]> Raphaƫl G. Git Repositories - youtubedl/blob - youtube_dl/YoutubeDL.py
f4324039c72ec656f67bf536ee7856aae464cc01
[youtubedl] / youtube_dl / YoutubeDL.py
1 #!/usr/bin/env python
2 # -*- coding: utf-8 -*-
3
4 from __future__ import absolute_import, unicode_literals
5
6 import collections
7 import contextlib
8 import datetime
9 import errno
10 import fileinput
11 import io
12 import itertools
13 import json
14 import locale
15 import operator
16 import os
17 import platform
18 import re
19 import shutil
20 import subprocess
21 import socket
22 import sys
23 import time
24 import tokenize
25 import traceback
26
27 if os.name == 'nt':
28 import ctypes
29
30 from .compat import (
31 compat_basestring,
32 compat_cookiejar,
33 compat_expanduser,
34 compat_get_terminal_size,
35 compat_http_client,
36 compat_kwargs,
37 compat_str,
38 compat_tokenize_tokenize,
39 compat_urllib_error,
40 compat_urllib_request,
41 compat_urllib_request_DataHandler,
42 )
43 from .utils import (
44 ContentTooShortError,
45 date_from_str,
46 DateRange,
47 DEFAULT_OUTTMPL,
48 determine_ext,
49 determine_protocol,
50 DownloadError,
51 encode_compat_str,
52 encodeFilename,
53 error_to_compat_str,
54 ExtractorError,
55 format_bytes,
56 formatSeconds,
57 locked_file,
58 make_HTTPS_handler,
59 MaxDownloadsReached,
60 PagedList,
61 parse_filesize,
62 PerRequestProxyHandler,
63 PostProcessingError,
64 platform_name,
65 preferredencoding,
66 render_table,
67 SameFileError,
68 sanitize_filename,
69 sanitize_path,
70 sanitized_Request,
71 std_headers,
72 subtitles_filename,
73 UnavailableVideoError,
74 url_basename,
75 version_tuple,
76 write_json_file,
77 write_string,
78 YoutubeDLCookieProcessor,
79 YoutubeDLHandler,
80 prepend_extension,
81 replace_extension,
82 args_to_str,
83 age_restricted,
84 )
85 from .cache import Cache
86 from .extractor import get_info_extractor, gen_extractors
87 from .downloader import get_suitable_downloader
88 from .downloader.rtmp import rtmpdump_version
89 from .postprocessor import (
90 FFmpegFixupM4aPP,
91 FFmpegFixupStretchedPP,
92 FFmpegMergerPP,
93 FFmpegPostProcessor,
94 get_postprocessor,
95 )
96 from .version import __version__
97
98
99 class 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.
128 videopassword: Password for accessing a video.
129 usenetrc: Use netrc for authentication instead.
130 verbose: Print additional info to stdout.
131 quiet: Do not print messages to stdout.
132 no_warnings: Do not print out anything for warnings.
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.
139 forceduration: Force printing duration.
140 forcejson: Force printing info_dict as JSON.
141 dump_single_json: Force printing the info_dict of the whole playlist
142 (or video) as a single JSON line.
143 simulate: Do not download the video files.
144 format: Video format code. See options.py for more information.
145 outtmpl: Template for output names.
146 restrictfilenames: Do not allow "&" and spaces in file names
147 ignoreerrors: Do not stop on download errors.
148 force_generic_extractor: Force downloader to use the generic extractor
149 nooverwrites: Prevent overwriting files.
150 playliststart: Playlist item to start at.
151 playlistend: Playlist item to end at.
152 playlist_items: Specific indices of playlist to download.
153 playlistreverse: Download playlist items in reverse order.
154 matchtitle: Download only matching titles.
155 rejecttitle: Reject downloads for matching titles.
156 logger: Log messages to a logging.Logger instance.
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
160 writeannotations: Write the video annotations to a .annotations.xml file
161 writethumbnail: Write the thumbnail image to a file
162 write_all_thumbnails: Write all thumbnail formats to files
163 writesubtitles: Write the video subtitles to a file
164 writeautomaticsub: Write the automatically generated subtitles to a file
165 allsubtitles: Downloads all the subtitles of the video
166 (requires writesubtitles or writeautomaticsub)
167 listsubtitles: Lists all available subtitles for the video
168 subtitlesformat: The format code for subtitles
169 subtitleslangs: List of languages of the subtitles to download
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
173 cachedir: Location of the cache files in the filesystem.
174 False to disable filesystem cache.
175 noplaylist: Download single video instead of a playlist if in doubt.
176 age_limit: An integer representing the user's age in years.
177 Unsuitable videos for the given age are skipped.
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.
188 Videos already present in the file are not downloaded
189 again.
190 cookiefile: File name where cookies should be read from and dumped to.
191 nocheckcertificate:Do not verify SSL certificates
192 prefer_insecure: Use HTTP instead of HTTPS to retrieve information.
