]> jfr.im git - yt-dlp.git/blame - youtube_dl/YoutubeDL.py
Merge pull request #8246 from dstftw/initial-json-ld-metadata-support
[yt-dlp.git] / youtube_dl / YoutubeDL.py
CommitLineData
8222d8de
JMF
1#!/usr/bin/env python
2# -*- coding: utf-8 -*-
3
6febd1c1 4from __future__ import absolute_import, unicode_literals
8222d8de 5
26e63931 6import collections
31bd3925 7import contextlib
9d2ecdbc 8import datetime
c1c9a79c 9import errno
31bd3925 10import fileinput
8222d8de 11import io
b82f815f 12import itertools
8694c600 13import json
62fec3b2 14import locale
083c9df9 15import operator
8222d8de 16import os
dca08720 17import platform
8222d8de
JMF
18import re
19import shutil
dca08720 20import subprocess
8222d8de
JMF
21import socket
22import sys
23import time
67134eab 24import tokenize
8222d8de
JMF
25import traceback
26
1e5b9a95
PH
27if os.name == 'nt':
28 import ctypes
29
8c25f81b 30from .compat import (
82d8a8b6 31 compat_basestring,
dca08720 32 compat_cookiejar,
4644ac55 33 compat_expanduser,
003c69a8 34 compat_get_terminal_size,
ce02ed60 35 compat_http_client,
4f026faf 36 compat_kwargs,
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 (
ce02ed60
PH
44 ContentTooShortError,
45 date_from_str,
46 DateRange,
acd69589 47 DEFAULT_OUTTMPL,
ce02ed60
PH
48 determine_ext,
49 DownloadError,
c0384f22 50 encode_compat_str,
ce02ed60 51 encodeFilename,
9b9c5355 52 error_to_compat_str,
ce02ed60 53 ExtractorError,
02dbf93f 54 format_bytes,
525ef922 55 formatSeconds,
ce02ed60 56 locked_file,
dca08720 57 make_HTTPS_handler,
ce02ed60 58 MaxDownloadsReached,
b7ab0590 59 PagedList,
083c9df9 60 parse_filesize,
91410c9b 61 PerRequestProxyHandler,
ce02ed60 62 PostProcessingError,
dca08720 63 platform_name,
ce02ed60 64 preferredencoding,
cfb56d1a 65 render_table,
ce02ed60
PH
66 SameFileError,
67 sanitize_filename,
1bb5c511 68 sanitize_path,
67dda517 69 sanitized_Request,
e5660ee6 70 std_headers,
ce02ed60 71 subtitles_filename,
ce02ed60 72 UnavailableVideoError,
29eb5174 73 url_basename,
58b1f00d 74 version_tuple,
ce02ed60
PH
75 write_json_file,
76 write_string,
6a3f4c3f 77 YoutubeDLCookieProcessor,
dca08720 78 YoutubeDLHandler,
6350728b 79 prepend_extension,
b29e0000 80 replace_extension,
7d4111ed 81 args_to_str,
05900629 82 age_restricted,
ce02ed60 83)
a0e07d31 84from .cache import Cache
023fa8c4 85from .extractor import get_info_extractor, gen_extractors
3bc2ddcc 86from .downloader import get_suitable_downloader
4c83c967 87from .downloader.rtmp import rtmpdump_version
4f026faf 88from .postprocessor import (
62cd676c 89 FFmpegFixupM4aPP,
6271f1ca 90 FFmpegFixupStretchedPP,
4f026faf
PH
91 FFmpegMergerPP,
92 FFmpegPostProcessor,
93 get_postprocessor,
94)
dca08720 95from .version import __version__
8222d8de
JMF
96
97
98class YoutubeDL(object):
99 """YoutubeDL class.
100
101 YoutubeDL objects are the ones responsible of downloading the
102 actual video file and writing it to disk if the user has requested
103 it, among some other tasks. In most cases there should be one per
104 program. As, given a video URL, the downloader doesn't know how to
105 extract all the needed information, task that InfoExtractors do, it
106 has to pass the URL to one of them.
107
108 For this, YoutubeDL objects have a method that allows
109 InfoExtractors to be registered in a given order. When it is passed
110 a URL, the YoutubeDL object handles it to the first InfoExtractor it
111 finds that reports being able to handle it. The InfoExtractor extracts
112 all the information about the video or videos the URL refers to, and
113 YoutubeDL process the extracted information, possibly using a File
114 Downloader to download the video.
115
116 YoutubeDL objects accept a lot of parameters. In order not to saturate
117 the object constructor with arguments, it receives a dictionary of
118 options instead. These options are available through the params
119 attribute for the InfoExtractors to use. The YoutubeDL also
120 registers itself as the downloader in charge for the InfoExtractors
121 that are added to it, so this is a "mutual registration".
122
123 Available options:
124
125 username: Username for authentication purposes.
126 password: Password for authentication purposes.
180940e0 127 videopassword: Password for accessing a video.
8222d8de
JMF
128 usenetrc: Use netrc for authentication instead.
129 verbose: Print additional info to stdout.
130 quiet: Do not print messages to stdout.
ad8915b7 131 no_warnings: Do not print out anything for warnings.
8222d8de
JMF
132 forceurl: Force printing final URL.
133 forcetitle: Force printing title.
134 forceid: Force printing ID.
135 forcethumbnail: Force printing thumbnail URL.
136 forcedescription: Force printing description.
137 forcefilename: Force printing final filename.
525ef922 138 forceduration: Force printing duration.
8694c600 139 forcejson: Force printing info_dict as JSON.
63e0be34
PH
140 dump_single_json: Force printing the info_dict of the whole playlist
141 (or video) as a single JSON line.
8222d8de 142 simulate: Do not download the video files.
d8600787 143 format: Video format code. See options.py for more information.
8222d8de
JMF
144 outtmpl: Template for output names.
145 restrictfilenames: Do not allow "&" and spaces in file names
146 ignoreerrors: Do not stop on download errors.
d22dec74 147 force_generic_extractor: Force downloader to use the generic extractor
8222d8de
JMF
148 nooverwrites: Prevent overwriting files.
149 playliststart: Playlist item to start at.
150 playlistend: Playlist item to end at.
c14e88f0 151 playlist_items: Specific indices of playlist to download.
ff815fe6 152 playlistreverse: Download playlist items in reverse order.
8222d8de
JMF
153 matchtitle: Download only matching titles.
154 rejecttitle: Reject downloads for matching titles.
8bf9319e 155 logger: Log messages to a logging.Logger instance.
8222d8de
JMF
156 logtostderr: Log messages to stderr instead of stdout.
157 writedescription: Write the video description to a .description file
158 writeinfojson: Write the video description to a .info.json file
1fb07d10 159 writeannotations: Write the video annotations to a .annotations.xml file
8222d8de 160 writethumbnail: Write the thumbnail image to a file
ec82d85a 161 write_all_thumbnails: Write all thumbnail formats to files
8222d8de 162 writesubtitles: Write the video subtitles to a file
741dd8ea 163 writeautomaticsub: Write the automatically generated subtitles to a file
8222d8de 164 allsubtitles: Downloads all the subtitles of the video
0b7f3118 165 (requires writesubtitles or writeautomaticsub)
8222d8de 166 listsubtitles: Lists all available subtitles for the video
a504ced0 167 subtitlesformat: The format code for subtitles
aa6a10c4 168 subtitleslangs: List of languages of the subtitles to download
8222d8de
JMF
169 keepvideo: Keep the video file after post-processing
170 daterange: A DateRange object, download only if the upload_date is in the range.
171 skip_download: Skip the actual download of the video file
c35f9e72 172 cachedir: Location of the cache files in the filesystem.
a0e07d31 173 False to disable filesystem cache.
47192f92 174 noplaylist: Download single video instead of a playlist if in doubt.
8dbe9899
PH
175 age_limit: An integer representing the user's age in years.
176 Unsuitable videos for the given age are skipped.
5fe18bdb
PH
177 min_views: An integer representing the minimum view count the video
178 must have in order to not be skipped.
179 Videos without view count information are always
180 downloaded. None for no limit.
181 max_views: An integer representing the maximum view count.
182 Videos that are more popular than that are not
183 downloaded.
184 Videos without view count information are always
185 downloaded. None for no limit.
186 download_archive: File name of a file where all downloads are recorded.
c1c9a79c
PH
187 Videos already present in the file are not downloaded
188 again.
dca08720 189 cookiefile: File name where cookies should be read from and dumped to.
a1ee09e8 190 nocheckcertificate:Do not verify SSL certificates
7e8c0af0
PH
191 prefer_insecure: Use HTTP instead of HTTPS to retrieve information.
192 At the moment, this is only supported by YouTube.
a1ee09e8 193 proxy: URL of the proxy server to use
91410c9b
PH
194 cn_verification_proxy: URL of the proxy to use for IP address verification
195 on Chinese sites. (Experimental)
e344693b 196 socket_timeout: Time to wait for unresponsive hosts, in seconds
0783b09b
PH
197 bidi_workaround: Work around buggy terminals without bidirectional text
198 support, using fridibi
a0ddb8a2 199 debug_printtraffic:Print out sent and received HTTP traffic
7b0817e8 200 include_ads: Download ads as well
04b4d394
PH
201 default_search: Prepend this string if an input url is not valid.
202 'auto' for elaborate guessing
62fec3b2 203 encoding: Use this encoding instead of the system-specified.
e8ee972c 204 extract_flat: Do not resolve URLs, return the immediate result.
057a5206
PH
205 Pass in 'in_playlist' to only show this behavior for
206 playlist items.
4f026faf 207 postprocessors: A list of dictionaries, each with an entry
71b640cc
PH
208 * key: The name of the postprocessor. See
209 youtube_dl/postprocessor/__init__.py for a list.
4f026faf
PH
210 as well as any further keyword arguments for the
211 postprocessor.
71b640cc
PH
212 progress_hooks: A list of functions that get called on download
213 progress, with a dictionary with the entries
5cda4eda 214 * status: One of "downloading", "error", or "finished".
ee69b99a 215 Check this first and ignore unknown values.
71b640cc 216
5cda4eda 217 If status is one of "downloading", or "finished", the
ee69b99a
PH
218 following properties may also be present:
219 * filename: The final filename (always present)
5cda4eda 220 * tmpfilename: The filename we're currently writing to
71b640cc
PH
221 * downloaded_bytes: Bytes on disk
222 * total_bytes: Size of the whole file, None if unknown
5cda4eda
PH
223 * total_bytes_estimate: Guess of the eventual file size,
224 None if unavailable.
225 * elapsed: The number of seconds since download started.
71b640cc
PH
226 * eta: The estimated time in seconds, None if unknown
227 * speed: The download speed in bytes/second, None if
228 unknown
5cda4eda
PH
229 * fragment_index: The counter of the currently
230 downloaded video fragment.
231 * fragment_count: The number of fragments (= individual
232 files that will be merged)
71b640cc
PH
233
234 Progress hooks are guaranteed to be called at least once
235 (with status "finished") if the download is successful.
45598f15 236 merge_output_format: Extension to use when merging formats.
6271f1ca
PH
237 fixup: Automatically correct known faults of the file.
238 One of:
239 - "never": do nothing
240 - "warn": only emit a warning
241 - "detect_or_warn": check whether we can do anything
62cd676c 242 about it, warn otherwise (default)
be4a824d 243 source_address: (Experimental) Client-side IP address to bind to.
6ec6cb4e 244 call_home: Boolean, true iff we are allowed to contact the
8bfa7545 245 youtube-dl servers for debugging.
5f0d813d 246 sleep_interval: Number of seconds to sleep before each download.
cfb56d1a
PH
247 listformats: Print an overview of available video formats and exit.
248 list_thumbnails: Print a table of all thumbnails and exit.
347de493
PH
249 match_filter: A function that gets called with the info_dict of
250 every video.
251 If it returns a message, the video is ignored.
252 If it returns None, the video is downloaded.
253 match_filter_func in utils.py is one example for this.
7e5db8c9 254 no_color: Do not emit color codes in output.
71b640cc 255
85729c51
PH
256 The following options determine which downloader is picked:
257 external_downloader: Executable of the external downloader to call.
