]> jfr.im git - yt-dlp.git/blob - yt_dlp/YoutubeDL.py
0edbb4119fb393132a7f2b6191d79f9fbfd17261
[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 'Support for Python version %d.%d have been deprecated and will break in future versions of yt-dlp! '
476 'Update to Python 3.6 or above' % sys.version_info[:2])
477
478 def check_deprecated(param, option, suggestion):
479 if self.params.get(param) is not None:
480 self.report_warning('%s is deprecated. Use %s instead' % (option, suggestion))
481 return True
482 return False
483
484 if check_deprecated('cn_verification_proxy', '--cn-verification-proxy', '--geo-verification-proxy'):
485 if self.params.get('geo_verification_proxy') is None:
486 self.params['geo_verification_proxy'] = self.params['cn_verification_proxy']
487
488 check_deprecated('autonumber', '--auto-number', '-o "%(autonumber)s-%(title)s.%(ext)s"')
489 check_deprecated('usetitle', '--title', '-o "%(title)s-%(id)s.%(ext)s"')
490 check_deprecated('useid', '--id', '-o "%(id)s.%(ext)s"')
491
492 for msg in self.params.get('warnings', []):
493 self.report_warning(msg)
494
495 if self.params.get('final_ext'):
496 if self.params.get('merge_output_format'):
497 self.report_warning('--merge-output-format will be ignored since --remux-video or --recode-video is given')
498 self.params['merge_output_format'] = self.params['final_ext']
499
500 if 'overwrites' in self.params and self.params['overwrites'] is None:
501 del self.params['overwrites']
502
503 if params.get('bidi_workaround', False):
504 try:
505 import pty
506 master, slave = pty.openpty()
507 width = compat_get_terminal_size().columns
508 if width is None:
509 width_args = []
510 else:
511 width_args = ['-w', str(width)]
512 sp_kwargs = dict(
513 stdin=subprocess.PIPE,
514 stdout=slave,
515 stderr=self._err_file)
516 try:
517 self._output_process = subprocess.Popen(
518 ['bidiv'] + width_args, **sp_kwargs
519 )
520 except OSError:
521 self._output_process = subprocess.Popen(
522 ['fribidi', '-c', 'UTF-8'] + width_args, **sp_kwargs)
523 self._output_channel = os.fdopen(master, 'rb')
524 except OSError as ose:
525 if ose.errno == errno.ENOENT:
526 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 $PATH.')
527 else:
528 raise
529
530 if (sys.platform != 'win32'
531 and sys.getfilesystemencoding() in ['ascii', 'ANSI_X3.4-1968']
532 and not params.get('restrictfilenames', False)):
533 # Unicode filesystem API will throw errors (#1474, #13027)
534 self.report_warning(
535 'Assuming --restrict-filenames since file system encoding '
536 'cannot encode all characters. '
537 'Set the LC_ALL environment variable to fix this.')
538 self.params['restrictfilenames'] = True
539
540 self.outtmpl_dict = self.parse_outtmpl()
541
542 self._setup_opener()
543
544 """Preload the archive, if any is specified"""
545 def preload_download_archive(fn):
546 if fn is None:
547 return False
548 self.write_debug('Loading archive file %r\n' % fn)
549 try:
550 with locked_file(fn, 'r', encoding='utf-8') as archive_file:
551 for line in archive_file:
552 self.archive.add(line.strip())
553 except IOError as ioe:
554 if ioe.errno != errno.ENOENT:
555 raise
556 return False
557 return True
558
559 self.archive = set()
560 preload_download_archive(self.params.get('download_archive'))
561
562 if auto_init:
563 self.print_debug_header()
564 self.add_default_info_extractors()
565
566 for pp_def_raw in self.params.get('postprocessors', []):
567 pp_class = get_postprocessor(pp_def_raw['key'])
568 pp_def = dict(pp_def_raw)
569 del pp_def['key']
570 if 'when' in pp_def:
571 when = pp_def['when']
572 del pp_def['when']
573 else:
574 when = 'post_process'
575 pp = pp_class(self, **compat_kwargs(pp_def))
576 self.add_post_processor(pp, when=when)
577
578 for ph in self.params.get('post_hooks', []):
579 self.add_post_hook(ph)
580
581 for ph in self.params.get('progress_hooks', []):
582 self.add_progress_hook(ph)
583
584 register_socks_protocols()
585
586 def warn_if_short_id(self, argv):
587 # short YouTube ID starting with dash?
588 idxs = [
589 i for i, a in enumerate(argv)
590 if re.match(r'^-[0-9A-Za-z_-]{10}$', a)]
591 if idxs:
592 correct_argv = (
593 ['yt-dlp']
594 + [a for i, a in enumerate(argv) if i not in idxs]
595 + ['--'] + [argv[i] for i in idxs]
596 )
597 self.report_warning(
598 'Long argument string detected. '
599 'Use -- to separate parameters and URLs, like this:\n%s\n' %
600 args_to_str(correct_argv))
601
602 def add_info_extractor(self, ie):
603 """Add an InfoExtractor object to the end of the list."""
604 self._ies.append(ie)
605 if not isinstance(ie, type):
606 self._ies_instances[ie.ie_key()] = ie
607 ie.set_downloader(self)
608
609 def get_info_extractor(self, ie_key):
610 """
611 Get an instance of an IE with name ie_key, it will try to get one from
612 the _ies list, if there's no instance it will create a new one and add
613 it to the extractor list.
614 """
615 ie = self._ies_instances.get(ie_key)
616 if ie is None:
617 ie = get_info_extractor(ie_key)()
618 self.add_info_extractor(ie)
619 return ie
620
621 def add_default_info_extractors(self):
622 """
623 Add the InfoExtractors returned by gen_extractors to the end of the list
624 """
625 for ie in gen_extractor_classes():
626 self.add_info_extractor(ie)
627
628 def add_post_processor(self, pp, when='post_process'):
629 """Add a PostProcessor object to the end of the chain."""
630 self._pps[when].append(pp)
631 pp.set_downloader(self)
632
633 def add_post_hook(self, ph):
634 """Add the post hook"""
635 self._post_hooks.append(ph)
636
637 def add_progress_hook(self, ph):
638 """Add the progress hook (currently only for the file downloader)"""
639 self._progress_hooks.append(ph)
640
641 def _bidi_workaround(self, message):
642 if not hasattr(self, '_output_channel'):
643 return message
644
645 assert hasattr(self, '_output_process')
646 assert isinstance(message, compat_str)
647 line_count = message.count('\n') + 1
648 self._output_process.stdin.write((message + '\n').encode('utf-8'))
649 self._output_process.stdin.flush()
650 res = ''.join(self._output_channel.readline().decode('utf-8')
651 for _ in range(line_count))
652 return res[:-len('\n')]
653
654 def _write_string(self, s, out=None):
655 write_string(s, out=out, encoding=self.params.get('encoding'))
656
657 def to_stdout(self, message, skip_eol=False, quiet=False):
658 """Print message to stdout"""
659 if self.params.get('logger'):
660 self.params['logger'].debug(message)
661 elif not quiet or self.params.get('verbose'):
662 self._write_string(
663 '%s%s' % (self._bidi_workaround(message), ('' if skip_eol else '\n')),
664 self._err_file if quiet else self._screen_file)
665
666 def to_stderr(self, message):
667 """Print message to stderr"""
668 assert isinstance(message, compat_str)
669 if self.params.get('logger'):
670 self.params['logger'].error(message)
671 else:
672 self._write_string('%s\n' % self._bidi_workaround(message), self._err_file)
673
674 def to_console_title(self, message):
675 if not self.params.get('consoletitle', False):
676 return
677 if compat_os_name == 'nt':
678 if ctypes.windll.kernel32.GetConsoleWindow():
679 # c_wchar_p() might not be necessary if `message` is
680 # already of type unicode()
681 ctypes.windll.kernel32.SetConsoleTitleW(ctypes.c_wchar_p(message))
682 elif 'TERM' in os.environ:
683 self._write_string('\033]0;%s\007' % message, self._screen_file)
684
685 def save_console_title(self):
686 if not self.params.get('consoletitle', False):
687 return
688 if self.params.get('simulate', False):
689 return
690 if compat_os_name != 'nt' and 'TERM' in os.environ:
691 # Save the title on stack
692 self._write_string('\033[22;0t', self._screen_file)
693
694 def restore_console_title(self):
695 if not self.params.get('consoletitle', False):
696 return
697 if self.params.get('simulate', False):
698 return
699 if compat_os_name != 'nt' and 'TERM' in os.environ:
700 # Restore the title from stack
701 self._write_string('\033[23;0t', self._screen_file)
702
703 def __enter__(self):
704 self.save_console_title()
705 return self
706
707 def __exit__(self, *args):
708 self.restore_console_title()
709
710 if self.params.get('cookiefile') is not None:
711 self.cookiejar.save(ignore_discard=True, ignore_expires=True)
712
713 def trouble(self, message=None, tb=None):
714 """Determine action to take when a download problem appears.
715
716 Depending on if the downloader has been configured to ignore
717 download errors or not, this method may throw an exception or
718 not when errors are found, after printing the message.
719
720 tb, if given, is additional traceback information.
721 """
722 if message is not None:
723 self.to_stderr(message)
724 if self.params.get('verbose'):
725 if tb is None:
726 if sys.exc_info()[0]: # if .trouble has been called from an except block
727 tb = ''
728 if hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
729 tb += ''.join(traceback.format_exception(*sys.exc_info()[1].exc_info))
730 tb += encode_compat_str(traceback.format_exc())
731 else:
732 tb_data = traceback.format_list(traceback.extract_stack())
733 tb = ''.join(tb_data)
734 if tb:
735 self.to_stderr(tb)
736 if not self.params.get('ignoreerrors', False):
737 if sys.exc_info()[0] and hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
738 exc_info = sys.exc_info()[1].exc_info
739 else:
740 exc_info = sys.exc_info()
741 raise DownloadError(message, exc_info)
742 self._download_retcode = 1
743
744 def to_screen(self, message, skip_eol=False):
745 """Print message to stdout if not in quiet mode"""
746 self.to_stdout(
747 message, skip_eol, quiet=self.params.get('quiet', False))
748
749 def report_warning(self, message):
750 '''
751 Print the message to stderr, it will be prefixed with 'WARNING:'
752 If stderr is a tty file the 'WARNING:' will be colored
753 '''
754 if self.params.get('logger') is not None:
755 self.params['logger'].warning(message)
756 else:
757 if self.params.get('no_warnings'):
758 return
759 if not self.params.get('no_color') and self._err_file.isatty() and compat_os_name != 'nt':
760 _msg_header = '\033[0;33mWARNING:\033[0m'
761 else:
762 _msg_header = 'WARNING:'
763 warning_message = '%s %s' % (_msg_header, message)
764 self.to_stderr(warning_message)
765
766 def report_error(self, message, tb=None):
767 '''
768 Do the same as trouble, but prefixes the message with 'ERROR:', colored
769 in red if stderr is a tty file.
770 '''
771 if not self.params.get('no_color') and self._err_file.isatty() and compat_os_name != 'nt':
772 _msg_header = '\033[0;31mERROR:\033[0m'
773 else:
774 _msg_header = 'ERROR:'
775 error_message = '%s %s' % (_msg_header, message)
776 self.trouble(error_message, tb)
777
778 def write_debug(self, message):
779 '''Log debug message or Print message to stderr'''
780 if not self.params.get('verbose', False):
781 return
782 message = '[debug] %s' % message
783 if self.params.get('logger'):
784 self.params['logger'].debug(message)
785 else:
786 self._write_string('%s\n' % message)
787
788 def report_file_already_downloaded(self, file_name):
789 """Report file has already been fully downloaded."""
790 try:
791 self.to_screen('[download] %s has already been downloaded' % file_name)
792 except UnicodeEncodeError:
793 self.to_screen('[download] The file has already been downloaded')
794
795 def report_file_delete(self, file_name):
796 """Report that existing file will be deleted."""
797 try:
798 self.to_screen('Deleting existing file %s' % file_name)
799 except UnicodeEncodeError:
800 self.to_screen('Deleting existing file')
801
802 def parse_outtmpl(self):
803 outtmpl_dict = self.params.get('outtmpl', {})
804 if not isinstance(outtmpl_dict, dict):
805 outtmpl_dict = {'default': outtmpl_dict}
806 outtmpl_dict.update({
807 k: v for k, v in DEFAULT_OUTTMPL.items()
808 if not outtmpl_dict.get(k)})
809 for key, val in outtmpl_dict.items():
810 if isinstance(val, bytes):
811 self.report_warning(
812 'Parameter outtmpl is bytes, but should be a unicode string. '
813 'Put from __future__ import unicode_literals at the top of your code file or consider switching to Python 3.x.')
814 return outtmpl_dict
815
816 @staticmethod
817 def validate_outtmpl(tmpl):
818 ''' @return None or Exception object '''
819 try:
820 re.sub(
821 STR_FORMAT_RE.format(''),
822 lambda mobj: ('%' if not mobj.group('has_key') else '') + mobj.group(0),
823 tmpl
824 ) % collections.defaultdict(int)
825 return None
826 except ValueError as err:
827 return err
828
829 def prepare_outtmpl(self, outtmpl, info_dict, sanitize=None):
830 """ Make the template and info_dict suitable for substitution (outtmpl % info_dict)"""
831 info_dict = dict(info_dict)
832 na = self.params.get('outtmpl_na_placeholder', 'NA')
833
834 info_dict['duration_string'] = ( # %(duration>%H-%M-%S)s is wrong if duration > 24hrs
835 formatSeconds(info_dict['duration'], '-' if sanitize else ':')
836 if info_dict.get('duration', None) is not None
837 else None)
838 info_dict['epoch'] = int(time.time())
839 info_dict['autonumber'] = self.params.get('autonumber_start', 1) - 1 + self._num_downloads
840 if info_dict.get('resolution') is None:
841 info_dict['resolution'] = self.format_resolution(info_dict, default=None)
842
843 # For fields playlist_index and autonumber convert all occurrences
844 # of %(field)s to %(field)0Nd for backward compatibility
845 field_size_compat_map = {
846 'playlist_index': len(str(info_dict.get('_last_playlist_index') or '')),
847 'autonumber': self.params.get('autonumber_size') or 5,
848 }
849
850 TMPL_DICT = {}
851 EXTERNAL_FORMAT_RE = re.compile(STR_FORMAT_RE.format('[^)]*'))
852 MATH_FUNCTIONS = {
853 '+': float.__add__,
854 '-': float.__sub__,
855 }
856 # Field is of the form key1.key2...
