]> jfr.im git - yt-dlp.git/blob - yt_dlp/YoutubeDL.py
[build] Build Windows x86 version with py3.7
[yt-dlp.git] / yt_dlp / YoutubeDL.py
1 #!/usr/bin/env python3
2 # coding: utf-8
3
4 from __future__ import absolute_import, unicode_literals
5
6 import collections
7 import contextlib
8 import copy
9 import datetime
10 import errno
11 import fileinput
12 import io
13 import itertools
14 import json
15 import locale
16 import operator
17 import os
18 import platform
19 import re
20 import shutil
21 import subprocess
22 import sys
23 import time
24 import tokenize
25 import traceback
26 import random
27
28 from string import ascii_letters
29 from zipimport import zipimporter
30
31 from .compat import (
32 compat_basestring,
33 compat_cookiejar,
34 compat_get_terminal_size,
35 compat_kwargs,
36 compat_numeric_types,
37 compat_os_name,
38 compat_str,
39 compat_tokenize_tokenize,
40 compat_urllib_error,
41 compat_urllib_request,
42 compat_urllib_request_DataHandler,
43 )
44 from .utils import (
45 age_restricted,
46 args_to_str,
47 ContentTooShortError,
48 date_from_str,
49 DateRange,
50 DEFAULT_OUTTMPL,
51 determine_ext,
52 determine_protocol,
53 DOT_DESKTOP_LINK_TEMPLATE,
54 DOT_URL_LINK_TEMPLATE,
55 DOT_WEBLOC_LINK_TEMPLATE,
56 DownloadError,
57 encode_compat_str,
58 encodeFilename,
59 EntryNotInPlaylist,
60 error_to_compat_str,
61 ExistingVideoReached,
62 expand_path,
63 ExtractorError,
64 float_or_none,
65 format_bytes,
66 format_field,
67 STR_FORMAT_RE,
68 formatSeconds,
69 GeoRestrictedError,
70 int_or_none,
71 iri_to_uri,
72 ISO3166Utils,
73 LazyList,
74 locked_file,
75 make_dir,
76 make_HTTPS_handler,
77 MaxDownloadsReached,
78 network_exceptions,
79 orderedSet,
80 OUTTMPL_TYPES,
81 PagedList,
82 parse_filesize,
83 PerRequestProxyHandler,
84 platform_name,
85 PostProcessingError,
86 preferredencoding,
87 prepend_extension,
88 process_communicate_or_kill,
89 random_uuidv4,
90 register_socks_protocols,
91 RejectedVideoReached,
92 render_table,
93 replace_extension,
94 SameFileError,
95 sanitize_filename,
96 sanitize_path,
97 sanitize_url,
98 sanitized_Request,
99 std_headers,
100 str_or_none,
101 strftime_or_none,
102 subtitles_filename,
103 to_high_limit_path,
104 traverse_obj,
105 UnavailableVideoError,
106 url_basename,
107 version_tuple,
108 write_json_file,
109 write_string,
110 YoutubeDLCookieJar,
111 YoutubeDLCookieProcessor,
112 YoutubeDLHandler,
113 YoutubeDLRedirectHandler,
114 )
115 from .cache import Cache
116 from .extractor import (
117 gen_extractor_classes,
118 get_info_extractor,
119 _LAZY_LOADER,
120 _PLUGIN_CLASSES
121 )
122 from .extractor.openload import PhantomJSwrapper
123 from .downloader import (
124 get_suitable_downloader,
125 shorten_protocol_name
126 )
127 from .downloader.rtmp import rtmpdump_version
128 from .postprocessor import (
129 FFmpegFixupM3u8PP,
130 FFmpegFixupM4aPP,
131 FFmpegFixupStretchedPP,
132 FFmpegMergerPP,
133 FFmpegPostProcessor,
134 # FFmpegSubtitlesConvertorPP,
135 get_postprocessor,
136 MoveFilesAfterDownloadPP,
137 )
138 from .version import __version__
139
140 if compat_os_name == 'nt':
141 import ctypes
142
143
144 class YoutubeDL(object):
145 """YoutubeDL class.
146
147 YoutubeDL objects are the ones responsible of downloading the
148 actual video file and writing it to disk if the user has requested
149 it, among some other tasks. In most cases there should be one per
150 program. As, given a video URL, the downloader doesn't know how to
151 extract all the needed information, task that InfoExtractors do, it
152 has to pass the URL to one of them.
153
154 For this, YoutubeDL objects have a method that allows
155 InfoExtractors to be registered in a given order. When it is passed
156 a URL, the YoutubeDL object handles it to the first InfoExtractor it
157 finds that reports being able to handle it. The InfoExtractor extracts
158 all the information about the video or videos the URL refers to, and
159 YoutubeDL process the extracted information, possibly using a File
160 Downloader to download the video.
161
162 YoutubeDL objects accept a lot of parameters. In order not to saturate
163 the object constructor with arguments, it receives a dictionary of
164 options instead. These options are available through the params
165 attribute for the InfoExtractors to use. The YoutubeDL also
166 registers itself as the downloader in charge for the InfoExtractors
167 that are added to it, so this is a "mutual registration".
168
169 Available options:
170
171 username: Username for authentication purposes.
172 password: Password for authentication purposes.
173 videopassword: Password for accessing a video.
174 ap_mso: Adobe Pass multiple-system operator identifier.
175 ap_username: Multiple-system operator account username.
176 ap_password: Multiple-system operator account password.
177 usenetrc: Use netrc for authentication instead.
178 verbose: Print additional info to stdout.
179 quiet: Do not print messages to stdout.
180 no_warnings: Do not print out anything for warnings.
181 forceprint: A list of templates to force print
182 forceurl: Force printing final URL. (Deprecated)
183 forcetitle: Force printing title. (Deprecated)
184 forceid: Force printing ID. (Deprecated)
185 forcethumbnail: Force printing thumbnail URL. (Deprecated)
186 forcedescription: Force printing description. (Deprecated)
187 forcefilename: Force printing final filename. (Deprecated)
188 forceduration: Force printing duration. (Deprecated)
189 forcejson: Force printing info_dict as JSON.
190 dump_single_json: Force printing the info_dict of the whole playlist
191 (or video) as a single JSON line.
192 force_write_download_archive: Force writing download archive regardless
193 of 'skip_download' or 'simulate'.
194 simulate: Do not download the video files.
195 format: Video format code. see "FORMAT SELECTION" for more details.
196 allow_unplayable_formats: Allow unplayable formats to be extracted and downloaded.
197 ignore_no_formats_error: Ignore "No video formats" error. Usefull for
198 extracting metadata even if the video is not actually
199 available for download (experimental)
200 format_sort: How to sort the video formats. see "Sorting Formats"
201 for more details.
202 format_sort_force: Force the given format_sort. see "Sorting Formats"
203 for more details.
204 allow_multiple_video_streams: Allow multiple video streams to be merged
205 into a single file
206 allow_multiple_audio_streams: Allow multiple audio streams to be merged
207 into a single file
208 paths: Dictionary of output paths. The allowed keys are 'home'
209 'temp' and the keys of OUTTMPL_TYPES (in utils.py)
210 outtmpl: Dictionary of templates for output names. Allowed keys
211 are 'default' and the keys of OUTTMPL_TYPES (in utils.py).
212 A string a also accepted for backward compatibility
213 outtmpl_na_placeholder: Placeholder for unavailable meta fields.
214 restrictfilenames: Do not allow "&" and spaces in file names
215 trim_file_name: Limit length of filename (extension excluded)
216 windowsfilenames: Force the filenames to be windows compatible
217 ignoreerrors: Do not stop on download errors
218 (Default True when running yt-dlp,
219 but False when directly accessing YoutubeDL class)
220 skip_playlist_after_errors: Number of allowed failures until the rest of
221 the playlist is skipped
222 force_generic_extractor: Force downloader to use the generic extractor
223 overwrites: Overwrite all video and metadata files if True,
224 overwrite only non-video files if None
225 and don't overwrite any file if False
226 playliststart: Playlist item to start at.
227 playlistend: Playlist item to end at.
228 playlist_items: Specific indices of playlist to download.
229 playlistreverse: Download playlist items in reverse order.
230 playlistrandom: Download playlist items in random order.
231 matchtitle: Download only matching titles.
232 rejecttitle: Reject downloads for matching titles.
233 logger: Log messages to a logging.Logger instance.
234 logtostderr: Log messages to stderr instead of stdout.
235 writedescription: Write the video description to a .description file
236 writeinfojson: Write the video description to a .info.json file
237 clean_infojson: Remove private fields from the infojson
238 writecomments: Extract video comments. This will not be written to disk
239 unless writeinfojson is also given
240 writeannotations: Write the video annotations to a .annotations.xml file
241 writethumbnail: Write the thumbnail image to a file
242 allow_playlist_files: Whether to write playlists' description, infojson etc
243 also to disk when using the 'write*' options
244 write_all_thumbnails: Write all thumbnail formats to files
245 writelink: Write an internet shortcut file, depending on the
246 current platform (.url/.webloc/.desktop)
247 writeurllink: Write a Windows internet shortcut file (.url)
248 writewebloclink: Write a macOS internet shortcut file (.webloc)
249 writedesktoplink: Write a Linux internet shortcut file (.desktop)
250 writesubtitles: Write the video subtitles to a file
251 writeautomaticsub: Write the automatically generated subtitles to a file
252 allsubtitles: Deprecated - Use subtitlelangs = ['all']
253 Downloads all the subtitles of the video
254 (requires writesubtitles or writeautomaticsub)
255 listsubtitles: Lists all available subtitles for the video
256 subtitlesformat: The format code for subtitles
257 subtitleslangs: List of languages of the subtitles to download (can be regex).
258 The list may contain "all" to refer to all the available
259 subtitles. The language can be prefixed with a "-" to
260 exclude it from the requested languages. Eg: ['all', '-live_chat']
261 keepvideo: Keep the video file after post-processing
262 daterange: A DateRange object, download only if the upload_date is in the range.
263 skip_download: Skip the actual download of the video file
264 cachedir: Location of the cache files in the filesystem.
265 False to disable filesystem cache.
266 noplaylist: Download single video instead of a playlist if in doubt.
267 age_limit: An integer representing the user's age in years.
268 Unsuitable videos for the given age are skipped.
269 min_views: An integer representing the minimum view count the video
270 must have in order to not be skipped.
271 Videos without view count information are always
272 downloaded. None for no limit.
273 max_views: An integer representing the maximum view count.
274 Videos that are more popular than that are not
275 downloaded.
276 Videos without view count information are always
277 downloaded. None for no limit.
278 download_archive: File name of a file where all downloads are recorded.
279 Videos already present in the file are not downloaded
280 again.
281 break_on_existing: Stop the download process after attempting to download a
282 file that is in the archive.
283 break_on_reject: Stop the download process when encountering a video that
284 has been filtered out.
285 cookiefile: File name where cookies should be read from and dumped to
286 nocheckcertificate:Do not verify SSL certificates
287 prefer_insecure: Use HTTP instead of HTTPS to retrieve information.
288 At the moment, this is only supported by YouTube.
289 proxy: URL of the proxy server to use
290 geo_verification_proxy: URL of the proxy to use for IP address verification
291 on geo-restricted sites.
292 socket_timeout: Time to wait for unresponsive hosts, in seconds
293 bidi_workaround: Work around buggy terminals without bidirectional text
294 support, using fridibi
295 debug_printtraffic:Print out sent and received HTTP traffic
296 include_ads: Download ads as well
297 default_search: Prepend this string if an input url is not valid.
298 'auto' for elaborate guessing
299 encoding: Use this encoding instead of the system-specified.
300 extract_flat: Do not resolve URLs, return the immediate result.
301 Pass in 'in_playlist' to only show this behavior for
302 playlist items.
303 postprocessors: A list of dictionaries, each with an entry
304 * key: The name of the postprocessor. See
305 yt_dlp/postprocessor/__init__.py for a list.
306 * when: When to run the postprocessor. Can be one of
307 pre_process|before_dl|post_process|after_move.
308 Assumed to be 'post_process' if not given
309 post_hooks: A list of functions that get called as the final step
310 for each video file, after all postprocessors have been
311 called. The filename will be passed as the only argument.
312 progress_hooks: A list of functions that get called on download
313 progress, with a dictionary with the entries
314 * status: One of "downloading", "error", or "finished".
315 Check this first and ignore unknown values.
316
317 If status is one of "downloading", or "finished", the
318 following properties may also be present:
319 * filename: The final filename (always present)
320 * tmpfilename: The filename we're currently writing to
321 * downloaded_bytes: Bytes on disk
322 * total_bytes: Size of the whole file, None if unknown
323 * total_bytes_estimate: Guess of the eventual file size,
324 None if unavailable.
325 * elapsed: The number of seconds since download started.
326 * eta: The estimated time in seconds, None if unknown
327 * speed: The download speed in bytes/second, None if
328 unknown
329 * fragment_index: The counter of the currently
330 downloaded video fragment.
331 * fragment_count: The number of fragments (= individual
332 files that will be merged)
333
334 Progress hooks are guaranteed to be called at least once
335 (with status "finished") if the download is successful.
336 merge_output_format: Extension to use when merging formats.
337 final_ext: Expected final extension; used to detect when the file was
338 already downloaded and converted. "merge_output_format" is
339 replaced by this extension when given
340 fixup: Automatically correct known faults of the file.
341 One of:
342 - "never": do nothing
343 - "warn": only emit a warning
344 - "detect_or_warn": check whether we can do anything
345 about it, warn otherwise (default)
346 source_address: Client-side IP address to bind to.
347 call_home: Boolean, true iff we are allowed to contact the
348 yt-dlp servers for debugging. (BROKEN)
349 sleep_interval_requests: Number of seconds to sleep between requests
350 during extraction
351 sleep_interval: Number of seconds to sleep before each download when
352 used alone or a lower bound of a range for randomized
353 sleep before each download (minimum possible number
354 of seconds to sleep) when used along with
355 max_sleep_interval.
356 max_sleep_interval:Upper bound of a range for randomized sleep before each
357 download (maximum possible number of seconds to sleep).
358 Must only be used along with sleep_interval.
359 Actual sleep time will be a random float from range
360 [sleep_interval; max_sleep_interval].
361 sleep_interval_subtitles: Number of seconds to sleep before each subtitle download
362 listformats: Print an overview of available video formats and exit.
363 list_thumbnails: Print a table of all thumbnails and exit.
364 match_filter: A function that gets called with the info_dict of
365 every video.
366 If it returns a message, the video is ignored.
367 If it returns None, the video is downloaded.
368 match_filter_func in utils.py is one example for this.
369 no_color: Do not emit color codes in output.
370 geo_bypass: Bypass geographic restriction via faking X-Forwarded-For
371 HTTP header
372 geo_bypass_country:
373 Two-letter ISO 3166-2 country code that will be used for
374 explicit geographic restriction bypassing via faking
375 X-Forwarded-For HTTP header
376 geo_bypass_ip_block:
377 IP range in CIDR notation that will be used similarly to
378 geo_bypass_country
379
380 The following options determine which downloader is picked:
381 external_downloader: A dictionary of protocol keys and the executable of the
382 external downloader to use for it. The allowed protocols
383 are default|http|ftp|m3u8|dash|rtsp|rtmp|mms.
384 Set the value to 'native' to use the native downloader
385 hls_prefer_native: Deprecated - Use external_downloader = {'m3u8': 'native'}
386 or {'m3u8': 'ffmpeg'} instead.
387 Use the native HLS downloader instead of ffmpeg/avconv
388 if True, otherwise use ffmpeg/avconv if False, otherwise
389 use downloader suggested by extractor if None.
390 compat_opts: Compatibility options. See "Differences in default behavior".
391 Note that only format-sort, format-spec, no-live-chat,
392 no-attach-info-json, playlist-index, list-formats,
393 no-direct-merge, no-youtube-channel-redirect,
394 and no-youtube-unavailable-videos works when used via the API
395
396 The following parameters are not used by YoutubeDL itself, they are used by
397 the downloader (see yt_dlp/downloader/common.py):
398 nopart, updatetime, buffersize, ratelimit, min_filesize, max_filesize, test,
399 noresizebuffer, retries, continuedl, noprogress, consoletitle,
400 xattr_set_filesize, external_downloader_args, hls_use_mpegts,
401 http_chunk_size.
402
403 The following options are used by the post processors:
404 prefer_ffmpeg: If False, use avconv instead of ffmpeg if both are available,
405 otherwise prefer ffmpeg. (avconv support is deprecated)
406 ffmpeg_location: Location of the ffmpeg/avconv binary; either the path
407 to the binary or its containing directory.
408 postprocessor_args: A dictionary of postprocessor/executable keys (in lower case)
409 and a list of additional command-line arguments for the
410 postprocessor/executable. The dict can also have "PP+EXE" keys
411 which are used when the given exe is used by the given PP.