258 None or unset for standard (built-in) downloader.
259 hls_prefer_native: Use the native HLS downloader instead of ffmpeg/avconv.
fe7e0c98 260
8222d8de 261 The following parameters are not used by YoutubeDL itself, they are used by
c75f0b36 262 the downloader (see youtube_dl/downloader/common.py):
8222d8de 263 nopart, updatetime, buffersize, ratelimit, min_filesize, max_filesize, test,
881e6a1f 264 noresizebuffer, retries, continuedl, noprogress, consoletitle,
c75f0b36 265 xattr_set_filesize, external_downloader_args.
76b1bd67
JMF
266
267 The following options are used by the post processors:
268 prefer_ffmpeg: If True, use ffmpeg instead of avconv if both are available,
269 otherwise prefer avconv.
f72b0a60
S
270 postprocessor_args: A list of additional command-line arguments for the
271 postprocessor.
8222d8de
JMF
272 """
273
274 params = None
275 _ies = []
276 _pps = []
277 _download_retcode = None
278 _num_downloads = None
279 _screen_file = None
280
3511266b 281 def __init__(self, params=None, auto_init=True):
8222d8de 282 """Create a FileDownloader object with the given options."""
e9f9a10f
JMF
283 if params is None:
284 params = {}
8222d8de 285 self._ies = []
56c73665 286 self._ies_instances = {}
8222d8de 287 self._pps = []
933605d7 288 self._progress_hooks = []
8222d8de
JMF
289 self._download_retcode = 0
290 self._num_downloads = 0
291 self._screen_file = [sys.stdout, sys.stderr][params.get('logtostderr', False)]
0783b09b 292 self._err_file = sys.stderr
4abf617b
S
293 self.params = {
294 # Default parameters
295 'nocheckcertificate': False,
296 }
297 self.params.update(params)
a0e07d31 298 self.cache = Cache(self)
34308b30 299
0783b09b 300 if params.get('bidi_workaround', False):
1c088fa8
PH
301 try:
302 import pty
303 master, slave = pty.openpty()
003c69a8 304 width = compat_get_terminal_size().columns
1c088fa8
PH
305 if width is None:
306 width_args = []
307 else:
308 width_args = ['-w', str(width)]
5d681e96 309 sp_kwargs = dict(
1c088fa8
PH
310 stdin=subprocess.PIPE,
311 stdout=slave,
312 stderr=self._err_file)
5d681e96
PH
313 try:
314 self._output_process = subprocess.Popen(
315 ['bidiv'] + width_args, **sp_kwargs
316 )
317 except OSError:
5d681e96
PH
318 self._output_process = subprocess.Popen(
319 ['fribidi', '-c', 'UTF-8'] + width_args, **sp_kwargs)
320 self._output_channel = os.fdopen(master, 'rb')
1c088fa8
PH
321 except OSError as ose:
322 if ose.errno == 2:
6febd1c1 323 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 $PATH.')
1c088fa8
PH
324 else:
325 raise
0783b09b 326
34308b30 327 if (sys.version_info >= (3,) and sys.platform != 'win32' and
8fb3ac36
PH
328 sys.getfilesystemencoding() in ['ascii', 'ANSI_X3.4-1968'] and
329 not params.get('restrictfilenames', False)):
34308b30
PH
330 # On Python 3, the Unicode filesystem API will throw errors (#1474)
331 self.report_warning(
6febd1c1 332 'Assuming --restrict-filenames since file system encoding '
1b725173 333 'cannot encode all characters. '
6febd1c1 334 'Set the LC_ALL environment variable to fix this.')
4a98cdbf 335 self.params['restrictfilenames'] = True
34308b30 336
486dd09e
PH
337 if isinstance(params.get('outtmpl'), bytes):
338 self.report_warning(
339 'Parameter outtmpl is bytes, but should be a unicode string. '
340 'Put from __future__ import unicode_literals at the top of your code file or consider switching to Python 3.x.')
341
dca08720
PH
342 self._setup_opener()
343
3511266b
PH
344 if auto_init:
345 self.print_debug_header()
346 self.add_default_info_extractors()
347
4f026faf
PH
348 for pp_def_raw in self.params.get('postprocessors', []):
349 pp_class = get_postprocessor(pp_def_raw['key'])
350 pp_def = dict(pp_def_raw)
351 del pp_def['key']
352 pp = pp_class(self, **compat_kwargs(pp_def))
353 self.add_post_processor(pp)
354
71b640cc
PH
355 for ph in self.params.get('progress_hooks', []):
356 self.add_progress_hook(ph)
357
7d4111ed
PH
358 def warn_if_short_id(self, argv):
359 # short YouTube ID starting with dash?
360 idxs = [
361 i for i, a in enumerate(argv)
362 if re.match(r'^-[0-9A-Za-z_-]{10}$', a)]
363 if idxs:
364 correct_argv = (
365 ['youtube-dl'] +
366 [a for i, a in enumerate(argv) if i not in idxs] +
367 ['--'] + [argv[i] for i in idxs]
368 )
369 self.report_warning(
370 'Long argument string detected. '
371 'Use -- to separate parameters and URLs, like this:\n%s\n' %
372 args_to_str(correct_argv))
373
8222d8de
JMF
374 def add_info_extractor(self, ie):
375 """Add an InfoExtractor object to the end of the list."""
376 self._ies.append(ie)
56c73665 377 self._ies_instances[ie.ie_key()] = ie
8222d8de
JMF
378 ie.set_downloader(self)
379
56c73665
JMF
380 def get_info_extractor(self, ie_key):
381 """
382 Get an instance of an IE with name ie_key, it will try to get one from
383 the _ies list, if there's no instance it will create a new one and add
384 it to the extractor list.
385 """
386 ie = self._ies_instances.get(ie_key)
387 if ie is None:
388 ie = get_info_extractor(ie_key)()
389 self.add_info_extractor(ie)
390 return ie
391
023fa8c4
JMF
392 def add_default_info_extractors(self):
393 """
394 Add the InfoExtractors returned by gen_extractors to the end of the list
395 """
396 for ie in gen_extractors():
397 self.add_info_extractor(ie)
398
8222d8de
JMF
399 def add_post_processor(self, pp):
400 """Add a PostProcessor object to the end of the chain."""
401 self._pps.append(pp)
402 pp.set_downloader(self)
403
933605d7
JMF
404 def add_progress_hook(self, ph):
405 """Add the progress hook (currently only for the file downloader)"""
406 self._progress_hooks.append(ph)
8ab470f1 407
1c088fa8 408 def _bidi_workaround(self, message):
5d681e96 409 if not hasattr(self, '_output_channel'):
1c088fa8
PH
410 return message
411
5d681e96 412 assert hasattr(self, '_output_process')
11b85ce6 413 assert isinstance(message, compat_str)
6febd1c1
PH
414 line_count = message.count('\n') + 1
415 self._output_process.stdin.write((message + '\n').encode('utf-8'))
5d681e96 416 self._output_process.stdin.flush()
6febd1c1 417 res = ''.join(self._output_channel.readline().decode('utf-8')
9e1a5b84 418 for _ in range(line_count))
6febd1c1 419 return res[:-len('\n')]
1c088fa8 420
8222d8de 421 def to_screen(self, message, skip_eol=False):
0783b09b
PH
422 """Print message to stdout if not in quiet mode."""
423 return self.to_stdout(message, skip_eol, check_quiet=True)
424
734f90bb 425 def _write_string(self, s, out=None):
b58ddb32 426 write_string(s, out=out, encoding=self.params.get('encoding'))
734f90bb 427
0783b09b 428 def to_stdout(self, message, skip_eol=False, check_quiet=False):
8222d8de 429 """Print message to stdout if not in quiet mode."""
8bf9319e 430 if self.params.get('logger'):
43afe285 431 self.params['logger'].debug(message)
0783b09b 432 elif not check_quiet or not self.params.get('quiet', False):
1c088fa8 433 message = self._bidi_workaround(message)
6febd1c1 434 terminator = ['\n', ''][skip_eol]
8222d8de 435 output = message + terminator
1c088fa8 436
734f90bb 437 self._write_string(output, self._screen_file)
8222d8de
JMF
438
439 def to_stderr(self, message):
440 """Print message to stderr."""
11b85ce6 441 assert isinstance(message, compat_str)
8bf9319e 442 if self.params.get('logger'):
43afe285
IB
443 self.params['logger'].error(message)
444 else:
1c088fa8 445 message = self._bidi_workaround(message)
6febd1c1 446 output = message + '\n'
734f90bb 447 self._write_string(output, self._err_file)
8222d8de 448
1e5b9a95
PH
449 def to_console_title(self, message):
450 if not self.params.get('consoletitle', False):
451 return
452 if os.name == 'nt' and ctypes.windll.kernel32.GetConsoleWindow():
453 # c_wchar_p() might not be necessary if `message` is
454 # already of type unicode()
455 ctypes.windll.kernel32.SetConsoleTitleW(ctypes.c_wchar_p(message))
456 elif 'TERM' in os.environ:
734f90bb 457 self._write_string('\033]0;%s\007' % message, self._screen_file)
1e5b9a95 458
bdde425c
PH
459 def save_console_title(self):
460 if not self.params.get('consoletitle', False):
461 return
462 if 'TERM' in os.environ:
efd6c574 463 # Save the title on stack
734f90bb 464 self._write_string('\033[22;0t', self._screen_file)
bdde425c
PH
465
466 def restore_console_title(self):
467 if not self.params.get('consoletitle', False):
468 return
469 if 'TERM' in os.environ:
efd6c574 470 # Restore the title from stack
734f90bb 471 self._write_string('\033[23;0t', self._screen_file)
bdde425c
PH
472
473 def __enter__(self):
474 self.save_console_title()
475 return self
476
477 def __exit__(self, *args):
478 self.restore_console_title()
f89197d7 479
dca08720
PH
480 if self.params.get('cookiefile') is not None:
481 self.cookiejar.save()
bdde425c 482
8222d8de
JMF
483 def trouble(self, message=None, tb=None):
484 """Determine action to take when a download problem appears.
485
486 Depending on if the downloader has been configured to ignore
487 download errors or not, this method may throw an exception or
488 not when errors are found, after printing the message.
489
490 tb, if given, is additional traceback information.
491 """
492 if message is not None:
493 self.to_stderr(message)
494 if self.params.get('verbose'):
495 if tb is None:
496 if sys.exc_info()[0]: # if .trouble has been called from an except block
6febd1c1 497 tb = ''
8222d8de 498 if hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
6febd1c1 499 tb += ''.join(traceback.format_exception(*sys.exc_info()[1].exc_info))
c0384f22 500 tb += encode_compat_str(traceback.format_exc())
8222d8de
JMF
501 else:
502 tb_data = traceback.format_list(traceback.extract_stack())
6febd1c1 503 tb = ''.join(tb_data)
8222d8de
JMF
504 self.to_stderr(tb)
505 if not self.params.get('ignoreerrors', False):
506 if sys.exc_info()[0] and hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
507 exc_info = sys.exc_info()[1].exc_info
508 else:
509 exc_info = sys.exc_info()
510 raise DownloadError(message, exc_info)
511 self._download_retcode = 1
512
513 def report_warning(self, message):
514 '''
515 Print the message to stderr, it will be prefixed with 'WARNING:'
516 If stderr is a tty file the 'WARNING:' will be colored
517 '''
6d07ce01
JMF
518 if self.params.get('logger') is not None:
519 self.params['logger'].warning(message)
8222d8de 520 else:
ad8915b7
PH
521 if self.params.get('no_warnings'):
522 return
7e5db8c9 523 if not self.params.get('no_color') and self._err_file.isatty() and os.name != 'nt':
6d07ce01
JMF
524 _msg_header = '\033[0;33mWARNING:\033[0m'
525 else:
526 _msg_header = 'WARNING:'
527 warning_message = '%s %s' % (_msg_header, message)
528 self.to_stderr(warning_message)
8222d8de
JMF
529
530 def report_error(self, message, tb=None):
531 '''
532 Do the same as trouble, but prefixes the message with 'ERROR:', colored
533 in red if stderr is a tty file.
