]> jfr.im git - yt-dlp.git/blame - youtube_dl/YoutubeDL.py
Remove the SubtitlesInfoExtractor class
[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
9d2ecdbc 7import datetime
c1c9a79c 8import errno
8222d8de 9import io
b82f815f 10import itertools
8694c600 11import json
62fec3b2 12import locale
083c9df9 13import operator
8222d8de 14import os
dca08720 15import platform
8222d8de
JMF
16import re
17import shutil
dca08720 18import subprocess
8222d8de
JMF
19import socket
20import sys
21import time
22import traceback
23
1e5b9a95
PH
24if os.name == 'nt':
25 import ctypes
26
8c25f81b 27from .compat import (
8f9312c3 28 compat_basestring,
dca08720 29 compat_cookiejar,
4644ac55 30 compat_expanduser,
ce02ed60 31 compat_http_client,
4f026faf 32 compat_kwargs,
ce02ed60
PH
33 compat_str,
34 compat_urllib_error,
35 compat_urllib_request,
8c25f81b
PH
36)
37from .utils import (
d05cfe06 38 escape_url,
ce02ed60
PH
39 ContentTooShortError,
40 date_from_str,
41 DateRange,
acd69589 42 DEFAULT_OUTTMPL,
ce02ed60
PH
43 determine_ext,
44 DownloadError,
45 encodeFilename,
46 ExtractorError,
02dbf93f 47 format_bytes,
525ef922 48 formatSeconds,
1c088fa8 49 get_term_width,
ce02ed60 50 locked_file,
dca08720 51 make_HTTPS_handler,
ce02ed60 52 MaxDownloadsReached,
b7ab0590 53 PagedList,
083c9df9 54 parse_filesize,
ce02ed60 55 PostProcessingError,
dca08720 56 platform_name,
ce02ed60 57 preferredencoding,
cfb56d1a 58 render_table,
ce02ed60
PH
59 SameFileError,
60 sanitize_filename,
e5660ee6 61 std_headers,
ce02ed60
PH
62 subtitles_filename,
63 takewhile_inclusive,
64 UnavailableVideoError,
29eb5174 65 url_basename,
58b1f00d 66 version_tuple,
ce02ed60
PH
67 write_json_file,
68 write_string,
dca08720 69 YoutubeDLHandler,
6350728b 70 prepend_extension,
7d4111ed 71 args_to_str,
05900629 72 age_restricted,
ce02ed60 73)
a0e07d31 74from .cache import Cache
023fa8c4 75from .extractor import get_info_extractor, gen_extractors
3bc2ddcc 76from .downloader import get_suitable_downloader
4c83c967 77from .downloader.rtmp import rtmpdump_version
4f026faf 78from .postprocessor import (
62cd676c 79 FFmpegFixupM4aPP,
6271f1ca 80 FFmpegFixupStretchedPP,
4f026faf
PH
81 FFmpegMergerPP,
82 FFmpegPostProcessor,
83 get_postprocessor,
84)
dca08720 85from .version import __version__
8222d8de
JMF
86
87
88class YoutubeDL(object):
89 """YoutubeDL class.
90
91 YoutubeDL objects are the ones responsible of downloading the
92 actual video file and writing it to disk if the user has requested
93 it, among some other tasks. In most cases there should be one per
94 program. As, given a video URL, the downloader doesn't know how to
95 extract all the needed information, task that InfoExtractors do, it
96 has to pass the URL to one of them.
97
98 For this, YoutubeDL objects have a method that allows
99 InfoExtractors to be registered in a given order. When it is passed
100 a URL, the YoutubeDL object handles it to the first InfoExtractor it
101 finds that reports being able to handle it. The InfoExtractor extracts
102 all the information about the video or videos the URL refers to, and
103 YoutubeDL process the extracted information, possibly using a File
104 Downloader to download the video.
105
106 YoutubeDL objects accept a lot of parameters. In order not to saturate
107 the object constructor with arguments, it receives a dictionary of
108 options instead. These options are available through the params
109 attribute for the InfoExtractors to use. The YoutubeDL also
110 registers itself as the downloader in charge for the InfoExtractors
111 that are added to it, so this is a "mutual registration".
112
113 Available options:
114
115 username: Username for authentication purposes.
116 password: Password for authentication purposes.
c6c19746 117 videopassword: Password for acces a video.
8222d8de
JMF
118 usenetrc: Use netrc for authentication instead.
119 verbose: Print additional info to stdout.
120 quiet: Do not print messages to stdout.
ad8915b7 121 no_warnings: Do not print out anything for warnings.
8222d8de
JMF
122 forceurl: Force printing final URL.
123 forcetitle: Force printing title.
124 forceid: Force printing ID.
125 forcethumbnail: Force printing thumbnail URL.
126 forcedescription: Force printing description.
127 forcefilename: Force printing final filename.
525ef922 128 forceduration: Force printing duration.
8694c600 129 forcejson: Force printing info_dict as JSON.
63e0be34
PH
130 dump_single_json: Force printing the info_dict of the whole playlist
131 (or video) as a single JSON line.
8222d8de 132 simulate: Do not download the video files.
d8600787 133 format: Video format code. See options.py for more information.
8222d8de
JMF
134 format_limit: Highest quality format to try.
135 outtmpl: Template for output names.
136 restrictfilenames: Do not allow "&" and spaces in file names
137 ignoreerrors: Do not stop on download errors.
138 nooverwrites: Prevent overwriting files.
139 playliststart: Playlist item to start at.
140 playlistend: Playlist item to end at.
c14e88f0 141 playlist_items: Specific indices of playlist to download.
ff815fe6 142 playlistreverse: Download playlist items in reverse order.
8222d8de
JMF
143 matchtitle: Download only matching titles.
144 rejecttitle: Reject downloads for matching titles.
8bf9319e 145 logger: Log messages to a logging.Logger instance.
8222d8de
JMF
146 logtostderr: Log messages to stderr instead of stdout.
147 writedescription: Write the video description to a .description file
148 writeinfojson: Write the video description to a .info.json file
1fb07d10 149 writeannotations: Write the video annotations to a .annotations.xml file
8222d8de 150 writethumbnail: Write the thumbnail image to a file
ec82d85a 151 write_all_thumbnails: Write all thumbnail formats to files
8222d8de 152 writesubtitles: Write the video subtitles to a file
b004821f 153 writeautomaticsub: Write the automatic subtitles to a file
8222d8de 154 allsubtitles: Downloads all the subtitles of the video
0b7f3118 155 (requires writesubtitles or writeautomaticsub)
8222d8de 156 listsubtitles: Lists all available subtitles for the video
a504ced0 157 subtitlesformat: The format code for subtitles
aa6a10c4 158 subtitleslangs: List of languages of the subtitles to download
8222d8de
JMF
159 keepvideo: Keep the video file after post-processing
160 daterange: A DateRange object, download only if the upload_date is in the range.
161 skip_download: Skip the actual download of the video file
c35f9e72 162 cachedir: Location of the cache files in the filesystem.
a0e07d31 163 False to disable filesystem cache.
47192f92 164 noplaylist: Download single video instead of a playlist if in doubt.
8dbe9899
PH
165 age_limit: An integer representing the user's age in years.
166 Unsuitable videos for the given age are skipped.
5fe18bdb
PH
167 min_views: An integer representing the minimum view count the video
168 must have in order to not be skipped.
169 Videos without view count information are always
170 downloaded. None for no limit.
171 max_views: An integer representing the maximum view count.
172 Videos that are more popular than that are not
173 downloaded.
174 Videos without view count information are always
175 downloaded. None for no limit.
176 download_archive: File name of a file where all downloads are recorded.
c1c9a79c
PH
177 Videos already present in the file are not downloaded
178 again.
dca08720 179 cookiefile: File name where cookies should be read from and dumped to.
a1ee09e8 180 nocheckcertificate:Do not verify SSL certificates
7e8c0af0
PH
181 prefer_insecure: Use HTTP instead of HTTPS to retrieve information.
182 At the moment, this is only supported by YouTube.
a1ee09e8 183 proxy: URL of the proxy server to use
e344693b 184 socket_timeout: Time to wait for unresponsive hosts, in seconds
0783b09b
PH
185 bidi_workaround: Work around buggy terminals without bidirectional text
186 support, using fridibi
a0ddb8a2 187 debug_printtraffic:Print out sent and received HTTP traffic
7b0817e8 188 include_ads: Download ads as well
04b4d394
PH
189 default_search: Prepend this string if an input url is not valid.
190 'auto' for elaborate guessing
62fec3b2 191 encoding: Use this encoding instead of the system-specified.
e8ee972c 192 extract_flat: Do not resolve URLs, return the immediate result.
057a5206
PH
193 Pass in 'in_playlist' to only show this behavior for
194 playlist items.
4f026faf 195 postprocessors: A list of dictionaries, each with an entry
71b640cc
PH
196 * key: The name of the postprocessor. See
197 youtube_dl/postprocessor/__init__.py for a list.
4f026faf
PH
198 as well as any further keyword arguments for the
199 postprocessor.
71b640cc
PH
200 progress_hooks: A list of functions that get called on download
201 progress, with a dictionary with the entries
ee69b99a
PH
202 * status: One of "downloading" and "finished".
203 Check this first and ignore unknown values.
71b640cc 204
ee69b99a
PH
205 If status is one of "downloading" or "finished", the
206 following properties may also be present:
207 * filename: The final filename (always present)
71b640cc
PH
208 * downloaded_bytes: Bytes on disk
209 * total_bytes: Size of the whole file, None if unknown
210 * tmpfilename: The filename we're currently writing to
211 * eta: The estimated time in seconds, None if unknown
212 * speed: The download speed in bytes/second, None if
213 unknown
214
215 Progress hooks are guaranteed to be called at least once
216 (with status "finished") if the download is successful.
45598f15 217 merge_output_format: Extension to use when merging formats.
6271f1ca
PH
218 fixup: Automatically correct known faults of the file.
219 One of:
220 - "never": do nothing
221 - "warn": only emit a warning
222 - "detect_or_warn": check whether we can do anything
62cd676c 223 about it, warn otherwise (default)
be4a824d 224 source_address: (Experimental) Client-side IP address to bind to.
8bfa7545
PH
225 call_home: Boolean, true iff we are allowed to contact the
226 youtube-dl servers for debugging.
5f0d813d 227 sleep_interval: Number of seconds to sleep before each download.
222516d9 228 external_downloader: Executable of the external downloader to call.
cfb56d1a
PH
229 listformats: Print an overview of available video formats and exit.
230 list_thumbnails: Print a table of all thumbnails and exit.
347de493
PH
231 match_filter: A function that gets called with the info_dict of
232 every video.
