat_str)
598 line_count = message.count('\n') + 1
599 self._output_process.stdin.write((message + '\n').encode('utf-8'))
600 self._output_process.stdin.flush()
601 res = ''.join(self._output_channel.readline().decode('utf-8')
602 for _ in range(line_count))
603 return res[:-len('\n')]
604
605 def to_screen(self, message, skip_eol=False):
606 """Print message to stdout if not in quiet mode."""
607 return self.to_stdout(message, skip_eol, check_quiet=True)
608
609 def _write_string(self, s, out=None):
610 write_string(s, out=out, encoding=self.params.get('encoding'))
611
612 def to_stdout(self, message, skip_eol=False, check_quiet=False):
613 """Print message to stdout if not in quiet mode."""
614 if self.params.get('logger'):
615 self.params['logger'].debug(message)
616 elif not check_quiet or not self.params.get('quiet', False):
617 message = self._bidi_workaround(message)
618 terminator = ['\n', ''][skip_eol]
619 output = message + terminator
620
621 self._write_string(output, self._screen_file)
622
623 def to_stderr(self, message):
624 """Print message to stderr."""
625 assert isinstance(message, compat_str)
626 if self.params.get('logger'):
627 self.params['logger'].error(message)
628 else:
629 message = self._bidi_workaround(message)
630 output = message + '\n'
631 self._write_string(output, self._err_file)
632
633 def to_console_title(self, message):
634 if not self.params.get('consoletitle', False):
635 return
636 if compat_os_name == 'nt':
637 if ctypes.windll.kernel32.GetConsoleWindow():
638 # c_wchar_p() might not be necessary if `message` is
639 # already of type unicode()
640 ctypes.windll.kernel32.SetConsoleTitleW(ctypes.c_wchar_p(message))
641 elif 'TERM' in os.environ:
642 self._write_string('\033]0;%s\007' % message, self._screen_file)
643
644 def save_console_title(self):
645 if not self.params.get('consoletitle', False):
646 return
647 if self.params.get('simulate', False):
648 return
649 if compat_os_name != 'nt' and 'TERM' in os.environ:
650 # Save the title on stack
651 self._write_string('\033[22;0t', self._screen_file)
652
653 def restore_console_title(self):
654 if not self.params.get('consoletitle', False):
655 return
656 if self.params.get('simulate', False):
657 return
658 if compat_os_name != 'nt' and 'TERM' in os.environ:
659 # Restore the title from stack
660 self._write_string('\033[23;0t', self._screen_file)
661
662 def __enter__(self):
663 self.save_console_title()
664 return self
665
666 def __exit__(self, *args):
667 self.restore_console_title()
668
669 if self.params.get('cookiefile') is not None:
670 self.cookiejar.save(ignore_discard=True, ignore_expires=True)
671
672 def trouble(self, message=None, tb=None):
673 """Determine action to take when a download problem appears.
674
675 Depending on if the downloader has been configured to ignore
676 download errors or not, this method may throw an exception or
677 not when errors are found, after printing the message.
678
679 tb, if given, is additional traceback information.
680 """
681 if message is not None:
682 self.to_stderr(message)
683 if self.params.get('verbose'):
684 if tb is None:
685 if sys.exc_info()[0]: # if .trouble has been called from an except block
686 tb = ''
687 if hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
688 tb += ''.join(traceback.format_exception(*sys.exc_info()[1].exc_info))
689 tb += encode_compat_str(traceback.format_exc())
690 else:
691 tb_data = traceback.format_list(traceback.extract_stack())
692 tb = ''.join(tb_data)
693 self.to_stderr(tb)
694 if not self.params.get('ignoreerrors', False):
695 if sys.exc_info()[0] and hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
696 exc_info = sys.exc_info()[1].exc_info
697 else:
698 exc_info = sys.exc_info()
699 raise DownloadError(message, exc_info)
700 self._download_retcode = 1
701
702 def report_warning(self, message):
703 '''
704 Print the message to stderr, it will be prefixed with 'WARNING:'
705 If stderr is a tty file the 'WARNING:' will be colored
706 '''
707 if self.params.get('logger') is not None:
708 self.params['logger'].warning(message)
709 else:
710 if self.params.get('no_warnings'):
711 return
712 if not self.params.get('no_color') and self._err_file.isatty() and compat_os_name != 'nt':
713 _msg_header = '\033[0;33mWARNING:\033[0m'
714 else:
715 _msg_header = 'WARNING:'
716 warning_message = '%s %s' % (_msg_header, message)
717 self.to_stderr(warning_message)
718
719 def report_error(self, message, tb=None):
720 '''
721 Do the same as trouble, but prefixes the message with 'ERROR:', colored
722 in red if stderr is a tty file.
723 '''
724 if not self.params.get('no_color') and self._err_file.isatty() and compat_os_name != 'nt':
725 _msg_header = '\033[0;31mERROR:\033[0m'
726 else:
727 _msg_header = 'ERROR:'
728 error_message = '%s %s' % (_msg_header, message)
729 self.trouble(error_message, tb)
730
731 def report_file_already_downloaded(self, file_name):
732 """Report file has already been fully downloaded."""
733 try:
734 self.to_screen('[download] %s has already been downloaded' % file_name)
735 except UnicodeEncodeError:
736 self.to_screen('[download] The file has already been downloaded')
737
738 def report_file_delete(self, file_name):
739 """Report that existing file will be deleted."""
740 try:
741 self.to_screen('Deleting existing file %s' % file_name)
742 except UnicodeEncodeError:
743 self.to_screen('Deleting existing file')
744
745 def parse_outtmpl(self):
746 outtmpl_dict = self.params.get('outtmpl', {})
747 if not isinstance(outtmpl_dict, dict):
748 outtmpl_dict = {'default': outtmpl_dict}
749 outtmpl_dict.update({
750 k: v for k, v in DEFAULT_OUTTMPL.items()
751 if not outtmpl_dict.get(k)})
752 for key, val in outtmpl_dict.items():
753 if isinstance(val, bytes):
754 self.report_warning(
755 'Parameter outtmpl is bytes, but should be a unicode string. '
756 'Put from __future__ import unicode_literals at the top of your code file or consider switching to Python 3.x.')
757 return outtmpl_dict
758
759 def _prepare_filename(self, info_dict, tmpl_type='default'):
760 try:
761 template_dict = dict(info_dict)
762
763 template_dict['duration_string'] = ( # %(duration>%H-%M-%S)s is wrong if duration > 24hrs
764 formatSeconds(info_dict['duration'], '-')
765 if info_dict.get('duration', None) is not None
766 else None)
767
768 template_dict['epoch'] = int(time.time())
769 autonumber_size = self.params.get('autonumber_size')
770 if autonumber_size is None:
771 autonumber_size = 5
772 template_dict['autonumber'] = self.params.get('autonumber_start', 1) - 1 + self._num_downloads
773 if template_dict.get('resolution') is None:
774 if template_dict.get('width') and template_dict.get('height'):
775 template_dict['resolution'] = '%dx%d' % (template_dict['width'], template_dict['height'])
776 elif template_dict.get('height'):
777 template_dict['resolution'] = '%sp' % template_dict['height']
778 elif template_dict.get('width'):
779 template_dict['resolution'] = '%dx?' % template_dict['width']
780
781 sanitize = lambda k, v: sanitize_filename(
782 compat_str(v),
783 restricted=self.params.get('restrictfilenames'),
784 is_id=(k == 'id' or k.endswith('_id')))
785 template_dict = dict((k, v if isinstance(v, compat_numeric_types) else sanitize(k, v))
786 for k, v in template_dict.items()
787 if v is not None and not isinstance(v, (list, tuple, dict)))
788 na = self.params.get('outtmpl_na_placeholder', 'NA')
789 template_dict = collections.defaultdict(lambda: na, template_dict)
790
791 outtmpl = self.outtmpl_dict.get(tmpl_type, self.outtmpl_dict['default'])
792 force_ext = OUTTMPL_TYPES.get(tmpl_type)
793
794 # For fields playlist_index and autonumber convert all occurrences
795 # of %(field)s to %(field)0Nd for backward compatibility
796 field_size_compat_map = {
797 'playlist_index': len(str(template_dict['n_entries'])),
798 'autonumber': autonumber_size,
799 }
800 FIELD_SIZE_COMPAT_RE = r'(?<!%)%\((?P<field>autonumber|playlist_index)\)s'
801 mobj = re.search(FIELD_SIZE_COMPAT_RE, outtmpl)
802 if mobj:
803 outtmpl = re.sub(
804 FIELD_SIZE_COMPAT_RE,
805 r'%%(\1)0%dd' % field_size_compat_map[mobj.group('field')],
806 outtmpl)
807
808 # As of [1] format syntax is:
809 # %[mapping_key][conversion_flags][minimum_width][.precision][length_modifier]type
810 # 1. https://docs.python.org/2/library/stdtypes.html#string-formatting
811 FORMAT_RE = r'''(?x)
812 (?<!%)
813 %
814 \({0}\) # mapping key
815 (?:[#0\-+ ]+)? # conversion flags (optional)
816 (?:\d+)? # minimum field width (optional)
817 (?:\.\d+)? # precision (optional)
818 [hlL]? # length modifier (optional)
819 (?P<type>[diouxXeEfFgGcrs%]) # conversion type
820 '''
821
822 numeric_fields = list(self._NUMERIC_FIELDS)
823
824 # Format date
825 FORMAT_DATE_RE = FORMAT_RE.format(r'(?P<key>(?P<field>\w+)>(?P<format>.+?))')
826 for mobj in re.finditer(FORMAT_DATE_RE, outtmpl):
827 conv_type, field, frmt, key = mobj.group('type', 'field', 'format', 'key')
828 if key in template_dict:
829 continue
830 value = strftime_or_none(template_dict.get(field), frmt, na)
831 if conv_type in 'crs': # string
832 value = sanitize(field, value)
833 else: # number
834 numeric_fields.append(key)
835 value = float_or_none(value, default=None)
836 if value is not None:
837 template_dict[key] = value
838
839 # Missing numeric fields used together with integer presentation types
840 # in format specification will break the argument substitution since
841 # string NA placeholder is returned for missing fields. We will patch
842 # output template for missing fields to meet string presentation type.
843 for numeric_field in numeric_fields:
844 if numeric_field not in template_dict:
845 outtmpl = re.sub(
846 FORMAT_RE.format(re.escape(numeric_field)),
847 r'%({0})s'.format(numeric_field), outtmpl)
848
849 # expand_path translates '%%' into '%' and '$$' into '$'
850 # correspondingly that is not what we want since we need to keep
851 # '%%' intact for template dict substitution step. Working around
852 # with boundary-alike separator hack.
853 sep = ''.join([random.choice(ascii_letters) for _ in range(32)])
854 outtmpl = outtmpl.replace('%%', '%{0}%'.format(sep)).replace('$$', '${0}$'.format(sep))
855
856 # outtmpl should be expand_path'ed before template dict substitution
857 # because meta fields may contain env variables we don't want to
858 # be expanded. For example, for outtmpl "%(title)s.%(ext)s" and
859 # title "Hello $PATH", we don't want `$PATH` to be expanded.
860 filename = expand_path(outtmpl).replace(sep, '') % template_dict
861
862 if force_ext is not None:
863 filename = replace_extension(filename, force_ext, template_dict.get('ext'))
864
865 # https://github.com/blackjack4494/youtube-dlc/issues/85
866 trim_file_name = self.params.get('trim_file_name', False)
867 if trim_file_name:
868 fn_groups = filename.rsplit('.')
869 ext = fn_groups[-1]
870 sub_ext = ''
871 if len(fn_groups) > 2:
872 sub_ext = fn_groups[-2]
873 filename = '.'.join(filter(None, [fn_groups[0][:trim_file_name], sub_ext, ext]))
874
875 return filename
876 except ValueError as err:
877 self.report_error('Error in output template: ' + str(err) + ' (encoding: ' + repr(preferredencoding()) + ')')
878 return None
879
880 def prepare_filename(self, info_dict, dir_type='', warn=False):
881 """Generate the output filename."""