412 Use 'default' as the name for arguments to passed to all PP
413
414 The following options are used by the extractors:
415 extractor_retries: Number of times to retry for known errors
416 dynamic_mpd: Whether to process dynamic DASH manifests (default: True)
417 hls_split_discontinuity: Split HLS playlists to different formats at
418 discontinuities such as ad breaks (default: False)
419 youtube_include_dash_manifest: If True (default), DASH manifests and related
420 data will be downloaded and processed by extractor.
421 You can reduce network I/O by disabling it if you don't
422 care about DASH. (only for youtube)
423 youtube_include_hls_manifest: If True (default), HLS manifests and related
424 data will be downloaded and processed by extractor.
425 You can reduce network I/O by disabling it if you don't
426 care about HLS. (only for youtube)
427 """
428
429 _NUMERIC_FIELDS = set((
430 'width', 'height', 'tbr', 'abr', 'asr', 'vbr', 'fps', 'filesize', 'filesize_approx',
431 'timestamp', 'upload_year', 'upload_month', 'upload_day',
432 'duration', 'view_count', 'like_count', 'dislike_count', 'repost_count',
433 'average_rating', 'comment_count', 'age_limit',
434 'start_time', 'end_time',
435 'chapter_number', 'season_number', 'episode_number',
436 'track_number', 'disc_number', 'release_year',
437 'playlist_index',
438 ))
439
440 params = None
441 _ies = []
442 _pps = {'pre_process': [], 'before_dl': [], 'after_move': [], 'post_process': []}
443 __prepare_filename_warned = False
444 _first_webpage_request = True
445 _download_retcode = None
446 _num_downloads = None
447 _playlist_level = 0
448 _playlist_urls = set()
449 _screen_file = None
450
451 def __init__(self, params=None, auto_init=True):
452 """Create a FileDownloader object with the given options."""
453 if params is None:
454 params = {}
455 self._ies = []
456 self._ies_instances = {}
457 self._pps = {'pre_process': [], 'before_dl': [], 'after_move': [], 'post_process': []}
458 self.__prepare_filename_warned = False
459 self._first_webpage_request = True
460 self._post_hooks = []
461 self._progress_hooks = []
462 self._download_retcode = 0
463 self._num_downloads = 0
464 self._screen_file = [sys.stdout, sys.stderr][params.get('logtostderr', False)]
465 self._err_file = sys.stderr
466 self.params = {
467 # Default parameters
468 'nocheckcertificate': False,
469 }
470 self.params.update(params)
471 self.cache = Cache(self)
472
473 if sys.version_info < (3, 6):
474 self.report_warning(
475 'Python version %d.%d is not supported! Please update to Python 3.6 or above' % sys.version_info[:2])
476
477 def check_deprecated(param, option, suggestion):
478 if self.params.get(param) is not None:
479 self.report_warning('%s is deprecated. Use %s instead' % (option, suggestion))
480 return True
481 return False
482
483 if check_deprecated('cn_verification_proxy', '--cn-verification-proxy', '--geo-verification-proxy'):
484 if self.params.get('geo_verification_proxy') is None:
485 self.params['geo_verification_proxy'] = self.params['cn_verification_proxy']
486
487 check_deprecated('autonumber', '--auto-number', '-o "%(autonumber)s-%(title)s.%(ext)s"')
488 check_deprecated('usetitle', '--title', '-o "%(title)s-%(id)s.%(ext)s"')
489 check_deprecated('useid', '--id', '-o "%(id)s.%(ext)s"')
490
491 for msg in self.params.get('warnings', []):
492 self.report_warning(msg)
493
494 if self.params.get('final_ext'):
495 if self.params.get('merge_output_format'):
496 self.report_warning('--merge-output-format will be ignored since --remux-video or --recode-video is given')
497 self.params['merge_output_format'] = self.params['final_ext']
498
499 if 'overwrites' in self.params and self.params['overwrites'] is None:
500 del self.params['overwrites']
501
502 if params.get('bidi_workaround', False):
503 try:
504 import pty
505 master, slave = pty.openpty()
506 width = compat_get_terminal_size().columns
507 if width is None:
508 width_args = []
509 else:
510 width_args = ['-w', str(width)]
511 sp_kwargs = dict(
512 stdin=subprocess.PIPE,
513 stdout=slave,
514 stderr=self._err_file)
515 try:
516 self._output_process = subprocess.Popen(
517 ['bidiv'] + width_args, **sp_kwargs
518 )
519 except OSError:
520 self._output_process = subprocess.Popen(
521 ['fribidi', '-c', 'UTF-8'] + width_args, **sp_kwargs)
522 self._output_channel = os.fdopen(master, 'rb')
523 except OSError as ose:
524 if ose.errno == errno.ENOENT:
525 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 $PATH.')
526 else:
527 raise
528
529 if (sys.platform != 'win32'
530 and sys.getfilesystemencoding() in ['ascii', 'ANSI_X3.4-1968']
531 and not params.get('restrictfilenames', False)):
532 # Unicode filesystem API will throw errors (#1474, #13027)
533 self.report_warning(
534 'Assuming --restrict-filenames since file system encoding '
535 'cannot encode all characters. '
536 'Set the LC_ALL environment variable to fix this.')
537 self.params['restrictfilenames'] = True
538
539 self.outtmpl_dict = self.parse_outtmpl()
540
541 self._setup_opener()
542
543 """Preload the archive, if any is specified"""
544 def preload_download_archive(fn):
545 if fn is None:
546 return False
547 self.write_debug('Loading archive file %r\n' % fn)
548 try:
549 with locked_file(fn, 'r', encoding='utf-8') as archive_file:
550 for line in archive_file:
551 self.archive.add(line.strip())
552 except IOError as ioe:
553 if ioe.errno != errno.ENOENT:
554 raise
555 return False
556 return True
557
558 self.archive = set()
559 preload_download_archive(self.params.get('download_archive'))
560
561 if auto_init:
562 self.print_debug_header()
563 self.add_default_info_extractors()
564
565 for pp_def_raw in self.params.get('postprocessors', []):
566 pp_class = get_postprocessor(pp_def_raw['key'])
567 pp_def = dict(pp_def_raw)
568 del pp_def['key']
569 if 'when' in pp_def:
570 when = pp_def['when']
571 del pp_def['when']
572 else:
573 when = 'post_process'
574 pp = pp_class(self, **compat_kwargs(pp_def))
575 self.add_post_processor(pp, when=when)
576
577 for ph in self.params.get('post_hooks', []):
578 self.add_post_hook(ph)
579
580 for ph in self.params.get('progress_hooks', []):
581 self.add_progress_hook(ph)
582
583 register_socks_protocols()
584
585 def warn_if_short_id(self, argv):
586 # short YouTube ID starting with dash?
587 idxs = [
588 i for i, a in enumerate(argv)
589 if re.match(r'^-[0-9A-Za-z_-]{10}$', a)]
590 if idxs:
591 correct_argv = (
592 ['yt-dlp']
593 + [a for i, a in enumerate(argv) if i not in idxs]
594 + ['--'] + [argv[i] for i in idxs]
595 )
596 self.report_warning(
597 'Long argument string detected. '
598 'Use -- to separate parameters and URLs, like this:\n%s\n' %
599 args_to_str(correct_argv))
600
601 def add_info_extractor(self, ie):
602 """Add an InfoExtractor object to the end of the list."""
603 self._ies.append(ie)
604 if not isinstance(ie, type):
605 self._ies_instances[ie.ie_key()] = ie
606 ie.set_downloader(self)
607
608 def get_info_extractor(self, ie_key):
609 """
610 Get an instance of an IE with name ie_key, it will try to get one from
611 the _ies list, if there's no instance it will create a new one and add
612 it to the extractor list.
613 """
614 ie = self._ies_instances.get(ie_key)
615 if ie is None:
616 ie = get_info_extractor(ie_key)()
617 self.add_info_extractor(ie)
618 return ie
619
620 def add_default_info_extractors(self):
621 """
622 Add the InfoExtractors returned by gen_extractors to the end of the list
623 """
624 for ie in gen_extractor_classes():
625 self.add_info_extractor(ie)
626
627 def add_post_processor(self, pp, when='post_process'):
628 """Add a PostProcessor object to the end of the chain."""
629 self._pps[when].append(pp)
630 pp.set_downloader(self)
631
632 def add_post_hook(self, ph):
633 """Add the post hook"""
634 self._post_hooks.append(ph)
635
636 def add_progress_hook(self, ph):
637 """Add the progress hook (currently only for the file downloader)"""
638 self._progress_hooks.append(ph)
639
640 def _bidi_workaround(self, message):
641 if not hasattr(self, '_output_channel'):
642 return message
643
644 assert hasattr(self, '_output_process')
645 assert isinstance(message, compat_str)
646 line_count = message.count('\n') + 1
647 self._output_process.stdin.write((message + '\n').encode('utf-8'))
648 self._output_process.stdin.flush()
649 res = ''.join(self._output_channel.readline().decode('utf-8')
650 for _ in range(line_count))
651 return res[:-len('\n')]
652
653 def _write_string(self, s, out=None):
654 write_string(s, out=out, encoding=self.params.get('encoding'))
655
656 def to_stdout(self, message, skip_eol=False, quiet=False):
657 """Print message to stdout"""
658 if self.params.get('logger'):
659 self.params['logger'].debug(message)
660 elif not quiet or self.params.get('verbose'):
661 self._write_string(
662 '%s%s' % (self._bidi_workaround(message), ('' if skip_eol else '\n')),
663 self._err_file if quiet else self._screen_file)
664
665 def to_stderr(self, message):
666 """Print message to stderr"""
667 assert isinstance(message, compat_str)
668 if self.params.get('logger'):
669 self.params['logger'].error(message)
670 else:
671 self._write_string('%s\n' % self._bidi_workaround(message), self._err_file)
672
673 def to_console_title(self, message):
674 if not self.params.get('consoletitle', False):
675 return
676 if compat_os_name == 'nt':
677 if ctypes.windll.kernel32.GetConsoleWindow():
678 # c_wchar_p() might not be necessary if `message` is
679 # already of type unicode()
680 ctypes.windll.kernel32.SetConsoleTitleW(ctypes.c_wchar_p(message))
681 elif 'TERM' in os.environ:
682 self._write_string('\033]0;%s\007' % message, self._screen_file)
683
684 def save_console_title(self):
685 if not self.params.get('consoletitle', False):
686 return
687 if self.params.get('simulate', False):
688 return
689 if compat_os_name != 'nt' and 'TERM' in os.environ:
690 # Save the title on stack
691 self._write_string('\033[22;0t', self._screen_file)
692
693 def restore_console_title(self):
694 if not self.params.get('consoletitle', False):
695 return
696 if self.params.get('simulate', False):
697 return
698 if compat_os_name != 'nt' and 'TERM' in os.environ:
699 # Restore the title from stack
700 self._write_string('\033[23;0t', self._screen_file)
701
702 def __enter__(self):
703 self.save_console_title()
704 return self
705
706 def __exit__(self, *args):
707 self.restore_console_title()
708
709 if self.params.get('cookiefile') is not None:
710 self.cookiejar.save(ignore_discard=True, ignore_expires=True)
711
712 def trouble(self, message=None, tb=None):
713 """Determine action to take when a download problem appears.
714
715 Depending on if the downloader has been configured to ignore
716 download errors or not, this method may throw an exception or
717 not when errors are found, after printing the message.
718
719 tb, if given, is additional traceback information.
720 """
721 if message is not None:
722 self.to_stderr(message)
723 if self.params.get('verbose'):
724 if tb is None:
725 if sys.exc_info()[0]: # if .trouble has been called from an except block
726 tb = ''
727 if hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
728 tb += ''.join(traceback.format_exception(*sys.exc_info()[1].exc_info))
729 tb += encode_compat_str(traceback.format_exc())
730 else:
731 tb_data = traceback.format_list(traceback.extract_stack())
732 tb = ''.join(tb_data)
733 if tb:
734 self.to_stderr(tb)
735 if not self.params.get('ignoreerrors', False):
736 if sys.exc_info()[0] and hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
737 exc_info = sys.exc_info()[1].exc_info
738 else:
739 exc_info = sys.exc_info()
740 raise DownloadError(message, exc_info)
741 self._download_retcode = 1
742
743 def to_screen(self, message, skip_eol=False):
744 """Print message to stdout if not in quiet mode"""
745 self.to_stdout(
746 message, skip_eol, quiet=self.params.get('quiet', False))
747
748 def report_warning(self, message):
749 '''
750 Print the message to stderr, it will be prefixed with 'WARNING:'
751 If stderr is a tty file the 'WARNING:' will be colored
752 '''
753 if self.params.get('logger') is not None:
754 self.params['logger'].warning(message)
755 else:
756 if self.params.get('no_warnings'):
757 return
758 if not self.params.get('no_color') and self._err_file.isatty() and compat_os_name != 'nt':
759 _msg_header = '\033[0;33mWARNING:\033[0m'
760 else:
761 _msg_header = 'WARNING:'
762 warning_message = '%s %s' % (_msg_header, message)
763 self.to_stderr(warning_message)
764
765 def report_error(self, message, tb=None):
766 '''
767 Do the same as trouble, but prefixes the message with 'ERROR:', colored
768 in red if stderr is a tty file.
769 '''
770 if not self.params.get('no_color') and self._err_file.isatty() and compat_os_name != 'nt':
771 _msg_header = '\033[0;31mERROR:\033[0m'
772 else:
773 _msg_header = 'ERROR:'
774 error_message = '%s %s' % (_msg_header, message)
775 self.trouble(error_message, tb)
776
777 def write_debug(self, message):
778 '''Log debug message or Print message to stderr'''
779 if not self.params.get('verbose', False):
780 return
781 message = '[debug] %s' % message
782 if self.params.get('logger'):
783 self.params['logger'].debug(message)
784 else:
785 self._write_string('%s\n' % message)
786
787 def report_file_already_downloaded(self, file_name):
788 """Report file has already been fully downloaded."""
789 try:
790 self.to_screen('[download] %s has already been downloaded' % file_name)
791 except UnicodeEncodeError:
792 self.to_screen('[download] The file has already been downloaded')
793
794 def report_file_delete(self, file_name):
795 """Report that existing file will be deleted."""
796 try:
797 self.to_screen('Deleting existing file %s' % file_name)
798 except UnicodeEncodeError:
799 self.to_screen('Deleting existing file')
800
801 def parse_outtmpl(self):
802 outtmpl_dict = self.params.get('outtmpl', {})
803 if not isinstance(outtmpl_dict, dict):
804 outtmpl_dict = {'default': outtmpl_dict}
805 outtmpl_dict.update({
806 k: v for k, v in DEFAULT_OUTTMPL.items()
807 if not outtmpl_dict.get(k)})
808 for key, val in outtmpl_dict.items():
809 if isinstance(val, bytes):
810 self.report_warning(
811 'Parameter outtmpl is bytes, but should be a unicode string. '
812 'Put from __future__ import unicode_literals at the top of your code file or consider switching to Python 3.x.')
813 return outtmpl_dict
814
815 @staticmethod
816 def validate_outtmpl(tmpl):
817 ''' @return None or Exception object '''
818 try:
819 re.sub(
820 STR_FORMAT_RE.format(''),
821 lambda mobj: ('%' if not mobj.group('has_key') else '') + mobj.group(0),
822 tmpl
823 ) % collections.defaultdict(int)
824 return None
825 except ValueError as err:
826 return err
827
828 def prepare_outtmpl(self, outtmpl, info_dict, sanitize=None):
829 """ Make the template and info_dict suitable for substitution (outtmpl % info_dict)"""
830 info_dict = dict(info_dict)
831 na = self.params.get('outtmpl_na_placeholder', 'NA')
832
833 info_dict['duration_string'] = ( # %(duration>%H-%M-%S)s is wrong if duration > 24hrs
834 formatSeconds(info_dict['duration'], '-' if sanitize else ':')
835 if info_dict.get('duration', None) is not None
836 else None)
837 info_dict['epoch'] = int(time.time())
838 info_dict['autonumber'] = self.params.get('autonumber_start', 1) - 1 + self._num_downloads
839 if info_dict.get('resolution') is None:
840 info_dict['resolution'] = self.format_resolution(info_dict, default=None)
841
842 # For fields playlist_index and autonumber convert all occurrences
843 # of %(field)s to %(field)0Nd for backward compatibility
844 field_size_compat_map = {
845 'playlist_index': len(str(info_dict.get('_last_playlist_index') or '')),
846 'autonumber': self.params.get('autonumber_size') or 5,
847 }
848
849 TMPL_DICT = {}
850 EXTERNAL_FORMAT_RE = re.compile(STR_FORMAT_RE.format('[^)]*'))
851 MATH_FUNCTIONS = {
852 '+': float.__add__,
853 '-': float.__sub__,
854 }
855 # Field is of the form key1.key2...