193 At the moment, this is only supported by YouTube.
194 proxy: URL of the proxy server to use
195 cn_verification_proxy: URL of the proxy to use for IP address verification
196 on Chinese sites. (Experimental)
197 socket_timeout: Time to wait for unresponsive hosts, in seconds
198 bidi_workaround: Work around buggy terminals without bidirectional text
199 support, using fridibi
200 debug_printtraffic:Print out sent and received HTTP traffic
201 include_ads: Download ads as well
202 default_search: Prepend this string if an input url is not valid.
203 'auto' for elaborate guessing
204 encoding: Use this encoding instead of the system-specified.
205 extract_flat: Do not resolve URLs, return the immediate result.
206 Pass in 'in_playlist' to only show this behavior for
207 playlist items.
208 postprocessors: A list of dictionaries, each with an entry
209 * key: The name of the postprocessor. See
210 youtube_dl/postprocessor/__init__.py for a list.
211 as well as any further keyword arguments for the
212 postprocessor.
213 progress_hooks: A list of functions that get called on download
214 progress, with a dictionary with the entries
215 * status: One of "downloading", "error", or "finished".
216 Check this first and ignore unknown values.
217
218 If status is one of "downloading", or "finished", the
219 following properties may also be present:
220 * filename: The final filename (always present)
221 * tmpfilename: The filename we're currently writing to
222 * downloaded_bytes: Bytes on disk
223 * total_bytes: Size of the whole file, None if unknown
224 * total_bytes_estimate: Guess of the eventual file size,
225 None if unavailable.
226 * elapsed: The number of seconds since download started.
227 * eta: The estimated time in seconds, None if unknown
228 * speed: The download speed in bytes/second, None if
229 unknown
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)
234
235 Progress hooks are guaranteed to be called at least once
236 (with status "finished") if the download is successful.
237 merge_output_format: Extension to use when merging formats.
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
243 about it, warn otherwise (default)
244 source_address: (Experimental) Client-side IP address to bind to.
245 call_home: Boolean, true iff we are allowed to contact the
246 youtube-dl servers for debugging.
247 sleep_interval: Number of seconds to sleep before each download.
248 listformats: Print an overview of available video formats and exit.
249 list_thumbnails: Print a table of all thumbnails and exit.
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.
255 no_color: Do not emit color codes in output.
256
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.
261
262 The following parameters are not used by YoutubeDL itself, they are used by
263 the downloader (see youtube_dl/downloader/common.py):
264 nopart, updatetime, buffersize, ratelimit, min_filesize, max_filesize, test,
265 noresizebuffer, retries, continuedl, noprogress, consoletitle,
266 xattr_set_filesize, external_downloader_args, hls_use_mpegts.
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.
271 postprocessor_args: A list of additional command-line arguments for the
272 postprocessor.
273 """
274
275 params = None
276 _ies = []
277 _pps = []
278 _download_retcode = None
279 _num_downloads = None
280 _screen_file = None
281
282 def __init__(self, params=None, auto_init=True):
283 """Create a FileDownloader object with the given options."""
284 if params is None:
285 params = {}
286 self._ies = []
287 self._ies_instances = {}
288 self._pps = []
289 self._progress_hooks = []
290 self._download_retcode = 0
291 self._num_downloads = 0
292 self._screen_file = [sys.stdout, sys.stderr][params.get('logtostderr', False)]
293 self._err_file = sys.stderr
294 self.params = {
295 # Default parameters
296 'nocheckcertificate': False,
297 }
298 self.params.update(params)
299 self.cache = Cache(self)
300
301 if params.get('bidi_workaround', False):
302 try:
303 import pty
304 master, slave = pty.openpty()
305 width = compat_get_terminal_size().columns
306 if width is None:
307 width_args = []
308 else:
309 width_args = ['-w', str(width)]
310 sp_kwargs = dict(
311 stdin=subprocess.PIPE,
312 stdout=slave,
313 stderr=self._err_file)
314 try:
315 self._output_process = subprocess.Popen(
316 ['bidiv'] + width_args, **sp_kwargs
317 )
318 except OSError:
319 self._output_process = subprocess.Popen(
320 ['fribidi', '-c', 'UTF-8'] + width_args, **sp_kwargs)
321 self._output_channel = os.fdopen(master, 'rb')
322 except OSError as ose:
323 if ose.errno == 2:
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 $PATH.')
325 else:
326 raise
327
328 if (sys.version_info >= (3,) and sys.platform != 'win32' and
329 sys.getfilesystemencoding() in ['ascii', 'ANSI_X3.4-1968'] and
330 not params.get('restrictfilenames', False)):
331 # On Python 3, the Unicode filesystem API will throw errors (#1474)
332 self.report_warning(
333 'Assuming --restrict-filenames since file system encoding '
334 'cannot encode all characters. '
335 'Set the LC_ALL environment variable to fix this.')