233 If it returns a message, the video is ignored.
234 If it returns None, the video is downloaded.
235 match_filter_func in utils.py is one example for this.
7e5db8c9 236 no_color: Do not emit color codes in output.
71b640cc 237
fe7e0c98 238
8222d8de
JMF
239 The following parameters are not used by YoutubeDL itself, they are used by
240 the FileDownloader:
241 nopart, updatetime, buffersize, ratelimit, min_filesize, max_filesize, test,
881e6a1f
PH
242 noresizebuffer, retries, continuedl, noprogress, consoletitle,
243 xattr_set_filesize.
76b1bd67
JMF
244
245 The following options are used by the post processors:
246 prefer_ffmpeg: If True, use ffmpeg instead of avconv if both are available,
247 otherwise prefer avconv.
8d31fa3c 248 exec_cmd: Arbitrary command to run after downloading
8222d8de
JMF
249 """
250
251 params = None
252 _ies = []
253 _pps = []
254 _download_retcode = None
255 _num_downloads = None
256 _screen_file = None
257
3511266b 258 def __init__(self, params=None, auto_init=True):
8222d8de 259 """Create a FileDownloader object with the given options."""
e9f9a10f
JMF
260 if params is None:
261 params = {}
8222d8de 262 self._ies = []
56c73665 263 self._ies_instances = {}
8222d8de 264 self._pps = []
933605d7 265 self._progress_hooks = []
8222d8de
JMF
266 self._download_retcode = 0
267 self._num_downloads = 0
268 self._screen_file = [sys.stdout, sys.stderr][params.get('logtostderr', False)]
0783b09b 269 self._err_file = sys.stderr
e9f9a10f 270 self.params = params
a0e07d31 271 self.cache = Cache(self)
34308b30 272
0783b09b 273 if params.get('bidi_workaround', False):
1c088fa8
PH
274 try:
275 import pty
276 master, slave = pty.openpty()
277 width = get_term_width()
278 if width is None:
279 width_args = []
280 else:
281 width_args = ['-w', str(width)]
5d681e96 282 sp_kwargs = dict(
1c088fa8
PH
283 stdin=subprocess.PIPE,
284 stdout=slave,
285 stderr=self._err_file)
5d681e96
PH
286 try:
287 self._output_process = subprocess.Popen(
288 ['bidiv'] + width_args, **sp_kwargs
289 )
290 except OSError:
5d681e96
PH
291 self._output_process = subprocess.Popen(
292 ['fribidi', '-c', 'UTF-8'] + width_args, **sp_kwargs)
293 self._output_channel = os.fdopen(master, 'rb')
1c088fa8
PH
294 except OSError as ose:
295 if ose.errno == 2:
6febd1c1 296 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 $PATH.')
1c088fa8
PH
297 else:
298 raise
0783b09b 299
34308b30
PH
300 if (sys.version_info >= (3,) and sys.platform != 'win32' and
301 sys.getfilesystemencoding() in ['ascii', 'ANSI_X3.4-1968']
53d9009b 302 and not params.get('restrictfilenames', False)):
34308b30
PH
303 # On Python 3, the Unicode filesystem API will throw errors (#1474)
304 self.report_warning(
6febd1c1 305 'Assuming --restrict-filenames since file system encoding '
1b725173 306 'cannot encode all characters. '
6febd1c1 307 'Set the LC_ALL environment variable to fix this.')
4a98cdbf 308 self.params['restrictfilenames'] = True
34308b30 309
a3927cf7 310 if '%(stitle)s' in self.params.get('outtmpl', ''):
6febd1c1 311 self.report_warning('%(stitle)s is deprecated. Use the %(title)s and the --restrict-filenames flag(which also secures %(uploader)s et al) instead.')
8222d8de 312
dca08720
PH
313 self._setup_opener()
314
3511266b
PH
315 if auto_init:
316 self.print_debug_header()
317 self.add_default_info_extractors()
318
4f026faf
PH
319 for pp_def_raw in self.params.get('postprocessors', []):
320 pp_class = get_postprocessor(pp_def_raw['key'])
321 pp_def = dict(pp_def_raw)
322 del pp_def['key']
323 pp = pp_class(self, **compat_kwargs(pp_def))
324 self.add_post_processor(pp)
325
71b640cc
PH
326 for ph in self.params.get('progress_hooks', []):
327 self.add_progress_hook(ph)
328
7d4111ed
PH
329 def warn_if_short_id(self, argv):
330 # short YouTube ID starting with dash?
331 idxs = [
332 i for i, a in enumerate(argv)
333 if re.match(r'^-[0-9A-Za-z_-]{10}$', a)]
334 if idxs:
335 correct_argv = (
336 ['youtube-dl'] +
337 [a for i, a in enumerate(argv) if i not in idxs] +
338 ['--'] + [argv[i] for i in idxs]
339 )
340 self.report_warning(
341 'Long argument string detected. '
342 'Use -- to separate parameters and URLs, like this:\n%s\n' %
343 args_to_str(correct_argv))
344
8222d8de
JMF
345 def add_info_extractor(self, ie):
346 """Add an InfoExtractor object to the end of the list."""
347 self._ies.append(ie)
56c73665 348 self._ies_instances[ie.ie_key()] = ie
8222d8de
JMF
349 ie.set_downloader(self)
350
56c73665
JMF
351 def get_info_extractor(self, ie_key):
352 """
353 Get an instance of an IE with name ie_key, it will try to get one from
354 the _ies list, if there's no instance it will create a new one and add
355 it to the extractor list.
356 """
357 ie = self._ies_instances.get(ie_key)
358 if ie is None:
359 ie = get_info_extractor(ie_key)()
360 self.add_info_extractor(ie)
361 return ie
362
023fa8c4
JMF
363 def add_default_info_extractors(self):
364 """
365 Add the InfoExtractors returned by gen_extractors to the end of the list
366 """
367 for ie in gen_extractors():
368 self.add_info_extractor(ie)
369
8222d8de
JMF
370 def add_post_processor(self, pp):
371 """Add a PostProcessor object to the end of the chain."""
372 self._pps.append(pp)
373 pp.set_downloader(self)
374
933605d7
JMF
375 def add_progress_hook(self, ph):
376 """Add the progress hook (currently only for the file downloader)"""
377 self._progress_hooks.append(ph)
8ab470f1 378
1c088fa8 379 def _bidi_workaround(self, message):
5d681e96 380 if not hasattr(self, '_output_channel'):
1c088fa8
PH
381 return message
382
5d681e96 383 assert hasattr(self, '_output_process')
11b85ce6 384 assert isinstance(message, compat_str)
6febd1c1
PH
385 line_count = message.count('\n') + 1
386 self._output_process.stdin.write((message + '\n').encode('utf-8'))
5d681e96 387 self._output_process.stdin.flush()
6febd1c1 388 res = ''.join(self._output_channel.readline().decode('utf-8')
9e1a5b84 389 for _ in range(line_count))
6febd1c1 390 return res[:-len('\n')]
1c088fa8 391
8222d8de 392 def to_screen(self, message, skip_eol=False):
0783b09b
PH
393 """Print message to stdout if not in quiet mode."""
394 return self.to_stdout(message, skip_eol, check_quiet=True)
395
734f90bb 396 def _write_string(self, s, out=None):
b58ddb32 397 write_string(s, out=out, encoding=self.params.get('encoding'))
734f90bb 398
0783b09b 399 def to_stdout(self, message, skip_eol=False, check_quiet=False):
8222d8de 400 """Print message to stdout if not in quiet mode."""
8bf9319e 401 if self.params.get('logger'):
43afe285 402 self.params['logger'].debug(message)
0783b09b 403 elif not check_quiet or not self.params.get('quiet', False):
1c088fa8 404 message = self._bidi_workaround(message)
6febd1c1 405 terminator = ['\n', ''][skip_eol]
8222d8de 406 output = message + terminator
1c088fa8 407
734f90bb 408 self._write_string(output, self._screen_file)
8222d8de
JMF
409
410 def to_stderr(self, message):
411 """Print message to stderr."""
11b85ce6 412 assert isinstance(message, compat_str)
8bf9319e 413 if self.params.get('logger'):
43afe285
IB
414 self.params['logger'].error(message)
415 else:
1c088fa8 416 message = self._bidi_workaround(message)
6febd1c1 417 output = message + '\n'
734f90bb 418 self._write_string(output, self._err_file)
8222d8de 419
1e5b9a95
PH
420 def to_console_title(self, message):
421 if not self.params.get('consoletitle', False):
422 return
423 if os.name == 'nt' and ctypes.windll.kernel32.GetConsoleWindow():
424 # c_wchar_p() might not be necessary if `message` is
425 # already of type unicode()
426 ctypes.windll.kernel32.SetConsoleTitleW(ctypes.c_wchar_p(message))
427 elif 'TERM' in os.environ:
734f90bb 428 self._write_string('\033]0;%s\007' % message, self._screen_file)
1e5b9a95 429
bdde425c
PH
430 def save_console_title(self):
431 if not self.params.get('consoletitle', False):
432 return
433 if 'TERM' in os.environ:
efd6c574 434 # Save the title on stack
734f90bb 435 self._write_string('\033[22;0t', self._screen_file)
bdde425c
PH
436
437 def restore_console_title(self):
438 if not self.params.get('consoletitle', False):
439 return
440 if 'TERM' in os.environ:
efd6c574 441 # Restore the title from stack
734f90bb 442 self._write_string('\033[23;0t', self._screen_file)
bdde425c
PH
443
444 def __enter__(self):
445 self.save_console_title()
446 return self
447
448 def __exit__(self, *args):
449 self.restore_console_title()
f89197d7 450
dca08720
PH
451 if self.params.get('cookiefile') is not None:
452 self.cookiejar.save()
bdde425c 453
8222d8de
JMF
454 def trouble(self, message=None, tb=None):
455 """Determine action to take when a download problem appears.
456
457 Depending on if the downloader has been configured to ignore
458 download errors or not, this method may throw an exception or
459 not when errors are found, after printing the message.
460
461 tb, if given, is additional traceback information.