882 paths = self.params.get('paths', {})
883 assert isinstance(paths, dict)
884 filename = self._prepare_filename(info_dict, dir_type or 'default')
885
886 if warn and not self.__prepare_filename_warned:
887 if not paths:
888 pass
889 elif filename == '-':
890 self.report_warning('--paths is ignored when an outputting to stdout')
891 elif os.path.isabs(filename):
892 self.report_warning('--paths is ignored since an absolute path is given in output template')
893 self.__prepare_filename_warned = True
894 if filename == '-' or not filename:
895 return filename
896
897 homepath = expand_path(paths.get('home', '').strip())
898 assert isinstance(homepath, compat_str)
899 subdir = expand_path(paths.get(dir_type, '').strip()) if dir_type else ''
900 assert isinstance(subdir, compat_str)
901 path = os.path.join(homepath, subdir, filename)
902
903 # Temporary fix for #4787
904 # 'Treat' all problem characters by passing filename through preferredencoding
905 # to workaround encoding issues with subprocess on python2 @ Windows
906 if sys.version_info < (3, 0) and sys.platform == 'win32':
907 path = encodeFilename(path, True).decode(preferredencoding())
908 return sanitize_path(path, force=self.params.get('windowsfilenames'))
909
910 def _match_entry(self, info_dict, incomplete):
911 """ Returns None if the file should be downloaded """
912
913 def check_filter():
914 video_title = info_dict.get('title', info_dict.get('id', 'video'))
915 if 'title' in info_dict:
916 # This can happen when we're just evaluating the playlist
917 title = info_dict['title']
918 matchtitle = self.params.get('matchtitle', False)
919 if matchtitle:
920 if not re.search(matchtitle, title, re.IGNORECASE):
921 return '"' + title + '" title did not match pattern "' + matchtitle + '"'
922 rejecttitle = self.params.get('rejecttitle', False)
923 if rejecttitle:
924 if re.search(rejecttitle, title, re.IGNORECASE):
925 return '"' + title + '" title matched reject pattern "' + rejecttitle + '"'
926 date = info_dict.get('upload_date')
927 if date is not None:
928 dateRange = self.params.get('daterange', DateRange())
929 if date not in dateRange:
930 return '%s upload date is not in range %s' % (date_from_str(date).isoformat(), dateRange)
931 view_count = info_dict.get('view_count')
932 if view_count is not None:
933 min_views = self.params.get('min_views')
934 if min_views is not None and view_count < min_views:
935 return 'Skipping %s, because it has not reached minimum view count (%d/%d)' % (video_title, view_count, min_views)
936 max_views = self.params.get('max_views')
937 if max_views is not None and view_count > max_views:
938 return 'Skipping %s, because it has exceeded the maximum view count (%d/%d)' % (video_title, view_count, max_views)
939 if age_restricted(info_dict.get('age_limit'), self.params.get('age_limit')):
940 return 'Skipping "%s" because it is age restricted' % video_title
941 if self.in_download_archive(info_dict):
942 return '%s has already been recorded in archive' % video_title
943
944 if not incomplete:
945 match_filter = self.params.get('match_filter')
946 if match_filter is not None:
947 ret = match_filter(info_dict)
948 if ret is not None:
949 return ret
950 return None
951
952 reason = check_filter()
953 if reason is not None:
954 self.to_screen('[download] ' + reason)
955 if reason.endswith('has already been recorded in the archive') and self.params.get('break_on_existing', False):
956 raise ExistingVideoReached()
957 elif self.params.get('break_on_reject', False):
958 raise RejectedVideoReached()
959 return reason
960
961 @staticmethod
962 def add_extra_info(info_dict, extra_info):
963 '''Set the keys from extra_info in info dict if they are missing'''
964 for key, value in extra_info.items():
965 info_dict.setdefault(key, value)
966
967 def extract_info(self, url, download=True, ie_key=None, info_dict=None, extra_info={},
968 process=True, force_generic_extractor=False):
969 '''
970 Returns a list with a dictionary for each video we find.
971 If 'download', also downloads the videos.
972 extra_info is a dict containing the extra values to add to each result
973 '''
974
975 if not ie_key and force_generic_extractor:
976 ie_key = 'Generic'
977
978 if ie_key:
979 ies = [self.get_info_extractor(ie_key)]
980 else:
981 ies = self._ies
982
983 for ie in ies:
984 if not ie.suitable(url):
985 continue
986
987 ie_key = ie.ie_key()
988 ie = self.get_info_extractor(ie_key)
989 if not ie.working():
990 self.report_warning('The program functionality for this site has been marked as broken, '
991 'and will probably not work.')
992
993 try:
994 temp_id = str_or_none(
995 ie.extract_id(url) if callable(getattr(ie, 'extract_id', None))
996 else ie._match_id(url))
997 except (AssertionError, IndexError, AttributeError):
998 temp_id = None
999 if temp_id is not None and self.in_download_archive({'id': temp_id, 'ie_key': ie_key}):
1000 self.to_screen("[%s] %s: has already been recorded in archive" % (
1001 ie_key, temp_id))
1002 break
1003 return self.__extract_info(url, ie, download, extra_info, process, info_dict)
1004 else:
1005 self.report_error('no suitable InfoExtractor for URL %s' % url)
1006
1007 def __handle_extraction_exceptions(func):
1008 def wrapper(self, *args, **kwargs):
1009 try:
1010 return func(self, *args, **kwargs)
1011 except GeoRestrictedError as e:
1012 msg = e.msg
1013 if e.countries:
1014 msg += '\nThis video is available in %s.' % ', '.join(
1015 map(ISO3166Utils.short2full, e.countries))
1016 msg += '\nYou might want to use a VPN or a proxy server (with --proxy) to workaround.'
1017 self.report_error(msg)
1018 except ExtractorError as e: # An error we somewhat expected
1019 self.report_error(compat_str(e), e.format_traceback())
1020 except (MaxDownloadsReached, ExistingVideoReached, RejectedVideoReached):
1021 raise
1022 except Exception as e:
1023 if self.params.get('ignoreerrors', False):
1024 self.report_error(error_to_compat_str(e), tb=encode_compat_str(traceback.format_exc()))
1025 else:
1026 raise
1027 return wrapper
1028
1029 @__handle_extraction_exceptions
1030 def __extract_info(self, url, ie, download, extra_info, process, info_dict):
1031 ie_result = ie.extract(url)
1032 if ie_result is None: # Finished already (backwards compatibility; listformats and friends should be moved here)
1033 return
1034 if isinstance(ie_result, list):
1035 # Backwards compatibility: old IE result format
1036 ie_result = {
1037 '_type': 'compat_list',
1038 'entries': ie_result,
1039 }
1040 if info_dict:
1041 if info_dict.get('id'):
1042 ie_result['id'] = info_dict['id']
1043 if info_dict.get('title'):
1044 ie_result['title'] = info_dict['title']
1045 self.add_default_extra_info(ie_result, ie, url)
1046 if process:
1047 return self.process_ie_result(ie_result, download, extra_info)
1048 else:
1049 return ie_result
1050
1051 def add_default_extra_info(self, ie_result, ie, url):
1052 self.add_extra_info(ie_result, {
1053 'extractor': ie.IE_NAME,
1054 'webpage_url': url,
1055 'webpage_url_basename': url_basename(url),
1056 'extractor_key': ie.ie_key(),
1057 })
1058
1059 def process_ie_result(self, ie_result, download=True, extra_info={}):
1060 """
1061 Take the result of the ie(may be modified) and resolve all unresolved
1062 references (URLs, playlist items).
1063
1064 It will also download the videos if 'download'.
1065 Returns the resolved ie_result.
1066 """
1067 result_type = ie_result.get('_type', 'video')
1068
1069 if result_type in ('url', 'url_transparent'):
1070 ie_result['url'] = sanitize_url(ie_result['url'])
1071 extract_flat = self.params.get('extract_flat', False)
1072 if ((extract_flat == 'in_playlist' and 'playlist' in extra_info)
1073 or extract_flat is True):
1074 self.__forced_printings(ie_result, self.prepare_filename(ie_result), incomplete=True)
1075 return ie_result
1076
1077 if result_type == 'video':
1078 self.add_extra_info(ie_result, extra_info)
1079 return self.process_video_result(ie_result, download=download)
1080 elif result_type == 'url':
1081 # We have to add extra_info to the results because it may be
1082 # contained in a playlist
1083 return self.extract_info(ie_result['url'],
1084 download, info_dict=ie_result,
1085 ie_key=ie_result.get('ie_key'),
1086 extra_info=extra_info)
1087 elif result_type == 'url_transparent':
1088 # Use the information from the embedding page
1089 info = self.extract_info(
1090 ie_result['url'], ie_key=ie_result.get('ie_key'),
1091 extra_info=extra_info, download=False, process=False)
1092
1093 # extract_info may return None when ignoreerrors is enabled and
1094 # extraction failed with an error, don't crash and return early
1095 # in this case
1096 if not info:
1097 return info
1098
1099 force_properties = dict(
1100 (k, v) for k, v in ie_result.items() if v is not None)
1101 for f in ('_type', 'url', 'id', 'extractor', 'extractor_key', 'ie_key'):
1102 if f in force_properties:
1103 del force_properties[f]
1104 new_result = info.copy()
1105 new_result.update(force_properties)
1106
1107 # Extracted info may not be a video result (i.e.
1108 # info.get('_type', 'video') != video) but rather an url or
1109 # url_transparent. In such cases outer metadata (from ie_result)
1110 # should be propagated to inner one (info). For this to happen
1111 # _type of info should be overridden with url_transparent. This
1112 # fixes issue from https://github.com/ytdl-org/youtube-dl/pull/11163.
1113 if new_result.get('_type') == 'url':
1114 new_result['_type'] = 'url_transparent'
1115
1116 return self.process_ie_result(
1117 new_result, download=download, extra_info=extra_info)
1118 elif result_type in ('playlist', 'multi_video'):
1119 # Protect from infinite recursion due to recursively nested playlists
1120 # (see https://github.com/ytdl-org/youtube-dl/issues/27833)
1121 webpage_url = ie_result['webpage_url']
1122 if webpage_url in self._playlist_urls:
1123 self.to_screen(
1124 '[download] Skipping already downloaded playlist: %s'
1125 % ie_result.get('title') or ie_result.get('id'))
1126 return
1127
1128 self._playlist_level += 1
1129 self._playlist_urls.add(webpage_url)
1130 try:
1131 return self.__process_playlist(ie_result, download)
1132 finally:
1133 self._playlist_level -= 1
1134 if not self._playlist_level:
1135 self._playlist_urls.clear()
1136 elif result_type == 'compat_list':
1137 self.report_warning(
1138 'Extractor %s returned a compat_list result. '
1139 'It needs to be updated.' % ie_result.get('extractor'))
1140
1141 def _fixup(r):
1142 self.add_extra_info(
1143 r,
1144 {
1145 'extractor': ie_result['extractor'],
1146 'webpage_url': ie_result['webpage_url'],
1147 'webpage_url_basename': url_basename(ie_result['webpage_url']),
1148 'extractor_key': ie_result['extractor_key'],
1149 }
1150 )
1151 return r
1152 ie_result['entries'] = [
1153 self.process_ie_result(_fixup(r), download, extra_info)
1154 for r in ie_result['entries']
1155 ]
1156 return ie_result
1157 else:
1158 raise Exception('Invalid result type: %s' % result_type)
1159
1160 def __process_playlist(self, ie_result, download):
1161 # We process each entry in the playlist
1162 playlist = ie_result.get('title') or ie_result.get('id')
1163 self.to_screen('[download] Downloading playlist: %s' % playlist)
1164
1165 if self.params.get('allow_playlist_files', True):
1166 ie_copy = {
1167 'playlist': playlist,
1168 'playlist_id': ie_result.get('id'),
1169 'playlist_title': ie_result.get('title'),
1170 'playlist_uploader': ie_result.get('uploader'),
1171 'playlist_uploader_id': ie_result.get('uploader_id'),
1172 'playlist_index': 0
1173 }
1174 ie_copy.update(dict(ie_result))
1175
1176 def ensure_dir_exists(path):
1177 return make_dir(path, self.report_error)
1178
1179 if self.params.get('writeinfojson', False):
1180 infofn = self.prepare_filename(ie_copy, 'pl_infojson')
1181 if not ensure_dir_exists(encodeFilename(infofn)):
1182 return
1183 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(infofn)):
1184 self.to_screen('[info] Playlist metadata is already present')
1185 else:
1186 playlist_info = dict(ie_result)
1187 # playlist_info['entries'] = list(playlist_info['entries']) # Entries is a generator which shouldnot be resolved here
1188 del playlist_info['entries']
1189 self.to_screen('[info] Writing playlist metadata as JSON to: ' + infofn)
1190 try:
1191 write_json_file(self.filter_requested_info(playlist_info), infofn)
1192 except (OSError, IOError):
1193 self.report_error('Cannot write playlist metadata to JSON file ' + infofn)
1194
1195 if self.params.get('writedescription', False):
1196 descfn = self.prepare_filename(ie_copy, 'pl_description')
1197 if not ensure_dir_exists(encodeFilename(descfn)):
1198 return
1199 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(descfn)):
1200 self.to_screen('[info] Playlist description is already present')
1201 elif ie_result.get('description') is None:
1202 self.report_warning('There\'s no playlist description to write.')
