]> jfr.im git - yt-dlp.git/blob - youtube_dlc/YoutubeDL.py
Fix typos (Closes #14)
[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
31 from .compat import (
32 compat_basestring,
33 compat_cookiejar,
34 compat_get_terminal_size,
35 compat_http_client,
36 compat_kwargs,
37 compat_numeric_types,
38 compat_os_name,
39 compat_str,
40 compat_tokenize_tokenize,
41 compat_urllib_error,
42 compat_urllib_request,
43 compat_urllib_request_DataHandler,
44 )
45 from .utils import (
46 age_restricted,
47 args_to_str,
48 ContentTooShortError,
49 date_from_str,
50 DateRange,
51 DEFAULT_OUTTMPL,
52 determine_ext,
53 determine_protocol,
54 DOT_DESKTOP_LINK_TEMPLATE,
55 DOT_URL_LINK_TEMPLATE,
56 DOT_WEBLOC_LINK_TEMPLATE,
57 DownloadError,
58 encode_compat_str,
59 encodeFilename,
60 error_to_compat_str,
61 expand_path,
62 ExtractorError,
63 format_bytes,
64 format_field,
65 formatSeconds,
66 GeoRestrictedError,
67 int_or_none,
68 iri_to_uri,
69 ISO3166Utils,
70 locked_file,
71 make_HTTPS_handler,
72 MaxDownloadsReached,
73 orderedSet,
74 PagedList,
75 parse_filesize,
76 PerRequestProxyHandler,
77 platform_name,
78 PostProcessingError,
79 preferredencoding,
80 prepend_extension,
81 register_socks_protocols,
82 render_table,
83 replace_extension,
84 SameFileError,
85 sanitize_filename,
86 sanitize_path,
87 sanitize_url,
88 sanitized_Request,
89 std_headers,
90 str_or_none,
91 subtitles_filename,
92 to_high_limit_path,
93 UnavailableVideoError,
94 url_basename,
95 version_tuple,
96 write_json_file,
97 write_string,
98 YoutubeDLCookieJar,
99 YoutubeDLCookieProcessor,
100 YoutubeDLHandler,
101 YoutubeDLRedirectHandler,
102 process_communicate_or_kill,
103 )
104 from .cache import Cache
105 from .extractor import get_info_extractor, gen_extractor_classes, _LAZY_LOADER
106 from .extractor.openload import PhantomJSwrapper
107 from .downloader import get_suitable_downloader
108 from .downloader.rtmp import rtmpdump_version
109 from .postprocessor import (
110 FFmpegFixupM3u8PP,
111 FFmpegFixupM4aPP,
112 FFmpegFixupStretchedPP,
113 FFmpegMergerPP,
114 FFmpegPostProcessor,
115 FFmpegSubtitlesConvertorPP,
116 get_postprocessor,
117 )
118 from .version import __version__
119
120 if compat_os_name == 'nt':
121 import ctypes
122
123
124 class YoutubeDL(object):
125 """YoutubeDL class.
126
127 YoutubeDL objects are the ones responsible of downloading the
128 actual video file and writing it to disk if the user has requested
129 it, among some other tasks. In most cases there should be one per
130 program. As, given a video URL, the downloader doesn't know how to
131 extract all the needed information, task that InfoExtractors do, it
132 has to pass the URL to one of them.
133
134 For this, YoutubeDL objects have a method that allows
135 InfoExtractors to be registered in a given order. When it is passed
136 a URL, the YoutubeDL object handles it to the first InfoExtractor it
137 finds that reports being able to handle it. The InfoExtractor extracts
138 all the information about the video or videos the URL refers to, and
139 YoutubeDL process the extracted information, possibly using a File
140 Downloader to download the video.
141
142 YoutubeDL objects accept a lot of parameters. In order not to saturate
143 the object constructor with arguments, it receives a dictionary of
144 options instead. These options are available through the params
145 attribute for the InfoExtractors to use. The YoutubeDL also
146 registers itself as the downloader in charge for the InfoExtractors
147 that are added to it, so this is a "mutual registration".
148
149 Available options:
150
151 username: Username for authentication purposes.
152 password: Password for authentication purposes.
153 videopassword: Password for accessing a video.
154 ap_mso: Adobe Pass multiple-system operator identifier.
155 ap_username: Multiple-system operator account username.
156 ap_password: Multiple-system operator account password.
157 usenetrc: Use netrc for authentication instead.
158 verbose: Print additional info to stdout.
159 quiet: Do not print messages to stdout.
160 no_warnings: Do not print out anything for warnings.
161 forceurl: Force printing final URL.
162 forcetitle: Force printing title.
163 forceid: Force printing ID.
164 forcethumbnail: Force printing thumbnail URL.
165 forcedescription: Force printing description.
166 forcefilename: Force printing final filename.
167 forceduration: Force printing duration.
168 forcejson: Force printing info_dict as JSON.
169 dump_single_json: Force printing the info_dict of the whole playlist
170 (or video) as a single JSON line.
171 force_write_download_archive: Force writing download archive regardless of
172 'skip_download' or 'simulate'.
173 simulate: Do not download the video files.
174 format: Video format code. see "FORMAT SELECTION" for more details.
175 format_sort: How to sort the video formats. see "Sorting Formats" for more details.
176 format_sort_force: Force the given format_sort. see "Sorting Formats" for more details.
177 allow_multiple_video_streams: Allow multiple video streams to be merged into a single file
178 allow_multiple_audio_streams: Allow multiple audio streams to be merged into a single file
179 outtmpl: Template for output names.
180 restrictfilenames: Do not allow "&" and spaces in file names.
181 trim_file_name: Limit length of filename (extension excluded).
182 ignoreerrors: Do not stop on download errors. (Default True when running youtube-dlc, but False when directly accessing YoutubeDL class)
183 force_generic_extractor: Force downloader to use the generic extractor
184 nooverwrites: Prevent overwriting files.
185 playliststart: Playlist item to start at.
186 playlistend: Playlist item to end at.
187 playlist_items: Specific indices of playlist to download.
188 playlistreverse: Download playlist items in reverse order.
189 playlistrandom: Download playlist items in random order.
190 matchtitle: Download only matching titles.
191 rejecttitle: Reject downloads for matching titles.
192 logger: Log messages to a logging.Logger instance.
193 logtostderr: Log messages to stderr instead of stdout.
194 writedescription: Write the video description to a .description file
195 writeinfojson: Write the video description to a .info.json file
196 writeannotations: Write the video annotations to a .annotations.xml file
197 writethumbnail: Write the thumbnail image to a file
198 write_all_thumbnails: Write all thumbnail formats to files
199 writelink: Write an internet shortcut file, depending on the
200 current platform (.url/.webloc/.desktop)
201 writeurllink: Write a Windows internet shortcut file (.url)
202 writewebloclink: Write a macOS internet shortcut file (.webloc)
203 writedesktoplink: Write a Linux internet shortcut file (.desktop)
204 writesubtitles: Write the video subtitles to a file
205 writeautomaticsub: Write the automatically generated subtitles to a file
206 allsubtitles: Downloads all the subtitles of the video
207 (requires writesubtitles or writeautomaticsub)
208 listsubtitles: Lists all available subtitles for the video
209 subtitlesformat: The format code for subtitles
210 subtitleslangs: List of languages of the subtitles to download
211 keepvideo: Keep the video file after post-processing
212 daterange: A DateRange object, download only if the upload_date is in the range.
213 skip_download: Skip the actual download of the video file
214 cachedir: Location of the cache files in the filesystem.
215 False to disable filesystem cache.
216 noplaylist: Download single video instead of a playlist if in doubt.
217 age_limit: An integer representing the user's age in years.
218 Unsuitable videos for the given age are skipped.
219 min_views: An integer representing the minimum view count the video
220 must have in order to not be skipped.
221 Videos without view count information are always
222 downloaded. None for no limit.
223 max_views: An integer representing the maximum view count.
224 Videos that are more popular than that are not
225 downloaded.
226 Videos without view count information are always
227 downloaded. None for no limit.
228 download_archive: File name of a file where all downloads are recorded.
229 Videos already present in the file are not downloaded
230 again.
231 break_on_existing: Stop the download process after attempting to download a file that's
232 in the archive.
233 cookiefile: File name where cookies should be read from and dumped to.
234 nocheckcertificate:Do not verify SSL certificates
235 prefer_insecure: Use HTTP instead of HTTPS to retrieve information.
236 At the moment, this is only supported by YouTube.
237 proxy: URL of the proxy server to use
238 geo_verification_proxy: URL of the proxy to use for IP address verification
239 on geo-restricted sites.
240 socket_timeout: Time to wait for unresponsive hosts, in seconds
241 bidi_workaround: Work around buggy terminals without bidirectional text
242 support, using fridibi
243 debug_printtraffic:Print out sent and received HTTP traffic
244 include_ads: Download ads as well
245 default_search: Prepend this string if an input url is not valid.
246 'auto' for elaborate guessing
247 encoding: Use this encoding instead of the system-specified.
248 extract_flat: Do not resolve URLs, return the immediate result.
249 Pass in 'in_playlist' to only show this behavior for
250 playlist items.
251 postprocessors: A list of dictionaries, each with an entry
252 * key: The name of the postprocessor. See
253 youtube_dlc/postprocessor/__init__.py for a list.
254 as well as any further keyword arguments for the
255 postprocessor.
256 post_hooks: A list of functions that get called as the final step
257 for each video file, after all postprocessors have been
258 called. The filename will be passed as the only argument.
259 progress_hooks: A list of functions that get called on download
260 progress, with a dictionary with the entries
261 * status: One of "downloading", "error", or "finished".
262 Check this first and ignore unknown values.
263
264 If status is one of "downloading", or "finished", the
265 following properties may also be present:
266 * filename: The final filename (always present)
267 * tmpfilename: The filename we're currently writing to
268 * downloaded_bytes: Bytes on disk
269 * total_bytes: Size of the whole file, None if unknown
270 * total_bytes_estimate: Guess of the eventual file size,
271 None if unavailable.
272 * elapsed: The number of seconds since download started.
273 * eta: The estimated time in seconds, None if unknown
274 * speed: The download speed in bytes/second, None if
275 unknown
276 * fragment_index: The counter of the currently
277 downloaded video fragment.
278 * fragment_count: The number of fragments (= individual
279 files that will be merged)
280
281 Progress hooks are guaranteed to be called at least once
282 (with status "finished") if the download is successful.
283 merge_output_format: Extension to use when merging formats.
284 fixup: Automatically correct known faults of the file.
285 One of:
286 - "never": do nothing
287 - "warn": only emit a warning
288 - "detect_or_warn": check whether we can do anything
289 about it, warn otherwise (default)
290 source_address: Client-side IP address to bind to.
291 call_home: Boolean, true iff we are allowed to contact the
292 youtube-dlc servers for debugging.
293 sleep_interval: Number of seconds to sleep before each download when
294 used alone or a lower bound of a range for randomized
295 sleep before each download (minimum possible number
296 of seconds to sleep) when used along with
297 max_sleep_interval.
298 max_sleep_interval:Upper bound of a range for randomized sleep before each
299 download (maximum possible number of seconds to sleep).
300 Must only be used along with sleep_interval.
301 Actual sleep time will be a random float from range
302 [sleep_interval; max_sleep_interval].
303 listformats: Print an overview of available video formats and exit.
304 list_thumbnails: Print a table of all thumbnails and exit.
305 match_filter: A function that gets called with the info_dict of
306 every video.
307 If it returns a message, the video is ignored.
308 If it returns None, the video is downloaded.
309 match_filter_func in utils.py is one example for this.
310 no_color: Do not emit color codes in output.
311 geo_bypass: Bypass geographic restriction via faking X-Forwarded-For
312 HTTP header
313 geo_bypass_country:
314 Two-letter ISO 3166-2 country code that will be used for
315 explicit geographic restriction bypassing via faking
316 X-Forwarded-For HTTP header
317 geo_bypass_ip_block:
318 IP range in CIDR notation that will be used similarly to
319 geo_bypass_country
320
321 The following options determine which downloader is picked:
322 external_downloader: Executable of the external downloader to call.
323 None or unset for standard (built-in) downloader.
324 hls_prefer_native: Use the native HLS downloader instead of ffmpeg/avconv
325 if True, otherwise use ffmpeg/avconv if False, otherwise
326 use downloader suggested by extractor if None.
327
328 The following parameters are not used by YoutubeDL itself, they are used by
329 the downloader (see youtube_dlc/downloader/common.py):
330 nopart, updatetime, buffersize, ratelimit, min_filesize, max_filesize, test,
331 noresizebuffer, retries, continuedl, noprogress, consoletitle,
332 xattr_set_filesize, external_downloader_args, hls_use_mpegts,
333 http_chunk_size.
334
335 The following options are used by the post processors:
336 prefer_ffmpeg: If False, use avconv instead of ffmpeg if both are available,
337 otherwise prefer ffmpeg.
338 ffmpeg_location: Location of the ffmpeg/avconv binary; either the path
339 to the binary or its containing directory.
340 postprocessor_args: A dictionary of postprocessor names (in lower case) and a list
341 of additional command-line arguments for the postprocessor.
342 Use 'default' as the name for arguments to passed to all PP.
343
344 The following options are used by the Youtube extractor:
345 youtube_include_dash_manifest: If True (default), DASH manifests and related
346 data will be downloaded and processed by extractor.
347 You can reduce network I/O by disabling it if you don't
348 care about DASH.
