]> jfr.im git - yt-dlp.git/blame - youtube_dl/YoutubeDL.py
[flickr] Ensure format id is string
[yt-dlp.git] / youtube_dl / YoutubeDL.py
CommitLineData
8222d8de 1#!/usr/bin/env python
dcdb292f 2# coding: utf-8
8222d8de 3
6febd1c1 4from __future__ import absolute_import, unicode_literals
8222d8de 5
26e63931 6import collections
31bd3925 7import contextlib
317f7ab6 8import copy
9d2ecdbc 9import datetime
c1c9a79c 10import errno
31bd3925 11import fileinput
8222d8de 12import io
b82f815f 13import itertools
8694c600 14import json
62fec3b2 15import locale
083c9df9 16import operator
8222d8de 17import os
dca08720 18import platform
8222d8de
JMF
19import re
20import shutil
dca08720 21import subprocess
8222d8de
JMF
22import socket
23import sys
24import time
67134eab 25import tokenize
8222d8de 26import traceback
75822ca7 27import random
8222d8de 28
8c25f81b 29from .compat import (
82d8a8b6 30 compat_basestring,
dca08720 31 compat_cookiejar,
003c69a8 32 compat_get_terminal_size,
ce02ed60 33 compat_http_client,
4f026faf 34 compat_kwargs,
d0d9ade4 35 compat_numeric_types,
e9c0cdd3 36 compat_os_name,
ce02ed60 37 compat_str,
67134eab 38 compat_tokenize_tokenize,
ce02ed60
PH
39 compat_urllib_error,
40 compat_urllib_request,
8b172c2e 41 compat_urllib_request_DataHandler,
8c25f81b
PH
42)
43from .utils import (
eedb7ba5
S
44 age_restricted,
45 args_to_str,
ce02ed60
PH
46 ContentTooShortError,
47 date_from_str,
48 DateRange,
acd69589 49 DEFAULT_OUTTMPL,
ce02ed60 50 determine_ext,
b5559424 51 determine_protocol,
ce02ed60 52 DownloadError,
c0384f22 53 encode_compat_str,
ce02ed60 54 encodeFilename,
9b9c5355 55 error_to_compat_str,
590bc6f6 56 expand_path,
ce02ed60 57 ExtractorError,
02dbf93f 58 format_bytes,
525ef922 59 formatSeconds,
773f291d 60 GeoRestrictedError,
c9969434 61 int_or_none,
773f291d 62 ISO3166Utils,
ce02ed60 63 locked_file,
dca08720 64 make_HTTPS_handler,
ce02ed60 65 MaxDownloadsReached,
b7ab0590 66 PagedList,
083c9df9 67 parse_filesize,
91410c9b 68 PerRequestProxyHandler,
dca08720 69 platform_name,
eedb7ba5 70 PostProcessingError,
ce02ed60 71 preferredencoding,
eedb7ba5 72 prepend_extension,
51fb4995 73 register_socks_protocols,
cfb56d1a 74 render_table,
eedb7ba5 75 replace_extension,
ce02ed60
PH
76 SameFileError,
77 sanitize_filename,
1bb5c511 78 sanitize_path,
dcf77cf1 79 sanitize_url,
67dda517 80 sanitized_Request,
e5660ee6 81 std_headers,
ce02ed60 82 subtitles_filename,
ce02ed60 83 UnavailableVideoError,
29eb5174 84 url_basename,
58b1f00d 85 version_tuple,
ce02ed60
PH
86 write_json_file,
87 write_string,
6a3f4c3f 88 YoutubeDLCookieProcessor,
dca08720 89 YoutubeDLHandler,
ce02ed60 90)
a0e07d31 91from .cache import Cache
e0986e31 92from .extractor import get_info_extractor, gen_extractor_classes, _LAZY_LOADER
3bc2ddcc 93from .downloader import get_suitable_downloader
4c83c967 94from .downloader.rtmp import rtmpdump_version
4f026faf 95from .postprocessor import (
f17f8651 96 FFmpegFixupM3u8PP,
62cd676c 97 FFmpegFixupM4aPP,
6271f1ca 98 FFmpegFixupStretchedPP,
4f026faf
PH
99 FFmpegMergerPP,
100 FFmpegPostProcessor,
101 get_postprocessor,
102)
dca08720 103from .version import __version__
8222d8de 104
e9c0cdd3
YCH
105if compat_os_name == 'nt':
106 import ctypes
107
8222d8de
JMF
108
109class YoutubeDL(object):
110 """YoutubeDL class.
111
112 YoutubeDL objects are the ones responsible of downloading the
113 actual video file and writing it to disk if the user has requested
114 it, among some other tasks. In most cases there should be one per
115 program. As, given a video URL, the downloader doesn't know how to
116 extract all the needed information, task that InfoExtractors do, it
117 has to pass the URL to one of them.
118
119 For this, YoutubeDL objects have a method that allows
120 InfoExtractors to be registered in a given order. When it is passed
121 a URL, the YoutubeDL object handles it to the first InfoExtractor it
122 finds that reports being able to handle it. The InfoExtractor extracts
123 all the information about the video or videos the URL refers to, and
124 YoutubeDL process the extracted information, possibly using a File
125 Downloader to download the video.
126
127 YoutubeDL objects accept a lot of parameters. In order not to saturate
128 the object constructor with arguments, it receives a dictionary of
129 options instead. These options are available through the params
130 attribute for the InfoExtractors to use. The YoutubeDL also
131 registers itself as the downloader in charge for the InfoExtractors
132 that are added to it, so this is a "mutual registration".
133
134 Available options:
135
136 username: Username for authentication purposes.
137 password: Password for authentication purposes.
180940e0 138 videopassword: Password for accessing a video.
1da50aa3
S
139 ap_mso: Adobe Pass multiple-system operator identifier.
140 ap_username: Multiple-system operator account username.
141 ap_password: Multiple-system operator account password.
8222d8de
JMF
142 usenetrc: Use netrc for authentication instead.
143 verbose: Print additional info to stdout.
144 quiet: Do not print messages to stdout.
ad8915b7 145 no_warnings: Do not print out anything for warnings.
8222d8de
JMF
146 forceurl: Force printing final URL.
147 forcetitle: Force printing title.
148 forceid: Force printing ID.
149 forcethumbnail: Force printing thumbnail URL.
150 forcedescription: Force printing description.
151 forcefilename: Force printing final filename.
525ef922 152 forceduration: Force printing duration.
8694c600 153 forcejson: Force printing info_dict as JSON.
63e0be34
PH
154 dump_single_json: Force printing the info_dict of the whole playlist
155 (or video) as a single JSON line.
8222d8de 156 simulate: Do not download the video files.
d8600787 157 format: Video format code. See options.py for more information.
8222d8de
JMF
158 outtmpl: Template for output names.
159 restrictfilenames: Do not allow "&" and spaces in file names
160 ignoreerrors: Do not stop on download errors.
d22dec74 161 force_generic_extractor: Force downloader to use the generic extractor
8222d8de
JMF
162 nooverwrites: Prevent overwriting files.
163 playliststart: Playlist item to start at.
164 playlistend: Playlist item to end at.
c14e88f0 165 playlist_items: Specific indices of playlist to download.
ff815fe6 166 playlistreverse: Download playlist items in reverse order.
75822ca7 167 playlistrandom: Download playlist items in random order.
8222d8de
JMF
168 matchtitle: Download only matching titles.
169 rejecttitle: Reject downloads for matching titles.
8bf9319e 170 logger: Log messages to a logging.Logger instance.
8222d8de
JMF
171 logtostderr: Log messages to stderr instead of stdout.
172 writedescription: Write the video description to a .description file
173 writeinfojson: Write the video description to a .info.json file
1fb07d10 174 writeannotations: Write the video annotations to a .annotations.xml file
8222d8de 175 writethumbnail: Write the thumbnail image to a file
ec82d85a 176 write_all_thumbnails: Write all thumbnail formats to files
8222d8de 177 writesubtitles: Write the video subtitles to a file
741dd8ea 178 writeautomaticsub: Write the automatically generated subtitles to a file
8222d8de 179 allsubtitles: Downloads all the subtitles of the video
0b7f3118 180 (requires writesubtitles or writeautomaticsub)
8222d8de 181 listsubtitles: Lists all available subtitles for the video
a504ced0 182 subtitlesformat: The format code for subtitles
aa6a10c4 183 subtitleslangs: List of languages of the subtitles to download
8222d8de
JMF
184 keepvideo: Keep the video file after post-processing
185 daterange: A DateRange object, download only if the upload_date is in the range.
186 skip_download: Skip the actual download of the video file
c35f9e72 187 cachedir: Location of the cache files in the filesystem.
a0e07d31 188 False to disable filesystem cache.
47192f92 189 noplaylist: Download single video instead of a playlist if in doubt.
8dbe9899
PH
190 age_limit: An integer representing the user's age in years.
191 Unsuitable videos for the given age are skipped.
5fe18bdb
PH
192 min_views: An integer representing the minimum view count the video
193 must have in order to not be skipped.
194 Videos without view count information are always
195 downloaded. None for no limit.
196 max_views: An integer representing the maximum view count.
197 Videos that are more popular than that are not
198 downloaded.
199 Videos without view count information are always
200 downloaded. None for no limit.
201 download_archive: File name of a file where all downloads are recorded.
c1c9a79c
PH
202 Videos already present in the file are not downloaded
203 again.
dca08720 204 cookiefile: File name where cookies should be read from and dumped to.
a1ee09e8 205 nocheckcertificate:Do not verify SSL certificates
7e8c0af0
PH
206 prefer_insecure: Use HTTP instead of HTTPS to retrieve information.
207 At the moment, this is only supported by YouTube.
a1ee09e8 208 proxy: URL of the proxy server to use
38cce791
YCH
209 geo_verification_proxy: URL of the proxy to use for IP address verification
210 on geo-restricted sites. (Experimental)
e344693b 211 socket_timeout: Time to wait for unresponsive hosts, in seconds
0783b09b
PH
212 bidi_workaround: Work around buggy terminals without bidirectional text
213 support, using fridibi
a0ddb8a2 214 debug_printtraffic:Print out sent and received HTTP traffic
7b0817e8 215 include_ads: Download ads as well
04b4d394
PH
216 default_search: Prepend this string if an input url is not valid.
217 'auto' for elaborate guessing
62fec3b2 218 encoding: Use this encoding instead of the system-specified.
e8ee972c 219 extract_flat: Do not resolve URLs, return the immediate result.
057a5206
PH
220 Pass in 'in_playlist' to only show this behavior for
221 playlist items.
4f026faf 222 postprocessors: A list of dictionaries, each with an entry
71b640cc
PH
223 * key: The name of the postprocessor. See
224 youtube_dl/postprocessor/__init__.py for a list.
4f026faf
PH
225 as well as any further keyword arguments for the
226 postprocessor.
71b640cc
PH
227 progress_hooks: A list of functions that get called on download
228 progress, with a dictionary with the entries
5cda4eda 229 * status: One of "downloading", "error", or "finished".
ee69b99a 230 Check this first and ignore unknown values.
71b640cc 231
5cda4eda 232 If status is one of "downloading", or "finished", the
ee69b99a
PH
233 following properties may also be present:
234 * filename: The final filename (always present)
5cda4eda 235 * tmpfilename: The filename we're currently writing to
71b640cc
PH
236 * downloaded_bytes: Bytes on disk
237 * total_bytes: Size of the whole file, None if unknown
5cda4eda
PH
238 * total_bytes_estimate: Guess of the eventual file size,
239 None if unavailable.
240 * elapsed: The number of seconds since download started.
71b640cc
PH
241 * eta: The estimated time in seconds, None if unknown
242 * speed: The download speed in bytes/second, None if
243 unknown
5cda4eda
PH
244 * fragment_index: The counter of the currently
245 downloaded video fragment.
246 * fragment_count: The number of fragments (= individual
247 files that will be merged)
71b640cc
PH
248
249 Progress hooks are guaranteed to be called at least once
250 (with status "finished") if the download is successful.
45598f15 251 merge_output_format: Extension to use when merging formats.
6271f1ca
PH
252 fixup: Automatically correct known faults of the file.
253 One of:
254 - "never": do nothing
255 - "warn": only emit a warning
256 - "detect_or_warn": check whether we can do anything
62cd676c 257 about it, warn otherwise (default)
be4a824d 258 source_address: (Experimental) Client-side IP address to bind to.
6ec6cb4e 259 call_home: Boolean, true iff we are allowed to contact the
8bfa7545 260 youtube-dl servers for debugging.
7aa589a5
S
261 sleep_interval: Number of seconds to sleep before each download when
262 used alone or a lower bound of a range for randomized
263 sleep before each download (minimum possible number
264 of seconds to sleep) when used along with
265 max_sleep_interval.
266 max_sleep_interval:Upper bound of a range for randomized sleep before each
267 download (maximum possible number of seconds to sleep).
268 Must only be used along with sleep_interval.
269 Actual sleep time will be a random float from range
270 [sleep_interval; max_sleep_interval].
cfb56d1a
PH
271 listformats: Print an overview of available video formats and exit.
272 list_thumbnails: Print a table of all thumbnails and exit.
347de493
PH
273 match_filter: A function that gets called with the info_dict of
274 every video.
275 If it returns a message, the video is ignored.
276 If it returns None, the video is downloaded.
277 match_filter_func in utils.py is one example for this.
7e5db8c9 278 no_color: Do not emit color codes in output.
0a840f58 279 geo_bypass: Bypass geographic restriction via faking X-Forwarded-For
773f291d 280 HTTP header (experimental)
0a840f58 281 geo_bypass_country:
773f291d
S
282 Two-letter ISO 3166-2 country code that will be used for
283 explicit geographic restriction bypassing via faking
284 X-Forwarded-For HTTP header (experimental)
71b640cc 285
85729c51
PH
286 The following options determine which downloader is picked:
287 external_downloader: Executable of the external downloader to call.