1203 else:
1204 try:
1205 self.to_screen('[info] Writing playlist description to: ' + descfn)
1206 with io.open(encodeFilename(descfn), 'w', encoding='utf-8') as descfile:
1207 descfile.write(ie_result['description'])
1208 except (OSError, IOError):
1209 self.report_error('Cannot write playlist description file ' + descfn)
1210 return
1211
1212 playlist_results = []
1213
1214 playliststart = self.params.get('playliststart', 1) - 1
1215 playlistend = self.params.get('playlistend')
1216 # For backwards compatibility, interpret -1 as whole list
1217 if playlistend == -1:
1218 playlistend = None
1219
1220 playlistitems_str = self.params.get('playlist_items')
1221 playlistitems = None
1222 if playlistitems_str is not None:
1223 def iter_playlistitems(format):
1224 for string_segment in format.split(','):
1225 if '-' in string_segment:
1226 start, end = string_segment.split('-')
1227 for item in range(int(start), int(end) + 1):
1228 yield int(item)
1229 else:
1230 yield int(string_segment)
1231 playlistitems = orderedSet(iter_playlistitems(playlistitems_str))
1232
1233 ie_entries = ie_result['entries']
1234
1235 def make_playlistitems_entries(list_ie_entries):
1236 num_entries = len(list_ie_entries)
1237 return [
1238 list_ie_entries[i - 1] for i in playlistitems
1239 if -num_entries <= i - 1 < num_entries]
1240
1241 def report_download(num_entries):
1242 self.to_screen(
1243 '[%s] playlist %s: Downloading %d videos' %
1244 (ie_result['extractor'], playlist, num_entries))
1245
1246 if isinstance(ie_entries, list):
1247 n_all_entries = len(ie_entries)
1248 if playlistitems:
1249 entries = make_playlistitems_entries(ie_entries)
1250 else:
1251 entries = ie_entries[playliststart:playlistend]
1252 n_entries = len(entries)
1253 self.to_screen(
1254 '[%s] playlist %s: Collected %d video ids (downloading %d of them)' %
1255 (ie_result['extractor'], playlist, n_all_entries, n_entries))
1256 elif isinstance(ie_entries, PagedList):
1257 if playlistitems:
1258 entries = []
1259 for item in playlistitems:
1260 entries.extend(ie_entries.getslice(
1261 item - 1, item
1262 ))
1263 else:
1264 entries = ie_entries.getslice(
1265 playliststart, playlistend)
1266 n_entries = len(entries)
1267 report_download(n_entries)
1268 else: # iterable
1269 if playlistitems:
1270 entries = make_playlistitems_entries(list(itertools.islice(
1271 ie_entries, 0, max(playlistitems))))
1272 else:
1273 entries = list(itertools.islice(
1274 ie_entries, playliststart, playlistend))
1275 n_entries = len(entries)
1276 report_download(n_entries)
1277
1278 if self.params.get('playlistreverse', False):
1279 entries = entries[::-1]
1280
1281 if self.params.get('playlistrandom', False):
1282 random.shuffle(entries)
1283
1284 x_forwarded_for = ie_result.get('__x_forwarded_for_ip')
1285
1286 for i, entry in enumerate(entries, 1):
1287 self.to_screen('[download] Downloading video %s of %s' % (i, n_entries))
1288 # This __x_forwarded_for_ip thing is a bit ugly but requires
1289 # minimal changes
1290 if x_forwarded_for:
1291 entry['__x_forwarded_for_ip'] = x_forwarded_for
1292 extra = {
1293 'n_entries': n_entries,
1294 'playlist': playlist,
1295 'playlist_id': ie_result.get('id'),
1296 'playlist_title': ie_result.get('title'),
1297 'playlist_uploader': ie_result.get('uploader'),
1298 'playlist_uploader_id': ie_result.get('uploader_id'),
1299 'playlist_index': playlistitems[i - 1] if playlistitems else i + playliststart,
1300 'extractor': ie_result['extractor'],
1301 'webpage_url': ie_result['webpage_url'],
1302 'webpage_url_basename': url_basename(ie_result['webpage_url']),
1303 'extractor_key': ie_result['extractor_key'],
1304 }
1305
1306 if self._match_entry(entry, incomplete=True) is not None:
1307 continue
1308
1309 entry_result = self.__process_iterable_entry(entry, download, extra)
1310 # TODO: skip failed (empty) entries?
1311 playlist_results.append(entry_result)
1312 ie_result['entries'] = playlist_results
1313 self.to_screen('[download] Finished downloading playlist: %s' % playlist)
1314 return ie_result
1315
1316 @__handle_extraction_exceptions
1317 def __process_iterable_entry(self, entry, download, extra_info):
1318 return self.process_ie_result(
1319 entry, download=download, extra_info=extra_info)
1320
1321 def _build_format_filter(self, filter_spec):
1322 " Returns a function to filter the formats according to the filter_spec "
1323
1324 OPERATORS = {
1325 '<': operator.lt,
1326 '<=': operator.le,
1327 '>': operator.gt,
1328 '>=': operator.ge,
1329 '=': operator.eq,
1330 '!=': operator.ne,
1331 }
1332 operator_rex = re.compile(r'''(?x)\s*
1333 (?P<key>width|height|tbr|abr|vbr|asr|filesize|filesize_approx|fps)
1334 \s*(?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
1335 (?P<value>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?)
1336 $
1337 ''' % '|'.join(map(re.escape, OPERATORS.keys())))
1338 m = operator_rex.search(filter_spec)
1339 if m:
1340 try:
1341 comparison_value = int(m.group('value'))
1342 except ValueError:
1343 comparison_value = parse_filesize(m.group('value'))
1344 if comparison_value is None:
1345 comparison_value = parse_filesize(m.group('value') + 'B')
1346 if comparison_value is None:
1347 raise ValueError(
1348 'Invalid value %r in format specification %r' % (
1349 m.group('value'), filter_spec))
1350 op = OPERATORS[m.group('op')]
1351
1352 if not m:
1353 STR_OPERATORS = {
1354 '=': operator.eq,
1355 '^=': lambda attr, value: attr.startswith(value),
1356 '$=': lambda attr, value: attr.endswith(value),
1357 '*=': lambda attr, value: value in attr,
1358 }
1359 str_operator_rex = re.compile(r'''(?x)
1360 \s*(?P<key>[a-zA-Z0-9._-]+)
1361 \s*(?P<negation>!\s*)?(?P<op>%s)(?P<none_inclusive>\s*\?)?
1362 \s*(?P<value>[a-zA-Z0-9._-]+)
1363 \s*$
1364 ''' % '|'.join(map(re.escape, STR_OPERATORS.keys())))
1365 m = str_operator_rex.search(filter_spec)
1366 if m:
1367 comparison_value = m.group('value')
1368 str_op = STR_OPERATORS[m.group('op')]
1369 if m.group('negation'):
1370 op = lambda attr, value: not str_op(attr, value)
1371 else:
1372 op = str_op
1373
1374 if not m:
1375 raise ValueError('Invalid filter specification %r' % filter_spec)
1376
1377 def _filter(f):
1378 actual_value = f.get(m.group('key'))
1379 if actual_value is None:
1380 return m.group('none_inclusive')
1381 return op(actual_value, comparison_value)
1382 return _filter
1383
1384 def _default_format_spec(self, info_dict, download=True):
1385
1386 def can_merge():
1387 merger = FFmpegMergerPP(self)
1388 return merger.available and merger.can_merge()
1389
1390 prefer_best = (
1391 not self.params.get('simulate', False)
1392 and download
1393 and (
1394 not can_merge()
1395 or info_dict.get('is_live', False)
1396 or self.outtmpl_dict['default'] == '-'))
1397
1398 return (
1399 'best/bestvideo+bestaudio'
1400 if prefer_best
1401 else 'bestvideo*+bestaudio/best'
1402 if not self.params.get('allow_multiple_audio_streams', False)
1403 else 'bestvideo+bestaudio/best')
1404
1405 def build_format_selector(self, format_spec):
1406 def syntax_error(note, start):
1407 message = (
1408 'Invalid format specification: '
1409 '{0}\n\t{1}\n\t{2}^'.format(note, format_spec, ' ' * start[1]))
1410 return SyntaxError(message)
1411
1412 PICKFIRST = 'PICKFIRST'
1413 MERGE = 'MERGE'
1414 SINGLE = 'SINGLE'
1415 GROUP = 'GROUP'
1416 FormatSelector = collections.namedtuple('FormatSelector', ['type', 'selector', 'filters'])
1417
1418 allow_multiple_streams = {'audio': self.params.get('allow_multiple_audio_streams', False),
1419 'video': self.params.get('allow_multiple_video_streams', False)}
1420
1421 def _parse_filter(tokens):
1422 filter_parts = []
1423 for type, string, start, _, _ in tokens:
1424 if type == tokenize.OP and string == ']':
1425 return ''.join(filter_parts)
1426 else:
1427 filter_parts.append(string)
1428
1429 def _remove_unused_ops(tokens):
1430 # Remove operators that we don't use and join them with the surrounding strings
1431 # for example: 'mp4' '-' 'baseline' '-' '16x9' is converted to 'mp4-baseline-16x9'
1432 ALLOWED_OPS = ('/', '+', ',', '(', ')')
1433 last_string, last_start, last_end, last_line = None, None, None, None
1434 for type, string, start, end, line in tokens:
1435 if type == tokenize.OP and string == '[':
1436 if last_string:
1437 yield tokenize.NAME, last_string, last_start, last_end, last_line
1438 last_string = None
1439 yield type, string, start, end, line
1440 # everything inside brackets will be handled by _parse_filter
1441 for type, string, start, end, line in tokens:
1442 yield type, string, start, end, line
1443 if type == tokenize.OP and string == ']':
1444 break
1445 elif type == tokenize.OP and string in ALLOWED_OPS:
1446 if last_string:
1447 yield tokenize.NAME, last_string, last_start, last_end, last_line
1448 last_string = None
1449 yield type, string, start, end, line
1450 elif type in [tokenize.NAME, tokenize.NUMBER, tokenize.OP]:
1451 if not last_string:
1452 last_string = string
1453 last_start = start
1454 last_end = end
1455 else:
1456 last_string += string
1457 if last_string:
1458 yield tokenize.NAME, last_string, last_start, last_end, last_line
1459
1460 def _parse_format_selection(tokens, inside_merge=False, inside_choice=False, inside_group=False):
1461 selectors = []
1462 current_selector = None
1463 for type, string, start, _, _ in tokens:
1464 # ENCODING is only defined in python 3.x
1465 if type == getattr(tokenize, 'ENCODING', None):
1466 continue
1467 elif type in [tokenize.NAME, tokenize.NUMBER]:
1468 current_selector = FormatSelector(SINGLE, string, [])
1469 elif type == tokenize.OP:
1470 if string == ')':
1471 if not inside_group:
1472 # ')' will be handled by the parentheses group
1473 tokens.restore_last_token()
1474 break
1475 elif inside_merge and string in ['/', ',']:
1476 tokens.restore_last_token()
1477 break
1478 elif inside_choice and string == ',':
1479 tokens.restore_last_token()
1480 break
1481 elif string == ',':
1482 if not current_selector:
1483 raise syntax_error('"," must follow a format selector', start)
1484 selectors.append(current_selector)
1485 current_selector = None
1486 elif string == '/':
1487 if not current_selector:
1488 raise syntax_error('"/" must follow a format selector', start)
1489 first_choice = current_selector
1490 second_choice = _parse_format_selection(tokens, inside_choice=True)
1491 current_selector = FormatSelector(PICKFIRST, (first_choice, second_choice), [])
1492 elif string == '[':
1493 if not current_selector:
1494 current_selector = FormatSelector(SINGLE, 'best', [])
1495 format_filter = _parse_filter(tokens)
1496 current_selector.filters.append(format_filter)
1497 elif string == '(':
1498 if current_selector:
1499 raise syntax_error('Unexpected "("', start)