349 """
350
351 _NUMERIC_FIELDS = set((
352 'width', 'height', 'tbr', 'abr', 'asr', 'vbr', 'fps', 'filesize', 'filesize_approx',
353 'timestamp', 'upload_year', 'upload_month', 'upload_day',
354 'duration', 'view_count', 'like_count', 'dislike_count', 'repost_count',
355 'average_rating', 'comment_count', 'age_limit',
356 'start_time', 'end_time',
357 'chapter_number', 'season_number', 'episode_number',
358 'track_number', 'disc_number', 'release_year',
359 'playlist_index',
360 ))
361
362 params = None
363 _ies = []
364 _pps = []
365 _download_retcode = None
366 _num_downloads = None
367 _screen_file = None
368
369 def __init__(self, params=None, auto_init=True):
370 """Create a FileDownloader object with the given options."""
371 if params is None:
372 params = {}
373 self._ies = []
374 self._ies_instances = {}
375 self._pps = []
376 self._post_hooks = []
377 self._progress_hooks = []
378 self._download_retcode = 0
379 self._num_downloads = 0
380 self._screen_file = [sys.stdout, sys.stderr][params.get('logtostderr', False)]
381 self._err_file = sys.stderr
382 self.params = {
383 # Default parameters
384 'nocheckcertificate': False,
385 }
386 self.params.update(params)
387 self.cache = Cache(self)
388 self.archive = set()
389
390 """Preload the archive, if any is specified"""
391 def preload_download_archive(self):
392 fn = self.params.get('download_archive')
393 if fn is None:
394 return False
395 try:
396 with locked_file(fn, 'r', encoding='utf-8') as archive_file:
397 for line in archive_file:
398 self.archive.add(line.strip())
399 except IOError as ioe:
400 if ioe.errno != errno.ENOENT:
401 raise
402 return False
403 return True
404
405 def check_deprecated(param, option, suggestion):
406 if self.params.get(param) is not None:
407 self.report_warning(
408 '%s is deprecated. Use %s instead.' % (option, suggestion))
409 return True
410 return False
411
412 if self.params.get('verbose'):
413 self.to_stdout('[debug] Loading archive file %r' % self.params.get('download_archive'))
414
415 preload_download_archive(self)
416
417 if check_deprecated('cn_verification_proxy', '--cn-verification-proxy', '--geo-verification-proxy'):
418 if self.params.get('geo_verification_proxy') is None:
419 self.params['geo_verification_proxy'] = self.params['cn_verification_proxy']
420
421 check_deprecated('autonumber_size', '--autonumber-size', 'output template with %(autonumber)0Nd, where N in the number of digits')
422 check_deprecated('autonumber', '--auto-number', '-o "%(autonumber)s-%(title)s.%(ext)s"')
423 check_deprecated('usetitle', '--title', '-o "%(title)s-%(id)s.%(ext)s"')
424
425 if params.get('bidi_workaround', False):
426 try:
427 import pty
428 master, slave = pty.openpty()
429 width = compat_get_terminal_size().columns
430 if width is None:
431 width_args = []
432 else:
433 width_args = ['-w', str(width)]
434 sp_kwargs = dict(
435 stdin=subprocess.PIPE,
436 stdout=slave,
437 stderr=self._err_file)
438 try:
439 self._output_process = subprocess.Popen(
440 ['bidiv'] + width_args, **sp_kwargs
441 )
442 except OSError:
443 self._output_process = subprocess.Popen(
444 ['fribidi', '-c', 'UTF-8'] + width_args, **sp_kwargs)
445 self._output_channel = os.fdopen(master, 'rb')
446 except OSError as ose:
447 if ose.errno == errno.ENOENT:
448 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 $PATH.')
449 else:
450 raise
451
452 if (sys.platform != 'win32'
453 and sys.getfilesystemencoding() in ['ascii', 'ANSI_X3.4-1968']
454 and not params.get('restrictfilenames', False)):
455 # Unicode filesystem API will throw errors (#1474, #13027)
456 self.report_warning(
457 'Assuming --restrict-filenames since file system encoding '
458 'cannot encode all characters. '
459 'Set the LC_ALL environment variable to fix this.')
460 self.params['restrictfilenames'] = True
461
462 if isinstance(params.get('outtmpl'), bytes):
463 self.report_warning(
464 'Parameter outtmpl is bytes, but should be a unicode string. '
465 'Put from __future__ import unicode_literals at the top of your code file or consider switching to Python 3.x.')
466
467 self._setup_opener()
468
469 if auto_init:
470 self.print_debug_header()
471 self.add_default_info_extractors()
472
473 for pp_def_raw in self.params.get('postprocessors', []):
474 pp_class = get_postprocessor(pp_def_raw['key'])
475 pp_def = dict(pp_def_raw)
476 del pp_def['key']
477 pp = pp_class(self, **compat_kwargs(pp_def))
478 self.add_post_processor(pp)
479
480 for ph in self.params.get('post_hooks', []):
481 self.add_post_hook(ph)
482
483 for ph in self.params.get('progress_hooks', []):
484 self.add_progress_hook(ph)
485
486 register_socks_protocols()
487
488 def warn_if_short_id(self, argv):
489 # short YouTube ID starting with dash?
490 idxs = [
491 i for i, a in enumerate(argv)
492 if re.match(r'^-[0-9A-Za-z_-]{10}$', a)]
493 if idxs:
494 correct_argv = (
495 ['youtube-dlc']
496 + [a for i, a in enumerate(argv) if i not in idxs]
497 + ['--'] + [argv[i] for i in idxs]
498 )
499 self.report_warning(
500 'Long argument string detected. '
501 'Use -- to separate parameters and URLs, like this:\n%s\n' %
502 args_to_str(correct_argv))
503
504 def add_info_extractor(self, ie):
505 """Add an InfoExtractor object to the end of the list."""
506 self._ies.append(ie)
507 if not isinstance(ie, type):
508 self._ies_instances[ie.ie_key()] = ie
509 ie.set_downloader(self)
510
511 def get_info_extractor(self, ie_key):
512 """
513 Get an instance of an IE with name ie_key, it will try to get one from
514 the _ies list, if there's no instance it will create a new one and add
515 it to the extractor list.
516 """
517 ie = self._ies_instances.get(ie_key)
518 if ie is None:
519 ie = get_info_extractor(ie_key)()
520 self.add_info_extractor(ie)
521 return ie
522
523 def add_default_info_extractors(self):
524 """
525 Add the InfoExtractors returned by gen_extractors to the end of the list
526 """
527 for ie in gen_extractor_classes():
528 self.add_info_extractor(ie)
529
530 def add_post_processor(self, pp):
531 """Add a PostProcessor object to the end of the chain."""
532 self._pps.append(pp)
533 pp.set_downloader(self)
534
535 def add_post_hook(self, ph):
536 """Add the post hook"""
537 self._post_hooks.append(ph)
538
539 def add_progress_hook(self, ph):
540 """Add the progress hook (currently only for the file downloader)"""
541 self._progress_hooks.append(ph)
542
543 def _bidi_workaround(self, message):
544 if not hasattr(self, '_output_channel'):
545 return message
546
547 assert hasattr(self, '_output_process')
548 assert isinstance(message, compat_str)
549 line_count = message.count('\n') + 1
550 self._output_process.stdin.write((message + '\n').encode('utf-8'))
551 self._output_process.stdin.flush()
552 res = ''.join(self._output_channel.readline().decode('utf-8')
553 for _ in range(line_count))
554 return res[:-len('\n')]
555
556 def to_screen(self, message, skip_eol=False):
557 """Print message to stdout if not in quiet mode."""
558 return self.to_stdout(message, skip_eol, check_quiet=True)
559
560 def _write_string(self, s, out=None):
561 write_string(s, out=out, encoding=self.params.get('encoding'))
562
563 def to_stdout(self, message, skip_eol=False, check_quiet=False):
564 """Print message to stdout if not in quiet mode."""
565 if self.params.get('logger'):
566 self.params['logger'].debug(message)
567 elif not check_quiet or not self.params.get('quiet', False):
568 message = self._bidi_workaround(message)
569 terminator = ['\n', ''][skip_eol]
570 output = message + terminator
571
572 self._write_string(output, self._screen_file)
573
574 def to_stderr(self, message):
575 """Print message to stderr."""
576 assert isinstance(message, compat_str)
577 if self.params.get('logger'):
578 self.params['logger'].error(message)
579 else:
580 message = self._bidi_workaround(message)
581 output = message + '\n'
582 self._write_string(output, self._err_file)
583
584 def to_console_title(self, message):
585 if not self.params.get('consoletitle', False):
586 return
587 if compat_os_name == 'nt':
588 if ctypes.windll.kernel32.GetConsoleWindow():
589 # c_wchar_p() might not be necessary if `message` is
590 # already of type unicode()
591 ctypes.windll.kernel32.SetConsoleTitleW(ctypes.c_wchar_p(message))
592 elif 'TERM' in os.environ:
593 self._write_string('\033[0;%s\007' % message, self._screen_file)
594
595 def save_console_title(self):
596 if not self.params.get('consoletitle', False):
597 return
598 if self.params.get('simulate', False):
599 return
600 if compat_os_name != 'nt' and 'TERM' in os.environ:
601 # Save the title on stack
602 self._write_string('\033[22;0t', self._screen_file)
603
604 def restore_console_title(self):
605 if not self.params.get('consoletitle', False):
606 return
607 if self.params.get('simulate', False):
608 return
609 if compat_os_name != 'nt' and 'TERM' in os.environ:
610 # Restore the title from stack
611 self._write_string('\033[23;0t', self._screen_file)
612
613 def __enter__(self):
614 self.save_console_title()
615 return self
616
617 def __exit__(self, *args):
618 self.restore_console_title()
619
620 if self.params.get('cookiefile') is not None:
621 self.cookiejar.save(ignore_discard=True, ignore_expires=True)
622
623 def trouble(self, message=None, tb=None):
624 """Determine action to take when a download problem appears.
625
626 Depending on if the downloader has been configured to ignore
627 download errors or not, this method may throw an exception or
628 not when errors are found, after printing the message.
629
630 tb, if given, is additional traceback information.
631 """
632 if message is not None:
633 self.to_stderr(message)
634 if self.params.get('verbose'):
635 if tb is None:
636 if sys.exc_info()[0]: # if .trouble has been called from an except block
637 tb = ''
638 if hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
639 tb += ''.join(traceback.format_exception(*sys.exc_info()[1].exc_info))
640 tb += encode_compat_str(traceback.format_exc())
641 else:
642 tb_data = traceback.format_list(traceback.extract_stack())
643 tb = ''.join(tb_data)
644 self.to_stderr(tb)
645 if not self.params.get('ignoreerrors', False):
646 if sys.exc_info()[0] and hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
647 exc_info = sys.exc_info()[1].exc_info
648 else:
649 exc_info = sys.exc_info()
650 raise DownloadError(message, exc_info)
651 self._download_retcode = 1
652
653 def report_warning(self, message):
654 '''
655 Print the message to stderr, it will be prefixed with 'WARNING:'
656 If stderr is a tty file the 'WARNING:' will be colored
657 '''
658 if self.params.get('logger') is not None:
659 self.params['logger'].warning(message)
660 else:
661 if self.params.get('no_warnings'):
662 return
663 if not self.params.get('no_color') and self._err_file.isatty() and compat_os_name != 'nt':
664 _msg_header = '\033[0;33mWARNING:\033[0m'
665 else:
666 _msg_header = 'WARNING:'
667 warning_message = '%s %s' % (_msg_header, message)
668 self.to_stderr(warning_message)
669
670 def report_error(self, message, tb=None):
671 '''
672 Do the same as trouble, but prefixes the message with 'ERROR:', colored
673 in red if stderr is a tty file.
674 '''
675 if not self.params.get('no_color') and self._err_file.isatty() and compat_os_name != 'nt':
676 _msg_header = '\033[0;31mERROR:\033[0m'
677 else:
678 _msg_header = 'ERROR:'
679 error_message = '%s %s' % (_msg_header, message)
680 self.trouble(error_message, tb)
681
682 def report_file_already_downloaded(self, file_name):
683 """Report file has already been fully downloaded."""
684 try:
685 self.to_screen('[download] %s has already been downloaded' % file_name)
686 except UnicodeEncodeError:
687 self.to_screen('[download] The file has already been downloaded')
688
689 def prepare_filename(self, info_dict):
690 """Generate the output filename."""
691 try:
692 template_dict = dict(info_dict)
693
694 template_dict['epoch'] = int(time.time())
695 autonumber_size = self.params.get('autonumber_size')
696 if autonumber_size is None:
697 autonumber_size = 5
698 template_dict['autonumber'] = self.params.get('autonumber_start', 1) - 1 + self._num_downloads
699 if template_dict.get('resolution') is None:
700 if template_dict.get('width') and template_dict.get('height'):
701 template_dict['resolution'] = '%dx%d' % (template_dict['width'], template_dict['height'])
702 elif template_dict.get('height'):
703 template_dict['resolution'] = '%sp' % template_dict['height']
704 elif template_dict.get('width'):
705 template_dict['resolution'] = '%dx?' % template_dict['width']
706
707 sanitize = lambda k, v: sanitize_filename(
708 compat_str(v),
709 restricted=self.params.get('restrictfilenames'),
710 is_id=(k == 'id' or k.endswith('_id')))
711 template_dict = dict((k, v if isinstance(v, compat_numeric_types) else sanitize(k, v))
712 for k, v in template_dict.items()
713 if v is not None and not isinstance(v, (list, tuple, dict)))
714 template_dict = collections.defaultdict(lambda: 'NA', template_dict)
715
716 outtmpl = self.params.get('outtmpl', DEFAULT_OUTTMPL)
717
718 # For fields playlist_index and autonumber convert all occurrences
719 # of %(field)s to %(field)0Nd for backward compatibility
720 field_size_compat_map = {
721 'playlist_index': len(str(template_dict['n_entries'])),
722 'autonumber': autonumber_size,
723 }
724 FIELD_SIZE_COMPAT_RE = r'(?<!%)%\((?P<field>autonumber|playlist_index)\)s'
725 mobj = re.search(FIELD_SIZE_COMPAT_RE, outtmpl)
726 if mobj:
727 outtmpl = re.sub(
728 FIELD_SIZE_COMPAT_RE,
729 r'%%(\1)0%dd' % field_size_compat_map[mobj.group('field')],
730 outtmpl)
731
732 # Missing numeric fields used together with integer presentation types
733 # in format specification will break the argument substitution since
734 # string 'NA' is returned for missing fields. We will patch output
735 # template for missing fields to meet string presentation type.