336 self.params['restrictfilenames'] = True
337
338 if isinstance(params.get('outtmpl'), bytes):
339 self.report_warning(
340 'Parameter outtmpl is bytes, but should be a unicode string. '
341 'Put from __future__ import unicode_literals at the top of your code file or consider switching to Python 3.x.')
342
343 self._setup_opener()
344
345 if auto_init:
346 self.print_debug_header()
347 self.add_default_info_extractors()
348
349 for pp_def_raw in self.params.get('postprocessors', []):
350 pp_class = get_postprocessor(pp_def_raw['key'])
351 pp_def = dict(pp_def_raw)
352 del pp_def['key']
353 pp = pp_class(self, **compat_kwargs(pp_def))
354 self.add_post_processor(pp)
355
356 for ph in self.params.get('progress_hooks', []):
357 self.add_progress_hook(ph)
358
359 def warn_if_short_id(self, argv):
360 # short YouTube ID starting with dash?
361 idxs = [
362 i for i, a in enumerate(argv)
363 if re.match(r'^-[0-9A-Za-z_-]{10}$', 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
375 def add_info_extractor(self, ie):
376 """Add an InfoExtractor object to the end of the list."""
377 self._ies.append(ie)
378 self._ies_instances[ie.ie_key()] = ie
379 ie.set_downloader(self)
380
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
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
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
405 def add_progress_hook(self, ph):
406 """Add the progress hook (currently only for the file downloader)"""
407 self._progress_hooks.append(ph)
408
409 def _bidi_workaround(self, message):
410 if not hasattr(self, '_output_channel'):
411 return message
412
413 assert hasattr(self, '_output_process')
414 assert isinstance(message, compat_str)
415 line_count = message.count('\n') + 1
416 self._output_process.stdin.write((message + '\n').encode('utf-8'))
417 self._output_process.stdin.flush()
418 res = ''.join(self._output_channel.readline().decode('utf-8')
419 for _ in range(line_count))
420 return res[:-len('\n')]
421
422 def to_screen(self, message, skip_eol=False):
423 """Print message to stdout if not in quiet mode."""
424 return self.to_stdout(message, skip_eol, check_quiet=True)
425
426 def _write_string(self, s, out=None):
427 write_string(s, out=out, encoding=self.params.get('encoding'))
428
429 def to_stdout(self, message, skip_eol=False, check_quiet=False):
430 """Print message to stdout if not in quiet mode."""
431 if self.params.get('logger'):
432 self.params['logger'].debug(message)
433 elif not check_quiet or not self.params.get('quiet', False):
434 message = self._bidi_workaround(message)
435 terminator = ['\n', ''][skip_eol]
436 output = message + terminator
437
438 self._write_string(output, self._screen_file)
439
440 def to_stderr(self, message):
441 """Print message to stderr."""
442 assert isinstance(message, compat_str)
443 if self.params.get('logger'):
444 self.params['logger'].error(message)
445 else:
446 message = self._bidi_workaround(message)
447 output = message + '\n'
448 self._write_string(output, self._err_file)
449
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:
458 self._write_string('\033]0;%s\007' % message, self._screen_file)
459
460 def save_console_title(self):
461 if not self.params.get('consoletitle', False):
462 return
463 if 'TERM' in os.environ:
464 # Save the title on stack
465 self._write_string('\033[22;0t', self._screen_file)
466
467 def restore_console_title(self):
468 if not self.params.get('consoletitle', False):
469 return
470 if 'TERM' in os.environ:
471 # Restore the title from stack
472 self._write_string('\033[23;0t', self._screen_file)
473
474 def __enter__(self):
475 self.save_console_title()
476 return self
477
478 def __exit__(self, *args):
479 self.restore_console_title()
480
481 if self.params.get('cookiefile') is not None:
482 self.cookiejar.save()
483
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
498 tb = ''
499 if hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
500 tb += ''.join(traceback.format_exception(*sys.exc_info()[1].exc_info))
501 tb += encode_compat_str(traceback.format_exc())
502 else:
503 tb_data = traceback.format_list(traceback.extract_stack())
504 tb = ''.join(tb_data)
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 '''
519 if self.params.get('logger') is not None:
520 self.params['logger'].warning(message)
521 else:
522 if self.params.get('no_warnings'):
523 return
524 if not self.params.get('no_color') and self._err_file.isatty() and os.name != 'nt':
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)
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 '''
536 if not self.params.get('no_color') and self._err_file.isatty() and os.name != 'nt':
537 _msg_header = '\033[0;31mERROR:\033[0m'
538 else:
539 _msg_header = 'ERROR:'
540 error_message = '%s %s' % (_msg_header, message)
541 self.trouble(error_message, tb)
542
543 def report_file_already_downloaded(self, file_name):
544 """Report file has already been fully downloaded."""