1500 group = _parse_format_selection(tokens, inside_group=True)
1501 current_selector = FormatSelector(GROUP, group, [])
1502 elif string == '+':
1503 if not current_selector:
1504 raise syntax_error('Unexpected "+"', start)
1505 selector_1 = current_selector
1506 selector_2 = _parse_format_selection(tokens, inside_merge=True)
1507 if not selector_2:
1508 raise syntax_error('Expected a selector', start)
1509 current_selector = FormatSelector(MERGE, (selector_1, selector_2), [])
1510 else:
1511 raise syntax_error('Operator not recognized: "{0}"'.format(string), start)
1512 elif type == tokenize.ENDMARKER:
1513 break
1514 if current_selector:
1515 selectors.append(current_selector)
1516 return selectors
1517
1518 def _build_selector_function(selector):
1519 if isinstance(selector, list): # ,
1520 fs = [_build_selector_function(s) for s in selector]
1521
1522 def selector_function(ctx):
1523 for f in fs:
1524 for format in f(ctx):
1525 yield format
1526 return selector_function
1527
1528 elif selector.type == GROUP: # ()
1529 selector_function = _build_selector_function(selector.selector)
1530
1531 elif selector.type == PICKFIRST: # /
1532 fs = [_build_selector_function(s) for s in selector.selector]
1533
1534 def selector_function(ctx):
1535 for f in fs:
1536 picked_formats = list(f(ctx))
1537 if picked_formats:
1538 return picked_formats
1539 return []
1540
1541 elif selector.type == SINGLE: # atom
1542 format_spec = selector.selector if selector.selector is not None else 'best'
1543
1544 if format_spec == 'all':
1545 def selector_function(ctx):
1546 formats = list(ctx['formats'])
1547 if formats:
1548 for f in formats:
1549 yield f
1550
1551 else:
1552 format_fallback = False
1553 format_spec_obj = re.match(r'(best|worst|b|w)(video|audio|v|a)?(\*)?$', format_spec)
1554 if format_spec_obj is not None:
1555 format_idx = 0 if format_spec_obj.group(1)[0] == 'w' else -1
1556 format_type = format_spec_obj.group(2)[0] if format_spec_obj.group(2) else False
1557 not_format_type = 'v' if format_type == 'a' else 'a'
1558 format_modified = format_spec_obj.group(3) is not None
1559
1560 format_fallback = not format_type and not format_modified # for b, w
1561 filter_f = ((lambda f: f.get(format_type + 'codec') != 'none')
1562 if format_type and format_modified # bv*, ba*, wv*, wa*
1563 else (lambda f: f.get(not_format_type + 'codec') == 'none')
1564 if format_type # bv, ba, wv, wa
1565 else (lambda f: f.get('vcodec') != 'none' and f.get('acodec') != 'none')
1566 if not format_modified # b, w
1567 else None) # b*, w*
1568 else:
1569 format_idx = -1
1570 filter_f = ((lambda f: f.get('ext') == format_spec)
1571 if format_spec in ['mp4', 'flv', 'webm', '3gp', 'm4a', 'mp3', 'ogg', 'aac', 'wav'] # extension
1572 else (lambda f: f.get('format_id') == format_spec)) # id
1573
1574 def selector_function(ctx):
1575 formats = list(ctx['formats'])
1576 if not formats:
1577 return
1578 matches = list(filter(filter_f, formats)) if filter_f is not None else formats
1579 if matches:
1580 yield matches[format_idx]
1581 elif format_fallback == 'force' or (format_fallback and ctx['incomplete_formats']):
1582 # for extractors with incomplete formats (audio only (soundcloud)
1583 # or video only (imgur)) best/worst will fallback to
1584 # best/worst {video,audio}-only format
1585 yield formats[format_idx]
1586
1587 elif selector.type == MERGE: # +
1588 def _merge(formats_pair):
1589 format_1, format_2 = formats_pair
1590
1591 formats_info = []
1592 formats_info.extend(format_1.get('requested_formats', (format_1,)))
1593 formats_info.extend(format_2.get('requested_formats', (format_2,)))
1594
1595 if not allow_multiple_streams['video'] or not allow_multiple_streams['audio']:
1596 get_no_more = {"video": False, "audio": False}
1597 for (i, fmt_info) in enumerate(formats_info):
1598 for aud_vid in ["audio", "video"]:
1599 if not allow_multiple_streams[aud_vid] and fmt_info.get(aud_vid[0] + 'codec') != 'none':
1600 if get_no_more[aud_vid]:
1601 formats_info.pop(i)
1602 get_no_more[aud_vid] = True
1603
1604 if len(formats_info) == 1:
1605 return formats_info[0]
1606
1607 video_fmts = [fmt_info for fmt_info in formats_info if fmt_info.get('vcodec') != 'none']
1608 audio_fmts = [fmt_info for fmt_info in formats_info if fmt_info.get('acodec') != 'none']
1609
1610 the_only_video = video_fmts[0] if len(video_fmts) == 1 else None
1611 the_only_audio = audio_fmts[0] if len(audio_fmts) == 1 else None
1612
1613 output_ext = self.params.get('merge_output_format')
1614 if not output_ext:
1615 if the_only_video:
1616 output_ext = the_only_video['ext']
1617 elif the_only_audio and not video_fmts:
1618 output_ext = the_only_audio['ext']
1619 else:
1620 output_ext = 'mkv'
1621
1622 new_dict = {
1623 'requested_formats': formats_info,
1624 'format': '+'.join(fmt_info.get('format') for fmt_info in formats_info),
1625 'format_id': '+'.join(fmt_info.get('format_id') for fmt_info in formats_info),
1626 'ext': output_ext,
1627 }
1628
1629 if the_only_video:
1630 new_dict.update({
1631 'width': the_only_video.get('width'),
1632 'height': the_only_video.get('height'),
1633 'resolution': the_only_video.get('resolution'),
1634 'fps': the_only_video.get('fps'),
1635 'vcodec': the_only_video.get('vcodec'),
1636 'vbr': the_only_video.get('vbr'),
1637 'stretched_ratio': the_only_video.get('stretched_ratio'),
1638 })
1639
1640 if the_only_audio:
1641 new_dict.update({
1642 'acodec': the_only_audio.get('acodec'),
1643 'abr': the_only_audio.get('abr'),
1644 })
1645
1646 return new_dict
1647
1648 selector_1, selector_2 = map(_build_selector_function, selector.selector)
1649
1650 def selector_function(ctx):
1651 for pair in itertools.product(
1652 selector_1(copy.deepcopy(ctx)), selector_2(copy.deepcopy(ctx))):
1653 yield _merge(pair)
1654
1655 filters = [self._build_format_filter(f) for f in selector.filters]
1656
1657 def final_selector(ctx):
1658 ctx_copy = copy.deepcopy(ctx)
1659 for _filter in filters:
1660 ctx_copy['formats'] = list(filter(_filter, ctx_copy['formats']))
1661 return selector_function(ctx_copy)
1662 return final_selector
1663
1664 stream = io.BytesIO(format_spec.encode('utf-8'))
1665 try:
1666 tokens = list(_remove_unused_ops(compat_tokenize_tokenize(stream.readline)))
1667 except tokenize.TokenError:
1668 raise syntax_error('Missing closing/opening brackets or parenthesis', (0, len(format_spec)))
1669
1670 class TokenIterator(object):
1671 def __init__(self, tokens):
1672 self.tokens = tokens
1673 self.counter = 0
1674
1675 def __iter__(self):
1676 return self
1677
1678 def __next__(self):
1679 if self.counter >= len(self.tokens):
1680 raise StopIteration()
1681 value = self.tokens[self.counter]
1682 self.counter += 1
1683 return value
1684
1685 next = __next__
1686
1687 def restore_last_token(self):
1688 self.counter -= 1
1689
1690 parsed_selector = _parse_format_selection(iter(TokenIterator(tokens)))
1691 return _build_selector_function(parsed_selector)
1692
1693 def _calc_headers(self, info_dict):
1694 res = std_headers.copy()
1695
1696 add_headers = info_dict.get('http_headers')
1697 if add_headers:
1698 res.update(add_headers)
1699
1700 cookies = self._calc_cookies(info_dict)
1701 if cookies:
1702 res['Cookie'] = cookies
1703
1704 if 'X-Forwarded-For' not in res:
1705 x_forwarded_for_ip = info_dict.get('__x_forwarded_for_ip')
1706 if x_forwarded_for_ip:
1707 res['X-Forwarded-For'] = x_forwarded_for_ip
1708
1709 return res
1710
1711 def _calc_cookies(self, info_dict):
1712 pr = sanitized_Request(info_dict['url'])
1713 self.cookiejar.add_cookie_header(pr)
1714 return pr.get_header('Cookie')
1715
1716 def process_video_result(self, info_dict, download=True):
1717 assert info_dict.get('_type', 'video') == 'video'
1718
1719 if 'id' not in info_dict:
1720 raise ExtractorError('Missing "id" field in extractor result')
1721 if 'title' not in info_dict:
1722 raise ExtractorError('Missing "title" field in extractor result')
1723
1724 def report_force_conversion(field, field_not, conversion):
1725 self.report_warning(
1726 '"%s" field is not %s - forcing %s conversion, there is an error in extractor'
1727 % (field, field_not, conversion))
1728
1729 def sanitize_string_field(info, string_field):
1730 field = info.get(string_field)
1731 if field is None or isinstance(field, compat_str):
1732 return
1733 report_force_conversion(string_field, 'a string', 'string')
1734 info[string_field] = compat_str(field)
1735
1736 def sanitize_numeric_fields(info):
1737 for numeric_field in self._NUMERIC_FIELDS:
1738 field = info.get(numeric_field)
1739 if field is None or isinstance(field, compat_numeric_types):
1740 continue
1741 report_force_conversion(numeric_field, 'numeric', 'int')
1742 info[numeric_field] = int_or_none(field)
1743
1744 sanitize_string_field(info_dict, 'id')
1745 sanitize_numeric_fields(info_dict)
1746
1747 if 'playlist' not in info_dict:
1748 # It isn't part of a playlist
1749 info_dict['playlist'] = None
1750 info_dict['playlist_index'] = None
1751
1752 thumbnails = info_dict.get('thumbnails')
1753 if thumbnails is None:
1754 thumbnail = info_dict.get('thumbnail')
1755 if thumbnail:
1756 info_dict['thumbnails'] = thumbnails = [{'url': thumbnail}]
1757 if thumbnails:
1758 thumbnails.sort(key=lambda t: (
1759 t.get('preference') if t.get('preference') is not None else -1,
1760 t.get('width') if t.get('width') is not None else -1,
1761 t.get('height') if t.get('height') is not None else -1,
1762 t.get('id') if t.get('id') is not None else '', t.get('url')))
1763 for i, t in enumerate(thumbnails):
1764 t['url'] = sanitize_url(t['url'])
1765 if t.get('width') and t.get('height'):
1766 t['resolution'] = '%dx%d' % (t['width'], t['height'])
1767 if t.get('id') is None:
1768 t['id'] = '%d' % i
1769
1770 if self.params.get('list_thumbnails'):
1771 self.list_thumbnails(info_dict)
1772 return
1773
1774 thumbnail = info_dict.get('thumbnail')
1775 if thumbnail:
1776 info_dict['thumbnail'] = sanitize_url(thumbnail)
1777 elif thumbnails:
1778 info_dict['thumbnail'] = thumbnails[-1]['url']
1779
1780 if 'display_id' not in info_dict and 'id' in info_dict:
1781 info_dict['display_id'] = info_dict['id']
1782
1783 if info_dict.get('upload_date') is None and info_dict.get('timestamp') is not None:
1784 # Working around out-of-range timestamp values (e.g. negative ones on Windows,
1785 # see http://bugs.python.org/issue1646728)
1786 try:
1787 upload_date = datetime.datetime.utcfromtimestamp(info_dict['timestamp'])
1788 info_dict['upload_date'] = upload_date.strftime('%Y%m%d')
1789 except (ValueError, OverflowError, OSError):
1790 pass
1791
1792 # Auto generate title fields corresponding to the *_number fields when missing
1793 # in order to always have clean titles. This is very common for TV series.