462 """
463 if message is not None:
464 self.to_stderr(message)
465 if self.params.get('verbose'):
466 if tb is None:
467 if sys.exc_info()[0]: # if .trouble has been called from an except block
6febd1c1 468 tb = ''
8222d8de 469 if hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
6febd1c1 470 tb += ''.join(traceback.format_exception(*sys.exc_info()[1].exc_info))
8222d8de
JMF
471 tb += compat_str(traceback.format_exc())
472 else:
473 tb_data = traceback.format_list(traceback.extract_stack())
6febd1c1 474 tb = ''.join(tb_data)
8222d8de
JMF
475 self.to_stderr(tb)
476 if not self.params.get('ignoreerrors', False):
477 if sys.exc_info()[0] and hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
478 exc_info = sys.exc_info()[1].exc_info
479 else:
480 exc_info = sys.exc_info()
481 raise DownloadError(message, exc_info)
482 self._download_retcode = 1
483
484 def report_warning(self, message):
485 '''
486 Print the message to stderr, it will be prefixed with 'WARNING:'
487 If stderr is a tty file the 'WARNING:' will be colored
488 '''
6d07ce01
JMF
489 if self.params.get('logger') is not None:
490 self.params['logger'].warning(message)
8222d8de 491 else:
ad8915b7
PH
492 if self.params.get('no_warnings'):
493 return
7e5db8c9 494 if not self.params.get('no_color') and self._err_file.isatty() and os.name != 'nt':
6d07ce01
JMF
495 _msg_header = '\033[0;33mWARNING:\033[0m'
496 else:
497 _msg_header = 'WARNING:'
498 warning_message = '%s %s' % (_msg_header, message)
499 self.to_stderr(warning_message)
8222d8de
JMF
500
501 def report_error(self, message, tb=None):
502 '''
503 Do the same as trouble, but prefixes the message with 'ERROR:', colored
504 in red if stderr is a tty file.
505 '''
7e5db8c9 506 if not self.params.get('no_color') and self._err_file.isatty() and os.name != 'nt':
6febd1c1 507 _msg_header = '\033[0;31mERROR:\033[0m'
8222d8de 508 else:
6febd1c1
PH
509 _msg_header = 'ERROR:'
510 error_message = '%s %s' % (_msg_header, message)
8222d8de
JMF
511 self.trouble(error_message, tb)
512
8222d8de
JMF
513 def report_file_already_downloaded(self, file_name):
514 """Report file has already been fully downloaded."""
515 try:
6febd1c1 516 self.to_screen('[download] %s has already been downloaded' % file_name)
ce02ed60 517 except UnicodeEncodeError:
6febd1c1 518 self.to_screen('[download] The file has already been downloaded')
8222d8de 519
8222d8de
JMF
520 def prepare_filename(self, info_dict):
521 """Generate the output filename."""
522 try:
523 template_dict = dict(info_dict)
524
525 template_dict['epoch'] = int(time.time())
526 autonumber_size = self.params.get('autonumber_size')
527 if autonumber_size is None:
528 autonumber_size = 5
6febd1c1 529 autonumber_templ = '%0' + str(autonumber_size) + 'd'
8222d8de 530 template_dict['autonumber'] = autonumber_templ % self._num_downloads
702665c0 531 if template_dict.get('playlist_index') is not None:
c6b4132a 532 template_dict['playlist_index'] = '%0*d' % (len(str(template_dict['n_entries'])), template_dict['playlist_index'])
17b75c0d
PH
533 if template_dict.get('resolution') is None:
534 if template_dict.get('width') and template_dict.get('height'):
535 template_dict['resolution'] = '%dx%d' % (template_dict['width'], template_dict['height'])
536 elif template_dict.get('height'):
805ef3c6 537 template_dict['resolution'] = '%sp' % template_dict['height']
17b75c0d 538 elif template_dict.get('width'):
805ef3c6 539 template_dict['resolution'] = '?x%d' % template_dict['width']
8222d8de 540
586a91b6 541 sanitize = lambda k, v: sanitize_filename(
45598aab 542 compat_str(v),
8222d8de 543 restricted=self.params.get('restrictfilenames'),
6febd1c1 544 is_id=(k == 'id'))
586a91b6 545 template_dict = dict((k, sanitize(k, v))
45598aab
PH
546 for k, v in template_dict.items()
547 if v is not None)
6febd1c1 548 template_dict = collections.defaultdict(lambda: 'NA', template_dict)
8222d8de 549
acd69589 550 outtmpl = self.params.get('outtmpl', DEFAULT_OUTTMPL)
4644ac55 551 tmpl = compat_expanduser(outtmpl)
586a91b6 552 filename = tmpl % template_dict
3a0d2f52
S
553 # Temporary fix for #4787
554 # 'Treat' all problem characters by passing filename through preferredencoding
555 # to workaround encoding issues with subprocess on python2 @ Windows
556 if sys.version_info < (3, 0) and sys.platform == 'win32':
557 filename = encodeFilename(filename, True).decode(preferredencoding())
8222d8de 558 return filename
8222d8de 559 except ValueError as err:
6febd1c1 560 self.report_error('Error in output template: ' + str(err) + ' (encoding: ' + repr(preferredencoding()) + ')')
8222d8de
JMF
561 return None
562
442c37b7 563 def _match_entry(self, info_dict, incomplete):
8222d8de
JMF
564 """ Returns None iff the file should be downloaded """
565
6febd1c1 566 video_title = info_dict.get('title', info_dict.get('id', 'video'))
7012b23c
PH
567 if 'title' in info_dict:
568 # This can happen when we're just evaluating the playlist
569 title = info_dict['title']
570 matchtitle = self.params.get('matchtitle', False)
571 if matchtitle:
572 if not re.search(matchtitle, title, re.IGNORECASE):
6febd1c1 573 return '"' + title + '" title did not match pattern "' + matchtitle + '"'
7012b23c
PH
574 rejecttitle = self.params.get('rejecttitle', False)
575 if rejecttitle:
576 if re.search(rejecttitle, title, re.IGNORECASE):
6febd1c1 577 return '"' + title + '" title matched reject pattern "' + rejecttitle + '"'
8222d8de
JMF
578 date = info_dict.get('upload_date', None)
579 if date is not None:
580 dateRange = self.params.get('daterange', DateRange())
581 if date not in dateRange:
6febd1c1 582 return '%s upload date is not in range %s' % (date_from_str(date).isoformat(), dateRange)
5fe18bdb
PH
583 view_count = info_dict.get('view_count', None)
584 if view_count is not None:
585 min_views = self.params.get('min_views')
586 if min_views is not None and view_count < min_views:
6febd1c1 587 return 'Skipping %s, because it has not reached minimum view count (%d/%d)' % (video_title, view_count, min_views)
5fe18bdb
PH
588 max_views = self.params.get('max_views')
589 if max_views is not None and view_count > max_views:
6febd1c1 590 return 'Skipping %s, because it has exceeded the maximum view count (%d/%d)' % (video_title, view_count, max_views)
05900629 591 if age_restricted(info_dict.get('age_limit'), self.params.get('age_limit')):
347de493 592 return 'Skipping "%s" because it is age restricted' % video_title
c1c9a79c 593 if self.in_download_archive(info_dict):
6febd1c1 594 return '%s has already been recorded in archive' % video_title
347de493 595
442c37b7
PH
596 if not incomplete:
597 match_filter = self.params.get('match_filter')
598 if match_filter is not None:
599 ret = match_filter(info_dict)
600 if ret is not None:
601 return ret
347de493 602
8222d8de 603 return None
fe7e0c98 604
b6c45014
JMF
605 @staticmethod
606 def add_extra_info(info_dict, extra_info):
607 '''Set the keys from extra_info in info dict if they are missing'''
608 for key, value in extra_info.items():
609 info_dict.setdefault(key, value)
610
7fc3fa05
PH
611 def extract_info(self, url, download=True, ie_key=None, extra_info={},
612 process=True):
8222d8de
JMF
613 '''
614 Returns a list with a dictionary for each video we find.
615 If 'download', also downloads the videos.
616 extra_info is a dict containing the extra values to add to each result
617 '''
fe7e0c98 618
8222d8de 619 if ie_key:
56c73665 620 ies = [self.get_info_extractor(ie_key)]
8222d8de
JMF
621 else:
622 ies = self._ies
623
624 for ie in ies:
625 if not ie.suitable(url):
626 continue
627
628 if not ie.working():
6febd1c1
PH
629 self.report_warning('The program functionality for this site has been marked as broken, '
630 'and will probably not work.')
8222d8de
JMF
631
632 try:
633 ie_result = ie.extract(url)
5f6a1245 634 if ie_result is None: # Finished already (backwards compatibility; listformats and friends should be moved here)
8222d8de
JMF
635 break
636 if isinstance(ie_result, list):
637 # Backwards compatibility: old IE result format
8222d8de
JMF
638 ie_result = {
639 '_type': 'compat_list',
640 'entries': ie_result,
641 }
ea38e55f 642 self.add_default_extra_info(ie_result, ie, url)
7fc3fa05
PH
643 if process:
644 return self.process_ie_result(ie_result, download, extra_info)
645 else:
646 return ie_result
5f6a1245 647 except ExtractorError as de: # An error we somewhat expected
8222d8de
JMF
648 self.report_error(compat_str(de), de.format_traceback())
649 break
d3e5bbf4
PH
650 except MaxDownloadsReached:
651 raise
8222d8de
JMF
652 except Exception as e:
653 if self.params.get('ignoreerrors', False):
654 self.report_error(compat_str(e), tb=compat_str(traceback.format_exc()))
655 break
656 else:
657 raise
658 else:
1a489545 659 self.report_error('no suitable InfoExtractor for URL %s' % url)
fe7e0c98 660
ea38e55f
PH
661 def add_default_extra_info(self, ie_result, ie, url):
662 self.add_extra_info(ie_result, {
663 'extractor': ie.IE_NAME,
664 'webpage_url': url,
665 'webpage_url_basename': url_basename(url),
666 'extractor_key': ie.ie_key(),
667 })
668
8222d8de
JMF
669 def process_ie_result(self, ie_result, download=True, extra_info={}):
670 """
671 Take the result of the ie(may be modified) and resolve all unresolved
672 references (URLs, playlist items).
