]> jfr.im git - yt-dlp.git/blame - youtube_dl/YoutubeDL.py
[daum] Add 'thumbnail' to all _TESTS
[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 48 determine_ext,
b5559424 49 determine_protocol,
ce02ed60 50 DownloadError,
c0384f22 51 encode_compat_str,
ce02ed60 52 encodeFilename,
9b9c5355 53 error_to_compat_str,
ce02ed60 54 ExtractorError,
02dbf93f 55 format_bytes,
525ef922 56 formatSeconds,
ce02ed60 57 locked_file,
dca08720 58 make_HTTPS_handler,
ce02ed60 59 MaxDownloadsReached,
b7ab0590 60 PagedList,
083c9df9 61 parse_filesize,
91410c9b 62 PerRequestProxyHandler,
ce02ed60 63 PostProcessingError,
dca08720 64 platform_name,
ce02ed60 65 preferredencoding,
cfb56d1a 66 render_table,
ce02ed60
PH
67 SameFileError,
68 sanitize_filename,
1bb5c511 69 sanitize_path,
67dda517 70 sanitized_Request,
e5660ee6 71 std_headers,
ce02ed60 72 subtitles_filename,
ce02ed60 73 UnavailableVideoError,
29eb5174 74 url_basename,
58b1f00d 75 version_tuple,
ce02ed60
PH
76 write_json_file,
77 write_string,
6a3f4c3f 78 YoutubeDLCookieProcessor,
dca08720 79 YoutubeDLHandler,
6350728b 80 prepend_extension,
b29e0000 81 replace_extension,
7d4111ed 82 args_to_str,
05900629 83 age_restricted,
ce02ed60 84)
a0e07d31 85from .cache import Cache
023fa8c4 86from .extractor import get_info_extractor, gen_extractors
3bc2ddcc 87from .downloader import get_suitable_downloader
4c83c967 88from .downloader.rtmp import rtmpdump_version
4f026faf 89from .postprocessor import (
62cd676c 90 FFmpegFixupM4aPP,
6271f1ca 91 FFmpegFixupStretchedPP,
4f026faf
PH
92 FFmpegMergerPP,
93 FFmpegPostProcessor,
94 get_postprocessor,
95)
dca08720 96from .version import __version__
8222d8de
JMF
97
98
99class YoutubeDL(object):
100 """YoutubeDL class.
101
102 YoutubeDL objects are the ones responsible of downloading the
103 actual video file and writing it to disk if the user has requested
104 it, among some other tasks. In most cases there should be one per
105 program. As, given a video URL, the downloader doesn't know how to
106 extract all the needed information, task that InfoExtractors do, it
107 has to pass the URL to one of them.
108
109 For this, YoutubeDL objects have a method that allows
110 InfoExtractors to be registered in a given order. When it is passed
111 a URL, the YoutubeDL object handles it to the first InfoExtractor it
112 finds that reports being able to handle it. The InfoExtractor extracts
113 all the information about the video or videos the URL refers to, and
114 YoutubeDL process the extracted information, possibly using a File
115 Downloader to download the video.
116
117 YoutubeDL objects accept a lot of parameters. In order not to saturate
118 the object constructor with arguments, it receives a dictionary of
119 options instead. These options are available through the params
120 attribute for the InfoExtractors to use. The YoutubeDL also
121 registers itself as the downloader in charge for the InfoExtractors
122 that are added to it, so this is a "mutual registration".
123
124 Available options:
125
126 username: Username for authentication purposes.
127 password: Password for authentication purposes.
180940e0 128 videopassword: Password for accessing a video.
8222d8de
JMF
129 usenetrc: Use netrc for authentication instead.
130 verbose: Print additional info to stdout.
131 quiet: Do not print messages to stdout.
ad8915b7 132 no_warnings: Do not print out anything for warnings.
8222d8de
JMF
133 forceurl: Force printing final URL.
134 forcetitle: Force printing title.
135 forceid: Force printing ID.
136 forcethumbnail: Force printing thumbnail URL.
137 forcedescription: Force printing description.
138 forcefilename: Force printing final filename.
525ef922 139 forceduration: Force printing duration.
8694c600 140 forcejson: Force printing info_dict as JSON.
63e0be34
PH
141 dump_single_json: Force printing the info_dict of the whole playlist
142 (or video) as a single JSON line.
8222d8de 143 simulate: Do not download the video files.
d8600787 144 format: Video format code. See options.py for more information.
8222d8de
JMF
145 outtmpl: Template for output names.
146 restrictfilenames: Do not allow "&" and spaces in file names
147 ignoreerrors: Do not stop on download errors.
d22dec74 148 force_generic_extractor: Force downloader to use the generic extractor
8222d8de
JMF
149 nooverwrites: Prevent overwriting files.
150 playliststart: Playlist item to start at.
151 playlistend: Playlist item to end at.
c14e88f0 152 playlist_items: Specific indices of playlist to download.
ff815fe6 153 playlistreverse: Download playlist items in reverse order.
8222d8de
JMF
154 matchtitle: Download only matching titles.
155 rejecttitle: Reject downloads for matching titles.
8bf9319e 156 logger: Log messages to a logging.Logger instance.
8222d8de
JMF
157 logtostderr: Log messages to stderr instead of stdout.
158 writedescription: Write the video description to a .description file
159 writeinfojson: Write the video description to a .info.json file
1fb07d10 160 writeannotations: Write the video annotations to a .annotations.xml file
8222d8de 161 writethumbnail: Write the thumbnail image to a file
ec82d85a 162 write_all_thumbnails: Write all thumbnail formats to files
8222d8de 163 writesubtitles: Write the video subtitles to a file
741dd8ea 164 writeautomaticsub: Write the automatically generated subtitles to a file
8222d8de 165 allsubtitles: Downloads all the subtitles of the video
0b7f3118 166 (requires writesubtitles or writeautomaticsub)
8222d8de 167 listsubtitles: Lists all available subtitles for the video
a504ced0 168 subtitlesformat: The format code for subtitles
aa6a10c4 169 subtitleslangs: List of languages of the subtitles to download
8222d8de
JMF
170 keepvideo: Keep the video file after post-processing
171 daterange: A DateRange object, download only if the upload_date is in the range.
172 skip_download: Skip the actual download of the video file
c35f9e72 173 cachedir: Location of the cache files in the filesystem.
a0e07d31 174 False to disable filesystem cache.
47192f92 175 noplaylist: Download single video instead of a playlist if in doubt.
8dbe9899
PH
176 age_limit: An integer representing the user's age in years.
177 Unsuitable videos for the given age are skipped.
5fe18bdb
PH
178 min_views: An integer representing the minimum view count the video
179 must have in order to not be skipped.
180 Videos without view count information are always
181 downloaded. None for no limit.
182 max_views: An integer representing the maximum view count.
183 Videos that are more popular than that are not
184 downloaded.
185 Videos without view count information are always
186 downloaded. None for no limit.
187 download_archive: File name of a file where all downloads are recorded.
c1c9a79c
PH
188 Videos already present in the file are not downloaded
189 again.
dca08720 190 cookiefile: File name where cookies should be read from and dumped to.
a1ee09e8 191 nocheckcertificate:Do not verify SSL certificates
7e8c0af0
PH
192 prefer_insecure: Use HTTP instead of HTTPS to retrieve information.
193 At the moment, this is only supported by YouTube.
a1ee09e8 194 proxy: URL of the proxy server to use
91410c9b
PH
195 cn_verification_proxy: URL of the proxy to use for IP address verification
196 on Chinese sites. (Experimental)
e344693b 197 socket_timeout: Time to wait for unresponsive hosts, in seconds
0783b09b
PH
198 bidi_workaround: Work around buggy terminals without bidirectional text
199 support, using fridibi
a0ddb8a2 200 debug_printtraffic:Print out sent and received HTTP traffic
7b0817e8 201 include_ads: Download ads as well
04b4d394
PH
202 default_search: Prepend this string if an input url is not valid.
203 'auto' for elaborate guessing
62fec3b2 204 encoding: Use this encoding instead of the system-specified.
e8ee972c 205 extract_flat: Do not resolve URLs, return the immediate result.
057a5206
PH
206 Pass in 'in_playlist' to only show this behavior for
207 playlist items.
4f026faf 208 postprocessors: A list of dictionaries, each with an entry
71b640cc
PH
209 * key: The name of the postprocessor. See
210 youtube_dl/postprocessor/__init__.py for a list.
4f026faf
PH
211 as well as any further keyword arguments for the
212 postprocessor.
71b640cc
PH
213 progress_hooks: A list of functions that get called on download
214 progress, with a dictionary with the entries
5cda4eda 215 * status: One of "downloading", "error", or "finished".
ee69b99a 216 Check this first and ignore unknown values.
71b640cc 217
5cda4eda 218 If status is one of "downloading", or "finished", the
ee69b99a
PH
219 following properties may also be present:
220 * filename: The final filename (always present)
5cda4eda 221 * tmpfilename: The filename we're currently writing to
71b640cc
PH
222 * downloaded_bytes: Bytes on disk
223 * total_bytes: Size of the whole file, None if unknown
5cda4eda
PH
224 * total_bytes_estimate: Guess of the eventual file size,
225 None if unavailable.
226 * elapsed: The number of seconds since download started.
71b640cc
PH
227 * eta: The estimated time in seconds, None if unknown
228 * speed: The download speed in bytes/second, None if
229 unknown
5cda4eda
PH
230 * fragment_index: The counter of the currently
231 downloaded video fragment.
232 * fragment_count: The number of fragments (= individual
233 files that will be merged)
71b640cc
PH
234
235 Progress hooks are guaranteed to be called at least once
236 (with status "finished") if the download is successful.
45598f15 237 merge_output_format: Extension to use when merging formats.
6271f1ca
PH
238 fixup: Automatically correct known faults of the file.
239 One of:
240 - "never": do nothing
241 - "warn": only emit a warning
242 - "detect_or_warn": check whether we can do anything
62cd676c 243 about it, warn otherwise (default)
be4a824d 244 source_address: (Experimental) Client-side IP address to bind to.
6ec6cb4e 245 call_home: Boolean, true iff we are allowed to contact the
8bfa7545 246 youtube-dl servers for debugging.
5f0d813d 247 sleep_interval: Number of seconds to sleep before each download.
cfb56d1a
PH
248 listformats: Print an overview of available video formats and exit.
249 list_thumbnails: Print a table of all thumbnails and exit.
347de493
PH
250 match_filter: A function that gets called with the info_dict of
251 every video.
252 If it returns a message, the video is ignored.
253 If it returns None, the video is downloaded.
254 match_filter_func in utils.py is one example for this.
7e5db8c9 255 no_color: Do not emit color codes in output.
71b640cc 256
85729c51
PH
257 The following options determine which downloader is picked:
258 external_downloader: Executable of the external downloader to call.
259 None or unset for standard (built-in) downloader.
260 hls_prefer_native: Use the native HLS downloader instead of ffmpeg/avconv.
fe7e0c98 261
8222d8de 262 The following parameters are not used by YoutubeDL itself, they are used by
c75f0b36 263 the downloader (see youtube_dl/downloader/common.py):
8222d8de 264 nopart, updatetime, buffersize, ratelimit, min_filesize, max_filesize, test,
881e6a1f 265 noresizebuffer, retries, continuedl, noprogress, consoletitle,
c75f0b36 266 xattr_set_filesize, external_downloader_args.
76b1bd67
JMF
267
268 The following options are used by the post processors:
269 prefer_ffmpeg: If True, use ffmpeg instead of avconv if both are available,
270 otherwise prefer avconv.
f72b0a60
S
271 postprocessor_args: A list of additional command-line arguments for the
272 postprocessor.