856 # where keys (except first) can be string, int or slice
857 FIELD_RE = r'\w+(?:\.(?:\w+|{num}|{num}?(?::{num}?){{1,2}}))*'.format(num=r'(?:-?\d+)')
858 MATH_FIELD_RE = r'''{field}|{num}'''.format(field=FIELD_RE, num=r'-?\d+(?:.\d+)?')
859 MATH_OPERATORS_RE = r'(?:%s)' % '|'.join(map(re.escape, MATH_FUNCTIONS.keys()))
860 INTERNAL_FORMAT_RE = re.compile(r'''(?x)
861 (?P<negate>-)?
862 (?P<fields>{field})
863 (?P<maths>(?:{math_op}{math_field})*)
864 (?:>(?P<strf_format>.+?))?
865 (?:\|(?P<default>.*?))?
866 $'''.format(field=FIELD_RE, math_op=MATH_OPERATORS_RE, math_field=MATH_FIELD_RE))
867
868 get_key = lambda k: traverse_obj(
869 info_dict, k.split('.'), is_user_input=True, traverse_string=True)
870
871 def get_value(mdict):
872 # Object traversal
873 value = get_key(mdict['fields'])
874 # Negative
875 if mdict['negate']:
876 value = float_or_none(value)
877 if value is not None:
878 value *= -1
879 # Do maths
880 offset_key = mdict['maths']
881 if offset_key:
882 value = float_or_none(value)
883 operator = None
884 while offset_key:
885 item = re.match(
886 MATH_FIELD_RE if operator else MATH_OPERATORS_RE,
887 offset_key).group(0)
888 offset_key = offset_key[len(item):]
889 if operator is None:
890 operator = MATH_FUNCTIONS[item]
891 continue
892 item, multiplier = (item[1:], -1) if item[0] == '-' else (item, 1)
893 offset = float_or_none(item)
894 if offset is None:
895 offset = float_or_none(get_key(item))
896 try:
897 value = operator(value, multiplier * offset)
898 except (TypeError, ZeroDivisionError):
899 return None
900 operator = None
901 # Datetime formatting
902 if mdict['strf_format']:
903 value = strftime_or_none(value, mdict['strf_format'])
904
905 return value
906
907 def create_key(outer_mobj):
908 if not outer_mobj.group('has_key'):
909 return '%{}'.format(outer_mobj.group(0))
910
911 key = outer_mobj.group('key')
912 fmt = outer_mobj.group('format')
913 mobj = re.match(INTERNAL_FORMAT_RE, key)
914 if mobj is None:
915 value, default = None, na
916 else:
917 mobj = mobj.groupdict()
918 default = mobj['default'] if mobj['default'] is not None else na
919 value = get_value(mobj)
920
921 if fmt == 's' and value is not None and key in field_size_compat_map.keys():
922 fmt = '0{:d}d'.format(field_size_compat_map[key])
923
924 value = default if value is None else value
925 key += '\0%s' % fmt
926
927 if fmt == 'c':
928 value = compat_str(value)
929 if value is None:
930 value, fmt = default, 's'
931 else:
932 value = value[0]
933 elif fmt[-1] not in 'rs': # numeric
934 value = float_or_none(value)
935 if value is None:
936 value, fmt = default, 's'
937 if sanitize:
938 if fmt[-1] == 'r':
939 # If value is an object, sanitize might convert it to a string
940 # So we convert it to repr first
941 value, fmt = repr(value), '%ss' % fmt[:-1]
942 if fmt[-1] in 'csr':
943 value = sanitize(key, value)
944 TMPL_DICT[key] = value
945 return '%({key}){fmt}'.format(key=key, fmt=fmt)
946
947 return EXTERNAL_FORMAT_RE.sub(create_key, outtmpl), TMPL_DICT
948
949 def _prepare_filename(self, info_dict, tmpl_type='default'):
950 try:
951 sanitize = lambda k, v: sanitize_filename(
952 compat_str(v),
953 restricted=self.params.get('restrictfilenames'),
954 is_id=(k == 'id' or k.endswith('_id')))
955 outtmpl = self.outtmpl_dict.get(tmpl_type, self.outtmpl_dict['default'])
956 outtmpl, template_dict = self.prepare_outtmpl(outtmpl, info_dict, sanitize)
957
958 # expand_path translates '%%' into '%' and '$$' into '$'
959 # correspondingly that is not what we want since we need to keep
960 # '%%' intact for template dict substitution step. Working around
961 # with boundary-alike separator hack.
962 sep = ''.join([random.choice(ascii_letters) for _ in range(32)])
963 outtmpl = outtmpl.replace('%%', '%{0}%'.format(sep)).replace('$$', '${0}$'.format(sep))
964
965 # outtmpl should be expand_path'ed before template dict substitution
966 # because meta fields may contain env variables we don't want to
967 # be expanded. For example, for outtmpl "%(title)s.%(ext)s" and
968 # title "Hello $PATH", we don't want `$PATH` to be expanded.
969 filename = expand_path(outtmpl).replace(sep, '') % template_dict
970
971 force_ext = OUTTMPL_TYPES.get(tmpl_type)
972 if force_ext is not None:
973 filename = replace_extension(filename, force_ext, info_dict.get('ext'))
974
975 # https://github.com/blackjack4494/youtube-dlc/issues/85
976 trim_file_name = self.params.get('trim_file_name', False)
977 if trim_file_name:
978 fn_groups = filename.rsplit('.')
979 ext = fn_groups[-1]
980 sub_ext = ''
981 if len(fn_groups) > 2:
982 sub_ext = fn_groups[-2]
983 filename = '.'.join(filter(None, [fn_groups[0][:trim_file_name], sub_ext, ext]))
984
985 return filename
986 except ValueError as err:
987 self.report_error('Error in output template: ' + str(err) + ' (encoding: ' + repr(preferredencoding()) + ')')
988 return None
989
990 def prepare_filename(self, info_dict, dir_type='', warn=False):
991 """Generate the output filename."""
992 paths = self.params.get('paths', {})
993 assert isinstance(paths, dict)
994 filename = self._prepare_filename(info_dict, dir_type or 'default')
995
996 if warn and not self.__prepare_filename_warned:
997 if not paths:
998 pass
999 elif filename == '-':
1000 self.report_warning('--paths is ignored when an outputting to stdout')
1001 elif os.path.isabs(filename):
1002 self.report_warning('--paths is ignored since an absolute path is given in output template')
1003 self.__prepare_filename_warned = True
1004 if filename == '-' or not filename:
1005 return filename
1006
1007 homepath = expand_path(paths.get('home', '').strip())
1008 assert isinstance(homepath, compat_str)
1009 subdir = expand_path(paths.get(dir_type, '').strip()) if dir_type else ''
1010 assert isinstance(subdir, compat_str)
1011 path = os.path.join(homepath, subdir, filename)
1012
1013 # Temporary fix for #4787
1014 # 'Treat' all problem characters by passing filename through preferredencoding
1015 # to workaround encoding issues with subprocess on python2 @ Windows
1016 if sys.version_info < (3, 0) and sys.platform == 'win32':
1017 path = encodeFilename(path, True).decode(preferredencoding())
1018 return sanitize_path(path, force=self.params.get('windowsfilenames'))
1019
1020 def _match_entry(self, info_dict, incomplete=False, silent=False):
1021 """ Returns None if the file should be downloaded """
1022
1023 video_title = info_dict.get('title', info_dict.get('id', 'video'))
1024
1025 def check_filter():
1026 if 'title' in info_dict:
1027 # This can happen when we're just evaluating the playlist
1028 title = info_dict['title']
1029 matchtitle = self.params.get('matchtitle', False)
1030 if matchtitle:
1031 if not re.search(matchtitle, title, re.IGNORECASE):
1032 return '"' + title + '" title did not match pattern "' + matchtitle + '"'
1033 rejecttitle = self.params.get('rejecttitle', False)
1034 if rejecttitle:
1035 if re.search(rejecttitle, title, re.IGNORECASE):
1036 return '"' + title + '" title matched reject pattern "' + rejecttitle + '"'
1037 date = info_dict.get('upload_date')
1038 if date is not None:
1039 dateRange = self.params.get('daterange', DateRange())
1040 if date not in dateRange:
1041 return '%s upload date is not in range %s' % (date_from_str(date).isoformat(), dateRange)
1042 view_count = info_dict.get('view_count')
1043 if view_count is not None:
1044 min_views = self.params.get('min_views')
1045 if min_views is not None and view_count < min_views:
1046 return 'Skipping %s, because it has not reached minimum view count (%d/%d)' % (video_title, view_count, min_views)
1047 max_views = self.params.get('max_views')
1048 if max_views is not None and view_count > max_views:
1049 return 'Skipping %s, because it has exceeded the maximum view count (%d/%d)' % (video_title, view_count, max_views)
1050 if age_restricted(info_dict.get('age_limit'), self.params.get('age_limit')):
1051 return 'Skipping "%s" because it is age restricted' % video_title
1052
1053 if not incomplete:
1054 match_filter = self.params.get('match_filter')
1055 if match_filter is not None:
1056 ret = match_filter(info_dict)
1057 if ret is not None:
1058 return ret
1059 return None
1060
1061 if self.in_download_archive(info_dict):
1062 reason = '%s has already been recorded in the archive' % video_title
1063 break_opt, break_err = 'break_on_existing', ExistingVideoReached
1064 else:
1065 reason = check_filter()
1066 break_opt, break_err = 'break_on_reject', RejectedVideoReached
1067 if reason is not None:
1068 if not silent:
1069 self.to_screen('[download] ' + reason)
1070 if self.params.get(break_opt, False):
1071 raise break_err()
1072 return reason
1073
1074 @staticmethod
1075 def add_extra_info(info_dict, extra_info):
1076 '''Set the keys from extra_info in info dict if they are missing'''
1077 for key, value in extra_info.items():
1078 info_dict.setdefault(key, value)
1079
1080 def extract_info(self, url, download=True, ie_key=None, extra_info={},
1081 process=True, force_generic_extractor=False):
1082 """
1083 Return a list with a dictionary for each video extracted.
1084
1085 Arguments:
1086 url -- URL to extract
1087
1088 Keyword arguments:
1089 download -- whether to download videos during extraction
1090 ie_key -- extractor key hint
1091 extra_info -- dictionary containing the extra values to add to each result
1092 process -- whether to resolve all unresolved references (URLs, playlist items),
1093 must be True for download to work.
1094 force_generic_extractor -- force using the generic extractor
1095 """
1096
1097 if not ie_key and force_generic_extractor:
1098 ie_key = 'Generic'
1099
1100 if ie_key:
1101 ies = [self.get_info_extractor(ie_key)]
1102 else:
1103 ies = self._ies
1104
1105 for ie in ies:
1106 if not ie.suitable(url):
1107 continue
1108
1109 ie_key = ie.ie_key()
1110 ie = self.get_info_extractor(ie_key)
1111 if not ie.working():
1112 self.report_warning('The program functionality for this site has been marked as broken, '
1113 'and will probably not work.')
1114
1115 try:
1116 temp_id = str_or_none(
1117 ie.extract_id(url) if callable(getattr(ie, 'extract_id', None))
1118 else ie._match_id(url))
1119 except (AssertionError, IndexError, AttributeError):
1120 temp_id = None
1121 if temp_id is not None and self.in_download_archive({'id': temp_id, 'ie_key': ie_key}):
1122 self.to_screen("[%s] %s: has already been recorded in archive" % (
1123 ie_key, temp_id))
1124 break
1125 return self.__extract_info(url, ie, download, extra_info, process)
1126 else:
1127 self.report_error('no suitable InfoExtractor for URL %s' % url)
1128
1129 def __handle_extraction_exceptions(func):
1130 def wrapper(self, *args, **kwargs):
1131 try:
1132 return func(self, *args, **kwargs)
1133 except GeoRestrictedError as e:
1134 msg = e.msg
1135 if e.countries:
1136 msg += '\nThis video is available in %s.' % ', '.join(
1137 map(ISO3166Utils.short2full, e.countries))
1138 msg += '\nYou might want to use a VPN or a proxy server (with --proxy) to workaround.'
1139 self.report_error(msg)
1140 except ExtractorError as e: # An error we somewhat expected
1141 self.report_error(compat_str(e), e.format_traceback())
1142 except (MaxDownloadsReached, ExistingVideoReached, RejectedVideoReached):
1143 raise
1144 except Exception as e:
1145 if self.params.get('ignoreerrors', False):
1146 self.report_error(error_to_compat_str(e), tb=encode_compat_str(traceback.format_exc()))
1147 else:
1148 raise
1149 return wrapper
1150
1151 @__handle_extraction_exceptions
1152 def __extract_info(self, url, ie, download, extra_info, process):
1153 ie_result = ie.extract(url)
1154 if ie_result is None: # Finished already (backwards compatibility; listformats and friends should be moved here)
1155 return
1156 if isinstance(ie_result, list):
1157 # Backwards compatibility: old IE result format
1158 ie_result = {
1159 '_type': 'compat_list',
1160 'entries': ie_result,
1161 }
1162 self.add_default_extra_info(ie_result, ie, url)
1163 if process:
1164 return self.process_ie_result(ie_result, download, extra_info)
1165 else:
1166 return ie_result
1167
1168 def add_default_extra_info(self, ie_result, ie, url):
1169 self.add_extra_info(ie_result, {
1170 'extractor': ie.IE_NAME,
1171 'webpage_url': url,
1172 'original_url': url,
1173 'webpage_url_basename': url_basename(url),
1174 'extractor_key': ie.ie_key(),
1175 })
1176
1177 def process_ie_result(self, ie_result, download=True, extra_info={}):
1178 """
1179 Take the result of the ie(may be modified) and resolve all unresolved
1180 references (URLs, playlist items).
1181
1182 It will also download the videos if 'download'.
1183 Returns the resolved ie_result.
1184 """
1185 result_type = ie_result.get('_type', 'video')
1186
1187 if result_type in ('url', 'url_transparent'):
1188 ie_result['url'] = sanitize_url(ie_result['url'])
1189 extract_flat = self.params.get('extract_flat', False)
1190 if ((extract_flat == 'in_playlist' and 'playlist' in extra_info)
1191 or extract_flat is True):
1192 info_copy = ie_result.copy()
1193 self.add_extra_info(info_copy, extra_info)
1194 self.add_default_extra_info(
1195 info_copy, self.get_info_extractor(ie_result.get('ie_key')), ie_result['url'])
1196 self.__forced_printings(info_copy, self.prepare_filename(info_copy), incomplete=True)
1197 return ie_result
1198
1199 if result_type == 'video':
1200 self.add_extra_info(ie_result, extra_info)
1201 ie_result = self.process_video_result(ie_result, download=download)
1202 additional_urls = (ie_result or {}).get('additional_urls')
1203 if additional_urls:
1204 # TODO: Improve MetadataFromFieldPP to allow setting a list
1205 if isinstance(additional_urls, compat_str):
1206 additional_urls = [additional_urls]
1207 self.to_screen(
1208 '[info] %s: %d additional URL(s) requested' % (ie_result['id'], len(additional_urls)))
1209 self.write_debug('Additional URLs: "%s"' % '", "'.join(additional_urls))
1210 ie_result['additional_entries'] = [
1211 self.extract_info(
1212 url, download, extra_info,
1213 force_generic_extractor=self.params.get('force_generic_extractor'))
1214 for url in additional_urls
1215 ]
1216 return ie_result
1217 elif result_type == 'url':
1218 # We have to add extra_info to the results because it may be
1219 # contained in a playlist
1220 return self.extract_info(
1221 ie_result['url'], download,
1222 ie_key=ie_result.get('ie_key'),
1223 extra_info=extra_info)
1224 elif result_type == 'url_transparent':
1225 # Use the information from the embedding page
1226 info = self.extract_info(
1227 ie_result['url'], ie_key=ie_result.get('ie_key'),
1228 extra_info=extra_info, download=False, process=False)
1229
1230 # extract_info may return None when ignoreerrors is enabled and
1231 # extraction failed with an error, don't crash and return early
1232 # in this case
1233 if not info:
1234 return info
1235
1236 force_properties = dict(
1237 (k, v) for k, v in ie_result.items() if v is not None)
1238 for f in ('_type', 'url', 'id', 'extractor', 'extractor_key', 'ie_key'):
1239 if f in force_properties:
1240 del force_properties[f]
1241 new_result = info.copy()
1242 new_result.update(force_properties)
1243
1244 # Extracted info may not be a video result (i.e.
1245 # info.get('_type', 'video') != video) but rather an url or
1246 # url_transparent. In such cases outer metadata (from ie_result)
1247 # should be propagated to inner one (info). For this to happen
1248 # _type of info should be overridden with url_transparent. This
1249 # fixes issue from https://github.com/ytdl-org/youtube-dl/pull/11163.