545 try:
546 self.to_screen('[download] %s has already been downloaded' % file_name)
547 except UnicodeEncodeError:
548 self.to_screen('[download] The file has already been downloaded')
549
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
559 autonumber_templ = '%0' + str(autonumber_size) + 'd'
560 template_dict['autonumber'] = autonumber_templ % self._num_downloads
561 if template_dict.get('playlist_index') is not None:
562 template_dict['playlist_index'] = '%0*d' % (len(str(template_dict['n_entries'])), template_dict['playlist_index'])
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'):
567 template_dict['resolution'] = '%sp' % template_dict['height']
568 elif template_dict.get('width'):
569 template_dict['resolution'] = '?x%d' % template_dict['width']
570
571 sanitize = lambda k, v: sanitize_filename(
572 compat_str(v),
573 restricted=self.params.get('restrictfilenames'),
574 is_id=(k == 'id'))
575 template_dict = dict((k, sanitize(k, v))
576 for k, v in template_dict.items()
577 if v is not None)
578 template_dict = collections.defaultdict(lambda: 'NA', template_dict)
579
580 outtmpl = self.params.get('outtmpl', DEFAULT_OUTTMPL)
581 tmpl = compat_expanduser(outtmpl)
582 filename = tmpl % template_dict
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())
588 return sanitize_path(filename)
589 except ValueError as err:
590 self.report_error('Error in output template: ' + str(err) + ' (encoding: ' + repr(preferredencoding()) + ')')
591 return None
592
593 def _match_entry(self, info_dict, incomplete):
594 """ Returns None iff the file should be downloaded """
595
596 video_title = info_dict.get('title', info_dict.get('id', 'video'))
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):
603 return '"' + title + '" title did not match pattern "' + matchtitle + '"'
604 rejecttitle = self.params.get('rejecttitle', False)
605 if rejecttitle:
606 if re.search(rejecttitle, title, re.IGNORECASE):
607 return '"' + title + '" title matched reject pattern "' + rejecttitle + '"'
608 date = info_dict.get('upload_date')
609 if date is not None:
610 dateRange = self.params.get('daterange', DateRange())
611 if date not in dateRange:
612 return '%s upload date is not in range %s' % (date_from_str(date).isoformat(), dateRange)
613 view_count = info_dict.get('view_count')
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:
617 return 'Skipping %s, because it has not reached minimum view count (%d/%d)' % (video_title, view_count, min_views)
618 max_views = self.params.get('max_views')
619 if max_views is not None and view_count > max_views:
620 return 'Skipping %s, because it has exceeded the maximum view count (%d/%d)' % (video_title, view_count, max_views)
621 if age_restricted(info_dict.get('age_limit'), self.params.get('age_limit')):
622 return 'Skipping "%s" because it is age restricted' % video_title
623 if self.in_download_archive(info_dict):
624 return '%s has already been recorded in archive' % video_title
625
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
632
633 return None
634
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
641 def extract_info(self, url, download=True, ie_key=None, extra_info={},
642 process=True, force_generic_extractor=False):
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
647 '''
648
649 if not ie_key and force_generic_extractor:
650 ie_key = 'Generic'
651
652 if ie_key:
653 ies = [self.get_info_extractor(ie_key)]
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():
662 self.report_warning('The program functionality for this site has been marked as broken, '
663 'and will probably not work.')
664
665 try:
666 ie_result = ie.extract(url)
667 if ie_result is None: # Finished already (backwards compatibility; listformats and friends should be moved here)
668 break
669 if isinstance(ie_result, list):
670 # Backwards compatibility: old IE result format
671 ie_result = {
672 '_type': 'compat_list',
673 'entries': ie_result,
674 }
675 self.add_default_extra_info(ie_result, ie, url)
676 if process:
677 return self.process_ie_result(ie_result, download, extra_info)
678 else:
679 return ie_result
680 except ExtractorError as e: # An error we somewhat expected
681 self.report_error(compat_str(e), e.format_traceback())
682 break
683 except MaxDownloadsReached:
684 raise
685 except Exception as e:
686 if self.params.get('ignoreerrors', False):
687 self.report_error(error_to_compat_str(e), tb=encode_compat_str(traceback.format_exc()))
688 break
689 else:
690 raise
691 else:
692 self.report_error('no suitable InfoExtractor for URL %s' % url)
693
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
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 result_type = ie_result.get('_type', 'video')
711
712 if result_type in ('url', 'url_transparent'):
713 extract_flat = self.params.get('extract_flat', False)
714 if ((extract_flat == 'in_playlist' and 'playlist' in extra_info) or
715 extract_flat is True):
716 if self.params.get('forcejson', False):
717 self.to_stdout(json.dumps(ie_result))
718 return ie_result
719
720 if result_type == 'video':
721 self.add_extra_info(ie_result, extra_info)
722 return self.process_video_result(ie_result, download=download)
723 elif result_type == 'url':
724 # We have to add extra_info to the results because it may be