8222d8de
JMF
273 """
274
275 params = None
276 _ies = []
277 _pps = []
278 _download_retcode = None
279 _num_downloads = None
280 _screen_file = None
281
3511266b 282 def __init__(self, params=None, auto_init=True):
8222d8de 283 """Create a FileDownloader object with the given options."""
e9f9a10f
JMF
284 if params is None:
285 params = {}
8222d8de 286 self._ies = []
56c73665 287 self._ies_instances = {}
8222d8de 288 self._pps = []
933605d7 289 self._progress_hooks = []
8222d8de
JMF
290 self._download_retcode = 0
291 self._num_downloads = 0
292 self._screen_file = [sys.stdout, sys.stderr][params.get('logtostderr', False)]
0783b09b 293 self._err_file = sys.stderr
4abf617b
S
294 self.params = {
295 # Default parameters
296 'nocheckcertificate': False,
297 }
298 self.params.update(params)
a0e07d31 299 self.cache = Cache(self)
34308b30 300
0783b09b 301 if params.get('bidi_workaround', False):
1c088fa8
PH
302 try:
303 import pty
304 master, slave = pty.openpty()
003c69a8 305 width = compat_get_terminal_size().columns
1c088fa8
PH
306 if width is None:
307 width_args = []
308 else:
309 width_args = ['-w', str(width)]
5d681e96 310 sp_kwargs = dict(
1c088fa8
PH
311 stdin=subprocess.PIPE,
312 stdout=slave,
313 stderr=self._err_file)
5d681e96
PH
314 try:
315 self._output_process = subprocess.Popen(
316 ['bidiv'] + width_args, **sp_kwargs
317 )
318 except OSError:
5d681e96
PH
319 self._output_process = subprocess.Popen(
320 ['fribidi', '-c', 'UTF-8'] + width_args, **sp_kwargs)
321 self._output_channel = os.fdopen(master, 'rb')
1c088fa8
PH
322 except OSError as ose:
323 if ose.errno == 2:
6febd1c1 324 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 $PATH.')
1c088fa8
PH
325 else:
326 raise
0783b09b 327
34308b30 328 if (sys.version_info >= (3,) and sys.platform != 'win32' and
8fb3ac36
PH
329 sys.getfilesystemencoding() in ['ascii', 'ANSI_X3.4-1968'] and
330 not params.get('restrictfilenames', False)):
34308b30
PH
331 # On Python 3, the Unicode filesystem API will throw errors (#1474)
332 self.report_warning(
6febd1c1 333 'Assuming --restrict-filenames since file system encoding '
1b725173 334 'cannot encode all characters. '
6febd1c1 335 'Set the LC_ALL environment variable to fix this.')
4a98cdbf 336 self.params['restrictfilenames'] = True
34308b30 337
486dd09e
PH
338 if isinstance(params.get('outtmpl'), bytes):
339 self.report_warning(
340 'Parameter outtmpl is bytes, but should be a unicode string. '
341 'Put from __future__ import unicode_literals at the top of your code file or consider switching to Python 3.x.')
342
dca08720
PH
343 self._setup_opener()
344
3511266b
PH
345 if auto_init:
346 self.print_debug_header()
347 self.add_default_info_extractors()
348
4f026faf
PH
349 for pp_def_raw in self.params.get('postprocessors', []):
350 pp_class = get_postprocessor(pp_def_raw['key'])
351 pp_def = dict(pp_def_raw)
352 del pp_def['key']
353 pp = pp_class(self, **compat_kwargs(pp_def))
354 self.add_post_processor(pp)
355
71b640cc
PH
356 for ph in self.params.get('progress_hooks', []):
357 self.add_progress_hook(ph)
358
7d4111ed
PH
359 def warn_if_short_id(self, argv):
360 # short YouTube ID starting with dash?
361 idxs = [
362 i for i, a in enumerate(argv)
363 if re.match(r'^-[0-9A-Za-z_-]{10}$', a)]
364 if idxs:
365 correct_argv = (
366 ['youtube-dl'] +
367 [a for i, a in enumerate(argv) if i not in idxs] +
368 ['--'] + [argv[i] for i in idxs]
369 )
370 self.report_warning(
371 'Long argument string detected. '
372 'Use -- to separate parameters and URLs, like this:\n%s\n' %
373 args_to_str(correct_argv))
374
8222d8de
JMF
375 def add_info_extractor(self, ie):
376 """Add an InfoExtractor object to the end of the list."""
377 self._ies.append(ie)
56c73665 378 self._ies_instances[ie.ie_key()] = ie
8222d8de
JMF
379 ie.set_downloader(self)
380
56c73665
JMF
381 def get_info_extractor(self, ie_key):
382 """
383 Get an instance of an IE with name ie_key, it will try to get one from
384 the _ies list, if there's no instance it will create a new one and add
385 it to the extractor list.
386 """
387 ie = self._ies_instances.get(ie_key)
388 if ie is None:
389 ie = get_info_extractor(ie_key)()
390 self.add_info_extractor(ie)
391 return ie
392
023fa8c4
JMF
393 def add_default_info_extractors(self):
394 """
395 Add the InfoExtractors returned by gen_extractors to the end of the list
396 """
397 for ie in gen_extractors():
398 self.add_info_extractor(ie)
399
8222d8de
JMF
400 def add_post_processor(self, pp):
401 """Add a PostProcessor object to the end of the chain."""
402 self._pps.append(pp)
403 pp.set_downloader(self)
404
933605d7
JMF
405 def add_progress_hook(self, ph):
406 """Add the progress hook (currently only for the file downloader)"""
407 self._progress_hooks.append(ph)
8ab470f1 408
1c088fa8 409 def _bidi_workaround(self, message):
5d681e96 410 if not hasattr(self, '_output_channel'):
1c088fa8
PH
411 return message
412
5d681e96 413 assert hasattr(self, '_output_process')
11b85ce6 414 assert isinstance(message, compat_str)
6febd1c1
PH
415 line_count = message.count('\n') + 1
416 self._output_process.stdin.write((message + '\n').encode('utf-8'))
5d681e96 417 self._output_process.stdin.flush()
6febd1c1 418 res = ''.join(self._output_channel.readline().decode('utf-8')
9e1a5b84 419 for _ in range(line_count))
6febd1c1 420 return res[:-len('\n')]
1c088fa8 421
8222d8de 422 def to_screen(self, message, skip_eol=False):
0783b09b
PH
423 """Print message to stdout if not in quiet mode."""
424 return self.to_stdout(message, skip_eol, check_quiet=True)
425
734f90bb 426 def _write_string(self, s, out=None):
b58ddb32 427 write_string(s, out=out, encoding=self.params.get('encoding'))
734f90bb 428
0783b09b 429 def to_stdout(self, message, skip_eol=False, check_quiet=False):
8222d8de 430 """Print message to stdout if not in quiet mode."""
8bf9319e 431 if self.params.get('logger'):
43afe285 432 self.params['logger'].debug(message)
0783b09b 433 elif not check_quiet or not self.params.get('quiet', False):
1c088fa8 434 message = self._bidi_workaround(message)
6febd1c1 435 terminator = ['\n', ''][skip_eol]
8222d8de 436 output = message + terminator
1c088fa8 437
734f90bb 438 self._write_string(output, self._screen_file)
8222d8de
JMF
439
440 def to_stderr(self, message):
441 """Print message to stderr."""
11b85ce6 442 assert isinstance(message, compat_str)
8bf9319e 443 if self.params.get('logger'):
43afe285
IB
444 self.params['logger'].error(message)
445 else:
1c088fa8 446 message = self._bidi_workaround(message)
6febd1c1 447 output = message + '\n'
734f90bb 448 self._write_string(output, self._err_file)
8222d8de 449
1e5b9a95
PH
450 def to_console_title(self, message):
451 if not self.params.get('consoletitle', False):
452 return
453 if os.name == 'nt' and ctypes.windll.kernel32.GetConsoleWindow():
454 # c_wchar_p() might not be necessary if `message` is
455 # already of type unicode()
456 ctypes.windll.kernel32.SetConsoleTitleW(ctypes.c_wchar_p(message))
457 elif 'TERM' in os.environ:
734f90bb 458 self._write_string('\033]0;%s\007' % message, self._screen_file)
1e5b9a95 459
bdde425c
PH
460 def save_console_title(self):
461 if not self.params.get('consoletitle', False):
462 return
463 if 'TERM' in os.environ:
efd6c574 464 # Save the title on stack
734f90bb 465 self._write_string('\033[22;0t', self._screen_file)
bdde425c
PH
466
467 def restore_console_title(self):
468 if not self.params.get('consoletitle', False):
469 return
470 if 'TERM' in os.environ:
efd6c574 471 # Restore the title from stack
734f90bb 472 self._write_string('\033[23;0t', self._screen_file)
bdde425c
PH
473
474 def __enter__(self):
475 self.save_console_title()
476 return self
477
478 def __exit__(self, *args):
479 self.restore_console_title()
f89197d7 480
dca08720
PH
481 if self.params.get('cookiefile') is not None:
482 self.cookiejar.save()
bdde425c 483
8222d8de
JMF
484 def trouble(self, message=None, tb=None):
485 """Determine action to take when a download problem appears.
486
487 Depending on if the downloader has been configured to ignore
488 download errors or not, this method may throw an exception or
489 not when errors are found, after printing the message.
490
491 tb, if given, is additional traceback information.
492 """
493 if message is not None:
494 self.to_stderr(message)
495 if self.params.get('verbose'):
496 if tb is None:
497 if sys.exc_info()[0]: # if .trouble has been called from an except block
6febd1c1 498 tb = ''
8222d8de 499 if hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
6febd1c1 500 tb += ''.join(traceback.format_exception(*sys.exc_info()[1].exc_info))
c0384f22 501 tb += encode_compat_str(traceback.format_exc())
8222d8de
JMF
502 else:
503 tb_data = traceback.format_list(traceback.extract_stack())
6febd1c1 504 tb = ''.join(tb_data)
8222d8de
JMF
505 self.to_stderr(tb)
506 if not self.params.get('ignoreerrors', False):
507 if sys.exc_info()[0] and hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
508 exc_info = sys.exc_info()[1].exc_info
509 else:
510 exc_info = sys.exc_info()
511 raise DownloadError(message, exc_info)
512 self._download_retcode = 1
513
514 def report_warning(self, message):
515 '''
516 Print the message to stderr, it will be prefixed with 'WARNING:'
517 If stderr is a tty file the 'WARNING:' will be colored
518 '''
6d07ce01
JMF
519 if self.params.get('logger') is not None:
520 self.params['logger'].warning(message)
8222d8de 521 else:
ad8915b7
PH
522 if self.params.get('no_warnings'):
523 return
7e5db8c9 524 if not self.params.get('no_color') and self._err_file.isatty() and os.name != 'nt':
6d07ce01
JMF
525 _msg_header = '\033[0;33mWARNING:\033[0m'
526 else:
527 _msg_header = 'WARNING:'
528 warning_message = '%s %s' % (_msg_header, message)
529 self.to_stderr(warning_message)
8222d8de
JMF
530
531 def report_error(self, message, tb=None):
532 '''
533 Do the same as trouble, but prefixes the message with 'ERROR:', colored
534 in red if stderr is a tty file.
535 '''
7e5db8c9 536 if not self.params.get('no_color') and self._err_file.isatty() and os.name != 'nt':
6febd1c1 537 _msg_header = '\033[0;31mERROR:\033[0m'
8222d8de 538 else:
6febd1c1
PH
539 _msg_header = 'ERROR:'
540 error_message = '%s %s' % (_msg_header, message)
8222d8de
JMF
541 self.trouble(error_message, tb)
542
8222d8de
JMF
543 def report_file_already_downloaded(self, file_name):
544 """Report file has already been fully downloaded."""
545 try:
6febd1c1 546 self.to_screen('[download] %s has already been downloaded' % file_name)
ce02ed60 547 except UnicodeEncodeError:
6febd1c1 548 self.to_screen('[download] The file has already been downloaded')
8222d8de 549
8222d8de
JMF
550 def prepare_filename(self, info_dict):
551 """Generate the output filename."""