288 None or unset for standard (built-in) downloader.
bf09af3a
S
289 hls_prefer_native: Use the native HLS downloader instead of ffmpeg/avconv
290 if True, otherwise use ffmpeg/avconv if False, otherwise
291 use downloader suggested by extractor if None.
fe7e0c98 292
8222d8de 293 The following parameters are not used by YoutubeDL itself, they are used by
c75f0b36 294 the downloader (see youtube_dl/downloader/common.py):
8222d8de 295 nopart, updatetime, buffersize, ratelimit, min_filesize, max_filesize, test,
881e6a1f 296 noresizebuffer, retries, continuedl, noprogress, consoletitle,
7d106a65 297 xattr_set_filesize, external_downloader_args, hls_use_mpegts.
76b1bd67
JMF
298
299 The following options are used by the post processors:
300 prefer_ffmpeg: If True, use ffmpeg instead of avconv if both are available,
301 otherwise prefer avconv.
f72b0a60
S
302 postprocessor_args: A list of additional command-line arguments for the
303 postprocessor.
8222d8de
JMF
304 """
305
c9969434
S
306 _NUMERIC_FIELDS = set((
307 'width', 'height', 'tbr', 'abr', 'asr', 'vbr', 'fps', 'filesize', 'filesize_approx',
308 'timestamp', 'upload_year', 'upload_month', 'upload_day',
309 'duration', 'view_count', 'like_count', 'dislike_count', 'repost_count',
310 'average_rating', 'comment_count', 'age_limit',
311 'start_time', 'end_time',
312 'chapter_number', 'season_number', 'episode_number',
313 'track_number', 'disc_number', 'release_year',
314 'playlist_index',
315 ))
316
8222d8de
JMF
317 params = None
318 _ies = []
319 _pps = []
320 _download_retcode = None
321 _num_downloads = None
322 _screen_file = None
323
3511266b 324 def __init__(self, params=None, auto_init=True):
8222d8de 325 """Create a FileDownloader object with the given options."""
e9f9a10f
JMF
326 if params is None:
327 params = {}
8222d8de 328 self._ies = []
56c73665 329 self._ies_instances = {}
8222d8de 330 self._pps = []
933605d7 331 self._progress_hooks = []
8222d8de
JMF
332 self._download_retcode = 0
333 self._num_downloads = 0
334 self._screen_file = [sys.stdout, sys.stderr][params.get('logtostderr', False)]
0783b09b 335 self._err_file = sys.stderr
4abf617b
S
336 self.params = {
337 # Default parameters
338 'nocheckcertificate': False,
339 }
340 self.params.update(params)
a0e07d31 341 self.cache = Cache(self)
34308b30 342
be5df5ee
S
343 def check_deprecated(param, option, suggestion):
344 if self.params.get(param) is not None:
345 self.report_warning(
346 '%s is deprecated. Use %s instead.' % (option, suggestion))
347 return True
348 return False
349
350 if check_deprecated('cn_verification_proxy', '--cn-verification-proxy', '--geo-verification-proxy'):
38cce791
YCH
351 if self.params.get('geo_verification_proxy') is None:
352 self.params['geo_verification_proxy'] = self.params['cn_verification_proxy']
353
be5df5ee
S
354 check_deprecated('autonumber_size', '--autonumber-size', 'output template with %(autonumber)0Nd, where N in the number of digits')
355 check_deprecated('autonumber', '--auto-number', '-o "%(autonumber)s-%(title)s.%(ext)s"')
356 check_deprecated('usetitle', '--title', '-o "%(title)s-%(id)s.%(ext)s"')
357
0783b09b 358 if params.get('bidi_workaround', False):
1c088fa8
PH
359 try:
360 import pty
361 master, slave = pty.openpty()
003c69a8 362 width = compat_get_terminal_size().columns
1c088fa8
PH
363 if width is None:
364 width_args = []
365 else:
366 width_args = ['-w', str(width)]
5d681e96 367 sp_kwargs = dict(
1c088fa8
PH
368 stdin=subprocess.PIPE,
369 stdout=slave,
370 stderr=self._err_file)
5d681e96
PH
371 try:
372 self._output_process = subprocess.Popen(
373 ['bidiv'] + width_args, **sp_kwargs
374 )
375 except OSError:
5d681e96
PH
376 self._output_process = subprocess.Popen(
377 ['fribidi', '-c', 'UTF-8'] + width_args, **sp_kwargs)
378 self._output_channel = os.fdopen(master, 'rb')
1c088fa8 379 except OSError as ose:
66e7ace1 380 if ose.errno == errno.ENOENT:
6febd1c1 381 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 $PATH.')
1c088fa8
PH
382 else:
383 raise
0783b09b 384
e9137224 385 if (sys.platform != 'win32' and
8fb3ac36
PH
386 sys.getfilesystemencoding() in ['ascii', 'ANSI_X3.4-1968'] and
387 not params.get('restrictfilenames', False)):
e9137224 388 # Unicode filesystem API will throw errors (#1474, #13027)
34308b30 389 self.report_warning(
6febd1c1 390 'Assuming --restrict-filenames since file system encoding '
1b725173 391 'cannot encode all characters. '
6febd1c1 392 'Set the LC_ALL environment variable to fix this.')
4a98cdbf 393 self.params['restrictfilenames'] = True
34308b30 394
486dd09e
PH
395 if isinstance(params.get('outtmpl'), bytes):
396 self.report_warning(
397 'Parameter outtmpl is bytes, but should be a unicode string. '
398 'Put from __future__ import unicode_literals at the top of your code file or consider switching to Python 3.x.')
399
dca08720
PH
400 self._setup_opener()
401
3511266b
PH
402 if auto_init:
403 self.print_debug_header()
404 self.add_default_info_extractors()
405
4f026faf
PH
406 for pp_def_raw in self.params.get('postprocessors', []):
407 pp_class = get_postprocessor(pp_def_raw['key'])
408 pp_def = dict(pp_def_raw)
409 del pp_def['key']
410 pp = pp_class(self, **compat_kwargs(pp_def))
411 self.add_post_processor(pp)
412
71b640cc
PH
413 for ph in self.params.get('progress_hooks', []):
414 self.add_progress_hook(ph)
415
51fb4995
YCH
416 register_socks_protocols()
417
7d4111ed
PH
418 def warn_if_short_id(self, argv):
419 # short YouTube ID starting with dash?
420 idxs = [
421 i for i, a in enumerate(argv)
422 if re.match(r'^-[0-9A-Za-z_-]{10}$', a)]
423 if idxs:
424 correct_argv = (
425 ['youtube-dl'] +
426 [a for i, a in enumerate(argv) if i not in idxs] +
427 ['--'] + [argv[i] for i in idxs]
428 )
429 self.report_warning(
430 'Long argument string detected. '
431 'Use -- to separate parameters and URLs, like this:\n%s\n' %
432 args_to_str(correct_argv))
433
8222d8de
JMF
434 def add_info_extractor(self, ie):
435 """Add an InfoExtractor object to the end of the list."""
436 self._ies.append(ie)
e52d7f85
JMF
437 if not isinstance(ie, type):
438 self._ies_instances[ie.ie_key()] = ie
439 ie.set_downloader(self)
8222d8de 440
56c73665
JMF
441 def get_info_extractor(self, ie_key):
442 """
443 Get an instance of an IE with name ie_key, it will try to get one from
444 the _ies list, if there's no instance it will create a new one and add
445 it to the extractor list.
446 """
447 ie = self._ies_instances.get(ie_key)
448 if ie is None:
449 ie = get_info_extractor(ie_key)()
450 self.add_info_extractor(ie)
451 return ie
452
023fa8c4
JMF
453 def add_default_info_extractors(self):
454 """
455 Add the InfoExtractors returned by gen_extractors to the end of the list
456 """
e52d7f85 457 for ie in gen_extractor_classes():
023fa8c4
JMF
458 self.add_info_extractor(ie)
459
8222d8de
JMF
460 def add_post_processor(self, pp):
461 """Add a PostProcessor object to the end of the chain."""
462 self._pps.append(pp)
463 pp.set_downloader(self)
464
933605d7
JMF
465 def add_progress_hook(self, ph):
466 """Add the progress hook (currently only for the file downloader)"""
467 self._progress_hooks.append(ph)
8ab470f1 468
1c088fa8 469 def _bidi_workaround(self, message):
5d681e96 470 if not hasattr(self, '_output_channel'):
1c088fa8
PH
471 return message
472
5d681e96 473 assert hasattr(self, '_output_process')
11b85ce6 474 assert isinstance(message, compat_str)
6febd1c1
PH
475 line_count = message.count('\n') + 1
476 self._output_process.stdin.write((message + '\n').encode('utf-8'))
5d681e96 477 self._output_process.stdin.flush()
6febd1c1 478 res = ''.join(self._output_channel.readline().decode('utf-8')
9e1a5b84 479 for _ in range(line_count))
6febd1c1 480 return res[:-len('\n')]
1c088fa8 481
8222d8de 482 def to_screen(self, message, skip_eol=False):
0783b09b
PH
483 """Print message to stdout if not in quiet mode."""
484 return self.to_stdout(message, skip_eol, check_quiet=True)
485
734f90bb 486 def _write_string(self, s, out=None):
b58ddb32 487 write_string(s, out=out, encoding=self.params.get('encoding'))
734f90bb 488
0783b09b 489 def to_stdout(self, message, skip_eol=False, check_quiet=False):
8222d8de 490 """Print message to stdout if not in quiet mode."""
8bf9319e 491 if self.params.get('logger'):
43afe285 492 self.params['logger'].debug(message)
0783b09b 493 elif not check_quiet or not self.params.get('quiet', False):
1c088fa8 494 message = self._bidi_workaround(message)
6febd1c1 495 terminator = ['\n', ''][skip_eol]
8222d8de 496 output = message + terminator
1c088fa8 497
734f90bb 498 self._write_string(output, self._screen_file)
8222d8de
JMF
499
500 def to_stderr(self, message):
501 """Print message to stderr."""
11b85ce6 502 assert isinstance(message, compat_str)
8bf9319e 503 if self.params.get('logger'):
43afe285
IB
504 self.params['logger'].error(message)
505 else:
1c088fa8 506 message = self._bidi_workaround(message)
6febd1c1 507 output = message + '\n'
734f90bb 508 self._write_string(output, self._err_file)
8222d8de 509
1e5b9a95
PH
510 def to_console_title(self, message):
511 if not self.params.get('consoletitle', False):
512 return
4bede0d8
C
513 if compat_os_name == 'nt':
514 if ctypes.windll.kernel32.GetConsoleWindow():
515 # c_wchar_p() might not be necessary if `message` is
516 # already of type unicode()
517 ctypes.windll.kernel32.SetConsoleTitleW(ctypes.c_wchar_p(message))
1e5b9a95 518 elif 'TERM' in os.environ:
734f90bb 519 self._write_string('\033]0;%s\007' % message, self._screen_file)
1e5b9a95 520
bdde425c
PH
521 def save_console_title(self):
522 if not self.params.get('consoletitle', False):
523 return
4bede0d8 524 if compat_os_name != 'nt' and 'TERM' in os.environ:
efd6c574 525 # Save the title on stack
734f90bb 526 self._write_string('\033[22;0t', self._screen_file)
bdde425c
PH
527
528 def restore_console_title(self):
529 if not self.params.get('consoletitle', False):
530 return
4bede0d8 531 if compat_os_name != 'nt' and 'TERM' in os.environ:
efd6c574 532 # Restore the title from stack
734f90bb 533 self._write_string('\033[23;0t', self._screen_file)
bdde425c
PH
534
535 def __enter__(self):
536 self.save_console_title()
537 return self
538
539 def __exit__(self, *args):
540 self.restore_console_title()
f89197d7 541
dca08720
PH
542 if self.params.get('cookiefile') is not None:
543 self.cookiejar.save()
bdde425c 544
8222d8de
JMF
545 def trouble(self, message=None, tb=None):
546 """Determine action to take when a download problem appears.
547
548 Depending on if the downloader has been configured to ignore
549 download errors or not, this method may throw an exception or
550 not when errors are found, after printing the message.
551
552 tb, if given, is additional traceback information.
553 """
554 if message is not None:
555 self.to_stderr(message)
556 if self.params.get('verbose'):
557 if tb is None:
558 if sys.exc_info()[0]: # if .trouble has been called from an except block
6febd1c1 559 tb = ''
8222d8de 560 if hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
6febd1c1 561 tb += ''.join(traceback.format_exception(*sys.exc_info()[1].exc_info))
c0384f22 562 tb += encode_compat_str(traceback.format_exc())
8222d8de
JMF
563 else:
564 tb_data = traceback.format_list(traceback.extract_stack())
6febd1c1 565 tb = ''.join(tb_data)
8222d8de
JMF
566 self.to_stderr(tb)
567 if not self.params.get('ignoreerrors', False):
568 if sys.exc_info()[0] and hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
569 exc_info = sys.exc_info()[1].exc_info
570 else:
571 exc_info = sys.exc_info()
572 raise DownloadError(message, exc_info)
573 self._download_retcode = 1
574
575 def report_warning(self, message):
576 '''
577 Print the message to stderr, it will be prefixed with 'WARNING:'
578 If stderr is a tty file the 'WARNING:' will be colored
579 '''
6d07ce01
JMF
580 if self.params.get('logger') is not None:
581 self.params['logger'].warning(message)
8222d8de 582 else:
ad8915b7
PH
583 if self.params.get('no_warnings'):
584 return
e9c0cdd3 585 if not self.params.get('no_color') and self._err_file.isatty() and compat_os_name != 'nt':
6d07ce01
JMF
586 _msg_header = '\033[0;33mWARNING:\033[0m'
587 else:
588 _msg_header = 'WARNING:'
589 warning_message = '%s %s' % (_msg_header, message)
590 self.to_stderr(warning_message)
8222d8de
JMF
591
592 def report_error(self, message, tb=None):
593 '''
594 Do the same as trouble, but prefixes the message with 'ERROR:', colored
595 in red if stderr is a tty file.