673
674 It will also download the videos if 'download'.
675 Returns the resolved ie_result.
676 """
677
e8ee972c
PH
678 result_type = ie_result.get('_type', 'video')
679
057a5206
PH
680 if result_type in ('url', 'url_transparent'):
681 extract_flat = self.params.get('extract_flat', False)
682 if ((extract_flat == 'in_playlist' and 'playlist' in extra_info) or
683 extract_flat is True):
057a5206
PH
684 if self.params.get('forcejson', False):
685 self.to_stdout(json.dumps(ie_result))
e8ee972c
PH
686 return ie_result
687
8222d8de 688 if result_type == 'video':
b6c45014 689 self.add_extra_info(ie_result, extra_info)
feee2ecf 690 return self.process_video_result(ie_result, download=download)
8222d8de
JMF
691 elif result_type == 'url':
692 # We have to add extra_info to the results because it may be
693 # contained in a playlist
694 return self.extract_info(ie_result['url'],
695 download,
696 ie_key=ie_result.get('ie_key'),
697 extra_info=extra_info)
7fc3fa05
PH
698 elif result_type == 'url_transparent':
699 # Use the information from the embedding page
700 info = self.extract_info(
701 ie_result['url'], ie_key=ie_result.get('ie_key'),
702 extra_info=extra_info, download=False, process=False)
703
412c617d
PH
704 force_properties = dict(
705 (k, v) for k, v in ie_result.items() if v is not None)
706 for f in ('_type', 'url'):
707 if f in force_properties:
708 del force_properties[f]
709 new_result = info.copy()
710 new_result.update(force_properties)
7fc3fa05
PH
711
712 assert new_result.get('_type') != 'url_transparent'
7fc3fa05
PH
713
714 return self.process_ie_result(
715 new_result, download=download, extra_info=extra_info)
42e12102 716 elif result_type == 'playlist' or result_type == 'multi_video':
8222d8de
JMF
717 # We process each entry in the playlist
718 playlist = ie_result.get('title', None) or ie_result.get('id', None)
6febd1c1 719 self.to_screen('[download] Downloading playlist: %s' % playlist)
8222d8de
JMF
720
721 playlist_results = []
722
8222d8de 723 playliststart = self.params.get('playliststart', 1) - 1
a19fd00c
PH
724 playlistend = self.params.get('playlistend', None)
725 # For backwards compatibility, interpret -1 as whole list
8222d8de 726 if playlistend == -1:
a19fd00c 727 playlistend = None
8222d8de 728
c14e88f0
PH
729 playlistitems_str = self.params.get('playlist_items', None)
730 playlistitems = None
731 if playlistitems_str is not None:
732 def iter_playlistitems(format):
733 for string_segment in format.split(','):
734 if '-' in string_segment:
735 start, end = string_segment.split('-')
736 for item in range(int(start), int(end) + 1):
737 yield int(item)
738 else:
739 yield int(string_segment)
740 playlistitems = iter_playlistitems(playlistitems_str)
741
b82f815f
PH
742 ie_entries = ie_result['entries']
743 if isinstance(ie_entries, list):
744 n_all_entries = len(ie_entries)
c14e88f0
PH
745 if playlistitems:
746 entries = [ie_entries[i - 1] for i in playlistitems]
747 else:
748 entries = ie_entries[playliststart:playlistend]
b7ab0590
PH
749 n_entries = len(entries)
750 self.to_screen(
751 "[%s] playlist %s: Collected %d video ids (downloading %d of them)" %
752 (ie_result['extractor'], playlist, n_all_entries, n_entries))
b82f815f 753 elif isinstance(ie_entries, PagedList):
c14e88f0
PH
754 if playlistitems:
755 entries = []
756 for item in playlistitems:
757 entries.extend(ie_entries.getslice(
758 item - 1, item
759 ))
760 else:
761 entries = ie_entries.getslice(
762 playliststart, playlistend)
b7ab0590
PH
763 n_entries = len(entries)
764 self.to_screen(
765 "[%s] playlist %s: Downloading %d videos" %
766 (ie_result['extractor'], playlist, n_entries))
b82f815f 767 else: # iterable
c14e88f0
PH
768 if playlistitems:
769 entry_list = list(ie_entries)
770 entries = [entry_list[i - 1] for i in playlistitems]
771 else:
772 entries = list(itertools.islice(
773 ie_entries, playliststart, playlistend))
b82f815f
PH
774 n_entries = len(entries)
775 self.to_screen(
776 "[%s] playlist %s: Downloading %d videos" %
777 (ie_result['extractor'], playlist, n_entries))
8222d8de 778
ff815fe6
MS
779 if self.params.get('playlistreverse', False):
780 entries = entries[::-1]
781
fe7e0c98 782 for i, entry in enumerate(entries, 1):
734ea11e 783 self.to_screen('[download] Downloading video %s of %s' % (i, n_entries))
8222d8de 784 extra = {
c6b4132a 785 'n_entries': n_entries,
fe7e0c98 786 'playlist': playlist,
a1cf99d0
PH
787 'playlist_id': ie_result.get('id'),
788 'playlist_title': ie_result.get('title'),
fe7e0c98 789 'playlist_index': i + playliststart,
b6c45014 790 'extractor': ie_result['extractor'],
9103bbc5 791 'webpage_url': ie_result['webpage_url'],
29eb5174 792 'webpage_url_basename': url_basename(ie_result['webpage_url']),
be97abc2 793 'extractor_key': ie_result['extractor_key'],
fe7e0c98 794 }
7012b23c 795
442c37b7 796 reason = self._match_entry(entry, incomplete=True)
7012b23c 797 if reason is not None:
6febd1c1 798 self.to_screen('[download] ' + reason)
7012b23c
PH
799 continue
800
8222d8de
JMF
801 entry_result = self.process_ie_result(entry,
802 download=download,
803 extra_info=extra)
804 playlist_results.append(entry_result)
805 ie_result['entries'] = playlist_results
806 return ie_result
807 elif result_type == 'compat_list':
c9bf4114
PH
808 self.report_warning(
809 'Extractor %s returned a compat_list result. '
810 'It needs to be updated.' % ie_result.get('extractor'))
5f6a1245 811
8222d8de 812 def _fixup(r):
9e1a5b84
JW
813 self.add_extra_info(
814 r,
9103bbc5
JMF
815 {
816 'extractor': ie_result['extractor'],
817 'webpage_url': ie_result['webpage_url'],
29eb5174 818 'webpage_url_basename': url_basename(ie_result['webpage_url']),
be97abc2 819 'extractor_key': ie_result['extractor_key'],
9e1a5b84
JW
820 }
821 )
8222d8de
JMF
822 return r
823 ie_result['entries'] = [
b6c45014 824 self.process_ie_result(_fixup(r), download, extra_info)
8222d8de
JMF
825 for r in ie_result['entries']
826 ]
827 return ie_result
828 else:
829 raise Exception('Invalid result type: %s' % result_type)
830
083c9df9
PH
831 def _apply_format_filter(self, format_spec, available_formats):
832 " Returns a tuple of the remaining format_spec and filtered formats "
833
834 OPERATORS = {
835 '<': operator.lt,
836 '<=': operator.le,
837 '>': operator.gt,
838 '>=': operator.ge,
839 '=': operator.eq,
840 '!=': operator.ne,
841 }
842 operator_rex = re.compile(r'''(?x)\s*\[
2ec19e95 843 (?P<key>width|height|tbr|abr|vbr|asr|filesize|fps)
083c9df9
PH
844 \s*(?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
845 (?P<value>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?)
846 \]$
847 ''' % '|'.join(map(re.escape, OPERATORS.keys())))
848 m = operator_rex.search(format_spec)
9ddb6925
S
849 if m:
850 try:
851 comparison_value = int(m.group('value'))
852 except ValueError:
853 comparison_value = parse_filesize(m.group('value'))
854 if comparison_value is None:
855 comparison_value = parse_filesize(m.group('value') + 'B')
856 if comparison_value is None:
857 raise ValueError(
858 'Invalid value %r in format specification %r' % (
859 m.group('value'), format_spec))
860 op = OPERATORS[m.group('op')]
861
083c9df9 862 if not m:
9ddb6925
S
863 STR_OPERATORS = {
864 '=': operator.eq,
865 '!=': operator.ne,
866 }
867 str_operator_rex = re.compile(r'''(?x)\s*\[
868 \s*(?P<key>ext|acodec|vcodec|container|protocol)
869 \s*(?P<op>%s)(?P<none_inclusive>\s*\?)?
870 \s*(?P<value>[a-zA-Z0-9_-]+)
871 \s*\]$
872 ''' % '|'.join(map(re.escape, STR_OPERATORS.keys())))
873 m = str_operator_rex.search(format_spec)
874 if m:
875 comparison_value = m.group('value')
876 op = STR_OPERATORS[m.group('op')]
083c9df9 877
9ddb6925
S
878 if not m:
879 raise ValueError('Invalid format specification %r' % format_spec)
083c9df9
PH
880
881 def _filter(f):
882 actual_value = f.get(m.group('key'))
883 if actual_value is None:
884 return m.group('none_inclusive')
885 return op(actual_value, comparison_value)
886 new_formats = [f for f in available_formats if _filter(f)]
887
888 new_format_spec = format_spec[:-len(m.group(0))]
889 if not new_format_spec:
890 new_format_spec = 'best'
891
892 return (new_format_spec, new_formats)
893
a9c58ad9 894 def select_format(self, format_spec, available_formats):
083c9df9
PH
895 while format_spec.endswith(']'):
896 format_spec, available_formats = self._apply_format_filter(
897 format_spec, available_formats)
898 if not available_formats:
899 return None
900
a9c58ad9
JMF
901 if format_spec == 'best' or format_spec is None:
902 return available_formats[-1]
903 elif format_spec == 'worst':
904 return available_formats[0]
ba7678f9
PH
905 elif format_spec == 'bestaudio':
906 audio_formats = [
907 f for f in available_formats
908 if f.get('vcodec') == 'none']
909 if audio_formats:
910 return audio_formats[-1]
911 elif format_spec == 'worstaudio':
912 audio_formats = [
913 f for f in available_formats
914 if f.get('vcodec') == 'none']
915 if audio_formats:
916 return audio_formats[0]
bc6d5978
JMF
917 elif format_spec == 'bestvideo':
918 video_formats = [
919 f for f in available_formats
920 if f.get('acodec') == 'none']
921 if video_formats:
922 return video_formats[-1]
923 elif format_spec == 'worstvideo':
924 video_formats = [
925 f for f in available_formats
926 if f.get('acodec') == 'none']
927 if video_formats:
928 return video_formats[0]
a9c58ad9 929 else:
0217c783 930 extensions = ['mp4', 'flv', 'webm', '3gp', 'm4a', 'mp3', 'ogg', 'aac', 'wav']
49e86983
JMF
931 if format_spec in extensions:
932 filter_f = lambda f: f['ext'] == format_spec
933 else:
934 filter_f = lambda f: f['format_id'] == format_spec
fe7e0c98 935 matches = list(filter(filter_f, available_formats))
a9c58ad9
JMF
936 if matches:
937 return matches[-1]
938 return None
939
e5660ee6
JMF
940 def _calc_headers(self, info_dict):
941 res = std_headers.copy()
942
943 add_headers = info_dict.get('http_headers')
944 if add_headers:
945 res.update(add_headers)
946
947 cookies = self._calc_cookies(info_dict)
948 if cookies:
949 res['Cookie'] = cookies
950
951 return res
952
953 def _calc_cookies(self, info_dict):
954 class _PseudoRequest(object):
955 def __init__(self, url):
956 self.url = url
957 self.headers = {}
958 self.unverifiable = False
959
960 def add_unredirected_header(self, k, v):
961 self.headers[k] = v
962