857 # where keys (except first) can be string, int or slice
858 FIELD_RE = r'\w+(?:\.(?:\w+|{num}|{num}?(?::{num}?){{1,2}}))*'.format(num=r'(?:-?\d+)')
859 MATH_FIELD_RE = r'''{field}|{num}'''.format(field=FIELD_RE, num=r'-?\d+(?:.\d+)?')
860 MATH_OPERATORS_RE = r'(?:%s)' % '|'.join(map(re.escape, MATH_FUNCTIONS.keys()))
861 INTERNAL_FORMAT_RE = re.compile(r'''(?x)
862 (?P<negate>-)?
863 (?P<fields>{field})
864 (?P<maths>(?:{math_op}{math_field})*)
865 (?:>(?P<strf_format>.+?))?
866 (?:\|(?P<default>.*?))?
867 $'''.format(field=FIELD_RE, math_op=MATH_OPERATORS_RE, math_field=MATH_FIELD_RE))
868
869 get_key = lambda k: traverse_obj(
870 info_dict, k.split('.'), is_user_input=True, traverse_string=True)
871
872 def get_value(mdict):
873 # Object traversal
874 value = get_key(mdict['fields'])
875 # Negative
876 if mdict['negate']:
877 value = float_or_none(value)
878 if value is not None:
879 value *= -1
880 # Do maths
881 offset_key = mdict['maths']
882 if offset_key:
883 value = float_or_none(value)
884 operator = None
885 while offset_key:
886 item = re.match(
887 MATH_FIELD_RE if operator else MATH_OPERATORS_RE,
888 offset_key).group(0)
889 offset_key = offset_key[len(item):]
890 if operator is None:
891 operator = MATH_FUNCTIONS[item]
892 continue
893 item, multiplier = (item[1:], -1) if item[0] == '-' else (item, 1)
894 offset = float_or_none(item)
895 if offset is None:
896 offset = float_or_none(get_key(item))
897 try:
898 value = operator(value, multiplier * offset)
899 except (TypeError, ZeroDivisionError):
900 return None
901 operator = None
902 # Datetime formatting
903 if mdict['strf_format']:
904 value = strftime_or_none(value, mdict['strf_format'])
905
906 return value
907
908 def create_key(outer_mobj):
909 if not outer_mobj.group('has_key'):
910 return '%{}'.format(outer_mobj.group(0))
911
912 key = outer_mobj.group('key')
913 fmt = outer_mobj.group('format')
914 mobj = re.match(INTERNAL_FORMAT_RE, key)
915 if mobj is None:
916 value, default = None, na
917 else:
918 mobj = mobj.groupdict()
919 default = mobj['default'] if mobj['default'] is not None else na
920 value = get_value(mobj)
921
922 if fmt == 's' and value is not None and key in field_size_compat_map.keys():
923 fmt = '0{:d}d'.format(field_size_compat_map[key])
924
925 value = default if value is None else value
926 key += '\0%s' % fmt
927
928 if fmt == 'c':
929 value = compat_str(value)
930 if value is None:
931 value, fmt = default, 's'
932 else:
933 value = value[0]
934 elif fmt[-1] not in 'rs': # numeric
935 value = float_or_none(value)
936 if value is None:
937 value, fmt = default, 's'
938 if sanitize:
939 if fmt[-1] == 'r':
940 # If value is an object, sanitize might convert it to a string
941 # So we convert it to repr first
942 value, fmt = repr(value), '%ss' % fmt[:-1]
943 if fmt[-1] in 'csr':
944 value = sanitize(key, value)
945 TMPL_DICT[key] = value
946 return '%({key}){fmt}'.format(key=key, fmt=fmt)
947
948 return EXTERNAL_FORMAT_RE.sub(create_key, outtmpl), TMPL_DICT
949
950 def _prepare_filename(self, info_dict, tmpl_type='default'):
951 try:
952 sanitize = lambda k, v: sanitize_filename(
953 compat_str(v),
954 restricted=self.params.get('restrictfilenames'),
955 is_id=(k == 'id' or k.endswith('_id')))
956 outtmpl = self.outtmpl_dict.get(tmpl_type, self.outtmpl_dict['default'])
957 outtmpl, template_dict = self.prepare_outtmpl(outtmpl, info_dict, sanitize)
958
959 # expand_path translates '%%' into '%' and '$$' into '$'
960 # correspondingly that is not what we want since we need to keep
961 # '%%' intact for template dict substitution step. Working around
962 # with boundary-alike separator hack.
963 sep = ''.join([random.choice(ascii_letters) for _ in range(32)])
964 outtmpl = outtmpl.replace('%%', '%{0}%'.format(sep)).replace('$$', '${0}$'.format(sep))
965
966 # outtmpl should be expand_path'ed before template dict substitution
967 # because meta fields may contain env variables we don't want to
968 # be expanded. For example, for outtmpl "%(title)s.%(ext)s" and
969 # title "Hello $PATH", we don't want `$PATH` to be expanded.
970 filename = expand_path(outtmpl).replace(sep, '') % template_dict
971
972 force_ext = OUTTMPL_TYPES.get(tmpl_type)
973 if force_ext is not None:
974 filename = replace_extension(filename, force_ext, info_dict.get('ext'))
975
976 # https://github.com/blackjack4494/youtube-dlc/issues/85
977 trim_file_name = self.params.get('trim_file_name', False)
978 if trim_file_name:
979 fn_groups = filename.rsplit('.')
980 ext = fn_groups[-1]
981 sub_ext = ''
982 if len(fn_groups) > 2:
983 sub_ext = fn_groups[-2]
984 filename = '.'.join(filter(None, [fn_groups[0][:trim_file_name], sub_ext, ext]))
985
986 return filename
987 except ValueError as err:
988 self.report_error('Error in output template: ' + str(err) + ' (encoding: ' + repr(preferredencoding()) + ')')
989 return None
990
991 def prepare_filename(self, info_dict, dir_type='', warn=False):
992 """Generate the output filename."""
993 paths = self.params.get('paths', {})
994 assert isinstance(paths, dict)
995 filename = self._prepare_filename(info_dict, dir_type or 'default')
996
997 if warn and not self.__prepare_filename_warned:
998 if not paths:
999 pass
1000 elif filename == '-':
1001 self.report_warning('--paths is ignored when an outputting to stdout')
1002 elif os.path.isabs(filename):
1003 self.report_warning('--paths is ignored since an absolute path is given in output template')
1004 self.__prepare_filename_warned = True
1005 if filename == '-' or not filename:
1006 return filename
1007
1008 homepath = expand_path(paths.get('home', '').strip())
1009 assert isinstance(homepath, compat_str)
1010 subdir = expand_path(paths.get(dir_type, '').strip()) if dir_type else ''
1011 assert isinstance(subdir, compat_str)
1012 path = os.path.join(homepath, subdir, filename)
1013
1014 # Temporary fix for #4787
1015 # 'Treat' all problem characters by passing filename through preferredencoding
1016 # to workaround encoding issues with subprocess on python2 @ Windows
1017 if sys.version_info < (3, 0) and sys.platform == 'win32':
1018 path = encodeFilename(path, True).decode(preferredencoding())
1019 return sanitize_path(path, force=self.params.get('windowsfilenames'))
1020
1021 def _match_entry(self, info_dict, incomplete=False, silent=False):
1022 """ Returns None if the file should be downloaded """
1023
1024 video_title = info_dict.get('title', info_dict.get('id', 'video'))
1025
1026 def check_filter():
1027 if 'title' in info_dict:
1028 # This can happen when we're just evaluating the playlist
1029 title = info_dict['title']
1030 matchtitle = self.params.get('matchtitle', False)
1031 if matchtitle:
1032 if not re.search(matchtitle, title, re.IGNORECASE):
1033 return '"' + title + '" title did not match pattern "' + matchtitle + '"'
1034 rejecttitle = self.params.get('rejecttitle', False)
1035 if rejecttitle:
1036 if re.search(rejecttitle, title, re.IGNORECASE):
1037 return '"' + title + '" title matched reject pattern "' + rejecttitle + '"'
1038 date = info_dict.get('upload_date')
1039 if date is not None:
1040 dateRange = self.params.get('daterange', DateRange())
1041 if date not in dateRange:
1042 return '%s upload date is not in range %s' % (date_from_str(date).isoformat(), dateRange)
1043 view_count = info_dict.get('view_count')
1044 if view_count is not None:
1045 min_views = self.params.get('min_views')
1046 if min_views is not None and view_count < min_views:
1047 return 'Skipping %s, because it has not reached minimum view count (%d/%d)' % (video_title, view_count, min_views)
1048 max_views = self.params.get('max_views')
1049 if max_views is not None and view_count > max_views:
1050 return 'Skipping %s, because it has exceeded the maximum view count (%d/%d)' % (video_title, view_count, max_views)
1051 if age_restricted(info_dict.get('age_limit'), self.params.get('age_limit')):
1052 return 'Skipping "%s" because it is age restricted' % video_title
1053
1054 if not incomplete:
1055 match_filter = self.params.get('match_filter')
1056 if match_filter is not None:
1057 ret = match_filter(info_dict)
1058 if ret is not None:
1059 return ret
1060 return None
1061
1062 if self.in_download_archive(info_dict):
1063 reason = '%s has already been recorded in the archive' % video_title
1064 break_opt, break_err = 'break_on_existing', ExistingVideoReached
1065 else:
1066 reason = check_filter()
1067 break_opt, break_err = 'break_on_reject', RejectedVideoReached
1068 if reason is not None:
1069 if not silent:
1070 self.to_screen('[download] ' + reason)
1071 if self.params.get(break_opt, False):
1072 raise break_err()
1073 return reason
1074
1075 @staticmethod
1076 def add_extra_info(info_dict, extra_info):
1077 '''Set the keys from extra_info in info dict if they are missing'''
1078 for key, value in extra_info.items():
1079 info_dict.setdefault(key, value)
1080
1081 def extract_info(self, url, download=True, ie_key=None, extra_info={},
1082 process=True, force_generic_extractor=False):
1083 """
1084 Return a list with a dictionary for each video extracted.
1085
1086 Arguments:
1087 url -- URL to extract
1088
1089 Keyword arguments:
1090 download -- whether to download videos during extraction
1091 ie_key -- extractor key hint
1092 extra_info -- dictionary containing the extra values to add to each result
1093 process -- whether to resolve all unresolved references (URLs, playlist items),
1094 must be True for download to work.
1095 force_generic_extractor -- force using the generic extractor
1096 """
1097
1098 if not ie_key and force_generic_extractor:
1099 ie_key = 'Generic'
1100
1101 if ie_key:
1102 ies = [self.get_info_extractor(ie_key)]
1103 else:
1104 ies = self._ies
1105
1106 for ie in ies:
1107 if not ie.suitable(url):
1108 continue
1109
1110 ie_key = ie.ie_key()
1111 ie = self.get_info_extractor(ie_key)
1112 if not ie.working():
1113 self.report_warning('The program functionality for this site has been marked as broken, '
1114 'and will probably not work.')
1115
1116 try:
1117 temp_id = str_or_none(
1118 ie.extract_id(url) if callable(getattr(ie, 'extract_id', None))
1119 else ie._match_id(url))
1120 except (AssertionError, IndexError, AttributeError):
1121 temp_id = None
1122 if temp_id is not None and self.in_download_archive({'id': temp_id, 'ie_key': ie_key}):
1123 self.to_screen("[%s] %s: has already been recorded in archive" % (
1124 ie_key, temp_id))
1125 break
1126 return self.__extract_info(url, ie, download, extra_info, process)
1127 else:
1128 self.report_error('no suitable InfoExtractor for URL %s' % url)
1129
1130 def __handle_extraction_exceptions(func):
1131 def wrapper(self, *args, **kwargs):
1132 try:
1133 return func(self, *args, **kwargs)
1134 except GeoRestrictedError as e:
1135 msg = e.msg
1136 if e.countries:
1137 msg += '\nThis video is available in %s.' % ', '.join(
1138 map(ISO3166Utils.short2full, e.countries))
1139 msg += '\nYou might want to use a VPN or a proxy server (with --proxy) to workaround.'
1140 self.report_error(msg)
1141 except ExtractorError as e: # An error we somewhat expected
1142 self.report_error(compat_str(e), e.format_traceback())
1143 except (MaxDownloadsReached, ExistingVideoReached, RejectedVideoReached):
1144 raise
1145 except Exception as e:
1146 if self.params.get('ignoreerrors', False):
1147 self.report_error(error_to_compat_str(e), tb=encode_compat_str(traceback.format_exc()))
1148 else:
1149 raise
1150 return wrapper
1151
1152 @__handle_extraction_exceptions
1153 def __extract_info(self, url, ie, download, extra_info, process):
1154 ie_result = ie.extract(url)
1155 if ie_result is None: # Finished already (backwards compatibility; listformats and friends should be moved here)
1156 return
1157 if isinstance(ie_result, list):
1158 # Backwards compatibility: old IE result format
1159 ie_result = {
1160 '_type': 'compat_list',
1161 'entries': ie_result,
1162 }
1163 self.add_default_extra_info(ie_result, ie, url)
1164 if process:
1165 return self.process_ie_result(ie_result, download, extra_info)
1166 else:
1167 return ie_result
1168
1169 def add_default_extra_info(self, ie_result, ie, url):
1170 self.add_extra_info(ie_result, {
1171 'extractor': ie.IE_NAME,
1172 'webpage_url': url,
1173 'original_url': url,
1174 'webpage_url_basename': url_basename(url),
1175 'extractor_key': ie.ie_key(),
1176 })
1177
1178 def process_ie_result(self, ie_result, download=True, extra_info={}):
1179 """
1180 Take the result of the ie(may be modified) and resolve all unresolved
1181 references (URLs, playlist items).