596 '''
e9c0cdd3 597 if not self.params.get('no_color') and self._err_file.isatty() and compat_os_name != 'nt':
6febd1c1 598 _msg_header = '\033[0;31mERROR:\033[0m'
8222d8de 599 else:
6febd1c1
PH
600 _msg_header = 'ERROR:'
601 error_message = '%s %s' % (_msg_header, message)
8222d8de
JMF
602 self.trouble(error_message, tb)
603
8222d8de
JMF
604 def report_file_already_downloaded(self, file_name):
605 """Report file has already been fully downloaded."""
606 try:
6febd1c1 607 self.to_screen('[download] %s has already been downloaded' % file_name)
ce02ed60 608 except UnicodeEncodeError:
6febd1c1 609 self.to_screen('[download] The file has already been downloaded')
8222d8de 610
8222d8de
JMF
611 def prepare_filename(self, info_dict):
612 """Generate the output filename."""
613 try:
614 template_dict = dict(info_dict)
615
616 template_dict['epoch'] = int(time.time())
617 autonumber_size = self.params.get('autonumber_size')
618 if autonumber_size is None:
619 autonumber_size = 5
89db639d 620 template_dict['autonumber'] = self.params.get('autonumber_start', 1) - 1 + self._num_downloads
17b75c0d
PH
621 if template_dict.get('resolution') is None:
622 if template_dict.get('width') and template_dict.get('height'):
623 template_dict['resolution'] = '%dx%d' % (template_dict['width'], template_dict['height'])
624 elif template_dict.get('height'):
805ef3c6 625 template_dict['resolution'] = '%sp' % template_dict['height']
17b75c0d 626 elif template_dict.get('width'):
51ce9117 627 template_dict['resolution'] = '%dx?' % template_dict['width']
8222d8de 628
586a91b6 629 sanitize = lambda k, v: sanitize_filename(
45598aab 630 compat_str(v),
1bb5c511 631 restricted=self.params.get('restrictfilenames'),
40df485f 632 is_id=(k == 'id' or k.endswith('_id')))
d0d9ade4 633 template_dict = dict((k, v if isinstance(v, compat_numeric_types) else sanitize(k, v))
45598aab 634 for k, v in template_dict.items()
f0e14fdd 635 if v is not None and not isinstance(v, (list, tuple, dict)))
6febd1c1 636 template_dict = collections.defaultdict(lambda: 'NA', template_dict)
8222d8de 637
b3613d36 638 outtmpl = self.params.get('outtmpl', DEFAULT_OUTTMPL)
d0d9ade4 639
89db639d
S
640 # For fields playlist_index and autonumber convert all occurrences
641 # of %(field)s to %(field)0Nd for backward compatibility
642 field_size_compat_map = {
643 'playlist_index': len(str(template_dict['n_entries'])),
644 'autonumber': autonumber_size,
645 }
646 FIELD_SIZE_COMPAT_RE = r'(?<!%)%\((?P<field>autonumber|playlist_index)\)s'
647 mobj = re.search(FIELD_SIZE_COMPAT_RE, outtmpl)
648 if mobj:
649 outtmpl = re.sub(
650 FIELD_SIZE_COMPAT_RE,
651 r'%%(\1)0%dd' % field_size_compat_map[mobj.group('field')],
652 outtmpl)
653
d0d9ade4
S
654 # Missing numeric fields used together with integer presentation types
655 # in format specification will break the argument substitution since
656 # string 'NA' is returned for missing fields. We will patch output
657 # template for missing fields to meet string presentation type.
c9969434 658 for numeric_field in self._NUMERIC_FIELDS:
d0d9ade4
S
659 if numeric_field not in template_dict:
660 # As of [1] format syntax is:
661 # %[mapping_key][conversion_flags][minimum_width][.precision][length_modifier]type
662 # 1. https://docs.python.org/2/library/stdtypes.html#string-formatting
663 FORMAT_RE = r'''(?x)
664 (?<!%)
665 %
666 \({0}\) # mapping key
667 (?:[#0\-+ ]+)? # conversion flags (optional)
668 (?:\d+)? # minimum field width (optional)
669 (?:\.\d+)? # precision (optional)
670 [hlL]? # length modifier (optional)
671 [diouxXeEfFgGcrs%] # conversion type
672 '''
673 outtmpl = re.sub(
674 FORMAT_RE.format(numeric_field),
675 r'%({0})s'.format(numeric_field), outtmpl)
676
d35dc344 677 filename = expand_path(outtmpl % template_dict)
3a0d2f52
S
678 # Temporary fix for #4787
679 # 'Treat' all problem characters by passing filename through preferredencoding
680 # to workaround encoding issues with subprocess on python2 @ Windows
681 if sys.version_info < (3, 0) and sys.platform == 'win32':
682 filename = encodeFilename(filename, True).decode(preferredencoding())
b3613d36 683 return sanitize_path(filename)
8222d8de 684 except ValueError as err:
6febd1c1 685 self.report_error('Error in output template: ' + str(err) + ' (encoding: ' + repr(preferredencoding()) + ')')
8222d8de
JMF
686 return None
687
442c37b7 688 def _match_entry(self, info_dict, incomplete):
6ec6cb4e 689 """ Returns None iff the file should be downloaded """
8222d8de 690
6febd1c1 691 video_title = info_dict.get('title', info_dict.get('id', 'video'))
7012b23c
PH
692 if 'title' in info_dict:
693 # This can happen when we're just evaluating the playlist
694 title = info_dict['title']
695 matchtitle = self.params.get('matchtitle', False)
696 if matchtitle:
697 if not re.search(matchtitle, title, re.IGNORECASE):
6febd1c1 698 return '"' + title + '" title did not match pattern "' + matchtitle + '"'
7012b23c
PH
699 rejecttitle = self.params.get('rejecttitle', False)
700 if rejecttitle:
701 if re.search(rejecttitle, title, re.IGNORECASE):
6febd1c1 702 return '"' + title + '" title matched reject pattern "' + rejecttitle + '"'
d800609c 703 date = info_dict.get('upload_date')
8222d8de
JMF
704 if date is not None:
705 dateRange = self.params.get('daterange', DateRange())
706 if date not in dateRange:
6febd1c1 707 return '%s upload date is not in range %s' % (date_from_str(date).isoformat(), dateRange)
d800609c 708 view_count = info_dict.get('view_count')
5fe18bdb
PH
709 if view_count is not None:
710 min_views = self.params.get('min_views')
711 if min_views is not None and view_count < min_views:
6febd1c1 712 return 'Skipping %s, because it has not reached minimum view count (%d/%d)' % (video_title, view_count, min_views)
5fe18bdb
PH
713 max_views = self.params.get('max_views')
714 if max_views is not None and view_count > max_views:
6febd1c1 715 return 'Skipping %s, because it has exceeded the maximum view count (%d/%d)' % (video_title, view_count, max_views)
05900629 716 if age_restricted(info_dict.get('age_limit'), self.params.get('age_limit')):
347de493 717 return 'Skipping "%s" because it is age restricted' % video_title
c1c9a79c 718 if self.in_download_archive(info_dict):
6febd1c1 719 return '%s has already been recorded in archive' % video_title
347de493 720
442c37b7
PH
721 if not incomplete:
722 match_filter = self.params.get('match_filter')
723 if match_filter is not None:
724 ret = match_filter(info_dict)
725 if ret is not None:
726 return ret
347de493 727
8222d8de 728 return None
fe7e0c98 729
b6c45014
JMF
730 @staticmethod
731 def add_extra_info(info_dict, extra_info):
732 '''Set the keys from extra_info in info dict if they are missing'''
733 for key, value in extra_info.items():
734 info_dict.setdefault(key, value)
735
7fc3fa05 736 def extract_info(self, url, download=True, ie_key=None, extra_info={},
61aa5ba3 737 process=True, force_generic_extractor=False):
8222d8de
JMF
738 '''
739 Returns a list with a dictionary for each video we find.
740 If 'download', also downloads the videos.
741 extra_info is a dict containing the extra values to add to each result
613b2d9d 742 '''
fe7e0c98 743
61aa5ba3 744 if not ie_key and force_generic_extractor:
d22dec74
S
745 ie_key = 'Generic'
746
8222d8de 747 if ie_key:
56c73665 748 ies = [self.get_info_extractor(ie_key)]
8222d8de
JMF
749 else:
750 ies = self._ies
751
752 for ie in ies:
753 if not ie.suitable(url):
754 continue
755
e52d7f85 756 ie = self.get_info_extractor(ie.ie_key())
8222d8de 757 if not ie.working():
6febd1c1
PH
758 self.report_warning('The program functionality for this site has been marked as broken, '
759 'and will probably not work.')
8222d8de
JMF
760
761 try:
762 ie_result = ie.extract(url)
5f6a1245 763 if ie_result is None: # Finished already (backwards compatibility; listformats and friends should be moved here)
8222d8de
JMF
764 break
765 if isinstance(ie_result, list):
766 # Backwards compatibility: old IE result format
8222d8de
JMF
767 ie_result = {
768 '_type': 'compat_list',
769 'entries': ie_result,
770 }
ea38e55f 771 self.add_default_extra_info(ie_result, ie, url)
7fc3fa05
PH
772 if process:
773 return self.process_ie_result(ie_result, download, extra_info)
774 else:
775 return ie_result
773f291d
S
776 except GeoRestrictedError as e:
777 msg = e.msg
778 if e.countries:
779 msg += '\nThis video is available in %s.' % ', '.join(
780 map(ISO3166Utils.short2full, e.countries))
781 msg += '\nYou might want to use a VPN or a proxy server (with --proxy) to workaround.'
782 self.report_error(msg)
783 break
fb043a6e 784 except ExtractorError as e: # An error we somewhat expected
2c74e6fa 785 self.report_error(compat_str(e), e.format_traceback())
8222d8de 786 break
d3e5bbf4
PH
787 except MaxDownloadsReached:
788 raise
8222d8de
JMF
789 except Exception as e:
790 if self.params.get('ignoreerrors', False):
9b9c5355 791 self.report_error(error_to_compat_str(e), tb=encode_compat_str(traceback.format_exc()))
8222d8de
JMF
792 break
793 else:
794 raise
795 else:
1a489545 796 self.report_error('no suitable InfoExtractor for URL %s' % url)
fe7e0c98 797
ea38e55f
PH
798 def add_default_extra_info(self, ie_result, ie, url):
799 self.add_extra_info(ie_result, {
800 'extractor': ie.IE_NAME,
801 'webpage_url': url,
802 'webpage_url_basename': url_basename(url),
803 'extractor_key': ie.ie_key(),
804 })
805
8222d8de
JMF
806 def process_ie_result(self, ie_result, download=True, extra_info={}):
807 """
808 Take the result of the ie(may be modified) and resolve all unresolved
809 references (URLs, playlist items).
810
811 It will also download the videos if 'download'.
812 Returns the resolved ie_result.
813 """
e8ee972c
PH
814 result_type = ie_result.get('_type', 'video')
815
057a5206 816 if result_type in ('url', 'url_transparent'):
134c6ea8 817 ie_result['url'] = sanitize_url(ie_result['url'])
057a5206
PH
818 extract_flat = self.params.get('extract_flat', False)
819 if ((extract_flat == 'in_playlist' and 'playlist' in extra_info) or
820 extract_flat is True):
057a5206
PH
821 if self.params.get('forcejson', False):
822 self.to_stdout(json.dumps(ie_result))
e8ee972c
PH
823 return ie_result
824
8222d8de 825 if result_type == 'video':
b6c45014 826 self.add_extra_info(ie_result, extra_info)
feee2ecf 827 return self.process_video_result(ie_result, download=download)
8222d8de
JMF
828 elif result_type == 'url':
829 # We have to add extra_info to the results because it may be
830 # contained in a playlist
831 return self.extract_info(ie_result['url'],
832 download,
833 ie_key=ie_result.get('ie_key'),
834 extra_info=extra_info)
7fc3fa05
PH
835 elif result_type == 'url_transparent':
836 # Use the information from the embedding page
837 info = self.extract_info(
838 ie_result['url'], ie_key=ie_result.get('ie_key'),
839 extra_info=extra_info, download=False, process=False)
840
1640eb09
S
841 # extract_info may return None when ignoreerrors is enabled and
842 # extraction failed with an error, don't crash and return early
843 # in this case
844 if not info:
845 return info
846
412c617d
PH
847 force_properties = dict(
848 (k, v) for k, v in ie_result.items() if v is not None)
b286f201 849 for f in ('_type', 'url', 'ie_key'):
412c617d
PH
850 if f in force_properties:
851 del force_properties[f]
852 new_result = info.copy()
853 new_result.update(force_properties)
7fc3fa05 854
0563f7ac
S
855 # Extracted info may not be a video result (i.e.
856 # info.get('_type', 'video') != video) but rather an url or
857 # url_transparent. In such cases outer metadata (from ie_result)
858 # should be propagated to inner one (info). For this to happen
859 # _type of info should be overridden with url_transparent. This
860 # fixes issue from https://github.com/rg3/youtube-dl/pull/11163.