736 for numeric_field in self._NUMERIC_FIELDS:
737 if numeric_field not in template_dict:
738 # As of [1] format syntax is:
739 # %[mapping_key][conversion_flags][minimum_width][.precision][length_modifier]type
740 # 1. https://docs.python.org/2/library/stdtypes.html#string-formatting
741 FORMAT_RE = r'''(?x)
742 (?<!%)
743 %
744 \({0}\) # mapping key
745 (?:[#0\-+ ]+)? # conversion flags (optional)
746 (?:\d+)? # minimum field width (optional)
747 (?:\.\d+)? # precision (optional)
748 [hlL]? # length modifier (optional)
749 [diouxXeEfFgGcrs%] # conversion type
750 '''
751 outtmpl = re.sub(
752 FORMAT_RE.format(numeric_field),
753 r'%({0})s'.format(numeric_field), outtmpl)
754
755 # expand_path translates '%%' into '%' and '$$' into '$'
756 # correspondingly that is not what we want since we need to keep
757 # '%%' intact for template dict substitution step. Working around
758 # with boundary-alike separator hack.
759 sep = ''.join([random.choice(ascii_letters) for _ in range(32)])
760 outtmpl = outtmpl.replace('%%', '%{0}%'.format(sep)).replace('$$', '${0}$'.format(sep))
761
762 # outtmpl should be expand_path'ed before template dict substitution
763 # because meta fields may contain env variables we don't want to
764 # be expanded. For example, for outtmpl "%(title)s.%(ext)s" and
765 # title "Hello $PATH", we don't want `$PATH` to be expanded.
766 filename = expand_path(outtmpl).replace(sep, '') % template_dict
767
768 # https://github.com/blackjack4494/youtube-dlc/issues/85
769 trim_file_name = self.params.get('trim_file_name', False)
770 if trim_file_name:
771 fn_groups = filename.rsplit('.')
772 ext = fn_groups[-1]
773 sub_ext = ''
774 if len(fn_groups) > 2:
775 sub_ext = fn_groups[-2]
776 filename = '.'.join(filter(None, [fn_groups[0][:trim_file_name], sub_ext, ext]))
777
778 # Temporary fix for #4787
779 # 'Treat' all problem characters by passing filename through preferredencoding
780 # to workaround encoding issues with subprocess on python2 @ Windows
781 if sys.version_info < (3, 0) and sys.platform == 'win32':
782 filename = encodeFilename(filename, True).decode(preferredencoding())
783 return sanitize_path(filename)
784 except ValueError as err:
785 self.report_error('Error in output template: ' + str(err) + ' (encoding: ' + repr(preferredencoding()) + ')')
786 return None
787
788 def _match_entry(self, info_dict, incomplete):
789 """ Returns None if the file should be downloaded """
790
791 video_title = info_dict.get('title', info_dict.get('id', 'video'))
792 if 'title' in info_dict:
793 # This can happen when we're just evaluating the playlist
794 title = info_dict['title']
795 matchtitle = self.params.get('matchtitle', False)
796 if matchtitle:
797 if not re.search(matchtitle, title, re.IGNORECASE):
798 return '"' + title + '" title did not match pattern "' + matchtitle + '"'
799 rejecttitle = self.params.get('rejecttitle', False)
800 if rejecttitle:
801 if re.search(rejecttitle, title, re.IGNORECASE):
802 return '"' + title + '" title matched reject pattern "' + rejecttitle + '"'
803 date = info_dict.get('upload_date')
804 if date is not None:
805 dateRange = self.params.get('daterange', DateRange())
806 if date not in dateRange:
807 return '%s upload date is not in range %s' % (date_from_str(date).isoformat(), dateRange)
808 view_count = info_dict.get('view_count')
809 if view_count is not None:
810 min_views = self.params.get('min_views')
811 if min_views is not None and view_count < min_views:
812 return 'Skipping %s, because it has not reached minimum view count (%d/%d)' % (video_title, view_count, min_views)
813 max_views = self.params.get('max_views')
814 if max_views is not None and view_count > max_views:
815 return 'Skipping %s, because it has exceeded the maximum view count (%d/%d)' % (video_title, view_count, max_views)
816 if age_restricted(info_dict.get('age_limit'), self.params.get('age_limit')):
817 return 'Skipping "%s" because it is age restricted' % video_title
818 if self.in_download_archive(info_dict):
819 return '%s has already been recorded in archive' % video_title
820
821 if not incomplete:
822 match_filter = self.params.get('match_filter')
823 if match_filter is not None:
824 ret = match_filter(info_dict)
825 if ret is not None:
826 return ret
827
828 return None
829
830 @staticmethod
831 def add_extra_info(info_dict, extra_info):
832 '''Set the keys from extra_info in info dict if they are missing'''
833 for key, value in extra_info.items():
834 info_dict.setdefault(key, value)
835
836 def extract_info(self, url, download=True, ie_key=None, info_dict=None, extra_info={},
837 process=True, force_generic_extractor=False):
838 '''
839 Returns a list with a dictionary for each video we find.
840 If 'download', also downloads the videos.
841 extra_info is a dict containing the extra values to add to each result
842 '''
843
844 if not ie_key and force_generic_extractor:
845 ie_key = 'Generic'
846
847 if ie_key:
848 ies = [self.get_info_extractor(ie_key)]
849 else:
850 ies = self._ies
851
852 for ie in ies:
853 if not ie.suitable(url):
854 continue
855
856 ie_key = ie.ie_key()
857 ie = self.get_info_extractor(ie_key)
858 if not ie.working():
859 self.report_warning('The program functionality for this site has been marked as broken, '
860 'and will probably not work.')
861
862 try:
863 temp_id = ie.extract_id(url) if callable(getattr(ie, 'extract_id', None)) else ie._match_id(url)
864 except (AssertionError, IndexError, AttributeError):
865 temp_id = None
866 if temp_id is not None and self.in_download_archive({'id': temp_id, 'ie_key': ie_key}):
867 self.to_screen("[%s] %s: has already been recorded in archive" % (
868 ie_key, temp_id))
869 break
870
871 return self.__extract_info(url, ie, download, extra_info, process, info_dict)
872
873 else:
874 self.report_error('no suitable InfoExtractor for URL %s' % url)
875
876 def __handle_extraction_exceptions(func):
877 def wrapper(self, *args, **kwargs):
878 try:
879 return func(self, *args, **kwargs)
880 except GeoRestrictedError as e:
881 msg = e.msg
882 if e.countries:
883 msg += '\nThis video is available in %s.' % ', '.join(
884 map(ISO3166Utils.short2full, e.countries))
885 msg += '\nYou might want to use a VPN or a proxy server (with --proxy) to workaround.'
886 self.report_error(msg)
887 except ExtractorError as e: # An error we somewhat expected
888 self.report_error(compat_str(e), e.format_traceback())
889 except MaxDownloadsReached:
890 raise
891 except Exception as e:
892 if self.params.get('ignoreerrors', False):
893 self.report_error(error_to_compat_str(e), tb=encode_compat_str(traceback.format_exc()))
894 else:
895 raise
896 return wrapper
897
898 @__handle_extraction_exceptions
899 def __extract_info(self, url, ie, download, extra_info, process, info_dict):
900 ie_result = ie.extract(url)
901 if ie_result is None: # Finished already (backwards compatibility; listformats and friends should be moved here)
902 return
903 if isinstance(ie_result, list):
904 # Backwards compatibility: old IE result format
905 ie_result = {
906 '_type': 'compat_list',
907 'entries': ie_result,
908 }
909 if info_dict:
910 if info_dict.get('id'):
911 ie_result['id'] = info_dict['id']
912 if info_dict.get('title'):
913 ie_result['title'] = info_dict['title']
914 self.add_default_extra_info(ie_result, ie, url)
915 if process:
916 return self.process_ie_result(ie_result, download, extra_info)
917 else:
918 return ie_result
919
920 def add_default_extra_info(self, ie_result, ie, url):
921 self.add_extra_info(ie_result, {
922 'extractor': ie.IE_NAME,
923 'webpage_url': url,
924 'duration_string': (
925 formatSeconds(ie_result['duration'], '-')
926 if ie_result.get('duration', None) is not None
927 else None),
928 'webpage_url_basename': url_basename(url),
929 'extractor_key': ie.ie_key(),
930 })
931
932 def process_ie_result(self, ie_result, download=True, extra_info={}):
933 """
934 Take the result of the ie(may be modified) and resolve all unresolved
935 references (URLs, playlist items).
936
937 It will also download the videos if 'download'.
938 Returns the resolved ie_result.
939 """
940 result_type = ie_result.get('_type', 'video')
941
942 if result_type in ('url', 'url_transparent'):
943 ie_result['url'] = sanitize_url(ie_result['url'])
944 extract_flat = self.params.get('extract_flat', False)
945 if ((extract_flat == 'in_playlist' and 'playlist' in extra_info)
946 or extract_flat is True):
947 self.__forced_printings(
948 ie_result, self.prepare_filename(ie_result),
949 incomplete=True)
950 return ie_result
951
952 if result_type == 'video':
953 self.add_extra_info(ie_result, extra_info)
954 return self.process_video_result(ie_result, download=download)
955 elif result_type == 'url':
956 # We have to add extra_info to the results because it may be
957 # contained in a playlist
958 return self.extract_info(ie_result['url'],
959 download, info_dict=ie_result,
960 ie_key=ie_result.get('ie_key'),
961 extra_info=extra_info)
962 elif result_type == 'url_transparent':
963 # Use the information from the embedding page
964 info = self.extract_info(
965 ie_result['url'], ie_key=ie_result.get('ie_key'),
966 extra_info=extra_info, download=False, process=False)
967
968 # extract_info may return None when ignoreerrors is enabled and
969 # extraction failed with an error, don't crash and return early
970 # in this case
971 if not info:
972 return info
973
974 force_properties = dict(
975 (k, v) for k, v in ie_result.items() if v is not None)
976 for f in ('_type', 'url', 'id', 'extractor', 'extractor_key', 'ie_key'):
977 if f in force_properties:
978 del force_properties[f]
979 new_result = info.copy()
980 new_result.update(force_properties)
981
982 # Extracted info may not be a video result (i.e.
983 # info.get('_type', 'video') != video) but rather an url or
984 # url_transparent. In such cases outer metadata (from ie_result)
985 # should be propagated to inner one (info). For this to happen
986 # _type of info should be overridden with url_transparent. This
987 # fixes issue from https://github.com/ytdl-org/youtube-dl/pull/11163.
988 if new_result.get('_type') == 'url':
989 new_result['_type'] = 'url_transparent'
990
991 return self.process_ie_result(
992 new_result, download=download, extra_info=extra_info)
993 elif result_type in ('playlist', 'multi_video'):
994 # We process each entry in the playlist
995 playlist = ie_result.get('title') or ie_result.get('id')
996 self.to_screen('[download] Downloading playlist: %s' % playlist)
997
998 playlist_results = []
999
1000 playliststart = self.params.get('playliststart', 1) - 1
1001 playlistend = self.params.get('playlistend')
1002 # For backwards compatibility, interpret -1 as whole list
1003 if playlistend == -1:
1004 playlistend = None
1005
1006 playlistitems_str = self.params.get('playlist_items')
1007 playlistitems = None
1008 if playlistitems_str is not None:
1009 def iter_playlistitems(format):
1010 for string_segment in format.split(','):
1011 if '-' in string_segment:
1012 start, end = string_segment.split('-')
1013 for item in range(int(start), int(end) + 1):
1014 yield int(item)
1015 else:
1016 yield int(string_segment)
1017 playlistitems = orderedSet(iter_playlistitems(playlistitems_str))
1018
1019 ie_entries = ie_result['entries']
1020
1021 def make_playlistitems_entries(list_ie_entries):
1022 num_entries = len(list_ie_entries)
1023 return [
1024 list_ie_entries[i - 1] for i in playlistitems
1025 if -num_entries <= i - 1 < num_entries]
1026
1027 def report_download(num_entries):
1028 self.to_screen(
1029 '[%s] playlist %s: Downloading %d videos' %
1030 (ie_result['extractor'], playlist, num_entries))
1031
1032 if isinstance(ie_entries, list):
1033 n_all_entries = len(ie_entries)
1034 if playlistitems:
1035 entries = make_playlistitems_entries(ie_entries)
1036 else:
1037 entries = ie_entries[playliststart:playlistend]
1038 n_entries = len(entries)
1039 self.to_screen(
1040 '[%s] playlist %s: Collected %d video ids (downloading %d of them)' %
1041 (ie_result['extractor'], playlist, n_all_entries, n_entries))
1042 elif isinstance(ie_entries, PagedList):
1043 if playlistitems:
1044 entries = []
1045 for item in playlistitems:
1046 entries.extend(ie_entries.getslice(
1047 item - 1, item
1048 ))
1049 else:
1050 entries = ie_entries.getslice(
1051 playliststart, playlistend)
1052 n_entries = len(entries)
1053 report_download(n_entries)
1054 else: # iterable
1055 if playlistitems:
1056 entries = make_playlistitems_entries(list(itertools.islice(
1057 ie_entries, 0, max(playlistitems))))
1058 else:
1059 entries = list(itertools.islice(
1060 ie_entries, playliststart, playlistend))
1061 n_entries = len(entries)
1062 report_download(n_entries)
1063
1064 if self.params.get('playlistreverse', False):
1065 entries = entries[::-1]
1066
1067 if self.params.get('playlistrandom', False):
1068 random.shuffle(entries)
1069
1070 x_forwarded_for = ie_result.get('__x_forwarded_for_ip')
1071
1072 for i, entry in enumerate(entries, 1):
1073 self.to_screen('[download] Downloading video %s of %s' % (i, n_entries))
1074 # This __x_forwarded_for_ip thing is a bit ugly but requires
1075 # minimal changes
1076 if x_forwarded_for:
1077 entry['__x_forwarded_for_ip'] = x_forwarded_for
1078 extra = {
1079 'n_entries': n_entries,
1080 'playlist': playlist,
1081 'playlist_id': ie_result.get('id'),
1082 'playlist_title': ie_result.get('title'),
1083 'playlist_uploader': ie_result.get('uploader'),
1084 'playlist_uploader_id': ie_result.get('uploader_id'),
1085 'playlist_index': playlistitems[i - 1] if playlistitems else i + playliststart,
1086 'extractor': ie_result['extractor'],
1087 'webpage_url': ie_result['webpage_url'],
1088 'webpage_url_basename': url_basename(ie_result['webpage_url']),
1089 'extractor_key': ie_result['extractor_key'],
1090 }
1091
1092 reason = self._match_entry(entry, incomplete=True)
1093 if reason is not None:
1094 if reason.endswith('has already been recorded in the archive') and self.params.get('break_on_existing'):
1095 print('[download] tried downloading a file that\'s already in the archive, stopping since --break-on-existing is set.')