1250 if new_result.get('_type') == 'url':
1251 new_result['_type'] = 'url_transparent'
1252
1253 return self.process_ie_result(
1254 new_result, download=download, extra_info=extra_info)
1255 elif result_type in ('playlist', 'multi_video'):
1256 # Protect from infinite recursion due to recursively nested playlists
1257 # (see https://github.com/ytdl-org/youtube-dl/issues/27833)
1258 webpage_url = ie_result['webpage_url']
1259 if webpage_url in self._playlist_urls:
1260 self.to_screen(
1261 '[download] Skipping already downloaded playlist: %s'
1262 % ie_result.get('title') or ie_result.get('id'))
1263 return
1264
1265 self._playlist_level += 1
1266 self._playlist_urls.add(webpage_url)
1267 self._sanitize_thumbnails(ie_result)
1268 try:
1269 return self.__process_playlist(ie_result, download)
1270 finally:
1271 self._playlist_level -= 1
1272 if not self._playlist_level:
1273 self._playlist_urls.clear()
1274 elif result_type == 'compat_list':
1275 self.report_warning(
1276 'Extractor %s returned a compat_list result. '
1277 'It needs to be updated.' % ie_result.get('extractor'))
1278
1279 def _fixup(r):
1280 self.add_extra_info(
1281 r,
1282 {
1283 'extractor': ie_result['extractor'],
1284 'webpage_url': ie_result['webpage_url'],
1285 'webpage_url_basename': url_basename(ie_result['webpage_url']),
1286 'extractor_key': ie_result['extractor_key'],
1287 }
1288 )
1289 return r
1290 ie_result['entries'] = [
1291 self.process_ie_result(_fixup(r), download, extra_info)
1292 for r in ie_result['entries']
1293 ]
1294 return ie_result
1295 else:
1296 raise Exception('Invalid result type: %s' % result_type)
1297
1298 def _ensure_dir_exists(self, path):
1299 return make_dir(path, self.report_error)
1300
1301 def __process_playlist(self, ie_result, download):
1302 # We process each entry in the playlist
1303 playlist = ie_result.get('title') or ie_result.get('id')
1304 self.to_screen('[download] Downloading playlist: %s' % playlist)
1305
1306 if 'entries' not in ie_result:
1307 raise EntryNotInPlaylist()
1308 incomplete_entries = bool(ie_result.get('requested_entries'))
1309 if incomplete_entries:
1310 def fill_missing_entries(entries, indexes):
1311 ret = [None] * max(*indexes)
1312 for i, entry in zip(indexes, entries):
1313 ret[i - 1] = entry
1314 return ret
1315 ie_result['entries'] = fill_missing_entries(ie_result['entries'], ie_result['requested_entries'])
1316
1317 playlist_results = []
1318
1319 playliststart = self.params.get('playliststart', 1)
1320 playlistend = self.params.get('playlistend')
1321 # For backwards compatibility, interpret -1 as whole list
1322 if playlistend == -1:
1323 playlistend = None
1324
1325 playlistitems_str = self.params.get('playlist_items')
1326 playlistitems = None
1327 if playlistitems_str is not None:
1328 def iter_playlistitems(format):
1329 for string_segment in format.split(','):
1330 if '-' in string_segment:
1331 start, end = string_segment.split('-')
1332 for item in range(int(start), int(end) + 1):
1333 yield int(item)
1334 else:
1335 yield int(string_segment)
1336 playlistitems = orderedSet(iter_playlistitems(playlistitems_str))
1337
1338 ie_entries = ie_result['entries']
1339 msg = (
1340 'Downloading %d videos' if not isinstance(ie_entries, list)
1341 else 'Collected %d videos; downloading %%d of them' % len(ie_entries))
1342 if not isinstance(ie_entries, (list, PagedList)):
1343 ie_entries = LazyList(ie_entries)
1344
1345 entries = []
1346 for i in playlistitems or itertools.count(playliststart):
1347 if playlistitems is None and playlistend is not None and playlistend < i:
1348 break
1349 entry = None
1350 try:
1351 entry = ie_entries[i - 1]
1352 if entry is None:
1353 raise EntryNotInPlaylist()
1354 except (IndexError, EntryNotInPlaylist):
1355 if incomplete_entries:
1356 raise EntryNotInPlaylist()
1357 elif not playlistitems:
1358 break
1359 entries.append(entry)
1360 try:
1361 if entry is not None:
1362 self._match_entry(entry, incomplete=True, silent=True)
1363 except (ExistingVideoReached, RejectedVideoReached):
1364 break
1365 ie_result['entries'] = entries
1366
1367 # Save playlist_index before re-ordering
1368 entries = [
1369 ((playlistitems[i - 1] if playlistitems else i), entry)
1370 for i, entry in enumerate(entries, 1)
1371 if entry is not None]
1372 n_entries = len(entries)
1373
1374 if not playlistitems and (playliststart or playlistend):
1375 playlistitems = list(range(playliststart, playliststart + n_entries))
1376 ie_result['requested_entries'] = playlistitems
1377
1378 if self.params.get('allow_playlist_files', True):
1379 ie_copy = {
1380 'playlist': playlist,
1381 'playlist_id': ie_result.get('id'),
1382 'playlist_title': ie_result.get('title'),
1383 'playlist_uploader': ie_result.get('uploader'),
1384 'playlist_uploader_id': ie_result.get('uploader_id'),
1385 'playlist_index': 0,
1386 }
1387 ie_copy.update(dict(ie_result))
1388
1389 if self.params.get('writeinfojson', False):
1390 infofn = self.prepare_filename(ie_copy, 'pl_infojson')
1391 if not self._ensure_dir_exists(encodeFilename(infofn)):
1392 return
1393 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(infofn)):
1394 self.to_screen('[info] Playlist metadata is already present')
1395 else:
1396 self.to_screen('[info] Writing playlist metadata as JSON to: ' + infofn)
1397 try:
1398 write_json_file(self.filter_requested_info(ie_result, self.params.get('clean_infojson', True)), infofn)
1399 except (OSError, IOError):
1400 self.report_error('Cannot write playlist metadata to JSON file ' + infofn)
1401
1402 # TODO: This should be passed to ThumbnailsConvertor if necessary
1403 self._write_thumbnails(ie_copy, self.prepare_filename(ie_copy, 'pl_thumbnail'))
1404
1405 if self.params.get('writedescription', False):
1406 descfn = self.prepare_filename(ie_copy, 'pl_description')
1407 if not self._ensure_dir_exists(encodeFilename(descfn)):
1408 return
1409 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(descfn)):
1410 self.to_screen('[info] Playlist description is already present')
1411 elif ie_result.get('description') is None:
1412 self.report_warning('There\'s no playlist description to write.')
1413 else:
1414 try:
1415 self.to_screen('[info] Writing playlist description to: ' + descfn)
1416 with io.open(encodeFilename(descfn), 'w', encoding='utf-8') as descfile:
1417 descfile.write(ie_result['description'])
1418 except (OSError, IOError):
1419 self.report_error('Cannot write playlist description file ' + descfn)
1420 return
1421
1422 if self.params.get('playlistreverse', False):
1423 entries = entries[::-1]
1424 if self.params.get('playlistrandom', False):
1425 random.shuffle(entries)
1426
1427 x_forwarded_for = ie_result.get('__x_forwarded_for_ip')
1428
1429 self.to_screen('[%s] playlist %s: %s' % (ie_result['extractor'], playlist, msg % n_entries))
1430 failures = 0
1431 max_failures = self.params.get('skip_playlist_after_errors') or float('inf')
1432 for i, entry_tuple in enumerate(entries, 1):
1433 playlist_index, entry = entry_tuple
1434 if 'playlist_index' in self.params.get('compat_options', []):
1435 playlist_index = playlistitems[i - 1] if playlistitems else i
1436 self.to_screen('[download] Downloading video %s of %s' % (i, n_entries))
1437 # This __x_forwarded_for_ip thing is a bit ugly but requires
1438 # minimal changes
1439 if x_forwarded_for:
1440 entry['__x_forwarded_for_ip'] = x_forwarded_for
1441 extra = {
1442 'n_entries': n_entries,
1443 '_last_playlist_index': max(playlistitems) if playlistitems else (playlistend or n_entries),
1444 'playlist_index': playlist_index,
1445 'playlist_autonumber': i,
1446 'playlist': playlist,
1447 'playlist_id': ie_result.get('id'),
1448 'playlist_title': ie_result.get('title'),
1449 'playlist_uploader': ie_result.get('uploader'),
1450 'playlist_uploader_id': ie_result.get('uploader_id'),
1451 'extractor': ie_result['extractor'],
1452 'webpage_url': ie_result['webpage_url'],
1453 'webpage_url_basename': url_basename(ie_result['webpage_url']),
1454 'extractor_key': ie_result['extractor_key'],
1455 }
1456
1457 if self._match_entry(entry, incomplete=True) is not None:
1458 continue
1459
1460 entry_result = self.__process_iterable_entry(entry, download, extra)
1461 if not entry_result:
1462 failures += 1
1463 if failures >= max_failures:
1464 self.report_error(
1465 'Skipping the remaining entries in playlist "%s" since %d items failed extraction' % (playlist, failures))
1466 break
1467 # TODO: skip failed (empty) entries?
1468 playlist_results.append(entry_result)
1469 ie_result['entries'] = playlist_results
1470 self.to_screen('[download] Finished downloading playlist: %s' % playlist)
1471 return ie_result
1472
1473 @__handle_extraction_exceptions
1474 def __process_iterable_entry(self, entry, download, extra_info):
1475 return self.process_ie_result(
1476 entry, download=download, extra_info=extra_info)
1477
1478 def _build_format_filter(self, filter_spec):
1479 " Returns a function to filter the formats according to the filter_spec "
1480
1481 OPERATORS = {
1482 '<': operator.lt,
1483 '<=': operator.le,
1484 '>': operator.gt,
1485 '>=': operator.ge,
1486 '=': operator.eq,
1487 '!=': operator.ne,
1488 }
1489 operator_rex = re.compile(r'''(?x)\s*
1490 (?P<key>width|height|tbr|abr|vbr|asr|filesize|filesize_approx|fps)
1491 \s*(?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
1492 (?P<value>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?)
1493 $
1494 ''' % '|'.join(map(re.escape, OPERATORS.keys())))
1495 m = operator_rex.search(filter_spec)
1496 if m:
1497 try:
1498 comparison_value = int(m.group('value'))
1499 except ValueError:
1500 comparison_value = parse_filesize(m.group('value'))
1501 if comparison_value is None:
1502 comparison_value = parse_filesize(m.group('value') + 'B')
1503 if comparison_value is None:
1504 raise ValueError(
1505 'Invalid value %r in format specification %r' % (
1506 m.group('value'), filter_spec))
1507 op = OPERATORS[m.group('op')]
1508
1509 if not m:
1510 STR_OPERATORS = {
1511 '=': operator.eq,
1512 '^=': lambda attr, value: attr.startswith(value),
1513 '$=': lambda attr, value: attr.endswith(value),
1514 '*=': lambda attr, value: value in attr,
1515 }
1516 str_operator_rex = re.compile(r'''(?x)
1517 \s*(?P<key>[a-zA-Z0-9._-]+)
1518 \s*(?P<negation>!\s*)?(?P<op>%s)(?P<none_inclusive>\s*\?)?
1519 \s*(?P<value>[a-zA-Z0-9._-]+)
1520 \s*$
1521 ''' % '|'.join(map(re.escape, STR_OPERATORS.keys())))
1522 m = str_operator_rex.search(filter_spec)
1523 if m:
1524 comparison_value = m.group('value')
1525 str_op = STR_OPERATORS[m.group('op')]
1526 if m.group('negation'):
1527 op = lambda attr, value: not str_op(attr, value)
1528 else:
1529 op = str_op
1530
1531 if not m:
1532 raise ValueError('Invalid filter specification %r' % filter_spec)
1533
1534 def _filter(f):
1535 actual_value = f.get(m.group('key'))
1536 if actual_value is None:
1537 return m.group('none_inclusive')
1538 return op(actual_value, comparison_value)
1539 return _filter
1540
1541 def _default_format_spec(self, info_dict, download=True):
1542
1543 def can_merge():
1544 merger = FFmpegMergerPP(self)
1545 return merger.available and merger.can_merge()
1546
1547 prefer_best = (
1548 not self.params.get('simulate', False)
1549 and download
1550 and (
1551 not can_merge()
1552 or info_dict.get('is_live', False)
1553 or self.outtmpl_dict['default'] == '-'))
1554 compat = (
1555 prefer_best
1556 or self.params.get('allow_multiple_audio_streams', False)
1557 or 'format-spec' in self.params.get('compat_opts', []))
1558
1559 return (
1560 'best/bestvideo+bestaudio' if prefer_best
1561 else 'bestvideo*+bestaudio/best' if not compat
1562 else 'bestvideo+bestaudio/best')
1563
1564 def build_format_selector(self, format_spec):
1565 def syntax_error(note, start):
1566 message = (
1567 'Invalid format specification: '
1568 '{0}\n\t{1}\n\t{2}^'.format(note, format_spec, ' ' * start[1]))
1569 return SyntaxError(message)
1570
1571 PICKFIRST = 'PICKFIRST'
1572 MERGE = 'MERGE'
1573 SINGLE = 'SINGLE'
1574 GROUP = 'GROUP'
1575 FormatSelector = collections.namedtuple('FormatSelector', ['type', 'selector', 'filters'])
1576
1577 allow_multiple_streams = {'audio': self.params.get('allow_multiple_audio_streams', False),
1578 'video': self.params.get('allow_multiple_video_streams', False)}
1579
1580 check_formats = self.params.get('check_formats')
1581
1582 def _parse_filter(tokens):
1583 filter_parts = []
1584 for type, string, start, _, _ in tokens:
1585 if type == tokenize.OP and string == ']':
1586 return ''.join(filter_parts)
1587 else:
1588 filter_parts.append(string)
1589
1590 def _remove_unused_ops(tokens):
1591 # Remove operators that we don't use and join them with the surrounding strings
1592 # for example: 'mp4' '-' 'baseline' '-' '16x9' is converted to 'mp4-baseline-16x9'
1593 ALLOWED_OPS = ('/', '+', ',', '(', ')')
1594 last_string, last_start, last_end, last_line = None, None, None, None
1595 for type, string, start, end, line in tokens:
1596 if type == tokenize.OP and string == '[':
1597 if last_string:
1598 yield tokenize.NAME, last_string, last_start, last_end, last_line
1599 last_string = None
1600 yield type, string, start, end, line
1601 # everything inside brackets will be handled by _parse_filter
1602 for type, string, start, end, line in tokens:
1603 yield type, string, start, end, line
1604 if type == tokenize.OP and string == ']':
1605 break
1606 elif type == tokenize.OP and string in ALLOWED_OPS:
1607 if last_string:
1608 yield tokenize.NAME, last_string, last_start, last_end, last_line
1609 last_string = None
1610 yield type, string, start, end, line
1611 elif type in [tokenize.NAME, tokenize.NUMBER, tokenize.OP]:
1612 if not last_string:
1613 last_string = string
1614 last_start = start
1615 last_end = end
1616 else:
1617 last_string += string
1618 if last_string:
1619 yield tokenize.NAME, last_string, last_start, last_end, last_line
1620
1621 def _parse_format_selection(tokens, inside_merge=False, inside_choice=False, inside_group=False):
1622 selectors = []
1623 current_selector = None
1624 for type, string, start, _, _ in tokens:
1625 # ENCODING is only defined in python 3.x
1626 if type == getattr(tokenize, 'ENCODING', None):
1627 continue
1628 elif type in [tokenize.NAME, tokenize.NUMBER]:
1629 current_selector = FormatSelector(SINGLE, string, [])
1630 elif type == tokenize.OP:
1631 if string == ')':
1632 if not inside_group:
1633 # ')' will be handled by the parentheses group
1634 tokens.restore_last_token()
1635 break
1636 elif inside_merge and string in ['/', ',']:
1637 tokens.restore_last_token()
1638 break
1639 elif inside_choice and string == ',':
1640 tokens.restore_last_token()
1641 break
1642 elif string == ',':
1643 if not current_selector:
1644 raise syntax_error('"," must follow a format selector', start)
1645 selectors.append(current_selector)
1646 current_selector = None
1647 elif string == '/':
1648 if not current_selector:
1649 raise syntax_error('"/" must follow a format selector', start)
1650 first_choice = current_selector
1651 second_choice = _parse_format_selection(tokens, inside_choice=True)
1652 current_selector = FormatSelector(PICKFIRST, (first_choice, second_choice), [])
1653 elif string == '[':
1654 if not current_selector:
1655 current_selector = FormatSelector(SINGLE, 'best', [])
1656 format_filter = _parse_filter(tokens)
1657 current_selector.filters.append(format_filter)
1658 elif string == '(':
1659 if current_selector:
1660 raise syntax_error('Unexpected "("', start)
1661 group = _parse_format_selection(tokens, inside_group=True)
1662 current_selector = FormatSelector(GROUP, group, [])
1663 elif string == '+':
1664 if not current_selector:
1665 raise syntax_error('Unexpected "+"', start)
1666 selector_1 = current_selector
1667 selector_2 = _parse_format_selection(tokens, inside_merge=True)
1668 if not selector_2:
1669 raise syntax_error('Expected a selector', start)
1670 current_selector = FormatSelector(MERGE, (selector_1, selector_2), [])
1671 else:
1672 raise syntax_error('Operator not recognized: "{0}"'.format(string), start)
1673 elif type == tokenize.ENDMARKER:
1674 break
1675 if current_selector:
1676 selectors.append(current_selector)
1677 return selectors
1678
1679 def _merge(formats_pair):
1680 format_1, format_2 = formats_pair
1681
1682 formats_info = []
1683 formats_info.extend(format_1.get('requested_formats', (format_1,)))
1684 formats_info.extend(format_2.get('requested_formats', (format_2,)))
1685
1686 if not allow_multiple_streams['video'] or not allow_multiple_streams['audio']:
1687 get_no_more = {"video": False, "audio": False}
1688 for (i, fmt_info) in enumerate(formats_info):
1689 for aud_vid in ["audio", "video"]:
1690 if not allow_multiple_streams[aud_vid] and fmt_info.get(aud_vid[0] + 'codec') != 'none':
1691 if get_no_more[aud_vid]:
1692 formats_info.pop(i)
1693 get_no_more[aud_vid] = True
1694
1695 if len(formats_info) == 1:
1696 return formats_info[0]
1697
1698 video_fmts = [fmt_info for fmt_info in formats_info if fmt_info.get('vcodec') != 'none']
1699 audio_fmts = [fmt_info for fmt_info in formats_info if fmt_info.get('acodec') != 'none']
1700
1701 the_only_video = video_fmts[0] if len(video_fmts) == 1 else None
1702 the_only_audio = audio_fmts[0] if len(audio_fmts) == 1 else None
1703
1704 output_ext = self.params.get('merge_output_format')
1705 if not output_ext:
1706 if the_only_video:
1707 output_ext = the_only_video['ext']
1708 elif the_only_audio and not video_fmts:
1709 output_ext = the_only_audio['ext']
1710 else:
1711 output_ext = 'mkv'
1712
1713 new_dict = {
1714 'requested_formats': formats_info,
1715 'format': '+'.join(fmt_info.get('format') for fmt_info in formats_info),
1716 'format_id': '+'.join(fmt_info.get('format_id') for fmt_info in formats_info),
1717 'ext': output_ext,
1718 }
1719
1720 if the_only_video:
1721 new_dict.update({
1722 'width': the_only_video.get('width'),
1723 'height': the_only_video.get('height'),
1724 'resolution': the_only_video.get('resolution') or self.format_resolution(the_only_video),
1725 'fps': the_only_video.get('fps'),
1726 'vcodec': the_only_video.get('vcodec'),
1727 'vbr': the_only_video.get('vbr'),
1728 'stretched_ratio': the_only_video.get('stretched_ratio'),
1729 })
1730
1731 if the_only_audio:
1732 new_dict.update({
1733 'acodec': the_only_audio.get('acodec'),
1734 'abr': the_only_audio.get('abr'),
1735 })
1736
1737 return new_dict
1738
1739 def _check_formats(formats):
1740 for f in formats:
1741 self.to_screen('[info] Testing format %s' % f['format_id'])
1742 paths = self.params.get('paths', {})
1743 temp_file = os.path.join(
1744 expand_path(paths.get('home', '').strip()),
1745 expand_path(paths.get('temp', '').strip()),
1746 'ytdl.%s.f%s.check-format' % (random_uuidv4(), f['format_id']))
1747 try:
1748 dl, _ = self.dl(temp_file, f, test=True)
1749 except (ExtractorError, IOError, OSError, ValueError) + network_exceptions:
1750 dl = False
1751 finally:
1752 if os.path.exists(temp_file):
1753 os.remove(temp_file)
1754 if dl:
1755 yield f
1756 else:
1757 self.to_screen('[info] Unable to download format %s. Skipping...' % f['format_id'])
1758
1759 def _build_selector_function(selector):
1760 if isinstance(selector, list): # ,
1761 fs = [_build_selector_function(s) for s in selector]
1762
1763 def selector_function(ctx):
1764 for f in fs:
1765 for format in f(ctx):
1766 yield format
1767 return selector_function
1768
1769 elif selector.type == GROUP: # ()
1770 selector_function = _build_selector_function(selector.selector)
1771
1772 elif selector.type == PICKFIRST: # /
1773 fs = [_build_selector_function(s) for s in selector.selector]
1774
1775 def selector_function(ctx):
1776 for f in fs:
1777 picked_formats = list(f(ctx))
1778 if picked_formats:
1779 return picked_formats
1780 return []
1781
1782 elif selector.type == SINGLE: # atom
1783 format_spec = selector.selector or 'best'
1784
1785 # TODO: Add allvideo, allaudio etc by generalizing the code with best/worst selector
1786 if format_spec == 'all':
1787 def selector_function(ctx):
1788 formats = list(ctx['formats'])
1789 if check_formats:
1790 formats = _check_formats(formats)
1791 for f in formats:
1792 yield f
1793 elif format_spec == 'mergeall':
1794 def selector_function(ctx):
1795 formats = list(_check_formats(ctx['formats']))
1796 if not formats:
1797 return
1798 merged_format = formats[-1]
1799 for f in formats[-2::-1]:
1800 merged_format = _merge((merged_format, f))
1801 yield merged_format
1802
1803 else:
1804 format_fallback, format_reverse, format_idx = False, True, 1
1805 mobj = re.match(
1806 r'(?P<bw>best|worst|b|w)(?P<type>video|audio|v|a)?(?P<mod>\*)?(?:\.(?P<n>[1-9]\d*))?$',
1807 format_spec)
1808 if mobj is not None:
1809 format_idx = int_or_none(mobj.group('n'), default=1)
1810 format_reverse = mobj.group('bw')[0] == 'b'
1811 format_type = (mobj.group('type') or [None])[0]
1812 not_format_type = {'v': 'a', 'a': 'v'}.get(format_type)
1813 format_modified = mobj.group('mod') is not None
1814
1815 format_fallback = not format_type and not format_modified # for b, w
1816 filter_f = (
1817 (lambda f: f.get('%scodec' % format_type) != 'none')
1818 if format_type and format_modified # bv*, ba*, wv*, wa*
1819 else (lambda f: f.get('%scodec' % not_format_type) == 'none')
1820 if format_type # bv, ba, wv, wa
1821 else (lambda f: f.get('vcodec') != 'none' and f.get('acodec') != 'none')
1822 if not format_modified # b, w
1823 else None) # b*, w*
1824 else:
1825 filter_f = ((lambda f: f.get('ext') == format_spec)
1826 if format_spec in ['mp4', 'flv', 'webm', '3gp', 'm4a', 'mp3', 'ogg', 'aac', 'wav'] # extension
1827 else (lambda f: f.get('format_id') == format_spec)) # id
1828
1829 def selector_function(ctx):
1830 formats = list(ctx['formats'])
1831 if not formats:
1832 return
1833 matches = list(filter(filter_f, formats)) if filter_f is not None else formats
1834 if format_fallback and ctx['incomplete_formats'] and not matches:
1835 # for extractors with incomplete formats (audio only (soundcloud)
1836 # or video only (imgur)) best/worst will fallback to
1837 # best/worst {video,audio}-only format
1838 matches = formats
1839 if format_reverse:
1840 matches = matches[::-1]
1841 if check_formats:
1842 matches = list(itertools.islice(_check_formats(matches), format_idx))
1843 n = len(matches)
1844 if -n <= format_idx - 1 < n:
1845 yield matches[format_idx - 1]
1846
1847 elif selector.type == MERGE: # +
1848 selector_1, selector_2 = map(_build_selector_function, selector.selector)
1849
1850 def selector_function(ctx):
1851 for pair in itertools.product(
1852 selector_1(copy.deepcopy(ctx)), selector_2(copy.deepcopy(ctx))):
1853 yield _merge(pair)
1854
1855 filters = [self._build_format_filter(f) for f in selector.filters]
1856
1857 def final_selector(ctx):
1858 ctx_copy = copy.deepcopy(ctx)
1859 for _filter in filters:
1860 ctx_copy['formats'] = list(filter(_filter, ctx_copy['formats']))
1861 return selector_function(ctx_copy)
1862 return final_selector
1863
1864 stream = io.BytesIO(format_spec.encode('utf-8'))
1865 try:
1866 tokens = list(_remove_unused_ops(compat_tokenize_tokenize(stream.readline)))
1867 except tokenize.TokenError:
1868 raise syntax_error('Missing closing/opening brackets or parenthesis', (0, len(format_spec)))
1869
1870 class TokenIterator(object):
1871 def __init__(self, tokens):
1872 self.tokens = tokens
1873 self.counter = 0
1874
1875 def __iter__(self):
1876 return self
1877
1878 def __next__(self):
1879 if self.counter >= len(self.tokens):
1880 raise StopIteration()
1881 value = self.tokens[self.counter]
1882 self.counter += 1
1883 return value
1884
1885 next = __next__
1886
1887 def restore_last_token(self):
1888 self.counter -= 1
1889
1890 parsed_selector = _parse_format_selection(iter(TokenIterator(tokens)))
1891 return _build_selector_function(parsed_selector)
1892
1893 def _calc_headers(self, info_dict):
1894 res = std_headers.copy()
1895
1896 add_headers = info_dict.get('http_headers')
1897 if add_headers:
1898 res.update(add_headers)
1899
1900 cookies = self._calc_cookies(info_dict)
1901 if cookies:
1902 res['Cookie'] = cookies
1903
1904 if 'X-Forwarded-For' not in res:
1905 x_forwarded_for_ip = info_dict.get('__x_forwarded_for_ip')
1906 if x_forwarded_for_ip:
1907 res['X-Forwarded-For'] = x_forwarded_for_ip
1908
1909 return res
1910
1911 def _calc_cookies(self, info_dict):
1912 pr = sanitized_Request(info_dict['url'])
1913 self.cookiejar.add_cookie_header(pr)
1914 return pr.get_header('Cookie')
1915
1916 @staticmethod
1917 def _sanitize_thumbnails(info_dict):
1918 thumbnails = info_dict.get('thumbnails')
1919 if thumbnails is None:
1920 thumbnail = info_dict.get('thumbnail')
1921 if thumbnail:
1922 info_dict['thumbnails'] = thumbnails = [{'url': thumbnail}]
1923 if thumbnails:
1924 thumbnails.sort(key=lambda t: (
1925 t.get('preference') if t.get('preference') is not None else -1,
1926 t.get('width') if t.get('width') is not None else -1,
1927 t.get('height') if t.get('height') is not None else -1,
1928 t.get('id') if t.get('id') is not None else '',
1929 t.get('url')))
1930 for i, t in enumerate(thumbnails):
1931 t['url'] = sanitize_url(t['url'])
1932 if t.get('width') and t.get('height'):
1933 t['resolution'] = '%dx%d' % (t['width'], t['height'])
1934 if t.get('id') is None:
1935 t['id'] = '%d' % i
1936
1937 def process_video_result(self, info_dict, download=True):
1938 assert info_dict.get('_type', 'video') == 'video'
1939
1940 if 'id' not in info_dict:
1941 raise ExtractorError('Missing "id" field in extractor result')
1942 if 'title' not in info_dict:
1943 raise ExtractorError('Missing "title" field in extractor result')
1944
1945 def report_force_conversion(field, field_not, conversion):
1946 self.report_warning(
1947 '"%s" field is not %s - forcing %s conversion, there is an error in extractor'
1948 % (field, field_not, conversion))
1949
1950 def sanitize_string_field(info, string_field):
1951 field = info.get(string_field)
1952 if field is None or isinstance(field, compat_str):
1953 return
1954 report_force_conversion(string_field, 'a string', 'string')
1955 info[string_field] = compat_str(field)
1956
1957 def sanitize_numeric_fields(info):
1958 for numeric_field in self._NUMERIC_FIELDS:
1959 field = info.get(numeric_field)
1960 if field is None or isinstance(field, compat_numeric_types):
1961 continue
1962 report_force_conversion(numeric_field, 'numeric', 'int')
1963 info[numeric_field] = int_or_none(field)
1964
1965 sanitize_string_field(info_dict, 'id')
1966 sanitize_numeric_fields(info_dict)
1967
1968 if 'playlist' not in info_dict:
1969 # It isn't part of a playlist
1970 info_dict['playlist'] = None
1971 info_dict['playlist_index'] = None
1972
1973 self._sanitize_thumbnails(info_dict)
1974
1975 if self.params.get('list_thumbnails'):
1976 self.list_thumbnails(info_dict)
1977 return
1978
1979 thumbnail = info_dict.get('thumbnail')
1980 thumbnails = info_dict.get('thumbnails')
1981 if thumbnail:
1982 info_dict['thumbnail'] = sanitize_url(thumbnail)
1983 elif thumbnails:
1984 info_dict['thumbnail'] = thumbnails[-1]['url']
1985
1986 if 'display_id' not in info_dict and 'id' in info_dict:
1987 info_dict['display_id'] = info_dict['id']
1988
1989 for ts_key, date_key in (
1990 ('timestamp', 'upload_date'),
1991 ('release_timestamp', 'release_date'),
1992 ):
1993 if info_dict.get(date_key) is None and info_dict.get(ts_key) is not None:
1994 # Working around out-of-range timestamp values (e.g. negative ones on Windows,
1995 # see http://bugs.python.org/issue1646728)
1996 try:
1997 upload_date = datetime.datetime.utcfromtimestamp(info_dict[ts_key])
1998 info_dict[date_key] = upload_date.strftime('%Y%m%d')
1999 except (ValueError, OverflowError, OSError):
2000 pass
2001
2002 # Auto generate title fields corresponding to the *_number fields when missing
2003 # in order to always have clean titles. This is very common for TV series.
2004 for field in ('chapter', 'season', 'episode'):
2005 if info_dict.get('%s_number' % field) is not None and not info_dict.get(field):
2006 info_dict[field] = '%s %d' % (field.capitalize(), info_dict['%s_number' % field])
2007
2008 for cc_kind in ('subtitles', 'automatic_captions'):
2009 cc = info_dict.get(cc_kind)
2010 if cc:
2011 for _, subtitle in cc.items():
2012 for subtitle_format in subtitle:
2013 if subtitle_format.get('url'):
2014 subtitle_format['url'] = sanitize_url(subtitle_format['url'])
2015 if subtitle_format.get('ext') is None:
2016 subtitle_format['ext'] = determine_ext(subtitle_format['url']).lower()
2017
2018 automatic_captions = info_dict.get('automatic_captions')
2019 subtitles = info_dict.get('subtitles')
2020
2021 if self.params.get('listsubtitles', False):
2022 if 'automatic_captions' in info_dict:
2023 self.list_subtitles(
2024 info_dict['id'], automatic_captions, 'automatic captions')
2025 self.list_subtitles(info_dict['id'], subtitles, 'subtitles')
2026 return
2027
2028 info_dict['requested_subtitles'] = self.process_subtitles(
2029 info_dict['id'], subtitles, automatic_captions)
2030
2031 # We now pick which formats have to be downloaded
2032 if info_dict.get('formats') is None:
2033 # There's only one format available
2034 formats = [info_dict]
2035 else:
2036 formats = info_dict['formats']
2037
2038 if not formats:
2039 if not self.params.get('ignore_no_formats_error'):
2040 raise ExtractorError('No video formats found!')
2041 else:
2042 self.report_warning('No video formats found!')