1182
1183 It will also download the videos if 'download'.
1184 Returns the resolved ie_result.
1185 """
1186 result_type = ie_result.get('_type', 'video')
1187
1188 if result_type in ('url', 'url_transparent'):
1189 ie_result['url'] = sanitize_url(ie_result['url'])
1190 extract_flat = self.params.get('extract_flat', False)
1191 if ((extract_flat == 'in_playlist' and 'playlist' in extra_info)
1192 or extract_flat is True):
1193 info_copy = ie_result.copy()
1194 self.add_extra_info(info_copy, extra_info)
1195 self.add_default_extra_info(
1196 info_copy, self.get_info_extractor(ie_result.get('ie_key')), ie_result['url'])
1197 self.__forced_printings(info_copy, self.prepare_filename(info_copy), incomplete=True)
1198 return ie_result
1199
1200 if result_type == 'video':
1201 self.add_extra_info(ie_result, extra_info)
1202 ie_result = self.process_video_result(ie_result, download=download)
1203 additional_urls = (ie_result or {}).get('additional_urls')
1204 if additional_urls:
1205 # TODO: Improve MetadataFromFieldPP to allow setting a list
1206 if isinstance(additional_urls, compat_str):
1207 additional_urls = [additional_urls]
1208 self.to_screen(
1209 '[info] %s: %d additional URL(s) requested' % (ie_result['id'], len(additional_urls)))
1210 self.write_debug('Additional URLs: "%s"' % '", "'.join(additional_urls))
1211 ie_result['additional_entries'] = [
1212 self.extract_info(
1213 url, download, extra_info,
1214 force_generic_extractor=self.params.get('force_generic_extractor'))
1215 for url in additional_urls
1216 ]
1217 return ie_result
1218 elif result_type == 'url':
1219 # We have to add extra_info to the results because it may be
1220 # contained in a playlist
1221 return self.extract_info(
1222 ie_result['url'], download,
1223 ie_key=ie_result.get('ie_key'),
1224 extra_info=extra_info)
1225 elif result_type == 'url_transparent':
1226 # Use the information from the embedding page
1227 info = self.extract_info(
1228 ie_result['url'], ie_key=ie_result.get('ie_key'),
1229 extra_info=extra_info, download=False, process=False)
1230
1231 # extract_info may return None when ignoreerrors is enabled and
1232 # extraction failed with an error, don't crash and return early
1233 # in this case
1234 if not info:
1235 return info
1236
1237 force_properties = dict(
1238 (k, v) for k, v in ie_result.items() if v is not None)
1239 for f in ('_type', 'url', 'id', 'extractor', 'extractor_key', 'ie_key'):
1240 if f in force_properties:
1241 del force_properties[f]
1242 new_result = info.copy()
1243 new_result.update(force_properties)
1244
1245 # Extracted info may not be a video result (i.e.
1246 # info.get('_type', 'video') != video) but rather an url or
1247 # url_transparent. In such cases outer metadata (from ie_result)
1248 # should be propagated to inner one (info). For this to happen
1249 # _type of info should be overridden with url_transparent. This
1250 # fixes issue from https://github.com/ytdl-org/youtube-dl/pull/11163.
1251 if new_result.get('_type') == 'url':
1252 new_result['_type'] = 'url_transparent'
1253
1254 return self.process_ie_result(
1255 new_result, download=download, extra_info=extra_info)
1256 elif result_type in ('playlist', 'multi_video'):
1257 # Protect from infinite recursion due to recursively nested playlists
1258 # (see https://github.com/ytdl-org/youtube-dl/issues/27833)
1259 webpage_url = ie_result['webpage_url']
1260 if webpage_url in self._playlist_urls:
1261 self.to_screen(
1262 '[download] Skipping already downloaded playlist: %s'
1263 % ie_result.get('title') or ie_result.get('id'))
1264 return
1265
1266 self._playlist_level += 1
1267 self._playlist_urls.add(webpage_url)
1268 self._sanitize_thumbnails(ie_result)
1269 try:
1270 return self.__process_playlist(ie_result, download)
1271 finally:
1272 self._playlist_level -= 1
1273 if not self._playlist_level:
1274 self._playlist_urls.clear()
1275 elif result_type == 'compat_list':
1276 self.report_warning(
1277 'Extractor %s returned a compat_list result. '
1278 'It needs to be updated.' % ie_result.get('extractor'))
1279
1280 def _fixup(r):
1281 self.add_extra_info(
1282 r,
1283 {
1284 'extractor': ie_result['extractor'],
1285 'webpage_url': ie_result['webpage_url'],
1286 'webpage_url_basename': url_basename(ie_result['webpage_url']),
1287 'extractor_key': ie_result['extractor_key'],
1288 }
1289 )
1290 return r
1291 ie_result['entries'] = [
1292 self.process_ie_result(_fixup(r), download, extra_info)
1293 for r in ie_result['entries']
1294 ]
1295 return ie_result
1296 else:
1297 raise Exception('Invalid result type: %s' % result_type)
1298
1299 def _ensure_dir_exists(self, path):
1300 return make_dir(path, self.report_error)
1301
1302 def __process_playlist(self, ie_result, download):
1303 # We process each entry in the playlist
1304 playlist = ie_result.get('title') or ie_result.get('id')
1305 self.to_screen('[download] Downloading playlist: %s' % playlist)
1306
1307 if 'entries' not in ie_result:
1308 raise EntryNotInPlaylist()
1309 incomplete_entries = bool(ie_result.get('requested_entries'))
1310 if incomplete_entries:
1311 def fill_missing_entries(entries, indexes):
1312 ret = [None] * max(*indexes)
1313 for i, entry in zip(indexes, entries):
1314 ret[i - 1] = entry
1315 return ret
1316 ie_result['entries'] = fill_missing_entries(ie_result['entries'], ie_result['requested_entries'])
1317
1318 playlist_results = []
1319
1320 playliststart = self.params.get('playliststart', 1)
1321 playlistend = self.params.get('playlistend')
1322 # For backwards compatibility, interpret -1 as whole list
1323 if playlistend == -1:
1324 playlistend = None
1325
1326 playlistitems_str = self.params.get('playlist_items')
1327 playlistitems = None
1328 if playlistitems_str is not None:
1329 def iter_playlistitems(format):
1330 for string_segment in format.split(','):
1331 if '-' in string_segment:
1332 start, end = string_segment.split('-')
1333 for item in range(int(start), int(end) + 1):
1334 yield int(item)
1335 else:
1336 yield int(string_segment)
1337 playlistitems = orderedSet(iter_playlistitems(playlistitems_str))
1338
1339 ie_entries = ie_result['entries']
1340 msg = (
1341 'Downloading %d videos' if not isinstance(ie_entries, list)
1342 else 'Collected %d videos; downloading %%d of them' % len(ie_entries))
1343 if not isinstance(ie_entries, (list, PagedList)):
1344 ie_entries = LazyList(ie_entries)
1345
1346 entries = []
1347 for i in playlistitems or itertools.count(playliststart):
1348 if playlistitems is None and playlistend is not None and playlistend < i:
1349 break
1350 entry = None
1351 try:
1352 entry = ie_entries[i - 1]
1353 if entry is None:
1354 raise EntryNotInPlaylist()
1355 except (IndexError, EntryNotInPlaylist):
1356 if incomplete_entries:
1357 raise EntryNotInPlaylist()
1358 elif not playlistitems:
1359 break
1360 entries.append(entry)
1361 try:
1362 if entry is not None:
1363 self._match_entry(entry, incomplete=True, silent=True)
1364 except (ExistingVideoReached, RejectedVideoReached):
1365 break
1366 ie_result['entries'] = entries
1367
1368 # Save playlist_index before re-ordering
1369 entries = [
1370 ((playlistitems[i - 1] if playlistitems else i), entry)
1371 for i, entry in enumerate(entries, 1)
1372 if entry is not None]
1373 n_entries = len(entries)
1374
1375 if not playlistitems and (playliststart or playlistend):
1376 playlistitems = list(range(playliststart, playliststart + n_entries))
1377 ie_result['requested_entries'] = playlistitems
1378
1379 if self.params.get('allow_playlist_files', True):
1380 ie_copy = {
1381 'playlist': playlist,
1382 'playlist_id': ie_result.get('id'),
1383 'playlist_title': ie_result.get('title'),
1384 'playlist_uploader': ie_result.get('uploader'),
1385 'playlist_uploader_id': ie_result.get('uploader_id'),
1386 'playlist_index': 0,
1387 }
1388 ie_copy.update(dict(ie_result))
1389
1390 if self.params.get('writeinfojson', False):
1391 infofn = self.prepare_filename(ie_copy, 'pl_infojson')
1392 if not self._ensure_dir_exists(encodeFilename(infofn)):
1393 return
1394 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(infofn)):
1395 self.to_screen('[info] Playlist metadata is already present')
1396 else:
1397 self.to_screen('[info] Writing playlist metadata as JSON to: ' + infofn)
1398 try:
1399 write_json_file(self.filter_requested_info(ie_result, self.params.get('clean_infojson', True)), infofn)
1400 except (OSError, IOError):
1401 self.report_error('Cannot write playlist metadata to JSON file ' + infofn)
1402
1403 # TODO: This should be passed to ThumbnailsConvertor if necessary
1404 self._write_thumbnails(ie_copy, self.prepare_filename(ie_copy, 'pl_thumbnail'))
1405
1406 if self.params.get('writedescription', False):
1407 descfn = self.prepare_filename(ie_copy, 'pl_description')
1408 if not self._ensure_dir_exists(encodeFilename(descfn)):
1409 return
1410 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(descfn)):
1411 self.to_screen('[info] Playlist description is already present')
1412 elif ie_result.get('description') is None:
1413 self.report_warning('There\'s no playlist description to write.')
1414 else:
1415 try:
1416 self.to_screen('[info] Writing playlist description to: ' + descfn)
1417 with io.open(encodeFilename(descfn), 'w', encoding='utf-8') as descfile:
1418 descfile.write(ie_result['description'])
1419 except (OSError, IOError):
1420 self.report_error('Cannot write playlist description file ' + descfn)
1421 return
1422
1423 if self.params.get('playlistreverse', False):
1424 entries = entries[::-1]
1425 if self.params.get('playlistrandom', False):
1426 random.shuffle(entries)
1427
1428 x_forwarded_for = ie_result.get('__x_forwarded_for_ip')
1429
1430 self.to_screen('[%s] playlist %s: %s' % (ie_result['extractor'], playlist, msg % n_entries))
1431 failures = 0
1432 max_failures = self.params.get('skip_playlist_after_errors') or float('inf')
1433 for i, entry_tuple in enumerate(entries, 1):
1434 playlist_index, entry = entry_tuple
1435 if 'playlist_index' in self.params.get('compat_options', []):
1436 playlist_index = playlistitems[i - 1] if playlistitems else i
1437 self.to_screen('[download] Downloading video %s of %s' % (i, n_entries))
1438 # This __x_forwarded_for_ip thing is a bit ugly but requires
1439 # minimal changes
1440 if x_forwarded_for:
1441 entry['__x_forwarded_for_ip'] = x_forwarded_for
1442 extra = {
1443 'n_entries': n_entries,
1444 '_last_playlist_index': max(playlistitems) if playlistitems else (playlistend or n_entries),
1445 'playlist_index': playlist_index,
1446 'playlist_autonumber': i,
1447 'playlist': playlist,
1448 'playlist_id': ie_result.get('id'),
1449 'playlist_title': ie_result.get('title'),
1450 'playlist_uploader': ie_result.get('uploader'),
1451 'playlist_uploader_id': ie_result.get('uploader_id'),
1452 'extractor': ie_result['extractor'],
1453 'webpage_url': ie_result['webpage_url'],
1454 'webpage_url_basename': url_basename(ie_result['webpage_url']),
1455 'extractor_key': ie_result['extractor_key'],
1456 }
1457
1458 if self._match_entry(entry, incomplete=True) is not None:
1459 continue
1460
1461 entry_result = self.__process_iterable_entry(entry, download, extra)
1462 if not entry_result:
1463 failures += 1
1464 if failures >= max_failures:
1465 self.report_error(
1466 'Skipping the remaining entries in playlist "%s" since %d items failed extraction' % (playlist, failures))
1467 break
1468 # TODO: skip failed (empty) entries?
1469 playlist_results.append(entry_result)
1470 ie_result['entries'] = playlist_results
1471 self.to_screen('[download] Finished downloading playlist: %s' % playlist)
1472 return ie_result
1473
1474 @__handle_extraction_exceptions
1475 def __process_iterable_entry(self, entry, download, extra_info):
1476 return self.process_ie_result(
1477 entry, download=download, extra_info=extra_info)
1478
1479 def _build_format_filter(self, filter_spec):
1480 " Returns a function to filter the formats according to the filter_spec "
1481
1482 OPERATORS = {
1483 '<': operator.lt,
1484 '<=': operator.le,
1485 '>': operator.gt,
1486 '>=': operator.ge,
1487 '=': operator.eq,
1488 '!=': operator.ne,
1489 }
1490 operator_rex = re.compile(r'''(?x)\s*
1491 (?P<key>width|height|tbr|abr|vbr|asr|filesize|filesize_approx|fps)
1492 \s*(?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
1493 (?P<value>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?)
1494 $
1495 ''' % '|'.join(map(re.escape, OPERATORS.keys())))
1496 m = operator_rex.search(filter_spec)
1497 if m:
1498 try:
1499 comparison_value = int(m.group('value'))
1500 except ValueError:
1501 comparison_value = parse_filesize(m.group('value'))
1502 if comparison_value is None:
1503 comparison_value = parse_filesize(m.group('value') + 'B')
1504 if comparison_value is None:
1505 raise ValueError(
1506 'Invalid value %r in format specification %r' % (
1507 m.group('value'), filter_spec))
1508 op = OPERATORS[m.group('op')]
1509
1510 if not m:
1511 STR_OPERATORS = {
1512 '=': operator.eq,
1513 '^=': lambda attr, value: attr.startswith(value),
1514 '$=': lambda attr, value: attr.endswith(value),
1515 '*=': lambda attr, value: value in attr,
1516 }
1517 str_operator_rex = re.compile(r'''(?x)
1518 \s*(?P<key>[a-zA-Z0-9._-]+)
1519 \s*(?P<negation>!\s*)?(?P<op>%s)(?P<none_inclusive>\s*\?)?