552 try:
553 template_dict = dict(info_dict)
554
555 template_dict['epoch'] = int(time.time())
556 autonumber_size = self.params.get('autonumber_size')
557 if autonumber_size is None:
558 autonumber_size = 5
6febd1c1 559 autonumber_templ = '%0' + str(autonumber_size) + 'd'
8222d8de 560 template_dict['autonumber'] = autonumber_templ % self._num_downloads
702665c0 561 if template_dict.get('playlist_index') is not None:
c6b4132a 562 template_dict['playlist_index'] = '%0*d' % (len(str(template_dict['n_entries'])), template_dict['playlist_index'])
17b75c0d
PH
563 if template_dict.get('resolution') is None:
564 if template_dict.get('width') and template_dict.get('height'):
565 template_dict['resolution'] = '%dx%d' % (template_dict['width'], template_dict['height'])
566 elif template_dict.get('height'):
805ef3c6 567 template_dict['resolution'] = '%sp' % template_dict['height']
17b75c0d 568 elif template_dict.get('width'):
805ef3c6 569 template_dict['resolution'] = '?x%d' % template_dict['width']
8222d8de 570
586a91b6 571 sanitize = lambda k, v: sanitize_filename(
45598aab 572 compat_str(v),
1bb5c511 573 restricted=self.params.get('restrictfilenames'),
6febd1c1 574 is_id=(k == 'id'))
586a91b6 575 template_dict = dict((k, sanitize(k, v))
45598aab
PH
576 for k, v in template_dict.items()
577 if v is not None)
6febd1c1 578 template_dict = collections.defaultdict(lambda: 'NA', template_dict)
8222d8de 579
b3613d36 580 outtmpl = self.params.get('outtmpl', DEFAULT_OUTTMPL)
4644ac55 581 tmpl = compat_expanduser(outtmpl)
586a91b6 582 filename = tmpl % template_dict
3a0d2f52
S
583 # Temporary fix for #4787
584 # 'Treat' all problem characters by passing filename through preferredencoding
585 # to workaround encoding issues with subprocess on python2 @ Windows
586 if sys.version_info < (3, 0) and sys.platform == 'win32':
587 filename = encodeFilename(filename, True).decode(preferredencoding())
b3613d36 588 return sanitize_path(filename)
8222d8de 589 except ValueError as err:
6febd1c1 590 self.report_error('Error in output template: ' + str(err) + ' (encoding: ' + repr(preferredencoding()) + ')')
8222d8de
JMF
591 return None
592
442c37b7 593 def _match_entry(self, info_dict, incomplete):
6ec6cb4e 594 """ Returns None iff the file should be downloaded """
8222d8de 595
6febd1c1 596 video_title = info_dict.get('title', info_dict.get('id', 'video'))
7012b23c
PH
597 if 'title' in info_dict:
598 # This can happen when we're just evaluating the playlist
599 title = info_dict['title']
600 matchtitle = self.params.get('matchtitle', False)
601 if matchtitle:
602 if not re.search(matchtitle, title, re.IGNORECASE):
6febd1c1 603 return '"' + title + '" title did not match pattern "' + matchtitle + '"'
7012b23c
PH
604 rejecttitle = self.params.get('rejecttitle', False)
605 if rejecttitle:
606 if re.search(rejecttitle, title, re.IGNORECASE):
6febd1c1 607 return '"' + title + '" title matched reject pattern "' + rejecttitle + '"'
8222d8de
JMF
608 date = info_dict.get('upload_date', None)
609 if date is not None:
610 dateRange = self.params.get('daterange', DateRange())
611 if date not in dateRange:
6febd1c1 612 return '%s upload date is not in range %s' % (date_from_str(date).isoformat(), dateRange)
5fe18bdb
PH
613 view_count = info_dict.get('view_count', None)
614 if view_count is not None:
615 min_views = self.params.get('min_views')
616 if min_views is not None and view_count < min_views:
6febd1c1 617 return 'Skipping %s, because it has not reached minimum view count (%d/%d)' % (video_title, view_count, min_views)
5fe18bdb
PH
618 max_views = self.params.get('max_views')
619 if max_views is not None and view_count > max_views:
6febd1c1 620 return 'Skipping %s, because it has exceeded the maximum view count (%d/%d)' % (video_title, view_count, max_views)
05900629 621 if age_restricted(info_dict.get('age_limit'), self.params.get('age_limit')):
347de493 622 return 'Skipping "%s" because it is age restricted' % video_title
c1c9a79c 623 if self.in_download_archive(info_dict):
6febd1c1 624 return '%s has already been recorded in archive' % video_title
347de493 625
442c37b7
PH
626 if not incomplete:
627 match_filter = self.params.get('match_filter')
628 if match_filter is not None:
629 ret = match_filter(info_dict)
630 if ret is not None:
631 return ret
347de493 632
8222d8de 633 return None
fe7e0c98 634
b6c45014
JMF
635 @staticmethod
636 def add_extra_info(info_dict, extra_info):
637 '''Set the keys from extra_info in info dict if they are missing'''
638 for key, value in extra_info.items():
639 info_dict.setdefault(key, value)
640
7fc3fa05 641 def extract_info(self, url, download=True, ie_key=None, extra_info={},
61aa5ba3 642 process=True, force_generic_extractor=False):
8222d8de
JMF
643 '''
644 Returns a list with a dictionary for each video we find.
645 If 'download', also downloads the videos.
646 extra_info is a dict containing the extra values to add to each result
613b2d9d 647 '''
fe7e0c98 648
61aa5ba3 649 if not ie_key and force_generic_extractor:
d22dec74
S
650 ie_key = 'Generic'
651
8222d8de 652 if ie_key:
56c73665 653 ies = [self.get_info_extractor(ie_key)]
8222d8de
JMF
654 else:
655 ies = self._ies
656
657 for ie in ies:
658 if not ie.suitable(url):
659 continue
660
661 if not ie.working():
6febd1c1
PH
662 self.report_warning('The program functionality for this site has been marked as broken, '
663 'and will probably not work.')
8222d8de
JMF
664
665 try:
666 ie_result = ie.extract(url)
5f6a1245 667 if ie_result is None: # Finished already (backwards compatibility; listformats and friends should be moved here)
8222d8de
JMF
668 break
669 if isinstance(ie_result, list):
670 # Backwards compatibility: old IE result format
8222d8de
JMF
671 ie_result = {
672 '_type': 'compat_list',
673 'entries': ie_result,
674 }
ea38e55f 675 self.add_default_extra_info(ie_result, ie, url)
7fc3fa05
PH
676 if process:
677 return self.process_ie_result(ie_result, download, extra_info)
678 else:
679 return ie_result
fb043a6e 680 except ExtractorError as e: # An error we somewhat expected
2c74e6fa 681 self.report_error(compat_str(e), e.format_traceback())
8222d8de 682 break
d3e5bbf4
PH
683 except MaxDownloadsReached:
684 raise
8222d8de
JMF
685 except Exception as e:
686 if self.params.get('ignoreerrors', False):
9b9c5355 687 self.report_error(error_to_compat_str(e), tb=encode_compat_str(traceback.format_exc()))
8222d8de
JMF
688 break
689 else:
690 raise
691 else:
1a489545 692 self.report_error('no suitable InfoExtractor for URL %s' % url)
fe7e0c98 693
ea38e55f
PH
694 def add_default_extra_info(self, ie_result, ie, url):
695 self.add_extra_info(ie_result, {
696 'extractor': ie.IE_NAME,
697 'webpage_url': url,
698 'webpage_url_basename': url_basename(url),
699 'extractor_key': ie.ie_key(),
700 })
701
8222d8de
JMF
702 def process_ie_result(self, ie_result, download=True, extra_info={}):
703 """
704 Take the result of the ie(may be modified) and resolve all unresolved
705 references (URLs, playlist items).
706
707 It will also download the videos if 'download'.
708 Returns the resolved ie_result.
709 """
710
e8ee972c
PH
711 result_type = ie_result.get('_type', 'video')
712
057a5206
PH
713 if result_type in ('url', 'url_transparent'):
714 extract_flat = self.params.get('extract_flat', False)
715 if ((extract_flat == 'in_playlist' and 'playlist' in extra_info) or
716 extract_flat is True):
057a5206
PH
717 if self.params.get('forcejson', False):
718 self.to_stdout(json.dumps(ie_result))
e8ee972c
PH
719 return ie_result
720
8222d8de 721 if result_type == 'video':
b6c45014 722 self.add_extra_info(ie_result, extra_info)
feee2ecf 723 return self.process_video_result(ie_result, download=download)
8222d8de
JMF
724 elif result_type == 'url':
725 # We have to add extra_info to the results because it may be
726 # contained in a playlist
727 return self.extract_info(ie_result['url'],
728 download,
729 ie_key=ie_result.get('ie_key'),
730 extra_info=extra_info)
7fc3fa05
PH
731 elif result_type == 'url_transparent':
732 # Use the information from the embedding page
733 info = self.extract_info(
734 ie_result['url'], ie_key=ie_result.get('ie_key'),
735 extra_info=extra_info, download=False, process=False)
736
412c617d
PH
737 force_properties = dict(
738 (k, v) for k, v in ie_result.items() if v is not None)
739 for f in ('_type', 'url'):
740 if f in force_properties:
741 del force_properties[f]
742 new_result = info.copy()
743 new_result.update(force_properties)
7fc3fa05
PH
744
745 assert new_result.get('_type') != 'url_transparent'
7fc3fa05
PH
746
747 return self.process_ie_result(
748 new_result, download=download, extra_info=extra_info)
42e12102 749 elif result_type == 'playlist' or result_type == 'multi_video':
8222d8de
JMF
750 # We process each entry in the playlist
751 playlist = ie_result.get('title', None) or ie_result.get('id', None)
6febd1c1 752 self.to_screen('[download] Downloading playlist: %s' % playlist)
8222d8de
JMF
753
754 playlist_results = []
755
8222d8de 756 playliststart = self.params.get('playliststart', 1) - 1
a19fd00c
PH
757 playlistend = self.params.get('playlistend', None)
758 # For backwards compatibility, interpret -1 as whole list
8222d8de 759 if playlistend == -1:
a19fd00c 760 playlistend = None
8222d8de 761
c14e88f0
PH
762 playlistitems_str = self.params.get('playlist_items', None)
763 playlistitems = None
764 if playlistitems_str is not None:
765 def iter_playlistitems(format):
766 for string_segment in format.split(','):
767 if '-' in string_segment:
768 start, end = string_segment.split('-')
769 for item in range(int(start), int(end) + 1):
770 yield int(item)
771 else:
772 yield int(string_segment)
773 playlistitems = iter_playlistitems(playlistitems_str)
774
b82f815f
PH
775 ie_entries = ie_result['entries']
776 if isinstance(ie_entries, list):
777 n_all_entries = len(ie_entries)
c14e88f0 778 if playlistitems:
3884dcf3
JMF
779 entries = [
780 ie_entries[i - 1] for i in playlistitems
781 if -n_all_entries <= i - 1 < n_all_entries]
c14e88f0
PH
782 else:
783 entries = ie_entries[playliststart:playlistend]
b7ab0590
PH
784 n_entries = len(entries)
785 self.to_screen(
786 "[%s] playlist %s: Collected %d video ids (downloading %d of them)" %
787 (ie_result['extractor'], playlist, n_all_entries, n_entries))
b82f815f 788 elif isinstance(ie_entries, PagedList):
c14e88f0
PH
789 if playlistitems:
790 entries = []
791 for item in playlistitems:
792 entries.extend(ie_entries.getslice(
793 item - 1, item
794 ))
795 else:
796 entries = ie_entries.getslice(
797 playliststart, playlistend)
b7ab0590
PH
798 n_entries = len(entries)
799 self.to_screen(
800 "[%s] playlist %s: Downloading %d videos" %
801 (ie_result['extractor'], playlist, n_entries))
b82f815f 802 else: # iterable
c14e88f0
PH
803 if playlistitems:
804 entry_list = list(ie_entries)
805 entries = [entry_list[i - 1] for i in playlistitems]
806 else:
807 entries = list(itertools.islice(
808 ie_entries, playliststart, playlistend))
b82f815f
PH
809 n_entries = len(entries)
810 self.to_screen(
811 "[%s] playlist %s: Downloading %d videos" %
812 (ie_result['extractor'], playlist, n_entries))
8222d8de 813
ff815fe6
MS
814 if self.params.get('playlistreverse', False):
815 entries = entries[::-1]
816
fe7e0c98 817 for i, entry in enumerate(entries, 1):
734ea11e 818 self.to_screen('[download] Downloading video %s of %s' % (i, n_entries))
8222d8de 819 extra = {
c6b4132a 820 'n_entries': n_entries,
fe7e0c98 821 'playlist': playlist,
a1cf99d0
PH
822 'playlist_id': ie_result.get('id'),
823 'playlist_title': ie_result.get('title'),
fe7e0c98 824 'playlist_index': i + playliststart,
b6c45014 825 'extractor': ie_result['extractor'],
9103bbc5 826 'webpage_url': ie_result['webpage_url'],
29eb5174 827 'webpage_url_basename': url_basename(ie_result['webpage_url']),
be97abc2 828 'extractor_key': ie_result['extractor_key'],
fe7e0c98 829 }
7012b23c 830
442c37b7 831 reason = self._match_entry(entry, incomplete=True)
7012b23c 832 if reason is not None:
6febd1c1 833 self.to_screen('[download] ' + reason)
7012b23c
PH
834 continue
835
8222d8de
JMF
836 entry_result = self.process_ie_result(entry,
837 download=download,
838 extra_info=extra)
839 playlist_results.append(entry_result)
840 ie_result['entries'] = playlist_results
371c3b79 841 self.to_screen('[download] Finished downloading playlist: %s' % playlist)
8222d8de
JMF
842 return ie_result
843 elif result_type == 'compat_list':
c9bf4114
PH
844 self.report_warning(
845 'Extractor %s returned a compat_list result. '
846 'It needs to be updated.' % ie_result.get('extractor'))
5f6a1245 847
8222d8de 848 def _fixup(r):
9e1a5b84
JW
849 self.add_extra_info(
850 r,
9103bbc5
JMF
851 {
852 'extractor': ie_result['extractor'],
853 'webpage_url': ie_result['webpage_url'],
29eb5174 854 'webpage_url_basename': url_basename(ie_result['webpage_url']),
be97abc2 855 'extractor_key': ie_result['extractor_key'],
9e1a5b84
JW
856 }
857 )
8222d8de
JMF
858 return r
859 ie_result['entries'] = [
b6c45014 860 self.process_ie_result(_fixup(r), download, extra_info)
8222d8de
JMF
861 for r in ie_result['entries']
862 ]
863 return ie_result
864 else:
865 raise Exception('Invalid result type: %s' % result_type)
866
67134eab
JMF
867 def _build_format_filter(self, filter_spec):
868 " Returns a function to filter the formats according to the filter_spec "
083c9df9
PH
869
870 OPERATORS = {
871 '<': operator.lt,
872 '<=': operator.le,
873 '>': operator.gt,
874 '>=': operator.ge,
875 '=': operator.eq,
876 '!=': operator.ne,
877 }
67134eab 878 operator_rex = re.compile(r'''(?x)\s*
2ec19e95 879 (?P<key>width|height|tbr|abr|vbr|asr|filesize|fps)
083c9df9
PH
880 \s*(?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
881 (?P<value>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?)