534 '''
7e5db8c9 535 if not self.params.get('no_color') and self._err_file.isatty() and os.name != 'nt':
6febd1c1 536 _msg_header = '\033[0;31mERROR:\033[0m'
8222d8de 537 else:
6febd1c1
PH
538 _msg_header = 'ERROR:'
539 error_message = '%s %s' % (_msg_header, message)
8222d8de
JMF
540 self.trouble(error_message, tb)
541
8222d8de
JMF
542 def report_file_already_downloaded(self, file_name):
543 """Report file has already been fully downloaded."""
544 try:
6febd1c1 545 self.to_screen('[download] %s has already been downloaded' % file_name)
ce02ed60 546 except UnicodeEncodeError:
6febd1c1 547 self.to_screen('[download] The file has already been downloaded')
8222d8de 548
8222d8de
JMF
549 def prepare_filename(self, info_dict):
550 """Generate the output filename."""
551 try:
552 template_dict = dict(info_dict)
553
554 template_dict['epoch'] = int(time.time())
555 autonumber_size = self.params.get('autonumber_size')
556 if autonumber_size is None:
557 autonumber_size = 5
6febd1c1 558 autonumber_templ = '%0' + str(autonumber_size) + 'd'
8222d8de 559 template_dict['autonumber'] = autonumber_templ % self._num_downloads
702665c0 560 if template_dict.get('playlist_index') is not None:
c6b4132a 561 template_dict['playlist_index'] = '%0*d' % (len(str(template_dict['n_entries'])), template_dict['playlist_index'])
17b75c0d
PH
562 if template_dict.get('resolution') is None:
563 if template_dict.get('width') and template_dict.get('height'):
564 template_dict['resolution'] = '%dx%d' % (template_dict['width'], template_dict['height'])
565 elif template_dict.get('height'):
805ef3c6 566 template_dict['resolution'] = '%sp' % template_dict['height']
17b75c0d 567 elif template_dict.get('width'):
805ef3c6 568 template_dict['resolution'] = '?x%d' % template_dict['width']
8222d8de 569
586a91b6 570 sanitize = lambda k, v: sanitize_filename(
45598aab 571 compat_str(v),
1bb5c511 572 restricted=self.params.get('restrictfilenames'),
6febd1c1 573 is_id=(k == 'id'))
586a91b6 574 template_dict = dict((k, sanitize(k, v))
45598aab
PH
575 for k, v in template_dict.items()
576 if v is not None)
6febd1c1 577 template_dict = collections.defaultdict(lambda: 'NA', template_dict)
8222d8de 578
b3613d36 579 outtmpl = self.params.get('outtmpl', DEFAULT_OUTTMPL)
4644ac55 580 tmpl = compat_expanduser(outtmpl)
586a91b6 581 filename = tmpl % template_dict
3a0d2f52
S
582 # Temporary fix for #4787
583 # 'Treat' all problem characters by passing filename through preferredencoding
584 # to workaround encoding issues with subprocess on python2 @ Windows
585 if sys.version_info < (3, 0) and sys.platform == 'win32':
586 filename = encodeFilename(filename, True).decode(preferredencoding())
b3613d36 587 return sanitize_path(filename)
8222d8de 588 except ValueError as err:
6febd1c1 589 self.report_error('Error in output template: ' + str(err) + ' (encoding: ' + repr(preferredencoding()) + ')')
8222d8de
JMF
590 return None
591
442c37b7 592 def _match_entry(self, info_dict, incomplete):
6ec6cb4e 593 """ Returns None iff the file should be downloaded """
8222d8de 594
6febd1c1 595 video_title = info_dict.get('title', info_dict.get('id', 'video'))
7012b23c
PH
596 if 'title' in info_dict:
597 # This can happen when we're just evaluating the playlist
598 title = info_dict['title']
599 matchtitle = self.params.get('matchtitle', False)
600 if matchtitle:
601 if not re.search(matchtitle, title, re.IGNORECASE):
6febd1c1 602 return '"' + title + '" title did not match pattern "' + matchtitle + '"'
7012b23c
PH
603 rejecttitle = self.params.get('rejecttitle', False)
604 if rejecttitle:
605 if re.search(rejecttitle, title, re.IGNORECASE):
6febd1c1 606 return '"' + title + '" title matched reject pattern "' + rejecttitle + '"'
8222d8de
JMF
607 date = info_dict.get('upload_date', None)
608 if date is not None:
609 dateRange = self.params.get('daterange', DateRange())
610 if date not in dateRange:
6febd1c1 611 return '%s upload date is not in range %s' % (date_from_str(date).isoformat(), dateRange)
5fe18bdb
PH
612 view_count = info_dict.get('view_count', None)
613 if view_count is not None:
614 min_views = self.params.get('min_views')
615 if min_views is not None and view_count < min_views:
6febd1c1 616 return 'Skipping %s, because it has not reached minimum view count (%d/%d)' % (video_title, view_count, min_views)
5fe18bdb
PH
617 max_views = self.params.get('max_views')
618 if max_views is not None and view_count > max_views:
6febd1c1 619 return 'Skipping %s, because it has exceeded the maximum view count (%d/%d)' % (video_title, view_count, max_views)
05900629 620 if age_restricted(info_dict.get('age_limit'), self.params.get('age_limit')):
347de493 621 return 'Skipping "%s" because it is age restricted' % video_title
c1c9a79c 622 if self.in_download_archive(info_dict):
6febd1c1 623 return '%s has already been recorded in archive' % video_title
347de493 624
442c37b7
PH
625 if not incomplete:
626 match_filter = self.params.get('match_filter')
627 if match_filter is not None:
628 ret = match_filter(info_dict)
629 if ret is not None:
630 return ret
347de493 631
8222d8de 632 return None
fe7e0c98 633
b6c45014
JMF
634 @staticmethod
635 def add_extra_info(info_dict, extra_info):
636 '''Set the keys from extra_info in info dict if they are missing'''
637 for key, value in extra_info.items():
638 info_dict.setdefault(key, value)
639
7fc3fa05 640 def extract_info(self, url, download=True, ie_key=None, extra_info={},
61aa5ba3 641 process=True, force_generic_extractor=False):
8222d8de
JMF
642 '''
643 Returns a list with a dictionary for each video we find.
644 If 'download', also downloads the videos.
645 extra_info is a dict containing the extra values to add to each result
613b2d9d 646 '''
fe7e0c98 647
61aa5ba3 648 if not ie_key and force_generic_extractor:
d22dec74
S
649 ie_key = 'Generic'
650
8222d8de 651 if ie_key:
56c73665 652 ies = [self.get_info_extractor(ie_key)]
8222d8de
JMF
653 else:
654 ies = self._ies
655
656 for ie in ies:
657 if not ie.suitable(url):
658 continue
659
660 if not ie.working():
6febd1c1
PH
661 self.report_warning('The program functionality for this site has been marked as broken, '
662 'and will probably not work.')
8222d8de
JMF
663
664 try:
665 ie_result = ie.extract(url)
5f6a1245 666 if ie_result is None: # Finished already (backwards compatibility; listformats and friends should be moved here)
8222d8de
JMF
667 break
668 if isinstance(ie_result, list):
669 # Backwards compatibility: old IE result format
8222d8de
JMF
670 ie_result = {
671 '_type': 'compat_list',
672 'entries': ie_result,
673 }
ea38e55f 674 self.add_default_extra_info(ie_result, ie, url)
7fc3fa05
PH
675 if process:
676 return self.process_ie_result(ie_result, download, extra_info)
677 else:
678 return ie_result
fb043a6e 679 except ExtractorError as e: # An error we somewhat expected
2c74e6fa 680 self.report_error(compat_str(e), e.format_traceback())
8222d8de 681 break
d3e5bbf4
PH
682 except MaxDownloadsReached:
683 raise
8222d8de
JMF
684 except Exception as e:
685 if self.params.get('ignoreerrors', False):
9b9c5355 686 self.report_error(error_to_compat_str(e), tb=encode_compat_str(traceback.format_exc()))
8222d8de
JMF
687 break
688 else:
689 raise
690 else:
1a489545 691 self.report_error('no suitable InfoExtractor for URL %s' % url)
fe7e0c98 692
ea38e55f
PH
693 def add_default_extra_info(self, ie_result, ie, url):
694 self.add_extra_info(ie_result, {
695 'extractor': ie.IE_NAME,
696 'webpage_url': url,
697 'webpage_url_basename': url_basename(url),
698 'extractor_key': ie.ie_key(),
699 })
700
8222d8de
JMF
701 def process_ie_result(self, ie_result, download=True, extra_info={}):
702 """
703 Take the result of the ie(may be modified) and resolve all unresolved
704 references (URLs, playlist items).