861 if new_result.get('_type') == 'url':
862 new_result['_type'] = 'url_transparent'
7fc3fa05
PH
863
864 return self.process_ie_result(
865 new_result, download=download, extra_info=extra_info)
40fcba5e 866 elif result_type in ('playlist', 'multi_video'):
8222d8de 867 # We process each entry in the playlist
d800609c 868 playlist = ie_result.get('title') or ie_result.get('id')
6febd1c1 869 self.to_screen('[download] Downloading playlist: %s' % playlist)
8222d8de
JMF
870
871 playlist_results = []
872
8222d8de 873 playliststart = self.params.get('playliststart', 1) - 1
d800609c 874 playlistend = self.params.get('playlistend')
a19fd00c 875 # For backwards compatibility, interpret -1 as whole list
8222d8de 876 if playlistend == -1:
a19fd00c 877 playlistend = None
8222d8de 878
d800609c 879 playlistitems_str = self.params.get('playlist_items')
c14e88f0
PH
880 playlistitems = None
881 if playlistitems_str is not None:
882 def iter_playlistitems(format):
883 for string_segment in format.split(','):
884 if '-' in string_segment:
885 start, end = string_segment.split('-')
886 for item in range(int(start), int(end) + 1):
887 yield int(item)
888 else:
889 yield int(string_segment)
890 playlistitems = iter_playlistitems(playlistitems_str)
891
b82f815f
PH
892 ie_entries = ie_result['entries']
893 if isinstance(ie_entries, list):
894 n_all_entries = len(ie_entries)
c14e88f0 895 if playlistitems:
3884dcf3
JMF
896 entries = [
897 ie_entries[i - 1] for i in playlistitems
898 if -n_all_entries <= i - 1 < n_all_entries]
c14e88f0
PH
899 else:
900 entries = ie_entries[playliststart:playlistend]
b7ab0590
PH
901 n_entries = len(entries)
902 self.to_screen(
611c1dd9 903 '[%s] playlist %s: Collected %d video ids (downloading %d of them)' %
b7ab0590 904 (ie_result['extractor'], playlist, n_all_entries, n_entries))
b82f815f 905 elif isinstance(ie_entries, PagedList):
c14e88f0
PH
906 if playlistitems:
907 entries = []
908 for item in playlistitems:
909 entries.extend(ie_entries.getslice(
910 item - 1, item
911 ))
912 else:
913 entries = ie_entries.getslice(
914 playliststart, playlistend)
b7ab0590
PH
915 n_entries = len(entries)
916 self.to_screen(
611c1dd9 917 '[%s] playlist %s: Downloading %d videos' %
b7ab0590 918 (ie_result['extractor'], playlist, n_entries))
b82f815f 919 else: # iterable
c14e88f0
PH
920 if playlistitems:
921 entry_list = list(ie_entries)
922 entries = [entry_list[i - 1] for i in playlistitems]
923 else:
924 entries = list(itertools.islice(
925 ie_entries, playliststart, playlistend))
b82f815f
PH
926 n_entries = len(entries)
927 self.to_screen(
611c1dd9 928 '[%s] playlist %s: Downloading %d videos' %
b82f815f 929 (ie_result['extractor'], playlist, n_entries))
8222d8de 930
ff815fe6
MS
931 if self.params.get('playlistreverse', False):
932 entries = entries[::-1]
933
75822ca7
TC
934 if self.params.get('playlistrandom', False):
935 random.shuffle(entries)
936
0016b84e
S
937 x_forwarded_for = ie_result.get('__x_forwarded_for_ip')
938
fe7e0c98 939 for i, entry in enumerate(entries, 1):
734ea11e 940 self.to_screen('[download] Downloading video %s of %s' % (i, n_entries))
0016b84e
S
941 # This __x_forwarded_for_ip thing is a bit ugly but requires
942 # minimal changes
943 if x_forwarded_for:
944 entry['__x_forwarded_for_ip'] = x_forwarded_for
8222d8de 945 extra = {
c6b4132a 946 'n_entries': n_entries,
fe7e0c98 947 'playlist': playlist,
a1cf99d0
PH
948 'playlist_id': ie_result.get('id'),
949 'playlist_title': ie_result.get('title'),
fe7e0c98 950 'playlist_index': i + playliststart,
b6c45014 951 'extractor': ie_result['extractor'],
9103bbc5 952 'webpage_url': ie_result['webpage_url'],
29eb5174 953 'webpage_url_basename': url_basename(ie_result['webpage_url']),
be97abc2 954 'extractor_key': ie_result['extractor_key'],
fe7e0c98 955 }
7012b23c 956
442c37b7 957 reason = self._match_entry(entry, incomplete=True)
7012b23c 958 if reason is not None:
6febd1c1 959 self.to_screen('[download] ' + reason)
7012b23c
PH
960 continue
961
8222d8de
JMF
962 entry_result = self.process_ie_result(entry,
963 download=download,
964 extra_info=extra)
965 playlist_results.append(entry_result)
966 ie_result['entries'] = playlist_results
371c3b79 967 self.to_screen('[download] Finished downloading playlist: %s' % playlist)
8222d8de
JMF
968 return ie_result
969 elif result_type == 'compat_list':
c9bf4114
PH
970 self.report_warning(
971 'Extractor %s returned a compat_list result. '
972 'It needs to be updated.' % ie_result.get('extractor'))
5f6a1245 973
8222d8de 974 def _fixup(r):
9e1a5b84
JW
975 self.add_extra_info(
976 r,
9103bbc5
JMF
977 {
978 'extractor': ie_result['extractor'],
979 'webpage_url': ie_result['webpage_url'],
29eb5174 980 'webpage_url_basename': url_basename(ie_result['webpage_url']),
be97abc2 981 'extractor_key': ie_result['extractor_key'],
9e1a5b84
JW
982 }
983 )
8222d8de
JMF
984 return r
985 ie_result['entries'] = [
b6c45014 986 self.process_ie_result(_fixup(r), download, extra_info)
8222d8de
JMF
987 for r in ie_result['entries']
988 ]
989 return ie_result
990 else:
991 raise Exception('Invalid result type: %s' % result_type)
992
67134eab
JMF
993 def _build_format_filter(self, filter_spec):
994 " Returns a function to filter the formats according to the filter_spec "
083c9df9
PH
995
996 OPERATORS = {
997 '<': operator.lt,
998 '<=': operator.le,
999 '>': operator.gt,
1000 '>=': operator.ge,
1001 '=': operator.eq,
1002 '!=': operator.ne,
1003 }
67134eab 1004 operator_rex = re.compile(r'''(?x)\s*
2ec19e95 1005 (?P<key>width|height|tbr|abr|vbr|asr|filesize|fps)
083c9df9
PH
1006 \s*(?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
1007 (?P<value>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?)
67134eab 1008 $
083c9df9 1009 ''' % '|'.join(map(re.escape, OPERATORS.keys())))
67134eab 1010 m = operator_rex.search(filter_spec)
9ddb6925
S
1011 if m:
1012 try:
1013 comparison_value = int(m.group('value'))
1014 except ValueError:
1015 comparison_value = parse_filesize(m.group('value'))
1016 if comparison_value is None:
1017 comparison_value = parse_filesize(m.group('value') + 'B')
1018 if comparison_value is None:
1019 raise ValueError(
1020 'Invalid value %r in format specification %r' % (
67134eab 1021 m.group('value'), filter_spec))
9ddb6925
S
1022 op = OPERATORS[m.group('op')]
1023
083c9df9 1024 if not m:
9ddb6925
S
1025 STR_OPERATORS = {
1026 '=': operator.eq,
1027 '!=': operator.ne,
10d33b34
YCH
1028 '^=': lambda attr, value: attr.startswith(value),
1029 '$=': lambda attr, value: attr.endswith(value),
1030 '*=': lambda attr, value: value in attr,
9ddb6925 1031 }
67134eab 1032 str_operator_rex = re.compile(r'''(?x)
d5aacf9a 1033 \s*(?P<key>ext|acodec|vcodec|container|protocol|format_id)
9ddb6925 1034 \s*(?P<op>%s)(?P<none_inclusive>\s*\?)?
b0df5223 1035 \s*(?P<value>[a-zA-Z0-9._-]+)
67134eab 1036 \s*$
9ddb6925 1037 ''' % '|'.join(map(re.escape, STR_OPERATORS.keys())))
67134eab 1038 m = str_operator_rex.search(filter_spec)
9ddb6925
S
1039 if m:
1040 comparison_value = m.group('value')
1041 op = STR_OPERATORS[m.group('op')]
083c9df9 1042
9ddb6925 1043 if not m:
67134eab 1044 raise ValueError('Invalid filter specification %r' % filter_spec)
083c9df9
PH
1045
1046 def _filter(f):
1047 actual_value = f.get(m.group('key'))
1048 if actual_value is None:
1049 return m.group('none_inclusive')
1050 return op(actual_value, comparison_value)
67134eab
JMF
1051 return _filter
1052
1053 def build_format_selector(self, format_spec):
1054 def syntax_error(note, start):
1055 message = (
1056 'Invalid format specification: '
1057 '{0}\n\t{1}\n\t{2}^'.format(note, format_spec, ' ' * start[1]))
1058 return SyntaxError(message)
1059
1060 PICKFIRST = 'PICKFIRST'
1061 MERGE = 'MERGE'
1062 SINGLE = 'SINGLE'
0130afb7 1063 GROUP = 'GROUP'
67134eab
JMF
1064 FormatSelector = collections.namedtuple('FormatSelector', ['type', 'selector', 'filters'])
1065
1066 def _parse_filter(tokens):
1067 filter_parts = []
1068 for type, string, start, _, _ in tokens:
1069 if type == tokenize.OP and string == ']':
1070 return ''.join(filter_parts)
1071 else:
1072 filter_parts.append(string)
1073
232541df 1074 def _remove_unused_ops(tokens):
17cc1534 1075 # Remove operators that we don't use and join them with the surrounding strings
232541df
JMF
1076 # for example: 'mp4' '-' 'baseline' '-' '16x9' is converted to 'mp4-baseline-16x9'
1077 ALLOWED_OPS = ('/', '+', ',', '(', ')')
1078 last_string, last_start, last_end, last_line = None, None, None, None
1079 for type, string, start, end, line in tokens:
1080 if type == tokenize.OP and string == '[':
1081 if last_string:
1082 yield tokenize.NAME, last_string, last_start, last_end, last_line
1083 last_string = None
1084 yield type, string, start, end, line
1085 # everything inside brackets will be handled by _parse_filter
1086 for type, string, start, end, line in tokens:
1087 yield type, string, start, end, line
1088 if type == tokenize.OP and string == ']':
1089 break
1090 elif type == tokenize.OP and string in ALLOWED_OPS:
1091 if last_string:
1092 yield tokenize.NAME, last_string, last_start, last_end, last_line
1093 last_string = None
1094 yield type, string, start, end, line
1095 elif type in [tokenize.NAME, tokenize.NUMBER, tokenize.OP]:
1096 if not last_string:
1097 last_string = string
1098 last_start = start
1099 last_end = end
1100 else:
1101 last_string += string
1102 if last_string:
1103 yield tokenize.NAME, last_string, last_start, last_end, last_line
1104
cf2ac6df 1105 def _parse_format_selection(tokens, inside_merge=False, inside_choice=False, inside_group=False):
67134eab
JMF
1106 selectors = []
1107 current_selector = None
1108 for type, string, start, _, _ in tokens:
1109 # ENCODING is only defined in python 3.x
1110 if type == getattr(tokenize, 'ENCODING', None):
1111 continue
1112 elif type in [tokenize.NAME, tokenize.NUMBER]:
1113 current_selector = FormatSelector(SINGLE, string, [])
1114 elif type == tokenize.OP:
cf2ac6df
JMF
1115 if string == ')':
1116 if not inside_group:
1117 # ')' will be handled by the parentheses group
1118 tokens.restore_last_token()
67134eab 1119 break
cf2ac6df 1120 elif inside_merge and string in ['/', ',']:
0130afb7
JMF
1121 tokens.restore_last_token()
1122 break
cf2ac6df
JMF
1123 elif inside_choice and string == ',':
1124 tokens.restore_last_token()
1125 break
1126 elif string == ',':
0a31a350
JMF
1127 if not current_selector:
1128 raise syntax_error('"," must follow a format selector', start)
67134eab
JMF
1129 selectors.append(current_selector)
1130 current_selector = None
1131 elif string == '/':
d96d604e
JMF
1132 if not current_selector:
1133 raise syntax_error('"/" must follow a format selector', start)
67134eab 1134 first_choice = current_selector
cf2ac6df 1135 second_choice = _parse_format_selection(tokens, inside_choice=True)
f5f4a27a 1136 current_selector = FormatSelector(PICKFIRST, (first_choice, second_choice), [])
67134eab
JMF
1137 elif string == '[':
1138 if not current_selector:
1139 current_selector = FormatSelector(SINGLE, 'best', [])
1140 format_filter = _parse_filter(tokens)
1141 current_selector.filters.append(format_filter)
0130afb7
JMF
1142 elif string == '(':
1143 if current_selector:
1144 raise syntax_error('Unexpected "("', start)
cf2ac6df
JMF
1145 group = _parse_format_selection(tokens, inside_group=True)
1146 current_selector = FormatSelector(GROUP, group, [])
67134eab
JMF
1147 elif string == '+':
1148 video_selector = current_selector
cf2ac6df 1149 audio_selector = _parse_format_selection(tokens, inside_merge=True)
0a31a350
JMF
1150 if not video_selector or not audio_selector:
1151 raise syntax_error('"+" must be between two format selectors', start)