963 def get_full_url(self):
964 return self.url
1070711d
JMF
965
966 def is_unverifiable(self):
967 return self.unverifiable
e5660ee6 968
4b405cfc
JMF
969 def has_header(self, h):
970 return h in self.headers
971
b3a286d6
JMF
972 def get_header(self, h, default=None):
973 return self.headers.get(h, default)
974
e5660ee6
JMF
975 pr = _PseudoRequest(info_dict['url'])
976 self.cookiejar.add_cookie_header(pr)
977 return pr.headers.get('Cookie')
978
dd82ffea
JMF
979 def process_video_result(self, info_dict, download=True):
980 assert info_dict.get('_type', 'video') == 'video'
981
bec1fad2
PH
982 if 'id' not in info_dict:
983 raise ExtractorError('Missing "id" field in extractor result')
984 if 'title' not in info_dict:
985 raise ExtractorError('Missing "title" field in extractor result')
986
dd82ffea
JMF
987 if 'playlist' not in info_dict:
988 # It isn't part of a playlist
989 info_dict['playlist'] = None
990 info_dict['playlist_index'] = None
991
d5519808 992 thumbnails = info_dict.get('thumbnails')
cfb56d1a
PH
993 if thumbnails is None:
994 thumbnail = info_dict.get('thumbnail')
995 if thumbnail:
a7a14d95 996 info_dict['thumbnails'] = thumbnails = [{'url': thumbnail}]
d5519808 997 if thumbnails:
be6d7229 998 thumbnails.sort(key=lambda t: (
cfb56d1a
PH
999 t.get('preference'), t.get('width'), t.get('height'),
1000 t.get('id'), t.get('url')))
f6c24009 1001 for i, t in enumerate(thumbnails):
d5519808
PH
1002 if 'width' in t and 'height' in t:
1003 t['resolution'] = '%dx%d' % (t['width'], t['height'])
f6c24009
PH
1004 if t.get('id') is None:
1005 t['id'] = '%d' % i
d5519808
PH
1006
1007 if thumbnails and 'thumbnail' not in info_dict:
1008 info_dict['thumbnail'] = thumbnails[-1]['url']
1009
c9ae7b95 1010 if 'display_id' not in info_dict and 'id' in info_dict:
0afef30b
PH
1011 info_dict['display_id'] = info_dict['id']
1012
955c4514 1013 if info_dict.get('upload_date') is None and info_dict.get('timestamp') is not None:
706d7d4e
S
1014 # Working around negative timestamps in Windows
1015 # (see http://bugs.python.org/issue1646728)
1016 if info_dict['timestamp'] < 0 and os.name == 'nt':
1017 info_dict['timestamp'] = 0
9d2ecdbc 1018 upload_date = datetime.datetime.utcfromtimestamp(
955c4514 1019 info_dict['timestamp'])
9d2ecdbc
PH
1020 info_dict['upload_date'] = upload_date.strftime('%Y%m%d')
1021
a504ced0 1022 if self.params.get('listsubtitles', False):
360e1ca5
JMF
1023 if 'automatic_captions' in info_dict:
1024 self.list_subtitles(info_dict['id'], info_dict.get('automatic_captions'), 'automatic captions')
1025 self.list_subtitles(info_dict['id'], info_dict.get('subtitles'), 'subtitles')
a504ced0 1026 return
360e1ca5
JMF
1027 info_dict['requested_subtitles'] = self.process_subtitles(
1028 info_dict['id'], info_dict.get('subtitles'),
1029 info_dict.get('automatic_captions'))
a504ced0 1030
6ff000b8 1031 # This extractors handle format selection themselves
6febd1c1 1032 if info_dict['extractor'] in ['Youku']:
12893efe
JMF
1033 if download:
1034 self.process_info(info_dict)
6ff000b8
JMF
1035 return info_dict
1036
dd82ffea
JMF
1037 # We now pick which formats have to be downloaded
1038 if info_dict.get('formats') is None:
1039 # There's only one format available
1040 formats = [info_dict]
1041 else:
1042 formats = info_dict['formats']
1043
db95dc13
PH
1044 if not formats:
1045 raise ExtractorError('No video formats found!')
1046
dd82ffea 1047 # We check that all the formats have the format and format_id fields
db95dc13 1048 for i, format in enumerate(formats):
bec1fad2
PH
1049 if 'url' not in format:
1050 raise ExtractorError('Missing "url" key in result (index %d)' % i)
1051
dd82ffea 1052 if format.get('format_id') is None:
8016c922 1053 format['format_id'] = compat_str(i)
8c51aa65 1054 if format.get('format') is None:
6febd1c1 1055 format['format'] = '{id} - {res}{note}'.format(
8c51aa65
JMF
1056 id=format['format_id'],
1057 res=self.format_resolution(format),
6febd1c1 1058 note=' ({0})'.format(format['format_note']) if format.get('format_note') is not None else '',
8c51aa65 1059 )
c1002e96
PH
1060 # Automatically determine file extension if missing
1061 if 'ext' not in format:
cce929ea 1062 format['ext'] = determine_ext(format['url']).lower()
e5660ee6
JMF
1063 # Add HTTP headers, so that external programs can use them from the
1064 # json output
1065 full_format_info = info_dict.copy()
1066 full_format_info.update(format)
1067 format['http_headers'] = self._calc_headers(full_format_info)
dd82ffea 1068
99e206d5
JMF
1069 format_limit = self.params.get('format_limit', None)
1070 if format_limit:
f4d96df0
PH
1071 formats = list(takewhile_inclusive(
1072 lambda f: f['format_id'] != format_limit, formats
1073 ))
4bcc7bd1
PH
1074
1075 # TODO Central sorting goes here
99e206d5 1076
f89197d7 1077 if formats[0] is not info_dict:
b3d9ef88
JMF
1078 # only set the 'formats' fields if the original info_dict list them
1079 # otherwise we end up with a circular reference, the first (and unique)
f89197d7 1080 # element in the 'formats' field in info_dict is info_dict itself,
b3d9ef88
JMF
1081 # wich can't be exported to json
1082 info_dict['formats'] = formats
cfb56d1a 1083 if self.params.get('listformats'):
bfaae0a7 1084 self.list_formats(info_dict)
1085 return
cfb56d1a
PH
1086 if self.params.get('list_thumbnails'):
1087 self.list_thumbnails(info_dict)
1088 return
bfaae0a7 1089
de3ef3ed 1090 req_format = self.params.get('format')
a9c58ad9
JMF
1091 if req_format is None:
1092 req_format = 'best'
dd82ffea 1093 formats_to_download = []
dd82ffea 1094 # The -1 is for supporting YoutubeIE
a9c58ad9 1095 if req_format in ('-1', 'all'):
dd82ffea
JMF
1096 formats_to_download = formats
1097 else:
1de33faf
PH
1098 for rfstr in req_format.split(','):
1099 # We can accept formats requested in the format: 34/5/best, we pick
1100 # the first that is available, starting from left
1101 req_formats = rfstr.split('/')
1102 for rf in req_formats:
1103 if re.match(r'.+?\+.+?', rf) is not None:
1104 # Two formats have been requested like '137+139'
1105 format_1, format_2 = rf.split('+')
1106 formats_info = (self.select_format(format_1, formats),
9e1a5b84 1107 self.select_format(format_2, formats))
1de33faf 1108 if all(formats_info):
c2954908
JMF
1109 # The first format must contain the video and the
1110 # second the audio
1111 if formats_info[0].get('vcodec') == 'none':
1112 self.report_error('The first format must '
9e1a5b84
JW
1113 'contain the video, try using '
1114 '"-f %s+%s"' % (format_2, format_1))
c2954908 1115 return
45598f15
PH
1116 output_ext = (
1117 formats_info[0]['ext']
1118 if self.params.get('merge_output_format') is None
1119 else self.params['merge_output_format'])
1de33faf
PH
1120 selected_format = {
1121 'requested_formats': formats_info,
300753a0
S
1122 'format': '%s+%s' % (formats_info[0].get('format'),
1123 formats_info[1].get('format')),
6d593c32
S
1124 'format_id': '%s+%s' % (formats_info[0].get('format_id'),
1125 formats_info[1].get('format_id')),
f90ad273
PH
1126 'width': formats_info[0].get('width'),
1127 'height': formats_info[0].get('height'),
1128 'resolution': formats_info[0].get('resolution'),
1129 'fps': formats_info[0].get('fps'),
1130 'vcodec': formats_info[0].get('vcodec'),
1131 'vbr': formats_info[0].get('vbr'),
6271f1ca 1132 'stretched_ratio': formats_info[0].get('stretched_ratio'),
f90ad273
PH
1133 'acodec': formats_info[1].get('acodec'),
1134 'abr': formats_info[1].get('abr'),
45598f15 1135 'ext': output_ext,
1de33faf
PH
1136 }
1137 else:
1138 selected_format = None
6350728b 1139 else:
1de33faf
PH
1140 selected_format = self.select_format(rf, formats)
1141 if selected_format is not None:
1142 formats_to_download.append(selected_format)
1143 break
dd82ffea 1144 if not formats_to_download:
6febd1c1 1145 raise ExtractorError('requested format not available',
78a3a9f8 1146 expected=True)
dd82ffea
JMF
1147
1148 if download:
1149 if len(formats_to_download) > 1:
6febd1c1 1150 self.to_screen('[info] %s: downloading video in %s formats' % (info_dict['id'], len(formats_to_download)))
dd82ffea
JMF
1151 for format in formats_to_download:
1152 new_info = dict(info_dict)
1153 new_info.update(format)
1154 self.process_info(new_info)
1155 # We update the info dict with the best quality format (backwards compatibility)
1156 info_dict.update(formats_to_download[-1])
1157 return info_dict
1158
360e1ca5 1159 def process_subtitles(self, video_id, available_subs, available_autocaps):
a504ced0 1160 """Select the requested subtitles and their format"""
360e1ca5
JMF
1161 if available_autocaps and self.params.get('writeautomaticsub'):
1162 available_subs = available_subs.copy()
1163 for lang, cap_info in available_autocaps.items():
1164 if lang not in available_subs:
1165 available_subs[lang] = cap_info
1166
a504ced0
JMF
1167 if not available_subs:
1168 return available_subs
1169
1170 if self.params.get('allsubtitles', False):
1171 requested_langs = available_subs.keys()
1172 else:
1173 if self.params.get('subtitleslangs', False):
1174 requested_langs = self.params.get('subtitleslangs')
1175 elif 'en' in available_subs:
1176 requested_langs = ['en']
1177 else:
1178 requested_langs = [list(available_subs.keys())[0]]
1179
1180 formats_query = self.params.get('subtitlesformat', 'best')
1181 formats_preference = formats_query.split('/') if formats_query else []
1182 subs = {}
1183 for lang in requested_langs:
1184 formats = available_subs.get(lang)
1185 if formats is None:
1186 self.report_warning('%s subtitles not available for %s' % (lang, video_id))
1187 continue
1188 if isinstance(formats, compat_str):
1189 # TODO: convert all IE with subtitles support to the new format
1190 # and remove this
1191 subs[lang] = {
1192 'ext': formats_preference[0],
1193 'data': formats,
1194 }
1195 continue
1196 for ext in formats_preference:
1197 if ext == 'best':
1198 f = formats[-1]
1199 break
1200 matches = list(filter(lambda f: f['ext'] == ext, formats))
1201 if matches:
1202 f = matches[-1]
1203 break
1204 else:
1205 f = formats[-1]
1206 self.report_warning(
1207 'No subtitle format found matching "%s" for language %s, '
1208 'using %s' % (formats_query, lang, f['ext']))
1209 subs[lang] = f
1210 return subs
1211
8222d8de
JMF
1212 def process_info(self, info_dict):
1213 """Process a single resolved IE result."""