1794 for field in ('chapter', 'season', 'episode'):
1795 if info_dict.get('%s_number' % field) is not None and not info_dict.get(field):
1796 info_dict[field] = '%s %d' % (field.capitalize(), info_dict['%s_number' % field])
1797
1798 for cc_kind in ('subtitles', 'automatic_captions'):
1799 cc = info_dict.get(cc_kind)
1800 if cc:
1801 for _, subtitle in cc.items():
1802 for subtitle_format in subtitle:
1803 if subtitle_format.get('url'):
1804 subtitle_format['url'] = sanitize_url(subtitle_format['url'])
1805 if subtitle_format.get('ext') is None:
1806 subtitle_format['ext'] = determine_ext(subtitle_format['url']).lower()
1807
1808 automatic_captions = info_dict.get('automatic_captions')
1809 subtitles = info_dict.get('subtitles')
1810
1811 if self.params.get('listsubtitles', False):
1812 if 'automatic_captions' in info_dict:
1813 self.list_subtitles(
1814 info_dict['id'], automatic_captions, 'automatic captions')
1815 self.list_subtitles(info_dict['id'], subtitles, 'subtitles')
1816 return
1817
1818 info_dict['requested_subtitles'] = self.process_subtitles(
1819 info_dict['id'], subtitles, automatic_captions)
1820
1821 # We now pick which formats have to be downloaded
1822 if info_dict.get('formats') is None:
1823 # There's only one format available
1824 formats = [info_dict]
1825 else:
1826 formats = info_dict['formats']
1827
1828 if not formats:
1829 raise ExtractorError('No video formats found!')
1830
1831 def is_wellformed(f):
1832 url = f.get('url')
1833 if not url:
1834 self.report_warning(
1835 '"url" field is missing or empty - skipping format, '
1836 'there is an error in extractor')
1837 return False
1838 if isinstance(url, bytes):
1839 sanitize_string_field(f, 'url')
1840 return True
1841
1842 # Filter out malformed formats for better extraction robustness
1843 formats = list(filter(is_wellformed, formats))
1844
1845 formats_dict = {}
1846
1847 # We check that all the formats have the format and format_id fields
1848 for i, format in enumerate(formats):
1849 sanitize_string_field(format, 'format_id')
1850 sanitize_numeric_fields(format)
1851 format['url'] = sanitize_url(format['url'])
1852 if not format.get('format_id'):
1853 format['format_id'] = compat_str(i)
1854 else:
1855 # Sanitize format_id from characters used in format selector expression
1856 format['format_id'] = re.sub(r'[\s,/+\[\]()]', '_', format['format_id'])
1857 format_id = format['format_id']
1858 if format_id not in formats_dict:
1859 formats_dict[format_id] = []
1860 formats_dict[format_id].append(format)
1861
1862 # Make sure all formats have unique format_id
1863 for format_id, ambiguous_formats in formats_dict.items():
1864 if len(ambiguous_formats) > 1:
1865 for i, format in enumerate(ambiguous_formats):
1866 format['format_id'] = '%s-%d' % (format_id, i)
1867
1868 for i, format in enumerate(formats):
1869 if format.get('format') is None:
1870 format['format'] = '{id} - {res}{note}'.format(
1871 id=format['format_id'],
1872 res=self.format_resolution(format),
1873 note=' ({0})'.format(format['format_note']) if format.get('format_note') is not None else '',
1874 )
1875 # Automatically determine file extension if missing
1876 if format.get('ext') is None:
1877 format['ext'] = determine_ext(format['url']).lower()
1878 # Automatically determine protocol if missing (useful for format
1879 # selection purposes)
1880 if format.get('protocol') is None:
1881 format['protocol'] = determine_protocol(format)
1882 # Add HTTP headers, so that external programs can use them from the
1883 # json output
1884 full_format_info = info_dict.copy()
1885 full_format_info.update(format)
1886 format['http_headers'] = self._calc_headers(full_format_info)
1887 # Remove private housekeeping stuff
1888 if '__x_forwarded_for_ip' in info_dict:
1889 del info_dict['__x_forwarded_for_ip']
1890
1891 # TODO Central sorting goes here
1892
1893 if formats[0] is not info_dict:
1894 # only set the 'formats' fields if the original info_dict list them
1895 # otherwise we end up with a circular reference, the first (and unique)
1896 # element in the 'formats' field in info_dict is info_dict itself,
1897 # which can't be exported to json
1898 info_dict['formats'] = formats
1899 if self.params.get('listformats'):
1900 self.list_formats(info_dict)
1901 return
1902
1903 req_format = self.params.get('format')
1904 if req_format is None:
1905 req_format = self._default_format_spec(info_dict, download=download)
1906 if self.params.get('verbose'):
1907 self.to_screen('[debug] Default format spec: %s' % req_format)
1908
1909 format_selector = self.build_format_selector(req_format)
1910
1911 # While in format selection we may need to have an access to the original
1912 # format set in order to calculate some metrics or do some processing.
1913 # For now we need to be able to guess whether original formats provided
1914 # by extractor are incomplete or not (i.e. whether extractor provides only
1915 # video-only or audio-only formats) for proper formats selection for
1916 # extractors with such incomplete formats (see
1917 # https://github.com/ytdl-org/youtube-dl/pull/5556).
1918 # Since formats may be filtered during format selection and may not match
1919 # the original formats the results may be incorrect. Thus original formats
1920 # or pre-calculated metrics should be passed to format selection routines
1921 # as well.
1922 # We will pass a context object containing all necessary additional data
1923 # instead of just formats.
1924 # This fixes incorrect format selection issue (see
1925 # https://github.com/ytdl-org/youtube-dl/issues/10083).
1926 incomplete_formats = (
1927 # All formats are video-only or
1928 all(f.get('vcodec') != 'none' and f.get('acodec') == 'none' for f in formats)
1929 # all formats are audio-only
1930 or all(f.get('vcodec') == 'none' and f.get('acodec') != 'none' for f in formats))
1931
1932 ctx = {
1933 'formats': formats,
1934 'incomplete_formats': incomplete_formats,
1935 }
1936
1937 formats_to_download = list(format_selector(ctx))
1938 if not formats_to_download:
1939 raise ExtractorError('requested format not available',
1940 expected=True)
1941
1942 if download:
1943 self.to_screen('[info] Downloading format(s) %s' % ", ".join([f['format_id'] for f in formats_to_download]))
1944 if len(formats_to_download) > 1:
1945 self.to_screen('[info] %s: downloading video in %s formats' % (info_dict['id'], len(formats_to_download)))
1946 for format in formats_to_download:
1947 new_info = dict(info_dict)
1948 new_info.update(format)
1949 self.process_info(new_info)
1950 # We update the info dict with the best quality format (backwards compatibility)
1951 info_dict.update(formats_to_download[-1])
1952 return info_dict
1953
1954 def process_subtitles(self, video_id, normal_subtitles, automatic_captions):
1955 """Select the requested subtitles and their format"""
1956 available_subs = {}
1957 if normal_subtitles and self.params.get('writesubtitles'):
1958 available_subs.update(normal_subtitles)
1959 if automatic_captions and self.params.get('writeautomaticsub'):
1960 for lang, cap_info in automatic_captions.items():
1961 if lang not in available_subs:
1962 available_subs[lang] = cap_info
1963
1964 if (not self.params.get('writesubtitles') and not
1965 self.params.get('writeautomaticsub') or not
1966 available_subs):
1967 return None
1968
1969 if self.params.get('allsubtitles', False):
1970 requested_langs = available_subs.keys()
1971 else:
1972 if self.params.get('subtitleslangs', False):
1973 requested_langs = self.params.get('subtitleslangs')
1974 elif 'en' in available_subs:
1975 requested_langs = ['en']
1976 else:
1977 requested_langs = [list(available_subs.keys())[0]]
1978
1979 formats_query = self.params.get('subtitlesformat', 'best')
1980 formats_preference = formats_query.split('/') if formats_query else []
1981 subs = {}
1982 for lang in requested_langs:
1983 formats = available_subs.get(lang)
1984 if formats is None:
1985 self.report_warning('%s subtitles not available for %s' % (lang, video_id))
1986 continue
1987 for ext in formats_preference:
1988 if ext == 'best':
1989 f = formats[-1]
1990 break
1991 matches = list(filter(lambda f: f['ext'] == ext, formats))
1992 if matches:
1993 f = matches[-1]
1994 break
1995 else:
1996 f = formats[-1]
1997 self.report_warning(
1998 'No subtitle format found matching "%s" for language %s, '
1999 'using %s' % (formats_query, lang, f['ext']))
2000 subs[lang] = f
2001 return subs
2002
2003 def __forced_printings(self, info_dict, filename, incomplete):
2004 def print_mandatory(field):
2005 if (self.params.get('force%s' % field, False)
2006 and (not incomplete or info_dict.get(field) is not None)):
2007 self.to_stdout(info_dict[field])
2008
2009 def print_optional(field):
2010 if (self.params.get('force%s' % field, False)
2011 and info_dict.get(field) is not None):
2012 self.to_stdout(info_dict[field])
2013
2014 print_mandatory('title')
2015 print_mandatory('id')
2016 if self.params.get('forceurl', False) and not incomplete:
2017 if info_dict.get('requested_formats') is not None:
2018 for f in info_dict['requested_formats']:
2019 self.to_stdout(f['url'] + f.get('play_path', ''))
2020 else:
2021 # For RTMP URLs, also include the playpath
2022 self.to_stdout(info_dict['url'] + info_dict.get('play_path', ''))
2023 print_optional('thumbnail')
2024 print_optional('description')
2025 if self.params.get('forcefilename', False) and filename is not None:
2026 self.to_stdout(filename)
2027 if self.params.get('forceduration', False) and info_dict.get('duration') is not None:
2028 self.to_stdout(formatSeconds(info_dict['duration']))
2029 print_mandatory('format')
2030 if self.params.get('forcejson', False):
2031 self.to_stdout(json.dumps(info_dict))
2032
2033 def process_info(self, info_dict):
2034 """Process a single resolved IE result."""
2035
2036 assert info_dict.get('_type', 'video') == 'video'
2037
2038 info_dict.setdefault('__postprocessors', [])
2039
2040 max_downloads = self.params.get('max_downloads')
2041 if max_downloads is not None:
2042 if self._num_downloads >= int(max_downloads):
2043 raise MaxDownloadsReached()
2044
2045 # TODO: backward compatibility, to be removed
2046 info_dict['fulltitle'] = info_dict['title']
2047
2048 if 'format' not in info_dict:
2049 info_dict['format'] = info_dict['ext']
2050
2051 if self._match_entry(info_dict, incomplete=False) is not None:
2052 return
2053
2054 self._num_downloads += 1
2055
2056 info_dict = self.pre_process(info_dict)
2057
2058 info_dict['_filename'] = full_filename = self.prepare_filename(info_dict, warn=True)
2059 temp_filename = self.prepare_filename(info_dict, 'temp')
2060 files_to_move = {}
2061 skip_dl = self.params.get('skip_download', False)
2062
2063 # Forced printings
2064 self.__forced_printings(info_dict, full_filename, incomplete=False)
2065
2066 if self.params.get('simulate', False):
2067 if self.params.get('force_write_download_archive', False):
2068 self.record_download_archive(info_dict)
2069
2070 # Do nothing else if in simulate mode
2071 return
2072
2073 if full_filename is None:
2074 return
2075
2076 def ensure_dir_exists(path):
2077 return make_dir(path, self.report_error)
2078
2079 if not ensure_dir_exists(encodeFilename(full_filename)):
2080 return
2081 if not ensure_dir_exists(encodeFilename(temp_filename)):
2082 return
2083
2084 if self.params.get('writedescription', False):
2085 descfn = self.prepare_filename(info_dict, 'description')
2086 if not ensure_dir_exists(encodeFilename(descfn)):
2087 return
2088 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(descfn)):
2089 self.to_screen('[info] Video description is already present')
2090 elif info_dict.get('description') is None:
2091 self.report_warning('There\'s no description to write.')
2092 else:
2093 try:
2094 self.to_screen('[info] Writing video description to: ' + descfn)
2095 with io.open(encodeFilename(descfn), 'w', encoding='utf-8') as descfile:
2096 descfile.write(info_dict['description'])
2097 except (OSError, IOError):
2098 self.report_error('Cannot write description file ' + descfn)
2099 return
2100
2101 if self.params.get('writeannotations', False):
2102 annofn = self.prepare_filename(info_dict, 'annotation')
2103 if not ensure_dir_exists(encodeFilename(annofn)):
2104 return
2105 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(annofn)):
2106 self.to_screen('[info] Video annotations are already present')
2107 elif not info_dict.get('annotations'):
2108 self.report_warning('There are no annotations to write.')
