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