705
706 It will also download the videos if 'download'.
707 Returns the resolved ie_result.
708 """
709
e8ee972c
PH
710 result_type = ie_result.get('_type', 'video')
711
057a5206
PH
712 if result_type in ('url', 'url_transparent'):
713 extract_flat = self.params.get('extract_flat', False)
714 if ((extract_flat == 'in_playlist' and 'playlist' in extra_info) or
715 extract_flat is True):
057a5206
PH
716 if self.params.get('forcejson', False):
717 self.to_stdout(json.dumps(ie_result))
e8ee972c
PH
718 return ie_result
719
8222d8de 720 if result_type == 'video':
b6c45014 721 self.add_extra_info(ie_result, extra_info)
feee2ecf 722 return self.process_video_result(ie_result, download=download)
8222d8de
JMF
723 elif result_type == 'url':
724 # We have to add extra_info to the results because it may be
725 # contained in a playlist
726 return self.extract_info(ie_result['url'],
727 download,
728 ie_key=ie_result.get('ie_key'),
729 extra_info=extra_info)
7fc3fa05
PH
730 elif result_type == 'url_transparent':
731 # Use the information from the embedding page
732 info = self.extract_info(
733 ie_result['url'], ie_key=ie_result.get('ie_key'),
734 extra_info=extra_info, download=False, process=False)
735
412c617d
PH
736 force_properties = dict(
737 (k, v) for k, v in ie_result.items() if v is not None)
738 for f in ('_type', 'url'):
739 if f in force_properties:
740 del force_properties[f]
741 new_result = info.copy()
742 new_result.update(force_properties)
7fc3fa05
PH
743
744 assert new_result.get('_type') != 'url_transparent'
7fc3fa05
PH
745
746 return self.process_ie_result(
747 new_result, download=download, extra_info=extra_info)
42e12102 748 elif result_type == 'playlist' or result_type == 'multi_video':
8222d8de
JMF
749 # We process each entry in the playlist
750 playlist = ie_result.get('title', None) or ie_result.get('id', None)
6febd1c1 751 self.to_screen('[download] Downloading playlist: %s' % playlist)
8222d8de
JMF
752
753 playlist_results = []
754
8222d8de 755 playliststart = self.params.get('playliststart', 1) - 1
a19fd00c
PH
756 playlistend = self.params.get('playlistend', None)
757 # For backwards compatibility, interpret -1 as whole list
8222d8de 758 if playlistend == -1:
a19fd00c 759 playlistend = None
8222d8de 760
c14e88f0
PH
761 playlistitems_str = self.params.get('playlist_items', None)
762 playlistitems = None
763 if playlistitems_str is not None:
764 def iter_playlistitems(format):
765 for string_segment in format.split(','):
766 if '-' in string_segment:
767 start, end = string_segment.split('-')
768 for item in range(int(start), int(end) + 1):
769 yield int(item)
770 else:
771 yield int(string_segment)
772 playlistitems = iter_playlistitems(playlistitems_str)
773
b82f815f
PH
774 ie_entries = ie_result['entries']
775 if isinstance(ie_entries, list):
776 n_all_entries = len(ie_entries)
c14e88f0 777 if playlistitems:
3884dcf3
JMF
778 entries = [
779 ie_entries[i - 1] for i in playlistitems
780 if -n_all_entries <= i - 1 < n_all_entries]
c14e88f0
PH
781 else:
782 entries = ie_entries[playliststart:playlistend]
b7ab0590
PH
783 n_entries = len(entries)
784 self.to_screen(
785 "[%s] playlist %s: Collected %d video ids (downloading %d of them)" %
786 (ie_result['extractor'], playlist, n_all_entries, n_entries))
b82f815f 787 elif isinstance(ie_entries, PagedList):
c14e88f0
PH
788 if playlistitems:
789 entries = []
790 for item in playlistitems:
791 entries.extend(ie_entries.getslice(
792 item - 1, item
793 ))
794 else:
795 entries = ie_entries.getslice(
796 playliststart, playlistend)
b7ab0590
PH
797 n_entries = len(entries)
798 self.to_screen(
799 "[%s] playlist %s: Downloading %d videos" %
800 (ie_result['extractor'], playlist, n_entries))
b82f815f 801 else: # iterable
c14e88f0
PH
802 if playlistitems:
803 entry_list = list(ie_entries)
804 entries = [entry_list[i - 1] for i in playlistitems]
805 else:
806 entries = list(itertools.islice(
807 ie_entries, playliststart, playlistend))
b82f815f
PH
808 n_entries = len(entries)
809 self.to_screen(
810 "[%s] playlist %s: Downloading %d videos" %
811 (ie_result['extractor'], playlist, n_entries))
8222d8de 812
ff815fe6
MS
813 if self.params.get('playlistreverse', False):
814 entries = entries[::-1]
815
fe7e0c98 816 for i, entry in enumerate(entries, 1):
734ea11e 817 self.to_screen('[download] Downloading video %s of %s' % (i, n_entries))
8222d8de 818 extra = {
c6b4132a 819 'n_entries': n_entries,
fe7e0c98 820 'playlist': playlist,
a1cf99d0
PH
821 'playlist_id': ie_result.get('id'),
822 'playlist_title': ie_result.get('title'),
fe7e0c98 823 'playlist_index': i + playliststart,
b6c45014 824 'extractor': ie_result['extractor'],
9103bbc5 825 'webpage_url': ie_result['webpage_url'],
29eb5174 826 'webpage_url_basename': url_basename(ie_result['webpage_url']),
be97abc2 827 'extractor_key': ie_result['extractor_key'],
fe7e0c98 828 }
7012b23c 829
442c37b7 830 reason = self._match_entry(entry, incomplete=True)
7012b23c 831 if reason is not None:
6febd1c1 832 self.to_screen('[download] ' + reason)
7012b23c
PH
833 continue
834
8222d8de
JMF
835 entry_result = self.process_ie_result(entry,
836 download=download,
837 extra_info=extra)
838 playlist_results.append(entry_result)
839 ie_result['entries'] = playlist_results
371c3b79 840 self.to_screen('[download] Finished downloading playlist: %s' % playlist)
8222d8de
JMF
841 return ie_result
842 elif result_type == 'compat_list':
c9bf4114
PH
843 self.report_warning(
844 'Extractor %s returned a compat_list result. '
845 'It needs to be updated.' % ie_result.get('extractor'))
5f6a1245 846
8222d8de 847 def _fixup(r):
9e1a5b84
JW
848 self.add_extra_info(
849 r,
9103bbc5
JMF
850 {
851 'extractor': ie_result['extractor'],
852 'webpage_url': ie_result['webpage_url'],
29eb5174 853 'webpage_url_basename': url_basename(ie_result['webpage_url']),
be97abc2 854 'extractor_key': ie_result['extractor_key'],
9e1a5b84
JW
855 }
856 )
8222d8de
JMF
857 return r
858 ie_result['entries'] = [
b6c45014 859 self.process_ie_result(_fixup(r), download, extra_info)
8222d8de
JMF
860 for r in ie_result['entries']
861 ]
862 return ie_result
863 else:
864 raise Exception('Invalid result type: %s' % result_type)
865
67134eab
JMF
866 def _build_format_filter(self, filter_spec):
867 " Returns a function to filter the formats according to the filter_spec "
083c9df9
PH
868
869 OPERATORS = {
870 '<': operator.lt,
871 '<=': operator.le,
872 '>': operator.gt,
873 '>=': operator.ge,
874 '=': operator.eq,
875 '!=': operator.ne,
876 }
67134eab 877 operator_rex = re.compile(r'''(?x)\s*
2ec19e95 878 (?P<key>width|height|tbr|abr|vbr|asr|filesize|fps)
083c9df9
PH
879 \s*(?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
880 (?P<value>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?)
67134eab 881 $
083c9df9 882 ''' % '|'.join(map(re.escape, OPERATORS.keys())))
67134eab 883 m = operator_rex.search(filter_spec)
9ddb6925
S
884 if m:
885 try:
886 comparison_value = int(m.group('value'))
887 except ValueError:
888 comparison_value = parse_filesize(m.group('value'))
889 if comparison_value is None:
890 comparison_value = parse_filesize(m.group('value') + 'B')
891 if comparison_value is None:
892 raise ValueError(
893 'Invalid value %r in format specification %r' % (
67134eab 894 m.group('value'), filter_spec))
9ddb6925
S
895 op = OPERATORS[m.group('op')]
896
083c9df9 897 if not m:
9ddb6925
S
898 STR_OPERATORS = {
899 '=': operator.eq,
900 '!=': operator.ne,
901 }
67134eab 902 str_operator_rex = re.compile(r'''(?x)
9ddb6925
S
903 \s*(?P<key>ext|acodec|vcodec|container|protocol)
904 \s*(?P<op>%s)(?P<none_inclusive>\s*\?)?
905 \s*(?P<value>[a-zA-Z0-9_-]+)
67134eab 906 \s*$
9ddb6925 907 ''' % '|'.join(map(re.escape, STR_OPERATORS.keys())))
67134eab 908 m = str_operator_rex.search(filter_spec)
9ddb6925
S
909 if m:
910 comparison_value = m.group('value')
911 op = STR_OPERATORS[m.group('op')]
083c9df9 912
9ddb6925 913 if not m:
67134eab 914 raise ValueError('Invalid filter specification %r' % filter_spec)
083c9df9
PH
915
916 def _filter(f):
917 actual_value = f.get(m.group('key'))
918 if actual_value is None:
919 return m.group('none_inclusive')
920 return op(actual_value, comparison_value)
67134eab
JMF
921 return _filter
922
923 def build_format_selector(self, format_spec):
924 def syntax_error(note, start):
925 message = (
926 'Invalid format specification: '
927 '{0}\n\t{1}\n\t{2}^'.format(note, format_spec, ' ' * start[1]))
928 return SyntaxError(message)
929
930 PICKFIRST = 'PICKFIRST'
931 MERGE = 'MERGE'
932 SINGLE = 'SINGLE'
0130afb7 933 GROUP = 'GROUP'
67134eab
JMF
934 FormatSelector = collections.namedtuple('FormatSelector', ['type', 'selector', 'filters'])
935
936 def _parse_filter(tokens):
937 filter_parts = []
938 for type, string, start, _, _ in tokens:
939 if type == tokenize.OP and string == ']':
940 return ''.join(filter_parts)
941 else:
942 filter_parts.append(string)
943
232541df 944 def _remove_unused_ops(tokens):
17cc1534 945 # Remove operators that we don't use and join them with the surrounding strings
232541df
JMF
946 # for example: 'mp4' '-' 'baseline' '-' '16x9' is converted to 'mp4-baseline-16x9'
947 ALLOWED_OPS = ('/', '+', ',', '(', ')')
948 last_string, last_start, last_end, last_line = None, None, None, None
949 for type, string, start, end, line in tokens:
950 if type == tokenize.OP and string == '[':
951 if last_string:
952 yield tokenize.NAME, last_string, last_start, last_end, last_line
953 last_string = None
954 yield type, string, start, end, line
955 # everything inside brackets will be handled by _parse_filter
956 for type, string, start, end, line in tokens:
957 yield type, string, start, end, line
958 if type == tokenize.OP and string == ']':
959 break
960 elif type == tokenize.OP and string in ALLOWED_OPS:
961 if last_string:
962 yield tokenize.NAME, last_string, last_start, last_end, last_line
963 last_string = None
964 yield type, string, start, end, line
965 elif type in [tokenize.NAME, tokenize.NUMBER, tokenize.OP]:
966 if not last_string:
967 last_string = string
968 last_start = start
969 last_end = end
970 else:
971 last_string += string
972 if last_string:
973 yield tokenize.NAME, last_string, last_start, last_end, last_line
974
cf2ac6df 975 def _parse_format_selection(tokens, inside_merge=False, inside_choice=False, inside_group=False):
67134eab
JMF
976 selectors = []
977 current_selector = None
978 for type, string, start, _, _ in tokens:
979 # ENCODING is only defined in python 3.x
980 if type == getattr(tokenize, 'ENCODING', None):
981 continue
982 elif type in [tokenize.NAME, tokenize.NUMBER]:
983 current_selector = FormatSelector(SINGLE, string, [])
984 elif type == tokenize.OP:
cf2ac6df
JMF
985 if string == ')':
986 if not inside_group:
987 # ')' will be handled by the parentheses group
988 tokens.restore_last_token()
67134eab 989 break
cf2ac6df 990 elif inside_merge and string in ['/', ',']:
0130afb7
JMF
991 tokens.restore_last_token()
992 break
cf2ac6df
JMF
993 elif inside_choice and string == ',':
994 tokens.restore_last_token()
995 break
996 elif string == ',':
0a31a350
JMF
997 if not current_selector:
998 raise syntax_error('"," must follow a format selector', start)
67134eab
JMF
999 selectors.append(current_selector)
1000 current_selector = None
1001 elif string == '/':
d96d604e
JMF
1002 if not current_selector:
1003 raise syntax_error('"/" must follow a format selector', start)
67134eab 1004 first_choice = current_selector
cf2ac6df 1005 second_choice = _parse_format_selection(tokens, inside_choice=True)
f5f4a27a 1006 current_selector = FormatSelector(PICKFIRST, (first_choice, second_choice), [])
67134eab
JMF
1007 elif string == '[':
1008 if not current_selector:
1009 current_selector = FormatSelector(SINGLE, 'best', [])
1010 format_filter = _parse_filter(tokens)
1011 current_selector.filters.append(format_filter)
0130afb7
JMF
1012 elif string == '(':
1013 if current_selector:
1014 raise syntax_error('Unexpected "("', start)
cf2ac6df
JMF
1015 group = _parse_format_selection(tokens, inside_group=True)
1016 current_selector = FormatSelector(GROUP, group, [])
67134eab
JMF
1017 elif string == '+':
1018 video_selector = current_selector
cf2ac6df 1019 audio_selector = _parse_format_selection(tokens, inside_merge=True)
0a31a350
JMF
1020 if not video_selector or not audio_selector:
1021 raise syntax_error('"+" must be between two format selectors', start)
cf2ac6df 1022 current_selector = FormatSelector(MERGE, (video_selector, audio_selector), [])
67134eab
JMF
1023 else:
1024 raise syntax_error('Operator not recognized: "{0}"'.format(string), start)
1025 elif type == tokenize.ENDMARKER:
1026 break
1027 if current_selector:
1028 selectors.append(current_selector)
1029 return selectors
1030
1031 def _build_selector_function(selector):
1032 if isinstance(selector, list):
1033 fs = [_build_selector_function(s) for s in selector]
1034
1035 def selector_function(formats):
1036 for f in fs:
1037 for format in f(formats):
1038 yield format
1039 return selector_function
0130afb7
JMF
1040 elif selector.type == GROUP:
1041 selector_function = _build_selector_function(selector.selector)
67134eab
JMF
1042 elif selector.type == PICKFIRST:
1043 fs = [_build_selector_function(s) for s in selector.selector]
1044
1045 def selector_function(formats):
1046 for f in fs:
1047 picked_formats = list(f(formats))
1048 if picked_formats:
1049 return picked_formats
1050 return []
1051 elif selector.type == SINGLE:
1052 format_spec = selector.selector
1053
1054 def selector_function(formats):
bb8e5536
JMF
1055 formats = list(formats)
1056 if not formats:
1057 return
5acfa126
JMF
1058 if format_spec == 'all':
1059 for f in formats:
1060 yield f
1061 elif format_spec in ['best', 'worst', None]:
67134eab
JMF
1062 format_idx = 0 if format_spec == 'worst' else -1
1063 audiovideo_formats = [
1064 f for f in formats
1065 if f.get('vcodec') != 'none' and f.get('acodec') != 'none']
1066 if audiovideo_formats:
1067 yield audiovideo_formats[format_idx]
1068 # for audio only (soundcloud) or video only (imgur) urls, select the best/worst audio format
1069 elif (all(f.get('acodec') != 'none' for f in formats) or
1070 all(f.get('vcodec') != 'none' for f in formats)):
1071 yield formats[format_idx]
1072 elif format_spec == 'bestaudio':
1073 audio_formats = [
1074 f for f in formats
1075 if f.get('vcodec') == 'none']
1076 if audio_formats:
1077 yield audio_formats[-1]
1078 elif format_spec == 'worstaudio':
1079 audio_formats = [
1080 f for f in formats
1081 if f.get('vcodec') == 'none']
1082 if audio_formats:
1083 yield audio_formats[0]
1084 elif format_spec == 'bestvideo':
1085 video_formats = [
1086 f for f in formats
1087 if f.get('acodec') == 'none']
1088 if video_formats:
1089 yield video_formats[-1]
1090 elif format_spec == 'worstvideo':
1091 video_formats = [
1092 f for f in formats
1093 if f.get('acodec') == 'none']
1094 if video_formats:
1095 yield video_formats[0]
1096 else:
1097 extensions = ['mp4', 'flv', 'webm', '3gp', 'm4a', 'mp3', 'ogg', 'aac', 'wav']
1098 if format_spec in extensions:
1099 filter_f = lambda f: f['ext'] == format_spec
1100 else:
1101 filter_f = lambda f: f['format_id'] == format_spec
1102 matches = list(filter(filter_f, formats))
1103 if matches:
1104 yield matches[-1]
1105 elif selector.type == MERGE:
1106 def _merge(formats_info):
1107 format_1, format_2 = [f['format_id'] for f in formats_info]
1108 # The first format must contain the video and the
1109 # second the audio
1110 if formats_info[0].get('vcodec') == 'none':
1111 self.report_error('The first format must '
1112 'contain the video, try using '
1113 '"-f %s+%s"' % (format_2, format_1))
1114 return
3d24bbfb
S
1115 # Formats must be opposite (video+audio)
1116 if formats_info[0].get('acodec') == 'none' and formats_info[1].get('acodec') == 'none':
1117 self.report_error(
1118 'Both formats %s and %s are video-only, you must specify "-f video+audio"'
1119 % (format_1, format_2))
1120 return
67134eab
JMF
1121 output_ext = (
1122 formats_info[0]['ext']
1123 if self.params.get('merge_output_format') is None
1124 else self.params['merge_output_format'])
1125 return {
1126 'requested_formats': formats_info,
1127 'format': '%s+%s' % (formats_info[0].get('format'),
1128 formats_info[1].get('format')),
1129 'format_id': '%s+%s' % (formats_info[0].get('format_id'),
1130 formats_info[1].get('format_id')),
1131 'width': formats_info[0].get('width'),
1132 'height': formats_info[0].get('height'),
1133 'resolution': formats_info[0].get('resolution'),
1134 'fps': formats_info[0].get('fps'),
1135 'vcodec': formats_info[0].get('vcodec'),
1136 'vbr': formats_info[0].get('vbr'),
1137 'stretched_ratio': formats_info[0].get('stretched_ratio'),
1138 'acodec': formats_info[1].get('acodec'),
1139 'abr': formats_info[1].get('abr'),
1140 'ext': output_ext,
1141 }
1142 video_selector, audio_selector = map(_build_selector_function, selector.selector)
083c9df9 1143
67134eab
JMF
1144 def selector_function(formats):
1145 formats = list(formats)
1146 for pair in itertools.product(video_selector(formats), audio_selector(formats)):
1147 yield _merge(pair)
083c9df9 1148
67134eab 1149 filters = [self._build_format_filter(f) for f in selector.filters]
083c9df9 1150
67134eab
JMF
1151 def final_selector(formats):
1152 for _filter in filters:
1153 formats = list(filter(_filter, formats))
1154 return selector_function(formats)
1155 return final_selector
083c9df9 1156
67134eab 1157 stream = io.BytesIO(format_spec.encode('utf-8'))
0130afb7 1158 try:
232541df 1159 tokens = list(_remove_unused_ops(compat_tokenize_tokenize(stream.readline)))
0130afb7
JMF
1160 except tokenize.TokenError:
1161 raise syntax_error('Missing closing/opening brackets or parenthesis', (0, len(format_spec)))
1162
1163 class TokenIterator(object):
1164 def __init__(self, tokens):
1165 self.tokens = tokens
1166 self.counter = 0
1167
1168 def __iter__(self):
1169 return self
1170
1171 def __next__(self):
1172 if self.counter >= len(self.tokens):
1173 raise StopIteration()
1174 value = self.tokens[self.counter]
1175 self.counter += 1
1176 return value
1177
1178 next = __next__
1179
1180 def restore_last_token(self):
1181 self.counter -= 1
1182
1183 parsed_selector = _parse_format_selection(iter(TokenIterator(tokens)))
67134eab 1184 return _build_selector_function(parsed_selector)
a9c58ad9 1185
e5660ee6
JMF
1186 def _calc_headers(self, info_dict):
1187 res = std_headers.copy()
1188
1189 add_headers = info_dict.get('http_headers')
1190 if add_headers:
1191 res.update(add_headers)
1192
1193 cookies = self._calc_cookies(info_dict)
1194 if cookies:
1195 res['Cookie'] = cookies
1196
1197 return res
1198
1199 def _calc_cookies(self, info_dict):
5c2266df 1200 pr = sanitized_Request(info_dict['url'])
e5660ee6 1201 self.cookiejar.add_cookie_header(pr)
662435f7 1202 return pr.get_header('Cookie')
e5660ee6 1203
dd82ffea
JMF
1204 def process_video_result(self, info_dict, download=True):
1205 assert info_dict.get('_type', 'video') == 'video'
1206
bec1fad2
PH
1207 if 'id' not in info_dict:
1208 raise ExtractorError('Missing "id" field in extractor result')
1209 if 'title' not in info_dict:
1210 raise ExtractorError('Missing "title" field in extractor result')
1211
dd82ffea
JMF
1212 if 'playlist' not in info_dict:
1213 # It isn't part of a playlist
1214 info_dict['playlist'] = None
1215 info_dict['playlist_index'] = None
1216
d5519808 1217 thumbnails = info_dict.get('thumbnails')
cfb56d1a
PH
1218 if thumbnails is None:
1219 thumbnail = info_dict.get('thumbnail')
1220 if thumbnail:
a7a14d95 1221 info_dict['thumbnails'] = thumbnails = [{'url': thumbnail}]
d5519808 1222 if thumbnails:
be6d7229 1223 thumbnails.sort(key=lambda t: (
cfb56d1a
PH
1224 t.get('preference'), t.get('width'), t.get('height'),
1225 t.get('id'), t.get('url')))
f6c24009 1226 for i, t in enumerate(thumbnails):
9603e8a7 1227 if t.get('width') and t.get('height'):
d5519808 1228 t['resolution'] = '%dx%d' % (t['width'], t['height'])
f6c24009
PH
1229 if t.get('id') is None:
1230 t['id'] = '%d' % i
d5519808
PH
1231
1232 if thumbnails and 'thumbnail' not in info_dict:
1233 info_dict['thumbnail'] = thumbnails[-1]['url']
1234
c9ae7b95 1235 if 'display_id' not in info_dict and 'id' in info_dict:
0afef30b
PH
1236 info_dict['display_id'] = info_dict['id']
1237
955c4514 1238 if info_dict.get('upload_date') is None and info_dict.get('timestamp') is not None:
a55e36f4
S
1239 # Working around out-of-range timestamp values (e.g. negative ones on Windows,
1240 # see http://bugs.python.org/issue1646728)
1241 try:
1242 upload_date = datetime.datetime.utcfromtimestamp(info_dict['timestamp'])
1243 info_dict['upload_date'] = upload_date.strftime('%Y%m%d')
1244 except (ValueError, OverflowError, OSError):
1245 pass
9d2ecdbc 1246
33d2fc2f
S
1247 # Auto generate title fields corresponding to the *_number fields when missing
1248 # in order to always have clean titles. This is very common for TV series.
1249 for field in ('chapter', 'season', 'episode'):
1250 if info_dict.get('%s_number' % field) is not None and not info_dict.get(field):
1251 info_dict[field] = '%s %d' % (field.capitalize(), info_dict['%s_number' % field])
1252
4bba3716
S
1253 subtitles = info_dict.get('subtitles')
1254 if subtitles:
1255 for _, subtitle in subtitles.items():
1256 for subtitle_format in subtitle:
1257 if 'ext' not in subtitle_format:
1258 subtitle_format['ext'] = determine_ext(subtitle_format['url']).lower()
1259
a504ced0 1260 if self.params.get('listsubtitles', False):
360e1ca5
JMF
1261 if 'automatic_captions' in info_dict:
1262 self.list_subtitles(info_dict['id'], info_dict.get('automatic_captions'), 'automatic captions')
4bba3716 1263 self.list_subtitles(info_dict['id'], subtitles, 'subtitles')
a504ced0 1264 return
360e1ca5 1265 info_dict['requested_subtitles'] = self.process_subtitles(
4bba3716 1266 info_dict['id'], subtitles,
360e1ca5 1267 info_dict.get('automatic_captions'))
a504ced0 1268
dd82ffea
JMF
1269 # We now pick which formats have to be downloaded
1270 if info_dict.get('formats') is None:
1271 # There's only one format available
1272 formats = [info_dict]
1273 else:
1274 formats = info_dict['formats']
1275
db95dc13
PH
1276 if not formats:
1277 raise ExtractorError('No video formats found!')