2109 else:
2110 try:
2111 self.to_screen('[info] Writing video annotations to: ' + annofn)
2112 with io.open(encodeFilename(annofn), 'w', encoding='utf-8') as annofile:
2113 annofile.write(info_dict['annotations'])
2114 except (KeyError, TypeError):
2115 self.report_warning('There are no annotations to write.')
2116 except (OSError, IOError):
2117 self.report_error('Cannot write annotations file: ' + annofn)
2118 return
2119
2120 def dl(name, info, subtitle=False):
2121 fd = get_suitable_downloader(info, self.params)(self, self.params)
2122 for ph in self._progress_hooks:
2123 fd.add_progress_hook(ph)
2124 if self.params.get('verbose'):
2125 self.to_screen('[debug] Invoking downloader on %r' % info.get('url'))
2126 return fd.download(name, info, subtitle)
2127
2128 subtitles_are_requested = any([self.params.get('writesubtitles', False),
2129 self.params.get('writeautomaticsub')])
2130
2131 if subtitles_are_requested and info_dict.get('requested_subtitles'):
2132 # subtitles download errors are already managed as troubles in relevant IE
2133 # that way it will silently go on when used with unsupporting IE
2134 subtitles = info_dict['requested_subtitles']
2135 # ie = self.get_info_extractor(info_dict['extractor_key'])
2136 for sub_lang, sub_info in subtitles.items():
2137 sub_format = sub_info['ext']
2138 sub_fn = self.prepare_filename(info_dict, 'subtitle')
2139 sub_filename = subtitles_filename(
2140 temp_filename if not skip_dl else sub_fn,
2141 sub_lang, sub_format, info_dict.get('ext'))
2142 sub_filename_final = subtitles_filename(sub_fn, sub_lang, sub_format, info_dict.get('ext'))
2143 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(sub_filename)):
2144 self.to_screen('[info] Video subtitle %s.%s is already present' % (sub_lang, sub_format))
2145 files_to_move[sub_filename] = sub_filename_final
2146 else:
2147 self.to_screen('[info] Writing video subtitles to: ' + sub_filename)
2148 if sub_info.get('data') is not None:
2149 try:
2150 # Use newline='' to prevent conversion of newline characters
2151 # See https://github.com/ytdl-org/youtube-dl/issues/10268
2152 with io.open(encodeFilename(sub_filename), 'w', encoding='utf-8', newline='') as subfile:
2153 subfile.write(sub_info['data'])
2154 files_to_move[sub_filename] = sub_filename_final
2155 except (OSError, IOError):
2156 self.report_error('Cannot write subtitles file ' + sub_filename)
2157 return
2158 else:
2159 try:
2160 dl(sub_filename, sub_info, subtitle=True)
2161 '''
2162 if self.params.get('sleep_interval_subtitles', False):
2163 dl(sub_filename, sub_info)
2164 else:
2165 sub_data = ie._request_webpage(
2166 sub_info['url'], info_dict['id'], note=False).read()
2167 with io.open(encodeFilename(sub_filename), 'wb') as subfile:
2168 subfile.write(sub_data)
2169 '''
2170 files_to_move[sub_filename] = sub_filename_final
2171 except (ExtractorError, IOError, OSError, ValueError, compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
2172 self.report_warning('Unable to download subtitle for "%s": %s' %
2173 (sub_lang, error_to_compat_str(err)))
2174 continue
2175
2176 if skip_dl:
2177 if self.params.get('convertsubtitles', False):
2178 # subconv = FFmpegSubtitlesConvertorPP(self, format=self.params.get('convertsubtitles'))
2179 filename_real_ext = os.path.splitext(full_filename)[1][1:]
2180 filename_wo_ext = (
2181 os.path.splitext(full_filename)[0]
2182 if filename_real_ext == info_dict['ext']
2183 else full_filename)
2184 afilename = '%s.%s' % (filename_wo_ext, self.params.get('convertsubtitles'))
2185 # if subconv.available:
2186 # info_dict['__postprocessors'].append(subconv)
2187 if os.path.exists(encodeFilename(afilename)):
2188 self.to_screen(
2189 '[download] %s has already been downloaded and '
2190 'converted' % afilename)
2191 else:
2192 try:
2193 self.post_process(full_filename, info_dict, files_to_move)
2194 except PostProcessingError as err:
2195 self.report_error('Postprocessing: %s' % str(err))
2196 return
2197
2198 if self.params.get('writeinfojson', False):
2199 infofn = self.prepare_filename(info_dict, 'infojson')
2200 if not ensure_dir_exists(encodeFilename(infofn)):
2201 return
2202 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(infofn)):
2203 self.to_screen('[info] Video metadata is already present')
2204 else:
2205 self.to_screen('[info] Writing video metadata as JSON to: ' + infofn)
2206 try:
2207 write_json_file(self.filter_requested_info(info_dict), infofn)
2208 except (OSError, IOError):
2209 self.report_error('Cannot write video metadata to JSON file ' + infofn)
2210 return
2211 info_dict['__infojson_filename'] = infofn
2212
2213 thumbfn = self.prepare_filename(info_dict, 'thumbnail')
2214 thumb_fn_temp = temp_filename if not skip_dl else thumbfn
2215 for thumb_ext in self._write_thumbnails(info_dict, thumb_fn_temp):
2216 thumb_filename_temp = replace_extension(thumb_fn_temp, thumb_ext, info_dict.get('ext'))
2217 thumb_filename = replace_extension(thumbfn, thumb_ext, info_dict.get('ext'))
2218 files_to_move[thumb_filename_temp] = info_dict['__thumbnail_filename'] = thumb_filename
2219
2220 # Write internet shortcut files
2221 url_link = webloc_link = desktop_link = False
2222 if self.params.get('writelink', False):
2223 if sys.platform == "darwin": # macOS.
2224 webloc_link = True
2225 elif sys.platform.startswith("linux"):
2226 desktop_link = True
2227 else: # if sys.platform in ['win32', 'cygwin']:
2228 url_link = True
2229 if self.params.get('writeurllink', False):
2230 url_link = True
2231 if self.params.get('writewebloclink', False):
2232 webloc_link = True
2233 if self.params.get('writedesktoplink', False):
2234 desktop_link = True
2235
2236 if url_link or webloc_link or desktop_link:
2237 if 'webpage_url' not in info_dict:
2238 self.report_error('Cannot write internet shortcut file because the "webpage_url" field is missing in the media information')
2239 return
2240 ascii_url = iri_to_uri(info_dict['webpage_url'])
2241
2242 def _write_link_file(extension, template, newline, embed_filename):
2243 linkfn = replace_extension(full_filename, extension, info_dict.get('ext'))
2244 if self.params.get('overwrites', True) and os.path.exists(encodeFilename(linkfn)):
2245 self.to_screen('[info] Internet shortcut is already present')
2246 else:
2247 try:
2248 self.to_screen('[info] Writing internet shortcut to: ' + linkfn)
2249 with io.open(encodeFilename(to_high_limit_path(linkfn)), 'w', encoding='utf-8', newline=newline) as linkfile:
2250 template_vars = {'url': ascii_url}
2251 if embed_filename:
2252 template_vars['filename'] = linkfn[:-(len(extension) + 1)]
2253 linkfile.write(template % template_vars)
2254 except (OSError, IOError):
2255 self.report_error('Cannot write internet shortcut ' + linkfn)
2256 return False
2257 return True
2258
2259 if url_link:
2260 if not _write_link_file('url', DOT_URL_LINK_TEMPLATE, '\r\n', embed_filename=False):
2261 return
2262 if webloc_link:
2263 if not _write_link_file('webloc', DOT_WEBLOC_LINK_TEMPLATE, '\n', embed_filename=False):
2264 return
2265 if desktop_link:
2266 if not _write_link_file('desktop', DOT_DESKTOP_LINK_TEMPLATE, '\n', embed_filename=True):
2267 return
2268
2269 # Download
2270 must_record_download_archive = False
2271 if not skip_dl:
2272 try:
2273
2274 def existing_file(*filepaths):
2275 ext = info_dict.get('ext')
2276 final_ext = self.params.get('final_ext', ext)
2277 existing_files = []
2278 for file in orderedSet(filepaths):
2279 if final_ext != ext:
2280 converted = replace_extension(file, final_ext, ext)
2281 if os.path.exists(encodeFilename(converted)):
2282 existing_files.append(converted)
2283 if os.path.exists(encodeFilename(file)):
2284 existing_files.append(file)
2285
2286 if not existing_files or self.params.get('overwrites', False):
2287 for file in orderedSet(existing_files):
2288 self.report_file_delete(file)
2289 os.remove(encodeFilename(file))
2290 return None
2291
2292 self.report_file_already_downloaded(existing_files[0])
2293 info_dict['ext'] = os.path.splitext(existing_files[0])[1][1:]
2294 return existing_files[0]
2295
2296 success = True
2297 if info_dict.get('requested_formats') is not None:
2298 downloaded = []
2299 merger = FFmpegMergerPP(self)
2300 if self.params.get('allow_unplayable_formats'):
2301 self.report_warning(
2302 'You have requested merging of multiple formats '
2303 'while also allowing unplayable formats to be downloaded. '
2304 'The formats won\'t be merged to prevent data corruption.')
2305 elif not merger.available:
2306 self.report_warning(
2307 'You have requested merging of multiple formats but ffmpeg is not installed. '
2308 'The formats won\'t be merged.')
2309
2310 def compatible_formats(formats):
2311 # TODO: some formats actually allow this (mkv, webm, ogg, mp4), but not all of them.
2312 video_formats = [format for format in formats if format.get('vcodec') != 'none']
2313 audio_formats = [format for format in formats if format.get('acodec') != 'none']
2314 if len(video_formats) > 2 or len(audio_formats) > 2:
2315 return False
2316
2317 # Check extension
2318 exts = set(format.get('ext') for format in formats)
2319 COMPATIBLE_EXTS = (
2320 set(('mp3', 'mp4', 'm4a', 'm4p', 'm4b', 'm4r', 'm4v', 'ismv', 'isma')),
2321 set(('webm',)),
2322 )
2323 for ext_sets in COMPATIBLE_EXTS:
2324 if ext_sets.issuperset(exts):
2325 return True
2326 # TODO: Check acodec/vcodec
2327 return False
2328
2329 requested_formats = info_dict['requested_formats']
2330 old_ext = info_dict['ext']
2331 if self.params.get('merge_output_format') is None and not compatible_formats(requested_formats):
2332 info_dict['ext'] = 'mkv'
2333 self.report_warning(
2334 'Requested formats are incompatible for merge and will be merged into mkv.')