1096 break
1097 else:
1098 self.to_screen('[download] ' + reason)
1099 continue
1100
1101 entry_result = self.__process_iterable_entry(entry, download, extra)
1102 # TODO: skip failed (empty) entries?
1103 playlist_results.append(entry_result)
1104 ie_result['entries'] = playlist_results
1105 self.to_screen('[download] Finished downloading playlist: %s' % playlist)
1106 return ie_result
1107 elif result_type == 'compat_list':
1108 self.report_warning(
1109 'Extractor %s returned a compat_list result. '
1110 'It needs to be updated.' % ie_result.get('extractor'))
1111
1112 def _fixup(r):
1113 self.add_extra_info(
1114 r,
1115 {
1116 'extractor': ie_result['extractor'],
1117 'webpage_url': ie_result['webpage_url'],
1118 'webpage_url_basename': url_basename(ie_result['webpage_url']),
1119 'extractor_key': ie_result['extractor_key'],
1120 }
1121 )
1122 return r
1123 ie_result['entries'] = [
1124 self.process_ie_result(_fixup(r), download, extra_info)
1125 for r in ie_result['entries']
1126 ]
1127 return ie_result
1128 else:
1129 raise Exception('Invalid result type: %s' % result_type)
1130
1131 @__handle_extraction_exceptions
1132 def __process_iterable_entry(self, entry, download, extra_info):
1133 return self.process_ie_result(
1134 entry, download=download, extra_info=extra_info)
1135
1136 def _build_format_filter(self, filter_spec):
1137 " Returns a function to filter the formats according to the filter_spec "
1138
1139 OPERATORS = {
1140 '<': operator.lt,
1141 '<=': operator.le,
1142 '>': operator.gt,
1143 '>=': operator.ge,
1144 '=': operator.eq,
1145 '!=': operator.ne,
1146 }
1147 operator_rex = re.compile(r'''(?x)\s*
1148 (?P<key>width|height|tbr|abr|vbr|asr|filesize|filesize_approx|fps)
1149 \s*(?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
1150 (?P<value>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?)
1151 $
1152 ''' % '|'.join(map(re.escape, OPERATORS.keys())))
1153 m = operator_rex.search(filter_spec)
1154 if m:
1155 try:
1156 comparison_value = int(m.group('value'))
1157 except ValueError:
1158 comparison_value = parse_filesize(m.group('value'))
1159 if comparison_value is None:
1160 comparison_value = parse_filesize(m.group('value') + 'B')
1161 if comparison_value is None:
1162 raise ValueError(
1163 'Invalid value %r in format specification %r' % (
1164 m.group('value'), filter_spec))
1165 op = OPERATORS[m.group('op')]
1166
1167 if not m:
1168 STR_OPERATORS = {
1169 '=': operator.eq,
1170 '^=': lambda attr, value: attr.startswith(value),
1171 '$=': lambda attr, value: attr.endswith(value),
1172 '*=': lambda attr, value: value in attr,
1173 }
1174 str_operator_rex = re.compile(r'''(?x)
1175 \s*(?P<key>[a-zA-Z0-9._-]+)
1176 \s*(?P<negation>!\s*)?(?P<op>%s)(?P<none_inclusive>\s*\?)?
1177 \s*(?P<value>[a-zA-Z0-9._-]+)
1178 \s*$
1179 ''' % '|'.join(map(re.escape, STR_OPERATORS.keys())))
1180 m = str_operator_rex.search(filter_spec)
1181 if m:
1182 comparison_value = m.group('value')
1183 str_op = STR_OPERATORS[m.group('op')]
1184 if m.group('negation'):
1185 op = lambda attr, value: not str_op(attr, value)
1186 else:
1187 op = str_op
1188
1189 if not m:
1190 raise ValueError('Invalid filter specification %r' % filter_spec)
1191
1192 def _filter(f):
1193 actual_value = f.get(m.group('key'))
1194 if actual_value is None:
1195 return m.group('none_inclusive')
1196 return op(actual_value, comparison_value)
1197 return _filter
1198
1199 def _default_format_spec(self, info_dict, download=True):
1200
1201 def can_merge():
1202 merger = FFmpegMergerPP(self)
1203 return merger.available and merger.can_merge()
1204
1205 prefer_best = (
1206 not self.params.get('simulate', False)
1207 and download
1208 and (
1209 not can_merge()
1210 or info_dict.get('is_live', False)
1211 or self.params.get('outtmpl', DEFAULT_OUTTMPL) == '-'))
1212
1213 return (
1214 'best/bestvideo+bestaudio'
1215 if prefer_best
1216 else 'bestvideo*+bestaudio/best'
1217 if not self.params.get('allow_multiple_audio_streams', False)
1218 else 'bestvideo+bestaudio/best')
1219
1220 def build_format_selector(self, format_spec):
1221 def syntax_error(note, start):
1222 message = (
1223 'Invalid format specification: '
1224 '{0}\n\t{1}\n\t{2}^'.format(note, format_spec, ' ' * start[1]))
1225 return SyntaxError(message)
1226
1227 PICKFIRST = 'PICKFIRST'
1228 MERGE = 'MERGE'
1229 SINGLE = 'SINGLE'
1230 GROUP = 'GROUP'
1231 FormatSelector = collections.namedtuple('FormatSelector', ['type', 'selector', 'filters'])
1232
1233 allow_multiple_streams = {'audio': self.params.get('allow_multiple_audio_streams', False),
1234 'video': self.params.get('allow_multiple_video_streams', False)}
1235
1236 def _parse_filter(tokens):
1237 filter_parts = []
1238 for type, string, start, _, _ in tokens:
1239 if type == tokenize.OP and string == ']':
1240 return ''.join(filter_parts)
1241 else:
1242 filter_parts.append(string)
1243
1244 def _remove_unused_ops(tokens):
1245 # Remove operators that we don't use and join them with the surrounding strings
1246 # for example: 'mp4' '-' 'baseline' '-' '16x9' is converted to 'mp4-baseline-16x9'
1247 ALLOWED_OPS = ('/', '+', ',', '(', ')')
1248 last_string, last_start, last_end, last_line = None, None, None, None
1249 for type, string, start, end, line in tokens:
1250 if type == tokenize.OP and string == '[':
1251 if last_string:
1252 yield tokenize.NAME, last_string, last_start, last_end, last_line
1253 last_string = None
1254 yield type, string, start, end, line
1255 # everything inside brackets will be handled by _parse_filter
1256 for type, string, start, end, line in tokens:
1257 yield type, string, start, end, line
1258 if type == tokenize.OP and string == ']':
1259 break
1260 elif type == tokenize.OP and string in ALLOWED_OPS:
1261 if last_string:
1262 yield tokenize.NAME, last_string, last_start, last_end, last_line
1263 last_string = None
1264 yield type, string, start, end, line
1265 elif type in [tokenize.NAME, tokenize.NUMBER, tokenize.OP]:
1266 if not last_string:
1267 last_string = string
1268 last_start = start
1269 last_end = end
1270 else:
1271 last_string += string
1272 if last_string:
1273 yield tokenize.NAME, last_string, last_start, last_end, last_line
1274
1275 def _parse_format_selection(tokens, inside_merge=False, inside_choice=False, inside_group=False):
1276 selectors = []
1277 current_selector = None
1278 for type, string, start, _, _ in tokens:
1279 # ENCODING is only defined in python 3.x
1280 if type == getattr(tokenize, 'ENCODING', None):
1281 continue
1282 elif type in [tokenize.NAME, tokenize.NUMBER]:
1283 current_selector = FormatSelector(SINGLE, string, [])
1284 elif type == tokenize.OP:
1285 if string == ')':
1286 if not inside_group:
1287 # ')' will be handled by the parentheses group
1288 tokens.restore_last_token()
1289 break
1290 elif inside_merge and string in ['/', ',']:
1291 tokens.restore_last_token()
1292 break
1293 elif inside_choice and string == ',':
1294 tokens.restore_last_token()
1295 break
1296 elif string == ',':
1297 if not current_selector:
1298 raise syntax_error('"," must follow a format selector', start)
1299 selectors.append(current_selector)
1300 current_selector = None
1301 elif string == '/':
1302 if not current_selector:
1303 raise syntax_error('"/" must follow a format selector', start)
1304 first_choice = current_selector
1305 second_choice = _parse_format_selection(tokens, inside_choice=True)
1306 current_selector = FormatSelector(PICKFIRST, (first_choice, second_choice), [])
1307 elif string == '[':
1308 if not current_selector:
1309 current_selector = FormatSelector(SINGLE, 'best', [])
1310 format_filter = _parse_filter(tokens)
1311 current_selector.filters.append(format_filter)
1312 elif string == '(':
1313 if current_selector:
1314 raise syntax_error('Unexpected "("', start)
1315 group = _parse_format_selection(tokens, inside_group=True)
1316 current_selector = FormatSelector(GROUP, group, [])
1317 elif string == '+':
1318 if not current_selector:
1319 raise syntax_error('Unexpected "+"', start)
1320 selector_1 = current_selector
1321 selector_2 = _parse_format_selection(tokens, inside_merge=True)
1322 if not selector_2:
1323 raise syntax_error('Expected a selector', start)
1324 current_selector = FormatSelector(MERGE, (selector_1, selector_2), [])
1325 else:
1326 raise syntax_error('Operator not recognized: "{0}"'.format(string), start)
1327 elif type == tokenize.ENDMARKER:
1328 break
1329 if current_selector:
1330 selectors.append(current_selector)
1331 return selectors
1332
1333 def _build_selector_function(selector):
1334 if isinstance(selector, list): # ,
1335 fs = [_build_selector_function(s) for s in selector]
1336
1337 def selector_function(ctx):
1338 for f in fs:
1339 for format in f(ctx):
1340 yield format
1341 return selector_function
1342
1343 elif selector.type == GROUP: # ()
1344 selector_function = _build_selector_function(selector.selector)
1345
1346 elif selector.type == PICKFIRST: # /
1347 fs = [_build_selector_function(s) for s in selector.selector]
1348
1349 def selector_function(ctx):
1350 for f in fs:
1351 picked_formats = list(f(ctx))
1352 if picked_formats:
1353 return picked_formats
1354 return []
1355
1356 elif selector.type == SINGLE: # atom
1357 format_spec = selector.selector if selector.selector is not None else 'best'
1358
1359 if format_spec == 'all':
1360 def selector_function(ctx):
1361 formats = list(ctx['formats'])
1362 if formats:
1363 for f in formats:
1364 yield f
1365
1366 else:
1367 format_fallback = False
1368 format_spec_obj = re.match(r'(best|worst|b|w)(video|audio|v|a)?(\*)?$', format_spec)
1369 if format_spec_obj is not None:
1370 format_idx = 0 if format_spec_obj.group(1)[0] == 'w' else -1
1371 format_type = format_spec_obj.group(2)[0] if format_spec_obj.group(2) else False
1372 not_format_type = 'v' if format_type == 'a' else 'a'
1373 format_modified = format_spec_obj.group(3) is not None
1374
1375 format_fallback = not format_type and not format_modified # for b, w
1376 filter_f = ((lambda f: f.get(format_type + 'codec') != 'none')
1377 if format_type and format_modified # bv*, ba*, wv*, wa*
1378 else (lambda f: f.get(not_format_type + 'codec') == 'none')
1379 if format_type # bv, ba, wv, wa
1380 else (lambda f: f.get('vcodec') != 'none' and f.get('acodec') != 'none')
1381 if not format_modified # b, w
1382 else None) # b*, w*
1383 else:
1384 format_idx = -1
1385 filter_f = ((lambda f: f.get('ext') == format_spec)
1386 if format_spec in ['mp4', 'flv', 'webm', '3gp', 'm4a', 'mp3', 'ogg', 'aac', 'wav'] # extension
1387 else (lambda f: f.get('format_id') == format_spec)) # id
1388
1389 def selector_function(ctx):
1390 formats = list(ctx['formats'])
1391 if not formats:
1392 return
1393 matches = list(filter(filter_f, formats)) if filter_f is not None else formats
1394 if matches:
1395 yield matches[format_idx]
1396 elif format_fallback == 'force' or (format_fallback and ctx['incomplete_formats']):
1397 # for extractors with incomplete formats (audio only (soundcloud)
1398 # or video only (imgur)) best/worst will fallback to
1399 # best/worst {video,audio}-only format
1400 yield formats[format_idx]
1401
1402 elif selector.type == MERGE: # +
1403 def _merge(formats_pair):
1404 format_1, format_2 = formats_pair
1405
1406 formats_info = []
1407 formats_info.extend(format_1.get('requested_formats', (format_1,)))
1408 formats_info.extend(format_2.get('requested_formats', (format_2,)))
1409
1410 if not allow_multiple_streams['video'] or not allow_multiple_streams['audio']:
1411 get_no_more = {"video": False, "audio": False}
1412 for (i, fmt_info) in enumerate(formats_info):
1413 for aud_vid in ["audio", "video"]:
1414 if not allow_multiple_streams[aud_vid] and fmt_info.get(aud_vid[0] + 'codec') != 'none':
1415 if get_no_more[aud_vid]:
1416 formats_info.pop(i)
1417 get_no_more[aud_vid] = True
1418
1419 if len(formats_info) == 1:
1420 return formats_info[0]
1421
1422 video_fmts = [fmt_info for fmt_info in formats_info if fmt_info.get('vcodec') != 'none']
1423 audio_fmts = [fmt_info for fmt_info in formats_info if fmt_info.get('acodec') != 'none']
1424
1425 the_only_video = video_fmts[0] if len(video_fmts) == 1 else None
1426 the_only_audio = audio_fmts[0] if len(audio_fmts) == 1 else None
1427
1428 output_ext = self.params.get('merge_output_format')
1429 if not output_ext:
1430 if the_only_video:
1431 output_ext = the_only_video['ext']
1432 elif the_only_audio and not video_fmts:
1433 output_ext = the_only_audio['ext']
1434 else:
1435 output_ext = 'mkv'
1436
1437 new_dict = {
1438 'requested_formats': formats_info,
1439 'format': '+'.join(fmt_info.get('format') for fmt_info in formats_info),
1440 'format_id': '+'.join(fmt_info.get('format_id') for fmt_info in formats_info),
1441 'ext': output_ext,
1442 }
1443
1444 if the_only_video:
1445 new_dict.update({
1446 'width': the_only_video.get('width'),
1447 'height': the_only_video.get('height'),
1448 'resolution': the_only_video.get('resolution'),
1449 'fps': the_only_video.get('fps'),
1450 'vcodec': the_only_video.get('vcodec'),
1451 'vbr': the_only_video.get('vbr'),
1452 'stretched_ratio': the_only_video.get('stretched_ratio'),
1453 })
1454
1455 if the_only_audio:
1456 new_dict.update({
1457 'acodec': the_only_audio.get('acodec'),
1458 'abr': the_only_audio.get('abr'),
1459 })
1460
1461 return new_dict
1462
1463 selector_1, selector_2 = map(_build_selector_function, selector.selector)
1464
1465 def selector_function(ctx):
1466 for pair in itertools.product(
1467 selector_1(copy.deepcopy(ctx)), selector_2(copy.deepcopy(ctx))):
1468 yield _merge(pair)
1469
1470 filters = [self._build_format_filter(f) for f in selector.filters]
1471
1472 def final_selector(ctx):
1473 ctx_copy = copy.deepcopy(ctx)
1474 for _filter in filters:
1475 ctx_copy['formats'] = list(filter(_filter, ctx_copy['formats']))
1476 return selector_function(ctx_copy)
1477 return final_selector
1478
1479 stream = io.BytesIO(format_spec.encode('utf-8'))
1480 try:
1481 tokens = list(_remove_unused_ops(compat_tokenize_tokenize(stream.readline)))
1482 except tokenize.TokenError:
1483 raise syntax_error('Missing closing/opening brackets or parenthesis', (0, len(format_spec)))
1484
1485 class TokenIterator(object):
1486 def __init__(self, tokens):
1487 self.tokens = tokens
1488 self.counter = 0
1489
1490 def __iter__(self):
1491 return self
1492
1493 def __next__(self):
1494 if self.counter >= len(self.tokens):
1495 raise StopIteration()
1496 value = self.tokens[self.counter]
1497 self.counter += 1
1498 return value
1499
1500 next = __next__
1501
1502 def restore_last_token(self):
1503 self.counter -= 1
1504
1505 parsed_selector = _parse_format_selection(iter(TokenIterator(tokens)))
1506 return _build_selector_function(parsed_selector)
1507
1508 def _calc_headers(self, info_dict):
1509 res = std_headers.copy()
1510
1511 add_headers = info_dict.get('http_headers')
1512 if add_headers:
1513 res.update(add_headers)
1514
1515 cookies = self._calc_cookies(info_dict)
1516 if cookies:
1517 res['Cookie'] = cookies
1518
1519 if 'X-Forwarded-For' not in res:
1520 x_forwarded_for_ip = info_dict.get('__x_forwarded_for_ip')
1521 if x_forwarded_for_ip:
1522 res['X-Forwarded-For'] = x_forwarded_for_ip
1523
1524 return res
1525
1526 def _calc_cookies(self, info_dict):
1527 pr = sanitized_Request(info_dict['url'])
1528 self.cookiejar.add_cookie_header(pr)
1529 return pr.get_header('Cookie')
1530
1531 def process_video_result(self, info_dict, download=True):
1532 assert info_dict.get('_type', 'video') == 'video'
1533
1534 if 'id' not in info_dict:
1535 raise ExtractorError('Missing "id" field in extractor result')
1536 if 'title' not in info_dict:
1537 raise ExtractorError('Missing "title" field in extractor result')
1538
1539 def report_force_conversion(field, field_not, conversion):
1540 self.report_warning(
1541 '"%s" field is not %s - forcing %s conversion, there is an error in extractor'
1542 % (field, field_not, conversion))
1543
1544 def sanitize_string_field(info, string_field):
1545 field = info.get(string_field)
1546 if field is None or isinstance(field, compat_str):
1547 return
1548 report_force_conversion(string_field, 'a string', 'string')
1549 info[string_field] = compat_str(field)
1550
1551 def sanitize_numeric_fields(info):
1552 for numeric_field in self._NUMERIC_FIELDS:
1553 field = info.get(numeric_field)
1554 if field is None or isinstance(field, compat_numeric_types):
1555 continue
1556 report_force_conversion(numeric_field, 'numeric', 'int')
1557 info[numeric_field] = int_or_none(field)
1558
1559 sanitize_string_field(info_dict, 'id')
1560 sanitize_numeric_fields(info_dict)
1561
1562 if 'playlist' not in info_dict:
1563 # It isn't part of a playlist
1564 info_dict['playlist'] = None
1565 info_dict['playlist_index'] = None
1566
1567 thumbnails = info_dict.get('thumbnails')
1568 if thumbnails is None:
1569 thumbnail = info_dict.get('thumbnail')
1570 if thumbnail:
1571 info_dict['thumbnails'] = thumbnails = [{'url': thumbnail}]
1572 if thumbnails:
1573 thumbnails.sort(key=lambda t: (
1574 t.get('preference') if t.get('preference') is not None else -1,
1575 t.get('width') if t.get('width') is not None else -1,
1576 t.get('height') if t.get('height') is not None else -1,
1577 t.get('id') if t.get('id') is not None else '', t.get('url')))
1578 for i, t in enumerate(thumbnails):
1579 t['url'] = sanitize_url(t['url'])
1580 if t.get('width') and t.get('height'):
1581 t['resolution'] = '%dx%d' % (t['width'], t['height'])
1582 if t.get('id') is None:
1583 t['id'] = '%d' % i
1584
1585 if self.params.get('list_thumbnails'):
1586 self.list_thumbnails(info_dict)
1587 return
1588
1589 thumbnail = info_dict.get('thumbnail')
1590 if thumbnail:
1591 info_dict['thumbnail'] = sanitize_url(thumbnail)
1592 elif thumbnails:
1593 info_dict['thumbnail'] = thumbnails[-1]['url']
1594
1595 if 'display_id' not in info_dict and 'id' in info_dict:
1596 info_dict['display_id'] = info_dict['id']
1597
1598 if info_dict.get('upload_date') is None and info_dict.get('timestamp') is not None:
1599 # Working around out-of-range timestamp values (e.g. negative ones on Windows,
1600 # see http://bugs.python.org/issue1646728)
1601 try:
1602 upload_date = datetime.datetime.utcfromtimestamp(info_dict['timestamp'])
1603 info_dict['upload_date'] = upload_date.strftime('%Y%m%d')
1604 except (ValueError, OverflowError, OSError):
1605 pass
1606
1607 # Auto generate title fields corresponding to the *_number fields when missing
1608 # in order to always have clean titles. This is very common for TV series.
1609 for field in ('chapter', 'season', 'episode'):
1610 if info_dict.get('%s_number' % field) is not None and not info_dict.get(field):
1611 info_dict[field] = '%s %d' % (field.capitalize(), info_dict['%s_number' % field])
1612
1613 for cc_kind in ('subtitles', 'automatic_captions'):
1614 cc = info_dict.get(cc_kind)
1615 if cc:
1616 for _, subtitle in cc.items():
1617 for subtitle_format in subtitle:
1618 if subtitle_format.get('url'):
1619 subtitle_format['url'] = sanitize_url(subtitle_format['url'])
1620 if subtitle_format.get('ext') is None:
1621 subtitle_format['ext'] = determine_ext(subtitle_format['url']).lower()
1622
1623 automatic_captions = info_dict.get('automatic_captions')
1624 subtitles = info_dict.get('subtitles')
1625
1626 if self.params.get('listsubtitles', False):
1627 if 'automatic_captions' in info_dict:
1628 self.list_subtitles(
1629 info_dict['id'], automatic_captions, 'automatic captions')
1630 self.list_subtitles(info_dict['id'], subtitles, 'subtitles')
1631 return
1632
1633 info_dict['requested_subtitles'] = self.process_subtitles(
1634 info_dict['id'], subtitles, automatic_captions)
1635
1636 # We now pick which formats have to be downloaded
1637 if info_dict.get('formats') is None:
1638 # There's only one format available
1639 formats = [info_dict]
1640 else:
1641 formats = info_dict['formats']
1642
1643 if not formats:
1644 raise ExtractorError('No video formats found!')
1645
1646 def is_wellformed(f):
1647 url = f.get('url')
1648 if not url:
1649 self.report_warning(
1650 '"url" field is missing or empty - skipping format, '
1651 'there is an error in extractor')
1652 return False
1653 if isinstance(url, bytes):
1654 sanitize_string_field(f, 'url')
1655 return True
1656
1657 # Filter out malformed formats for better extraction robustness
1658 formats = list(filter(is_wellformed, formats))
1659
1660 formats_dict = {}
1661
1662 # We check that all the formats have the format and format_id fields
1663 for i, format in enumerate(formats):
1664 sanitize_string_field(format, 'format_id')
1665 sanitize_numeric_fields(format)
1666 format['url'] = sanitize_url(format['url'])
1667 if not format.get('format_id'):
1668 format['format_id'] = compat_str(i)
1669 else:
1670 # Sanitize format_id from characters used in format selector expression
1671 format['format_id'] = re.sub(r'[\s,/+\[\]()]', '_', format['format_id'])
1672 format_id = format['format_id']
1673 if format_id not in formats_dict:
1674 formats_dict[format_id] = []
1675 formats_dict[format_id].append(format)
1676
1677 # Make sure all formats have unique format_id
1678 for format_id, ambiguous_formats in formats_dict.items():
1679 if len(ambiguous_formats) > 1:
1680 for i, format in enumerate(ambiguous_formats):
1681 format['format_id'] = '%s-%d' % (format_id, i)
1682
1683 for i, format in enumerate(formats):
1684 if format.get('format') is None:
1685 format['format'] = '{id} - {res}{note}'.format(
1686 id=format['format_id'],
1687 res=self.format_resolution(format),
1688 note=' ({0})'.format(format['format_note']) if format.get('format_note') is not None else '',
1689 )
1690 # Automatically determine file extension if missing
1691 if format.get('ext') is None:
1692 format['ext'] = determine_ext(format['url']).lower()
1693 # Automatically determine protocol if missing (useful for format
1694 # selection purposes)
1695 if format.get('protocol') is None:
1696 format['protocol'] = determine_protocol(format)
1697 # Add HTTP headers, so that external programs can use them from the
1698 # json output
1699 full_format_info = info_dict.copy()
1700 full_format_info.update(format)
1701 format['http_headers'] = self._calc_headers(full_format_info)
1702 # Remove private housekeeping stuff
1703 if '__x_forwarded_for_ip' in info_dict:
1704 del info_dict['__x_forwarded_for_ip']
1705
1706 # TODO Central sorting goes here
1707
1708 if formats[0] is not info_dict:
1709 # only set the 'formats' fields if the original info_dict list them
1710 # otherwise we end up with a circular reference, the first (and unique)
1711 # element in the 'formats' field in info_dict is info_dict itself,
1712 # which can't be exported to json
1713 info_dict['formats'] = formats
1714 if self.params.get('listformats'):
1715 self.list_formats(info_dict)
1716 return
1717
1718 req_format = self.params.get('format')
1719 if req_format is None:
1720 req_format = self._default_format_spec(info_dict, download=download)
1721 if self.params.get('verbose'):
1722 self._write_string('[debug] Default format spec: %s\n' % req_format)
1723
1724 format_selector = self.build_format_selector(req_format)
1725
1726 # While in format selection we may need to have an access to the original
1727 # format set in order to calculate some metrics or do some processing.