2043
2044 def is_wellformed(f):
2045 url = f.get('url')
2046 if not url:
2047 self.report_warning(
2048 '"url" field is missing or empty - skipping format, '
2049 'there is an error in extractor')
2050 return False
2051 if isinstance(url, bytes):
2052 sanitize_string_field(f, 'url')
2053 return True
2054
2055 # Filter out malformed formats for better extraction robustness
2056 formats = list(filter(is_wellformed, formats))
2057
2058 formats_dict = {}
2059
2060 # We check that all the formats have the format and format_id fields
2061 for i, format in enumerate(formats):
2062 sanitize_string_field(format, 'format_id')
2063 sanitize_numeric_fields(format)
2064 format['url'] = sanitize_url(format['url'])
2065 if not format.get('format_id'):
2066 format['format_id'] = compat_str(i)
2067 else:
2068 # Sanitize format_id from characters used in format selector expression
2069 format['format_id'] = re.sub(r'[\s,/+\[\]()]', '_', format['format_id'])
2070 format_id = format['format_id']
2071 if format_id not in formats_dict:
2072 formats_dict[format_id] = []
2073 formats_dict[format_id].append(format)
2074
2075 # Make sure all formats have unique format_id
2076 for format_id, ambiguous_formats in formats_dict.items():
2077 if len(ambiguous_formats) > 1:
2078 for i, format in enumerate(ambiguous_formats):
2079 format['format_id'] = '%s-%d' % (format_id, i)
2080
2081 for i, format in enumerate(formats):
2082 if format.get('format') is None:
2083 format['format'] = '{id} - {res}{note}'.format(
2084 id=format['format_id'],
2085 res=self.format_resolution(format),
2086 note=' ({0})'.format(format['format_note']) if format.get('format_note') is not None else '',
2087 )
2088 # Automatically determine file extension if missing
2089 if format.get('ext') is None:
2090 format['ext'] = determine_ext(format['url']).lower()
2091 # Automatically determine protocol if missing (useful for format
2092 # selection purposes)
2093 if format.get('protocol') is None:
2094 format['protocol'] = determine_protocol(format)
2095 # Add HTTP headers, so that external programs can use them from the
2096 # json output
2097 full_format_info = info_dict.copy()
2098 full_format_info.update(format)
2099 format['http_headers'] = self._calc_headers(full_format_info)
2100 # Remove private housekeeping stuff
2101 if '__x_forwarded_for_ip' in info_dict:
2102 del info_dict['__x_forwarded_for_ip']
2103
2104 # TODO Central sorting goes here
2105
2106 if formats and formats[0] is not info_dict:
2107 # only set the 'formats' fields if the original info_dict list them
2108 # otherwise we end up with a circular reference, the first (and unique)
2109 # element in the 'formats' field in info_dict is info_dict itself,
2110 # which can't be exported to json
2111 info_dict['formats'] = formats
2112
2113 info_dict, _ = self.pre_process(info_dict)
2114
2115 if self.params.get('listformats'):
2116 if not info_dict.get('formats'):
2117 raise ExtractorError('No video formats found', expected=True)
2118 self.list_formats(info_dict)
2119 return
2120
2121 req_format = self.params.get('format')
2122 if req_format is None:
2123 req_format = self._default_format_spec(info_dict, download=download)
2124 self.write_debug('Default format spec: %s' % req_format)
2125
2126 format_selector = self.build_format_selector(req_format)
2127
2128 # While in format selection we may need to have an access to the original
2129 # format set in order to calculate some metrics or do some processing.
2130 # For now we need to be able to guess whether original formats provided
2131 # by extractor are incomplete or not (i.e. whether extractor provides only
2132 # video-only or audio-only formats) for proper formats selection for
2133 # extractors with such incomplete formats (see
2134 # https://github.com/ytdl-org/youtube-dl/pull/5556).
2135 # Since formats may be filtered during format selection and may not match
2136 # the original formats the results may be incorrect. Thus original formats
2137 # or pre-calculated metrics should be passed to format selection routines
2138 # as well.
2139 # We will pass a context object containing all necessary additional data
2140 # instead of just formats.
2141 # This fixes incorrect format selection issue (see
2142 # https://github.com/ytdl-org/youtube-dl/issues/10083).
2143 incomplete_formats = (
2144 # All formats are video-only or
2145 all(f.get('vcodec') != 'none' and f.get('acodec') == 'none' for f in formats)
2146 # all formats are audio-only
2147 or all(f.get('vcodec') == 'none' and f.get('acodec') != 'none' for f in formats))
2148
2149 ctx = {
2150 'formats': formats,
2151 'incomplete_formats': incomplete_formats,
2152 }
2153
2154 formats_to_download = list(format_selector(ctx))
2155 if not formats_to_download:
2156 if not self.params.get('ignore_no_formats_error'):
2157 raise ExtractorError('Requested format is not available', expected=True)
2158 else:
2159 self.report_warning('Requested format is not available')
2160 elif download:
2161 self.to_screen(
2162 '[info] %s: Downloading %d format(s): %s' % (
2163 info_dict['id'], len(formats_to_download),
2164 ", ".join([f['format_id'] for f in formats_to_download])))
2165 for fmt in formats_to_download:
2166 new_info = dict(info_dict)
2167 # Save a reference to the original info_dict so that it can be modified in process_info if needed
2168 new_info['__original_infodict'] = info_dict
2169 new_info.update(fmt)
2170 self.process_info(new_info)
2171 # We update the info dict with the best quality format (backwards compatibility)
2172 if formats_to_download:
2173 info_dict.update(formats_to_download[-1])
2174 return info_dict
2175
2176 def process_subtitles(self, video_id, normal_subtitles, automatic_captions):
2177 """Select the requested subtitles and their format"""
2178 available_subs = {}
2179 if normal_subtitles and self.params.get('writesubtitles'):
2180 available_subs.update(normal_subtitles)
2181 if automatic_captions and self.params.get('writeautomaticsub'):
2182 for lang, cap_info in automatic_captions.items():
2183 if lang not in available_subs:
2184 available_subs[lang] = cap_info
2185
2186 if (not self.params.get('writesubtitles') and not
2187 self.params.get('writeautomaticsub') or not
2188 available_subs):
2189 return None
2190
2191 all_sub_langs = available_subs.keys()
2192 if self.params.get('allsubtitles', False):
2193 requested_langs = all_sub_langs
2194 elif self.params.get('subtitleslangs', False):
2195 requested_langs = set()
2196 for lang in self.params.get('subtitleslangs'):
2197 if lang == 'all':
2198 requested_langs.update(all_sub_langs)
2199 continue
2200 discard = lang[0] == '-'
2201 if discard:
2202 lang = lang[1:]
2203 current_langs = filter(re.compile(lang + '$').match, all_sub_langs)
2204 if discard:
2205 for lang in current_langs:
2206 requested_langs.discard(lang)
2207 else:
2208 requested_langs.update(current_langs)
2209 elif 'en' in available_subs:
2210 requested_langs = ['en']
2211 else:
2212 requested_langs = [list(all_sub_langs)[0]]
2213 self.write_debug('Downloading subtitles: %s' % ', '.join(requested_langs))
2214
2215 formats_query = self.params.get('subtitlesformat', 'best')
2216 formats_preference = formats_query.split('/') if formats_query else []
2217 subs = {}
2218 for lang in requested_langs:
2219 formats = available_subs.get(lang)
2220 if formats is None:
2221 self.report_warning('%s subtitles not available for %s' % (lang, video_id))
2222 continue
2223 for ext in formats_preference:
2224 if ext == 'best':
2225 f = formats[-1]
2226 break
2227 matches = list(filter(lambda f: f['ext'] == ext, formats))
2228 if matches:
2229 f = matches[-1]
2230 break
2231 else:
2232 f = formats[-1]
2233 self.report_warning(
2234 'No subtitle format found matching "%s" for language %s, '
2235 'using %s' % (formats_query, lang, f['ext']))
2236 subs[lang] = f
2237 return subs
2238
2239 def __forced_printings(self, info_dict, filename, incomplete):
2240 def print_mandatory(field, actual_field=None):
2241 if actual_field is None:
2242 actual_field = field
2243 if (self.params.get('force%s' % field, False)
2244 and (not incomplete or info_dict.get(actual_field) is not None)):
2245 self.to_stdout(info_dict[actual_field])
2246
2247 def print_optional(field):
2248 if (self.params.get('force%s' % field, False)
2249 and info_dict.get(field) is not None):
2250 self.to_stdout(info_dict[field])
2251
2252 info_dict = info_dict.copy()
2253 if filename is not None:
2254 info_dict['filename'] = filename
2255 if info_dict.get('requested_formats') is not None:
2256 # For RTMP URLs, also include the playpath
2257 info_dict['urls'] = '\n'.join(f['url'] + f.get('play_path', '') for f in info_dict['requested_formats'])
2258 elif 'url' in info_dict:
2259 info_dict['urls'] = info_dict['url'] + info_dict.get('play_path', '')
2260
2261 for tmpl in self.params.get('forceprint', []):
2262 if re.match(r'\w+$', tmpl):
2263 tmpl = '%({})s'.format(tmpl)
2264 tmpl, info_copy = self.prepare_outtmpl(tmpl, info_dict)
2265 self.to_stdout(tmpl % info_copy)
2266
2267 print_mandatory('title')
2268 print_mandatory('id')
2269 print_mandatory('url', 'urls')
2270 print_optional('thumbnail')
2271 print_optional('description')
2272 print_optional('filename')
2273 if self.params.get('forceduration', False) and info_dict.get('duration') is not None:
2274 self.to_stdout(formatSeconds(info_dict['duration']))
2275 print_mandatory('format')
2276
2277 if self.params.get('forcejson', False):
2278 self.post_extract(info_dict)
2279 self.to_stdout(json.dumps(info_dict, default=repr))
2280
2281 def dl(self, name, info, subtitle=False, test=False):
2282
2283 if test:
2284 verbose = self.params.get('verbose')
2285 params = {
2286 'test': True,
2287 'quiet': not verbose,
2288 'verbose': verbose,
2289 'noprogress': not verbose,
2290 'nopart': True,
2291 'skip_unavailable_fragments': False,
2292 'keep_fragments': False,
2293 'overwrites': True,
2294 '_no_ytdl_file': True,
2295 }
2296 else:
2297 params = self.params
2298 fd = get_suitable_downloader(info, params)(self, params)
2299 if not test:
2300 for ph in self._progress_hooks:
2301 fd.add_progress_hook(ph)
2302 urls = '", "'.join([f['url'] for f in info.get('requested_formats', [])] or [info['url']])
2303 self.write_debug('Invoking downloader on "%s"' % urls)
2304 new_info = dict(info)
2305 if new_info.get('http_headers') is None:
2306 new_info['http_headers'] = self._calc_headers(new_info)
2307 return fd.download(name, new_info, subtitle)
2308
2309 def process_info(self, info_dict):
2310 """Process a single resolved IE result."""
2311
2312 assert info_dict.get('_type', 'video') == 'video'
2313
2314 info_dict.setdefault('__postprocessors', [])
2315
2316 max_downloads = self.params.get('max_downloads')
2317 if max_downloads is not None:
2318 if self._num_downloads >= int(max_downloads):
2319 raise MaxDownloadsReached()
2320
2321 # TODO: backward compatibility, to be removed
2322 info_dict['fulltitle'] = info_dict['title']
2323
2324 if 'format' not in info_dict:
2325 info_dict['format'] = info_dict['ext']
2326
2327 if self._match_entry(info_dict) is not None:
2328 return
2329
2330 self.post_extract(info_dict)
2331 self._num_downloads += 1
2332
2333 # info_dict['_filename'] needs to be set for backward compatibility
2334 info_dict['_filename'] = full_filename = self.prepare_filename(info_dict, warn=True)
2335 temp_filename = self.prepare_filename(info_dict, 'temp')
2336 files_to_move = {}
2337
2338 # Forced printings
2339 self.__forced_printings(info_dict, full_filename, incomplete=False)
2340
2341 if self.params.get('simulate', False):
2342 if self.params.get('force_write_download_archive', False):
2343 self.record_download_archive(info_dict)
2344
2345 # Do nothing else if in simulate mode
2346 return
2347
2348 if full_filename is None:
2349 return
2350
2351 if not self._ensure_dir_exists(encodeFilename(full_filename)):
2352 return
2353 if not self._ensure_dir_exists(encodeFilename(temp_filename)):
2354 return
2355
2356 if self.params.get('writedescription', False):
2357 descfn = self.prepare_filename(info_dict, 'description')
2358 if not self._ensure_dir_exists(encodeFilename(descfn)):
2359 return
2360 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(descfn)):
2361 self.to_screen('[info] Video description is already present')
2362 elif info_dict.get('description') is None:
2363 self.report_warning('There\'s no description to write.')
2364 else:
2365 try:
2366 self.to_screen('[info] Writing video description to: ' + descfn)
2367 with io.open(encodeFilename(descfn), 'w', encoding='utf-8') as descfile:
2368 descfile.write(info_dict['description'])
2369 except (OSError, IOError):
2370 self.report_error('Cannot write description file ' + descfn)
2371 return
2372
2373 if self.params.get('writeannotations', False):
2374 annofn = self.prepare_filename(info_dict, 'annotation')
2375 if not self._ensure_dir_exists(encodeFilename(annofn)):
2376 return
2377 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(annofn)):
2378 self.to_screen('[info] Video annotations are already present')
2379 elif not info_dict.get('annotations'):
2380 self.report_warning('There are no annotations to write.')
2381 else:
2382 try:
2383 self.to_screen('[info] Writing video annotations to: ' + annofn)
2384 with io.open(encodeFilename(annofn), 'w', encoding='utf-8') as annofile:
2385 annofile.write(info_dict['annotations'])
2386 except (KeyError, TypeError):
2387 self.report_warning('There are no annotations to write.')
2388 except (OSError, IOError):
2389 self.report_error('Cannot write annotations file: ' + annofn)
2390 return
2391
2392 subtitles_are_requested = any([self.params.get('writesubtitles', False),
2393 self.params.get('writeautomaticsub')])
2394
2395 if subtitles_are_requested and info_dict.get('requested_subtitles'):
2396 # subtitles download errors are already managed as troubles in relevant IE
2397 # that way it will silently go on when used with unsupporting IE
2398 subtitles = info_dict['requested_subtitles']
2399 # ie = self.get_info_extractor(info_dict['extractor_key'])
2400 for sub_lang, sub_info in subtitles.items():
2401 sub_format = sub_info['ext']
2402 sub_filename = subtitles_filename(temp_filename, sub_lang, sub_format, info_dict.get('ext'))
2403 sub_filename_final = subtitles_filename(
2404 self.prepare_filename(info_dict, 'subtitle'), sub_lang, sub_format, info_dict.get('ext'))
2405 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(sub_filename)):
2406 self.to_screen('[info] Video subtitle %s.%s is already present' % (sub_lang, sub_format))
2407 sub_info['filepath'] = sub_filename
2408 files_to_move[sub_filename] = sub_filename_final
2409 else:
2410 self.to_screen('[info] Writing video subtitles to: ' + sub_filename)
2411 if sub_info.get('data') is not None:
2412 try:
2413 # Use newline='' to prevent conversion of newline characters
2414 # See https://github.com/ytdl-org/youtube-dl/issues/10268
2415 with io.open(encodeFilename(sub_filename), 'w', encoding='utf-8', newline='') as subfile:
2416 subfile.write(sub_info['data'])
2417 sub_info['filepath'] = sub_filename
2418 files_to_move[sub_filename] = sub_filename_final
2419 except (OSError, IOError):
2420 self.report_error('Cannot write subtitles file ' + sub_filename)
2421 return
2422 else:
2423 try:
2424 self.dl(sub_filename, sub_info.copy(), subtitle=True)
2425 sub_info['filepath'] = sub_filename
2426 files_to_move[sub_filename] = sub_filename_final
2427 except (ExtractorError, IOError, OSError, ValueError) + network_exceptions as err:
2428 self.report_warning('Unable to download subtitle for "%s": %s' %
2429 (sub_lang, error_to_compat_str(err)))
2430 continue
2431
2432 if self.params.get('writeinfojson', False):
2433 infofn = self.prepare_filename(info_dict, 'infojson')
2434 if not self._ensure_dir_exists(encodeFilename(infofn)):
2435 return
2436 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(infofn)):
2437 self.to_screen('[info] Video metadata is already present')
2438 else:
2439 self.to_screen('[info] Writing video metadata as JSON to: ' + infofn)
2440 try:
2441 write_json_file(self.filter_requested_info(info_dict, self.params.get('clean_infojson', True)), infofn)
2442 except (OSError, IOError):
2443 self.report_error('Cannot write video metadata to JSON file ' + infofn)
2444 return
2445 info_dict['__infojson_filename'] = infofn
2446
2447 for thumb_ext in self._write_thumbnails(info_dict, temp_filename):
2448 thumb_filename_temp = replace_extension(temp_filename, thumb_ext, info_dict.get('ext'))
2449 thumb_filename = replace_extension(
2450 self.prepare_filename(info_dict, 'thumbnail'), thumb_ext, info_dict.get('ext'))
2451 files_to_move[thumb_filename_temp] = thumb_filename
2452
2453 # Write internet shortcut files
2454 url_link = webloc_link = desktop_link = False
2455 if self.params.get('writelink', False):
2456 if sys.platform == "darwin": # macOS.