67134eab 882 $
083c9df9 883 ''' % '|'.join(map(re.escape, OPERATORS.keys())))
67134eab 884 m = operator_rex.search(filter_spec)
9ddb6925
S
885 if m:
886 try:
887 comparison_value = int(m.group('value'))
888 except ValueError:
889 comparison_value = parse_filesize(m.group('value'))
890 if comparison_value is None:
891 comparison_value = parse_filesize(m.group('value') + 'B')
892 if comparison_value is None:
893 raise ValueError(
894 'Invalid value %r in format specification %r' % (
67134eab 895 m.group('value'), filter_spec))
9ddb6925
S
896 op = OPERATORS[m.group('op')]
897
083c9df9 898 if not m:
9ddb6925
S
899 STR_OPERATORS = {
900 '=': operator.eq,
901 '!=': operator.ne,
10d33b34
YCH
902 '^=': lambda attr, value: attr.startswith(value),
903 '$=': lambda attr, value: attr.endswith(value),
904 '*=': lambda attr, value: value in attr,
9ddb6925 905 }
67134eab 906 str_operator_rex = re.compile(r'''(?x)
9ddb6925
S
907 \s*(?P<key>ext|acodec|vcodec|container|protocol)
908 \s*(?P<op>%s)(?P<none_inclusive>\s*\?)?
b0df5223 909 \s*(?P<value>[a-zA-Z0-9._-]+)
67134eab 910 \s*$
9ddb6925 911 ''' % '|'.join(map(re.escape, STR_OPERATORS.keys())))
67134eab 912 m = str_operator_rex.search(filter_spec)
9ddb6925
S
913 if m:
914 comparison_value = m.group('value')
915 op = STR_OPERATORS[m.group('op')]
083c9df9 916
9ddb6925 917 if not m:
67134eab 918 raise ValueError('Invalid filter specification %r' % filter_spec)
083c9df9
PH
919
920 def _filter(f):
921 actual_value = f.get(m.group('key'))
922 if actual_value is None:
923 return m.group('none_inclusive')
924 return op(actual_value, comparison_value)
67134eab
JMF
925 return _filter
926
927 def build_format_selector(self, format_spec):
928 def syntax_error(note, start):
929 message = (
930 'Invalid format specification: '
931 '{0}\n\t{1}\n\t{2}^'.format(note, format_spec, ' ' * start[1]))
932 return SyntaxError(message)
933
934 PICKFIRST = 'PICKFIRST'
935 MERGE = 'MERGE'
936 SINGLE = 'SINGLE'
0130afb7 937 GROUP = 'GROUP'
67134eab
JMF
938 FormatSelector = collections.namedtuple('FormatSelector', ['type', 'selector', 'filters'])
939
940 def _parse_filter(tokens):
941 filter_parts = []
942 for type, string, start, _, _ in tokens:
943 if type == tokenize.OP and string == ']':
944 return ''.join(filter_parts)
945 else:
946 filter_parts.append(string)
947
232541df 948 def _remove_unused_ops(tokens):
17cc1534 949 # Remove operators that we don't use and join them with the surrounding strings
232541df
JMF
950 # for example: 'mp4' '-' 'baseline' '-' '16x9' is converted to 'mp4-baseline-16x9'
951 ALLOWED_OPS = ('/', '+', ',', '(', ')')
952 last_string, last_start, last_end, last_line = None, None, None, None
953 for type, string, start, end, line in tokens:
954 if type == tokenize.OP and string == '[':
955 if last_string:
956 yield tokenize.NAME, last_string, last_start, last_end, last_line
957 last_string = None
958 yield type, string, start, end, line
959 # everything inside brackets will be handled by _parse_filter
960 for type, string, start, end, line in tokens:
961 yield type, string, start, end, line
962 if type == tokenize.OP and string == ']':
963 break
964 elif type == tokenize.OP and string in ALLOWED_OPS:
965 if last_string:
966 yield tokenize.NAME, last_string, last_start, last_end, last_line
967 last_string = None
968 yield type, string, start, end, line
969 elif type in [tokenize.NAME, tokenize.NUMBER, tokenize.OP]:
970 if not last_string:
971 last_string = string
972 last_start = start
973 last_end = end
974 else:
975 last_string += string
976 if last_string:
977 yield tokenize.NAME, last_string, last_start, last_end, last_line
978
cf2ac6df 979 def _parse_format_selection(tokens, inside_merge=False, inside_choice=False, inside_group=False):
67134eab
JMF
980 selectors = []
981 current_selector = None
982 for type, string, start, _, _ in tokens:
983 # ENCODING is only defined in python 3.x
984 if type == getattr(tokenize, 'ENCODING', None):
985 continue
986 elif type in [tokenize.NAME, tokenize.NUMBER]:
987 current_selector = FormatSelector(SINGLE, string, [])
988 elif type == tokenize.OP:
cf2ac6df
JMF
989 if string == ')':
990 if not inside_group:
991 # ')' will be handled by the parentheses group
992 tokens.restore_last_token()
67134eab 993 break
cf2ac6df 994 elif inside_merge and string in ['/', ',']:
0130afb7
JMF
995 tokens.restore_last_token()
996 break
cf2ac6df
JMF
997 elif inside_choice and string == ',':
998 tokens.restore_last_token()
999 break
1000 elif string == ',':
0a31a350
JMF
1001 if not current_selector:
1002 raise syntax_error('"," must follow a format selector', start)
67134eab
JMF
1003 selectors.append(current_selector)
1004 current_selector = None
1005 elif string == '/':
d96d604e
JMF
1006 if not current_selector:
1007 raise syntax_error('"/" must follow a format selector', start)
67134eab 1008 first_choice = current_selector
cf2ac6df 1009 second_choice = _parse_format_selection(tokens, inside_choice=True)
f5f4a27a 1010 current_selector = FormatSelector(PICKFIRST, (first_choice, second_choice), [])
67134eab
JMF
1011 elif string == '[':
1012 if not current_selector:
1013 current_selector = FormatSelector(SINGLE, 'best', [])
1014 format_filter = _parse_filter(tokens)
1015 current_selector.filters.append(format_filter)
0130afb7
JMF
1016 elif string == '(':
1017 if current_selector:
1018 raise syntax_error('Unexpected "("', start)
cf2ac6df
JMF
1019 group = _parse_format_selection(tokens, inside_group=True)
1020 current_selector = FormatSelector(GROUP, group, [])
67134eab
JMF
1021 elif string == '+':
1022 video_selector = current_selector
cf2ac6df 1023 audio_selector = _parse_format_selection(tokens, inside_merge=True)
0a31a350
JMF
1024 if not video_selector or not audio_selector:
1025 raise syntax_error('"+" must be between two format selectors', start)
cf2ac6df 1026 current_selector = FormatSelector(MERGE, (video_selector, audio_selector), [])
67134eab
JMF
1027 else:
1028 raise syntax_error('Operator not recognized: "{0}"'.format(string), start)
1029 elif type == tokenize.ENDMARKER:
1030 break
1031 if current_selector:
1032 selectors.append(current_selector)
1033 return selectors
1034
1035 def _build_selector_function(selector):
1036 if isinstance(selector, list):
1037 fs = [_build_selector_function(s) for s in selector]
1038
1039 def selector_function(formats):
1040 for f in fs:
1041 for format in f(formats):
1042 yield format
1043 return selector_function
0130afb7
JMF
1044 elif selector.type == GROUP:
1045 selector_function = _build_selector_function(selector.selector)
67134eab
JMF
1046 elif selector.type == PICKFIRST:
1047 fs = [_build_selector_function(s) for s in selector.selector]
1048
1049 def selector_function(formats):
1050 for f in fs:
1051 picked_formats = list(f(formats))
1052 if picked_formats:
1053 return picked_formats
1054 return []
1055 elif selector.type == SINGLE:
1056 format_spec = selector.selector
1057
1058 def selector_function(formats):
bb8e5536
JMF
1059 formats = list(formats)
1060 if not formats:
1061 return
5acfa126
JMF
1062 if format_spec == 'all':
1063 for f in formats:
1064 yield f
1065 elif format_spec in ['best', 'worst', None]:
67134eab
JMF
1066 format_idx = 0 if format_spec == 'worst' else -1
1067 audiovideo_formats = [
1068 f for f in formats
1069 if f.get('vcodec') != 'none' and f.get('acodec') != 'none']
1070 if audiovideo_formats:
1071 yield audiovideo_formats[format_idx]
1072 # for audio only (soundcloud) or video only (imgur) urls, select the best/worst audio format
1073 elif (all(f.get('acodec') != 'none' for f in formats) or
1074 all(f.get('vcodec') != 'none' for f in formats)):
1075 yield formats[format_idx]
1076 elif format_spec == 'bestaudio':
1077 audio_formats = [
1078 f for f in formats
1079 if f.get('vcodec') == 'none']
1080 if audio_formats:
1081 yield audio_formats[-1]