1728 # For now we need to be able to guess whether original formats provided
1729 # by extractor are incomplete or not (i.e. whether extractor provides only
1730 # video-only or audio-only formats) for proper formats selection for
1731 # extractors with such incomplete formats (see
1732 # https://github.com/ytdl-org/youtube-dl/pull/5556).
1733 # Since formats may be filtered during format selection and may not match
1734 # the original formats the results may be incorrect. Thus original formats
1735 # or pre-calculated metrics should be passed to format selection routines
1736 # as well.
1737 # We will pass a context object containing all necessary additional data
1738 # instead of just formats.
1739 # This fixes incorrect format selection issue (see
1740 # https://github.com/ytdl-org/youtube-dl/issues/10083).
1741 incomplete_formats = (
1742 # All formats are video-only or
1743 all(f.get('vcodec') != 'none' and f.get('acodec') == 'none' for f in formats)
1744 # all formats are audio-only
1745 or all(f.get('vcodec') == 'none' and f.get('acodec') != 'none' for f in formats))
1746
1747 ctx = {
1748 'formats': formats,
1749 'incomplete_formats': incomplete_formats,
1750 }
1751
1752 formats_to_download = list(format_selector(ctx))
1753 if not formats_to_download:
1754 raise ExtractorError('requested format not available',
1755 expected=True)
1756
1757 if download:
1758 self.to_screen('[info] Downloading format(s) %s' % ", ".join([f['format_id'] for f in formats_to_download]))
1759 if len(formats_to_download) > 1:
1760 self.to_screen('[info] %s: downloading video in %s formats' % (info_dict['id'], len(formats_to_download)))
1761 for format in formats_to_download:
1762 new_info = dict(info_dict)
1763 new_info.update(format)
1764 self.process_info(new_info)
1765 # We update the info dict with the best quality format (backwards compatibility)
1766 info_dict.update(formats_to_download[-1])
1767 return info_dict
1768
1769 def process_subtitles(self, video_id, normal_subtitles, automatic_captions):
1770 """Select the requested subtitles and their format"""
1771 available_subs = {}
1772 if normal_subtitles and self.params.get('writesubtitles'):
1773 available_subs.update(normal_subtitles)
1774 if automatic_captions and self.params.get('writeautomaticsub'):
1775 for lang, cap_info in automatic_captions.items():
1776 if lang not in available_subs:
1777 available_subs[lang] = cap_info
1778
1779 if (not self.params.get('writesubtitles') and not
1780 self.params.get('writeautomaticsub') or not
1781 available_subs):
1782 return None
1783
1784 if self.params.get('allsubtitles', False):
1785 requested_langs = available_subs.keys()
1786 else:
1787 if self.params.get('subtitleslangs', False):
1788 requested_langs = self.params.get('subtitleslangs')
1789 elif 'en' in available_subs:
1790 requested_langs = ['en']
1791 else:
1792 requested_langs = [list(available_subs.keys())[0]]
1793
1794 formats_query = self.params.get('subtitlesformat', 'best')
1795 formats_preference = formats_query.split('/') if formats_query else []
1796 subs = {}
1797 for lang in requested_langs:
1798 formats = available_subs.get(lang)
1799 if formats is None:
1800 self.report_warning('%s subtitles not available for %s' % (lang, video_id))
1801 continue
1802 for ext in formats_preference:
1803 if ext == 'best':
1804 f = formats[-1]
1805 break
1806 matches = list(filter(lambda f: f['ext'] == ext, formats))
1807 if matches:
1808 f = matches[-1]
1809 break
1810 else:
1811 f = formats[-1]
1812 self.report_warning(
1813 'No subtitle format found matching "%s" for language %s, '
1814 'using %s' % (formats_query, lang, f['ext']))
1815 subs[lang] = f
1816 return subs
1817
1818 def __forced_printings(self, info_dict, filename, incomplete):
1819 def print_mandatory(field):
1820 if (self.params.get('force%s' % field, False)
1821 and (not incomplete or info_dict.get(field) is not None)):
1822 self.to_stdout(info_dict[field])
1823
1824 def print_optional(field):
1825 if (self.params.get('force%s' % field, False)
1826 and info_dict.get(field) is not None):
1827 self.to_stdout(info_dict[field])
1828
1829 print_mandatory('title')
1830 print_mandatory('id')
1831 if self.params.get('forceurl', False) and not incomplete:
1832 if info_dict.get('requested_formats') is not None:
1833 for f in info_dict['requested_formats']:
1834 self.to_stdout(f['url'] + f.get('play_path', ''))
1835 else:
1836 # For RTMP URLs, also include the playpath
1837 self.to_stdout(info_dict['url'] + info_dict.get('play_path', ''))
1838 print_optional('thumbnail')
1839 print_optional('description')
1840 if self.params.get('forcefilename', False) and filename is not None:
1841 self.to_stdout(filename)
1842 if self.params.get('forceduration', False) and info_dict.get('duration') is not None:
1843 self.to_stdout(formatSeconds(info_dict['duration']))
1844 print_mandatory('format')
1845 if self.params.get('forcejson', False):
1846 self.to_stdout(json.dumps(info_dict))
1847
1848 def process_info(self, info_dict):
1849 """Process a single resolved IE result."""
1850
1851 assert info_dict.get('_type', 'video') == 'video'
1852
1853 max_downloads = self.params.get('max_downloads')
1854 if max_downloads is not None:
1855 if self._num_downloads >= int(max_downloads):
1856 raise MaxDownloadsReached()
1857
1858 # TODO: backward compatibility, to be removed
1859 info_dict['fulltitle'] = info_dict['title']
1860
1861 if 'format' not in info_dict:
1862 info_dict['format'] = info_dict['ext']
1863
1864 reason = self._match_entry(info_dict, incomplete=False)
1865 if reason is not None:
1866 self.to_screen('[download] ' + reason)
1867 return
1868
1869 self._num_downloads += 1
1870
1871 info_dict['_filename'] = filename = self.prepare_filename(info_dict)
1872
1873 # Forced printings
1874 self.__forced_printings(info_dict, filename, incomplete=False)
1875
1876 if self.params.get('simulate', False):
1877 if self.params.get('force_write_download_archive', False):
1878 self.record_download_archive(info_dict)
1879
1880 # Do nothing else if in simulate mode
1881 return
1882
1883 if filename is None:
1884 return
1885
1886 def ensure_dir_exists(path):
1887 try:
1888 dn = os.path.dirname(path)
1889 if dn and not os.path.exists(dn):
1890 os.makedirs(dn)
1891 return True
1892 except (OSError, IOError) as err:
1893 self.report_error('unable to create directory ' + error_to_compat_str(err))
1894 return False
1895
1896 if not ensure_dir_exists(sanitize_path(encodeFilename(filename))):
1897 return
1898
1899 if self.params.get('writedescription', False):
1900 descfn = replace_extension(filename, 'description', info_dict.get('ext'))
1901 if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(descfn)):
1902 self.to_screen('[info] Video description is already present')
1903 elif info_dict.get('description') is None:
1904 self.report_warning('There\'s no description to write.')
1905 else:
1906 try:
1907 self.to_screen('[info] Writing video description to: ' + descfn)
1908 with io.open(encodeFilename(descfn), 'w', encoding='utf-8') as descfile:
1909 descfile.write(info_dict['description'])
1910 except (OSError, IOError):
1911 self.report_error('Cannot write description file ' + descfn)
1912 return
1913
1914 if self.params.get('writeannotations', False):
1915 annofn = replace_extension(filename, 'annotations.xml', info_dict.get('ext'))
1916 if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(annofn)):
1917 self.to_screen('[info] Video annotations are already present')
1918 elif not info_dict.get('annotations'):
1919 self.report_warning('There are no annotations to write.')
1920 else:
1921 try:
1922 self.to_screen('[info] Writing video annotations to: ' + annofn)
1923 with io.open(encodeFilename(annofn), 'w', encoding='utf-8') as annofile:
1924 annofile.write(info_dict['annotations'])
1925 except (KeyError, TypeError):
1926 self.report_warning('There are no annotations to write.')
1927 except (OSError, IOError):
1928 self.report_error('Cannot write annotations file: ' + annofn)
1929 return
1930
1931 def dl(name, info, subtitle=False):
1932 fd = get_suitable_downloader(info, self.params)(self, self.params)
1933 for ph in self._progress_hooks:
1934 fd.add_progress_hook(ph)
1935 if self.params.get('verbose'):
1936 self.to_screen('[debug] Invoking downloader on %r' % info.get('url'))
1937 return fd.download(name, info, subtitle)
1938
1939 subtitles_are_requested = any([self.params.get('writesubtitles', False),
1940 self.params.get('writeautomaticsub')])
1941
1942 if subtitles_are_requested and info_dict.get('requested_subtitles'):
1943 # subtitles download errors are already managed as troubles in relevant IE
1944 # that way it will silently go on when used with unsupporting IE
1945 subtitles = info_dict['requested_subtitles']
1946 # ie = self.get_info_extractor(info_dict['extractor_key'])
1947 for sub_lang, sub_info in subtitles.items():
1948 sub_format = sub_info['ext']
1949 sub_filename = subtitles_filename(filename, sub_lang, sub_format, info_dict.get('ext'))
1950 if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(sub_filename)):
1951 self.to_screen('[info] Video subtitle %s.%s is already present' % (sub_lang, sub_format))
1952 else:
1953 self.to_screen('[info] Writing video subtitles to: ' + sub_filename)
1954 if sub_info.get('data') is not None:
1955 try:
1956 # Use newline='' to prevent conversion of newline characters
1957 # See https://github.com/ytdl-org/youtube-dl/issues/10268
1958 with io.open(encodeFilename(sub_filename), 'w', encoding='utf-8', newline='') as subfile:
1959 subfile.write(sub_info['data'])
1960 except (OSError, IOError):
1961 self.report_error('Cannot write subtitles file ' + sub_filename)
1962 return
1963 else:
1964 try:
1965 dl(sub_filename, sub_info, subtitle=True)
1966 '''
1967 if self.params.get('sleep_interval_subtitles', False):
1968 dl(sub_filename, sub_info)
1969 else:
1970 sub_data = ie._request_webpage(
1971 sub_info['url'], info_dict['id'], note=False).read()
1972 with io.open(encodeFilename(sub_filename), 'wb') as subfile:
1973 subfile.write(sub_data)
1974 '''
1975 except (ExtractorError, IOError, OSError, ValueError, compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
1976 self.report_warning('Unable to download subtitle for "%s": %s' %
1977 (sub_lang, error_to_compat_str(err)))
1978 continue
1979
1980 if self.params.get('skip_download', False):
1981 if self.params.get('convertsubtitles', False):
1982 subconv = FFmpegSubtitlesConvertorPP(self, format=self.params.get('convertsubtitles'))
1983 filename_real_ext = os.path.splitext(filename)[1][1:]
1984 filename_wo_ext = (
1985 os.path.splitext(filename)[0]
1986 if filename_real_ext == info_dict['ext']
1987 else filename)
1988 afilename = '%s.%s' % (filename_wo_ext, self.params.get('convertsubtitles'))
1989 if subconv.available:
1990 info_dict.setdefault('__postprocessors', [])
1991 # info_dict['__postprocessors'].append(subconv)
1992 if os.path.exists(encodeFilename(afilename)):
1993 self.to_screen(
1994 '[download] %s has already been downloaded and '
1995 'converted' % afilename)
1996 else:
1997 try:
1998 self.post_process(filename, info_dict)
1999 except (PostProcessingError) as err:
2000 self.report_error('postprocessing: %s' % str(err))
2001 return
2002
2003 if self.params.get('writeinfojson', False):
2004 infofn = replace_extension(filename, 'info.json', info_dict.get('ext'))
2005 if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(infofn)):
2006 self.to_screen('[info] Video description metadata is already present')
2007 else:
2008 self.to_screen('[info] Writing video description metadata as JSON to: ' + infofn)
2009 try:
2010 write_json_file(self.filter_requested_info(info_dict), infofn)
2011 except (OSError, IOError):
2012 self.report_error('Cannot write metadata to JSON file ' + infofn)
2013 return
2014
2015 self._write_thumbnails(info_dict, filename)
2016
2017 # Write internet shortcut files
2018 url_link = webloc_link = desktop_link = False
2019 if self.params.get('writelink', False):
2020 if sys.platform == "darwin": # macOS.