1520 \s*(?P<value>[a-zA-Z0-9._-]+)
1521 \s*$
1522 ''' % '|'.join(map(re.escape, STR_OPERATORS.keys())))
1523 m = str_operator_rex.search(filter_spec)
1524 if m:
1525 comparison_value = m.group('value')
1526 str_op = STR_OPERATORS[m.group('op')]
1527 if m.group('negation'):
1528 op = lambda attr, value: not str_op(attr, value)
1529 else:
1530 op = str_op
1531
1532 if not m:
1533 raise ValueError('Invalid filter specification %r' % filter_spec)
1534
1535 def _filter(f):
1536 actual_value = f.get(m.group('key'))
1537 if actual_value is None:
1538 return m.group('none_inclusive')
1539 return op(actual_value, comparison_value)
1540 return _filter
1541
1542 def _default_format_spec(self, info_dict, download=True):
1543
1544 def can_merge():
1545 merger = FFmpegMergerPP(self)
1546 return merger.available and merger.can_merge()
1547
1548 prefer_best = (
1549 not self.params.get('simulate', False)
1550 and download
1551 and (
1552 not can_merge()
1553 or info_dict.get('is_live', False)
1554 or self.outtmpl_dict['default'] == '-'))
1555 compat = (
1556 prefer_best
1557 or self.params.get('allow_multiple_audio_streams', False)
1558 or 'format-spec' in self.params.get('compat_opts', []))
1559
1560 return (
1561 'best/bestvideo+bestaudio' if prefer_best
1562 else 'bestvideo*+bestaudio/best' if not compat
1563 else 'bestvideo+bestaudio/best')
1564
1565 def build_format_selector(self, format_spec):
1566 def syntax_error(note, start):
1567 message = (
1568 'Invalid format specification: '
1569 '{0}\n\t{1}\n\t{2}^'.format(note, format_spec, ' ' * start[1]))
1570 return SyntaxError(message)
1571
1572 PICKFIRST = 'PICKFIRST'
1573 MERGE = 'MERGE'
1574 SINGLE = 'SINGLE'
1575 GROUP = 'GROUP'
1576 FormatSelector = collections.namedtuple('FormatSelector', ['type', 'selector', 'filters'])
1577
1578 allow_multiple_streams = {'audio': self.params.get('allow_multiple_audio_streams', False),
1579 'video': self.params.get('allow_multiple_video_streams', False)}
1580
1581 check_formats = self.params.get('check_formats')
1582
1583 def _parse_filter(tokens):
1584 filter_parts = []
1585 for type, string, start, _, _ in tokens:
1586 if type == tokenize.OP and string == ']':
1587 return ''.join(filter_parts)
1588 else:
1589 filter_parts.append(string)
1590
1591 def _remove_unused_ops(tokens):
1592 # Remove operators that we don't use and join them with the surrounding strings
1593 # for example: 'mp4' '-' 'baseline' '-' '16x9' is converted to 'mp4-baseline-16x9'
1594 ALLOWED_OPS = ('/', '+', ',', '(', ')')
1595 last_string, last_start, last_end, last_line = None, None, None, None
1596 for type, string, start, end, line in tokens:
1597 if type == tokenize.OP and string == '[':
1598 if last_string:
1599 yield tokenize.NAME, last_string, last_start, last_end, last_line
1600 last_string = None
1601 yield type, string, start, end, line
1602 # everything inside brackets will be handled by _parse_filter
1603 for type, string, start, end, line in tokens:
1604 yield type, string, start, end, line
1605 if type == tokenize.OP and string == ']':
1606 break
1607 elif type == tokenize.OP and string in ALLOWED_OPS:
1608 if last_string:
1609 yield tokenize.NAME, last_string, last_start, last_end, last_line
1610 last_string = None
1611 yield type, string, start, end, line
1612 elif type in [tokenize.NAME, tokenize.NUMBER, tokenize.OP]:
1613 if not last_string:
1614 last_string = string
1615 last_start = start
1616 last_end = end
1617 else:
1618 last_string += string
1619 if last_string:
1620 yield tokenize.NAME, last_string, last_start, last_end, last_line
1621
1622 def _parse_format_selection(tokens, inside_merge=False, inside_choice=False, inside_group=False):
1623 selectors = []
1624 current_selector = None
1625 for type, string, start, _, _ in tokens:
1626 # ENCODING is only defined in python 3.x
1627 if type == getattr(tokenize, 'ENCODING', None):
1628 continue
1629 elif type in [tokenize.NAME, tokenize.NUMBER]:
1630 current_selector = FormatSelector(SINGLE, string, [])
1631 elif type == tokenize.OP:
1632 if string == ')':
1633 if not inside_group:
1634 # ')' will be handled by the parentheses group
1635 tokens.restore_last_token()
1636 break
1637 elif inside_merge and string in ['/', ',']:
1638 tokens.restore_last_token()
1639 break
1640 elif inside_choice and string == ',':
1641 tokens.restore_last_token()
1642 break
1643 elif string == ',':
1644 if not current_selector:
1645 raise syntax_error('"," must follow a format selector', start)
1646 selectors.append(current_selector)
1647 current_selector = None
1648 elif string == '/':
1649 if not current_selector:
1650 raise syntax_error('"/" must follow a format selector', start)
1651 first_choice = current_selector
1652 second_choice = _parse_format_selection(tokens, inside_choice=True)
1653 current_selector = FormatSelector(PICKFIRST, (first_choice, second_choice), [])
1654 elif string == '[':
1655 if not current_selector:
1656 current_selector = FormatSelector(SINGLE, 'best', [])
1657 format_filter = _parse_filter(tokens)
1658 current_selector.filters.append(format_filter)
1659 elif string == '(':
1660 if current_selector:
1661 raise syntax_error('Unexpected "("', start)
1662 group = _parse_format_selection(tokens, inside_group=True)
1663 current_selector = FormatSelector(GROUP, group, [])
1664 elif string == '+':
1665 if not current_selector:
1666 raise syntax_error('Unexpected "+"', start)
1667 selector_1 = current_selector
1668 selector_2 = _parse_format_selection(tokens, inside_merge=True)
1669 if not selector_2:
1670 raise syntax_error('Expected a selector', start)
1671 current_selector = FormatSelector(MERGE, (selector_1, selector_2), [])
1672 else:
1673 raise syntax_error('Operator not recognized: "{0}"'.format(string), start)
1674 elif type == tokenize.ENDMARKER:
1675 break
1676 if current_selector:
1677 selectors.append(current_selector)
1678 return selectors
1679
1680 def _merge(formats_pair):
1681 format_1, format_2 = formats_pair
1682
1683 formats_info = []
1684 formats_info.extend(format_1.get('requested_formats', (format_1,)))
1685 formats_info.extend(format_2.get('requested_formats', (format_2,)))
1686
1687 if not allow_multiple_streams['video'] or not allow_multiple_streams['audio']:
1688 get_no_more = {"video": False, "audio": False}
1689 for (i, fmt_info) in enumerate(formats_info):
1690 for aud_vid in ["audio", "video"]:
1691 if not allow_multiple_streams[aud_vid] and fmt_info.get(aud_vid[0] + 'codec') != 'none':
1692 if get_no_more[aud_vid]:
1693 formats_info.pop(i)
1694 get_no_more[aud_vid] = True
1695
1696 if len(formats_info) == 1:
1697 return formats_info[0]
1698
1699 video_fmts = [fmt_info for fmt_info in formats_info if fmt_info.get('vcodec') != 'none']
1700 audio_fmts = [fmt_info for fmt_info in formats_info if fmt_info.get('acodec') != 'none']
1701
1702 the_only_video = video_fmts[0] if len(video_fmts) == 1 else None
1703 the_only_audio = audio_fmts[0] if len(audio_fmts) == 1 else None
1704
1705 output_ext = self.params.get('merge_output_format')
1706 if not output_ext:
1707 if the_only_video:
1708 output_ext = the_only_video['ext']
1709 elif the_only_audio and not video_fmts:
1710 output_ext = the_only_audio['ext']
1711 else:
1712 output_ext = 'mkv'
1713
1714 new_dict = {
1715 'requested_formats': formats_info,
1716 'format': '+'.join(fmt_info.get('format') for fmt_info in formats_info),
1717 'format_id': '+'.join(fmt_info.get('format_id') for fmt_info in formats_info),
1718 'ext': output_ext,
1719 }
1720
1721 if the_only_video:
1722 new_dict.update({
1723 'width': the_only_video.get('width'),
1724 'height': the_only_video.get('height'),
1725 'resolution': the_only_video.get('resolution') or self.format_resolution(the_only_video),
1726 'fps': the_only_video.get('fps'),
1727 'vcodec': the_only_video.get('vcodec'),
1728 'vbr': the_only_video.get('vbr'),
1729 'stretched_ratio': the_only_video.get('stretched_ratio'),
1730 })
1731
1732 if the_only_audio:
1733 new_dict.update({
1734 'acodec': the_only_audio.get('acodec'),
1735 'abr': the_only_audio.get('abr'),
1736 })
1737
1738 return new_dict
1739
1740 def _check_formats(formats):
1741 for f in formats:
1742 self.to_screen('[info] Testing format %s' % f['format_id'])
1743 paths = self.params.get('paths', {})
1744 temp_file = os.path.join(
1745 expand_path(paths.get('home', '').strip()),
1746 expand_path(paths.get('temp', '').strip()),
1747 'ytdl.%s.f%s.check-format' % (random_uuidv4(), f['format_id']))
1748 try:
1749 dl, _ = self.dl(temp_file, f, test=True)
1750 except (ExtractorError, IOError, OSError, ValueError) + network_exceptions:
1751 dl = False
1752 finally:
1753 if os.path.exists(temp_file):
1754 os.remove(temp_file)
1755 if dl:
1756 yield f
1757 else:
1758 self.to_screen('[info] Unable to download format %s. Skipping...' % f['format_id'])
1759
1760 def _build_selector_function(selector):
1761 if isinstance(selector, list): # ,
1762 fs = [_build_selector_function(s) for s in selector]
1763
1764 def selector_function(ctx):
1765 for f in fs:
1766 for format in f(ctx):
1767 yield format
1768 return selector_function
1769
1770 elif selector.type == GROUP: # ()
1771 selector_function = _build_selector_function(selector.selector)
1772
1773 elif selector.type == PICKFIRST: # /
1774 fs = [_build_selector_function(s) for s in selector.selector]
1775
1776 def selector_function(ctx):
1777 for f in fs:
1778 picked_formats = list(f(ctx))
1779 if picked_formats:
1780 return picked_formats
1781 return []
1782
1783 elif selector.type == SINGLE: # atom
1784 format_spec = selector.selector or 'best'
1785
1786 # TODO: Add allvideo, allaudio etc by generalizing the code with best/worst selector
1787 if format_spec == 'all':
1788 def selector_function(ctx):
1789 formats = list(ctx['formats'])
1790 if check_formats:
1791 formats = _check_formats(formats)
1792 for f in formats:
1793 yield f
1794 elif format_spec == 'mergeall':
1795 def selector_function(ctx):
1796 formats = list(_check_formats(ctx['formats']))
1797 if not formats:
1798 return
1799 merged_format = formats[-1]
1800 for f in formats[-2::-1]:
1801 merged_format = _merge((merged_format, f))
1802 yield merged_format
1803
1804 else:
1805 format_fallback, format_reverse, format_idx = False, True, 1
1806 mobj = re.match(
1807 r'(?P<bw>best|worst|b|w)(?P<type>video|audio|v|a)?(?P<mod>\*)?(?:\.(?P<n>[1-9]\d*))?$',
1808 format_spec)
1809 if mobj is not None:
1810 format_idx = int_or_none(mobj.group('n'), default=1)
1811 format_reverse = mobj.group('bw')[0] == 'b'
1812 format_type = (mobj.group('type') or [None])[0]
1813 not_format_type = {'v': 'a', 'a': 'v'}.get(format_type)
1814 format_modified = mobj.group('mod') is not None
1815
1816 format_fallback = not format_type and not format_modified # for b, w
1817 filter_f = (
1818 (lambda f: f.get('%scodec' % format_type) != 'none')
1819 if format_type and format_modified # bv*, ba*, wv*, wa*
1820 else (lambda f: f.get('%scodec' % not_format_type) == 'none')
1821 if format_type # bv, ba, wv, wa
1822 else (lambda f: f.get('vcodec') != 'none' and f.get('acodec') != 'none')
1823 if not format_modified # b, w
1824 else None) # b*, w*
1825 else:
1826 filter_f = ((lambda f: f.get('ext') == format_spec)
1827 if format_spec in ['mp4', 'flv', 'webm', '3gp', 'm4a', 'mp3', 'ogg', 'aac', 'wav'] # extension
1828 else (lambda f: f.get('format_id') == format_spec)) # id
1829
1830 def selector_function(ctx):
1831 formats = list(ctx['formats'])
1832 if not formats:
1833 return
1834 matches = list(filter(filter_f, formats)) if filter_f is not None else formats
1835 if format_fallback and ctx['incomplete_formats'] and not matches:
1836 # for extractors with incomplete formats (audio only (soundcloud)
1837 # or video only (imgur)) best/worst will fallback to
1838 # best/worst {video,audio}-only format
1839 matches = formats
1840 if format_reverse:
1841 matches = matches[::-1]
1842 if check_formats:
1843 matches = list(itertools.islice(_check_formats(matches), format_idx))
1844 n = len(matches)
1845 if -n <= format_idx - 1 < n:
1846 yield matches[format_idx - 1]
1847
1848 elif selector.type == MERGE: # +
1849 selector_1, selector_2 = map(_build_selector_function, selector.selector)
1850
1851 def selector_function(ctx):
1852 for pair in itertools.product(
1853 selector_1(copy.deepcopy(ctx)), selector_2(copy.deepcopy(ctx))):
1854 yield _merge(pair)
1855
1856 filters = [self._build_format_filter(f) for f in selector.filters]
1857
1858 def final_selector(ctx):
1859 ctx_copy = copy.deepcopy(ctx)
1860 for _filter in filters:
1861 ctx_copy['formats'] = list(filter(_filter, ctx_copy['formats']))
1862 return selector_function(ctx_copy)
1863 return final_selector
1864
1865 stream = io.BytesIO(format_spec.encode('utf-8'))
1866 try:
1867 tokens = list(_remove_unused_ops(compat_tokenize_tokenize(stream.readline)))
1868 except tokenize.TokenError:
1869 raise syntax_error('Missing closing/opening brackets or parenthesis', (0, len(format_spec)))
1870
1871 class TokenIterator(object):
1872 def __init__(self, tokens):
1873 self.tokens = tokens
1874 self.counter = 0
1875
1876 def __iter__(self):
1877 return self
1878
1879 def __next__(self):
1880 if self.counter >= len(self.tokens):
1881 raise StopIteration()
1882 value = self.tokens[self.counter]
1883 self.counter += 1
1884 return value
1885
1886 next = __next__
1887
1888 def restore_last_token(self):
1889 self.counter -= 1
1890
1891 parsed_selector = _parse_format_selection(iter(TokenIterator(tokens)))
1892 return _build_selector_function(parsed_selector)
1893
1894 def _calc_headers(self, info_dict):
1895 res = std_headers.copy()
1896
1897 add_headers = info_dict.get('http_headers')
1898 if add_headers:
1899 res.update(add_headers)
1900
1901 cookies = self._calc_cookies(info_dict)
1902 if cookies:
1903 res['Cookie'] = cookies
1904
1905 if 'X-Forwarded-For' not in res:
1906 x_forwarded_for_ip = info_dict.get('__x_forwarded_for_ip')
1907 if x_forwarded_for_ip:
1908 res['X-Forwarded-For'] = x_forwarded_for_ip
1909
1910 return res
1911
1912 def _calc_cookies(self, info_dict):
1913 pr = sanitized_Request(info_dict['url'])
1914 self.cookiejar.add_cookie_header(pr)
1915 return pr.get_header('Cookie')
1916
1917 @staticmethod
1918 def _sanitize_thumbnails(info_dict):
1919 thumbnails = info_dict.get('thumbnails')
1920 if thumbnails is None:
1921 thumbnail = info_dict.get('thumbnail')
1922 if thumbnail:
1923 info_dict['thumbnails'] = thumbnails = [{'url': thumbnail}]
1924 if thumbnails:
1925 thumbnails.sort(key=lambda t: (
1926 t.get('preference') if t.get('preference') is not None else -1,
1927 t.get('width') if t.get('width') is not None else -1,
1928 t.get('height') if t.get('height') is not None else -1,
1929 t.get('id') if t.get('id') is not None else '',
1930 t.get('url')))
1931 for i, t in enumerate(thumbnails):
1932 t['url'] = sanitize_url(t['url'])
1933 if t.get('width') and t.get('height'):
1934 t['resolution'] = '%dx%d' % (t['width'], t['height'])
1935 if t.get('id') is None:
1936 t['id'] = '%d' % i
1937
1938 def process_video_result(self, info_dict, download=True):
1939 assert info_dict.get('_type', 'video') == 'video'
1940
1941 if 'id' not in info_dict:
1942 raise ExtractorError('Missing "id" field in extractor result')
1943 if 'title' not in info_dict:
1944 raise ExtractorError('Missing "title" field in extractor result')
1945
1946 def report_force_conversion(field, field_not, conversion):
1947 self.report_warning(
1948 '"%s" field is not %s - forcing %s conversion, there is an error in extractor'
1949 % (field, field_not, conversion))
1950
1951 def sanitize_string_field(info, string_field):
1952 field = info.get(string_field)
1953 if field is None or isinstance(field, compat_str):
1954 return
1955 report_force_conversion(string_field, 'a string', 'string')
1956 info[string_field] = compat_str(field)
1957
1958 def sanitize_numeric_fields(info):
1959 for numeric_field in self._NUMERIC_FIELDS:
1960 field = info.get(numeric_field)
1961 if field is None or isinstance(field, compat_numeric_types):
1962 continue
1963 report_force_conversion(numeric_field, 'numeric', 'int')
1964 info[numeric_field] = int_or_none(field)
1965
1966 sanitize_string_field(info_dict, 'id')
1967 sanitize_numeric_fields(info_dict)
1968
1969 if 'playlist' not in info_dict:
1970 # It isn't part of a playlist
1971 info_dict['playlist'] = None
1972 info_dict['playlist_index'] = None
1973
1974 self._sanitize_thumbnails(info_dict)
1975
1976 if self.params.get('list_thumbnails'):
1977 self.list_thumbnails(info_dict)
1978 return
1979
1980 thumbnail = info_dict.get('thumbnail')
1981 thumbnails = info_dict.get('thumbnails')
1982 if thumbnail:
1983 info_dict['thumbnail'] = sanitize_url(thumbnail)
1984 elif thumbnails:
1985 info_dict['thumbnail'] = thumbnails[-1]['url']
1986
1987 if 'display_id' not in info_dict and 'id' in info_dict:
1988 info_dict['display_id'] = info_dict['id']
1989
1990 for ts_key, date_key in (
1991 ('timestamp', 'upload_date'),
1992 ('release_timestamp', 'release_date'),
1993 ):
1994 if info_dict.get(date_key) is None and info_dict.get(ts_key) is not None:
1995 # Working around out-of-range timestamp values (e.g. negative ones on Windows,
1996 # see http://bugs.python.org/issue1646728)
1997 try:
1998 upload_date = datetime.datetime.utcfromtimestamp(info_dict[ts_key])
1999 info_dict[date_key] = upload_date.strftime('%Y%m%d')
2000 except (ValueError, OverflowError, OSError):
2001 pass
2002
2003 # Auto generate title fields corresponding to the *_number fields when missing
2004 # in order to always have clean titles. This is very common for TV series.