cf2ac6df 1152 current_selector = FormatSelector(MERGE, (video_selector, audio_selector), [])
67134eab
JMF
1153 else:
1154 raise syntax_error('Operator not recognized: "{0}"'.format(string), start)
1155 elif type == tokenize.ENDMARKER:
1156 break
1157 if current_selector:
1158 selectors.append(current_selector)
1159 return selectors
1160
1161 def _build_selector_function(selector):
1162 if isinstance(selector, list):
1163 fs = [_build_selector_function(s) for s in selector]
1164
317f7ab6 1165 def selector_function(ctx):
67134eab 1166 for f in fs:
317f7ab6 1167 for format in f(ctx):
67134eab
JMF
1168 yield format
1169 return selector_function
0130afb7
JMF
1170 elif selector.type == GROUP:
1171 selector_function = _build_selector_function(selector.selector)
67134eab
JMF
1172 elif selector.type == PICKFIRST:
1173 fs = [_build_selector_function(s) for s in selector.selector]
1174
317f7ab6 1175 def selector_function(ctx):
67134eab 1176 for f in fs:
317f7ab6 1177 picked_formats = list(f(ctx))
67134eab
JMF
1178 if picked_formats:
1179 return picked_formats
1180 return []
1181 elif selector.type == SINGLE:
1182 format_spec = selector.selector
1183
317f7ab6
S
1184 def selector_function(ctx):
1185 formats = list(ctx['formats'])
bb8e5536
JMF
1186 if not formats:
1187 return
5acfa126
JMF
1188 if format_spec == 'all':
1189 for f in formats:
1190 yield f
1191 elif format_spec in ['best', 'worst', None]:
67134eab
JMF
1192 format_idx = 0 if format_spec == 'worst' else -1
1193 audiovideo_formats = [
1194 f for f in formats
1195 if f.get('vcodec') != 'none' and f.get('acodec') != 'none']
1196 if audiovideo_formats:
1197 yield audiovideo_formats[format_idx]
317f7ab6
S
1198 # for extractors with incomplete formats (audio only (soundcloud)
1199 # or video only (imgur)) we will fallback to best/worst
1200 # {video,audio}-only format
1201 elif ctx['incomplete_formats']:
67134eab
JMF
1202 yield formats[format_idx]
1203 elif format_spec == 'bestaudio':
1204 audio_formats = [
1205 f for f in formats
1206 if f.get('vcodec') == 'none']
1207 if audio_formats:
1208 yield audio_formats[-1]
1209 elif format_spec == 'worstaudio':
1210 audio_formats = [
1211 f for f in formats
1212 if f.get('vcodec') == 'none']
1213 if audio_formats:
1214 yield audio_formats[0]
1215 elif format_spec == 'bestvideo':
1216 video_formats = [
1217 f for f in formats
1218 if f.get('acodec') == 'none']
1219 if video_formats:
1220 yield video_formats[-1]
1221 elif format_spec == 'worstvideo':
1222 video_formats = [
1223 f for f in formats
1224 if f.get('acodec') == 'none']
1225 if video_formats:
1226 yield video_formats[0]
1227 else:
1228 extensions = ['mp4', 'flv', 'webm', '3gp', 'm4a', 'mp3', 'ogg', 'aac', 'wav']
1229 if format_spec in extensions:
1230 filter_f = lambda f: f['ext'] == format_spec
1231 else:
1232 filter_f = lambda f: f['format_id'] == format_spec
1233 matches = list(filter(filter_f, formats))
1234 if matches:
1235 yield matches[-1]
1236 elif selector.type == MERGE:
1237 def _merge(formats_info):
1238 format_1, format_2 = [f['format_id'] for f in formats_info]
1239 # The first format must contain the video and the
1240 # second the audio
1241 if formats_info[0].get('vcodec') == 'none':
1242 self.report_error('The first format must '
1243 'contain the video, try using '
1244 '"-f %s+%s"' % (format_2, format_1))
1245 return
3d24bbfb
S
1246 # Formats must be opposite (video+audio)
1247 if formats_info[0].get('acodec') == 'none' and formats_info[1].get('acodec') == 'none':
1248 self.report_error(
1249 'Both formats %s and %s are video-only, you must specify "-f video+audio"'
1250 % (format_1, format_2))
1251 return
67134eab
JMF
1252 output_ext = (
1253 formats_info[0]['ext']
1254 if self.params.get('merge_output_format') is None
1255 else self.params['merge_output_format'])
1256 return {
1257 'requested_formats': formats_info,
1258 'format': '%s+%s' % (formats_info[0].get('format'),
1259 formats_info[1].get('format')),
1260 'format_id': '%s+%s' % (formats_info[0].get('format_id'),
1261 formats_info[1].get('format_id')),
1262 'width': formats_info[0].get('width'),
1263 'height': formats_info[0].get('height'),
1264 'resolution': formats_info[0].get('resolution'),
1265 'fps': formats_info[0].get('fps'),
1266 'vcodec': formats_info[0].get('vcodec'),
1267 'vbr': formats_info[0].get('vbr'),
1268 'stretched_ratio': formats_info[0].get('stretched_ratio'),
1269 'acodec': formats_info[1].get('acodec'),
1270 'abr': formats_info[1].get('abr'),
1271 'ext': output_ext,
1272 }
1273 video_selector, audio_selector = map(_build_selector_function, selector.selector)
083c9df9 1274
317f7ab6
S
1275 def selector_function(ctx):
1276 for pair in itertools.product(
1277 video_selector(copy.deepcopy(ctx)), audio_selector(copy.deepcopy(ctx))):
67134eab 1278 yield _merge(pair)
083c9df9 1279
67134eab 1280 filters = [self._build_format_filter(f) for f in selector.filters]
083c9df9 1281
317f7ab6
S
1282 def final_selector(ctx):
1283 ctx_copy = copy.deepcopy(ctx)
67134eab 1284 for _filter in filters:
317f7ab6
S
1285 ctx_copy['formats'] = list(filter(_filter, ctx_copy['formats']))
1286 return selector_function(ctx_copy)
67134eab 1287 return final_selector
083c9df9 1288
67134eab 1289 stream = io.BytesIO(format_spec.encode('utf-8'))
0130afb7 1290 try:
232541df 1291 tokens = list(_remove_unused_ops(compat_tokenize_tokenize(stream.readline)))
0130afb7
JMF
1292 except tokenize.TokenError:
1293 raise syntax_error('Missing closing/opening brackets or parenthesis', (0, len(format_spec)))
1294
1295 class TokenIterator(object):
1296 def __init__(self, tokens):
1297 self.tokens = tokens
1298 self.counter = 0
1299
1300 def __iter__(self):
1301 return self
1302
1303 def __next__(self):
1304 if self.counter >= len(self.tokens):
1305 raise StopIteration()
1306 value = self.tokens[self.counter]
1307 self.counter += 1
1308 return value
1309
1310 next = __next__
1311
1312 def restore_last_token(self):
1313 self.counter -= 1
1314
1315 parsed_selector = _parse_format_selection(iter(TokenIterator(tokens)))
67134eab 1316 return _build_selector_function(parsed_selector)
a9c58ad9 1317
e5660ee6
JMF
1318 def _calc_headers(self, info_dict):
1319 res = std_headers.copy()
1320
1321 add_headers = info_dict.get('http_headers')
1322 if add_headers:
1323 res.update(add_headers)
1324
1325 cookies = self._calc_cookies(info_dict)
1326 if cookies:
1327 res['Cookie'] = cookies
1328
0016b84e
S
1329 if 'X-Forwarded-For' not in res:
1330 x_forwarded_for_ip = info_dict.get('__x_forwarded_for_ip')
1331 if x_forwarded_for_ip:
1332 res['X-Forwarded-For'] = x_forwarded_for_ip
1333
e5660ee6
JMF
1334 return res
1335
1336 def _calc_cookies(self, info_dict):
5c2266df 1337 pr = sanitized_Request(info_dict['url'])
e5660ee6 1338 self.cookiejar.add_cookie_header(pr)
662435f7 1339 return pr.get_header('Cookie')
e5660ee6 1340
dd82ffea
JMF
1341 def process_video_result(self, info_dict, download=True):
1342 assert info_dict.get('_type', 'video') == 'video'
1343
bec1fad2
PH
1344 if 'id' not in info_dict:
1345 raise ExtractorError('Missing "id" field in extractor result')
1346 if 'title' not in info_dict:
1347 raise ExtractorError('Missing "title" field in extractor result')
1348
c9969434
S
1349 def report_force_conversion(field, field_not, conversion):
1350 self.report_warning(
1351 '"%s" field is not %s - forcing %s conversion, there is an error in extractor'
1352 % (field, field_not, conversion))
1353
1354 def sanitize_string_field(info, string_field):
1355 field = info.get(string_field)
1356 if field is None or isinstance(field, compat_str):
1357 return
1358 report_force_conversion(string_field, 'a string', 'string')
1359 info[string_field] = compat_str(field)
1360
1361 def sanitize_numeric_fields(info):
1362 for numeric_field in self._NUMERIC_FIELDS:
1363 field = info.get(numeric_field)
1364 if field is None or isinstance(field, compat_numeric_types):
1365 continue
1366 report_force_conversion(numeric_field, 'numeric', 'int')
1367 info[numeric_field] = int_or_none(field)
1368
1369 sanitize_string_field(info_dict, 'id')
1370 sanitize_numeric_fields(info_dict)
be6217b2 1371
dd82ffea
JMF
1372 if 'playlist' not in info_dict:
1373 # It isn't part of a playlist
1374 info_dict['playlist'] = None
1375 info_dict['playlist_index'] = None
1376
d5519808 1377 thumbnails = info_dict.get('thumbnails')
cfb56d1a
PH
1378 if thumbnails is None:
1379 thumbnail = info_dict.get('thumbnail')
1380 if thumbnail:
a7a14d95 1381 info_dict['thumbnails'] = thumbnails = [{'url': thumbnail}]
d5519808 1382 if thumbnails:
be6d7229 1383 thumbnails.sort(key=lambda t: (
d37708fc
RA
1384 t.get('preference') if t.get('preference') is not None else -1,
1385 t.get('width') if t.get('width') is not None else -1,
1386 t.get('height') if t.get('height') is not None else -1,
1387 t.get('id') if t.get('id') is not None else '', t.get('url')))
f6c24009 1388 for i, t in enumerate(thumbnails):
dcf77cf1 1389 t['url'] = sanitize_url(t['url'])
9603e8a7 1390 if t.get('width') and t.get('height'):
d5519808 1391 t['resolution'] = '%dx%d' % (t['width'], t['height'])
f6c24009
PH
1392 if t.get('id') is None:
1393 t['id'] = '%d' % i
d5519808 1394
b7b72db9 1395 if self.params.get('list_thumbnails'):
1396 self.list_thumbnails(info_dict)
1397 return
1398
536a55da
S
1399 thumbnail = info_dict.get('thumbnail')
1400 if thumbnail:
1401 info_dict['thumbnail'] = sanitize_url(thumbnail)
1402 elif thumbnails:
d5519808
PH
1403 info_dict['thumbnail'] = thumbnails[-1]['url']
1404
c9ae7b95 1405 if 'display_id' not in info_dict and 'id' in info_dict:
0afef30b
PH
1406 info_dict['display_id'] = info_dict['id']
1407
955c4514 1408 if info_dict.get('upload_date') is None and info_dict.get('timestamp') is not None:
a55e36f4
S
1409 # Working around out-of-range timestamp values (e.g. negative ones on Windows,
1410 # see http://bugs.python.org/issue1646728)
1411 try:
1412 upload_date = datetime.datetime.utcfromtimestamp(info_dict['timestamp'])
1413 info_dict['upload_date'] = upload_date.strftime('%Y%m%d')
1414 except (ValueError, OverflowError, OSError):
1415 pass
9d2ecdbc 1416
33d2fc2f
S
1417 # Auto generate title fields corresponding to the *_number fields when missing
1418 # in order to always have clean titles. This is very common for TV series.
1419 for field in ('chapter', 'season', 'episode'):
1420 if info_dict.get('%s_number' % field) is not None and not info_dict.get(field):
1421 info_dict[field] = '%s %d' % (field.capitalize(), info_dict['%s_number' % field])
1422
4bba3716
S
1423 subtitles = info_dict.get('subtitles')
1424 if subtitles:
1425 for _, subtitle in subtitles.items():
1426 for subtitle_format in subtitle:
33f3040a
S
1427 if subtitle_format.get('url'):
1428 subtitle_format['url'] = sanitize_url(subtitle_format['url'])
5b1d8575 1429 if subtitle_format.get('ext') is None:
4bba3716
S
1430 subtitle_format['ext'] = determine_ext(subtitle_format['url']).lower()
1431
a504ced0 1432 if self.params.get('listsubtitles', False):
360e1ca5
JMF
1433 if 'automatic_captions' in info_dict:
1434 self.list_subtitles(info_dict['id'], info_dict.get('automatic_captions'), 'automatic captions')
4bba3716 1435 self.list_subtitles(info_dict['id'], subtitles, 'subtitles')
a504ced0 1436 return
360e1ca5 1437 info_dict['requested_subtitles'] = self.process_subtitles(
4bba3716 1438 info_dict['id'], subtitles,
360e1ca5 1439 info_dict.get('automatic_captions'))
a504ced0 1440
dd82ffea
JMF
1441 # We now pick which formats have to be downloaded
1442 if info_dict.get('formats') is None:
1443 # There's only one format available
1444 formats = [info_dict]
1445 else:
1446 formats = info_dict['formats']
1447
db95dc13
PH
1448 if not formats:
1449 raise ExtractorError('No video formats found!')