1214
1215 assert info_dict.get('_type', 'video') == 'video'
fd288278
PH
1216
1217 max_downloads = self.params.get('max_downloads')
1218 if max_downloads is not None:
1219 if self._num_downloads >= int(max_downloads):
1220 raise MaxDownloadsReached()
8222d8de
JMF
1221
1222 info_dict['fulltitle'] = info_dict['title']
1223 if len(info_dict['title']) > 200:
6febd1c1 1224 info_dict['title'] = info_dict['title'][:197] + '...'
8222d8de
JMF
1225
1226 # Keep for backwards compatibility
1227 info_dict['stitle'] = info_dict['title']
1228
11b85ce6 1229 if 'format' not in info_dict:
8222d8de
JMF
1230 info_dict['format'] = info_dict['ext']
1231
442c37b7 1232 reason = self._match_entry(info_dict, incomplete=False)
8222d8de 1233 if reason is not None:
6febd1c1 1234 self.to_screen('[download] ' + reason)
8222d8de
JMF
1235 return
1236
fd288278 1237 self._num_downloads += 1
8222d8de 1238
e72c7e41 1239 info_dict['_filename'] = filename = self.prepare_filename(info_dict)
8222d8de
JMF
1240
1241 # Forced printings
1242 if self.params.get('forcetitle', False):
0783b09b 1243 self.to_stdout(info_dict['fulltitle'])
8222d8de 1244 if self.params.get('forceid', False):
0783b09b 1245 self.to_stdout(info_dict['id'])
8222d8de 1246 if self.params.get('forceurl', False):
16ae61f6 1247 if info_dict.get('requested_formats') is not None:
1248 for f in info_dict['requested_formats']:
1249 self.to_stdout(f['url'] + f.get('play_path', ''))
1250 else:
1251 # For RTMP URLs, also include the playpath
1252 self.to_stdout(info_dict['url'] + info_dict.get('play_path', ''))
216d71d0 1253 if self.params.get('forcethumbnail', False) and info_dict.get('thumbnail') is not None:
0783b09b 1254 self.to_stdout(info_dict['thumbnail'])
216d71d0 1255 if self.params.get('forcedescription', False) and info_dict.get('description') is not None:
0783b09b 1256 self.to_stdout(info_dict['description'])
8222d8de 1257 if self.params.get('forcefilename', False) and filename is not None:
0783b09b 1258 self.to_stdout(filename)
525ef922
PH
1259 if self.params.get('forceduration', False) and info_dict.get('duration') is not None:
1260 self.to_stdout(formatSeconds(info_dict['duration']))
8222d8de 1261 if self.params.get('forceformat', False):
0783b09b 1262 self.to_stdout(info_dict['format'])
9d153818 1263 if self.params.get('forcejson', False):
0783b09b 1264 self.to_stdout(json.dumps(info_dict))
8222d8de
JMF
1265
1266 # Do nothing else if in simulate mode
1267 if self.params.get('simulate', False):
1268 return
1269
1270 if filename is None:
1271 return
1272
1273 try:
1274 dn = os.path.dirname(encodeFilename(filename))
d26e981d 1275 if dn and not os.path.exists(dn):
8222d8de
JMF
1276 os.makedirs(dn)
1277 except (OSError, IOError) as err:
6febd1c1 1278 self.report_error('unable to create directory ' + compat_str(err))
8222d8de
JMF
1279 return
1280
1281 if self.params.get('writedescription', False):
6febd1c1 1282 descfn = filename + '.description'
7b6fefc9 1283 if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(descfn)):
6febd1c1 1284 self.to_screen('[info] Video description is already present')
f00fd51d
JMF
1285 elif info_dict.get('description') is None:
1286 self.report_warning('There\'s no description to write.')
7b6fefc9
PH
1287 else:
1288 try:
6febd1c1 1289 self.to_screen('[info] Writing video description to: ' + descfn)
7b6fefc9
PH
1290 with io.open(encodeFilename(descfn), 'w', encoding='utf-8') as descfile:
1291 descfile.write(info_dict['description'])
7b6fefc9 1292 except (OSError, IOError):
6febd1c1 1293 self.report_error('Cannot write description file ' + descfn)
7b6fefc9 1294 return
8222d8de 1295
1fb07d10 1296 if self.params.get('writeannotations', False):
6febd1c1 1297 annofn = filename + '.annotations.xml'
7b6fefc9 1298 if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(annofn)):
6febd1c1 1299 self.to_screen('[info] Video annotations are already present')
7b6fefc9
PH
1300 else:
1301 try:
6febd1c1 1302 self.to_screen('[info] Writing video annotations to: ' + annofn)
7b6fefc9
PH
1303 with io.open(encodeFilename(annofn), 'w', encoding='utf-8') as annofile:
1304 annofile.write(info_dict['annotations'])
1305 except (KeyError, TypeError):
6febd1c1 1306 self.report_warning('There are no annotations to write.')
7b6fefc9 1307 except (OSError, IOError):
6febd1c1 1308 self.report_error('Cannot write annotations file: ' + annofn)
7b6fefc9 1309 return
1fb07d10 1310
c4a91be7 1311 subtitles_are_requested = any([self.params.get('writesubtitles', False),
0b7f3118 1312 self.params.get('writeautomaticsub')])
c4a91be7 1313
c84dd8a9 1314 if subtitles_are_requested and info_dict.get('requested_subtitles'):
8222d8de
JMF
1315 # subtitles download errors are already managed as troubles in relevant IE
1316 # that way it will silently go on when used with unsupporting IE
c84dd8a9 1317 subtitles = info_dict['requested_subtitles']
a504ced0
JMF
1318 for sub_lang, sub_info in subtitles.items():
1319 sub_format = sub_info['ext']
1320 if sub_info.get('data') is not None:
1321 sub_data = sub_info['data']
1322 else:
1323 try:
1324 uf = self.urlopen(sub_info['url'])
1325 sub_data = uf.read().decode('utf-8')
1326 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
1327 self.report_warning('Unable to download subtitle for "%s": %s' %
1328 (sub_lang, compat_str(err)))
1329 continue
8222d8de 1330 try:
d4051a8e 1331 sub_filename = subtitles_filename(filename, sub_lang, sub_format)
7b6fefc9 1332 if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(sub_filename)):
6febd1c1 1333 self.to_screen('[info] Video subtitle %s.%s is already_present' % (sub_lang, sub_format))
7b6fefc9 1334 else:
6febd1c1 1335 self.to_screen('[info] Writing video subtitles to: ' + sub_filename)
7b6fefc9 1336 with io.open(encodeFilename(sub_filename), 'w', encoding='utf-8') as subfile:
a504ced0 1337 subfile.write(sub_data)
8222d8de 1338 except (OSError, IOError):
e4db1951 1339 self.report_error('Cannot write subtitles file ' + sub_filename)
8222d8de
JMF
1340 return
1341
8222d8de 1342 if self.params.get('writeinfojson', False):
6febd1c1 1343 infofn = os.path.splitext(filename)[0] + '.info.json'
7b6fefc9 1344 if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(infofn)):
6febd1c1 1345 self.to_screen('[info] Video description metadata is already present')
7b6fefc9 1346 else:
6febd1c1 1347 self.to_screen('[info] Writing video description metadata as JSON to: ' + infofn)
7b6fefc9 1348 try:
92120217 1349 write_json_file(info_dict, infofn)
7b6fefc9 1350 except (OSError, IOError):
6febd1c1 1351 self.report_error('Cannot write metadata to JSON file ' + infofn)
7b6fefc9 1352 return
8222d8de 1353
ec82d85a 1354 self._write_thumbnails(info_dict, filename)
8222d8de
JMF
1355
1356 if not self.params.get('skip_download', False):
4340deca
P
1357 try:
1358 def dl(name, info):
a055469f 1359 fd = get_suitable_downloader(info, self.params)(self, self.params)
4340deca
P
1360 for ph in self._progress_hooks:
1361 fd.add_progress_hook(ph)
1362 if self.params.get('verbose'):
1363 self.to_stdout('[debug] Invoking downloader on %r' % info.get('url'))
1364 return fd.download(name, info)
ee69b99a 1365
4340deca
P
1366 if info_dict.get('requested_formats') is not None:
1367 downloaded = []
1368 success = True
1369 merger = FFmpegMergerPP(self, not self.params.get('keepvideo'))
1370 if not merger._executable:
1371 postprocessors = []
1372 self.report_warning('You have requested multiple '
1373 'formats but ffmpeg or avconv are not installed.'