2335
2336 def correct_ext(filename):
2337 filename_real_ext = os.path.splitext(filename)[1][1:]
2338 filename_wo_ext = (
2339 os.path.splitext(filename)[0]
2340 if filename_real_ext == old_ext
2341 else filename)
2342 return '%s.%s' % (filename_wo_ext, info_dict['ext'])
2343
2344 # Ensure filename always has a correct extension for successful merge
2345 full_filename = correct_ext(full_filename)
2346 temp_filename = correct_ext(temp_filename)
2347 dl_filename = existing_file(full_filename, temp_filename)
2348 info_dict['__real_download'] = False
2349 if dl_filename is None:
2350 for f in requested_formats:
2351 new_info = dict(info_dict)
2352 new_info.update(f)
2353 fname = prepend_extension(
2354 self.prepare_filename(new_info, 'temp'),
2355 'f%s' % f['format_id'], new_info['ext'])
2356 if not ensure_dir_exists(fname):
2357 return
2358 downloaded.append(fname)
2359 partial_success, real_download = dl(fname, new_info)
2360 info_dict['__real_download'] = info_dict['__real_download'] or real_download
2361 success = success and partial_success
2362 if merger.available and not self.params.get('allow_unplayable_formats'):
2363 info_dict['__postprocessors'].append(merger)
2364 info_dict['__files_to_merge'] = downloaded
2365 # Even if there were no downloads, it is being merged only now
2366 info_dict['__real_download'] = True
2367 else:
2368 for file in downloaded:
2369 files_to_move[file] = None
2370 else:
2371 # Just a single file
2372 dl_filename = existing_file(full_filename, temp_filename)
2373 if dl_filename is None:
2374 success, real_download = dl(temp_filename, info_dict)
2375 info_dict['__real_download'] = real_download
2376
2377 dl_filename = dl_filename or temp_filename
2378 info_dict['__finaldir'] = os.path.dirname(os.path.abspath(encodeFilename(full_filename)))
2379
2380 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
2381 self.report_error('unable to download video data: %s' % error_to_compat_str(err))
2382 return
2383 except (OSError, IOError) as err:
2384 raise UnavailableVideoError(err)
2385 except (ContentTooShortError, ) as err:
2386 self.report_error('content too short (expected %s bytes and served %s)' % (err.expected, err.downloaded))
2387 return
2388
2389 if success and full_filename != '-':
2390 # Fixup content
2391 fixup_policy = self.params.get('fixup')
2392 if fixup_policy is None:
2393 fixup_policy = 'detect_or_warn'
2394
2395 INSTALL_FFMPEG_MESSAGE = 'Install ffmpeg to fix this automatically.'
2396
2397 stretched_ratio = info_dict.get('stretched_ratio')
2398 if stretched_ratio is not None and stretched_ratio != 1:
2399 if fixup_policy == 'warn':
2400 self.report_warning('%s: Non-uniform pixel ratio (%s)' % (
2401 info_dict['id'], stretched_ratio))
2402 elif fixup_policy == 'detect_or_warn':
2403 stretched_pp = FFmpegFixupStretchedPP(self)
2404 if stretched_pp.available:
2405 info_dict['__postprocessors'].append(stretched_pp)
2406 else:
2407 self.report_warning(
2408 '%s: Non-uniform pixel ratio (%s). %s'
2409 % (info_dict['id'], stretched_ratio, INSTALL_FFMPEG_MESSAGE))
2410 else:
2411 assert fixup_policy in ('ignore', 'never')
2412
2413 if (info_dict.get('requested_formats') is None
2414 and info_dict.get('container') == 'm4a_dash'
2415 and info_dict.get('ext') == 'm4a'):
2416 if fixup_policy == 'warn':
2417 self.report_warning(
2418 '%s: writing DASH m4a. '
2419 'Only some players support this container.'
2420 % info_dict['id'])
2421 elif fixup_policy == 'detect_or_warn':
2422 fixup_pp = FFmpegFixupM4aPP(self)
2423 if fixup_pp.available:
2424 info_dict['__postprocessors'].append(fixup_pp)
2425 else:
2426 self.report_warning(
2427 '%s: writing DASH m4a. '
2428 'Only some players support this container. %s'
2429 % (info_dict['id'], INSTALL_FFMPEG_MESSAGE))
2430 else:
2431 assert fixup_policy in ('ignore', 'never')
2432
2433 if (info_dict.get('protocol') == 'm3u8_native'
2434 or info_dict.get('protocol') == 'm3u8'
2435 and self.params.get('hls_prefer_native')):
2436 if fixup_policy == 'warn':
2437 self.report_warning('%s: malformed AAC bitstream detected.' % (
2438 info_dict['id']))
2439 elif fixup_policy == 'detect_or_warn':
2440 fixup_pp = FFmpegFixupM3u8PP(self)
2441 if fixup_pp.available:
2442 info_dict['__postprocessors'].append(fixup_pp)
2443 else:
2444 self.report_warning(
2445 '%s: malformed AAC bitstream detected. %s'
2446 % (info_dict['id'], INSTALL_FFMPEG_MESSAGE))
2447 else:
2448 assert fixup_policy in ('ignore', 'never')
2449
2450 try:
2451 self.post_process(dl_filename, info_dict, files_to_move)
2452 except PostProcessingError as err:
2453 self.report_error('Postprocessing: %s' % str(err))
2454 return
2455 try:
2456 for ph in self._post_hooks:
2457 ph(full_filename)
2458 except Exception as err:
2459 self.report_error('post hooks: %s' % str(err))
2460 return
2461 must_record_download_archive = True
2462
2463 if must_record_download_archive or self.params.get('force_write_download_archive', False):
2464 self.record_download_archive(info_dict)
2465 max_downloads = self.params.get('max_downloads')
2466 if max_downloads is not None and self._num_downloads >= int(max_downloads):
2467 raise MaxDownloadsReached()
2468
2469 def download(self, url_list):
2470 """Download a given list of URLs."""
2471 outtmpl = self.outtmpl_dict['default']
2472 if (len(url_list) > 1
2473 and outtmpl != '-'
2474 and '%' not in outtmpl
2475 and self.params.get('max_downloads') != 1):
2476 raise SameFileError(outtmpl)
2477
2478 for url in url_list:
2479 try:
2480 # It also downloads the videos
2481 res = self.extract_info(
2482 url, force_generic_extractor=self.params.get('force_generic_extractor', False))
2483 except UnavailableVideoError:
2484 self.report_error('unable to download video')
2485 except MaxDownloadsReached:
2486 self.to_screen('[info] Maximum number of downloaded files reached')
2487 raise
2488 except ExistingVideoReached:
2489 self.to_screen('[info] Encountered a file that is already in the archive, stopping due to --break-on-existing')
2490 raise
2491 except RejectedVideoReached:
2492 self.to_screen('[info] Encountered a file that did not match filter, stopping due to --break-on-reject')
2493 raise
2494 else:
2495 if self.params.get('dump_single_json', False):
2496 self.to_stdout(json.dumps(res))
2497
2498 return self._download_retcode
2499
2500 def download_with_info_file(self, info_filename):
2501 with contextlib.closing(fileinput.FileInput(
2502 [info_filename], mode='r',
2503 openhook=fileinput.hook_encoded('utf-8'))) as f:
2504 # FileInput doesn't have a read method, we can't call json.load
2505 info = self.filter_requested_info(json.loads('\n'.join(f)))
2506 try:
2507 self.process_ie_result(info, download=True)
2508 except DownloadError:
2509 webpage_url = info.get('webpage_url')
2510 if webpage_url is not None:
2511 self.report_warning('The info failed to download, trying with "%s"' % webpage_url)
2512 return self.download([webpage_url])
2513 else:
2514 raise
2515 return self._download_retcode
2516
2517 @staticmethod
2518 def filter_requested_info(info_dict):
2519 fields_to_remove = ('requested_formats', 'requested_subtitles')
2520 return dict(
2521 (k, v) for k, v in info_dict.items()
2522 if (k[0] != '_' or k == '_type') and k not in fields_to_remove)
2523
2524 def run_pp(self, pp, infodict, files_to_move={}):
2525 files_to_delete = []
2526 files_to_delete, infodict = pp.run(infodict)
2527 if not files_to_delete:
2528 return files_to_move, infodict
2529
2530 if self.params.get('keepvideo', False):
2531 for f in files_to_delete:
2532 files_to_move.setdefault(f, '')
2533 else:
2534 for old_filename in set(files_to_delete):
2535 self.to_screen('Deleting original file %s (pass -k to keep)' % old_filename)
2536 try:
2537 os.remove(encodeFilename(old_filename))
2538 except (IOError, OSError):
2539 self.report_warning('Unable to remove downloaded original file')
2540 if old_filename in files_to_move:
2541 del files_to_move[old_filename]
2542 return files_to_move, infodict
2543
2544 def pre_process(self, ie_info):
2545 info = dict(ie_info)
2546 for pp in self._pps['beforedl']:
2547 info = self.run_pp(pp, info)[1]
2548 return info
2549
2550 def post_process(self, filename, ie_info, files_to_move={}):
2551 """Run all the postprocessors on the given file."""
2552 info = dict(ie_info)
2553 info['filepath'] = filename
2554 info['__files_to_move'] = {}
2555
2556 for pp in ie_info.get('__postprocessors', []) + self._pps['normal']:
2557 files_to_move, info = self.run_pp(pp, info, files_to_move)
2558 info = self.run_pp(MoveFilesAfterDownloadPP(self, files_to_move), info)[1]
2559 for pp in self._pps['aftermove']:
2560 info = self.run_pp(pp, info, {})[1]
2561
2562 def _make_archive_id(self, info_dict):
2563 video_id = info_dict.get('id')
2564 if not video_id:
2565 return
2566 # Future-proof against any change in case
2567 # and backwards compatibility with prior versions
2568 extractor = info_dict.get('extractor_key') or info_dict.get('ie_key') # key in a playlist
2569 if extractor is None:
2570 url = str_or_none(info_dict.get('url'))
2571 if not url:
2572 return
2573 # Try to find matching extractor for the URL and take its ie_key
2574 for ie in self._ies:
2575 if ie.suitable(url):
2576 extractor = ie.ie_key()
2577 break
2578 else:
2579 return
2580 return '%s %s' % (extractor.lower(), video_id)
2581
2582 def in_download_archive(self, info_dict):
2583 fn = self.params.get('download_archive')
2584 if fn is None:
2585 return False
2586
2587 vid_id = self._make_archive_id(info_dict)
2588 if not vid_id:
2589 return False # Incomplete video information
2590
2591 return vid_id in self.archive
2592
2593 def record_download_archive(self, info_dict):
2594 fn = self.params.get('download_archive')
2595 if fn is None:
2596 return
2597 vid_id = self._make_archive_id(info_dict)
2598 assert vid_id
2599 with locked_file(fn, 'a', encoding='utf-8') as archive_file:
2600 archive_file.write(vid_id + '\n')
2601 self.archive.add(vid_id)
2602
2603 @staticmethod
2604 def format_resolution(format, default='unknown'):
2605 if format.get('vcodec') == 'none':
2606 return 'audio only'
2607 if format.get('resolution') is not None:
2608 return format['resolution']
2609 if format.get('height') is not None:
2610 if format.get('width') is not None:
2611 res = '%sx%s' % (format['width'], format['height'])
2612 else:
2613 res = '%sp' % format['height']
2614 elif format.get('width') is not None:
2615 res = '%dx?' % format['width']
2616 else:
2617 res = default
2618 return res
2619
2620 def _format_note(self, fdict):
2621 res = ''
2622 if fdict.get('ext') in ['f4f', 'f4m']:
2623 res += '(unsupported) '
2624 if fdict.get('language'):
2625 if res:
2626 res += ' '
2627 res += '[%s] ' % fdict['language']
2628 if fdict.get('format_note') is not None:
2629 res += fdict['format_note'] + ' '
2630 if fdict.get('tbr') is not None:
2631 res += '%4dk ' % fdict['tbr']
2632 if fdict.get('container') is not None:
2633 if res:
2634 res += ', '
2635 res += '%s container' % fdict['container']
2636 if (fdict.get('vcodec') is not None
2637 and fdict.get('vcodec') != 'none'):
2638 if res:
2639 res += ', '
2640 res += fdict['vcodec']
2641 if fdict.get('vbr') is not None:
2642 res += '@'
2643 elif fdict.get('vbr') is not None and fdict.get('abr') is not None:
2644 res += 'video@'
2645 if fdict.get('vbr') is not None:
2646 res += '%4dk' % fdict['vbr']
2647 if fdict.get('fps') is not None:
2648 if res:
2649 res += ', '
2650 res += '%sfps' % fdict['fps']
2651 if fdict.get('acodec') is not None:
2652 if res:
2653 res += ', '