1450
181c7053
S
1451 formats_dict = {}
1452
dd82ffea 1453 # We check that all the formats have the format and format_id fields
db95dc13 1454 for i, format in enumerate(formats):
bec1fad2
PH
1455 if 'url' not in format:
1456 raise ExtractorError('Missing "url" key in result (index %d)' % i)
1457
c9969434
S
1458 sanitize_string_field(format, 'format_id')
1459 sanitize_numeric_fields(format)
dcf77cf1
S
1460 format['url'] = sanitize_url(format['url'])
1461
dd82ffea 1462 if format.get('format_id') is None:
8016c922 1463 format['format_id'] = compat_str(i)
e2effb08
S
1464 else:
1465 # Sanitize format_id from characters used in format selector expression
ec85ded8 1466 format['format_id'] = re.sub(r'[\s,/+\[\]()]', '_', format['format_id'])
181c7053
S
1467 format_id = format['format_id']
1468 if format_id not in formats_dict:
1469 formats_dict[format_id] = []
1470 formats_dict[format_id].append(format)
1471
1472 # Make sure all formats have unique format_id
1473 for format_id, ambiguous_formats in formats_dict.items():
1474 if len(ambiguous_formats) > 1:
1475 for i, format in enumerate(ambiguous_formats):
1476 format['format_id'] = '%s-%d' % (format_id, i)
1477
1478 for i, format in enumerate(formats):
8c51aa65 1479 if format.get('format') is None:
6febd1c1 1480 format['format'] = '{id} - {res}{note}'.format(
8c51aa65
JMF
1481 id=format['format_id'],
1482 res=self.format_resolution(format),
6febd1c1 1483 note=' ({0})'.format(format['format_note']) if format.get('format_note') is not None else '',
8c51aa65 1484 )
c1002e96 1485 # Automatically determine file extension if missing
5b1d8575 1486 if format.get('ext') is None:
cce929ea 1487 format['ext'] = determine_ext(format['url']).lower()
b5559424
S
1488 # Automatically determine protocol if missing (useful for format
1489 # selection purposes)
6f0be937 1490 if format.get('protocol') is None:
b5559424 1491 format['protocol'] = determine_protocol(format)
e5660ee6
JMF
1492 # Add HTTP headers, so that external programs can use them from the
1493 # json output
1494 full_format_info = info_dict.copy()
1495 full_format_info.update(format)
1496 format['http_headers'] = self._calc_headers(full_format_info)
0016b84e
S
1497 # Remove private housekeeping stuff
1498 if '__x_forwarded_for_ip' in info_dict:
1499 del info_dict['__x_forwarded_for_ip']
dd82ffea 1500
4bcc7bd1 1501 # TODO Central sorting goes here
99e206d5 1502
f89197d7 1503 if formats[0] is not info_dict:
b3d9ef88
JMF
1504 # only set the 'formats' fields if the original info_dict list them
1505 # otherwise we end up with a circular reference, the first (and unique)
f89197d7 1506 # element in the 'formats' field in info_dict is info_dict itself,
dfb1b146 1507 # which can't be exported to json
b3d9ef88 1508 info_dict['formats'] = formats
cfb56d1a 1509 if self.params.get('listformats'):
bfaae0a7 1510 self.list_formats(info_dict)
1511 return
1512
de3ef3ed 1513 req_format = self.params.get('format')
a9c58ad9 1514 if req_format is None:
feccf29c 1515 req_format_list = []
3749e36e 1516 if (self.params.get('outtmpl', DEFAULT_OUTTMPL) != '-' and
8250c32f 1517 not info_dict.get('is_live')):
7fcb605b 1518 merger = FFmpegMergerPP(self)
97fcf1bb 1519 if merger.available and merger.can_merge():
7fcb605b 1520 req_format_list.append('bestvideo+bestaudio')
feccf29c
S
1521 req_format_list.append('best')
1522 req_format = '/'.join(req_format_list)
5acfa126 1523 format_selector = self.build_format_selector(req_format)
317f7ab6
S
1524
1525 # While in format selection we may need to have an access to the original
1526 # format set in order to calculate some metrics or do some processing.
1527 # For now we need to be able to guess whether original formats provided
1528 # by extractor are incomplete or not (i.e. whether extractor provides only
1529 # video-only or audio-only formats) for proper formats selection for
1530 # extractors with such incomplete formats (see
1531 # https://github.com/rg3/youtube-dl/pull/5556).
1532 # Since formats may be filtered during format selection and may not match
1533 # the original formats the results may be incorrect. Thus original formats
1534 # or pre-calculated metrics should be passed to format selection routines
1535 # as well.
1536 # We will pass a context object containing all necessary additional data
1537 # instead of just formats.
1538 # This fixes incorrect format selection issue (see
1539 # https://github.com/rg3/youtube-dl/issues/10083).
2e221ca3 1540 incomplete_formats = (
317f7ab6 1541 # All formats are video-only or
2e221ca3 1542 all(f.get('vcodec') != 'none' and f.get('acodec') == 'none' for f in formats) or
317f7ab6 1543 # all formats are audio-only
2e221ca3 1544 all(f.get('vcodec') == 'none' and f.get('acodec') != 'none' for f in formats))
317f7ab6
S
1545
1546 ctx = {
1547 'formats': formats,
1548 'incomplete_formats': incomplete_formats,
1549 }
1550
1551 formats_to_download = list(format_selector(ctx))
dd82ffea 1552 if not formats_to_download:
6febd1c1 1553 raise ExtractorError('requested format not available',
78a3a9f8 1554 expected=True)
dd82ffea
JMF
1555
1556 if download:
1557 if len(formats_to_download) > 1:
6febd1c1 1558 self.to_screen('[info] %s: downloading video in %s formats' % (info_dict['id'], len(formats_to_download)))
dd82ffea
JMF
1559 for format in formats_to_download:
1560 new_info = dict(info_dict)
1561 new_info.update(format)
1562 self.process_info(new_info)
1563 # We update the info dict with the best quality format (backwards compatibility)
1564 info_dict.update(formats_to_download[-1])
1565 return info_dict
1566
98c70d6f 1567 def process_subtitles(self, video_id, normal_subtitles, automatic_captions):
a504ced0 1568 """Select the requested subtitles and their format"""
98c70d6f
JMF
1569 available_subs = {}
1570 if normal_subtitles and self.params.get('writesubtitles'):
1571 available_subs.update(normal_subtitles)
1572 if automatic_captions and self.params.get('writeautomaticsub'):
1573 for lang, cap_info in automatic_captions.items():
360e1ca5
JMF
1574 if lang not in available_subs:
1575 available_subs[lang] = cap_info
1576
4d171848
JMF
1577 if (not self.params.get('writesubtitles') and not
1578 self.params.get('writeautomaticsub') or not
1579 available_subs):
1580 return None
a504ced0
JMF
1581
1582 if self.params.get('allsubtitles', False):
1583 requested_langs = available_subs.keys()
1584 else:
1585 if self.params.get('subtitleslangs', False):
1586 requested_langs = self.params.get('subtitleslangs')
1587 elif 'en' in available_subs:
1588 requested_langs = ['en']
1589 else:
1590 requested_langs = [list(available_subs.keys())[0]]
1591
1592 formats_query = self.params.get('subtitlesformat', 'best')
1593 formats_preference = formats_query.split('/') if formats_query else []
1594 subs = {}
1595 for lang in requested_langs:
1596 formats = available_subs.get(lang)
1597 if formats is None:
1598 self.report_warning('%s subtitles not available for %s' % (lang, video_id))
1599 continue
a504ced0
JMF
1600 for ext in formats_preference:
1601 if ext == 'best':
1602 f = formats[-1]
1603 break
1604 matches = list(filter(lambda f: f['ext'] == ext, formats))
1605 if matches:
1606 f = matches[-1]
1607 break
1608 else:
1609 f = formats[-1]
1610 self.report_warning(
1611 'No subtitle format found matching "%s" for language %s, '
1612 'using %s' % (formats_query, lang, f['ext']))
1613 subs[lang] = f
1614 return subs
1615
8222d8de
JMF
1616 def process_info(self, info_dict):
1617 """Process a single resolved IE result."""
1618
1619 assert info_dict.get('_type', 'video') == 'video'
fd288278
PH
1620
1621 max_downloads = self.params.get('max_downloads')
1622 if max_downloads is not None:
1623 if self._num_downloads >= int(max_downloads):
1624 raise MaxDownloadsReached()
8222d8de
JMF
1625
1626 info_dict['fulltitle'] = info_dict['title']
1627 if len(info_dict['title']) > 200:
6febd1c1 1628 info_dict['title'] = info_dict['title'][:197] + '...'
8222d8de 1629
11b85ce6 1630 if 'format' not in info_dict:
8222d8de
JMF
1631 info_dict['format'] = info_dict['ext']
1632
442c37b7 1633 reason = self._match_entry(info_dict, incomplete=False)
8222d8de 1634 if reason is not None:
6febd1c1 1635 self.to_screen('[download] ' + reason)
8222d8de
JMF
1636 return
1637
fd288278 1638 self._num_downloads += 1
8222d8de 1639
e72c7e41 1640 info_dict['_filename'] = filename = self.prepare_filename(info_dict)
8222d8de
JMF
1641
1642 # Forced printings
1643 if self.params.get('forcetitle', False):
0783b09b 1644 self.to_stdout(info_dict['fulltitle'])
8222d8de 1645 if self.params.get('forceid', False):
0783b09b 1646 self.to_stdout(info_dict['id'])
8222d8de 1647 if self.params.get('forceurl', False):
16ae61f6 1648 if info_dict.get('requested_formats') is not None:
1649 for f in info_dict['requested_formats']:
1650 self.to_stdout(f['url'] + f.get('play_path', ''))
1651 else:
1652 # For RTMP URLs, also include the playpath
1653 self.to_stdout(info_dict['url'] + info_dict.get('play_path', ''))
216d71d0 1654 if self.params.get('forcethumbnail', False) and info_dict.get('thumbnail') is not None:
0783b09b 1655 self.to_stdout(info_dict['thumbnail'])
216d71d0 1656 if self.params.get('forcedescription', False) and info_dict.get('description') is not None:
0783b09b 1657 self.to_stdout(info_dict['description'])
8222d8de 1658 if self.params.get('forcefilename', False) and filename is not None:
0783b09b 1659 self.to_stdout(filename)
525ef922
PH
1660 if self.params.get('forceduration', False) and info_dict.get('duration') is not None:
1661 self.to_stdout(formatSeconds(info_dict['duration']))
8222d8de 1662 if self.params.get('forceformat', False):
0783b09b 1663 self.to_stdout(info_dict['format'])
9d153818 1664 if self.params.get('forcejson', False):
0783b09b 1665 self.to_stdout(json.dumps(info_dict))
8222d8de
JMF
1666
1667 # Do nothing else if in simulate mode
1668 if self.params.get('simulate', False):
1669 return
1670
1671 if filename is None:
1672 return
1673
1674 try:
e5a11a22 1675 dn = os.path.dirname(sanitize_path(encodeFilename(filename)))
d26e981d 1676 if dn and not os.path.exists(dn):
8222d8de
JMF
1677 os.makedirs(dn)
1678 except (OSError, IOError) as err:
9b9c5355 1679 self.report_error('unable to create directory ' + error_to_compat_str(err))
8222d8de
JMF
1680 return
1681
1682 if self.params.get('writedescription', False):
2699da80 1683 descfn = replace_extension(filename, 'description', info_dict.get('ext'))
7b6fefc9 1684 if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(descfn)):
6febd1c1 1685 self.to_screen('[info] Video description is already present')
f00fd51d
JMF
1686 elif info_dict.get('description') is None:
1687 self.report_warning('There\'s no description to write.')
7b6fefc9
PH
1688 else:
1689 try:
6febd1c1 1690 self.to_screen('[info] Writing video description to: ' + descfn)
7b6fefc9
PH
1691 with io.open(encodeFilename(descfn), 'w', encoding='utf-8') as descfile:
1692 descfile.write(info_dict['description'])
7b6fefc9 1693 except (OSError, IOError):
6febd1c1 1694 self.report_error('Cannot write description file ' + descfn)
7b6fefc9 1695 return
8222d8de 1696
1fb07d10 1697 if self.params.get('writeannotations', False):
98727e12 1698 annofn = replace_extension(filename, 'annotations.xml', info_dict.get('ext'))
7b6fefc9 1699 if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(annofn)):
6febd1c1 1700 self.to_screen('[info] Video annotations are already present')
7b6fefc9
PH
1701 else:
1702 try:
6febd1c1 1703 self.to_screen('[info] Writing video annotations to: ' + annofn)
7b6fefc9
PH
1704 with io.open(encodeFilename(annofn), 'w', encoding='utf-8') as annofile:
1705 annofile.write(info_dict['annotations'])
1706 except (KeyError, TypeError):
6febd1c1 1707 self.report_warning('There are no annotations to write.')