1374 ' The formats won\'t be merged')
6350728b 1375 else:
4340deca
P
1376 postprocessors = [merger]
1377 for f in info_dict['requested_formats']:
1378 new_info = dict(info_dict)
1379 new_info.update(f)
1380 fname = self.prepare_filename(new_info)
1381 fname = prepend_extension(fname, 'f%s' % f['format_id'])
1382 downloaded.append(fname)
1383 partial_success = dl(fname, new_info)
1384 success = success and partial_success
1385 info_dict['__postprocessors'] = postprocessors
1386 info_dict['__files_to_merge'] = downloaded
1387 else:
1388 # Just a single file
1389 success = dl(filename, info_dict)
1390 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
1391 self.report_error('unable to download video data: %s' % str(err))
1392 return
1393 except (OSError, IOError) as err:
1394 raise UnavailableVideoError(err)
1395 except (ContentTooShortError, ) as err:
1396 self.report_error('content too short (expected %s bytes and served %s)' % (err.expected, err.downloaded))
1397 return
8222d8de
JMF
1398
1399 if success:
6271f1ca 1400 # Fixup content
62cd676c
PH
1401 fixup_policy = self.params.get('fixup')
1402 if fixup_policy is None:
1403 fixup_policy = 'detect_or_warn'
1404
6271f1ca
PH
1405 stretched_ratio = info_dict.get('stretched_ratio')
1406 if stretched_ratio is not None and stretched_ratio != 1:
6271f1ca
PH
1407 if fixup_policy == 'warn':
1408 self.report_warning('%s: Non-uniform pixel ratio (%s)' % (
1409 info_dict['id'], stretched_ratio))
1410 elif fixup_policy == 'detect_or_warn':
1411 stretched_pp = FFmpegFixupStretchedPP(self)
1412 if stretched_pp.available:
1413 info_dict.setdefault('__postprocessors', [])
1414 info_dict['__postprocessors'].append(stretched_pp)
1415 else:
1416 self.report_warning(
1417 '%s: Non-uniform pixel ratio (%s). Install ffmpeg or avconv to fix this automatically.' % (
1418 info_dict['id'], stretched_ratio))
1419 else:
62cd676c
PH
1420 assert fixup_policy in ('ignore', 'never')
1421
1422 if info_dict.get('requested_formats') is None and info_dict.get('container') == 'm4a_dash':
1423 if fixup_policy == 'warn':
1424 self.report_warning('%s: writing DASH m4a. Only some players support this container.' % (
1425 info_dict['id']))
1426 elif fixup_policy == 'detect_or_warn':
1427 fixup_pp = FFmpegFixupM4aPP(self)
1428 if fixup_pp.available:
1429 info_dict.setdefault('__postprocessors', [])
1430 info_dict['__postprocessors'].append(fixup_pp)
1431 else:
1432 self.report_warning(
1433 '%s: writing DASH m4a. Only some players support this container. Install ffmpeg or avconv to fix this automatically.' % (
1434 info_dict['id']))
1435 else:
1436 assert fixup_policy in ('ignore', 'never')
6271f1ca 1437
8222d8de
JMF
1438 try:
1439 self.post_process(filename, info_dict)
1440 except (PostProcessingError) as err:
6febd1c1 1441 self.report_error('postprocessing: %s' % str(err))
8222d8de 1442 return
cd58dc3e 1443 self.record_download_archive(info_dict)
8222d8de
JMF
1444
1445 def download(self, url_list):
1446 """Download a given list of URLs."""
acd69589 1447 outtmpl = self.params.get('outtmpl', DEFAULT_OUTTMPL)
0c75c3fa 1448 if (len(url_list) > 1 and
acd69589 1449 '%' not in outtmpl
0c75c3fa 1450 and self.params.get('max_downloads') != 1):
acd69589 1451 raise SameFileError(outtmpl)
8222d8de
JMF
1452
1453 for url in url_list:
1454 try:
5f6a1245 1455 # It also downloads the videos
63e0be34 1456 res = self.extract_info(url)
8222d8de 1457 except UnavailableVideoError:
6febd1c1 1458 self.report_error('unable to download video')
8222d8de 1459 except MaxDownloadsReached:
6febd1c1 1460 self.to_screen('[info] Maximum number of downloaded files reached.')
8222d8de 1461 raise
63e0be34
PH
1462 else:
1463 if self.params.get('dump_single_json', False):
1464 self.to_stdout(json.dumps(res))
8222d8de
JMF
1465
1466 return self._download_retcode
1467
1dcc4c0c 1468 def download_with_info_file(self, info_filename):
395293a8 1469 with io.open(info_filename, 'r', encoding='utf-8') as f:
1dcc4c0c 1470 info = json.load(f)
d4943898
JMF
1471 try:
1472 self.process_ie_result(info, download=True)
1473 except DownloadError:
1474 webpage_url = info.get('webpage_url')
1475 if webpage_url is not None:
6febd1c1 1476 self.report_warning('The info failed to download, trying with "%s"' % webpage_url)
d4943898
JMF
1477 return self.download([webpage_url])
1478 else:
1479 raise
1480 return self._download_retcode
1dcc4c0c 1481
8222d8de
JMF
1482 def post_process(self, filename, ie_info):
1483 """Run all the postprocessors on the given file."""
1484 info = dict(ie_info)
1485 info['filepath'] = filename
6350728b
JMF
1486 pps_chain = []
1487 if ie_info.get('__postprocessors') is not None:
1488 pps_chain.extend(ie_info['__postprocessors'])
1489 pps_chain.extend(self._pps)
1490 for pp in pps_chain:
f3ff1a36
JMF
1491 keep_video = None
1492 old_filename = info['filepath']
8222d8de 1493 try:
f3ff1a36 1494 keep_video_wish, info = pp.run(info)
8222d8de
JMF
1495 if keep_video_wish is not None:
1496 if keep_video_wish:
1497 keep_video = keep_video_wish
1498 elif keep_video is None:
1499 # No clear decision yet, let IE decide
1500 keep_video = keep_video_wish
1501 except PostProcessingError as e:
bbcbf4d4 1502 self.report_error(e.msg)
f3ff1a36
JMF
1503 if keep_video is False and not self.params.get('keepvideo', False):
1504 try:
1505 self.to_screen('Deleting original file %s (pass -k to keep)' % old_filename)
1506 os.remove(encodeFilename(old_filename))
1507 except (IOError, OSError):
1508 self.report_warning('Unable to remove downloaded video file')
c1c9a79c 1509
5db07df6
PH
1510 def _make_archive_id(self, info_dict):
1511 # Future-proof against any change in case
1512 # and backwards compatibility with prior versions
d31209a1 1513 extractor = info_dict.get('extractor_key')
7012b23c
PH
1514 if extractor is None:
1515 if 'id' in info_dict:
1516 extractor = info_dict.get('ie_key') # key in a playlist
1517 if extractor is None:
5db07df6 1518 return None # Incomplete video information
6febd1c1 1519 return extractor.lower() + ' ' + info_dict['id']
5db07df6
PH
1520
1521 def in_download_archive(self, info_dict):
1522 fn = self.params.get('download_archive')
1523 if fn is None:
1524 return False
1525
1526 vid_id = self._make_archive_id(info_dict)
1527 if vid_id is None:
7012b23c 1528 return False # Incomplete video information
5db07df6 1529
c1c9a79c
PH
1530 try:
1531 with locked_file(fn, 'r', encoding='utf-8') as archive_file:
1532 for line in archive_file:
1533 if line.strip() == vid_id:
1534 return True
1535 except IOError as ioe:
1536 if ioe.errno != errno.ENOENT:
1537 raise
1538 return False
1539
1540 def record_download_archive(self, info_dict):
1541 fn = self.params.get('download_archive')
1542 if fn is None:
1543 return
5db07df6
PH
1544 vid_id = self._make_archive_id(info_dict)
1545 assert vid_id
c1c9a79c 1546 with locked_file(fn, 'a', encoding='utf-8') as archive_file:
6febd1c1 1547 archive_file.write(vid_id + '\n')
dd82ffea 1548
8c51aa65 1549 @staticmethod
8abeeb94 1550 def format_resolution(format, default='unknown'):
fb04e403
PH
1551 if format.get('vcodec') == 'none':
1552 return 'audio only'
f49d89ee
PH
1553 if format.get('resolution') is not None:
1554 return format['resolution']
8c51aa65
JMF
1555 if format.get('height') is not None:
1556 if format.get('width') is not None:
6febd1c1 1557 res = '%sx%s' % (format['width'], format['height'])
8c51aa65 1558 else:
6febd1c1 1559 res = '%sp' % format['height']
f49d89ee 1560 elif format.get('width') is not None:
6febd1c1 1561 res = '?x%d' % format['width']
8c51aa65 1562 else:
8abeeb94 1563 res = default
8c51aa65
JMF
1564 return res
1565
c57f7757
PH
1566 def _format_note(self, fdict):
1567 res = ''
1568 if fdict.get('ext') in ['f4f', 'f4m']:
1569 res += '(unsupported) '
1570 if fdict.get('format_note') is not None:
1571 res += fdict['format_note'] + ' '
1572 if fdict.get('tbr') is not None:
1573 res += '%4dk ' % fdict['tbr']
1574 if fdict.get('container') is not None:
1575 if res:
1576 res += ', '
1577 res += '%s container' % fdict['container']
1578 if (fdict.get('vcodec') is not None and
1579 fdict.get('vcodec') != 'none'):
1580 if res:
1581 res += ', '
1582 res += fdict['vcodec']
91c7271a 1583 if fdict.get('vbr') is not None:
c57f7757
PH
1584 res += '@'
1585 elif fdict.get('vbr') is not None and fdict.get('abr') is not None:
1586 res += 'video@'
1587 if fdict.get('vbr') is not None:
1588 res += '%4dk' % fdict['vbr']
fbb21cf5
PH
1589 if fdict.get('fps') is not None:
1590 res += ', %sfps' % fdict['fps']
c57f7757
PH
1591 if fdict.get('acodec') is not None:
1592 if res:
1593 res += ', '
1594 if fdict['acodec'] == 'none':
1595 res += 'video only'
1596 else:
1597 res += '%-5s' % fdict['acodec']
1598 elif fdict.get('abr') is not None:
1599 if res:
1600 res += ', '
1601 res += 'audio'
1602 if fdict.get('abr') is not None:
1603 res += '@%3dk' % fdict['abr']
1604 if fdict.get('asr') is not None:
1605 res += ' (%5dHz)' % fdict['asr']
1606 if fdict.get('filesize') is not None:
1607 if res:
1608 res += ', '
1609 res += format_bytes(fdict['filesize'])
9732d77e
PH
1610 elif fdict.get('filesize_approx') is not None:
1611 if res:
1612 res += ', '
1613 res += '~' + format_bytes(fdict['filesize_approx'])
c57f7757 1614 return res
91c7271a 1615
c57f7757 1616 def list_formats(self, info_dict):
02dbf93f 1617 def line(format, idlen=20):
6febd1c1 1618 return (('%-' + compat_str(idlen + 1) + 's%-10s%-12s%s') % (
8c51aa65
JMF
1619 format['format_id'],
1620 format['ext'],