2654 if fdict['acodec'] == 'none':
2655 res += 'video only'
2656 else:
2657 res += '%-5s' % fdict['acodec']
2658 elif fdict.get('abr') is not None:
2659 if res:
2660 res += ', '
2661 res += 'audio'
2662 if fdict.get('abr') is not None:
2663 res += '@%3dk' % fdict['abr']
2664 if fdict.get('asr') is not None:
2665 res += ' (%5dHz)' % fdict['asr']
2666 if fdict.get('filesize') is not None:
2667 if res:
2668 res += ', '
2669 res += format_bytes(fdict['filesize'])
2670 elif fdict.get('filesize_approx') is not None:
2671 if res:
2672 res += ', '
2673 res += '~' + format_bytes(fdict['filesize_approx'])
2674 return res
2675
2676 def _format_note_table(self, f):
2677 def join_fields(*vargs):
2678 return ', '.join((val for val in vargs if val != ''))
2679
2680 return join_fields(
2681 'UNSUPPORTED' if f.get('ext') in ('f4f', 'f4m') else '',
2682 format_field(f, 'language', '[%s]'),
2683 format_field(f, 'format_note'),
2684 format_field(f, 'container', ignore=(None, f.get('ext'))),
2685 format_field(f, 'asr', '%5dHz'))
2686
2687 def list_formats(self, info_dict):
2688 formats = info_dict.get('formats', [info_dict])
2689 new_format = self.params.get('listformats_table', False)
2690 if new_format:
2691 table = [
2692 [
2693 format_field(f, 'format_id'),
2694 format_field(f, 'ext'),
2695 self.format_resolution(f),
2696 format_field(f, 'fps', '%d'),
2697 '|',
2698 format_field(f, 'filesize', ' %s', func=format_bytes) + format_field(f, 'filesize_approx', '~%s', func=format_bytes),
2699 format_field(f, 'tbr', '%4dk'),
2700 f.get('protocol').replace('http_dash_segments', 'dash').replace("native", "n").replace('niconico_', ''),
2701 '|',
2702 format_field(f, 'vcodec', default='unknown').replace('none', ''),
2703 format_field(f, 'vbr', '%4dk'),
2704 format_field(f, 'acodec', default='unknown').replace('none', ''),
2705 format_field(f, 'abr', '%3dk'),
2706 format_field(f, 'asr', '%5dHz'),
2707 self._format_note_table(f)]
2708 for f in formats
2709 if f.get('preference') is None or f['preference'] >= -1000]
2710 header_line = ['ID', 'EXT', 'RESOLUTION', 'FPS', '|', ' FILESIZE', ' TBR', 'PROTO',
2711 '|', 'VCODEC', ' VBR', 'ACODEC', ' ABR', ' ASR', 'NOTE']
2712 else:
2713 table = [
2714 [
2715 format_field(f, 'format_id'),
2716 format_field(f, 'ext'),
2717 self.format_resolution(f),
2718 self._format_note(f)]
2719 for f in formats
2720 if f.get('preference') is None or f['preference'] >= -1000]
2721 header_line = ['format code', 'extension', 'resolution', 'note']
2722
2723 self.to_screen(
2724 '[info] Available formats for %s:\n%s' % (info_dict['id'], render_table(
2725 header_line,
2726 table,
2727 delim=new_format,
2728 extraGap=(0 if new_format else 1),
2729 hideEmpty=new_format)))
2730
2731 def list_thumbnails(self, info_dict):
2732 thumbnails = info_dict.get('thumbnails')
2733 if not thumbnails:
2734 self.to_screen('[info] No thumbnails present for %s' % info_dict['id'])
2735 return
2736
2737 self.to_screen(
2738 '[info] Thumbnails for %s:' % info_dict['id'])
2739 self.to_screen(render_table(
2740 ['ID', 'width', 'height', 'URL'],
2741 [[t['id'], t.get('width', 'unknown'), t.get('height', 'unknown'), t['url']] for t in thumbnails]))
2742
2743 def list_subtitles(self, video_id, subtitles, name='subtitles'):
2744 if not subtitles:
2745 self.to_screen('%s has no %s' % (video_id, name))
2746 return
2747 self.to_screen(
2748 'Available %s for %s:' % (name, video_id))
2749 self.to_screen(render_table(
2750 ['Language', 'formats'],
2751 [[lang, ', '.join(f['ext'] for f in reversed(formats))]
2752 for lang, formats in subtitles.items()]))
2753
2754 def urlopen(self, req):
2755 """ Start an HTTP download """
2756 if isinstance(req, compat_basestring):
2757 req = sanitized_Request(req)
2758 return self._opener.open(req, timeout=self._socket_timeout)
2759
2760 def print_debug_header(self):
2761 if not self.params.get('verbose'):
2762 return
2763
2764 if type('') is not compat_str:
2765 # Python 2.6 on SLES11 SP1 (https://github.com/ytdl-org/youtube-dl/issues/3326)
2766 self.report_warning(
2767 'Your Python is broken! Update to a newer and supported version')
2768
2769 stdout_encoding = getattr(
2770 sys.stdout, 'encoding', 'missing (%s)' % type(sys.stdout).__name__)
2771 encoding_str = (
2772 '[debug] Encodings: locale %s, fs %s, out %s, pref %s\n' % (
2773 locale.getpreferredencoding(),
2774 sys.getfilesystemencoding(),
2775 stdout_encoding,
2776 self.get_encoding()))
2777 write_string(encoding_str, encoding=None)
2778
2779 source = (
2780 '(exe)' if hasattr(sys, 'frozen')
2781 else '(zip)' if isinstance(globals().get('__loader__'), zipimporter)
2782 else '(source)' if os.path.basename(sys.argv[0]) == '__main__.py'
2783 else '')
2784 self._write_string('[debug] yt-dlp version %s %s\n' % (__version__, source))
2785 if _LAZY_LOADER:
2786 self._write_string('[debug] Lazy loading extractors enabled\n')
2787 if _PLUGIN_CLASSES:
2788 self._write_string(
2789 '[debug] Plugin Extractors: %s\n' % [ie.ie_key() for ie in _PLUGIN_CLASSES])
2790 try:
2791 sp = subprocess.Popen(
2792 ['git', 'rev-parse', '--short', 'HEAD'],
2793 stdout=subprocess.PIPE, stderr=subprocess.PIPE,
2794 cwd=os.path.dirname(os.path.abspath(__file__)))
2795 out, err = process_communicate_or_kill(sp)
2796 out = out.decode().strip()
2797 if re.match('[0-9a-f]+', out):
2798 self._write_string('[debug] Git HEAD: %s\n' % out)
2799 except Exception:
2800 try:
2801 sys.exc_clear()
2802 except Exception:
2803 pass
2804
2805 def python_implementation():
2806 impl_name = platform.python_implementation()
2807 if impl_name == 'PyPy' and hasattr(sys, 'pypy_version_info'):
2808 return impl_name + ' version %d.%d.%d' % sys.pypy_version_info[:3]
2809 return impl_name
2810
2811 self._write_string('[debug] Python version %s (%s %s) - %s\n' % (
2812 platform.python_version(),
2813 python_implementation(),
2814 platform.architecture()[0],
2815 platform_name()))
2816
2817 exe_versions = FFmpegPostProcessor.get_versions(self)
2818 exe_versions['rtmpdump'] = rtmpdump_version()
2819 exe_versions['phantomjs'] = PhantomJSwrapper._version()
2820 exe_str = ', '.join(
2821 '%s %s' % (exe, v)
2822 for exe, v in sorted(exe_versions.items())
2823 if v
2824 )
2825 if not exe_str:
2826 exe_str = 'none'
2827 self._write_string('[debug] exe versions: %s\n' % exe_str)
2828
2829 proxy_map = {}
2830 for handler in self._opener.handlers:
2831 if hasattr(handler, 'proxies'):
2832 proxy_map.update(handler.proxies)
2833 self._write_string('[debug] Proxy map: ' + compat_str(proxy_map) + '\n')
2834
2835 if self.params.get('call_home', False):
2836 ipaddr = self.urlopen('https://yt-dl.org/ip').read().decode('utf-8')
2837 self._write_string('[debug] Public IP address: %s\n' % ipaddr)
2838 return
2839 latest_version = self.urlopen(
2840 'https://yt-dl.org/latest/version').read().decode('utf-8')
2841 if version_tuple(latest_version) > version_tuple(__version__):
2842 self.report_warning(
2843 'You are using an outdated version (newest version: %s)! '
2844 'See https://yt-dl.org/update if you need help updating.' %
2845 latest_version)
2846
2847 def _setup_opener(self):
2848 timeout_val = self.params.get('socket_timeout')
2849 self._socket_timeout = 600 if timeout_val is None else float(timeout_val)
2850
2851 opts_cookiefile = self.params.get('cookiefile')
2852 opts_proxy = self.params.get('proxy')
2853
2854 if opts_cookiefile is None:
2855 self.cookiejar = compat_cookiejar.CookieJar()
2856 else:
2857 opts_cookiefile = expand_path(opts_cookiefile)
2858 self.cookiejar = YoutubeDLCookieJar(opts_cookiefile)
2859 if os.access(opts_cookiefile, os.R_OK):
2860 self.cookiejar.load(ignore_discard=True, ignore_expires=True)
2861
2862 cookie_processor = YoutubeDLCookieProcessor(self.cookiejar)
2863 if opts_proxy is not None:
2864 if opts_proxy == '':
2865 proxies = {}
2866 else:
2867 proxies = {'http': opts_proxy, 'https': opts_proxy}
2868 else:
2869 proxies = compat_urllib_request.getproxies()
2870 # Set HTTPS proxy to HTTP one if given (https://github.com/ytdl-org/youtube-dl/issues/805)
2871 if 'http' in proxies and 'https' not in proxies:
2872 proxies['https'] = proxies['http']
2873 proxy_handler = PerRequestProxyHandler(proxies)
2874
2875 debuglevel = 1 if self.params.get('debug_printtraffic') else 0
2876 https_handler = make_HTTPS_handler(self.params, debuglevel=debuglevel)
2877 ydlh = YoutubeDLHandler(self.params, debuglevel=debuglevel)
2878 redirect_handler = YoutubeDLRedirectHandler()
2879 data_handler = compat_urllib_request_DataHandler()
2880
2881 # When passing our own FileHandler instance, build_opener won't add the
2882 # default FileHandler and allows us to disable the file protocol, which
2883 # can be used for malicious purposes (see
2884 # https://github.com/ytdl-org/youtube-dl/issues/8227)
2885 file_handler = compat_urllib_request.FileHandler()
2886
2887 def file_open(*args, **kwargs):
2888 raise compat_urllib_error.URLError('file:// scheme is explicitly disabled in youtube-dlc for security reasons')
2889 file_handler.file_open = file_open
2890
2891 opener = compat_urllib_request.build_opener(
2892 proxy_handler, https_handler, cookie_processor, ydlh, redirect_handler, data_handler, file_handler)
2893
2894 # Delete the default user-agent header, which would otherwise apply in
2895 # cases where our custom HTTP handler doesn't come into play
2896 # (See https://github.com/ytdl-org/youtube-dl/issues/1309 for details)
2897 opener.addheaders = []
2898 self._opener = opener
2899
2900 def encode(self, s):
2901 if isinstance(s, bytes):
2902 return s # Already encoded
2903
2904 try:
2905 return s.encode(self.get_encoding())
2906 except UnicodeEncodeError as err:
2907 err.reason = err.reason + '. Check your system encoding configuration or use the --encoding option.'
2908 raise
2909
2910 def get_encoding(self):
2911 encoding = self.params.get('encoding')
2912 if encoding is None:
2913 encoding = preferredencoding()
2914 return encoding
2915
2916 def _write_thumbnails(self, info_dict, filename): # return the extensions
2917 write_all = self.params.get('write_all_thumbnails', False)
2918 thumbnails = []
2919 if write_all or self.params.get('writethumbnail', False):
2920 thumbnails = info_dict.get('thumbnails') or []
2921 multiple = write_all and len(thumbnails) > 1
2922
2923 ret = []
2924 for t in thumbnails[::1 if write_all else -1]:
2925 thumb_ext = determine_ext(t['url'], 'jpg')
2926 suffix = '%s.' % t['id'] if multiple else ''
2927 thumb_display_id = '%s ' % t['id'] if multiple else ''
2928 t['filename'] = thumb_filename = replace_extension(filename, suffix + thumb_ext, info_dict.get('ext'))
2929
2930 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(thumb_filename)):
2931 ret.append(suffix + thumb_ext)
2932 self.to_screen('[%s] %s: Thumbnail %sis already present' %
2933 (info_dict['extractor'], info_dict['id'], thumb_display_id))
2934 else:
2935 self.to_screen('[%s] %s: Downloading thumbnail %s...' %
2936 (info_dict['extractor'], info_dict['id'], thumb_display_id))
2937 try:
2938 uf = self.urlopen(t['url'])
2939 with open(encodeFilename(thumb_filename), 'wb') as thumbf:
2940 shutil.copyfileobj(uf, thumbf)
2941 ret.append(suffix + thumb_ext)
2942 self.to_screen('[%s] %s: Writing thumbnail %sto: %s' %
2943 (info_dict['extractor'], info_dict['id'], thumb_display_id, thumb_filename))
2944 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
2945 self.report_warning('Unable to download thumbnail "%s": %s' %
2946 (t['url'], error_to_compat_str(err)))
2947 if ret and not write_all:
2948 break
2949 return ret