7b6fefc9 1708 except (OSError, IOError):
6febd1c1 1709 self.report_error('Cannot write annotations file: ' + annofn)
7b6fefc9 1710 return
1fb07d10 1711
c4a91be7 1712 subtitles_are_requested = any([self.params.get('writesubtitles', False),
0b7f3118 1713 self.params.get('writeautomaticsub')])
c4a91be7 1714
c84dd8a9 1715 if subtitles_are_requested and info_dict.get('requested_subtitles'):
8222d8de
JMF
1716 # subtitles download errors are already managed as troubles in relevant IE
1717 # that way it will silently go on when used with unsupporting IE
c84dd8a9 1718 subtitles = info_dict['requested_subtitles']
0f2c0d33 1719 ie = self.get_info_extractor(info_dict['extractor_key'])
a504ced0
JMF
1720 for sub_lang, sub_info in subtitles.items():
1721 sub_format = sub_info['ext']
1722 if sub_info.get('data') is not None:
1723 sub_data = sub_info['data']
1724 else:
1725 try:
0f2c0d33
JMF
1726 sub_data = ie._download_webpage(
1727 sub_info['url'], info_dict['id'], note=False)
1728 except ExtractorError as err:
a504ced0 1729 self.report_warning('Unable to download subtitle for "%s": %s' %
9b9c5355 1730 (sub_lang, error_to_compat_str(err.cause)))
a504ced0 1731 continue
8222d8de 1732 try:
d4051a8e 1733 sub_filename = subtitles_filename(filename, sub_lang, sub_format)
7b6fefc9 1734 if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(sub_filename)):
6febd1c1 1735 self.to_screen('[info] Video subtitle %s.%s is already_present' % (sub_lang, sub_format))
7b6fefc9 1736 else:
6febd1c1 1737 self.to_screen('[info] Writing video subtitles to: ' + sub_filename)
b1927f4e
YCH
1738 # Use newline='' to prevent conversion of newline characters
1739 # See https://github.com/rg3/youtube-dl/issues/10268
7f832413 1740 with io.open(encodeFilename(sub_filename), 'w', encoding='utf-8', newline='') as subfile:
a504ced0 1741 subfile.write(sub_data)
8222d8de 1742 except (OSError, IOError):
e4db1951 1743 self.report_error('Cannot write subtitles file ' + sub_filename)
8222d8de
JMF
1744 return
1745
8222d8de 1746 if self.params.get('writeinfojson', False):
b29e0000 1747 infofn = replace_extension(filename, 'info.json', info_dict.get('ext'))
7b6fefc9 1748 if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(infofn)):
6febd1c1 1749 self.to_screen('[info] Video description metadata is already present')
7b6fefc9 1750 else:
6febd1c1 1751 self.to_screen('[info] Writing video description metadata as JSON to: ' + infofn)
7b6fefc9 1752 try:
cb202fd2 1753 write_json_file(self.filter_requested_info(info_dict), infofn)
7b6fefc9 1754 except (OSError, IOError):
6febd1c1 1755 self.report_error('Cannot write metadata to JSON file ' + infofn)
7b6fefc9 1756 return
8222d8de 1757
ec82d85a 1758 self._write_thumbnails(info_dict, filename)
8222d8de
JMF
1759
1760 if not self.params.get('skip_download', False):
4340deca
P
1761 try:
1762 def dl(name, info):
a055469f 1763 fd = get_suitable_downloader(info, self.params)(self, self.params)
4340deca
P
1764 for ph in self._progress_hooks:
1765 fd.add_progress_hook(ph)
1766 if self.params.get('verbose'):
1767 self.to_stdout('[debug] Invoking downloader on %r' % info.get('url'))
1768 return fd.download(name, info)
ee69b99a 1769
4340deca
P
1770 if info_dict.get('requested_formats') is not None:
1771 downloaded = []
1772 success = True
d47aeb22 1773 merger = FFmpegMergerPP(self)
f740fae2 1774 if not merger.available:
4340deca
P
1775 postprocessors = []
1776 self.report_warning('You have requested multiple '
1777 'formats but ffmpeg or avconv are not installed.'
4a5a898a 1778 ' The formats won\'t be merged.')
6350728b 1779 else:
4340deca 1780 postprocessors = [merger]
81cd954a
S
1781
1782 def compatible_formats(formats):
1783 video, audio = formats
1784 # Check extension
1785 video_ext, audio_ext = audio.get('ext'), video.get('ext')
1786 if video_ext and audio_ext:
1787 COMPATIBLE_EXTS = (
b2758123 1788 ('mp3', 'mp4', 'm4a', 'm4p', 'm4b', 'm4r', 'm4v', 'ismv', 'isma'),
81cd954a
S
1789 ('webm')
1790 )
1791 for exts in COMPATIBLE_EXTS:
1792 if video_ext in exts and audio_ext in exts:
1793 return True
1794 # TODO: Check acodec/vcodec
1795 return False
1796
38c6902b
S
1797 filename_real_ext = os.path.splitext(filename)[1][1:]
1798 filename_wo_ext = (
1799 os.path.splitext(filename)[0]
1800 if filename_real_ext == info_dict['ext']
1801 else filename)
81cd954a 1802 requested_formats = info_dict['requested_formats']
c0dea0a7 1803 if self.params.get('merge_output_format') is None and not compatible_formats(requested_formats):
38c6902b 1804 info_dict['ext'] = 'mkv'
4a5a898a
S
1805 self.report_warning(
1806 'Requested formats are incompatible for merge and will be merged into mkv.')
38c6902b
S
1807 # Ensure filename always has a correct extension for successful merge
1808 filename = '%s.%s' % (filename_wo_ext, info_dict['ext'])
5b5fbc08
JMF
1809 if os.path.exists(encodeFilename(filename)):
1810 self.to_screen(
1811 '[download] %s has already been downloaded and '
1812 'merged' % filename)
1813 else:
81cd954a 1814 for f in requested_formats:
5b5fbc08
JMF
1815 new_info = dict(info_dict)
1816 new_info.update(f)
1817 fname = self.prepare_filename(new_info)
666a9a2b 1818 fname = prepend_extension(fname, 'f%s' % f['format_id'], new_info['ext'])
5b5fbc08
JMF
1819 downloaded.append(fname)
1820 partial_success = dl(fname, new_info)
1821 success = success and partial_success
1822 info_dict['__postprocessors'] = postprocessors
1823 info_dict['__files_to_merge'] = downloaded
4340deca
P
1824 else:
1825 # Just a single file
1826 success = dl(filename, info_dict)
1827 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
7960b056 1828 self.report_error('unable to download video data: %s' % error_to_compat_str(err))
4340deca
P
1829 return
1830 except (OSError, IOError) as err:
1831 raise UnavailableVideoError(err)
1832 except (ContentTooShortError, ) as err:
1833 self.report_error('content too short (expected %s bytes and served %s)' % (err.expected, err.downloaded))
1834 return
8222d8de 1835
e38cafe9 1836 if success and filename != '-':
6271f1ca 1837 # Fixup content
62cd676c
PH
1838 fixup_policy = self.params.get('fixup')
1839 if fixup_policy is None:
1840 fixup_policy = 'detect_or_warn'
1841
d1e4a464
S
1842 INSTALL_FFMPEG_MESSAGE = 'Install ffmpeg or avconv to fix this automatically.'
1843
6271f1ca
PH
1844 stretched_ratio = info_dict.get('stretched_ratio')
1845 if stretched_ratio is not None and stretched_ratio != 1:
6271f1ca
PH
1846 if fixup_policy == 'warn':
1847 self.report_warning('%s: Non-uniform pixel ratio (%s)' % (
1848 info_dict['id'], stretched_ratio))
1849 elif fixup_policy == 'detect_or_warn':
1850 stretched_pp = FFmpegFixupStretchedPP(self)
1851 if stretched_pp.available:
1852 info_dict.setdefault('__postprocessors', [])
1853 info_dict['__postprocessors'].append(stretched_pp)
1854 else:
1855 self.report_warning(
d1e4a464
S
1856 '%s: Non-uniform pixel ratio (%s). %s'
1857 % (info_dict['id'], stretched_ratio, INSTALL_FFMPEG_MESSAGE))
6271f1ca 1858 else:
62cd676c
PH
1859 assert fixup_policy in ('ignore', 'never')
1860
d1e4a464
S
1861 if (info_dict.get('requested_formats') is None and
1862 info_dict.get('container') == 'm4a_dash'):
62cd676c 1863 if fixup_policy == 'warn':
d1e4a464
S
1864 self.report_warning(
1865 '%s: writing DASH m4a. '
1866 'Only some players support this container.'
1867 % info_dict['id'])
62cd676c
PH
1868 elif fixup_policy == 'detect_or_warn':
1869 fixup_pp = FFmpegFixupM4aPP(self)
1870 if fixup_pp.available:
1871 info_dict.setdefault('__postprocessors', [])
1872 info_dict['__postprocessors'].append(fixup_pp)
1873 else:
1874 self.report_warning(
d1e4a464
S
1875 '%s: writing DASH m4a. '
1876 'Only some players support this container. %s'
1877 % (info_dict['id'], INSTALL_FFMPEG_MESSAGE))
62cd676c
PH
1878 else:
1879 assert fixup_policy in ('ignore', 'never')
6271f1ca 1880
d1e4a464
S
1881 if (info_dict.get('protocol') == 'm3u8_native' or
1882 info_dict.get('protocol') == 'm3u8' and
1883 self.params.get('hls_prefer_native')):
f17f8651 1884 if fixup_policy == 'warn':
1885 self.report_warning('%s: malformated aac bitstream.' % (
1886 info_dict['id']))
1887 elif fixup_policy == 'detect_or_warn':
1888 fixup_pp = FFmpegFixupM3u8PP(self)
1889 if fixup_pp.available:
1890 info_dict.setdefault('__postprocessors', [])
1891 info_dict['__postprocessors'].append(fixup_pp)
1892 else:
1893 self.report_warning(
d1e4a464
S
1894 '%s: malformated aac bitstream. %s'
1895 % (info_dict['id'], INSTALL_FFMPEG_MESSAGE))
f17f8651 1896 else:
1897 assert fixup_policy in ('ignore', 'never')
1898
8222d8de
JMF
1899 try:
1900 self.post_process(filename, info_dict)
1901 except (PostProcessingError) as err:
6febd1c1 1902 self.report_error('postprocessing: %s' % str(err))
8222d8de 1903 return
cd58dc3e 1904 self.record_download_archive(info_dict)
8222d8de
JMF
1905
1906 def download(self, url_list):
1907 """Download a given list of URLs."""
acd69589 1908 outtmpl = self.params.get('outtmpl', DEFAULT_OUTTMPL)
0c75c3fa 1909 if (len(url_list) > 1 and
9487ce03 1910 outtmpl != '-' and
8fb3ac36
PH
1911 '%' not in outtmpl and
1912 self.params.get('max_downloads') != 1):
acd69589 1913 raise SameFileError(outtmpl)
8222d8de
JMF
1914
1915 for url in url_list:
1916 try:
5f6a1245 1917 # It also downloads the videos
61aa5ba3
S
1918 res = self.extract_info(
1919 url, force_generic_extractor=self.params.get('force_generic_extractor', False))
8222d8de 1920 except UnavailableVideoError:
6febd1c1 1921 self.report_error('unable to download video')
8222d8de 1922 except MaxDownloadsReached:
6febd1c1 1923 self.to_screen('[info] Maximum number of downloaded files reached.')
8222d8de 1924 raise
63e0be34
PH
1925 else:
1926 if self.params.get('dump_single_json', False):
1927 self.to_stdout(json.dumps(res))
8222d8de
JMF
1928
1929 return self._download_retcode
1930
1dcc4c0c 1931 def download_with_info_file(self, info_filename):
31bd3925
JMF
1932 with contextlib.closing(fileinput.FileInput(
1933 [info_filename], mode='r',
1934 openhook=fileinput.hook_encoded('utf-8'))) as f:
1935 # FileInput doesn't have a read method, we can't call json.load
cb202fd2 1936 info = self.filter_requested_info(json.loads('\n'.join(f)))
d4943898
JMF
1937 try:
1938 self.process_ie_result(info, download=True)
1939 except DownloadError:
1940 webpage_url = info.get('webpage_url')
1941 if webpage_url is not None:
6febd1c1 1942 self.report_warning('The info failed to download, trying with "%s"' % webpage_url)
d4943898
JMF
1943 return self.download([webpage_url])
1944 else:
1945 raise
1946 return self._download_retcode
1dcc4c0c 1947
cb202fd2
S
1948 @staticmethod
1949 def filter_requested_info(info_dict):
1950 return dict(
1951 (k, v) for k, v in info_dict.items()
1952 if k not in ['requested_formats', 'requested_subtitles'])
1953
8222d8de
JMF
1954 def post_process(self, filename, ie_info):
1955 """Run all the postprocessors on the given file."""