2005 for field in ('chapter', 'season', 'episode'):
2006 if info_dict.get('%s_number' % field) is not None and not info_dict.get(field):
2007 info_dict[field] = '%s %d' % (field.capitalize(), info_dict['%s_number' % field])
2008
2009 for cc_kind in ('subtitles', 'automatic_captions'):
2010 cc = info_dict.get(cc_kind)
2011 if cc:
2012 for _, subtitle in cc.items():
2013 for subtitle_format in subtitle:
2014 if subtitle_format.get('url'):
2015 subtitle_format['url'] = sanitize_url(subtitle_format['url'])
2016 if subtitle_format.get('ext') is None:
2017 subtitle_format['ext'] = determine_ext(subtitle_format['url']).lower()
2018
2019 automatic_captions = info_dict.get('automatic_captions')
2020 subtitles = info_dict.get('subtitles')
2021
2022 if self.params.get('listsubtitles', False):
2023 if 'automatic_captions' in info_dict:
2024 self.list_subtitles(
2025 info_dict['id'], automatic_captions, 'automatic captions')
2026 self.list_subtitles(info_dict['id'], subtitles, 'subtitles')
2027 return
2028
2029 info_dict['requested_subtitles'] = self.process_subtitles(
2030 info_dict['id'], subtitles, automatic_captions)
2031
2032 # We now pick which formats have to be downloaded
2033 if info_dict.get('formats') is None:
2034 # There's only one format available
2035 formats = [info_dict]
2036 else:
2037 formats = info_dict['formats']
2038
2039 if not formats:
2040 if not self.params.get('ignore_no_formats_error'):
2041 raise ExtractorError('No video formats found!')
2042 else:
2043 self.report_warning('No video formats found!')
2044
2045 def is_wellformed(f):
2046 url = f.get('url')
2047 if not url:
2048 self.report_warning(
2049 '"url" field is missing or empty - skipping format, '
2050 'there is an error in extractor')
2051 return False
2052 if isinstance(url, bytes):
2053 sanitize_string_field(f, 'url')
2054 return True
2055
2056 # Filter out malformed formats for better extraction robustness
2057 formats = list(filter(is_wellformed, formats))
2058
2059 formats_dict = {}
2060
2061 # We check that all the formats have the format and format_id fields
2062 for i, format in enumerate(formats):
2063 sanitize_string_field(format, 'format_id')
2064 sanitize_numeric_fields(format)
2065 format['url'] = sanitize_url(format['url'])
2066 if not format.get('format_id'):
2067 format['format_id'] = compat_str(i)
2068 else:
2069 # Sanitize format_id from characters used in format selector expression
2070 format['format_id'] = re.sub(r'[\s,/+\[\]()]', '_', format['format_id'])
2071 format_id = format['format_id']
2072 if format_id not in formats_dict:
2073 formats_dict[format_id] = []
2074 formats_dict[format_id].append(format)
2075
2076 # Make sure all formats have unique format_id
2077 for format_id, ambiguous_formats in formats_dict.items():
2078 if len(ambiguous_formats) > 1:
2079 for i, format in enumerate(ambiguous_formats):
2080 format['format_id'] = '%s-%d' % (format_id, i)
2081
2082 for i, format in enumerate(formats):
2083 if format.get('format') is None:
2084 format['format'] = '{id} - {res}{note}'.format(
2085 id=format['format_id'],
2086 res=self.format_resolution(format),
2087 note=' ({0})'.format(format['format_note']) if format.get('format_note') is not None else '',
2088 )
2089 # Automatically determine file extension if missing
2090 if format.get('ext') is None:
2091 format['ext'] = determine_ext(format['url']).lower()
2092 # Automatically determine protocol if missing (useful for format
2093 # selection purposes)
2094 if format.get('protocol') is None:
2095 format['protocol'] = determine_protocol(format)
2096 # Add HTTP headers, so that external programs can use them from the
2097 # json output
2098 full_format_info = info_dict.copy()
2099 full_format_info.update(format)
2100 format['http_headers'] = self._calc_headers(full_format_info)
2101 # Remove private housekeeping stuff
2102 if '__x_forwarded_for_ip' in info_dict:
2103 del info_dict['__x_forwarded_for_ip']
2104
2105 # TODO Central sorting goes here
2106
2107 if formats and formats[0] is not info_dict:
2108 # only set the 'formats' fields if the original info_dict list them
2109 # otherwise we end up with a circular reference, the first (and unique)
2110 # element in the 'formats' field in info_dict is info_dict itself,
2111 # which can't be exported to json
2112 info_dict['formats'] = formats
2113
2114 info_dict, _ = self.pre_process(info_dict)
2115
2116 if self.params.get('listformats'):
2117 if not info_dict.get('formats'):
2118 raise ExtractorError('No video formats found', expected=True)
2119 self.list_formats(info_dict)
2120 return
2121
2122 req_format = self.params.get('format')
2123 if req_format is None:
2124 req_format = self._default_format_spec(info_dict, download=download)
2125 self.write_debug('Default format spec: %s' % req_format)
2126
2127 format_selector = self.build_format_selector(req_format)
2128
2129 # While in format selection we may need to have an access to the original
2130 # format set in order to calculate some metrics or do some processing.
2131 # For now we need to be able to guess whether original formats provided
2132 # by extractor are incomplete or not (i.e. whether extractor provides only
2133 # video-only or audio-only formats) for proper formats selection for
2134 # extractors with such incomplete formats (see
2135 # https://github.com/ytdl-org/youtube-dl/pull/5556).
2136 # Since formats may be filtered during format selection and may not match
2137 # the original formats the results may be incorrect. Thus original formats
2138 # or pre-calculated metrics should be passed to format selection routines
2139 # as well.
2140 # We will pass a context object containing all necessary additional data
2141 # instead of just formats.
2142 # This fixes incorrect format selection issue (see
2143 # https://github.com/ytdl-org/youtube-dl/issues/10083).
2144 incomplete_formats = (
2145 # All formats are video-only or
2146 all(f.get('vcodec') != 'none' and f.get('acodec') == 'none' for f in formats)
2147 # all formats are audio-only
2148 or all(f.get('vcodec') == 'none' and f.get('acodec') != 'none' for f in formats))
2149
2150 ctx = {
2151 'formats': formats,
2152 'incomplete_formats': incomplete_formats,
2153 }
2154
2155 formats_to_download = list(format_selector(ctx))
2156 if not formats_to_download:
2157 if not self.params.get('ignore_no_formats_error'):
2158 raise ExtractorError('Requested format is not available', expected=True)
2159 else:
2160 self.report_warning('Requested format is not available')
2161 elif download:
2162 self.to_screen(
2163 '[info] %s: Downloading %d format(s): %s' % (
2164 info_dict['id'], len(formats_to_download),
2165 ", ".join([f['format_id'] for f in formats_to_download])))
2166 for fmt in formats_to_download:
2167 new_info = dict(info_dict)
2168 # Save a reference to the original info_dict so that it can be modified in process_info if needed
2169 new_info['__original_infodict'] = info_dict
2170 new_info.update(fmt)
2171 self.process_info(new_info)
2172 # We update the info dict with the best quality format (backwards compatibility)
2173 if formats_to_download:
2174 info_dict.update(formats_to_download[-1])
2175 return info_dict
2176
2177 def process_subtitles(self, video_id, normal_subtitles, automatic_captions):
2178 """Select the requested subtitles and their format"""
2179 available_subs = {}
2180 if normal_subtitles and self.params.get('writesubtitles'):
2181 available_subs.update(normal_subtitles)
2182 if automatic_captions and self.params.get('writeautomaticsub'):
2183 for lang, cap_info in automatic_captions.items():
2184 if lang not in available_subs:
2185 available_subs[lang] = cap_info
2186
2187 if (not self.params.get('writesubtitles') and not
2188 self.params.get('writeautomaticsub') or not
2189 available_subs):
2190 return None
2191
2192 all_sub_langs = available_subs.keys()
2193 if self.params.get('allsubtitles', False):
2194 requested_langs = all_sub_langs
2195 elif self.params.get('subtitleslangs', False):
2196 requested_langs = set()
2197 for lang in self.params.get('subtitleslangs'):
2198 if lang == 'all':
2199 requested_langs.update(all_sub_langs)
2200 continue
2201 discard = lang[0] == '-'
2202 if discard:
2203 lang = lang[1:]
2204 current_langs = filter(re.compile(lang + '$').match, all_sub_langs)
2205 if discard:
2206 for lang in current_langs:
2207 requested_langs.discard(lang)
2208 else:
2209 requested_langs.update(current_langs)
2210 elif 'en' in available_subs:
2211 requested_langs = ['en']
2212 else:
2213 requested_langs = [list(all_sub_langs)[0]]
2214 self.write_debug('Downloading subtitles: %s' % ', '.join(requested_langs))
2215
2216 formats_query = self.params.get('subtitlesformat', 'best')
2217 formats_preference = formats_query.split('/') if formats_query else []
2218 subs = {}
2219 for lang in requested_langs:
2220 formats = available_subs.get(lang)
2221 if formats is None:
2222 self.report_warning('%s subtitles not available for %s' % (lang, video_id))
2223 continue
2224 for ext in formats_preference:
2225 if ext == 'best':
2226 f = formats[-1]
2227 break
2228 matches = list(filter(lambda f: f['ext'] == ext, formats))
2229 if matches:
2230 f = matches[-1]
2231 break
2232 else:
2233 f = formats[-1]
2234 self.report_warning(
2235 'No subtitle format found matching "%s" for language %s, '
2236 'using %s' % (formats_query, lang, f['ext']))
2237 subs[lang] = f
2238 return subs
2239
2240 def __forced_printings(self, info_dict, filename, incomplete):
2241 def print_mandatory(field, actual_field=None):
2242 if actual_field is None:
2243 actual_field = field
2244 if (self.params.get('force%s' % field, False)
2245 and (not incomplete or info_dict.get(actual_field) is not None)):
2246 self.to_stdout(info_dict[actual_field])
2247
2248 def print_optional(field):
2249 if (self.params.get('force%s' % field, False)
2250 and info_dict.get(field) is not None):
2251 self.to_stdout(info_dict[field])
2252
2253 info_dict = info_dict.copy()
2254 if filename is not None:
2255 info_dict['filename'] = filename
2256 if info_dict.get('requested_formats') is not None:
2257 # For RTMP URLs, also include the playpath
2258 info_dict['urls'] = '\n'.join(f['url'] + f.get('play_path', '') for f in info_dict['requested_formats'])
2259 elif 'url' in info_dict:
2260 info_dict['urls'] = info_dict['url'] + info_dict.get('play_path', '')
2261
2262 for tmpl in self.params.get('forceprint', []):
2263 if re.match(r'\w+$', tmpl):
2264 tmpl = '%({})s'.format(tmpl)
2265 tmpl, info_copy = self.prepare_outtmpl(tmpl, info_dict)
2266 self.to_stdout(tmpl % info_copy)
2267
2268 print_mandatory('title')
2269 print_mandatory('id')
2270 print_mandatory('url', 'urls')
2271 print_optional('thumbnail')
2272 print_optional('description')
2273 print_optional('filename')
2274 if self.params.get('forceduration', False) and info_dict.get('duration') is not None:
2275 self.to_stdout(formatSeconds(info_dict['duration']))
2276 print_mandatory('format')
2277
2278 if self.params.get('forcejson', False):
2279 self.post_extract(info_dict)
2280 self.to_stdout(json.dumps(info_dict, default=repr))
2281
2282 def dl(self, name, info, subtitle=False, test=False):
2283
2284 if test:
2285 verbose = self.params.get('verbose')
2286 params = {
2287 'test': True,
2288 'quiet': not verbose,
2289 'verbose': verbose,
2290 'noprogress': not verbose,
2291 'nopart': True,
2292 'skip_unavailable_fragments': False,
2293 'keep_fragments': False,
2294 'overwrites': True,
2295 '_no_ytdl_file': True,
2296 }
2297 else:
2298 params = self.params
2299 fd = get_suitable_downloader(info, params)(self, params)
2300 if not test:
2301 for ph in self._progress_hooks:
2302 fd.add_progress_hook(ph)
2303 urls = '", "'.join([f['url'] for f in info.get('requested_formats', [])] or [info['url']])
2304 self.write_debug('Invoking downloader on "%s"' % urls)
2305 new_info = dict(info)
2306 if new_info.get('http_headers') is None:
2307 new_info['http_headers'] = self._calc_headers(new_info)
2308 return fd.download(name, new_info, subtitle)
2309
2310 def process_info(self, info_dict):
2311 """Process a single resolved IE result."""