8c51aa65 1621 self.format_resolution(format),
c57f7757 1622 self._format_note(format),
02dbf93f 1623 ))
57dd9a8f 1624
94badb25 1625 formats = info_dict.get('formats', [info_dict])
6febd1c1 1626 idlen = max(len('format code'),
02dbf93f 1627 max(len(f['format_id']) for f in formats))
e65566a9
PH
1628 formats_s = [
1629 line(f, idlen) for f in formats
1630 if f.get('preference') is None or f['preference'] >= -1000]
94badb25 1631 if len(formats) > 1:
c57f7757 1632 formats_s[-1] += (' ' if self._format_note(formats[-1]) else '') + '(best)'
57dd9a8f
PH
1633
1634 header_line = line({
6febd1c1
PH
1635 'format_id': 'format code', 'ext': 'extension',
1636 'resolution': 'resolution', 'format_note': 'note'}, idlen=idlen)
cfb56d1a
PH
1637 self.to_screen(
1638 '[info] Available formats for %s:\n%s\n%s' %
1639 (info_dict['id'], header_line, '\n'.join(formats_s)))
1640
1641 def list_thumbnails(self, info_dict):
1642 thumbnails = info_dict.get('thumbnails')
1643 if not thumbnails:
1644 tn_url = info_dict.get('thumbnail')
1645 if tn_url:
1646 thumbnails = [{'id': '0', 'url': tn_url}]
1647 else:
1648 self.to_screen(
1649 '[info] No thumbnails present for %s' % info_dict['id'])
1650 return
1651
1652 self.to_screen(
1653 '[info] Thumbnails for %s:' % info_dict['id'])
1654 self.to_screen(render_table(
1655 ['ID', 'width', 'height', 'URL'],
1656 [[t['id'], t.get('width', 'unknown'), t.get('height', 'unknown'), t['url']] for t in thumbnails]))
dca08720 1657
360e1ca5 1658 def list_subtitles(self, video_id, subtitles, name='subtitles'):
a504ced0 1659 if not subtitles:
360e1ca5 1660 self.to_screen('%s has no %s' % (video_id, name))
a504ced0 1661 return
a504ced0 1662 self.to_screen(
edab9dbf
JMF
1663 'Available %s for %s:' % (name, video_id))
1664 self.to_screen(render_table(
1665 ['Language', 'formats'],
1666 [[lang, ', '.join(f['ext'] for f in reversed(formats))]
1667 for lang, formats in subtitles.items()]))
a504ced0 1668
dca08720
PH
1669 def urlopen(self, req):
1670 """ Start an HTTP download """
37419b4f 1671
d05cfe06
S
1672 # According to RFC 3986, URLs can not contain non-ASCII characters, however this is not
1673 # always respected by websites, some tend to give out URLs with non percent-encoded
1674 # non-ASCII characters (see telemb.py, ard.py [#3412])
37419b4f 1675 # urllib chokes on URLs with non-ASCII characters (see http://bugs.python.org/issue3991)
d05cfe06
S
1676 # To work around aforementioned issue we will replace request's original URL with
1677 # percent-encoded one
8f9312c3 1678 req_is_string = isinstance(req, compat_basestring)
68b09730 1679 url = req if req_is_string else req.get_full_url()
d05cfe06 1680 url_escaped = escape_url(url)
37419b4f
S
1681
1682 # Substitute URL if any change after escaping
1683 if url != url_escaped:
68b09730 1684 if req_is_string:
37419b4f
S
1685 req = url_escaped
1686 else:
1687 req = compat_urllib_request.Request(
1688 url_escaped, data=req.data, headers=req.headers,
1689 origin_req_host=req.origin_req_host, unverifiable=req.unverifiable)
1690
19a41fc6 1691 return self._opener.open(req, timeout=self._socket_timeout)
dca08720
PH
1692
1693 def print_debug_header(self):
1694 if not self.params.get('verbose'):
1695 return
62fec3b2 1696
4192b51c
PH
1697 if type('') is not compat_str:
1698 # Python 2.6 on SLES11 SP1 (https://github.com/rg3/youtube-dl/issues/3326)
1699 self.report_warning(
1700 'Your Python is broken! Update to a newer and supported version')
1701
c6afed48
PH
1702 stdout_encoding = getattr(
1703 sys.stdout, 'encoding', 'missing (%s)' % type(sys.stdout).__name__)
b0472057 1704 encoding_str = (
734f90bb
PH
1705 '[debug] Encodings: locale %s, fs %s, out %s, pref %s\n' % (
1706 locale.getpreferredencoding(),
1707 sys.getfilesystemencoding(),
c6afed48 1708 stdout_encoding,
b0472057 1709 self.get_encoding()))
4192b51c 1710 write_string(encoding_str, encoding=None)
734f90bb
PH
1711
1712 self._write_string('[debug] youtube-dl version ' + __version__ + '\n')
dca08720
PH
1713 try:
1714 sp = subprocess.Popen(
1715 ['git', 'rev-parse', '--short', 'HEAD'],
1716 stdout=subprocess.PIPE, stderr=subprocess.PIPE,
1717 cwd=os.path.dirname(os.path.abspath(__file__)))
1718 out, err = sp.communicate()
1719 out = out.decode().strip()
1720 if re.match('[0-9a-f]+', out):
734f90bb 1721 self._write_string('[debug] Git HEAD: ' + out + '\n')
dca08720
PH
1722 except:
1723 try:
1724 sys.exc_clear()
1725 except:
1726 pass
d28b5171
PH
1727 self._write_string('[debug] Python version %s - %s\n' % (
1728 platform.python_version(), platform_name()))
1729
1730 exe_versions = FFmpegPostProcessor.get_versions()
4c83c967 1731 exe_versions['rtmpdump'] = rtmpdump_version()
d28b5171
PH
1732 exe_str = ', '.join(
1733 '%s %s' % (exe, v)
1734 for exe, v in sorted(exe_versions.items())
1735 if v
1736 )
1737 if not exe_str:
1738 exe_str = 'none'
1739 self._write_string('[debug] exe versions: %s\n' % exe_str)
dca08720
PH
1740
1741 proxy_map = {}
1742 for handler in self._opener.handlers:
1743 if hasattr(handler, 'proxies'):
1744 proxy_map.update(handler.proxies)
734f90bb 1745 self._write_string('[debug] Proxy map: ' + compat_str(proxy_map) + '\n')
dca08720 1746
58b1f00d
PH
1747 if self.params.get('call_home', False):
1748 ipaddr = self.urlopen('https://yt-dl.org/ip').read().decode('utf-8')
1749 self._write_string('[debug] Public IP address: %s\n' % ipaddr)
1750 latest_version = self.urlopen(
1751 'https://yt-dl.org/latest/version').read().decode('utf-8')
1752 if version_tuple(latest_version) > version_tuple(__version__):
1753 self.report_warning(
1754 'You are using an outdated version (newest version: %s)! '
1755 'See https://yt-dl.org/update if you need help updating.' %
1756 latest_version)
1757
e344693b 1758 def _setup_opener(self):
6ad14cab 1759 timeout_val = self.params.get('socket_timeout')
19a41fc6 1760 self._socket_timeout = 600 if timeout_val is None else float(timeout_val)
6ad14cab 1761
dca08720
PH
1762 opts_cookiefile = self.params.get('cookiefile')
1763 opts_proxy = self.params.get('proxy')
1764
1765 if opts_cookiefile is None:
1766 self.cookiejar = compat_cookiejar.CookieJar()
1767 else:
1768 self.cookiejar = compat_cookiejar.MozillaCookieJar(
1769 opts_cookiefile)
1770 if os.access(opts_cookiefile, os.R_OK):
1771 self.cookiejar.load()
1772
1773 cookie_processor = compat_urllib_request.HTTPCookieProcessor(
1774 self.cookiejar)
1775 if opts_proxy is not None:
1776 if opts_proxy == '':
1777 proxies = {}
1778 else:
1779 proxies = {'http': opts_proxy, 'https': opts_proxy}
1780 else:
1781 proxies = compat_urllib_request.getproxies()
1782 # Set HTTPS proxy to HTTP one if given (https://github.com/rg3/youtube-dl/issues/805)
1783 if 'http' in proxies and 'https' not in proxies:
1784 proxies['https'] = proxies['http']
1785 proxy_handler = compat_urllib_request.ProxyHandler(proxies)
a0ddb8a2
PH
1786
1787 debuglevel = 1 if self.params.get('debug_printtraffic') else 0
be4a824d
PH
1788 https_handler = make_HTTPS_handler(self.params, debuglevel=debuglevel)
1789 ydlh = YoutubeDLHandler(self.params, debuglevel=debuglevel)
dca08720 1790 opener = compat_urllib_request.build_opener(
a0ddb8a2 1791 https_handler, proxy_handler, cookie_processor, ydlh)
dca08720
PH
1792 # Delete the default user-agent header, which would otherwise apply in
1793 # cases where our custom HTTP handler doesn't come into play
1794 # (See https://github.com/rg3/youtube-dl/issues/1309 for details)
1795 opener.addheaders = []
1796 self._opener = opener
62fec3b2
PH
1797
1798 def encode(self, s):
1799 if isinstance(s, bytes):
1800 return s # Already encoded
1801
1802 try:
1803 return s.encode(self.get_encoding())
1804 except UnicodeEncodeError as err:
1805 err.reason = err.reason + '. Check your system encoding configuration or use the --encoding option.'
1806 raise
1807
1808 def get_encoding(self):
1809 encoding = self.params.get('encoding')
1810 if encoding is None:
1811 encoding = preferredencoding()
1812 return encoding
ec82d85a
PH
1813
1814 def _write_thumbnails(self, info_dict, filename):
1815 if self.params.get('writethumbnail', False):
1816 thumbnails = info_dict.get('thumbnails')
1817 if thumbnails:
1818 thumbnails = [thumbnails[-1]]
1819 elif self.params.get('write_all_thumbnails', False):
1820 thumbnails = info_dict.get('thumbnails')
1821 else:
1822 return
1823
1824 if not thumbnails:
1825 # No thumbnails present, so return immediately
1826 return
1827
1828 for t in thumbnails:
1829 thumb_ext = determine_ext(t['url'], 'jpg')
1830 suffix = '_%s' % t['id'] if len(thumbnails) > 1 else ''
1831 thumb_display_id = '%s ' % t['id'] if len(thumbnails) > 1 else ''
1832 thumb_filename = os.path.splitext(filename)[0] + suffix + '.' + thumb_ext
1833
1834 if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(thumb_filename)):
1835 self.to_screen('[%s] %s: Thumbnail %sis already present' %
1836 (info_dict['extractor'], info_dict['id'], thumb_display_id))
1837 else:
1838 self.to_screen('[%s] %s: Downloading thumbnail %s...' %
1839 (info_dict['extractor'], info_dict['id'], thumb_display_id))
1840 try:
1841 uf = self.urlopen(t['url'])
1842 with open(thumb_filename, 'wb') as thumbf:
1843 shutil.copyfileobj(uf, thumbf)
1844 self.to_screen('[%s] %s: Writing thumbnail %sto: %s' %
1845 (info_dict['extractor'], info_dict['id'], thumb_display_id, thumb_filename))
1846 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
1847 self.report_warning('Unable to download thumbnail "%s": %s' %
1848 (t['url'], compat_str(err)))