2457 webloc_link = True
2458 elif sys.platform.startswith("linux"):
2459 desktop_link = True
2460 else: # if sys.platform in ['win32', 'cygwin']:
2461 url_link = True
2462 if self.params.get('writeurllink', False):
2463 url_link = True
2464 if self.params.get('writewebloclink', False):
2465 webloc_link = True
2466 if self.params.get('writedesktoplink', False):
2467 desktop_link = True
2468
2469 if url_link or webloc_link or desktop_link:
2470 if 'webpage_url' not in info_dict:
2471 self.report_error('Cannot write internet shortcut file because the "webpage_url" field is missing in the media information')
2472 return
2473 ascii_url = iri_to_uri(info_dict['webpage_url'])
2474
2475 def _write_link_file(extension, template, newline, embed_filename):
2476 linkfn = replace_extension(full_filename, extension, info_dict.get('ext'))
2477 if self.params.get('overwrites', True) and os.path.exists(encodeFilename(linkfn)):
2478 self.to_screen('[info] Internet shortcut is already present')
2479 else:
2480 try:
2481 self.to_screen('[info] Writing internet shortcut to: ' + linkfn)
2482 with io.open(encodeFilename(to_high_limit_path(linkfn)), 'w', encoding='utf-8', newline=newline) as linkfile:
2483 template_vars = {'url': ascii_url}
2484 if embed_filename:
2485 template_vars['filename'] = linkfn[:-(len(extension) + 1)]
2486 linkfile.write(template % template_vars)
2487 except (OSError, IOError):
2488 self.report_error('Cannot write internet shortcut ' + linkfn)
2489 return False
2490 return True
2491
2492 if url_link:
2493 if not _write_link_file('url', DOT_URL_LINK_TEMPLATE, '\r\n', embed_filename=False):
2494 return
2495 if webloc_link:
2496 if not _write_link_file('webloc', DOT_WEBLOC_LINK_TEMPLATE, '\n', embed_filename=False):
2497 return
2498 if desktop_link:
2499 if not _write_link_file('desktop', DOT_DESKTOP_LINK_TEMPLATE, '\n', embed_filename=True):
2500 return
2501
2502 try:
2503 info_dict, files_to_move = self.pre_process(info_dict, 'before_dl', files_to_move)
2504 except PostProcessingError as err:
2505 self.report_error('Preprocessing: %s' % str(err))
2506 return
2507
2508 must_record_download_archive = False
2509 if self.params.get('skip_download', False):
2510 info_dict['filepath'] = temp_filename
2511 info_dict['__finaldir'] = os.path.dirname(os.path.abspath(encodeFilename(full_filename)))
2512 info_dict['__files_to_move'] = files_to_move
2513 info_dict = self.run_pp(MoveFilesAfterDownloadPP(self, False), info_dict)
2514 else:
2515 # Download
2516 try:
2517
2518 def existing_file(*filepaths):
2519 ext = info_dict.get('ext')
2520 final_ext = self.params.get('final_ext', ext)
2521 existing_files = []
2522 for file in orderedSet(filepaths):
2523 if final_ext != ext:
2524 converted = replace_extension(file, final_ext, ext)
2525 if os.path.exists(encodeFilename(converted)):
2526 existing_files.append(converted)
2527 if os.path.exists(encodeFilename(file)):
2528 existing_files.append(file)
2529
2530 if not existing_files or self.params.get('overwrites', False):
2531 for file in orderedSet(existing_files):
2532 self.report_file_delete(file)
2533 os.remove(encodeFilename(file))
2534 return None
2535
2536 self.report_file_already_downloaded(existing_files[0])
2537 info_dict['ext'] = os.path.splitext(existing_files[0])[1][1:]
2538 return existing_files[0]
2539
2540 success = True
2541 if info_dict.get('requested_formats') is not None:
2542
2543 def compatible_formats(formats):
2544 # TODO: some formats actually allow this (mkv, webm, ogg, mp4), but not all of them.
2545 video_formats = [format for format in formats if format.get('vcodec') != 'none']
2546 audio_formats = [format for format in formats if format.get('acodec') != 'none']
2547 if len(video_formats) > 2 or len(audio_formats) > 2:
2548 return False
2549
2550 # Check extension
2551 exts = set(format.get('ext') for format in formats)
2552 COMPATIBLE_EXTS = (
2553 set(('mp3', 'mp4', 'm4a', 'm4p', 'm4b', 'm4r', 'm4v', 'ismv', 'isma')),
2554 set(('webm',)),
2555 )
2556 for ext_sets in COMPATIBLE_EXTS:
2557 if ext_sets.issuperset(exts):
2558 return True
2559 # TODO: Check acodec/vcodec
2560 return False
2561
2562 requested_formats = info_dict['requested_formats']
2563 old_ext = info_dict['ext']
2564 if self.params.get('merge_output_format') is None:
2565 if not compatible_formats(requested_formats):
2566 info_dict['ext'] = 'mkv'
2567 self.report_warning(
2568 'Requested formats are incompatible for merge and will be merged into mkv.')
2569 if (info_dict['ext'] == 'webm'
2570 and self.params.get('writethumbnail', False)
2571 and info_dict.get('thumbnails')):
2572 info_dict['ext'] = 'mkv'
2573 self.report_warning(
2574 'webm doesn\'t support embedding a thumbnail, mkv will be used.')
2575
2576 def correct_ext(filename):
2577 filename_real_ext = os.path.splitext(filename)[1][1:]
2578 filename_wo_ext = (
2579 os.path.splitext(filename)[0]
2580 if filename_real_ext == old_ext
2581 else filename)
2582 return '%s.%s' % (filename_wo_ext, info_dict['ext'])
2583
2584 # Ensure filename always has a correct extension for successful merge
2585 full_filename = correct_ext(full_filename)
2586 temp_filename = correct_ext(temp_filename)
2587 dl_filename = existing_file(full_filename, temp_filename)
2588 info_dict['__real_download'] = False
2589
2590 _protocols = set(determine_protocol(f) for f in requested_formats)
2591 if len(_protocols) == 1:
2592 info_dict['protocol'] = _protocols.pop()
2593 directly_mergable = (
2594 'no-direct-merge' not in self.params.get('compat_opts', [])
2595 and info_dict.get('protocol') is not None # All requested formats have same protocol
2596 and not self.params.get('allow_unplayable_formats')
2597 and get_suitable_downloader(info_dict, self.params).__name__ == 'FFmpegFD')
2598 if directly_mergable:
2599 info_dict['url'] = requested_formats[0]['url']
2600 # Treat it as a single download
2601 dl_filename = existing_file(full_filename, temp_filename)
2602 if dl_filename is None:
2603 success, real_download = self.dl(temp_filename, info_dict)
2604 info_dict['__real_download'] = real_download
2605 else:
2606 downloaded = []
2607 merger = FFmpegMergerPP(self)
2608 if self.params.get('allow_unplayable_formats'):
2609 self.report_warning(
2610 'You have requested merging of multiple formats '
2611 'while also allowing unplayable formats to be downloaded. '
2612 'The formats won\'t be merged to prevent data corruption.')
2613 elif not merger.available:
2614 self.report_warning(
2615 'You have requested merging of multiple formats but ffmpeg is not installed. '
2616 'The formats won\'t be merged.')
2617
2618 if dl_filename is None:
2619 for f in requested_formats:
2620 new_info = dict(info_dict)
2621 del new_info['requested_formats']
2622 new_info.update(f)
2623 fname = prepend_extension(
2624 self.prepare_filename(new_info, 'temp'),
2625 'f%s' % f['format_id'], new_info['ext'])
2626 if not self._ensure_dir_exists(fname):
2627 return
2628 downloaded.append(fname)
2629 partial_success, real_download = self.dl(fname, new_info)
2630 info_dict['__real_download'] = info_dict['__real_download'] or real_download
2631 success = success and partial_success
2632 if merger.available and not self.params.get('allow_unplayable_formats'):
2633 info_dict['__postprocessors'].append(merger)
2634 info_dict['__files_to_merge'] = downloaded
2635 # Even if there were no downloads, it is being merged only now
2636 info_dict['__real_download'] = True
2637 else:
2638 for file in downloaded:
2639 files_to_move[file] = None
2640 else:
2641 # Just a single file
2642 dl_filename = existing_file(full_filename, temp_filename)
2643 if dl_filename is None:
2644 success, real_download = self.dl(temp_filename, info_dict)
2645 info_dict['__real_download'] = real_download
2646
2647 dl_filename = dl_filename or temp_filename
2648 info_dict['__finaldir'] = os.path.dirname(os.path.abspath(encodeFilename(full_filename)))
2649
2650 except network_exceptions as err:
2651 self.report_error('unable to download video data: %s' % error_to_compat_str(err))
2652 return
2653 except (OSError, IOError) as err:
2654 raise UnavailableVideoError(err)
2655 except (ContentTooShortError, ) as err:
2656 self.report_error('content too short (expected %s bytes and served %s)' % (err.expected, err.downloaded))
2657 return
2658
2659 if success and full_filename != '-':
2660 # Fixup content
2661 fixup_policy = self.params.get('fixup')
2662 if fixup_policy is None:
2663 fixup_policy = 'detect_or_warn'
2664
2665 INSTALL_FFMPEG_MESSAGE = 'Install ffmpeg to fix this automatically.'
2666
2667 stretched_ratio = info_dict.get('stretched_ratio')
2668 if stretched_ratio is not None and stretched_ratio != 1:
2669 if fixup_policy == 'warn':
2670 self.report_warning('%s: Non-uniform pixel ratio (%s)' % (
2671 info_dict['id'], stretched_ratio))
2672 elif fixup_policy == 'detect_or_warn':
2673 stretched_pp = FFmpegFixupStretchedPP(self)
2674 if stretched_pp.available:
2675 info_dict['__postprocessors'].append(stretched_pp)
2676 else:
2677 self.report_warning(
2678 '%s: Non-uniform pixel ratio (%s). %s'
2679 % (info_dict['id'], stretched_ratio, INSTALL_FFMPEG_MESSAGE))
2680 else:
2681 assert fixup_policy in ('ignore', 'never')
2682
2683 if (info_dict.get('requested_formats') is None
2684 and info_dict.get('container') == 'm4a_dash'
2685 and info_dict.get('ext') == 'm4a'):
2686 if fixup_policy == 'warn':
2687 self.report_warning(
2688 '%s: writing DASH m4a. '
2689 'Only some players support this container.'
2690 % info_dict['id'])
2691 elif fixup_policy == 'detect_or_warn':
2692 fixup_pp = FFmpegFixupM4aPP(self)
2693 if fixup_pp.available:
2694 info_dict['__postprocessors'].append(fixup_pp)
2695 else:
2696 self.report_warning(
2697 '%s: writing DASH m4a. '
2698 'Only some players support this container. %s'
2699 % (info_dict['id'], INSTALL_FFMPEG_MESSAGE))
2700 else:
2701 assert fixup_policy in ('ignore', 'never')
2702
2703 if ('protocol' in info_dict
2704 and get_suitable_downloader(info_dict, self.params).__name__ == 'HlsFD'):
2705 if fixup_policy == 'warn':
2706 self.report_warning('%s: malformed AAC bitstream detected.' % (
2707 info_dict['id']))
2708 elif fixup_policy == 'detect_or_warn':
2709 fixup_pp = FFmpegFixupM3u8PP(self)
2710 if fixup_pp.available:
2711 info_dict['__postprocessors'].append(fixup_pp)
2712 else:
2713 self.report_warning(
2714 '%s: malformed AAC bitstream detected. %s'
2715 % (info_dict['id'], INSTALL_FFMPEG_MESSAGE))
2716 else:
2717 assert fixup_policy in ('ignore', 'never')
2718
2719 try:
2720 info_dict = self.post_process(dl_filename, info_dict, files_to_move)
2721 except PostProcessingError as err:
2722 self.report_error('Postprocessing: %s' % str(err))
2723 return
2724 try:
2725 for ph in self._post_hooks:
2726 ph(info_dict['filepath'])
2727 except Exception as err:
2728 self.report_error('post hooks: %s' % str(err))
2729 return
2730 must_record_download_archive = True
2731
2732 if must_record_download_archive or self.params.get('force_write_download_archive', False):
2733 self.record_download_archive(info_dict)
2734 max_downloads = self.params.get('max_downloads')
2735 if max_downloads is not None and self._num_downloads >= int(max_downloads):
2736 raise MaxDownloadsReached()
2737
2738 def download(self, url_list):
2739 """Download a given list of URLs."""
2740 outtmpl = self.outtmpl_dict['default']
2741 if (len(url_list) > 1
2742 and outtmpl != '-'
2743 and '%' not in outtmpl
2744 and self.params.get('max_downloads') != 1):
2745 raise SameFileError(outtmpl)
2746
2747 for url in url_list:
2748 try:
2749 # It also downloads the videos
2750 res = self.extract_info(
2751 url, force_generic_extractor=self.params.get('force_generic_extractor', False))
2752 except UnavailableVideoError:
2753 self.report_error('unable to download video')
2754 except MaxDownloadsReached:
2755 self.to_screen('[info] Maximum number of downloaded files reached')
2756 raise
2757 except ExistingVideoReached:
2758 self.to_screen('[info] Encountered a file that is already in the archive, stopping due to --break-on-existing')
2759 raise
2760 except RejectedVideoReached:
2761 self.to_screen('[info] Encountered a file that did not match filter, stopping due to --break-on-reject')
2762 raise
2763 else:
2764 if self.params.get('dump_single_json', False):
2765 self.post_extract(res)
2766 self.to_stdout(json.dumps(res, default=repr))
2767
2768 return self._download_retcode
2769
2770 def download_with_info_file(self, info_filename):
2771 with contextlib.closing(fileinput.FileInput(
2772 [info_filename], mode='r',
2773 openhook=fileinput.hook_encoded('utf-8'))) as f:
2774 # FileInput doesn't have a read method, we can't call json.load
2775 info = self.filter_requested_info(json.loads('\n'.join(f)), self.params.get('clean_infojson', True))
2776 try:
2777 self.process_ie_result(info, download=True)
2778 except (DownloadError, EntryNotInPlaylist):
2779 webpage_url = info.get('webpage_url')
2780 if webpage_url is not None:
2781 self.report_warning('The info failed to download, trying with "%s"' % webpage_url)
2782 return self.download([webpage_url])
2783 else:
2784 raise
2785 return self._download_retcode
2786
2787 @staticmethod
2788 def filter_requested_info(info_dict, actually_filter=True):
2789 remove_keys = ['__original_infodict'] # Always remove this since this may contain a copy of the entire dict
2790 keep_keys = ['_type'], # Always keep this to facilitate load-info-json
2791 if actually_filter:
2792 remove_keys += ('requested_formats', 'requested_subtitles', 'requested_entries', 'filepath', 'entries', 'original_url')
2793 empty_values = (None, {}, [], set(), tuple())
2794 reject = lambda k, v: k not in keep_keys and (
2795 k.startswith('_') or k in remove_keys or v in empty_values)
2796 else:
2797 info_dict['epoch'] = int(time.time())
2798 reject = lambda k, v: k in remove_keys
2799 filter_fn = lambda obj: (
2800 list(map(filter_fn, obj)) if isinstance(obj, (list, tuple, set))
2801 else obj if not isinstance(obj, dict)
2802 else dict((k, filter_fn(v)) for k, v in obj.items() if not reject(k, v)))
2803 return filter_fn(info_dict)
2804
2805 def run_pp(self, pp, infodict):
2806 files_to_delete = []
2807 if '__files_to_move' not in infodict:
2808 infodict['__files_to_move'] = {}
2809 files_to_delete, infodict = pp.run(infodict)
2810 if not files_to_delete:
2811 return infodict
2812
2813 if self.params.get('keepvideo', False):
2814 for f in files_to_delete:
2815 infodict['__files_to_move'].setdefault(f, '')
2816 else:
2817 for old_filename in set(files_to_delete):
2818 self.to_screen('Deleting original file %s (pass -k to keep)' % old_filename)
2819 try:
2820 os.remove(encodeFilename(old_filename))
2821 except (IOError, OSError):
2822 self.report_warning('Unable to remove downloaded original file')
2823 if old_filename in infodict['__files_to_move']:
2824 del infodict['__files_to_move'][old_filename]
2825 return infodict
2826
2827 @staticmethod
2828 def post_extract(info_dict):
2829 def actual_post_extract(info_dict):
2830 if info_dict.get('_type') in ('playlist', 'multi_video'):
2831 for video_dict in info_dict.get('entries', {}):
2832 actual_post_extract(video_dict or {})
2833 return
2834
2835 post_extractor = info_dict.get('__post_extractor') or (lambda: {})
2836 extra = post_extractor().items()
2837 info_dict.update(extra)
2838 info_dict.pop('__post_extractor', None)
2839
2840 original_infodict = info_dict.get('__original_infodict') or {}
2841 original_infodict.update(extra)
2842 original_infodict.pop('__post_extractor', None)
2843
2844 actual_post_extract(info_dict or {})
2845
2846 def pre_process(self, ie_info, key='pre_process', files_to_move=None):
2847 info = dict(ie_info)
2848 info['__files_to_move'] = files_to_move or {}
2849 for pp in self._pps[key]:
2850 info = self.run_pp(pp, info)
2851 return info, info.pop('__files_to_move', None)
2852
2853 def post_process(self, filename, ie_info, files_to_move=None):
2854 """Run all the postprocessors on the given file."""