1082 elif format_spec == 'worstaudio':
1083 audio_formats = [
1084 f for f in formats
1085 if f.get('vcodec') == 'none']
1086 if audio_formats:
1087 yield audio_formats[0]
1088 elif format_spec == 'bestvideo':
1089 video_formats = [
1090 f for f in formats
1091 if f.get('acodec') == 'none']
1092 if video_formats:
1093 yield video_formats[-1]
1094 elif format_spec == 'worstvideo':
1095 video_formats = [
1096 f for f in formats
1097 if f.get('acodec') == 'none']
1098 if video_formats:
1099 yield video_formats[0]
1100 else:
1101 extensions = ['mp4', 'flv', 'webm', '3gp', 'm4a', 'mp3', 'ogg', 'aac', 'wav']
1102 if format_spec in extensions:
1103 filter_f = lambda f: f['ext'] == format_spec
1104 else:
1105 filter_f = lambda f: f['format_id'] == format_spec
1106 matches = list(filter(filter_f, formats))
1107 if matches:
1108 yield matches[-1]
1109 elif selector.type == MERGE:
1110 def _merge(formats_info):
1111 format_1, format_2 = [f['format_id'] for f in formats_info]
1112 # The first format must contain the video and the
1113 # second the audio
1114 if formats_info[0].get('vcodec') == 'none':
1115 self.report_error('The first format must '
1116 'contain the video, try using '
1117 '"-f %s+%s"' % (format_2, format_1))
1118 return
3d24bbfb
S
1119 # Formats must be opposite (video+audio)
1120 if formats_info[0].get('acodec') == 'none' and formats_info[1].get('acodec') == 'none':
1121 self.report_error(
1122 'Both formats %s and %s are video-only, you must specify "-f video+audio"'
1123 % (format_1, format_2))
1124 return
67134eab
JMF
1125 output_ext = (
1126 formats_info[0]['ext']
1127 if self.params.get('merge_output_format') is None
1128 else self.params['merge_output_format'])
1129 return {
1130 'requested_formats': formats_info,
1131 'format': '%s+%s' % (formats_info[0].get('format'),
1132 formats_info[1].get('format')),
1133 'format_id': '%s+%s' % (formats_info[0].get('format_id'),
1134 formats_info[1].get('format_id')),
1135 'width': formats_info[0].get('width'),
1136 'height': formats_info[0].get('height'),
1137 'resolution': formats_info[0].get('resolution'),
1138 'fps': formats_info[0].get('fps'),
1139 'vcodec': formats_info[0].get('vcodec'),
1140 'vbr': formats_info[0].get('vbr'),
1141 'stretched_ratio': formats_info[0].get('stretched_ratio'),
1142 'acodec': formats_info[1].get('acodec'),
1143 'abr': formats_info[1].get('abr'),
1144 'ext': output_ext,
1145 }
1146 video_selector, audio_selector = map(_build_selector_function, selector.selector)
083c9df9 1147
67134eab
JMF
1148 def selector_function(formats):
1149 formats = list(formats)
1150 for pair in itertools.product(video_selector(formats), audio_selector(formats)):
1151 yield _merge(pair)
083c9df9 1152
67134eab 1153 filters = [self._build_format_filter(f) for f in selector.filters]
083c9df9 1154
67134eab
JMF
1155 def final_selector(formats):
1156 for _filter in filters:
1157 formats = list(filter(_filter, formats))
1158 return selector_function(formats)
1159 return final_selector
083c9df9 1160
67134eab 1161 stream = io.BytesIO(format_spec.encode('utf-8'))
0130afb7 1162 try:
232541df 1163 tokens = list(_remove_unused_ops(compat_tokenize_tokenize(stream.readline)))
0130afb7
JMF
1164 except tokenize.TokenError:
1165 raise syntax_error('Missing closing/opening brackets or parenthesis', (0, len(format_spec)))
1166
1167 class TokenIterator(object):
1168 def __init__(self, tokens):
1169 self.tokens = tokens
1170 self.counter = 0
1171
1172 def __iter__(self):
1173 return self
1174
1175 def __next__(self):
1176 if self.counter >= len(self.tokens):
1177 raise StopIteration()
1178 value = self.tokens[self.counter]
1179 self.counter += 1
1180 return value
1181
1182 next = __next__
1183
1184 def restore_last_token(self):
1185 self.counter -= 1
1186
1187 parsed_selector = _parse_format_selection(iter(TokenIterator(tokens)))
67134eab 1188 return _build_selector_function(parsed_selector)
a9c58ad9 1189
e5660ee6
JMF
1190 def _calc_headers(self, info_dict):
1191 res = std_headers.copy()
1192
1193 add_headers = info_dict.get('http_headers')
1194 if add_headers:
1195 res.update(add_headers)
1196
1197 cookies = self._calc_cookies(info_dict)
1198 if cookies:
1199 res['Cookie'] = cookies
1200
1201 return res
1202
1203 def _calc_cookies(self, info_dict):
5c2266df 1204 pr = sanitized_Request(info_dict['url'])
e5660ee6 1205 self.cookiejar.add_cookie_header(pr)
662435f7 1206 return pr.get_header('Cookie')
e5660ee6 1207
dd82ffea
JMF
1208 def process_video_result(self, info_dict, download=True):
1209 assert info_dict.get('_type', 'video') == 'video'
1210
bec1fad2
PH
1211 if 'id' not in info_dict:
1212 raise ExtractorError('Missing "id" field in extractor result')
1213 if 'title' not in info_dict:
1214 raise ExtractorError('Missing "title" field in extractor result')
1215
dd82ffea
JMF
1216 if 'playlist' not in info_dict:
1217 # It isn't part of a playlist
1218 info_dict['playlist'] = None
1219 info_dict['playlist_index'] = None
1220
d5519808 1221 thumbnails = info_dict.get('thumbnails')
cfb56d1a
PH
1222 if thumbnails is None:
1223 thumbnail = info_dict.get('thumbnail')
1224 if thumbnail:
a7a14d95 1225 info_dict['thumbnails'] = thumbnails = [{'url': thumbnail}]
d5519808 1226 if thumbnails:
be6d7229 1227 thumbnails.sort(key=lambda t: (
cfb56d1a
PH
1228 t.get('preference'), t.get('width'), t.get('height'),
1229 t.get('id'), t.get('url')))
f6c24009 1230 for i, t in enumerate(thumbnails):
9603e8a7 1231 if t.get('width') and t.get('height'):
d5519808 1232 t['resolution'] = '%dx%d' % (t['width'], t['height'])
f6c24009
PH
1233 if t.get('id') is None:
1234 t['id'] = '%d' % i
d5519808
PH
1235
1236 if thumbnails and 'thumbnail' not in info_dict:
1237 info_dict['thumbnail'] = thumbnails[-1]['url']
1238
c9ae7b95 1239 if 'display_id' not in info_dict and 'id' in info_dict:
0afef30b
PH
1240 info_dict['display_id'] = info_dict['id']
1241
955c4514 1242 if info_dict.get('upload_date') is None and info_dict.get('timestamp') is not None:
a55e36f4
S
1243 # Working around out-of-range timestamp values (e.g. negative ones on Windows,
1244 # see http://bugs.python.org/issue1646728)
1245 try:
1246 upload_date = datetime.datetime.utcfromtimestamp(info_dict['timestamp'])
1247 info_dict['upload_date'] = upload_date.strftime('%Y%m%d')
1248 except (ValueError, OverflowError, OSError):
1249 pass
9d2ecdbc 1250
33d2fc2f
S
1251 # Auto generate title fields corresponding to the *_number fields when missing
1252 # in order to always have clean titles. This is very common for TV series.
1253 for field in ('chapter', 'season', 'episode'):
1254 if info_dict.get('%s_number' % field) is not None and not info_dict.get(field):
1255 info_dict[field] = '%s %d' % (field.capitalize(), info_dict['%s_number' % field])
1256
4bba3716
S
1257 subtitles = info_dict.get('subtitles')
1258 if subtitles:
1259 for _, subtitle in subtitles.items():
1260 for subtitle_format in subtitle:
1261 if 'ext' not in subtitle_format:
1262 subtitle_format['ext'] = determine_ext(subtitle_format['url']).lower()
1263
a504ced0 1264 if self.params.get('listsubtitles', False):
360e1ca5
JMF
1265 if 'automatic_captions' in info_dict:
1266 self.list_subtitles(info_dict['id'], info_dict.get('automatic_captions'), 'automatic captions')
4bba3716 1267 self.list_subtitles(info_dict['id'], subtitles, 'subtitles')
a504ced0 1268 return
360e1ca5 1269 info_dict['requested_subtitles'] = self.process_subtitles(
4bba3716 1270 info_dict['id'], subtitles,
360e1ca5 1271 info_dict.get('automatic_captions'))
a504ced0 1272
dd82ffea
JMF
1273 # We now pick which formats have to be downloaded
1274 if info_dict.get('formats') is None:
1275 # There's only one format available
1276 formats = [info_dict]
1277 else:
1278 formats = info_dict['formats']
1279
db95dc13
PH
1280 if not formats:
1281 raise ExtractorError('No video formats found!')