2312
2313 assert info_dict.get('_type', 'video') == 'video'
2314
2315 info_dict.setdefault('__postprocessors', [])
2316
2317 max_downloads = self.params.get('max_downloads')
2318 if max_downloads is not None:
2319 if self._num_downloads >= int(max_downloads):
2320 raise MaxDownloadsReached()
2321
2322 # TODO: backward compatibility, to be removed
2323 info_dict['fulltitle'] = info_dict['title']
2324
2325 if 'format' not in info_dict:
2326 info_dict['format'] = info_dict['ext']
2327
2328 if self._match_entry(info_dict) is not None:
2329 return
2330
2331 self.post_extract(info_dict)
2332 self._num_downloads += 1
2333
2334 # info_dict['_filename'] needs to be set for backward compatibility
2335 info_dict['_filename'] = full_filename = self.prepare_filename(info_dict, warn=True)
2336 temp_filename = self.prepare_filename(info_dict, 'temp')
2337 files_to_move = {}
2338
2339 # Forced printings
2340 self.__forced_printings(info_dict, full_filename, incomplete=False)
2341
2342 if self.params.get('simulate', False):
2343 if self.params.get('force_write_download_archive', False):
2344 self.record_download_archive(info_dict)
2345
2346 # Do nothing else if in simulate mode
2347 return
2348
2349 if full_filename is None:
2350 return
2351
2352 if not self._ensure_dir_exists(encodeFilename(full_filename)):
2353 return
2354 if not self._ensure_dir_exists(encodeFilename(temp_filename)):
2355 return
2356
2357 if self.params.get('writedescription', False):
2358 descfn = self.prepare_filename(info_dict, 'description')
2359 if not self._ensure_dir_exists(encodeFilename(descfn)):
2360 return
2361 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(descfn)):
2362 self.to_screen('[info] Video description is already present')
2363 elif info_dict.get('description') is None:
2364 self.report_warning('There\'s no description to write.')
2365 else:
2366 try:
2367 self.to_screen('[info] Writing video description to: ' + descfn)
2368 with io.open(encodeFilename(descfn), 'w', encoding='utf-8') as descfile:
2369 descfile.write(info_dict['description'])
2370 except (OSError, IOError):
2371 self.report_error('Cannot write description file ' + descfn)
2372 return
2373
2374 if self.params.get('writeannotations', False):
2375 annofn = self.prepare_filename(info_dict, 'annotation')
2376 if not self._ensure_dir_exists(encodeFilename(annofn)):
2377 return
2378 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(annofn)):
2379 self.to_screen('[info] Video annotations are already present')
2380 elif not info_dict.get('annotations'):
2381 self.report_warning('There are no annotations to write.')
2382 else:
2383 try:
2384 self.to_screen('[info] Writing video annotations to: ' + annofn)
2385 with io.open(encodeFilename(annofn), 'w', encoding='utf-8') as annofile:
2386 annofile.write(info_dict['annotations'])
2387 except (KeyError, TypeError):
2388 self.report_warning('There are no annotations to write.')
2389 except (OSError, IOError):
2390 self.report_error('Cannot write annotations file: ' + annofn)
2391 return
2392
2393 subtitles_are_requested = any([self.params.get('writesubtitles', False),
2394 self.params.get('writeautomaticsub')])
2395
2396 if subtitles_are_requested and info_dict.get('requested_subtitles'):
2397 # subtitles download errors are already managed as troubles in relevant IE
2398 # that way it will silently go on when used with unsupporting IE
2399 subtitles = info_dict['requested_subtitles']
2400 # ie = self.get_info_extractor(info_dict['extractor_key'])
2401 for sub_lang, sub_info in subtitles.items():
2402 sub_format = sub_info['ext']
2403 sub_filename = subtitles_filename(temp_filename, sub_lang, sub_format, info_dict.get('ext'))
2404 sub_filename_final = subtitles_filename(
2405 self.prepare_filename(info_dict, 'subtitle'), sub_lang, sub_format, info_dict.get('ext'))
2406 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(sub_filename)):
2407 self.to_screen('[info] Video subtitle %s.%s is already present' % (sub_lang, sub_format))
2408 sub_info['filepath'] = sub_filename
2409 files_to_move[sub_filename] = sub_filename_final
2410 else:
2411 self.to_screen('[info] Writing video subtitles to: ' + sub_filename)
2412 if sub_info.get('data') is not None:
2413 try:
2414 # Use newline='' to prevent conversion of newline characters
2415 # See https://github.com/ytdl-org/youtube-dl/issues/10268
2416 with io.open(encodeFilename(sub_filename), 'w', encoding='utf-8', newline='') as subfile:
2417 subfile.write(sub_info['data'])
2418 sub_info['filepath'] = sub_filename
2419 files_to_move[sub_filename] = sub_filename_final
2420 except (OSError, IOError):
2421 self.report_error('Cannot write subtitles file ' + sub_filename)
2422 return
2423 else:
2424 try:
2425 self.dl(sub_filename, sub_info.copy(), subtitle=True)
2426 sub_info['filepath'] = sub_filename
2427 files_to_move[sub_filename] = sub_filename_final
2428 except (ExtractorError, IOError, OSError, ValueError) + network_exceptions as err:
2429 self.report_warning('Unable to download subtitle for "%s": %s' %
2430 (sub_lang, error_to_compat_str(err)))
2431 continue
2432
2433 if self.params.get('writeinfojson', False):
2434 infofn = self.prepare_filename(info_dict, 'infojson')
2435 if not self._ensure_dir_exists(encodeFilename(infofn)):
2436 return
2437 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(infofn)):
2438 self.to_screen('[info] Video metadata is already present')
2439 else:
2440 self.to_screen('[info] Writing video metadata as JSON to: ' + infofn)
2441 try:
2442 write_json_file(self.filter_requested_info(info_dict, self.params.get('clean_infojson', True)), infofn)
2443 except (OSError, IOError):
2444 self.report_error('Cannot write video metadata to JSON file ' + infofn)
2445 return
2446 info_dict['__infojson_filename'] = infofn
2447
2448 for thumb_ext in self._write_thumbnails(info_dict, temp_filename):
2449 thumb_filename_temp = replace_extension(temp_filename, thumb_ext, info_dict.get('ext'))
2450 thumb_filename = replace_extension(
2451 self.prepare_filename(info_dict, 'thumbnail'), thumb_ext, info_dict.get('ext'))
2452 files_to_move[thumb_filename_temp] = thumb_filename
2453
2454 # Write internet shortcut files
2455 url_link = webloc_link = desktop_link = False
2456 if self.params.get('writelink', False):
2457 if sys.platform == "darwin": # macOS.
2458 webloc_link = True
2459 elif sys.platform.startswith("linux"):
2460 desktop_link = True
2461 else: # if sys.platform in ['win32', 'cygwin']:
2462 url_link = True
2463 if self.params.get('writeurllink', False):
2464 url_link = True
2465 if self.params.get('writewebloclink', False):
2466 webloc_link = True
2467 if self.params.get('writedesktoplink', False):
2468 desktop_link = True
2469
2470 if url_link or webloc_link or desktop_link:
2471 if 'webpage_url' not in info_dict:
2472 self.report_error('Cannot write internet shortcut file because the "webpage_url" field is missing in the media information')
2473 return
2474 ascii_url = iri_to_uri(info_dict['webpage_url'])
2475
2476 def _write_link_file(extension, template, newline, embed_filename):
2477 linkfn = replace_extension(full_filename, extension, info_dict.get('ext'))
2478 if self.params.get('overwrites', True) and os.path.exists(encodeFilename(linkfn)):
2479 self.to_screen('[info] Internet shortcut is already present')
2480 else:
2481 try:
2482 self.to_screen('[info] Writing internet shortcut to: ' + linkfn)
2483 with io.open(encodeFilename(to_high_limit_path(linkfn)), 'w', encoding='utf-8', newline=newline) as linkfile:
2484 template_vars = {'url': ascii_url}
2485 if embed_filename:
2486 template_vars['filename'] = linkfn[:-(len(extension) + 1)]
2487 linkfile.write(template % template_vars)
2488 except (OSError, IOError):
2489 self.report_error('Cannot write internet shortcut ' + linkfn)
2490 return False
2491 return True
2492
2493 if url_link:
2494 if not _write_link_file('url', DOT_URL_LINK_TEMPLATE, '\r\n', embed_filename=False):
2495 return
2496 if webloc_link:
2497 if not _write_link_file('webloc', DOT_WEBLOC_LINK_TEMPLATE, '\n', embed_filename=False):
2498 return
2499 if desktop_link:
2500 if not _write_link_file('desktop', DOT_DESKTOP_LINK_TEMPLATE, '\n', embed_filename=True):
2501 return
2502
2503 try:
2504 info_dict, files_to_move = self.pre_process(info_dict, 'before_dl', files_to_move)
2505 except PostProcessingError as err:
2506 self.report_error('Preprocessing: %s' % str(err))
2507 return
2508
2509 must_record_download_archive = False
2510 if self.params.get('skip_download', False):
2511 info_dict['filepath'] = temp_filename
2512 info_dict['__finaldir'] = os.path.dirname(os.path.abspath(encodeFilename(full_filename)))
2513 info_dict['__files_to_move'] = files_to_move
2514 info_dict = self.run_pp(MoveFilesAfterDownloadPP(self, False), info_dict)
2515 else:
2516 # Download
2517 try:
2518
2519 def existing_file(*filepaths):
2520 ext = info_dict.get('ext')
2521 final_ext = self.params.get('final_ext', ext)
2522 existing_files = []
2523 for file in orderedSet(filepaths):
2524 if final_ext != ext:
2525 converted = replace_extension(file, final_ext, ext)
2526 if os.path.exists(encodeFilename(converted)):
2527 existing_files.append(converted)
2528 if os.path.exists(encodeFilename(file)):
2529 existing_files.append(file)
2530
2531 if not existing_files or self.params.get('overwrites', False):
2532 for file in orderedSet(existing_files):
2533 self.report_file_delete(file)
2534 os.remove(encodeFilename(file))
2535 return None
2536
2537 self.report_file_already_downloaded(existing_files[0])
2538 info_dict['ext'] = os.path.splitext(existing_files[0])[1][1:]
2539 return existing_files[0]
2540
2541 success = True
2542 if info_dict.get('requested_formats') is not None:
2543
2544 def compatible_formats(formats):
2545 # TODO: some formats actually allow this (mkv, webm, ogg, mp4), but not all of them.
2546 video_formats = [format for format in formats if format.get('vcodec') != 'none']
2547 audio_formats = [format for format in formats if format.get('acodec') != 'none']
2548 if len(video_formats) > 2 or len(audio_formats) > 2:
2549 return False
2550
2551 # Check extension
2552 exts = set(format.get('ext') for format in formats)
2553 COMPATIBLE_EXTS = (
2554 set(('mp3', 'mp4', 'm4a', 'm4p', 'm4b', 'm4r', 'm4v', 'ismv', 'isma')),
2555 set(('webm',)),
2556 )
2557 for ext_sets in COMPATIBLE_EXTS:
2558 if ext_sets.issuperset(exts):
2559 return True
2560 # TODO: Check acodec/vcodec
2561 return False
2562
2563 requested_formats = info_dict['requested_formats']
2564 old_ext = info_dict['ext']
2565 if self.params.get('merge_output_format') is None:
2566 if not compatible_formats(requested_formats):
2567 info_dict['ext'] = 'mkv'
2568 self.report_warning(
2569 'Requested formats are incompatible for merge and will be merged into mkv.')
2570 if (info_dict['ext'] == 'webm'
2571 and self.params.get('writethumbnail', False)
2572 and info_dict.get('thumbnails')):
2573 info_dict['ext'] = 'mkv'
2574 self.report_warning(
2575 'webm doesn\'t support embedding a thumbnail, mkv will be used.')