2021 webloc_link = True
2022 elif sys.platform.startswith("linux"):
2023 desktop_link = True
2024 else: # if sys.platform in ['win32', 'cygwin']:
2025 url_link = True
2026 if self.params.get('writeurllink', False):
2027 url_link = True
2028 if self.params.get('writewebloclink', False):
2029 webloc_link = True
2030 if self.params.get('writedesktoplink', False):
2031 desktop_link = True
2032
2033 if url_link or webloc_link or desktop_link:
2034 if 'webpage_url' not in info_dict:
2035 self.report_error('Cannot write internet shortcut file because the "webpage_url" field is missing in the media information')
2036 return
2037 ascii_url = iri_to_uri(info_dict['webpage_url'])
2038
2039 def _write_link_file(extension, template, newline, embed_filename):
2040 linkfn = replace_extension(filename, extension, info_dict.get('ext'))
2041 if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(linkfn)):
2042 self.to_screen('[info] Internet shortcut is already present')
2043 else:
2044 try:
2045 self.to_screen('[info] Writing internet shortcut to: ' + linkfn)
2046 with io.open(encodeFilename(to_high_limit_path(linkfn)), 'w', encoding='utf-8', newline=newline) as linkfile:
2047 template_vars = {'url': ascii_url}
2048 if embed_filename:
2049 template_vars['filename'] = linkfn[:-(len(extension) + 1)]
2050 linkfile.write(template % template_vars)
2051 except (OSError, IOError):
2052 self.report_error('Cannot write internet shortcut ' + linkfn)
2053 return False
2054 return True
2055
2056 if url_link:
2057 if not _write_link_file('url', DOT_URL_LINK_TEMPLATE, '\r\n', embed_filename=False):
2058 return
2059 if webloc_link:
2060 if not _write_link_file('webloc', DOT_WEBLOC_LINK_TEMPLATE, '\n', embed_filename=False):
2061 return
2062 if desktop_link:
2063 if not _write_link_file('desktop', DOT_DESKTOP_LINK_TEMPLATE, '\n', embed_filename=True):
2064 return
2065
2066 # Download
2067 must_record_download_archive = False
2068 if not self.params.get('skip_download', False):
2069 try:
2070 if info_dict.get('requested_formats') is not None:
2071 downloaded = []
2072 success = True
2073 merger = FFmpegMergerPP(self)
2074 if not merger.available:
2075 postprocessors = []
2076 self.report_warning('You have requested multiple '
2077 'formats but ffmpeg or avconv are not installed.'
2078 ' The formats won\'t be merged.')
2079 else:
2080 postprocessors = [merger]
2081
2082 def compatible_formats(formats):
2083 # TODO: some formats actually allow this (mkv, webm, ogg, mp4), but not all of them.
2084 video_formats = [format for format in formats if format.get('vcodec') != 'none']
2085 audio_formats = [format for format in formats if format.get('acodec') != 'none']
2086 if len(video_formats) > 2 or len(audio_formats) > 2:
2087 return False
2088
2089 # Check extension
2090 exts = set(format.get('ext') for format in formats)
2091 COMPATIBLE_EXTS = (
2092 set(('mp3', 'mp4', 'm4a', 'm4p', 'm4b', 'm4r', 'm4v', 'ismv', 'isma')),
2093 set(('webm',)),
2094 )
2095 for ext_sets in COMPATIBLE_EXTS:
2096 if ext_sets.issuperset(exts):
2097 return True
2098 # TODO: Check acodec/vcodec
2099 return False
2100
2101 filename_real_ext = os.path.splitext(filename)[1][1:]
2102 filename_wo_ext = (
2103 os.path.splitext(filename)[0]
2104 if filename_real_ext == info_dict['ext']
2105 else filename)
2106 requested_formats = info_dict['requested_formats']
2107 if self.params.get('merge_output_format') is None and not compatible_formats(requested_formats):
2108 info_dict['ext'] = 'mkv'
2109 self.report_warning(
2110 'Requested formats are incompatible for merge and will be merged into mkv.')
2111 # Ensure filename always has a correct extension for successful merge
2112 filename = '%s.%s' % (filename_wo_ext, info_dict['ext'])
2113 if os.path.exists(encodeFilename(filename)):
2114 self.to_screen(
2115 '[download] %s has already been downloaded and '
2116 'merged' % filename)
2117 else:
2118 for f in requested_formats:
2119 new_info = dict(info_dict)
2120 new_info.update(f)
2121 fname = prepend_extension(
2122 self.prepare_filename(new_info),
2123 'f%s' % f['format_id'], new_info['ext'])
2124 if not ensure_dir_exists(fname):
2125 return
2126 downloaded.append(fname)
2127 partial_success, real_download = dl(fname, new_info)
2128 success = success and partial_success
2129 info_dict['__postprocessors'] = postprocessors
2130 info_dict['__files_to_merge'] = downloaded
2131 # Even if there were no downloads, it is being merged only now
2132 info_dict['__real_download'] = True
2133 else:
2134 # Just a single file
2135 success, real_download = dl(filename, info_dict)
2136 info_dict['__real_download'] = real_download
2137 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
2138 self.report_error('unable to download video data: %s' % error_to_compat_str(err))
2139 return
2140 except (OSError, IOError) as err:
2141 raise UnavailableVideoError(err)
2142 except (ContentTooShortError, ) as err:
2143 self.report_error('content too short (expected %s bytes and served %s)' % (err.expected, err.downloaded))
2144 return
2145
2146 if success and filename != '-':
2147 # Fixup content
2148 fixup_policy = self.params.get('fixup')
2149 if fixup_policy is None:
2150 fixup_policy = 'detect_or_warn'
2151
2152 INSTALL_FFMPEG_MESSAGE = 'Install ffmpeg or avconv to fix this automatically.'
2153
2154 stretched_ratio = info_dict.get('stretched_ratio')
2155 if stretched_ratio is not None and stretched_ratio != 1:
2156 if fixup_policy == 'warn':
2157 self.report_warning('%s: Non-uniform pixel ratio (%s)' % (
2158 info_dict['id'], stretched_ratio))
2159 elif fixup_policy == 'detect_or_warn':
2160 stretched_pp = FFmpegFixupStretchedPP(self)
2161 if stretched_pp.available:
2162 info_dict.setdefault('__postprocessors', [])
2163 info_dict['__postprocessors'].append(stretched_pp)
2164 else:
2165 self.report_warning(
2166 '%s: Non-uniform pixel ratio (%s). %s'
2167 % (info_dict['id'], stretched_ratio, INSTALL_FFMPEG_MESSAGE))
2168 else:
2169 assert fixup_policy in ('ignore', 'never')
2170
2171 if (info_dict.get('requested_formats') is None
2172 and info_dict.get('container') == 'm4a_dash'):
2173 if fixup_policy == 'warn':
2174 self.report_warning(
2175 '%s: writing DASH m4a. '
2176 'Only some players support this container.'
2177 % info_dict['id'])
2178 elif fixup_policy == 'detect_or_warn':
2179 fixup_pp = FFmpegFixupM4aPP(self)
2180 if fixup_pp.available:
2181 info_dict.setdefault('__postprocessors', [])
2182 info_dict['__postprocessors'].append(fixup_pp)
2183 else:
2184 self.report_warning(
2185 '%s: writing DASH m4a. '
2186 'Only some players support this container. %s'
2187 % (info_dict['id'], INSTALL_FFMPEG_MESSAGE))
2188 else:
2189 assert fixup_policy in ('ignore', 'never')
2190
2191 if (info_dict.get('protocol') == 'm3u8_native'
2192 or info_dict.get('protocol') == 'm3u8'
2193 and self.params.get('hls_prefer_native')):
2194 if fixup_policy == 'warn':
2195 self.report_warning('%s: malformed AAC bitstream detected.' % (
2196 info_dict['id']))
2197 elif fixup_policy == 'detect_or_warn':
2198 fixup_pp = FFmpegFixupM3u8PP(self)
2199 if fixup_pp.available:
2200 info_dict.setdefault('__postprocessors', [])
2201 info_dict['__postprocessors'].append(fixup_pp)
2202 else:
2203 self.report_warning(
2204 '%s: malformed AAC bitstream detected. %s'
2205 % (info_dict['id'], INSTALL_FFMPEG_MESSAGE))
2206 else:
2207 assert fixup_policy in ('ignore', 'never')
2208
2209 try:
2210 self.post_process(filename, info_dict)
2211 except (PostProcessingError) as err:
2212 self.report_error('postprocessing: %s' % str(err))
2213 return
2214 try:
2215 for ph in self._post_hooks:
2216 ph(filename)
2217 except Exception as err:
2218 self.report_error('post hooks: %s' % str(err))
2219 return
2220 must_record_download_archive = True
2221
2222 if must_record_download_archive or self.params.get('force_write_download_archive', False):
2223 self.record_download_archive(info_dict)
2224 max_downloads = self.params.get('max_downloads')
2225 if max_downloads is not None and self._num_downloads >= int(max_downloads):
2226 raise MaxDownloadsReached()
2227
2228 def download(self, url_list):
2229 """Download a given list of URLs."""
2230 outtmpl = self.params.get('outtmpl', DEFAULT_OUTTMPL)
2231 if (len(url_list) > 1
2232 and outtmpl != '-'
2233 and '%' not in outtmpl
2234 and self.params.get('max_downloads') != 1):
2235 raise SameFileError(outtmpl)
2236
2237 for url in url_list:
2238 try:
2239 # It also downloads the videos
2240 res = self.extract_info(
2241 url, force_generic_extractor=self.params.get('force_generic_extractor', False))
2242 except UnavailableVideoError:
2243 self.report_error('unable to download video')
2244 except MaxDownloadsReached:
2245 self.to_screen('[info] Maximum number of downloaded files reached.')
2246 raise
2247 else:
2248 if self.params.get('dump_single_json', False):
2249 self.to_stdout(json.dumps(res))
2250
2251 return self._download_retcode
2252
2253 def download_with_info_file(self, info_filename):
2254 with contextlib.closing(fileinput.FileInput(
2255 [info_filename], mode='r',
2256 openhook=fileinput.hook_encoded('utf-8'))) as f:
2257 # FileInput doesn't have a read method, we can't call json.load
2258 info = self.filter_requested_info(json.loads('\n'.join(f)))
2259 try:
2260 self.process_ie_result(info, download=True)
2261 except DownloadError:
2262 webpage_url = info.get('webpage_url')
2263 if webpage_url is not None:
2264 self.report_warning('The info failed to download, trying with "%s"' % webpage_url)
2265 return self.download([webpage_url])
2266 else:
2267 raise
2268 return self._download_retcode
2269
2270 @staticmethod
2271 def filter_requested_info(info_dict):
2272 return dict(
2273 (k, v) for k, v in info_dict.items()
2274 if k not in ['requested_formats', 'requested_subtitles'])
2275
2276 def post_process(self, filename, ie_info):
2277 """Run all the postprocessors on the given file."""