1282
181c7053
S
1283 formats_dict = {}
1284
dd82ffea 1285 # We check that all the formats have the format and format_id fields
db95dc13 1286 for i, format in enumerate(formats):
bec1fad2
PH
1287 if 'url' not in format:
1288 raise ExtractorError('Missing "url" key in result (index %d)' % i)
1289
dd82ffea 1290 if format.get('format_id') is None:
8016c922 1291 format['format_id'] = compat_str(i)
181c7053
S
1292 format_id = format['format_id']
1293 if format_id not in formats_dict:
1294 formats_dict[format_id] = []
1295 formats_dict[format_id].append(format)
1296
1297 # Make sure all formats have unique format_id
1298 for format_id, ambiguous_formats in formats_dict.items():
1299 if len(ambiguous_formats) > 1:
1300 for i, format in enumerate(ambiguous_formats):
1301 format['format_id'] = '%s-%d' % (format_id, i)
1302
1303 for i, format in enumerate(formats):
8c51aa65 1304 if format.get('format') is None:
6febd1c1 1305 format['format'] = '{id} - {res}{note}'.format(
8c51aa65
JMF
1306 id=format['format_id'],
1307 res=self.format_resolution(format),
6febd1c1 1308 note=' ({0})'.format(format['format_note']) if format.get('format_note') is not None else '',
8c51aa65 1309 )
c1002e96
PH
1310 # Automatically determine file extension if missing
1311 if 'ext' not in format:
cce929ea 1312 format['ext'] = determine_ext(format['url']).lower()
b5559424
S
1313 # Automatically determine protocol if missing (useful for format
1314 # selection purposes)
1315 if 'protocol' not in format:
1316 format['protocol'] = determine_protocol(format)
e5660ee6
JMF
1317 # Add HTTP headers, so that external programs can use them from the
1318 # json output
1319 full_format_info = info_dict.copy()
1320 full_format_info.update(format)
1321 format['http_headers'] = self._calc_headers(full_format_info)
dd82ffea 1322
4bcc7bd1 1323 # TODO Central sorting goes here
99e206d5 1324
f89197d7 1325 if formats[0] is not info_dict:
b3d9ef88
JMF
1326 # only set the 'formats' fields if the original info_dict list them
1327 # otherwise we end up with a circular reference, the first (and unique)
f89197d7 1328 # element in the 'formats' field in info_dict is info_dict itself,
dfb1b146 1329 # which can't be exported to json
b3d9ef88 1330 info_dict['formats'] = formats
cfb56d1a 1331 if self.params.get('listformats'):
bfaae0a7 1332 self.list_formats(info_dict)
1333 return
cfb56d1a
PH
1334 if self.params.get('list_thumbnails'):
1335 self.list_thumbnails(info_dict)
1336 return
bfaae0a7 1337
de3ef3ed 1338 req_format = self.params.get('format')
a9c58ad9 1339 if req_format is None:
feccf29c 1340 req_format_list = []
3749e36e 1341 if (self.params.get('outtmpl', DEFAULT_OUTTMPL) != '-' and
8250c32f
JMF
1342 info_dict['extractor'] in ['youtube', 'ted'] and
1343 not info_dict.get('is_live')):
7fcb605b 1344 merger = FFmpegMergerPP(self)
97fcf1bb 1345 if merger.available and merger.can_merge():
7fcb605b 1346 req_format_list.append('bestvideo+bestaudio')
feccf29c
S
1347 req_format_list.append('best')
1348 req_format = '/'.join(req_format_list)
5acfa126
JMF
1349 format_selector = self.build_format_selector(req_format)
1350 formats_to_download = list(format_selector(formats))
dd82ffea 1351 if not formats_to_download:
6febd1c1 1352 raise ExtractorError('requested format not available',
78a3a9f8 1353 expected=True)
dd82ffea
JMF
1354
1355 if download:
1356 if len(formats_to_download) > 1:
6febd1c1 1357 self.to_screen('[info] %s: downloading video in %s formats' % (info_dict['id'], len(formats_to_download)))
dd82ffea
JMF
1358 for format in formats_to_download:
1359 new_info = dict(info_dict)
1360 new_info.update(format)
1361 self.process_info(new_info)
1362 # We update the info dict with the best quality format (backwards compatibility)
1363 info_dict.update(formats_to_download[-1])
1364 return info_dict
1365
98c70d6f 1366 def process_subtitles(self, video_id, normal_subtitles, automatic_captions):
a504ced0 1367 """Select the requested subtitles and their format"""
98c70d6f
JMF
1368 available_subs = {}
1369 if normal_subtitles and self.params.get('writesubtitles'):
1370 available_subs.update(normal_subtitles)
1371 if automatic_captions and self.params.get('writeautomaticsub'):
1372 for lang, cap_info in automatic_captions.items():
360e1ca5
JMF
1373 if lang not in available_subs:
1374 available_subs[lang] = cap_info
1375
4d171848
JMF
1376 if (not self.params.get('writesubtitles') and not
1377 self.params.get('writeautomaticsub') or not
1378 available_subs):
1379 return None
a504ced0
JMF
1380
1381 if self.params.get('allsubtitles', False):
1382 requested_langs = available_subs.keys()
1383 else:
1384 if self.params.get('subtitleslangs', False):
1385 requested_langs = self.params.get('subtitleslangs')
1386 elif 'en' in available_subs:
1387 requested_langs = ['en']
1388 else:
1389 requested_langs = [list(available_subs.keys())[0]]
1390
1391 formats_query = self.params.get('subtitlesformat', 'best')
1392 formats_preference = formats_query.split('/') if formats_query else []
1393 subs = {}
1394 for lang in requested_langs:
1395 formats = available_subs.get(lang)
1396 if formats is None:
1397 self.report_warning('%s subtitles not available for %s' % (lang, video_id))
1398 continue
a504ced0
JMF
1399 for ext in formats_preference:
1400 if ext == 'best':
1401 f = formats[-1]
1402 break
1403 matches = list(filter(lambda f: f['ext'] == ext, formats))
1404 if matches:
1405 f = matches[-1]
1406 break
1407 else:
1408 f = formats[-1]
1409 self.report_warning(
1410 'No subtitle format found matching "%s" for language %s, '
1411 'using %s' % (formats_query, lang, f['ext']))
1412 subs[lang] = f
1413 return subs
1414
8222d8de
JMF
1415 def process_info(self, info_dict):
1416 """Process a single resolved IE result."""
1417
1418 assert info_dict.get('_type', 'video') == 'video'
fd288278
PH
1419
1420 max_downloads = self.params.get('max_downloads')
1421 if max_downloads is not None:
1422 if self._num_downloads >= int(max_downloads):
1423 raise MaxDownloadsReached()
8222d8de
JMF
1424
1425 info_dict['fulltitle'] = info_dict['title']
1426 if len(info_dict['title']) > 200:
6febd1c1 1427 info_dict['title'] = info_dict['title'][:197] + '...'
8222d8de 1428
11b85ce6 1429 if 'format' not in info_dict:
8222d8de
JMF
1430 info_dict['format'] = info_dict['ext']
1431
442c37b7 1432 reason = self._match_entry(info_dict, incomplete=False)
8222d8de 1433 if reason is not None:
6febd1c1 1434 self.to_screen('[download] ' + reason)
8222d8de
JMF
1435 return
1436
fd288278 1437 self._num_downloads += 1
8222d8de 1438
e72c7e41 1439 info_dict['_filename'] = filename = self.prepare_filename(info_dict)
8222d8de
JMF
1440
1441 # Forced printings
1442 if self.params.get('forcetitle', False):
0783b09b 1443 self.to_stdout(info_dict['fulltitle'])
8222d8de 1444 if self.params.get('forceid', False):
0783b09b 1445 self.to_stdout(info_dict['id'])
8222d8de 1446 if self.params.get('forceurl', False):
16ae61f6 1447 if info_dict.get('requested_formats') is not None:
1448 for f in info_dict['requested_formats']:
1449 self.to_stdout(f['url'] + f.get('play_path', ''))
1450 else:
1451 # For RTMP URLs, also include the playpath
1452 self.to_stdout(info_dict['url'] + info_dict.get('play_path', ''))
216d71d0 1453 if self.params.get('forcethumbnail', False) and info_dict.get('thumbnail') is not None:
0783b09b 1454 self.to_stdout(info_dict['thumbnail'])
216d71d0 1455 if self.params.get('forcedescription', False) and info_dict.get('description') is not None:
0783b09b 1456 self.to_stdout(info_dict['description'])
8222d8de 1457 if self.params.get('forcefilename', False) and filename is not None:
0783b09b 1458 self.to_stdout(filename)
525ef922
PH
1459 if self.params.get('forceduration', False) and info_dict.get('duration') is not None:
1460 self.to_stdout(formatSeconds(info_dict['duration']))
8222d8de 1461 if self.params.get('forceformat', False):
0783b09b 1462 self.to_stdout(info_dict['format'])
9d153818 1463 if self.params.get('forcejson', False):
0783b09b 1464 self.to_stdout(json.dumps(info_dict))
8222d8de
JMF
1465
1466 # Do nothing else if in simulate mode
1467 if self.params.get('simulate', False):
1468 return
1469
1470 if filename is None:
1471 return
1472
1473 try:
e5a11a22 1474 dn = os.path.dirname(sanitize_path(encodeFilename(filename)))
d26e981d 1475 if dn and not os.path.exists(dn):
8222d8de
JMF
1476 os.makedirs(dn)
1477 except (OSError, IOError) as err:
9b9c5355 1478 self.report_error('unable to create directory ' + error_to_compat_str(err))
8222d8de
JMF
1479 return
1480
1481 if self.params.get('writedescription', False):
2699da80 1482 descfn = replace_extension(filename, 'description', info_dict.get('ext'))
7b6fefc9 1483 if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(descfn)):
6febd1c1 1484 self.to_screen('[info] Video description is already present')
f00fd51d
JMF
1485 elif info_dict.get('description') is None:
1486 self.report_warning('There\'s no description to write.')
7b6fefc9
PH
1487 else:
1488 try:
6febd1c1 1489 self.to_screen('[info] Writing video description to: ' + descfn)
7b6fefc9
PH
1490 with io.open(encodeFilename(descfn), 'w', encoding='utf-8') as descfile:
1491 descfile.write(info_dict['description'])
7b6fefc9 1492 except (OSError, IOError):
6febd1c1 1493 self.report_error('Cannot write description file ' + descfn)
7b6fefc9 1494 return
8222d8de 1495
1fb07d10 1496 if self.params.get('writeannotations', False):
98727e12 1497 annofn = replace_extension(filename, 'annotations.xml', info_dict.get('ext'))
7b6fefc9 1498 if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(annofn)):
6febd1c1 1499 self.to_screen('[info] Video annotations are already present')
7b6fefc9
PH
1500 else:
1501 try:
6febd1c1 1502 self.to_screen('[info] Writing video annotations to: ' + annofn)
7b6fefc9
PH
1503 with io.open(encodeFilename(annofn), 'w', encoding='utf-8') as annofile:
1504 annofile.write(info_dict['annotations'])
1505 except (KeyError, TypeError):
6febd1c1 1506 self.report_warning('There are no annotations to write.')
7b6fefc9 1507 except (OSError, IOError):
6febd1c1 1508 self.report_error('Cannot write annotations file: ' + annofn)
7b6fefc9 1509 return
1fb07d10 1510
c4a91be7 1511 subtitles_are_requested = any([self.params.get('writesubtitles', False),
0b7f3118 1512 self.params.get('writeautomaticsub')])
c4a91be7 1513
c84dd8a9 1514 if subtitles_are_requested and info_dict.get('requested_subtitles'):
8222d8de
JMF
1515 # subtitles download errors are already managed as troubles in relevant IE
1516 # that way it will silently go on when used with unsupporting IE
c84dd8a9 1517 subtitles = info_dict['requested_subtitles']
0f2c0d33 1518 ie = self.get_info_extractor(info_dict['extractor_key'])
a504ced0
JMF
1519 for sub_lang, sub_info in subtitles.items():
1520 sub_format = sub_info['ext']
1521 if sub_info.get('data') is not None:
1522 sub_data = sub_info['data']
1523 else:
1524 try:
0f2c0d33
JMF
1525 sub_data = ie._download_webpage(
1526 sub_info['url'], info_dict['id'], note=False)
1527 except ExtractorError as err:
a504ced0 1528 self.report_warning('Unable to download subtitle for "%s": %s' %
9b9c5355 1529 (sub_lang, error_to_compat_str(err.cause)))
a504ced0 1530 continue
8222d8de 1531 try:
d4051a8e 1532 sub_filename = subtitles_filename(filename, sub_lang, sub_format)
7b6fefc9 1533 if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(sub_filename)):
6febd1c1 1534 self.to_screen('[info] Video subtitle %s.%s is already_present' % (sub_lang, sub_format))
7b6fefc9 1535 else:
6febd1c1 1536 self.to_screen('[info] Writing video subtitles to: ' + sub_filename)
7b6fefc9 1537 with io.open(encodeFilename(sub_filename), 'w', encoding='utf-8') as subfile:
a504ced0 1538 subfile.write(sub_data)
8222d8de 1539 except (OSError, IOError):
e4db1951 1540 self.report_error('Cannot write subtitles file ' + sub_filename)
8222d8de
JMF
1541 return
1542
8222d8de 1543 if self.params.get('writeinfojson', False):
b29e0000 1544 infofn = replace_extension(filename, 'info.json', info_dict.get('ext'))
7b6fefc9 1545 if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(infofn)):
6febd1c1 1546 self.to_screen('[info] Video description metadata is already present')
7b6fefc9 1547 else:
6febd1c1 1548 self.to_screen('[info] Writing video description metadata as JSON to: ' + infofn)
7b6fefc9 1549 try:
cb202fd2 1550 write_json_file(self.filter_requested_info(info_dict), infofn)
7b6fefc9 1551 except (OSError, IOError):
6febd1c1 1552 self.report_error('Cannot write metadata to JSON file ' + infofn)
7b6fefc9 1553 return
8222d8de 1554
ec82d85a 1555 self._write_thumbnails(info_dict, filename)
8222d8de
JMF
1556
1557 if not self.params.get('skip_download', False):
4340deca
P
1558 try:
1559 def dl(name, info):
a055469f 1560 fd = get_suitable_downloader(info, self.params)(self, self.params)
4340deca
P
1561 for ph in self._progress_hooks:
1562 fd.add_progress_hook(ph)
1563 if self.params.get('verbose'):
1564 self.to_stdout('[debug] Invoking downloader on %r' % info.get('url'))
1565 return fd.download(name, info)
ee69b99a 1566
4340deca
P
1567 if info_dict.get('requested_formats') is not None:
1568 downloaded = []
1569 success = True
d47aeb22 1570 merger = FFmpegMergerPP(self)
f740fae2 1571 if not merger.available:
4340deca
P
1572 postprocessors = []
1573 self.report_warning('You have requested multiple '
1574 'formats but ffmpeg or avconv are not installed.'