725 # contained in a playlist
726 return self.extract_info(ie_result['url'],
727 download,
728 ie_key=ie_result.get('ie_key'),
729 extra_info=extra_info)
730 elif result_type == 'url_transparent':
731 # Use the information from the embedding page
732 info = self.extract_info(
733 ie_result['url'], ie_key=ie_result.get('ie_key'),
734 extra_info=extra_info, download=False, process=False)
735
736 force_properties = dict(
737 (k, v) for k, v in ie_result.items() if v is not None)
738 for f in ('_type', 'url', 'ie_key'):
739 if f in force_properties:
740 del force_properties[f]
741 new_result = info.copy()
742 new_result.update(force_properties)
743
744 assert new_result.get('_type') != 'url_transparent'
745
746 return self.process_ie_result(
747 new_result, download=download, extra_info=extra_info)
748 elif result_type == 'playlist' or result_type == 'multi_video':
749 # We process each entry in the playlist
750 playlist = ie_result.get('title') or ie_result.get('id')
751 self.to_screen('[download] Downloading playlist: %s' % playlist)
752
753 playlist_results = []
754
755 playliststart = self.params.get('playliststart', 1) - 1
756 playlistend = self.params.get('playlistend')
757 # For backwards compatibility, interpret -1 as whole list
758 if playlistend == -1:
759 playlistend = None
760
761 playlistitems_str = self.params.get('playlist_items')
762 playlistitems = None
763 if playlistitems_str is not None:
764 def iter_playlistitems(format):
765 for string_segment in format.split(','):
766 if '-' in string_segment:
767 start, end = string_segment.split('-')
768 for item in range(int(start), int(end) + 1):
769 yield int(item)
770 else:
771 yield int(string_segment)
772 playlistitems = iter_playlistitems(playlistitems_str)
773
774 ie_entries = ie_result['entries']
775 if isinstance(ie_entries, list):
776 n_all_entries = len(ie_entries)
777 if playlistitems:
778 entries = [
779 ie_entries[i - 1] for i in playlistitems
780 if -n_all_entries <= i - 1 < n_all_entries]
781 else:
782 entries = ie_entries[playliststart:playlistend]
783 n_entries = len(entries)
784 self.to_screen(
785 '[%s] playlist %s: Collected %d video ids (downloading %d of them)' %
786 (ie_result['extractor'], playlist, n_all_entries, n_entries))
787 elif isinstance(ie_entries, PagedList):
788 if playlistitems:
789 entries = []
790 for item in playlistitems:
791 entries.extend(ie_entries.getslice(
792 item - 1, item
793 ))
794 else:
795 entries = ie_entries.getslice(
796 playliststart, playlistend)
797 n_entries = len(entries)
798 self.to_screen(
799 '[%s] playlist %s: Downloading %d videos' %
800 (ie_result['extractor'], playlist, n_entries))
801 else: # iterable
802 if playlistitems:
803 entry_list = list(ie_entries)
804 entries = [entry_list[i - 1] for i in playlistitems]
805 else:
806 entries = list(itertools.islice(
807 ie_entries, playliststart, playlistend))
808 n_entries = len(entries)
809 self.to_screen(
810 '[%s] playlist %s: Downloading %d videos' %
811 (ie_result['extractor'], playlist, n_entries))
812
813 if self.params.get('playlistreverse', False):
814 entries = entries[::-1]
815
816 for i, entry in enumerate(entries, 1):
817 self.to_screen('[download] Downloading video %s of %s' % (i, n_entries))
818 extra = {
819 'n_entries': n_entries,
820 'playlist': playlist,
821 'playlist_id': ie_result.get('id'),
822 'playlist_title': ie_result.get('title'),
823 'playlist_index': i + playliststart,
824 'extractor': ie_result['extractor'],
825 'webpage_url': ie_result['webpage_url'],
826 'webpage_url_basename': url_basename(ie_result['webpage_url']),
827 'extractor_key': ie_result['extractor_key'],
828 }
829
830 reason = self._match_entry(entry, incomplete=True)
831 if reason is not None:
832 self.to_screen('[download] ' + reason)
833 continue
834
835 entry_result = self.process_ie_result(entry,
836 download=download,
837 extra_info=extra)
838 playlist_results.append(entry_result)
839 ie_result['entries'] = playlist_results
840 self.to_screen('[download] Finished downloading playlist: %s' % playlist)
841 return ie_result
842 elif result_type == 'compat_list':
843 self.report_warning(
844 'Extractor %s returned a compat_list result. '
845 'It needs to be updated.' % ie_result.get('extractor'))
846
847 def _fixup(r):
848 self.add_extra_info(
849 r,
850 {
851 'extractor': ie_result['extractor'],
852 'webpage_url': ie_result['webpage_url'],
853 'webpage_url_basename': url_basename(ie_result['webpage_url']),
854 'extractor_key': ie_result['extractor_key'],
855 }
856 )
857 return r
858 ie_result['entries'] = [
859 self.process_ie_result(_fixup(r), download, extra_info)
860 for r in ie_result['entries']
861 ]
862 return ie_result
863 else:
864 raise Exception('Invalid result type: %s' % result_type)
865
866 def _build_format_filter(self, filter_spec):
867 " Returns a function to filter the formats according to the filter_spec "
868
869 OPERATORS = {
870 '<': operator.lt,
871 '<=': operator.le,
872 '>': operator.gt,
873 '>=': operator.ge,
874 '=': operator.eq,
875 '!=': operator.ne,
876 }
877 operator_rex = re.compile(r'''(?x)\s*
878 (?P<key>width|height|tbr|abr|vbr|asr|filesize|fps)
879 \s*(?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
880 (?P<value>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?)