2855 info = dict(ie_info)
2856 info['filepath'] = filename
2857 info['__files_to_move'] = files_to_move or {}
2858
2859 for pp in ie_info.get('__postprocessors', []) + self._pps['post_process']:
2860 info = self.run_pp(pp, info)
2861 info = self.run_pp(MoveFilesAfterDownloadPP(self), info)
2862 del info['__files_to_move']
2863 for pp in self._pps['after_move']:
2864 info = self.run_pp(pp, info)
2865 return info
2866
2867 def _make_archive_id(self, info_dict):
2868 video_id = info_dict.get('id')
2869 if not video_id:
2870 return
2871 # Future-proof against any change in case
2872 # and backwards compatibility with prior versions
2873 extractor = info_dict.get('extractor_key') or info_dict.get('ie_key') # key in a playlist
2874 if extractor is None:
2875 url = str_or_none(info_dict.get('url'))
2876 if not url:
2877 return
2878 # Try to find matching extractor for the URL and take its ie_key
2879 for ie in self._ies:
2880 if ie.suitable(url):
2881 extractor = ie.ie_key()
2882 break
2883 else:
2884 return
2885 return '%s %s' % (extractor.lower(), video_id)
2886
2887 def in_download_archive(self, info_dict):
2888 fn = self.params.get('download_archive')
2889 if fn is None:
2890 return False
2891
2892 vid_id = self._make_archive_id(info_dict)
2893 if not vid_id:
2894 return False # Incomplete video information
2895
2896 return vid_id in self.archive
2897
2898 def record_download_archive(self, info_dict):
2899 fn = self.params.get('download_archive')
2900 if fn is None:
2901 return
2902 vid_id = self._make_archive_id(info_dict)
2903 assert vid_id
2904 with locked_file(fn, 'a', encoding='utf-8') as archive_file:
2905 archive_file.write(vid_id + '\n')
2906 self.archive.add(vid_id)
2907
2908 @staticmethod
2909 def format_resolution(format, default='unknown'):
2910 if format.get('vcodec') == 'none':
2911 return 'audio only'
2912 if format.get('resolution') is not None:
2913 return format['resolution']
2914 if format.get('width') and format.get('height'):
2915 res = '%dx%d' % (format['width'], format['height'])
2916 elif format.get('height'):
2917 res = '%sp' % format['height']
2918 elif format.get('width'):
2919 res = '%dx?' % format['width']
2920 else:
2921 res = default
2922 return res
2923
2924 def _format_note(self, fdict):
2925 res = ''
2926 if fdict.get('ext') in ['f4f', 'f4m']:
2927 res += '(unsupported) '
2928 if fdict.get('language'):
2929 if res:
2930 res += ' '
2931 res += '[%s] ' % fdict['language']
2932 if fdict.get('format_note') is not None:
2933 res += fdict['format_note'] + ' '
2934 if fdict.get('tbr') is not None:
2935 res += '%4dk ' % fdict['tbr']
2936 if fdict.get('container') is not None:
2937 if res:
2938 res += ', '
2939 res += '%s container' % fdict['container']
2940 if (fdict.get('vcodec') is not None
2941 and fdict.get('vcodec') != 'none'):
2942 if res:
2943 res += ', '
2944 res += fdict['vcodec']
2945 if fdict.get('vbr') is not None:
2946 res += '@'
2947 elif fdict.get('vbr') is not None and fdict.get('abr') is not None:
2948 res += 'video@'
2949 if fdict.get('vbr') is not None:
2950 res += '%4dk' % fdict['vbr']
2951 if fdict.get('fps') is not None:
2952 if res:
2953 res += ', '
2954 res += '%sfps' % fdict['fps']
2955 if fdict.get('acodec') is not None:
2956 if res:
2957 res += ', '
2958 if fdict['acodec'] == 'none':
2959 res += 'video only'
2960 else:
2961 res += '%-5s' % fdict['acodec']
2962 elif fdict.get('abr') is not None:
2963 if res:
2964 res += ', '
2965 res += 'audio'
2966 if fdict.get('abr') is not None:
2967 res += '@%3dk' % fdict['abr']
2968 if fdict.get('asr') is not None:
2969 res += ' (%5dHz)' % fdict['asr']
2970 if fdict.get('filesize') is not None:
2971 if res:
2972 res += ', '
2973 res += format_bytes(fdict['filesize'])
2974 elif fdict.get('filesize_approx') is not None:
2975 if res:
2976 res += ', '
2977 res += '~' + format_bytes(fdict['filesize_approx'])
2978 return res
2979
2980 def _format_note_table(self, f):
2981 def join_fields(*vargs):
2982 return ', '.join((val for val in vargs if val != ''))
2983
2984 return join_fields(
2985 'UNSUPPORTED' if f.get('ext') in ('f4f', 'f4m') else '',
2986 format_field(f, 'language', '[%s]'),
2987 format_field(f, 'format_note'),
2988 format_field(f, 'container', ignore=(None, f.get('ext'))),
2989 format_field(f, 'asr', '%5dHz'))
2990
2991 def list_formats(self, info_dict):
2992 formats = info_dict.get('formats', [info_dict])
2993 new_format = (
2994 'list-formats' not in self.params.get('compat_opts', [])
2995 and self.params.get('list_formats_as_table', True) is not False)
2996 if new_format:
2997 table = [
2998 [
2999 format_field(f, 'format_id'),
3000 format_field(f, 'ext'),
3001 self.format_resolution(f),
3002 format_field(f, 'fps', '%d'),
3003 '|',
3004 format_field(f, 'filesize', ' %s', func=format_bytes) + format_field(f, 'filesize_approx', '~%s', func=format_bytes),
3005 format_field(f, 'tbr', '%4dk'),
3006 shorten_protocol_name(f.get('protocol', '').replace("native", "n")),
3007 '|',
3008 format_field(f, 'vcodec', default='unknown').replace('none', ''),
3009 format_field(f, 'vbr', '%4dk'),
3010 format_field(f, 'acodec', default='unknown').replace('none', ''),
3011 format_field(f, 'abr', '%3dk'),
3012 format_field(f, 'asr', '%5dHz'),
3013 self._format_note_table(f)]
3014 for f in formats
3015 if f.get('preference') is None or f['preference'] >= -1000]
3016 header_line = ['ID', 'EXT', 'RESOLUTION', 'FPS', '|', ' FILESIZE', ' TBR', 'PROTO',
3017 '|', 'VCODEC', ' VBR', 'ACODEC', ' ABR', ' ASR', 'NOTE']
3018 else:
3019 table = [
3020 [
3021 format_field(f, 'format_id'),
3022 format_field(f, 'ext'),
3023 self.format_resolution(f),
3024 self._format_note(f)]
3025 for f in formats
3026 if f.get('preference') is None or f['preference'] >= -1000]
3027 header_line = ['format code', 'extension', 'resolution', 'note']
3028
3029 self.to_screen(
3030 '[info] Available formats for %s:\n%s' % (info_dict['id'], render_table(
3031 header_line,
3032 table,
3033 delim=new_format,
3034 extraGap=(0 if new_format else 1),
3035 hideEmpty=new_format)))
3036
3037 def list_thumbnails(self, info_dict):
3038 thumbnails = info_dict.get('thumbnails')
3039 if not thumbnails:
3040 self.to_screen('[info] No thumbnails present for %s' % info_dict['id'])
3041 return
3042
3043 self.to_screen(
3044 '[info] Thumbnails for %s:' % info_dict['id'])
3045 self.to_screen(render_table(
3046 ['ID', 'width', 'height', 'URL'],
3047 [[t['id'], t.get('width', 'unknown'), t.get('height', 'unknown'), t['url']] for t in thumbnails]))
3048
3049 def list_subtitles(self, video_id, subtitles, name='subtitles'):
3050 if not subtitles:
3051 self.to_screen('%s has no %s' % (video_id, name))
3052 return
3053 self.to_screen(
3054 'Available %s for %s:' % (name, video_id))
3055
3056 def _row(lang, formats):
3057 exts, names = zip(*((f['ext'], f.get('name', 'unknown')) for f in reversed(formats)))
3058 if len(set(names)) == 1:
3059 names = [] if names[0] == 'unknown' else names[:1]
3060 return [lang, ', '.join(names), ', '.join(exts)]
3061
3062 self.to_screen(render_table(
3063 ['Language', 'Name', 'Formats'],
3064 [_row(lang, formats) for lang, formats in subtitles.items()],
3065 hideEmpty=True))
3066
3067 def urlopen(self, req):
3068 """ Start an HTTP download """
3069 if isinstance(req, compat_basestring):
3070 req = sanitized_Request(req)
3071 return self._opener.open(req, timeout=self._socket_timeout)
3072
3073 def print_debug_header(self):
3074 if not self.params.get('verbose'):
3075 return
3076
3077 if type('') is not compat_str:
3078 # Python 2.6 on SLES11 SP1 (https://github.com/ytdl-org/youtube-dl/issues/3326)
3079 self.report_warning(
3080 'Your Python is broken! Update to a newer and supported version')
3081
3082 stdout_encoding = getattr(
3083 sys.stdout, 'encoding', 'missing (%s)' % type(sys.stdout).__name__)
3084 encoding_str = (
3085 '[debug] Encodings: locale %s, fs %s, out %s, pref %s\n' % (
3086 locale.getpreferredencoding(),
3087 sys.getfilesystemencoding(),
3088 stdout_encoding,
3089 self.get_encoding()))
3090 write_string(encoding_str, encoding=None)
3091
3092 source = (
3093 '(exe)' if hasattr(sys, 'frozen')
3094 else '(zip)' if isinstance(globals().get('__loader__'), zipimporter)
3095 else '(source)' if os.path.basename(sys.argv[0]) == '__main__.py'
3096 else '')
3097 self._write_string('[debug] yt-dlp version %s %s\n' % (__version__, source))
3098 if _LAZY_LOADER:
3099 self._write_string('[debug] Lazy loading extractors enabled\n')
3100 if _PLUGIN_CLASSES:
3101 self._write_string(
3102 '[debug] Plugin Extractors: %s\n' % [ie.ie_key() for ie in _PLUGIN_CLASSES])
3103 if self.params.get('compat_opts'):
3104 self._write_string(
3105 '[debug] Compatibility options: %s\n' % ', '.join(self.params.get('compat_opts')))
3106 try:
3107 sp = subprocess.Popen(
3108 ['git', 'rev-parse', '--short', 'HEAD'],
3109 stdout=subprocess.PIPE, stderr=subprocess.PIPE,
3110 cwd=os.path.dirname(os.path.abspath(__file__)))
3111 out, err = process_communicate_or_kill(sp)
3112 out = out.decode().strip()
3113 if re.match('[0-9a-f]+', out):
3114 self._write_string('[debug] Git HEAD: %s\n' % out)
3115 except Exception:
3116 try:
3117 sys.exc_clear()
3118 except Exception:
3119 pass
3120
3121 def python_implementation():
3122 impl_name = platform.python_implementation()
3123 if impl_name == 'PyPy' and hasattr(sys, 'pypy_version_info'):
3124 return impl_name + ' version %d.%d.%d' % sys.pypy_version_info[:3]
3125 return impl_name
3126
3127 self._write_string('[debug] Python version %s (%s %s) - %s\n' % (
3128 platform.python_version(),
3129 python_implementation(),
3130 platform.architecture()[0],
3131 platform_name()))
3132
3133 exe_versions = FFmpegPostProcessor.get_versions(self)
3134 exe_versions['rtmpdump'] = rtmpdump_version()
3135 exe_versions['phantomjs'] = PhantomJSwrapper._version()
3136 exe_str = ', '.join(
3137 '%s %s' % (exe, v)
3138 for exe, v in sorted(exe_versions.items())
3139 if v
3140 )
3141 if not exe_str:
3142 exe_str = 'none'
3143 self._write_string('[debug] exe versions: %s\n' % exe_str)
3144
3145 proxy_map = {}
3146 for handler in self._opener.handlers:
3147 if hasattr(handler, 'proxies'):
3148 proxy_map.update(handler.proxies)
3149 self._write_string('[debug] Proxy map: ' + compat_str(proxy_map) + '\n')
3150
3151 if self.params.get('call_home', False):
3152 ipaddr = self.urlopen('https://yt-dl.org/ip').read().decode('utf-8')
3153 self._write_string('[debug] Public IP address: %s\n' % ipaddr)
3154 return
3155 latest_version = self.urlopen(
3156 'https://yt-dl.org/latest/version').read().decode('utf-8')
3157 if version_tuple(latest_version) > version_tuple(__version__):
3158 self.report_warning(
3159 'You are using an outdated version (newest version: %s)! '
3160 'See https://yt-dl.org/update if you need help updating.' %
3161 latest_version)
3162
3163 def _setup_opener(self):
3164 timeout_val = self.params.get('socket_timeout')
3165 self._socket_timeout = 600 if timeout_val is None else float(timeout_val)
3166
3167 opts_cookiefile = self.params.get('cookiefile')
3168 opts_proxy = self.params.get('proxy')
3169
3170 if opts_cookiefile is None:
3171 self.cookiejar = compat_cookiejar.CookieJar()
3172 else:
3173 opts_cookiefile = expand_path(opts_cookiefile)
3174 self.cookiejar = YoutubeDLCookieJar(opts_cookiefile)
3175 if os.access(opts_cookiefile, os.R_OK):
3176 self.cookiejar.load(ignore_discard=True, ignore_expires=True)
3177
3178 cookie_processor = YoutubeDLCookieProcessor(self.cookiejar)
3179 if opts_proxy is not None:
3180 if opts_proxy == '':
3181 proxies = {}
3182 else:
3183 proxies = {'http': opts_proxy, 'https': opts_proxy}
3184 else:
3185 proxies = compat_urllib_request.getproxies()
3186 # Set HTTPS proxy to HTTP one if given (https://github.com/ytdl-org/youtube-dl/issues/805)
3187 if 'http' in proxies and 'https' not in proxies:
3188 proxies['https'] = proxies['http']
3189 proxy_handler = PerRequestProxyHandler(proxies)
3190
3191 debuglevel = 1 if self.params.get('debug_printtraffic') else 0
3192 https_handler = make_HTTPS_handler(self.params, debuglevel=debuglevel)
3193 ydlh = YoutubeDLHandler(self.params, debuglevel=debuglevel)
3194 redirect_handler = YoutubeDLRedirectHandler()
3195 data_handler = compat_urllib_request_DataHandler()
3196
3197 # When passing our own FileHandler instance, build_opener won't add the
3198 # default FileHandler and allows us to disable the file protocol, which
3199 # can be used for malicious purposes (see
3200 # https://github.com/ytdl-org/youtube-dl/issues/8227)
3201 file_handler = compat_urllib_request.FileHandler()
3202
3203 def file_open(*args, **kwargs):
3204 raise compat_urllib_error.URLError('file:// scheme is explicitly disabled in yt-dlp for security reasons')
3205 file_handler.file_open = file_open
3206
3207 opener = compat_urllib_request.build_opener(
3208 proxy_handler, https_handler, cookie_processor, ydlh, redirect_handler, data_handler, file_handler)
3209
3210 # Delete the default user-agent header, which would otherwise apply in
3211 # cases where our custom HTTP handler doesn't come into play
3212 # (See https://github.com/ytdl-org/youtube-dl/issues/1309 for details)
3213 opener.addheaders = []
3214 self._opener = opener
3215
3216 def encode(self, s):
3217 if isinstance(s, bytes):
3218 return s # Already encoded
3219
3220 try:
3221 return s.encode(self.get_encoding())
3222 except UnicodeEncodeError as err:
3223 err.reason = err.reason + '. Check your system encoding configuration or use the --encoding option.'
3224 raise
3225
3226 def get_encoding(self):
3227 encoding = self.params.get('encoding')
3228 if encoding is None:
3229 encoding = preferredencoding()
3230 return encoding
3231
3232 def _write_thumbnails(self, info_dict, filename): # return the extensions
3233 write_all = self.params.get('write_all_thumbnails', False)
3234 thumbnails = []
3235 if write_all or self.params.get('writethumbnail', False):
3236 thumbnails = info_dict.get('thumbnails') or []
3237 multiple = write_all and len(thumbnails) > 1
3238
3239 ret = []
3240 for t in thumbnails[::1 if write_all else -1]:
3241 thumb_ext = determine_ext(t['url'], 'jpg')
3242 suffix = '%s.' % t['id'] if multiple else ''
3243 thumb_display_id = '%s ' % t['id'] if multiple else ''
3244 thumb_filename = replace_extension(filename, suffix + thumb_ext, info_dict.get('ext'))
3245
3246 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(thumb_filename)):
3247 ret.append(suffix + thumb_ext)
3248 self.to_screen('[%s] %s: Thumbnail %sis already present' %
3249 (info_dict['extractor'], info_dict['id'], thumb_display_id))
3250 else:
3251 self.to_screen('[%s] %s: Downloading thumbnail %s ...' %
3252 (info_dict['extractor'], info_dict['id'], thumb_display_id))
3253 try:
3254 uf = self.urlopen(t['url'])
3255 with open(encodeFilename(thumb_filename), 'wb') as thumbf:
3256 shutil.copyfileobj(uf, thumbf)
3257 ret.append(suffix + thumb_ext)
3258 self.to_screen('[%s] %s: Writing thumbnail %sto: %s' %
3259 (info_dict['extractor'], info_dict['id'], thumb_display_id, thumb_filename))
3260 t['filepath'] = thumb_filename
3261 except network_exceptions as err:
3262 self.report_warning('Unable to download thumbnail "%s": %s' %
3263 (t['url'], error_to_compat_str(err)))
3264 if ret and not write_all:
3265 break
3266 return ret