2576
2577 def correct_ext(filename):
2578 filename_real_ext = os.path.splitext(filename)[1][1:]
2579 filename_wo_ext = (
2580 os.path.splitext(filename)[0]
2581 if filename_real_ext == old_ext
2582 else filename)
2583 return '%s.%s' % (filename_wo_ext, info_dict['ext'])
2584
2585 # Ensure filename always has a correct extension for successful merge
2586 full_filename = correct_ext(full_filename)
2587 temp_filename = correct_ext(temp_filename)
2588 dl_filename = existing_file(full_filename, temp_filename)
2589 info_dict['__real_download'] = False
2590
2591 _protocols = set(determine_protocol(f) for f in requested_formats)
2592 if len(_protocols) == 1:
2593 info_dict['protocol'] = _protocols.pop()
2594 directly_mergable = (
2595 'no-direct-merge' not in self.params.get('compat_opts', [])
2596 and info_dict.get('protocol') is not None # All requested formats have same protocol
2597 and not self.params.get('allow_unplayable_formats')
2598 and get_suitable_downloader(info_dict, self.params).__name__ == 'FFmpegFD')
2599 if directly_mergable:
2600 info_dict['url'] = requested_formats[0]['url']
2601 # Treat it as a single download
2602 dl_filename = existing_file(full_filename, temp_filename)
2603 if dl_filename is None:
2604 success, real_download = self.dl(temp_filename, info_dict)
2605 info_dict['__real_download'] = real_download
2606 else:
2607 downloaded = []
2608 merger = FFmpegMergerPP(self)
2609 if self.params.get('allow_unplayable_formats'):
2610 self.report_warning(
2611 'You have requested merging of multiple formats '
2612 'while also allowing unplayable formats to be downloaded. '
2613 'The formats won\'t be merged to prevent data corruption.')
2614 elif not merger.available:
2615 self.report_warning(
2616 'You have requested merging of multiple formats but ffmpeg is not installed. '
2617 'The formats won\'t be merged.')
2618
2619 if dl_filename is None:
2620 for f in requested_formats:
2621 new_info = dict(info_dict)
2622 del new_info['requested_formats']
2623 new_info.update(f)
2624 fname = prepend_extension(
2625 self.prepare_filename(new_info, 'temp'),
2626 'f%s' % f['format_id'], new_info['ext'])
2627 if not self._ensure_dir_exists(fname):
2628 return
2629 downloaded.append(fname)
2630 partial_success, real_download = self.dl(fname, new_info)
2631 info_dict['__real_download'] = info_dict['__real_download'] or real_download
2632 success = success and partial_success
2633 if merger.available and not self.params.get('allow_unplayable_formats'):
2634 info_dict['__postprocessors'].append(merger)
2635 info_dict['__files_to_merge'] = downloaded
2636 # Even if there were no downloads, it is being merged only now
2637 info_dict['__real_download'] = True
2638 else:
2639 for file in downloaded:
2640 files_to_move[file] = None
2641 else:
2642 # Just a single file
2643 dl_filename = existing_file(full_filename, temp_filename)
2644 if dl_filename is None:
2645 success, real_download = self.dl(temp_filename, info_dict)
2646 info_dict['__real_download'] = real_download
2647
2648 dl_filename = dl_filename or temp_filename
2649 info_dict['__finaldir'] = os.path.dirname(os.path.abspath(encodeFilename(full_filename)))
2650
2651 except network_exceptions as err:
2652 self.report_error('unable to download video data: %s' % error_to_compat_str(err))
2653 return
2654 except (OSError, IOError) as err:
2655 raise UnavailableVideoError(err)
2656 except (ContentTooShortError, ) as err:
2657 self.report_error('content too short (expected %s bytes and served %s)' % (err.expected, err.downloaded))
2658 return
2659
2660 if success and full_filename != '-':
2661 # Fixup content
2662 fixup_policy = self.params.get('fixup')
2663 if fixup_policy is None:
2664 fixup_policy = 'detect_or_warn'
2665
2666 INSTALL_FFMPEG_MESSAGE = 'Install ffmpeg to fix this automatically.'
2667
2668 stretched_ratio = info_dict.get('stretched_ratio')
2669 if stretched_ratio is not None and stretched_ratio != 1:
2670 if fixup_policy == 'warn':
2671 self.report_warning('%s: Non-uniform pixel ratio (%s)' % (
2672 info_dict['id'], stretched_ratio))
2673 elif fixup_policy == 'detect_or_warn':
2674 stretched_pp = FFmpegFixupStretchedPP(self)
2675 if stretched_pp.available:
2676 info_dict['__postprocessors'].append(stretched_pp)
2677 else:
2678 self.report_warning(
2679 '%s: Non-uniform pixel ratio (%s). %s'
2680 % (info_dict['id'], stretched_ratio, INSTALL_FFMPEG_MESSAGE))
2681 else:
2682 assert fixup_policy in ('ignore', 'never')
2683
2684 if (info_dict.get('requested_formats') is None
2685 and info_dict.get('container') == 'm4a_dash'
2686 and info_dict.get('ext') == 'm4a'):
2687 if fixup_policy == 'warn':
2688 self.report_warning(
2689 '%s: writing DASH m4a. '
2690 'Only some players support this container.'
2691 % info_dict['id'])
2692 elif fixup_policy == 'detect_or_warn':
2693 fixup_pp = FFmpegFixupM4aPP(self)
2694 if fixup_pp.available:
2695 info_dict['__postprocessors'].append(fixup_pp)
2696 else:
2697 self.report_warning(
2698 '%s: writing DASH m4a. '
2699 'Only some players support this container. %s'
2700 % (info_dict['id'], INSTALL_FFMPEG_MESSAGE))
2701 else:
2702 assert fixup_policy in ('ignore', 'never')
2703
2704 if ('protocol' in info_dict
2705 and get_suitable_downloader(info_dict, self.params).__name__ == 'HlsFD'):
2706 if fixup_policy == 'warn':
2707 self.report_warning('%s: malformed AAC bitstream detected.' % (
2708 info_dict['id']))
2709 elif fixup_policy == 'detect_or_warn':
2710 fixup_pp = FFmpegFixupM3u8PP(self)
2711 if fixup_pp.available:
2712 info_dict['__postprocessors'].append(fixup_pp)
2713 else:
2714 self.report_warning(
2715 '%s: malformed AAC bitstream detected. %s'
2716 % (info_dict['id'], INSTALL_FFMPEG_MESSAGE))
2717 else:
2718 assert fixup_policy in ('ignore', 'never')
2719
2720 try:
2721 info_dict = self.post_process(dl_filename, info_dict, files_to_move)
2722 except PostProcessingError as err:
2723 self.report_error('Postprocessing: %s' % str(err))
2724 return
2725 try:
2726 for ph in self._post_hooks:
2727 ph(info_dict['filepath'])
2728 except Exception as err:
2729 self.report_error('post hooks: %s' % str(err))
2730 return
2731 must_record_download_archive = True
2732
2733 if must_record_download_archive or self.params.get('force_write_download_archive', False):
2734 self.record_download_archive(info_dict)
2735 max_downloads = self.params.get('max_downloads')
2736 if max_downloads is not None and self._num_downloads >= int(max_downloads):
2737 raise MaxDownloadsReached()
2738
2739 def download(self, url_list):
2740 """Download a given list of URLs."""
2741 outtmpl = self.outtmpl_dict['default']
2742 if (len(url_list) > 1
2743 and outtmpl != '-'
2744 and '%' not in outtmpl
2745 and self.params.get('max_downloads') != 1):
2746 raise SameFileError(outtmpl)
2747
2748 for url in url_list:
2749 try:
2750 # It also downloads the videos
2751 res = self.extract_info(
2752 url, force_generic_extractor=self.params.get('force_generic_extractor', False))
2753 except UnavailableVideoError:
2754 self.report_error('unable to download video')
2755 except MaxDownloadsReached:
2756 self.to_screen('[info] Maximum number of downloaded files reached')
2757 raise
2758 except ExistingVideoReached:
2759 self.to_screen('[info] Encountered a file that is already in the archive, stopping due to --break-on-existing')
2760 raise
2761 except RejectedVideoReached:
2762 self.to_screen('[info] Encountered a file that did not match filter, stopping due to --break-on-reject')
2763 raise
2764 else:
2765 if self.params.get('dump_single_json', False):
2766 self.post_extract(res)
2767 self.to_stdout(json.dumps(res, default=repr))
2768
2769 return self._download_retcode
2770
2771 def download_with_info_file(self, info_filename):
2772 with contextlib.closing(fileinput.FileInput(
2773 [info_filename], mode='r',
2774 openhook=fileinput.hook_encoded('utf-8'))) as f:
2775 # FileInput doesn't have a read method, we can't call json.load
2776 info = self.filter_requested_info(json.loads('\n'.join(f)), self.params.get('clean_infojson', True))
2777 try:
2778 self.process_ie_result(info, download=True)
2779 except (DownloadError, EntryNotInPlaylist):
2780 webpage_url = info.get('webpage_url')
2781 if webpage_url is not None:
2782 self.report_warning('The info failed to download, trying with "%s"' % webpage_url)
2783 return self.download([webpage_url])
2784 else:
2785 raise
2786 return self._download_retcode
2787
2788 @staticmethod
2789 def filter_requested_info(info_dict, actually_filter=True):
2790 remove_keys = ['__original_infodict'] # Always remove this since this may contain a copy of the entire dict
2791 keep_keys = ['_type'], # Always keep this to facilitate load-info-json
2792 if actually_filter:
2793 remove_keys += ('requested_formats', 'requested_subtitles', 'requested_entries', 'filepath', 'entries', 'original_url')
2794 empty_values = (None, {}, [], set(), tuple())
2795 reject = lambda k, v: k not in keep_keys and (
2796 k.startswith('_') or k in remove_keys or v in empty_values)
2797 else:
2798 info_dict['epoch'] = int(time.time())
2799 reject = lambda k, v: k in remove_keys
2800 filter_fn = lambda obj: (
2801 list(map(filter_fn, obj)) if isinstance(obj, (list, tuple, set))
2802 else obj if not isinstance(obj, dict)
2803 else dict((k, filter_fn(v)) for k, v in obj.items() if not reject(k, v)))
2804 return filter_fn(info_dict)
2805
2806 def run_pp(self, pp, infodict):
2807 files_to_delete = []
2808 if '__files_to_move' not in infodict:
2809 infodict['__files_to_move'] = {}
2810 files_to_delete, infodict = pp.run(infodict)
2811 if not files_to_delete:
2812 return infodict
2813
2814 if self.params.get('keepvideo', False):
2815 for f in files_to_delete:
2816 infodict['__files_to_move'].setdefault(f, '')
2817 else:
2818 for old_filename in set(files_to_delete):
2819 self.to_screen('Deleting original file %s (pass -k to keep)' % old_filename)
2820 try:
2821 os.remove(encodeFilename(old_filename))
2822 except (IOError, OSError):
2823 self.report_warning('Unable to remove downloaded original file')
2824 if old_filename in infodict['__files_to_move']:
2825 del infodict['__files_to_move'][old_filename]
2826 return infodict
2827
2828 @staticmethod
2829 def post_extract(info_dict):
2830 def actual_post_extract(info_dict):
2831 if info_dict.get('_type') in ('playlist', 'multi_video'):
2832 for video_dict in info_dict.get('entries', {}):
2833 actual_post_extract(video_dict or {})
2834 return
2835
2836 post_extractor = info_dict.get('__post_extractor') or (lambda: {})
2837 extra = post_extractor().items()
2838 info_dict.update(extra)
2839 info_dict.pop('__post_extractor', None)
2840
2841 original_infodict = info_dict.get('__original_infodict') or {}
2842 original_infodict.update(extra)
2843 original_infodict.pop('__post_extractor', None)
2844
2845 actual_post_extract(info_dict or {})
2846
2847 def pre_process(self, ie_info, key='pre_process', files_to_move=None):
2848 info = dict(ie_info)
2849 info['__files_to_move'] = files_to_move or {}
2850 for pp in self._pps[key]:
2851 info = self.run_pp(pp, info)
2852 return info, info.pop('__files_to_move', None)
2853
2854 def post_process(self, filename, ie_info, files_to_move=None):
2855 """Run all the postprocessors on the given file."""