881 $
882 ''' % '|'.join(map(re.escape, OPERATORS.keys())))
883 m = operator_rex.search(filter_spec)
884 if m:
885 try:
886 comparison_value = int(m.group('value'))
887 except ValueError:
888 comparison_value = parse_filesize(m.group('value'))
889 if comparison_value is None:
890 comparison_value = parse_filesize(m.group('value') + 'B')
891 if comparison_value is None:
892 raise ValueError(
893 'Invalid value %r in format specification %r' % (
894 m.group('value'), filter_spec))
895 op = OPERATORS[m.group('op')]
896
897 if not m:
898 STR_OPERATORS = {
899 '=': operator.eq,
900 '!=': operator.ne,
901 '^=': lambda attr, value: attr.startswith(value),
902 '$=': lambda attr, value: attr.endswith(value),
903 '*=': lambda attr, value: value in attr,
904 }
905 str_operator_rex = re.compile(r'''(?x)
906 \s*(?P<key>ext|acodec|vcodec|container|protocol)
907 \s*(?P<op>%s)(?P<none_inclusive>\s*\?)?
908 \s*(?P<value>[a-zA-Z0-9._-]+)
909 \s*$
910 ''' % '|'.join(map(re.escape, STR_OPERATORS.keys())))
911 m = str_operator_rex.search(filter_spec)
912 if m:
913 comparison_value = m.group('value')
914 op = STR_OPERATORS[m.group('op')]
915
916 if not m:
917 raise ValueError('Invalid filter specification %r' % filter_spec)
918
919 def _filter(f):
920 actual_value = f.get(m.group('key'))
921 if actual_value is None:
922 return m.group('none_inclusive')
923 return op(actual_value, comparison_value)
924 return _filter
925
926 def build_format_selector(self, format_spec):
927 def syntax_error(note, start):
928 message = (
929 'Invalid format specification: '
930 '{0}\n\t{1}\n\t{2}^'.format(note, format_spec, ' ' * start[1]))
931 return SyntaxError(message)
932
933 PICKFIRST = 'PICKFIRST'
934 MERGE = 'MERGE'
935 SINGLE = 'SINGLE'
936 GROUP = 'GROUP'
937 FormatSelector = collections.namedtuple('FormatSelector', ['type', 'selector', 'filters'])
938
939 def _parse_filter(tokens):
940 filter_parts = []
941 for type, string, start, _, _ in tokens:
942 if type == tokenize.OP and string == ']':
943 return ''.join(filter_parts)
944 else:
945 filter_parts.append(string)
946
947 def _remove_unused_ops(tokens):
948 # Remove operators that we don't use and join them with the surrounding strings
949 # for example: 'mp4' '-' 'baseline' '-' '16x9' is converted to 'mp4-baseline-16x9'
950 ALLOWED_OPS = ('/', '+', ',', '(', ')')
951 last_string, last_start, last_end, last_line = None, None, None, None
952 for type, string, start, end, line in tokens:
953 if type == tokenize.OP and string == '[':
954 if last_string:
955 yield tokenize.NAME, last_string, last_start, last_end, last_line
956 last_string = None
957 yield type, string, start, end, line
958 # everything inside brackets will be handled by _parse_filter
959 for type, string, start, end, line in tokens:
960 yield type, string, start, end, line
961 if type == tokenize.OP and string == ']':
962 break
963 elif type == tokenize.OP and string in ALLOWED_OPS:
964 if last_string:
965 yield tokenize.NAME, last_string, last_start, last_end, last_line
966 last_string = None
967 yield type, string, start, end, line
968 elif type in [tokenize.NAME, tokenize.NUMBER, tokenize.OP]:
969 if not last_string:
970 last_string = string
971 last_start = start
972 last_end = end
973 else:
974 last_string += string
975 if last_string:
976 yield tokenize.NAME, last_string, last_start, last_end, last_line
977
978 def _parse_format_selection(tokens, inside_merge=False, inside_choice=False, inside_group=False):
979 selectors = []
980 current_selector = None
981 for type, string, start, _, _ in tokens:
982 # ENCODING is only defined in python 3.x
983 if type == getattr(tokenize, 'ENCODING', None):
984 continue
985 elif type in [tokenize.NAME, tokenize.NUMBER]:
986 current_selector = FormatSelector(SINGLE, string, [])
987 elif type == tokenize.OP:
988 if string == ')':
989 if not inside_group:
990 # ')' will be handled by the parentheses group
991 tokens.restore_last_token()
992 break
993 elif inside_merge and string in ['/', ',']:
994 tokens.restore_last_token()
995 break
996 elif inside_choice and string == ',':
997 tokens.restore_last_token()