1956 info = dict(ie_info)
1957 info['filepath'] = filename
6350728b
JMF
1958 pps_chain = []
1959 if ie_info.get('__postprocessors') is not None:
1960 pps_chain.extend(ie_info['__postprocessors'])
1961 pps_chain.extend(self._pps)
1962 for pp in pps_chain:
71646e46 1963 files_to_delete = []
8222d8de 1964 try:
592e97e8 1965 files_to_delete, info = pp.run(info)
8222d8de 1966 except PostProcessingError as e:
bbcbf4d4 1967 self.report_error(e.msg)
592e97e8
JMF
1968 if files_to_delete and not self.params.get('keepvideo', False):
1969 for old_filename in files_to_delete:
f3ff1a36 1970 self.to_screen('Deleting original file %s (pass -k to keep)' % old_filename)
592e97e8
JMF
1971 try:
1972 os.remove(encodeFilename(old_filename))
1973 except (IOError, OSError):
1974 self.report_warning('Unable to remove downloaded original file')
c1c9a79c 1975
5db07df6
PH
1976 def _make_archive_id(self, info_dict):
1977 # Future-proof against any change in case
1978 # and backwards compatibility with prior versions
d31209a1 1979 extractor = info_dict.get('extractor_key')
7012b23c
PH
1980 if extractor is None:
1981 if 'id' in info_dict:
1982 extractor = info_dict.get('ie_key') # key in a playlist
1983 if extractor is None:
5db07df6 1984 return None # Incomplete video information
6febd1c1 1985 return extractor.lower() + ' ' + info_dict['id']
5db07df6
PH
1986
1987 def in_download_archive(self, info_dict):
1988 fn = self.params.get('download_archive')
1989 if fn is None:
1990 return False
1991
1992 vid_id = self._make_archive_id(info_dict)
1993 if vid_id is None:
7012b23c 1994 return False # Incomplete video information
5db07df6 1995
c1c9a79c
PH
1996 try:
1997 with locked_file(fn, 'r', encoding='utf-8') as archive_file:
1998 for line in archive_file:
1999 if line.strip() == vid_id:
2000 return True
2001 except IOError as ioe:
2002 if ioe.errno != errno.ENOENT:
2003 raise
2004 return False
2005
2006 def record_download_archive(self, info_dict):
2007 fn = self.params.get('download_archive')
2008 if fn is None:
2009 return
5db07df6
PH
2010 vid_id = self._make_archive_id(info_dict)
2011 assert vid_id
c1c9a79c 2012 with locked_file(fn, 'a', encoding='utf-8') as archive_file:
6febd1c1 2013 archive_file.write(vid_id + '\n')
dd82ffea 2014
8c51aa65 2015 @staticmethod
8abeeb94 2016 def format_resolution(format, default='unknown'):
fb04e403
PH
2017 if format.get('vcodec') == 'none':
2018 return 'audio only'
f49d89ee
PH
2019 if format.get('resolution') is not None:
2020 return format['resolution']
8c51aa65
JMF
2021 if format.get('height') is not None:
2022 if format.get('width') is not None:
6febd1c1 2023 res = '%sx%s' % (format['width'], format['height'])
8c51aa65 2024 else:
6febd1c1 2025 res = '%sp' % format['height']
f49d89ee 2026 elif format.get('width') is not None:
388ae76b 2027 res = '%dx?' % format['width']
8c51aa65 2028 else:
8abeeb94 2029 res = default
8c51aa65
JMF
2030 return res
2031
c57f7757
PH
2032 def _format_note(self, fdict):
2033 res = ''
2034 if fdict.get('ext') in ['f4f', 'f4m']:
2035 res += '(unsupported) '
32f90364
PH
2036 if fdict.get('language'):
2037 if res:
2038 res += ' '
9016d76f 2039 res += '[%s] ' % fdict['language']
c57f7757
PH
2040 if fdict.get('format_note') is not None:
2041 res += fdict['format_note'] + ' '
2042 if fdict.get('tbr') is not None:
2043 res += '%4dk ' % fdict['tbr']
2044 if fdict.get('container') is not None:
2045 if res:
2046 res += ', '
2047 res += '%s container' % fdict['container']
2048 if (fdict.get('vcodec') is not None and
2049 fdict.get('vcodec') != 'none'):
2050 if res:
2051 res += ', '
2052 res += fdict['vcodec']
91c7271a 2053 if fdict.get('vbr') is not None:
c57f7757
PH
2054 res += '@'
2055 elif fdict.get('vbr') is not None and fdict.get('abr') is not None:
2056 res += 'video@'
2057 if fdict.get('vbr') is not None:
2058 res += '%4dk' % fdict['vbr']
fbb21cf5 2059 if fdict.get('fps') is not None:
5d583bdf
S
2060 if res:
2061 res += ', '
2062 res += '%sfps' % fdict['fps']
c57f7757
PH
2063 if fdict.get('acodec') is not None:
2064 if res:
2065 res += ', '
2066 if fdict['acodec'] == 'none':
2067 res += 'video only'
2068 else:
2069 res += '%-5s' % fdict['acodec']
2070 elif fdict.get('abr') is not None:
2071 if res:
2072 res += ', '
2073 res += 'audio'
2074 if fdict.get('abr') is not None:
2075 res += '@%3dk' % fdict['abr']
2076 if fdict.get('asr') is not None:
2077 res += ' (%5dHz)' % fdict['asr']
2078 if fdict.get('filesize') is not None:
2079 if res:
2080 res += ', '
2081 res += format_bytes(fdict['filesize'])
9732d77e
PH
2082 elif fdict.get('filesize_approx') is not None:
2083 if res:
2084 res += ', '
2085 res += '~' + format_bytes(fdict['filesize_approx'])
c57f7757 2086 return res
91c7271a 2087
c57f7757 2088 def list_formats(self, info_dict):
94badb25 2089 formats = info_dict.get('formats', [info_dict])
b81a359e
PH
2090 table = [
2091 [f['format_id'], f['ext'], self.format_resolution(f), self._format_note(f)]
2092 for f in formats
e65566a9 2093 if f.get('preference') is None or f['preference'] >= -1000]
94badb25 2094 if len(formats) > 1:
b81a359e 2095 table[-1][-1] += (' ' if table[-1][-1] else '') + '(best)'
57dd9a8f 2096
b81a359e 2097 header_line = ['format code', 'extension', 'resolution', 'note']
cfb56d1a 2098 self.to_screen(
b81a359e
PH
2099 '[info] Available formats for %s:\n%s' %
2100 (info_dict['id'], render_table(header_line, table)))
cfb56d1a
PH
2101
2102 def list_thumbnails(self, info_dict):
2103 thumbnails = info_dict.get('thumbnails')
2104 if not thumbnails:
b7b72db9 2105 self.to_screen('[info] No thumbnails present for %s' % info_dict['id'])
2106 return
cfb56d1a
PH
2107
2108 self.to_screen(
2109 '[info] Thumbnails for %s:' % info_dict['id'])
2110 self.to_screen(render_table(
2111 ['ID', 'width', 'height', 'URL'],
2112 [[t['id'], t.get('width', 'unknown'), t.get('height', 'unknown'), t['url']] for t in thumbnails]))
dca08720 2113
360e1ca5 2114 def list_subtitles(self, video_id, subtitles, name='subtitles'):
a504ced0 2115 if not subtitles:
360e1ca5 2116 self.to_screen('%s has no %s' % (video_id, name))
a504ced0 2117 return
a504ced0 2118 self.to_screen(
edab9dbf
JMF
2119 'Available %s for %s:' % (name, video_id))
2120 self.to_screen(render_table(
2121 ['Language', 'formats'],
2122 [[lang, ', '.join(f['ext'] for f in reversed(formats))]
2123 for lang, formats in subtitles.items()]))
a504ced0 2124
dca08720
PH
2125 def urlopen(self, req):
2126 """ Start an HTTP download """
82d8a8b6 2127 if isinstance(req, compat_basestring):
67dda517 2128 req = sanitized_Request(req)
19a41fc6 2129 return self._opener.open(req, timeout=self._socket_timeout)
dca08720
PH
2130
2131 def print_debug_header(self):
2132 if not self.params.get('verbose'):
2133 return
62fec3b2 2134
4192b51c
PH
2135 if type('') is not compat_str:
2136 # Python 2.6 on SLES11 SP1 (https://github.com/rg3/youtube-dl/issues/3326)
2137 self.report_warning(
2138 'Your Python is broken! Update to a newer and supported version')
2139
c6afed48
PH
2140 stdout_encoding = getattr(
2141 sys.stdout, 'encoding', 'missing (%s)' % type(sys.stdout).__name__)
b0472057 2142 encoding_str = (
734f90bb
PH
2143 '[debug] Encodings: locale %s, fs %s, out %s, pref %s\n' % (
2144 locale.getpreferredencoding(),
2145 sys.getfilesystemencoding(),
c6afed48 2146 stdout_encoding,
b0472057 2147 self.get_encoding()))
4192b51c 2148 write_string(encoding_str, encoding=None)
734f90bb
PH
2149
2150 self._write_string('[debug] youtube-dl version ' + __version__ + '\n')
e0986e31
JMF
2151 if _LAZY_LOADER:
2152 self._write_string('[debug] Lazy loading extractors enabled' + '\n')
dca08720
PH
2153 try:
2154 sp = subprocess.Popen(
2155 ['git', 'rev-parse', '--short', 'HEAD'],
2156 stdout=subprocess.PIPE, stderr=subprocess.PIPE,
2157 cwd=os.path.dirname(os.path.abspath(__file__)))
2158 out, err = sp.communicate()
2159 out = out.decode().strip()
2160 if re.match('[0-9a-f]+', out):
734f90bb 2161 self._write_string('[debug] Git HEAD: ' + out + '\n')
70a1165b 2162 except Exception:
dca08720
PH
2163 try:
2164 sys.exc_clear()
70a1165b 2165 except Exception:
dca08720 2166 pass
d28b5171
PH
2167 self._write_string('[debug] Python version %s - %s\n' % (
2168 platform.python_version(), platform_name()))
2169
73fac4e9 2170 exe_versions = FFmpegPostProcessor.get_versions(self)
4c83c967 2171 exe_versions['rtmpdump'] = rtmpdump_version()
d28b5171
PH
2172 exe_str = ', '.join(
2173 '%s %s' % (exe, v)
2174 for exe, v in sorted(exe_versions.items())
2175 if v
2176 )
2177 if not exe_str:
2178 exe_str = 'none'
2179 self._write_string('[debug] exe versions: %s\n' % exe_str)
dca08720
PH
2180
2181 proxy_map = {}
2182 for handler in self._opener.handlers:
2183 if hasattr(handler, 'proxies'):
2184 proxy_map.update(handler.proxies)
734f90bb 2185 self._write_string('[debug] Proxy map: ' + compat_str(proxy_map) + '\n')
dca08720 2186
58b1f00d
PH
2187 if self.params.get('call_home', False):
2188 ipaddr = self.urlopen('https://yt-dl.org/ip').read().decode('utf-8')
2189 self._write_string('[debug] Public IP address: %s\n' % ipaddr)
2190 latest_version = self.urlopen(
2191 'https://yt-dl.org/latest/version').read().decode('utf-8')
2192 if version_tuple(latest_version) > version_tuple(__version__):
2193 self.report_warning(
2194 'You are using an outdated version (newest version: %s)! '
2195 'See https://yt-dl.org/update if you need help updating.' %
2196 latest_version)
2197
e344693b 2198 def _setup_opener(self):
6ad14cab 2199 timeout_val = self.params.get('socket_timeout')
19a41fc6 2200 self._socket_timeout = 600 if timeout_val is None else float(timeout_val)
6ad14cab 2201
dca08720
PH
2202 opts_cookiefile = self.params.get('cookiefile')
2203 opts_proxy = self.params.get('proxy')
2204
2205 if opts_cookiefile is None:
2206 self.cookiejar = compat_cookiejar.CookieJar()
2207 else:
590bc6f6 2208 opts_cookiefile = expand_path(opts_cookiefile)
dca08720
PH
2209 self.cookiejar = compat_cookiejar.MozillaCookieJar(
2210 opts_cookiefile)
2211 if os.access(opts_cookiefile, os.R_OK):
2212 self.cookiejar.load()
2213
6a3f4c3f 2214 cookie_processor = YoutubeDLCookieProcessor(self.cookiejar)
dca08720
PH
2215 if opts_proxy is not None:
2216 if opts_proxy == '':
2217 proxies = {}
2218 else:
2219 proxies = {'http': opts_proxy, 'https': opts_proxy}
2220 else:
2221 proxies = compat_urllib_request.getproxies()
2222 # Set HTTPS proxy to HTTP one if given (https://github.com/rg3/youtube-dl/issues/805)
2223 if 'http' in proxies and 'https' not in proxies:
2224 proxies['https'] = proxies['http']
91410c9b 2225 proxy_handler = PerRequestProxyHandler(proxies)
a0ddb8a2
PH
2226
2227 debuglevel = 1 if self.params.get('debug_printtraffic') else 0
be4a824d
PH
2228 https_handler = make_HTTPS_handler(self.params, debuglevel=debuglevel)
2229 ydlh = YoutubeDLHandler(self.params, debuglevel=debuglevel)
8b172c2e 2230 data_handler = compat_urllib_request_DataHandler()
6240b0a2
JMF
2231
2232 # When passing our own FileHandler instance, build_opener won't add the
2233 # default FileHandler and allows us to disable the file protocol, which
2234 # can be used for malicious purposes (see
e37afbe0 2235 # https://github.com/rg3/youtube-dl/issues/8227)
6240b0a2
JMF
2236 file_handler = compat_urllib_request.FileHandler()
2237
2238 def file_open(*args, **kwargs):
30e2f2d7 2239 raise compat_urllib_error.URLError('file:// scheme is explicitly disabled in youtube-dl for security reasons')
6240b0a2
JMF
2240 file_handler.file_open = file_open
2241
2242 opener = compat_urllib_request.build_opener(
2243 proxy_handler, https_handler, cookie_processor, ydlh, data_handler, file_handler)
2461f79d 2244
dca08720
PH
2245 # Delete the default user-agent header, which would otherwise apply in
2246 # cases where our custom HTTP handler doesn't come into play
2247 # (See https://github.com/rg3/youtube-dl/issues/1309 for details)
2248 opener.addheaders = []
2249 self._opener = opener
62fec3b2
PH
2250
2251 def encode(self, s):
2252 if isinstance(s, bytes):
2253 return s # Already encoded
2254
2255 try:
2256 return s.encode(self.get_encoding())
2257 except UnicodeEncodeError as err:
2258 err.reason = err.reason + '. Check your system encoding configuration or use the --encoding option.'
2259 raise
2260
2261 def get_encoding(self):
2262 encoding = self.params.get('encoding')
2263 if encoding is None:
2264 encoding = preferredencoding()
2265 return encoding
ec82d85a
PH
2266
2267 def _write_thumbnails(self, info_dict, filename):
2268 if self.params.get('writethumbnail', False):
2269 thumbnails = info_dict.get('thumbnails')
2270 if thumbnails:
2271 thumbnails = [thumbnails[-1]]
2272 elif self.params.get('write_all_thumbnails', False):
2273 thumbnails = info_dict.get('thumbnails')
2274 else:
2275 return
2276
2277 if not thumbnails:
2278 # No thumbnails present, so return immediately
2279 return
2280
2281 for t in thumbnails:
2282 thumb_ext = determine_ext(t['url'], 'jpg')
2283 suffix = '_%s' % t['id'] if len(thumbnails) > 1 else ''
2284 thumb_display_id = '%s ' % t['id'] if len(thumbnails) > 1 else ''
82245a6d 2285 t['filename'] = thumb_filename = os.path.splitext(filename)[0] + suffix + '.' + thumb_ext
ec82d85a
PH
2286
2287 if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(thumb_filename)):
2288 self.to_screen('[%s] %s: Thumbnail %sis already present' %
2289 (info_dict['extractor'], info_dict['id'], thumb_display_id))
2290 else:
2291 self.to_screen('[%s] %s: Downloading thumbnail %s...' %
2292 (info_dict['extractor'], info_dict['id'], thumb_display_id))
2293 try:
2294 uf = self.urlopen(t['url'])
d3d89c32 2295 with open(encodeFilename(thumb_filename), 'wb') as thumbf:
ec82d85a
PH
2296 shutil.copyfileobj(uf, thumbf)
2297 self.to_screen('[%s] %s: Writing thumbnail %sto: %s' %
2298 (info_dict['extractor'], info_dict['id'], thumb_display_id, thumb_filename))
2299 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
2300 self.report_warning('Unable to download thumbnail "%s": %s' %
9b9c5355 2301 (t['url'], error_to_compat_str(err)))