4a5a898a 1575 ' The formats won\'t be merged.')
6350728b 1576 else:
4340deca 1577 postprocessors = [merger]
81cd954a
S
1578
1579 def compatible_formats(formats):
1580 video, audio = formats
1581 # Check extension
1582 video_ext, audio_ext = audio.get('ext'), video.get('ext')
1583 if video_ext and audio_ext:
1584 COMPATIBLE_EXTS = (
6728187a 1585 ('mp3', 'mp4', 'm4a', 'm4p', 'm4b', 'm4r', 'm4v'),
81cd954a
S
1586 ('webm')
1587 )
1588 for exts in COMPATIBLE_EXTS:
1589 if video_ext in exts and audio_ext in exts:
1590 return True
1591 # TODO: Check acodec/vcodec
1592 return False
1593
38c6902b
S
1594 filename_real_ext = os.path.splitext(filename)[1][1:]
1595 filename_wo_ext = (
1596 os.path.splitext(filename)[0]
1597 if filename_real_ext == info_dict['ext']
1598 else filename)
81cd954a 1599 requested_formats = info_dict['requested_formats']
c0dea0a7 1600 if self.params.get('merge_output_format') is None and not compatible_formats(requested_formats):
38c6902b 1601 info_dict['ext'] = 'mkv'
4a5a898a
S
1602 self.report_warning(
1603 'Requested formats are incompatible for merge and will be merged into mkv.')
38c6902b
S
1604 # Ensure filename always has a correct extension for successful merge
1605 filename = '%s.%s' % (filename_wo_ext, info_dict['ext'])
5b5fbc08
JMF
1606 if os.path.exists(encodeFilename(filename)):
1607 self.to_screen(
1608 '[download] %s has already been downloaded and '
1609 'merged' % filename)
1610 else:
81cd954a 1611 for f in requested_formats:
5b5fbc08
JMF
1612 new_info = dict(info_dict)
1613 new_info.update(f)
1614 fname = self.prepare_filename(new_info)
666a9a2b 1615 fname = prepend_extension(fname, 'f%s' % f['format_id'], new_info['ext'])
5b5fbc08
JMF
1616 downloaded.append(fname)
1617 partial_success = dl(fname, new_info)
1618 success = success and partial_success
1619 info_dict['__postprocessors'] = postprocessors
1620 info_dict['__files_to_merge'] = downloaded
4340deca
P
1621 else:
1622 # Just a single file
1623 success = dl(filename, info_dict)
1624 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
1625 self.report_error('unable to download video data: %s' % str(err))
1626 return
1627 except (OSError, IOError) as err:
1628 raise UnavailableVideoError(err)
1629 except (ContentTooShortError, ) as err:
1630 self.report_error('content too short (expected %s bytes and served %s)' % (err.expected, err.downloaded))
1631 return
8222d8de
JMF
1632
1633 if success:
6271f1ca 1634 # Fixup content
62cd676c
PH
1635 fixup_policy = self.params.get('fixup')
1636 if fixup_policy is None:
1637 fixup_policy = 'detect_or_warn'
1638
6271f1ca
PH
1639 stretched_ratio = info_dict.get('stretched_ratio')
1640 if stretched_ratio is not None and stretched_ratio != 1:
6271f1ca
PH
1641 if fixup_policy == 'warn':
1642 self.report_warning('%s: Non-uniform pixel ratio (%s)' % (
1643 info_dict['id'], stretched_ratio))
1644 elif fixup_policy == 'detect_or_warn':
1645 stretched_pp = FFmpegFixupStretchedPP(self)
1646 if stretched_pp.available:
1647 info_dict.setdefault('__postprocessors', [])
1648 info_dict['__postprocessors'].append(stretched_pp)
1649 else:
1650 self.report_warning(
1651 '%s: Non-uniform pixel ratio (%s). Install ffmpeg or avconv to fix this automatically.' % (
1652 info_dict['id'], stretched_ratio))
1653 else:
62cd676c
PH
1654 assert fixup_policy in ('ignore', 'never')
1655
1656 if info_dict.get('requested_formats') is None and info_dict.get('container') == 'm4a_dash':
1657 if fixup_policy == 'warn':
1658 self.report_warning('%s: writing DASH m4a. Only some players support this container.' % (
1659 info_dict['id']))
1660 elif fixup_policy == 'detect_or_warn':
1661 fixup_pp = FFmpegFixupM4aPP(self)
1662 if fixup_pp.available:
1663 info_dict.setdefault('__postprocessors', [])
1664 info_dict['__postprocessors'].append(fixup_pp)
1665 else:
1666 self.report_warning(
1667 '%s: writing DASH m4a. Only some players support this container. Install ffmpeg or avconv to fix this automatically.' % (
1668 info_dict['id']))
1669 else:
1670 assert fixup_policy in ('ignore', 'never')
6271f1ca 1671
8222d8de
JMF
1672 try:
1673 self.post_process(filename, info_dict)
1674 except (PostProcessingError) as err:
6febd1c1 1675 self.report_error('postprocessing: %s' % str(err))
8222d8de 1676 return
cd58dc3e 1677 self.record_download_archive(info_dict)
8222d8de
JMF
1678
1679 def download(self, url_list):
1680 """Download a given list of URLs."""
acd69589 1681 outtmpl = self.params.get('outtmpl', DEFAULT_OUTTMPL)
0c75c3fa 1682 if (len(url_list) > 1 and
8fb3ac36
PH
1683 '%' not in outtmpl and
1684 self.params.get('max_downloads') != 1):
acd69589 1685 raise SameFileError(outtmpl)
8222d8de
JMF
1686
1687 for url in url_list:
1688 try:
5f6a1245 1689 # It also downloads the videos
61aa5ba3
S
1690 res = self.extract_info(
1691 url, force_generic_extractor=self.params.get('force_generic_extractor', False))
8222d8de 1692 except UnavailableVideoError:
6febd1c1 1693 self.report_error('unable to download video')
8222d8de 1694 except MaxDownloadsReached:
6febd1c1 1695 self.to_screen('[info] Maximum number of downloaded files reached.')
8222d8de 1696 raise
63e0be34
PH
1697 else:
1698 if self.params.get('dump_single_json', False):
1699 self.to_stdout(json.dumps(res))
8222d8de
JMF
1700
1701 return self._download_retcode
1702
1dcc4c0c 1703 def download_with_info_file(self, info_filename):
31bd3925
JMF
1704 with contextlib.closing(fileinput.FileInput(
1705 [info_filename], mode='r',
1706 openhook=fileinput.hook_encoded('utf-8'))) as f:
1707 # FileInput doesn't have a read method, we can't call json.load
cb202fd2 1708 info = self.filter_requested_info(json.loads('\n'.join(f)))
d4943898
JMF
1709 try:
1710 self.process_ie_result(info, download=True)
1711 except DownloadError:
1712 webpage_url = info.get('webpage_url')
1713 if webpage_url is not None:
6febd1c1 1714 self.report_warning('The info failed to download, trying with "%s"' % webpage_url)
d4943898
JMF
1715 return self.download([webpage_url])
1716 else:
1717 raise
1718 return self._download_retcode
1dcc4c0c 1719
cb202fd2
S
1720 @staticmethod
1721 def filter_requested_info(info_dict):
1722 return dict(
1723 (k, v) for k, v in info_dict.items()
1724 if k not in ['requested_formats', 'requested_subtitles'])
1725
8222d8de
JMF
1726 def post_process(self, filename, ie_info):
1727 """Run all the postprocessors on the given file."""
1728 info = dict(ie_info)
1729 info['filepath'] = filename
6350728b
JMF
1730 pps_chain = []
1731 if ie_info.get('__postprocessors') is not None:
1732 pps_chain.extend(ie_info['__postprocessors'])
1733 pps_chain.extend(self._pps)
1734 for pp in pps_chain:
71646e46 1735 files_to_delete = []
8222d8de 1736 try:
592e97e8 1737 files_to_delete, info = pp.run(info)
8222d8de 1738 except PostProcessingError as e:
bbcbf4d4 1739 self.report_error(e.msg)
592e97e8
JMF
1740 if files_to_delete and not self.params.get('keepvideo', False):
1741 for old_filename in files_to_delete:
f3ff1a36 1742 self.to_screen('Deleting original file %s (pass -k to keep)' % old_filename)
592e97e8
JMF
1743 try:
1744 os.remove(encodeFilename(old_filename))
1745 except (IOError, OSError):
1746 self.report_warning('Unable to remove downloaded original file')
c1c9a79c 1747
5db07df6
PH
1748 def _make_archive_id(self, info_dict):
1749 # Future-proof against any change in case
1750 # and backwards compatibility with prior versions
d31209a1 1751 extractor = info_dict.get('extractor_key')
7012b23c
PH
1752 if extractor is None:
1753 if 'id' in info_dict:
1754 extractor = info_dict.get('ie_key') # key in a playlist
1755 if extractor is None:
5db07df6 1756 return None # Incomplete video information
6febd1c1 1757 return extractor.lower() + ' ' + info_dict['id']
5db07df6
PH
1758
1759 def in_download_archive(self, info_dict):
1760 fn = self.params.get('download_archive')
1761 if fn is None:
1762 return False
1763
1764 vid_id = self._make_archive_id(info_dict)
1765 if vid_id is None:
7012b23c 1766 return False # Incomplete video information
5db07df6 1767
c1c9a79c
PH
1768 try:
1769 with locked_file(fn, 'r', encoding='utf-8') as archive_file:
1770 for line in archive_file:
1771 if line.strip() == vid_id:
1772 return True
1773 except IOError as ioe:
1774 if ioe.errno != errno.ENOENT:
1775 raise
1776 return False
1777
1778 def record_download_archive(self, info_dict):
1779 fn = self.params.get('download_archive')
1780 if fn is None:
1781 return
5db07df6
PH
1782 vid_id = self._make_archive_id(info_dict)
1783 assert vid_id
c1c9a79c 1784 with locked_file(fn, 'a', encoding='utf-8') as archive_file:
6febd1c1 1785 archive_file.write(vid_id + '\n')
dd82ffea 1786
8c51aa65 1787 @staticmethod
8abeeb94 1788 def format_resolution(format, default='unknown'):
fb04e403
PH
1789 if format.get('vcodec') == 'none':
1790 return 'audio only'
f49d89ee
PH
1791 if format.get('resolution') is not None:
1792 return format['resolution']
8c51aa65
JMF
1793 if format.get('height') is not None:
1794 if format.get('width') is not None:
6febd1c1 1795 res = '%sx%s' % (format['width'], format['height'])
8c51aa65 1796 else:
6febd1c1 1797 res = '%sp' % format['height']
f49d89ee 1798 elif format.get('width') is not None:
6febd1c1 1799 res = '?x%d' % format['width']
8c51aa65 1800 else:
8abeeb94 1801 res = default
8c51aa65
JMF
1802 return res
1803
c57f7757
PH
1804 def _format_note(self, fdict):
1805 res = ''
1806 if fdict.get('ext') in ['f4f', 'f4m']:
1807 res += '(unsupported) '
32f90364
PH
1808 if fdict.get('language'):
1809 if res:
1810 res += ' '
1811 res += '[%s]' % fdict['language']
c57f7757
PH
1812 if fdict.get('format_note') is not None:
1813 res += fdict['format_note'] + ' '
1814 if fdict.get('tbr') is not None:
1815 res += '%4dk ' % fdict['tbr']
1816 if fdict.get('container') is not None:
1817 if res:
1818 res += ', '
1819 res += '%s container' % fdict['container']
1820 if (fdict.get('vcodec') is not None and
1821 fdict.get('vcodec') != 'none'):
1822 if res:
1823 res += ', '
1824 res += fdict['vcodec']
91c7271a 1825 if fdict.get('vbr') is not None:
c57f7757
PH
1826 res += '@'
1827 elif fdict.get('vbr') is not None and fdict.get('abr') is not None:
1828 res += 'video@'
1829 if fdict.get('vbr') is not None:
1830 res += '%4dk' % fdict['vbr']
fbb21cf5
PH
1831 if fdict.get('fps') is not None:
1832 res += ', %sfps' % fdict['fps']
c57f7757
PH
1833 if fdict.get('acodec') is not None:
1834 if res:
1835 res += ', '
1836 if fdict['acodec'] == 'none':
1837 res += 'video only'
1838 else:
1839 res += '%-5s' % fdict['acodec']
1840 elif fdict.get('abr') is not None:
1841 if res:
1842 res += ', '
1843 res += 'audio'
1844 if fdict.get('abr') is not None:
1845 res += '@%3dk' % fdict['abr']
1846 if fdict.get('asr') is not None:
1847 res += ' (%5dHz)' % fdict['asr']
1848 if fdict.get('filesize') is not None:
1849 if res:
1850 res += ', '
1851 res += format_bytes(fdict['filesize'])
9732d77e
PH
1852 elif fdict.get('filesize_approx') is not None:
1853 if res:
1854 res += ', '
1855 res += '~' + format_bytes(fdict['filesize_approx'])
c57f7757 1856 return res
91c7271a 1857
c57f7757 1858 def list_formats(self, info_dict):
94badb25 1859 formats = info_dict.get('formats', [info_dict])
b81a359e
PH
1860 table = [
1861 [f['format_id'], f['ext'], self.format_resolution(f), self._format_note(f)]
1862 for f in formats
e65566a9 1863 if f.get('preference') is None or f['preference'] >= -1000]
94badb25 1864 if len(formats) > 1:
b81a359e 1865 table[-1][-1] += (' ' if table[-1][-1] else '') + '(best)'
57dd9a8f 1866
b81a359e 1867 header_line = ['format code', 'extension', 'resolution', 'note']
cfb56d1a 1868 self.to_screen(
b81a359e
PH
1869 '[info] Available formats for %s:\n%s' %
1870 (info_dict['id'], render_table(header_line, table)))
cfb56d1a
PH
1871
1872 def list_thumbnails(self, info_dict):
1873 thumbnails = info_dict.get('thumbnails')
1874 if not thumbnails:
1875 tn_url = info_dict.get('thumbnail')
1876 if tn_url:
1877 thumbnails = [{'id': '0', 'url': tn_url}]
1878 else:
1879 self.to_screen(
1880 '[info] No thumbnails present for %s' % info_dict['id'])
1881 return
1882
1883 self.to_screen(
1884 '[info] Thumbnails for %s:' % info_dict['id'])
1885 self.to_screen(render_table(
1886 ['ID', 'width', 'height', 'URL'],
1887 [[t['id'], t.get('width', 'unknown'), t.get('height', 'unknown'), t['url']] for t in thumbnails]))
dca08720 1888
360e1ca5 1889 def list_subtitles(self, video_id, subtitles, name='subtitles'):
a504ced0 1890 if not subtitles:
360e1ca5 1891 self.to_screen('%s has no %s' % (video_id, name))
a504ced0 1892 return
a504ced0 1893 self.to_screen(
edab9dbf
JMF
1894 'Available %s for %s:' % (name, video_id))
1895 self.to_screen(render_table(
1896 ['Language', 'formats'],
1897 [[lang, ', '.join(f['ext'] for f in reversed(formats))]
1898 for lang, formats in subtitles.items()]))
a504ced0 1899
dca08720
PH
1900 def urlopen(self, req):
1901 """ Start an HTTP download """
82d8a8b6 1902 if isinstance(req, compat_basestring):
67dda517 1903 req = sanitized_Request(req)
19a41fc6 1904 return self._opener.open(req, timeout=self._socket_timeout)
dca08720
PH
1905
1906 def print_debug_header(self):
1907 if not self.params.get('verbose'):
1908 return
62fec3b2 1909
4192b51c
PH
1910 if type('') is not compat_str:
1911 # Python 2.6 on SLES11 SP1 (https://github.com/rg3/youtube-dl/issues/3326)
1912 self.report_warning(
1913 'Your Python is broken! Update to a newer and supported version')
1914
c6afed48
PH
1915 stdout_encoding = getattr(
1916 sys.stdout, 'encoding', 'missing (%s)' % type(sys.stdout).__name__)
b0472057 1917 encoding_str = (
734f90bb
PH
1918 '[debug] Encodings: locale %s, fs %s, out %s, pref %s\n' % (
1919 locale.getpreferredencoding(),
1920 sys.getfilesystemencoding(),
c6afed48 1921 stdout_encoding,
b0472057 1922 self.get_encoding()))
4192b51c 1923 write_string(encoding_str, encoding=None)
734f90bb
PH
1924
1925 self._write_string('[debug] youtube-dl version ' + __version__ + '\n')
dca08720
PH
1926 try:
1927 sp = subprocess.Popen(
1928 ['git', 'rev-parse', '--short', 'HEAD'],
1929 stdout=subprocess.PIPE, stderr=subprocess.PIPE,
1930 cwd=os.path.dirname(os.path.abspath(__file__)))
1931 out, err = sp.communicate()
1932 out = out.decode().strip()
1933 if re.match('[0-9a-f]+', out):
734f90bb 1934 self._write_string('[debug] Git HEAD: ' + out + '\n')
70a1165b 1935 except Exception:
dca08720
PH
1936 try:
1937 sys.exc_clear()
70a1165b 1938 except Exception:
dca08720 1939 pass
d28b5171
PH
1940 self._write_string('[debug] Python version %s - %s\n' % (
1941 platform.python_version(), platform_name()))
1942
73fac4e9 1943 exe_versions = FFmpegPostProcessor.get_versions(self)
4c83c967 1944 exe_versions['rtmpdump'] = rtmpdump_version()
d28b5171
PH
1945 exe_str = ', '.join(
1946 '%s %s' % (exe, v)
1947 for exe, v in sorted(exe_versions.items())
1948 if v
1949 )
1950 if not exe_str:
1951 exe_str = 'none'
1952 self._write_string('[debug] exe versions: %s\n' % exe_str)
dca08720
PH
1953
1954 proxy_map = {}
1955 for handler in self._opener.handlers:
1956 if hasattr(handler, 'proxies'):
1957 proxy_map.update(handler.proxies)
734f90bb 1958 self._write_string('[debug] Proxy map: ' + compat_str(proxy_map) + '\n')
dca08720 1959
58b1f00d
PH
1960 if self.params.get('call_home', False):
1961 ipaddr = self.urlopen('https://yt-dl.org/ip').read().decode('utf-8')
1962 self._write_string('[debug] Public IP address: %s\n' % ipaddr)
1963 latest_version = self.urlopen(
1964 'https://yt-dl.org/latest/version').read().decode('utf-8')
1965 if version_tuple(latest_version) > version_tuple(__version__):
1966 self.report_warning(
1967 'You are using an outdated version (newest version: %s)! '
1968 'See https://yt-dl.org/update if you need help updating.' %
1969 latest_version)
1970
e344693b 1971 def _setup_opener(self):
6ad14cab 1972 timeout_val = self.params.get('socket_timeout')
19a41fc6 1973 self._socket_timeout = 600 if timeout_val is None else float(timeout_val)
6ad14cab 1974
dca08720
PH
1975 opts_cookiefile = self.params.get('cookiefile')
1976 opts_proxy = self.params.get('proxy')
1977
1978 if opts_cookiefile is None:
1979 self.cookiejar = compat_cookiejar.CookieJar()
1980 else:
1981 self.cookiejar = compat_cookiejar.MozillaCookieJar(
1982 opts_cookiefile)
1983 if os.access(opts_cookiefile, os.R_OK):
1984 self.cookiejar.load()
1985
6a3f4c3f 1986 cookie_processor = YoutubeDLCookieProcessor(self.cookiejar)
dca08720
PH
1987 if opts_proxy is not None:
1988 if opts_proxy == '':
1989 proxies = {}
1990 else:
1991 proxies = {'http': opts_proxy, 'https': opts_proxy}
1992 else:
1993 proxies = compat_urllib_request.getproxies()
1994 # Set HTTPS proxy to HTTP one if given (https://github.com/rg3/youtube-dl/issues/805)
1995 if 'http' in proxies and 'https' not in proxies:
1996 proxies['https'] = proxies['http']
91410c9b 1997 proxy_handler = PerRequestProxyHandler(proxies)
a0ddb8a2
PH
1998
1999 debuglevel = 1 if self.params.get('debug_printtraffic') else 0
be4a824d
PH
2000 https_handler = make_HTTPS_handler(self.params, debuglevel=debuglevel)
2001 ydlh = YoutubeDLHandler(self.params, debuglevel=debuglevel)
8b172c2e 2002 data_handler = compat_urllib_request_DataHandler()
6240b0a2
JMF
2003
2004 # When passing our own FileHandler instance, build_opener won't add the
2005 # default FileHandler and allows us to disable the file protocol, which
2006 # can be used for malicious purposes (see
e37afbe0 2007 # https://github.com/rg3/youtube-dl/issues/8227)
6240b0a2
JMF
2008 file_handler = compat_urllib_request.FileHandler()
2009
2010 def file_open(*args, **kwargs):
30e2f2d7 2011 raise compat_urllib_error.URLError('file:// scheme is explicitly disabled in youtube-dl for security reasons')
6240b0a2
JMF
2012 file_handler.file_open = file_open
2013
2014 opener = compat_urllib_request.build_opener(
2015 proxy_handler, https_handler, cookie_processor, ydlh, data_handler, file_handler)
2461f79d 2016
dca08720
PH
2017 # Delete the default user-agent header, which would otherwise apply in
2018 # cases where our custom HTTP handler doesn't come into play
2019 # (See https://github.com/rg3/youtube-dl/issues/1309 for details)
2020 opener.addheaders = []
2021 self._opener = opener
62fec3b2
PH
2022
2023 def encode(self, s):
2024 if isinstance(s, bytes):
2025 return s # Already encoded
2026
2027 try:
2028 return s.encode(self.get_encoding())
2029 except UnicodeEncodeError as err:
2030 err.reason = err.reason + '. Check your system encoding configuration or use the --encoding option.'
2031 raise
2032
2033 def get_encoding(self):
2034 encoding = self.params.get('encoding')
2035 if encoding is None:
2036 encoding = preferredencoding()
2037 return encoding
ec82d85a
PH
2038
2039 def _write_thumbnails(self, info_dict, filename):
2040 if self.params.get('writethumbnail', False):
2041 thumbnails = info_dict.get('thumbnails')
2042 if thumbnails:
2043 thumbnails = [thumbnails[-1]]
2044 elif self.params.get('write_all_thumbnails', False):
2045 thumbnails = info_dict.get('thumbnails')
2046 else:
2047 return
2048
2049 if not thumbnails:
2050 # No thumbnails present, so return immediately
2051 return
2052
2053 for t in thumbnails:
2054 thumb_ext = determine_ext(t['url'], 'jpg')
2055 suffix = '_%s' % t['id'] if len(thumbnails) > 1 else ''
2056 thumb_display_id = '%s ' % t['id'] if len(thumbnails) > 1 else ''
82245a6d 2057 t['filename'] = thumb_filename = os.path.splitext(filename)[0] + suffix + '.' + thumb_ext
ec82d85a
PH
2058
2059 if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(thumb_filename)):
2060 self.to_screen('[%s] %s: Thumbnail %sis already present' %
2061 (info_dict['extractor'], info_dict['id'], thumb_display_id))
2062 else:
2063 self.to_screen('[%s] %s: Downloading thumbnail %s...' %
2064 (info_dict['extractor'], info_dict['id'], thumb_display_id))
2065 try:
2066 uf = self.urlopen(t['url'])
d3d89c32 2067 with open(encodeFilename(thumb_filename), 'wb') as thumbf:
ec82d85a
PH
2068 shutil.copyfileobj(uf, thumbf)
2069 self.to_screen('[%s] %s: Writing thumbnail %sto: %s' %
2070 (info_dict['extractor'], info_dict['id'], thumb_display_id, thumb_filename))
2071 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
2072 self.report_warning('Unable to download thumbnail "%s": %s' %
9b9c5355 2073 (t['url'], error_to_compat_str(err)))