1278
181c7053
S
1279 formats_dict = {}
1280
dd82ffea 1281 # We check that all the formats have the format and format_id fields
db95dc13 1282 for i, format in enumerate(formats):
bec1fad2
PH
1283 if 'url' not in format:
1284 raise ExtractorError('Missing "url" key in result (index %d)' % i)
1285
dd82ffea 1286 if format.get('format_id') is None:
8016c922 1287 format['format_id'] = compat_str(i)
181c7053
S
1288 format_id = format['format_id']
1289 if format_id not in formats_dict:
1290 formats_dict[format_id] = []
1291 formats_dict[format_id].append(format)
1292
1293 # Make sure all formats have unique format_id
1294 for format_id, ambiguous_formats in formats_dict.items():
1295 if len(ambiguous_formats) > 1:
1296 for i, format in enumerate(ambiguous_formats):
1297 format['format_id'] = '%s-%d' % (format_id, i)
1298
1299 for i, format in enumerate(formats):
8c51aa65 1300 if format.get('format') is None:
6febd1c1 1301 format['format'] = '{id} - {res}{note}'.format(
8c51aa65
JMF
1302 id=format['format_id'],
1303 res=self.format_resolution(format),
6febd1c1 1304 note=' ({0})'.format(format['format_note']) if format.get('format_note') is not None else '',
8c51aa65 1305 )
c1002e96
PH
1306 # Automatically determine file extension if missing
1307 if 'ext' not in format:
cce929ea 1308 format['ext'] = determine_ext(format['url']).lower()
e5660ee6
JMF
1309 # Add HTTP headers, so that external programs can use them from the
1310 # json output
1311 full_format_info = info_dict.copy()
1312 full_format_info.update(format)
1313 format['http_headers'] = self._calc_headers(full_format_info)
dd82ffea 1314
4bcc7bd1 1315 # TODO Central sorting goes here
99e206d5 1316
f89197d7 1317 if formats[0] is not info_dict:
b3d9ef88
JMF
1318 # only set the 'formats' fields if the original info_dict list them
1319 # otherwise we end up with a circular reference, the first (and unique)
f89197d7 1320 # element in the 'formats' field in info_dict is info_dict itself,
dfb1b146 1321 # which can't be exported to json
b3d9ef88 1322 info_dict['formats'] = formats
cfb56d1a 1323 if self.params.get('listformats'):
bfaae0a7 1324 self.list_formats(info_dict)
1325 return
cfb56d1a
PH
1326 if self.params.get('list_thumbnails'):
1327 self.list_thumbnails(info_dict)
1328 return
bfaae0a7 1329
de3ef3ed 1330 req_format = self.params.get('format')
a9c58ad9 1331 if req_format is None:
feccf29c 1332 req_format_list = []
3749e36e 1333 if (self.params.get('outtmpl', DEFAULT_OUTTMPL) != '-' and
8250c32f
JMF
1334 info_dict['extractor'] in ['youtube', 'ted'] and
1335 not info_dict.get('is_live')):
7fcb605b 1336 merger = FFmpegMergerPP(self)
97fcf1bb 1337 if merger.available and merger.can_merge():
7fcb605b 1338 req_format_list.append('bestvideo+bestaudio')
feccf29c
S
1339 req_format_list.append('best')
1340 req_format = '/'.join(req_format_list)
5acfa126
JMF
1341 format_selector = self.build_format_selector(req_format)
1342 formats_to_download = list(format_selector(formats))
dd82ffea 1343 if not formats_to_download:
6febd1c1 1344 raise ExtractorError('requested format not available',
78a3a9f8 1345 expected=True)
dd82ffea
JMF
1346
1347 if download:
1348 if len(formats_to_download) > 1:
6febd1c1 1349 self.to_screen('[info] %s: downloading video in %s formats' % (info_dict['id'], len(formats_to_download)))
dd82ffea
JMF
1350 for format in formats_to_download:
1351 new_info = dict(info_dict)
1352 new_info.update(format)
1353 self.process_info(new_info)
1354 # We update the info dict with the best quality format (backwards compatibility)
1355 info_dict.update(formats_to_download[-1])
1356 return info_dict
1357
98c70d6f 1358 def process_subtitles(self, video_id, normal_subtitles, automatic_captions):
a504ced0 1359 """Select the requested subtitles and their format"""
98c70d6f
JMF
1360 available_subs = {}
1361 if normal_subtitles and self.params.get('writesubtitles'):
1362 available_subs.update(normal_subtitles)
1363 if automatic_captions and self.params.get('writeautomaticsub'):
1364 for lang, cap_info in automatic_captions.items():
360e1ca5
JMF
1365 if lang not in available_subs:
1366 available_subs[lang] = cap_info
1367
4d171848
JMF
1368 if (not self.params.get('writesubtitles') and not
1369 self.params.get('writeautomaticsub') or not
1370 available_subs):
1371 return None
a504ced0
JMF
1372
1373 if self.params.get('allsubtitles', False):
1374 requested_langs = available_subs.keys()
1375 else:
1376 if self.params.get('subtitleslangs', False):
1377 requested_langs = self.params.get('subtitleslangs')
1378 elif 'en' in available_subs:
1379 requested_langs = ['en']
1380 else:
1381 requested_langs = [list(available_subs.keys())[0]]
1382
1383 formats_query = self.params.get('subtitlesformat', 'best')
1384 formats_preference = formats_query.split('/') if formats_query else []
1385 subs = {}
1386 for lang in requested_langs:
1387 formats = available_subs.get(lang)
1388 if formats is None:
1389 self.report_warning('%s subtitles not available for %s' % (lang, video_id))
1390 continue
a504ced0
JMF
1391 for ext in formats_preference:
1392 if ext == 'best':
1393 f = formats[-1]
1394 break
1395 matches = list(filter(lambda f: f['ext'] == ext, formats))
1396 if matches:
1397 f = matches[-1]
1398 break
1399 else:
1400 f = formats[-1]
1401 self.report_warning(
1402 'No subtitle format found matching "%s" for language %s, '
1403 'using %s' % (formats_query, lang, f['ext']))
1404 subs[lang] = f
1405 return subs
1406
8222d8de
JMF
1407 def process_info(self, info_dict):
1408 """Process a single resolved IE result."""
1409
1410 assert info_dict.get('_type', 'video') == 'video'
fd288278
PH
1411
1412 max_downloads = self.params.get('max_downloads')
1413 if max_downloads is not None:
1414 if self._num_downloads >= int(max_downloads):
1415 raise MaxDownloadsReached()
8222d8de
JMF
1416
1417 info_dict['fulltitle'] = info_dict['title']
1418 if len(info_dict['title']) > 200:
6febd1c1 1419 info_dict['title'] = info_dict['title'][:197] + '...'
8222d8de 1420
11b85ce6 1421 if 'format' not in info_dict:
8222d8de
JMF
1422 info_dict['format'] = info_dict['ext']
1423
442c37b7 1424 reason = self._match_entry(info_dict, incomplete=False)
8222d8de 1425 if reason is not None:
6febd1c1 1426 self.to_screen('[download] ' + reason)
8222d8de
JMF
1427 return
1428
fd288278 1429 self._num_downloads += 1
8222d8de 1430
e72c7e41 1431 info_dict['_filename'] = filename = self.prepare_filename(info_dict)
8222d8de
JMF
1432
1433 # Forced printings
1434 if self.params.get('forcetitle', False):
0783b09b 1435 self.to_stdout(info_dict['fulltitle'])
8222d8de 1436 if self.params.get('forceid', False):
0783b09b 1437 self.to_stdout(info_dict['id'])
8222d8de 1438 if self.params.get('forceurl', False):
16ae61f6 1439 if info_dict.get('requested_formats') is not None:
1440 for f in info_dict['requested_formats']:
1441 self.to_stdout(f['url'] + f.get('play_path', ''))
1442 else:
1443 # For RTMP URLs, also include the playpath
1444 self.to_stdout(info_dict['url'] + info_dict.get('play_path', ''))
216d71d0 1445 if self.params.get('forcethumbnail', False) and info_dict.get('thumbnail') is not None:
0783b09b 1446 self.to_stdout(info_dict['thumbnail'])
216d71d0 1447 if self.params.get('forcedescription', False) and info_dict.get('description') is not None:
0783b09b 1448 self.to_stdout(info_dict['description'])
8222d8de 1449 if self.params.get('forcefilename', False) and filename is not None:
0783b09b 1450 self.to_stdout(filename)
525ef922
PH
1451 if self.params.get('forceduration', False) and info_dict.get('duration') is not None:
1452 self.to_stdout(formatSeconds(info_dict['duration']))
8222d8de 1453 if self.params.get('forceformat', False):
0783b09b 1454 self.to_stdout(info_dict['format'])
9d153818 1455 if self.params.get('forcejson', False):
0783b09b 1456 self.to_stdout(json.dumps(info_dict))
8222d8de
JMF
1457
1458 # Do nothing else if in simulate mode
1459 if self.params.get('simulate', False):
1460 return
1461
1462 if filename is None:
1463 return
1464
1465 try:
e5a11a22 1466 dn = os.path.dirname(sanitize_path(encodeFilename(filename)))
d26e981d 1467 if dn and not os.path.exists(dn):
8222d8de
JMF
1468 os.makedirs(dn)
1469 except (OSError, IOError) as err:
9b9c5355 1470 self.report_error('unable to create directory ' + error_to_compat_str(err))
8222d8de
JMF
1471 return
1472
1473 if self.params.get('writedescription', False):
2699da80 1474 descfn = replace_extension(filename, 'description', info_dict.get('ext'))
7b6fefc9 1475 if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(descfn)):
6febd1c1 1476 self.to_screen('[info] Video description is already present')
f00fd51d
JMF
1477 elif info_dict.get('description') is None:
1478 self.report_warning('There\'s no description to write.')
7b6fefc9
PH
1479 else:
1480 try:
6febd1c1 1481 self.to_screen('[info] Writing video description to: ' + descfn)
7b6fefc9
PH
1482 with io.open(encodeFilename(descfn), 'w', encoding='utf-8') as descfile:
1483 descfile.write(info_dict['description'])
7b6fefc9 1484 except (OSError, IOError):
6febd1c1 1485 self.report_error('Cannot write description file ' + descfn)
7b6fefc9 1486 return
8222d8de 1487
1fb07d10 1488 if self.params.get('writeannotations', False):
98727e12 1489 annofn = replace_extension(filename, 'annotations.xml', info_dict.get('ext'))
7b6fefc9 1490 if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(annofn)):
6febd1c1 1491 self.to_screen('[info] Video annotations are already present')
7b6fefc9
PH
1492 else:
1493 try:
6febd1c1 1494 self.to_screen('[info] Writing video annotations to: ' + annofn)
7b6fefc9
PH
1495 with io.open(encodeFilename(annofn), 'w', encoding='utf-8') as annofile:
1496 annofile.write(info_dict['annotations'])
1497 except (KeyError, TypeError):
6febd1c1 1498 self.report_warning('There are no annotations to write.')
7b6fefc9 1499 except (OSError, IOError):
6febd1c1 1500 self.report_error('Cannot write annotations file: ' + annofn)
7b6fefc9 1501 return
1fb07d10 1502
c4a91be7 1503 subtitles_are_requested = any([self.params.get('writesubtitles', False),
0b7f3118 1504 self.params.get('writeautomaticsub')])
c4a91be7 1505
c84dd8a9 1506 if subtitles_are_requested and info_dict.get('requested_subtitles'):
8222d8de
JMF
1507 # subtitles download errors are already managed as troubles in relevant IE
1508 # that way it will silently go on when used with unsupporting IE
c84dd8a9 1509 subtitles = info_dict['requested_subtitles']
0f2c0d33 1510 ie = self.get_info_extractor(info_dict['extractor_key'])
a504ced0
JMF
1511 for sub_lang, sub_info in subtitles.items():
1512 sub_format = sub_info['ext']
1513 if sub_info.get('data') is not None:
1514 sub_data = sub_info['data']
1515 else:
1516 try:
0f2c0d33
JMF
1517 sub_data = ie._download_webpage(
1518 sub_info['url'], info_dict['id'], note=False)
1519 except ExtractorError as err:
a504ced0 1520 self.report_warning('Unable to download subtitle for "%s": %s' %
9b9c5355 1521 (sub_lang, error_to_compat_str(err.cause)))
a504ced0 1522 continue
8222d8de 1523 try:
d4051a8e 1524 sub_filename = subtitles_filename(filename, sub_lang, sub_format)
7b6fefc9 1525 if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(sub_filename)):
6febd1c1 1526 self.to_screen('[info] Video subtitle %s.%s is already_present' % (sub_lang, sub_format))
7b6fefc9 1527 else:
6febd1c1 1528 self.to_screen('[info] Writing video subtitles to: ' + sub_filename)
7b6fefc9 1529 with io.open(encodeFilename(sub_filename), 'w', encoding='utf-8') as subfile:
a504ced0 1530 subfile.write(sub_data)
8222d8de 1531 except (OSError, IOError):
e4db1951 1532 self.report_error('Cannot write subtitles file ' + sub_filename)
8222d8de
JMF
1533 return
1534
8222d8de 1535 if self.params.get('writeinfojson', False):
b29e0000 1536 infofn = replace_extension(filename, 'info.json', info_dict.get('ext'))
7b6fefc9 1537 if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(infofn)):
6febd1c1 1538 self.to_screen('[info] Video description metadata is already present')
7b6fefc9 1539 else:
6febd1c1 1540 self.to_screen('[info] Writing video description metadata as JSON to: ' + infofn)
7b6fefc9 1541 try:
cb202fd2 1542 write_json_file(self.filter_requested_info(info_dict), infofn)
7b6fefc9 1543 except (OSError, IOError):
6febd1c1 1544 self.report_error('Cannot write metadata to JSON file ' + infofn)
7b6fefc9 1545 return
8222d8de 1546
ec82d85a 1547 self._write_thumbnails(info_dict, filename)
8222d8de
JMF
1548
1549 if not self.params.get('skip_download', False):
4340deca
P
1550 try:
1551 def dl(name, info):
a055469f 1552 fd = get_suitable_downloader(info, self.params)(self, self.params)
4340deca
P
1553 for ph in self._progress_hooks:
1554 fd.add_progress_hook(ph)
1555 if self.params.get('verbose'):
1556 self.to_stdout('[debug] Invoking downloader on %r' % info.get('url'))
1557 return fd.download(name, info)
ee69b99a 1558
4340deca
P
1559 if info_dict.get('requested_formats') is not None:
1560 downloaded = []
1561 success = True
d47aeb22 1562 merger = FFmpegMergerPP(self)
f740fae2 1563 if not merger.available:
4340deca
P
1564 postprocessors = []
1565 self.report_warning('You have requested multiple '
1566 'formats but ffmpeg or avconv are not installed.'