998 break
999 elif string == ',':
1000 if not current_selector:
1001 raise syntax_error('"," must follow a format selector', start)
1002 selectors.append(current_selector)
1003 current_selector = None
1004 elif string == '/':
1005 if not current_selector:
1006 raise syntax_error('"/" must follow a format selector', start)
1007 first_choice = current_selector
1008 second_choice = _parse_format_selection(tokens, inside_choice=True)
1009 current_selector = FormatSelector(PICKFIRST, (first_choice, second_choice), [])
1010 elif string == '[':
1011 if not current_selector:
1012 current_selector = FormatSelector(SINGLE, 'best', [])
1013 format_filter = _parse_filter(tokens)
1014 current_selector.filters.append(format_filter)
1015 elif string == '(':
1016 if current_selector:
1017 raise syntax_error('Unexpected "("', start)
1018 group = _parse_format_selection(tokens, inside_group=True)
1019 current_selector = FormatSelector(GROUP, group, [])
1020 elif string == '+':
1021 video_selector = current_selector
1022 audio_selector = _parse_format_selection(tokens, inside_merge=True)
1023 if not video_selector or not audio_selector:
1024 raise syntax_error('"+" must be between two format selectors', start)
1025 current_selector = FormatSelector(MERGE, (video_selector, audio_selector), [])
1026 else:
1027 raise syntax_error('Operator not recognized: "{0}"'.format(string), start)
1028 elif type == tokenize.ENDMARKER:
1029 break
1030 if current_selector:
1031 selectors.append(current_selector)
1032 return selectors
1033
1034 def _build_selector_function(selector):
1035 if isinstance(selector, list):
1036 fs = [_build_selector_function(s) for s in selector]
1037
1038 def selector_function(formats):
1039 for f in fs:
1040 for format in f(formats):
1041 yield format
1042 return selector_function
1043 elif selector.type == GROUP:
1044 selector_function = _build_selector_function(selector.selector)
1045 elif selector.type == PICKFIRST:
1046 fs = [_build_selector_function(s) for s in selector.selector]
1047
1048 def selector_function(formats):
1049 for f in fs:
1050 picked_formats = list(f(formats))
1051 if picked_formats:
1052 return picked_formats
1053 return []
1054 elif selector.type == SINGLE:
1055 format_spec = selector.selector
1056
1057 def selector_function(formats):
1058 formats = list(formats)
1059 if not formats:
1060 return
1061 if format_spec == 'all':
1062 for f in formats:
1063 yield f
1064 elif format_spec in ['best', 'worst', None]:
1065 format_idx = 0 if format_spec == 'worst' else -1
1066 audiovideo_formats = [
1067 f for f in formats
1068 if f.get('vcodec') != 'none' and f.get('acodec') != 'none']
1069 if audiovideo_formats:
1070 yield audiovideo_formats[format_idx]
1071 # for audio only (soundcloud) or video only (imgur) urls, select the best/worst audio format
1072 elif (all(f.get('acodec') != 'none' for f in formats) or
1073 all(f.get('vcodec') != 'none' for f in formats)):
1074 yield formats[format_idx]
1075 elif format_spec == 'bestaudio':
1076 audio_formats = [
1077 f for f in formats
1078 if f.get('vcodec') == 'none']
1079 if audio_formats:
1080 yield audio_formats[-1]
1081 elif format_spec == 'worstaudio':
1082 audio_formats = [
1083 f for f in formats
1084 if f.get('vcodec') == 'none']
1085 if audio_formats:
1086 yield audio_formats[0]
1087 elif format_spec == 'bestvideo':
1088 video_formats = [
1089 f for f in formats
1090 if f.get('acodec') == 'none']
1091 if video_formats:
1092 yield video_formats[-1]
1093 elif format_spec == 'worstvideo':
1094 video_formats = [
1095 f for f in formats
1096 if f.get('acodec') == 'none']
1097 if video_formats:
1098 yield video_formats[0]
1099 else:
1100 extensions = ['mp4', 'flv', 'webm', '3gp', 'm4a', 'mp3', 'ogg', 'aac', 'wav']
1101 if format_spec in extensions:
1102 filter_f = lambda f: f['ext'] == format_spec
1103 else:
1104 filter_f = lambda f: f['format_id'] == format_spec
1105 matches = list(filter(filter_f, formats))
1106 if matches:
1107 yield matches[-1]
1108 elif selector.type == MERGE:
1109 def _merge(formats_info):
1110 format_1, format_2 = [f['format_id'] for f in formats_info]