2278 info = dict(ie_info)
2279 info['filepath'] = filename
2280 pps_chain = []
2281 if ie_info.get('__postprocessors') is not None:
2282 pps_chain.extend(ie_info['__postprocessors'])
2283 pps_chain.extend(self._pps)
2284 for pp in pps_chain:
2285 files_to_delete = []
2286 try:
2287 files_to_delete, info = pp.run(info)
2288 except PostProcessingError as e:
2289 self.report_error(e.msg)
2290 if files_to_delete and not self.params.get('keepvideo', False):
2291 for old_filename in set(files_to_delete):
2292 self.to_screen('Deleting original file %s (pass -k to keep)' % old_filename)
2293 try:
2294 os.remove(encodeFilename(old_filename))
2295 except (IOError, OSError):
2296 self.report_warning('Unable to remove downloaded original file')
2297
2298 def _make_archive_id(self, info_dict):
2299 video_id = info_dict.get('id')
2300 if not video_id:
2301 return
2302 # Future-proof against any change in case
2303 # and backwards compatibility with prior versions
2304 extractor = info_dict.get('extractor_key') or info_dict.get('ie_key') # key in a playlist
2305 if extractor is None:
2306 url = str_or_none(info_dict.get('url'))
2307 if not url:
2308 return
2309 # Try to find matching extractor for the URL and take its ie_key
2310 for ie in self._ies:
2311 if ie.suitable(url):
2312 extractor = ie.ie_key()
2313 break
2314 else:
2315 return
2316 return extractor.lower() + ' ' + video_id
2317
2318 def in_download_archive(self, info_dict):
2319 fn = self.params.get('download_archive')
2320 if fn is None:
2321 return False
2322
2323 vid_id = self._make_archive_id(info_dict)
2324 if not vid_id:
2325 return False # Incomplete video information
2326
2327 return vid_id in self.archive
2328
2329 def record_download_archive(self, info_dict):
2330 fn = self.params.get('download_archive')
2331 if fn is None:
2332 return
2333 vid_id = self._make_archive_id(info_dict)
2334 assert vid_id
2335 with locked_file(fn, 'a', encoding='utf-8') as archive_file:
2336 archive_file.write(vid_id + '\n')
2337 self.archive.add(vid_id)
2338
2339 @staticmethod
2340 def format_resolution(format, default='unknown'):
2341 if format.get('vcodec') == 'none':
2342 return 'audio only'
2343 if format.get('resolution') is not None:
2344 return format['resolution']
2345 if format.get('height') is not None:
2346 if format.get('width') is not None:
2347 res = '%sx%s' % (format['width'], format['height'])
2348 else:
2349 res = '%sp' % format['height']
2350 elif format.get('width') is not None:
2351 res = '%dx?' % format['width']
2352 else:
2353 res = default
2354 return res
2355
2356 def _format_note(self, fdict):
2357 res = ''
2358 if fdict.get('ext') in ['f4f', 'f4m']:
2359 res += '(unsupported) '
2360 if fdict.get('language'):
2361 if res:
2362 res += ' '
2363 res += '[%s] ' % fdict['language']
2364 if fdict.get('format_note') is not None:
2365 res += fdict['format_note'] + ' '
2366 if fdict.get('tbr') is not None:
2367 res += '%4dk ' % fdict['tbr']
2368 if fdict.get('container') is not None:
2369 if res:
2370 res += ', '
2371 res += '%s container' % fdict['container']
2372 if (fdict.get('vcodec') is not None
2373 and fdict.get('vcodec') != 'none'):
2374 if res:
2375 res += ', '
2376 res += fdict['vcodec']
2377 if fdict.get('vbr') is not None:
2378 res += '@'
2379 elif fdict.get('vbr') is not None and fdict.get('abr') is not None:
2380 res += 'video@'
2381 if fdict.get('vbr') is not None:
2382 res += '%4dk' % fdict['vbr']
2383 if fdict.get('fps') is not None:
2384 if res:
2385 res += ', '
2386 res += '%sfps' % fdict['fps']
2387 if fdict.get('acodec') is not None:
2388 if res:
2389 res += ', '
2390 if fdict['acodec'] == 'none':
2391 res += 'video only'
2392 else:
2393 res += '%-5s' % fdict['acodec']
2394 elif fdict.get('abr') is not None:
2395 if res:
2396 res += ', '
2397 res += 'audio'
2398 if fdict.get('abr') is not None:
2399 res += '@%3dk' % fdict['abr']
2400 if fdict.get('asr') is not None:
2401 res += ' (%5dHz)' % fdict['asr']
2402 if fdict.get('filesize') is not None:
2403 if res:
2404 res += ', '
2405 res += format_bytes(fdict['filesize'])
2406 elif fdict.get('filesize_approx') is not None:
2407 if res:
2408 res += ', '
2409 res += '~' + format_bytes(fdict['filesize_approx'])
2410 return res
2411
2412 def _format_note_table(self, f):
2413 def join_fields(*vargs):
2414 return ', '.join((val for val in vargs if val != ''))
2415
2416 return join_fields(
2417 'UNSUPPORTED' if f.get('ext') in ('f4f', 'f4m') else '',
2418 format_field(f, 'language', '[%s]'),
2419 format_field(f, 'format_note'),
2420 format_field(f, 'container', ignore=(None, f.get('ext'))),
2421 format_field(f, 'asr', '%5dHz'))
2422
2423 def list_formats(self, info_dict):
2424 formats = info_dict.get('formats', [info_dict])
2425 new_format = self.params.get('listformats_table', False)
2426 if new_format:
2427 table = [
2428 [
2429 format_field(f, 'format_id'),
2430 format_field(f, 'ext'),
2431 self.format_resolution(f),
2432 format_field(f, 'fps', '%d'),
2433 '|',
2434 format_field(f, 'filesize', ' %s', func=format_bytes) + format_field(f, 'filesize_approx', '~%s', func=format_bytes),
2435 format_field(f, 'tbr', '%4dk'),
2436 f.get('protocol').replace('http_dash_segments', 'dash').replace("native", "n"),
2437 '|',
2438 format_field(f, 'vcodec', default='unknown').replace('none', ''),
2439 format_field(f, 'vbr', '%4dk'),
2440 format_field(f, 'acodec', default='unknown').replace('none', ''),
2441 format_field(f, 'abr', '%3dk'),
2442 format_field(f, 'asr', '%5dHz'),
2443 self._format_note_table(f)]
2444 for f in formats
2445 if f.get('preference') is None or f['preference'] >= -1000]
2446 header_line = ['ID', 'EXT', 'RESOLUTION', 'FPS', '|', ' FILESIZE', ' TBR', 'PROTO',
2447 '|', 'VCODEC', ' VBR', 'ACODEC', ' ABR', ' ASR', 'NOTE']
2448 else:
2449 table = [
2450 [
2451 format_field(f, 'format_id'),
2452 format_field(f, 'ext'),
2453 self.format_resolution(f),
2454 self._format_note(f)]
2455 for f in formats
2456 if f.get('preference') is None or f['preference'] >= -1000]
2457 header_line = ['format code', 'extension', 'resolution', 'note']
2458
2459 # if len(formats) > 1:
2460 # table[-1][-1] += (' ' if table[-1][-1] else '') + '(best)'
2461 self.to_screen(
2462 '[info] Available formats for %s:\n%s' % (info_dict['id'], render_table(
2463 header_line,
2464 table,
2465 delim=new_format,
2466 extraGap=(0 if new_format else 1),
2467 hideEmpty=new_format)))
2468
2469 def list_thumbnails(self, info_dict):
2470 thumbnails = info_dict.get('thumbnails')
2471 if not thumbnails:
2472 self.to_screen('[info] No thumbnails present for %s' % info_dict['id'])
2473 return
2474
2475 self.to_screen(
2476 '[info] Thumbnails for %s:' % info_dict['id'])
2477 self.to_screen(render_table(
2478 ['ID', 'width', 'height', 'URL'],
2479 [[t['id'], t.get('width', 'unknown'), t.get('height', 'unknown'), t['url']] for t in thumbnails]))
2480
2481 def list_subtitles(self, video_id, subtitles, name='subtitles'):
2482 if not subtitles:
2483 self.to_screen('%s has no %s' % (video_id, name))
2484 return
2485 self.to_screen(
2486 'Available %s for %s:' % (name, video_id))
2487 self.to_screen(render_table(
2488 ['Language', 'formats'],
2489 [[lang, ', '.join(f['ext'] for f in reversed(formats))]
2490 for lang, formats in subtitles.items()]))
2491
2492 def urlopen(self, req):
2493 """ Start an HTTP download """
2494 if isinstance(req, compat_basestring):
2495 req = sanitized_Request(req)
2496 return self._opener.open(req, timeout=self._socket_timeout)
2497
2498 def print_debug_header(self):
2499 if not self.params.get('verbose'):
2500 return
2501
2502 if type('') is not compat_str:
2503 # Python 2.6 on SLES11 SP1 (https://github.com/ytdl-org/youtube-dl/issues/3326)
2504 self.report_warning(
2505 'Your Python is broken! Update to a newer and supported version')
2506
2507 stdout_encoding = getattr(
2508 sys.stdout, 'encoding', 'missing (%s)' % type(sys.stdout).__name__)
2509 encoding_str = (
2510 '[debug] Encodings: locale %s, fs %s, out %s, pref %s\n' % (
2511 locale.getpreferredencoding(),
2512 sys.getfilesystemencoding(),
2513 stdout_encoding,
2514 self.get_encoding()))
2515 write_string(encoding_str, encoding=None)
2516
2517 self._write_string('[debug] youtube-dlc version ' + __version__ + '\n')
2518 if _LAZY_LOADER:
2519 self._write_string('[debug] Lazy loading extractors enabled' + '\n')
2520 try:
2521 sp = subprocess.Popen(
2522 ['git', 'rev-parse', '--short', 'HEAD'],
2523 stdout=subprocess.PIPE, stderr=subprocess.PIPE,
2524 cwd=os.path.dirname(os.path.abspath(__file__)))
2525 out, err = process_communicate_or_kill(sp)
2526 out = out.decode().strip()
2527 if re.match('[0-9a-f]+', out):
2528 self._write_string('[debug] Git HEAD: ' + out + '\n')
2529 except Exception:
2530 try:
2531 sys.exc_clear()
2532 except Exception:
2533 pass
2534
2535 def python_implementation():
2536 impl_name = platform.python_implementation()
2537 if impl_name == 'PyPy' and hasattr(sys, 'pypy_version_info'):
2538 return impl_name + ' version %d.%d.%d' % sys.pypy_version_info[:3]
2539 return impl_name
2540
2541 self._write_string('[debug] Python version %s (%s) - %s\n' % (
2542 platform.python_version(), python_implementation(),
2543 platform_name()))
2544
2545 exe_versions = FFmpegPostProcessor.get_versions(self)
2546 exe_versions['rtmpdump'] = rtmpdump_version()
2547 exe_versions['phantomjs'] = PhantomJSwrapper._version()
2548 exe_str = ', '.join(
2549 '%s %s' % (exe, v)
2550 for exe, v in sorted(exe_versions.items())
2551 if v
2552 )
2553 if not exe_str:
2554 exe_str = 'none'
2555 self._write_string('[debug] exe versions: %s\n' % exe_str)
2556
2557 proxy_map = {}
2558 for handler in self._opener.handlers:
2559 if hasattr(handler, 'proxies'):
2560 proxy_map.update(handler.proxies)
2561 self._write_string('[debug] Proxy map: ' + compat_str(proxy_map) + '\n')
2562
2563 if self.params.get('call_home', False):
2564 ipaddr = self.urlopen('https://yt-dl.org/ip').read().decode('utf-8')
2565 self._write_string('[debug] Public IP address: %s\n' % ipaddr)
2566 return
2567 latest_version = self.urlopen(
2568 'https://yt-dl.org/latest/version').read().decode('utf-8')
2569 if version_tuple(latest_version) > version_tuple(__version__):
2570 self.report_warning(
2571 'You are using an outdated version (newest version: %s)! '
2572 'See https://yt-dl.org/update if you need help updating.' %
2573 latest_version)
2574
2575 def _setup_opener(self):
2576 timeout_val = self.params.get('socket_timeout')
2577 self._socket_timeout = 600 if timeout_val is None else float(timeout_val)
2578
2579 opts_cookiefile = self.params.get('cookiefile')
2580 opts_proxy = self.params.get('proxy')
2581
2582 if opts_cookiefile is None:
2583 self.cookiejar = compat_cookiejar.CookieJar()
2584 else:
2585 opts_cookiefile = expand_path(opts_cookiefile)
2586 self.cookiejar = YoutubeDLCookieJar(opts_cookiefile)
2587 if os.access(opts_cookiefile, os.R_OK):
2588 self.cookiejar.load(ignore_discard=True, ignore_expires=True)
2589
2590 cookie_processor = YoutubeDLCookieProcessor(self.cookiejar)
2591 if opts_proxy is not None:
2592 if opts_proxy == '':
2593 proxies = {}
2594 else:
2595 proxies = {'http': opts_proxy, 'https': opts_proxy}
2596 else:
2597 proxies = compat_urllib_request.getproxies()
2598 # Set HTTPS proxy to HTTP one if given (https://github.com/ytdl-org/youtube-dl/issues/805)
2599 if 'http' in proxies and 'https' not in proxies:
2600 proxies['https'] = proxies['http']
2601 proxy_handler = PerRequestProxyHandler(proxies)
2602
2603 debuglevel = 1 if self.params.get('debug_printtraffic') else 0
2604 https_handler = make_HTTPS_handler(self.params, debuglevel=debuglevel)
2605 ydlh = YoutubeDLHandler(self.params, debuglevel=debuglevel)
2606 redirect_handler = YoutubeDLRedirectHandler()
2607 data_handler = compat_urllib_request_DataHandler()
2608
2609 # When passing our own FileHandler instance, build_opener won't add the
2610 # default FileHandler and allows us to disable the file protocol, which
2611 # can be used for malicious purposes (see
2612 # https://github.com/ytdl-org/youtube-dl/issues/8227)
2613 file_handler = compat_urllib_request.FileHandler()
2614
2615 def file_open(*args, **kwargs):
2616 raise compat_urllib_error.URLError('file:// scheme is explicitly disabled in youtube-dlc for security reasons')
2617 file_handler.file_open = file_open
2618
2619 opener = compat_urllib_request.build_opener(
2620 proxy_handler, https_handler, cookie_processor, ydlh, redirect_handler, data_handler, file_handler)
2621
2622 # Delete the default user-agent header, which would otherwise apply in
2623 # cases where our custom HTTP handler doesn't come into play
2624 # (See https://github.com/ytdl-org/youtube-dl/issues/1309 for details)
2625 opener.addheaders = []
2626 self._opener = opener
2627
2628 def encode(self, s):
2629 if isinstance(s, bytes):
2630 return s # Already encoded
2631
2632 try:
2633 return s.encode(self.get_encoding())
2634 except UnicodeEncodeError as err:
2635 err.reason = err.reason + '. Check your system encoding configuration or use the --encoding option.'
2636 raise
2637
2638 def get_encoding(self):
2639 encoding = self.params.get('encoding')
2640 if encoding is None:
2641 encoding = preferredencoding()
2642 return encoding
2643
2644 def _write_thumbnails(self, info_dict, filename):
2645 if self.params.get('writethumbnail', False):
2646 thumbnails = info_dict.get('thumbnails')
2647 if thumbnails:
2648 thumbnails = [thumbnails[-1]]
2649 elif self.params.get('write_all_thumbnails', False):
2650 thumbnails = info_dict.get('thumbnails')
2651 else:
2652 return
2653
2654 if not thumbnails:
2655 # No thumbnails present, so return immediately
2656 return
2657
2658 for t in thumbnails:
2659 thumb_ext = determine_ext(t['url'], 'jpg')
2660 suffix = '_%s' % t['id'] if len(thumbnails) > 1 else ''
2661 thumb_display_id = '%s ' % t['id'] if len(thumbnails) > 1 else ''
2662 t['filename'] = thumb_filename = replace_extension(filename + suffix, thumb_ext, info_dict.get('ext'))
2663
2664 if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(thumb_filename)):
2665 self.to_screen('[%s] %s: Thumbnail %sis already present' %
2666 (info_dict['extractor'], info_dict['id'], thumb_display_id))
2667 else:
2668 self.to_screen('[%s] %s: Downloading thumbnail %s...' %
2669 (info_dict['extractor'], info_dict['id'], thumb_display_id))
2670 try:
2671 uf = self.urlopen(t['url'])
2672 with open(encodeFilename(thumb_filename), 'wb') as thumbf:
2673 shutil.copyfileobj(uf, thumbf)
2674 self.to_screen('[%s] %s: Writing thumbnail %sto: %s' %
2675 (info_dict['extractor'], info_dict['id'], thumb_display_id, thumb_filename))
2676 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
2677 self.report_warning('Unable to download thumbnail "%s": %s' %
2678 (t['url'], error_to_compat_str(err)))