4a5a898a 1567 ' The formats won\'t be merged.')
6350728b 1568 else:
4340deca 1569 postprocessors = [merger]
81cd954a
S
1570
1571 def compatible_formats(formats):
1572 video, audio = formats
1573 # Check extension
1574 video_ext, audio_ext = audio.get('ext'), video.get('ext')
1575 if video_ext and audio_ext:
1576 COMPATIBLE_EXTS = (
6728187a 1577 ('mp3', 'mp4', 'm4a', 'm4p', 'm4b', 'm4r', 'm4v'),
81cd954a
S
1578 ('webm')
1579 )
1580 for exts in COMPATIBLE_EXTS:
1581 if video_ext in exts and audio_ext in exts:
1582 return True
1583 # TODO: Check acodec/vcodec
1584 return False
1585
38c6902b
S
1586 filename_real_ext = os.path.splitext(filename)[1][1:]
1587 filename_wo_ext = (
1588 os.path.splitext(filename)[0]
1589 if filename_real_ext == info_dict['ext']
1590 else filename)
81cd954a 1591 requested_formats = info_dict['requested_formats']
c0dea0a7 1592 if self.params.get('merge_output_format') is None and not compatible_formats(requested_formats):
38c6902b 1593 info_dict['ext'] = 'mkv'
4a5a898a
S
1594 self.report_warning(
1595 'Requested formats are incompatible for merge and will be merged into mkv.')
38c6902b
S
1596 # Ensure filename always has a correct extension for successful merge
1597 filename = '%s.%s' % (filename_wo_ext, info_dict['ext'])
5b5fbc08
JMF
1598 if os.path.exists(encodeFilename(filename)):
1599 self.to_screen(
1600 '[download] %s has already been downloaded and '
1601 'merged' % filename)
1602 else:
81cd954a 1603 for f in requested_formats:
5b5fbc08
JMF
1604 new_info = dict(info_dict)
1605 new_info.update(f)
1606 fname = self.prepare_filename(new_info)
666a9a2b 1607 fname = prepend_extension(fname, 'f%s' % f['format_id'], new_info['ext'])
5b5fbc08
JMF
1608 downloaded.append(fname)
1609 partial_success = dl(fname, new_info)
1610 success = success and partial_success
1611 info_dict['__postprocessors'] = postprocessors
1612 info_dict['__files_to_merge'] = downloaded
4340deca
P
1613 else:
1614 # Just a single file
1615 success = dl(filename, info_dict)
1616 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
1617 self.report_error('unable to download video data: %s' % str(err))
1618 return
1619 except (OSError, IOError) as err:
1620 raise UnavailableVideoError(err)
1621 except (ContentTooShortError, ) as err:
1622 self.report_error('content too short (expected %s bytes and served %s)' % (err.expected, err.downloaded))
1623 return
8222d8de
JMF
1624
1625 if success:
6271f1ca 1626 # Fixup content
62cd676c
PH
1627 fixup_policy = self.params.get('fixup')
1628 if fixup_policy is None:
1629 fixup_policy = 'detect_or_warn'
1630
6271f1ca
PH
1631 stretched_ratio = info_dict.get('stretched_ratio')
1632 if stretched_ratio is not None and stretched_ratio != 1:
6271f1ca
PH
1633 if fixup_policy == 'warn':
1634 self.report_warning('%s: Non-uniform pixel ratio (%s)' % (
1635 info_dict['id'], stretched_ratio))
1636 elif fixup_policy == 'detect_or_warn':
1637 stretched_pp = FFmpegFixupStretchedPP(self)
1638 if stretched_pp.available:
1639 info_dict.setdefault('__postprocessors', [])
1640 info_dict['__postprocessors'].append(stretched_pp)
1641 else:
1642 self.report_warning(
1643 '%s: Non-uniform pixel ratio (%s). Install ffmpeg or avconv to fix this automatically.' % (
1644 info_dict['id'], stretched_ratio))
1645 else:
62cd676c
PH
1646 assert fixup_policy in ('ignore', 'never')
1647
1648 if info_dict.get('requested_formats') is None and info_dict.get('container') == 'm4a_dash':
1649 if fixup_policy == 'warn':
1650 self.report_warning('%s: writing DASH m4a. Only some players support this container.' % (
1651 info_dict['id']))
1652 elif fixup_policy == 'detect_or_warn':
1653 fixup_pp = FFmpegFixupM4aPP(self)
1654 if fixup_pp.available:
1655 info_dict.setdefault('__postprocessors', [])
1656 info_dict['__postprocessors'].append(fixup_pp)
1657 else:
1658 self.report_warning(
1659 '%s: writing DASH m4a. Only some players support this container. Install ffmpeg or avconv to fix this automatically.' % (
1660 info_dict['id']))
1661 else:
1662 assert fixup_policy in ('ignore', 'never')
6271f1ca 1663
8222d8de
JMF
1664 try:
1665 self.post_process(filename, info_dict)
1666 except (PostProcessingError) as err:
6febd1c1 1667 self.report_error('postprocessing: %s' % str(err))
8222d8de 1668 return
cd58dc3e 1669 self.record_download_archive(info_dict)
8222d8de
JMF
1670
1671 def download(self, url_list):
1672 """Download a given list of URLs."""
acd69589 1673 outtmpl = self.params.get('outtmpl', DEFAULT_OUTTMPL)
0c75c3fa 1674 if (len(url_list) > 1 and
8fb3ac36
PH
1675 '%' not in outtmpl and
1676 self.params.get('max_downloads') != 1):
acd69589 1677 raise SameFileError(outtmpl)
8222d8de
JMF
1678
1679 for url in url_list:
1680 try:
5f6a1245 1681 # It also downloads the videos
61aa5ba3
S
1682 res = self.extract_info(
1683 url, force_generic_extractor=self.params.get('force_generic_extractor', False))
8222d8de 1684 except UnavailableVideoError:
6febd1c1 1685 self.report_error('unable to download video')
8222d8de 1686 except MaxDownloadsReached:
6febd1c1 1687 self.to_screen('[info] Maximum number of downloaded files reached.')
8222d8de 1688 raise
63e0be34
PH
1689 else:
1690 if self.params.get('dump_single_json', False):
1691 self.to_stdout(json.dumps(res))
8222d8de
JMF
1692
1693 return self._download_retcode
1694
1dcc4c0c 1695 def download_with_info_file(self, info_filename):
31bd3925
JMF
1696 with contextlib.closing(fileinput.FileInput(
1697 [info_filename], mode='r',
1698 openhook=fileinput.hook_encoded('utf-8'))) as f:
1699 # FileInput doesn't have a read method, we can't call json.load
cb202fd2 1700 info = self.filter_requested_info(json.loads('\n'.join(f)))
d4943898
JMF
1701 try:
1702 self.process_ie_result(info, download=True)
1703 except DownloadError:
1704 webpage_url = info.get('webpage_url')
1705 if webpage_url is not None:
6febd1c1 1706 self.report_warning('The info failed to download, trying with "%s"' % webpage_url)
d4943898
JMF
1707 return self.download([webpage_url])
1708 else:
1709 raise
1710 return self._download_retcode
1dcc4c0c 1711
cb202fd2
S
1712 @staticmethod
1713 def filter_requested_info(info_dict):
1714 return dict(
1715 (k, v) for k, v in info_dict.items()
1716 if k not in ['requested_formats', 'requested_subtitles'])
1717
8222d8de
JMF
1718 def post_process(self, filename, ie_info):
1719 """Run all the postprocessors on the given file."""
1720 info = dict(ie_info)
1721 info['filepath'] = filename
6350728b
JMF
1722 pps_chain = []
1723 if ie_info.get('__postprocessors') is not None:
1724 pps_chain.extend(ie_info['__postprocessors'])
1725 pps_chain.extend(self._pps)
1726 for pp in pps_chain:
71646e46 1727 files_to_delete = []
8222d8de 1728 try:
592e97e8 1729 files_to_delete, info = pp.run(info)
8222d8de 1730 except PostProcessingError as e:
bbcbf4d4 1731 self.report_error(e.msg)
592e97e8
JMF
1732 if files_to_delete and not self.params.get('keepvideo', False):
1733 for old_filename in files_to_delete:
f3ff1a36 1734 self.to_screen('Deleting original file %s (pass -k to keep)' % old_filename)
592e97e8
JMF
1735 try:
1736 os.remove(encodeFilename(old_filename))
1737 except (IOError, OSError):
1738 self.report_warning('Unable to remove downloaded original file')
c1c9a79c 1739
5db07df6
PH
1740 def _make_archive_id(self, info_dict):
1741 # Future-proof against any change in case
1742 # and backwards compatibility with prior versions
d31209a1 1743 extractor = info_dict.get('extractor_key')
7012b23c
PH
1744 if extractor is None:
1745 if 'id' in info_dict:
1746 extractor = info_dict.get('ie_key') # key in a playlist
1747 if extractor is None:
5db07df6 1748 return None # Incomplete video information
6febd1c1 1749 return extractor.lower() + ' ' + info_dict['id']
5db07df6
PH
1750
1751 def in_download_archive(self, info_dict):
1752 fn = self.params.get('download_archive')
1753 if fn is None:
1754 return False
1755
1756 vid_id = self._make_archive_id(info_dict)
1757 if vid_id is None:
7012b23c 1758 return False # Incomplete video information
5db07df6 1759
c1c9a79c
PH
1760 try:
1761 with locked_file(fn, 'r', encoding='utf-8') as archive_file:
1762 for line in archive_file:
1763 if line.strip() == vid_id:
1764 return True
1765 except IOError as ioe:
1766 if ioe.errno != errno.ENOENT:
1767 raise
1768 return False
1769
1770 def record_download_archive(self, info_dict):
1771 fn = self.params.get('download_archive')
1772 if fn is None:
1773 return
5db07df6
PH
1774 vid_id = self._make_archive_id(info_dict)
1775 assert vid_id
c1c9a79c 1776 with locked_file(fn, 'a', encoding='utf-8') as archive_file:
6febd1c1 1777 archive_file.write(vid_id + '\n')
dd82ffea 1778
8c51aa65 1779 @staticmethod
8abeeb94 1780 def format_resolution(format, default='unknown'):
fb04e403
PH
1781 if format.get('vcodec') == 'none':
1782 return 'audio only'
f49d89ee
PH
1783 if format.get('resolution') is not None:
1784 return format['resolution']
8c51aa65
JMF
1785 if format.get('height') is not None:
1786 if format.get('width') is not None:
6febd1c1 1787 res = '%sx%s' % (format['width'], format['height'])
8c51aa65 1788 else:
6febd1c1 1789 res = '%sp' % format['height']
f49d89ee 1790 elif format.get('width') is not None:
6febd1c1 1791 res = '?x%d' % format['width']
8c51aa65 1792 else:
8abeeb94 1793 res = default
8c51aa65
JMF
1794 return res
1795
c57f7757
PH
1796 def _format_note(self, fdict):
1797 res = ''
1798 if fdict.get('ext') in ['f4f', 'f4m']:
1799 res += '(unsupported) '
32f90364
PH
1800 if fdict.get('language'):
1801 if res:
1802 res += ' '
1803 res += '[%s]' % fdict['language']
c57f7757
PH
1804 if fdict.get('format_note') is not None:
1805 res += fdict['format_note'] + ' '
1806 if fdict.get('tbr') is not None:
1807 res += '%4dk ' % fdict['tbr']
1808 if fdict.get('container') is not None:
1809 if res:
1810 res += ', '
1811 res += '%s container' % fdict['container']
1812 if (fdict.get('vcodec') is not None and
1813 fdict.get('vcodec') != 'none'):
1814 if res:
1815 res += ', '
1816 res += fdict['vcodec']
91c7271a 1817 if fdict.get('vbr') is not None:
c57f7757
PH
1818 res += '@'
1819 elif fdict.get('vbr') is not None and fdict.get('abr') is not None:
1820 res += 'video@'
1821 if fdict.get('vbr') is not None:
1822 res += '%4dk' % fdict['vbr']
fbb21cf5
PH
1823 if fdict.get('fps') is not None:
1824 res += ', %sfps' % fdict['fps']
c57f7757
PH
1825 if fdict.get('acodec') is not None:
1826 if res:
1827 res += ', '
1828 if fdict['acodec'] == 'none':
1829 res += 'video only'
1830 else:
1831 res += '%-5s' % fdict['acodec']
1832 elif fdict.get('abr') is not None:
1833 if res:
1834 res += ', '
1835 res += 'audio'
1836 if fdict.get('abr') is not None:
1837 res += '@%3dk' % fdict['abr']
1838 if fdict.get('asr') is not None:
1839 res += ' (%5dHz)' % fdict['asr']
1840 if fdict.get('filesize') is not None:
1841 if res:
1842 res += ', '
1843 res += format_bytes(fdict['filesize'])
9732d77e
PH
1844 elif fdict.get('filesize_approx') is not None:
1845 if res:
1846 res += ', '
1847 res += '~' + format_bytes(fdict['filesize_approx'])
c57f7757 1848 return res
91c7271a 1849
c57f7757 1850 def list_formats(self, info_dict):
94badb25 1851 formats = info_dict.get('formats', [info_dict])
b81a359e
PH
1852 table = [
1853 [f['format_id'], f['ext'], self.format_resolution(f), self._format_note(f)]
1854 for f in formats
e65566a9 1855 if f.get('preference') is None or f['preference'] >= -1000]
94badb25 1856 if len(formats) > 1:
b81a359e 1857 table[-1][-1] += (' ' if table[-1][-1] else '') + '(best)'
57dd9a8f 1858
b81a359e 1859 header_line = ['format code', 'extension', 'resolution', 'note']
cfb56d1a 1860 self.to_screen(
b81a359e
PH
1861 '[info] Available formats for %s:\n%s' %
1862 (info_dict['id'], render_table(header_line, table)))
cfb56d1a
PH
1863
1864 def list_thumbnails(self, info_dict):
1865 thumbnails = info_dict.get('thumbnails')
1866 if not thumbnails:
1867 tn_url = info_dict.get('thumbnail')
1868 if tn_url:
1869 thumbnails = [{'id': '0', 'url': tn_url}]
1870 else:
1871 self.to_screen(
1872 '[info] No thumbnails present for %s' % info_dict['id'])
1873 return
1874
1875 self.to_screen(
1876 '[info] Thumbnails for %s:' % info_dict['id'])
1877 self.to_screen(render_table(
1878 ['ID', 'width', 'height', 'URL'],
1879 [[t['id'], t.get('width', 'unknown'), t.get('height', 'unknown'), t['url']] for t in thumbnails]))
dca08720 1880
360e1ca5 1881 def list_subtitles(self, video_id, subtitles, name='subtitles'):
a504ced0 1882 if not subtitles:
360e1ca5 1883 self.to_screen('%s has no %s' % (video_id, name))
a504ced0 1884 return
a504ced0 1885 self.to_screen(
edab9dbf
JMF
1886 'Available %s for %s:' % (name, video_id))
1887 self.to_screen(render_table(
1888 ['Language', 'formats'],
1889 [[lang, ', '.join(f['ext'] for f in reversed(formats))]
1890 for lang, formats in subtitles.items()]))
a504ced0 1891
dca08720
PH
1892 def urlopen(self, req):
1893 """ Start an HTTP download """
82d8a8b6 1894 if isinstance(req, compat_basestring):
67dda517 1895 req = sanitized_Request(req)
19a41fc6 1896 return self._opener.open(req, timeout=self._socket_timeout)
dca08720
PH
1897
1898 def print_debug_header(self):
1899 if not self.params.get('verbose'):
1900 return
62fec3b2 1901
4192b51c
PH
1902 if type('') is not compat_str:
1903 # Python 2.6 on SLES11 SP1 (https://github.com/rg3/youtube-dl/issues/3326)
1904 self.report_warning(
1905 'Your Python is broken! Update to a newer and supported version')
1906
c6afed48
PH
1907 stdout_encoding = getattr(
1908 sys.stdout, 'encoding', 'missing (%s)' % type(sys.stdout).__name__)
b0472057 1909 encoding_str = (
734f90bb
PH
1910 '[debug] Encodings: locale %s, fs %s, out %s, pref %s\n' % (
1911 locale.getpreferredencoding(),
1912 sys.getfilesystemencoding(),
c6afed48 1913 stdout_encoding,
b0472057 1914 self.get_encoding()))
4192b51c 1915 write_string(encoding_str, encoding=None)
734f90bb
PH
1916
1917 self._write_string('[debug] youtube-dl version ' + __version__ + '\n')
dca08720
PH
1918 try:
1919 sp = subprocess.Popen(
1920 ['git', 'rev-parse', '--short', 'HEAD'],
1921 stdout=subprocess.PIPE, stderr=subprocess.PIPE,
1922 cwd=os.path.dirname(os.path.abspath(__file__)))
1923 out, err = sp.communicate()
1924 out = out.decode().strip()
1925 if re.match('[0-9a-f]+', out):
734f90bb 1926 self._write_string('[debug] Git HEAD: ' + out + '\n')
70a1165b 1927 except Exception:
dca08720
PH
1928 try:
1929 sys.exc_clear()
70a1165b 1930 except Exception:
dca08720 1931 pass
d28b5171
PH
1932 self._write_string('[debug] Python version %s - %s\n' % (
1933 platform.python_version(), platform_name()))
1934
73fac4e9 1935 exe_versions = FFmpegPostProcessor.get_versions(self)
4c83c967 1936 exe_versions['rtmpdump'] = rtmpdump_version()
d28b5171
PH
1937 exe_str = ', '.join(
1938 '%s %s' % (exe, v)
1939 for exe, v in sorted(exe_versions.items())
1940 if v
1941 )
1942 if not exe_str:
1943 exe_str = 'none'
1944 self._write_string('[debug] exe versions: %s\n' % exe_str)
dca08720
PH
1945
1946 proxy_map = {}
1947 for handler in self._opener.handlers:
1948 if hasattr(handler, 'proxies'):
1949 proxy_map.update(handler.proxies)
734f90bb 1950 self._write_string('[debug] Proxy map: ' + compat_str(proxy_map) + '\n')
dca08720 1951
58b1f00d
PH
1952 if self.params.get('call_home', False):
1953 ipaddr = self.urlopen('https://yt-dl.org/ip').read().decode('utf-8')
1954 self._write_string('[debug] Public IP address: %s\n' % ipaddr)
1955 latest_version = self.urlopen(
1956 'https://yt-dl.org/latest/version').read().decode('utf-8')
1957 if version_tuple(latest_version) > version_tuple(__version__):
1958 self.report_warning(
1959 'You are using an outdated version (newest version: %s)! '
1960 'See https://yt-dl.org/update if you need help updating.' %
1961 latest_version)
1962
e344693b 1963 def _setup_opener(self):
6ad14cab 1964 timeout_val = self.params.get('socket_timeout')
19a41fc6 1965 self._socket_timeout = 600 if timeout_val is None else float(timeout_val)
6ad14cab 1966
dca08720
PH
1967 opts_cookiefile = self.params.get('cookiefile')
1968 opts_proxy = self.params.get('proxy')
1969
1970 if opts_cookiefile is None:
1971 self.cookiejar = compat_cookiejar.CookieJar()
1972 else:
1973 self.cookiejar = compat_cookiejar.MozillaCookieJar(
1974 opts_cookiefile)
1975 if os.access(opts_cookiefile, os.R_OK):
1976 self.cookiejar.load()
1977
6a3f4c3f 1978 cookie_processor = YoutubeDLCookieProcessor(self.cookiejar)
dca08720
PH
1979 if opts_proxy is not None:
1980 if opts_proxy == '':
1981 proxies = {}
1982 else:
1983 proxies = {'http': opts_proxy, 'https': opts_proxy}
1984 else:
1985 proxies = compat_urllib_request.getproxies()
1986 # Set HTTPS proxy to HTTP one if given (https://github.com/rg3/youtube-dl/issues/805)
1987 if 'http' in proxies and 'https' not in proxies:
1988 proxies['https'] = proxies['http']
91410c9b 1989 proxy_handler = PerRequestProxyHandler(proxies)
a0ddb8a2
PH
1990
1991 debuglevel = 1 if self.params.get('debug_printtraffic') else 0
be4a824d
PH
1992 https_handler = make_HTTPS_handler(self.params, debuglevel=debuglevel)
1993 ydlh = YoutubeDLHandler(self.params, debuglevel=debuglevel)
8b172c2e 1994 data_handler = compat_urllib_request_DataHandler()
6240b0a2
JMF
1995
1996 # When passing our own FileHandler instance, build_opener won't add the
1997 # default FileHandler and allows us to disable the file protocol, which
1998 # can be used for malicious purposes (see
e37afbe0 1999 # https://github.com/rg3/youtube-dl/issues/8227)
6240b0a2
JMF
2000 file_handler = compat_urllib_request.FileHandler()
2001
2002 def file_open(*args, **kwargs):
30e2f2d7 2003 raise compat_urllib_error.URLError('file:// scheme is explicitly disabled in youtube-dl for security reasons')
6240b0a2
JMF
2004 file_handler.file_open = file_open
2005
2006 opener = compat_urllib_request.build_opener(
2007 proxy_handler, https_handler, cookie_processor, ydlh, data_handler, file_handler)
2461f79d 2008
dca08720
PH
2009 # Delete the default user-agent header, which would otherwise apply in
2010 # cases where our custom HTTP handler doesn't come into play
2011 # (See https://github.com/rg3/youtube-dl/issues/1309 for details)
2012 opener.addheaders = []
2013 self._opener = opener
62fec3b2
PH
2014
2015 def encode(self, s):
2016 if isinstance(s, bytes):
2017 return s # Already encoded
2018
2019 try:
2020 return s.encode(self.get_encoding())
2021 except UnicodeEncodeError as err:
2022 err.reason = err.reason + '. Check your system encoding configuration or use the --encoding option.'
2023 raise
2024
2025 def get_encoding(self):
2026 encoding = self.params.get('encoding')
2027 if encoding is None:
2028 encoding = preferredencoding()
2029 return encoding
ec82d85a
PH
2030
2031 def _write_thumbnails(self, info_dict, filename):
2032 if self.params.get('writethumbnail', False):
2033 thumbnails = info_dict.get('thumbnails')
2034 if thumbnails:
2035 thumbnails = [thumbnails[-1]]
2036 elif self.params.get('write_all_thumbnails', False):
2037 thumbnails = info_dict.get('thumbnails')
2038 else:
2039 return
2040
2041 if not thumbnails:
2042 # No thumbnails present, so return immediately
2043 return
2044
2045 for t in thumbnails:
2046 thumb_ext = determine_ext(t['url'], 'jpg')
2047 suffix = '_%s' % t['id'] if len(thumbnails) > 1 else ''
2048 thumb_display_id = '%s ' % t['id'] if len(thumbnails) > 1 else ''
82245a6d 2049 t['filename'] = thumb_filename = os.path.splitext(filename)[0] + suffix + '.' + thumb_ext
ec82d85a
PH
2050
2051 if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(thumb_filename)):
2052 self.to_screen('[%s] %s: Thumbnail %sis already present' %
2053 (info_dict['extractor'], info_dict['id'], thumb_display_id))
2054 else:
2055 self.to_screen('[%s] %s: Downloading thumbnail %s...' %
2056 (info_dict['extractor'], info_dict['id'], thumb_display_id))
2057 try:
2058 uf = self.urlopen(t['url'])
d3d89c32 2059 with open(encodeFilename(thumb_filename), 'wb') as thumbf:
ec82d85a
PH
2060 shutil.copyfileobj(uf, thumbf)
2061 self.to_screen('[%s] %s: Writing thumbnail %sto: %s' %
2062 (info_dict['extractor'], info_dict['id'], thumb_display_id, thumb_filename))
2063 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
2064 self.report_warning('Unable to download thumbnail "%s": %s' %
9b9c5355 2065 (t['url'], error_to_compat_str(err)))