2856 info = dict(ie_info)
2857 info['filepath'] = filename
2858 info['__files_to_move'] = files_to_move or {}
2859
2860 for pp in ie_info.get('__postprocessors', []) + self._pps['post_process']:
2861 info = self.run_pp(pp, info)
2862 info = self.run_pp(MoveFilesAfterDownloadPP(self), info)
2863 del info['__files_to_move']
2864 for pp in self._pps['after_move']:
2865 info = self.run_pp(pp, info)
2866 return info
2867
2868 def _make_archive_id(self, info_dict):
2869 video_id = info_dict.get('id')
2870 if not video_id:
2871 return
2872 # Future-proof against any change in case
2873 # and backwards compatibility with prior versions
2874 extractor = info_dict.get('extractor_key') or info_dict.get('ie_key') # key in a playlist
2875 if extractor is None:
2876 url = str_or_none(info_dict.get('url'))
2877 if not url:
2878 return
2879 # Try to find matching extractor for the URL and take its ie_key
2880 for ie in self._ies:
2881 if ie.suitable(url):
2882 extractor = ie.ie_key()
2883 break
2884 else:
2885 return
2886 return '%s %s' % (extractor.lower(), video_id)
2887
2888 def in_download_archive(self, info_dict):
2889 fn = self.params.get('download_archive')
2890 if fn is None:
2891 return False
2892
2893 vid_id = self._make_archive_id(info_dict)
2894 if not vid_id:
2895 return False # Incomplete video information
2896
2897 return vid_id in self.archive
2898
2899 def record_download_archive(self, info_dict):
2900 fn = self.params.get('download_archive')
2901 if fn is None:
2902 return
2903 vid_id = self._make_archive_id(info_dict)
2904 assert vid_id
2905 with locked_file(fn, 'a', encoding='utf-8') as archive_file:
2906 archive_file.write(vid_id + '\n')
2907 self.archive.add(vid_id)
2908
2909 @staticmethod
2910 def format_resolution(format, default='unknown'):
2911 if format.get('vcodec') == 'none':
2912 return 'audio only'
2913 if format.get('resolution') is not None:
2914 return format['resolution']
2915 if format.get('width') and format.get('height'):
2916 res = '%dx%d' % (format['width'], format['height'])
2917 elif format.get('height'):
2918 res = '%sp' % format['height']
2919 elif format.get('width'):
2920 res = '%dx?' % format['width']
2921 else:
2922 res = default
2923 return res
2924
2925 def _format_note(self, fdict):
2926 res = ''
2927 if fdict.get('ext') in ['f4f', 'f4m']:
2928 res += '(unsupported) '
2929 if fdict.get('language'):
2930 if res:
2931 res += ' '
2932 res += '[%s] ' % fdict['language']
2933 if fdict.get('format_note') is not None:
2934 res += fdict['format_note'] + ' '
2935 if fdict.get('tbr') is not None:
2936 res += '%4dk ' % fdict['tbr']
2937 if fdict.get('container') is not None:
2938 if res:
2939 res += ', '
2940 res += '%s container' % fdict['container']
2941 if (fdict.get('vcodec') is not None
2942 and fdict.get('vcodec') != 'none'):
2943 if res:
2944 res += ', '
2945 res += fdict['vcodec']
2946 if fdict.get('vbr') is not None:
2947 res += '@'
2948 elif fdict.get('vbr') is not None and fdict.get('abr') is not None:
2949 res += 'video@'
2950 if fdict.get('vbr') is not None:
2951 res += '%4dk' % fdict['vbr']
2952 if fdict.get('fps') is not None:
2953 if res:
2954 res += ', '
2955 res += '%sfps' % fdict['fps']
2956 if fdict.get('acodec') is not None:
2957 if res:
2958 res += ', '
2959 if fdict['acodec'] == 'none':
2960 res += 'video only'
2961 else:
2962 res += '%-5s' % fdict['acodec']
2963 elif fdict.get('abr') is not None:
2964 if res:
2965 res += ', '
2966 res += 'audio'
2967 if fdict.get('abr') is not None:
2968 res += '@%3dk' % fdict['abr']
2969 if fdict.get('asr') is not None:
2970 res += ' (%5dHz)' % fdict['asr']
2971 if fdict.get('filesize') is not None:
2972 if res:
2973 res += ', '
2974 res += format_bytes(fdict['filesize'])
2975 elif fdict.get('filesize_approx') is not None:
2976 if res:
2977 res += ', '
2978 res += '~' + format_bytes(fdict['filesize_approx'])
2979 return res
2980
2981 def _format_note_table(self, f):
2982 def join_fields(*vargs):
2983 return ', '.join((val for val in vargs if val != ''))
2984
2985 return join_fields(
2986 'UNSUPPORTED' if f.get('ext') in ('f4f', 'f4m') else '',
2987 format_field(f, 'language', '[%s]'),
2988 format_field(f, 'format_note'),
2989 format_field(f, 'container', ignore=(None, f.get('ext'))),
2990 format_field(f, 'asr', '%5dHz'))
2991
2992 def list_formats(self, info_dict):
2993 formats = info_dict.get('formats', [info_dict])
2994 new_format = (
2995 'list-formats' not in self.params.get('compat_opts', [])
2996 and self.params.get('list_formats_as_table', True) is not False)
2997 if new_format:
2998 table = [
2999 [
3000 format_field(f, 'format_id'),
3001 format_field(f, 'ext'),
3002 self.format_resolution(f),
3003 format_field(f, 'fps', '%d'),
3004 '|',
3005 format_field(f, 'filesize', ' %s', func=format_bytes) + format_field(f, 'filesize_approx', '~%s', func=format_bytes),
3006 format_field(f, 'tbr', '%4dk'),
3007 shorten_protocol_name(f.get('protocol', '').replace("native", "n")),
3008 '|',
3009 format_field(f, 'vcodec', default='unknown').replace('none', ''),
3010 format_field(f, 'vbr', '%4dk'),
3011 format_field(f, 'acodec', default='unknown').replace('none', ''),
3012 format_field(f, 'abr', '%3dk'),
3013 format_field(f, 'asr', '%5dHz'),
3014 self._format_note_table(f)]
3015 for f in formats
3016 if f.get('preference') is None or f['preference'] >= -1000]
3017 header_line = ['ID', 'EXT', 'RESOLUTION', 'FPS', '|', ' FILESIZE', ' TBR', 'PROTO',
3018 '|', 'VCODEC', ' VBR', 'ACODEC', ' ABR', ' ASR', 'NOTE']
3019 else:
3020 table = [
3021 [
3022 format_field(f, 'format_id'),
3023 format_field(f, 'ext'),
3024 self.format_resolution(f),
3025 self._format_note(f)]
3026 for f in formats
3027 if f.get('preference') is None or f['preference'] >= -1000]
3028 header_line = ['format code', 'extension', 'resolution', 'note']
3029
3030 self.to_screen(
3031 '[info] Available formats for %s:\n%s' % (info_dict['id'], render_table(
3032 header_line,
3033 table,
3034 delim=new_format,
3035 extraGap=(0 if new_format else 1),
3036 hideEmpty=new_format)))
3037
3038 def list_thumbnails(self, info_dict):
3039 thumbnails = info_dict.get('thumbnails')
3040 if not thumbnails:
3041 self.to_screen('[info] No thumbnails present for %s' % info_dict['id'])
3042 return
3043
3044 self.to_screen(
3045 '[info] Thumbnails for %s:' % info_dict['id'])
3046 self.to_screen(render_table(
3047 ['ID', 'width', 'height', 'URL'],
3048 [[t['id'], t.get('width', 'unknown'), t.get('height', 'unknown'), t['url']] for t in thumbnails]))
3049
3050 def list_subtitles(self, video_id, subtitles, name='subtitles'):
3051 if not subtitles:
3052 self.to_screen('%s has no %s' % (video_id, name))
3053 return
3054 self.to_screen(
3055 'Available %s for %s:' % (name, video_id))
3056
3057 def _row(lang, formats):
3058 exts, names = zip(*((f['ext'], f.get('name', 'unknown')) for f in reversed(formats)))
3059 if len(set(names)) == 1:
3060 names = [] if names[0] == 'unknown' else names[:1]
3061 return [lang, ', '.join(names), ', '.join(exts)]
3062
3063 self.to_screen(render_table(
3064 ['Language', 'Name', 'Formats'],
3065 [_row(lang, formats) for lang, formats in subtitles.items()],
3066 hideEmpty=True))
3067
3068 def urlopen(self, req):
3069 """ Start an HTTP download """
3070 if isinstance(req, compat_basestring):
3071 req = sanitized_Request(req)
3072 return self._opener.open(req, timeout=self._socket_timeout)
3073
3074 def print_debug_header(self):
3075 if not self.params.get('verbose'):
3076 return
3077
3078 if type('') is not compat_str:
3079 # Python 2.6 on SLES11 SP1 (https://github.com/ytdl-org/youtube-dl/issues/3326)
3080 self.report_warning(
3081 'Your Python is broken! Update to a newer and supported version')
3082
3083 stdout_encoding = getattr(
3084 sys.stdout, 'encoding', 'missing (%s)' % type(sys.stdout).__name__)
3085 encoding_str = (
3086 '[debug] Encodings: locale %s, fs %s, out %s, pref %s\n' % (
3087 locale.getpreferredencoding(),
3088 sys.getfilesystemencoding(),
3089 stdout_encoding,
3090 self.get_encoding()))
3091 write_string(encoding_str, encoding=None)
3092
3093 source = (
3094 '(exe)' if hasattr(sys, 'frozen')
3095 else '(zip)' if isinstance(globals().get('__loader__'), zipimporter)
3096 else '(source)' if os.path.basename(sys.argv[0]) == '__main__.py'
3097 else '')
3098 self._write_string('[debug] yt-dlp version %s %s\n' % (__version__, source))
3099 if _LAZY_LOADER:
3100 self._write_string('[debug] Lazy loading extractors enabled\n')
3101 if _PLUGIN_CLASSES:
3102 self._write_string(
3103 '[debug] Plugin Extractors: %s\n' % [ie.ie_key() for ie in _PLUGIN_CLASSES])
3104 if self.params.get('compat_opts'):
3105 self._write_string(
3106 '[debug] Compatibility options: %s\n' % ', '.join(self.params.get('compat_opts')))
3107 try:
3108 sp = subprocess.Popen(
3109 ['git', 'rev-parse', '--short', 'HEAD'],
3110 stdout=subprocess.PIPE, stderr=subprocess.PIPE,
3111 cwd=os.path.dirname(os.path.abspath(__file__)))
3112 out, err = process_communicate_or_kill(sp)
3113 out = out.decode().strip()
3114 if re.match('[0-9a-f]+', out):
3115 self._write_string('[debug] Git HEAD: %s\n' % out)
3116 except Exception:
3117 try:
3118 sys.exc_clear()
3119 except Exception:
3120 pass
3121
3122 def python_implementation():
3123 impl_name = platform.python_implementation()
3124 if impl_name == 'PyPy' and hasattr(sys, 'pypy_version_info'):
3125 return impl_name + ' version %d.%d.%d' % sys.pypy_version_info[:3]
3126 return impl_name
3127
3128 self._write_string('[debug] Python version %s (%s %s) - %s\n' % (
3129 platform.python_version(),
3130 python_implementation(),
3131 platform.architecture()[0],
3132 platform_name()))
3133
3134 exe_versions = FFmpegPostProcessor.get_versions(self)
3135 exe_versions['rtmpdump'] = rtmpdump_version()
3136 exe_versions['phantomjs'] = PhantomJSwrapper._version()
3137 exe_str = ', '.join(
3138 '%s %s' % (exe, v)
3139 for exe, v in sorted(exe_versions.items())
3140 if v
3141 )
3142 if not exe_str:
3143 exe_str = 'none'
3144 self._write_string('[debug] exe versions: %s\n' % exe_str)
3145
3146 proxy_map = {}
3147 for handler in self._opener.handlers:
3148 if hasattr(handler, 'proxies'):
3149 proxy_map.update(handler.proxies)
3150 self._write_string('[debug] Proxy map: ' + compat_str(proxy_map) + '\n')
3151
3152 if self.params.get('call_home', False):
3153 ipaddr = self.urlopen('https://yt-dl.org/ip').read().decode('utf-8')
3154 self._write_string('[debug] Public IP address: %s\n' % ipaddr)
3155 return
3156 latest_version = self.urlopen(
3157 'https://yt-dl.org/latest/version').read().decode('utf-8')
3158 if version_tuple(latest_version) > version_tuple(__version__):
3159 self.report_warning(
3160 'You are using an outdated version (newest version: %s)! '
3161 'See https://yt-dl.org/update if you need help updating.' %
3162 latest_version)
3163
3164 def _setup_opener(self):
3165 timeout_val = self.params.get('socket_timeout')
3166 self._socket_timeout = 600 if timeout_val is None else float(timeout_val)
3167
3168 opts_cookiefile = self.params.get('cookiefile')
3169 opts_proxy = self.params.get('proxy')
3170
3171 if opts_cookiefile is None:
3172 self.cookiejar = compat_cookiejar.CookieJar()
3173 else:
3174 opts_cookiefile = expand_path(opts_cookiefile)
3175 self.cookiejar = YoutubeDLCookieJar(opts_cookiefile)
3176 if os.access(opts_cookiefile, os.R_OK):
3177 self.cookiejar.load(ignore_discard=True, ignore_expires=True)
3178
3179 cookie_processor = YoutubeDLCookieProcessor(self.cookiejar)
3180 if opts_proxy is not None:
3181 if opts_proxy == '':
3182 proxies = {}
3183 else:
3184 proxies = {'http': opts_proxy, 'https': opts_proxy}
3185 else:
3186 proxies = compat_urllib_request.getproxies()
3187 # Set HTTPS proxy to HTTP one if given (https://github.com/ytdl-org/youtube-dl/issues/805)
3188 if 'http' in proxies and 'https' not in proxies:
3189 proxies['https'] = proxies['http']
3190 proxy_handler = PerRequestProxyHandler(proxies)
3191
3192 debuglevel = 1 if self.params.get('debug_printtraffic') else 0
3193 https_handler = make_HTTPS_handler(self.params, debuglevel=debuglevel)
3194 ydlh = YoutubeDLHandler(self.params, debuglevel=debuglevel)
3195 redirect_handler = YoutubeDLRedirectHandler()
3196 data_handler = compat_urllib_request_DataHandler()
3197
3198 # When passing our own FileHandler instance, build_opener won't add the
3199 # default FileHandler and allows us to disable the file protocol, which
3200 # can be used for malicious purposes (see
3201 # https://github.com/ytdl-org/youtube-dl/issues/8227)
3202 file_handler = compat_urllib_request.FileHandler()
3203
3204 def file_open(*args, **kwargs):
3205 raise compat_urllib_error.URLError('file:// scheme is explicitly disabled in yt-dlp for security reasons')
3206 file_handler.file_open = file_open
3207
3208 opener = compat_urllib_request.build_opener(
3209 proxy_handler, https_handler, cookie_processor, ydlh, redirect_handler, data_handler, file_handler)
3210
3211 # Delete the default user-agent header, which would otherwise apply in
3212 # cases where our custom HTTP handler doesn't come into play
3213 # (See https://github.com/ytdl-org/youtube-dl/issues/1309 for details)
3214 opener.addheaders = []
3215 self._opener = opener
3216
3217 def encode(self, s):
3218 if isinstance(s, bytes):
3219 return s # Already encoded
3220
3221 try:
3222 return s.encode(self.get_encoding())
3223 except UnicodeEncodeError as err:
3224 err.reason = err.reason + '. Check your system encoding configuration or use the --encoding option.'
3225 raise
3226
3227 def get_encoding(self):
3228 encoding = self.params.get('encoding')
3229 if encoding is None:
3230 encoding = preferredencoding()
3231 return encoding
3232
3233 def _write_thumbnails(self, info_dict, filename): # return the extensions
3234 write_all = self.params.get('write_all_thumbnails', False)
3235 thumbnails = []
3236 if write_all or self.params.get('writethumbnail', False):
3237 thumbnails = info_dict.get('thumbnails') or []
3238 multiple = write_all and len(thumbnails) > 1
3239
3240 ret = []
3241 for t in thumbnails[::1 if write_all else -1]:
3242 thumb_ext = determine_ext(t['url'], 'jpg')
3243 suffix = '%s.' % t['id'] if multiple else ''
3244 thumb_display_id = '%s ' % t['id'] if multiple else ''
3245 thumb_filename = replace_extension(filename, suffix + thumb_ext, info_dict.get('ext'))
3246
3247 if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(thumb_filename)):
3248 ret.append(suffix + thumb_ext)
3249 self.to_screen('[%s] %s: Thumbnail %sis already present' %
3250 (info_dict['extractor'], info_dict['id'], thumb_display_id))
3251 else:
3252 self.to_screen('[%s] %s: Downloading thumbnail %s ...' %
3253 (info_dict['extractor'], info_dict['id'], thumb_display_id))
3254 try:
3255 uf = self.urlopen(t['url'])
3256 with open(encodeFilename(thumb_filename), 'wb') as thumbf:
3257 shutil.copyfileobj(uf, thumbf)
3258 ret.append(suffix + thumb_ext)
3259 self.to_screen('[%s] %s: Writing thumbnail %sto: %s' %
3260 (info_dict['extractor'], info_dict['id'], thumb_display_id, thumb_filename))
3261 t['filepath'] = thumb_filename
3262 except network_exceptions as err:
3263 self.report_warning('Unable to download thumbnail "%s": %s' %
3264 (t['url'], error_to_compat_str(err)))
3265 if ret and not write_all:
3266 break
3267 return ret