1111 # The first format must contain the video and the
1112 # second the audio
1113 if formats_info[0].get('vcodec') == 'none':
1114 self.report_error('The first format must '
1115 'contain the video, try using '
1116 '"-f %s+%s"' % (format_2, format_1))
1117 return
1118 # Formats must be opposite (video+audio)
1119 if formats_info[0].get('acodec') == 'none' and formats_info[1].get('acodec') == 'none':
1120 self.report_error(
1121 'Both formats %s and %s are video-only, you must specify "-f video+audio"'
1122 % (format_1, format_2))
1123 return
1124 output_ext = (
1125 formats_info[0]['ext']
1126 if self.params.get('merge_output_format') is None
1127 else self.params['merge_output_format'])
1128 return {
1129 'requested_formats': formats_info,
1130 'format': '%s+%s' % (formats_info[0].get('format'),
1131 formats_info[1].get('format')),
1132 'format_id': '%s+%s' % (formats_info[0].get('format_id'),
1133 formats_info[1].get('format_id')),
1134 'width': formats_info[0].get('width'),
1135 'height': formats_info[0].get('height'),
1136 'resolution': formats_info[0].get('resolution'),
1137 'fps': formats_info[0].get('fps'),
1138 'vcodec': formats_info[0].get('vcodec'),
1139 'vbr': formats_info[0].get('vbr'),
1140 'stretched_ratio': formats_info[0].get('stretched_ratio'),
1141 'acodec': formats_info[1].get('acodec'),
1142 'abr': formats_info[1].get('abr'),
1143 'ext': output_ext,
1144 }
1145 video_selector, audio_selector = map(_build_selector_function, selector.selector)
1146
1147 def selector_function(formats):
1148 formats = list(formats)
1149 for pair in itertools.product(video_selector(formats), audio_selector(formats)):
1150 yield _merge(pair)
1151
1152 filters = [self._build_format_filter(f) for f in selector.filters]
1153
1154 def final_selector(formats):
1155 for _filter in filters:
1156 formats = list(filter(_filter, formats))
1157 return selector_function(formats)
1158 return final_selector
1159
1160 stream = io.BytesIO(format_spec.encode('utf-8'))
1161 try:
1162 tokens = list(_remove_unused_ops(compat_tokenize_tokenize(stream.readline)))
1163 except tokenize.TokenError:
1164 raise syntax_error('Missing closing/opening brackets or parenthesis', (0, len(format_spec)))
1165
1166 class TokenIterator(object):
1167 def __init__(self, tokens):
1168 self.tokens = tokens
1169 self.counter = 0
1170
1171 def __iter__(self):
1172 return self
1173
1174 def __next__(self):
1175 if self.counter >= len(self.tokens):
1176 raise StopIteration()
1177 value = self.tokens[self.counter]
1178 self.counter += 1
1179 return value
1180
1181 next = __next__
1182
1183 def restore_last_token(self):
1184 self.counter -= 1
1185
1186 parsed_selector = _parse_format_selection(iter(TokenIterator(tokens)))
1187 return _build_selector_function(parsed_selector)
1188
1189 def _calc_headers(self, info_dict):
1190 res = std_headers.copy()
1191
1192 add_headers = info_dict.get('http_headers')
1193 if add_headers:
1194 res.update(add_headers)
1195
1196 cookies = self._calc_cookies(info_dict)
1197 if cookies:
1198 res['Cookie'] = cookies
1199
1200 return res
1201
1202 def _calc_cookies(self, info_dict):
1203 pr = sanitized_Request(info_dict['url'])
1204 self.cookiejar.add_cookie_header(pr)
1205 return pr.get_header('Cookie')
1206
1207 def process_video_result(self, info_dict, download=True):
1208 assert info_dict.get('_type', 'video') == 'video'
1209
1210 if 'id' not in info_dict:
1211 raise ExtractorError('Missing "id" field in extractor result')
1212 if 'title' not in info_dict:
1213 raise ExtractorError('Missing "title" field in extractor result')
1214
1215 if 'playlist' not in info_dict:
1216 # It isn't part of a playlist
1217 info_dict['playlist'] = None
1218 info_dict['playlist_index'] = None
1219
1220 thumbnails = info_dict.get('thumbnails')
1221 if thumbnails is None:
1222 thumbnail = info_dict.get('thumbnail')
1223 if thumbnail:
1224 info_dict['thumbnails'] = thumbnails = [{'url': thumbnail}]
1225 if thumbnails:
1226 thumbnails.sort(key=lambda t: (
1227 t.get('preference'), t.get('width'), t.get('height'),
1228 t.get('id'), t.get('url')))
1229 for i, t in enumerate(thumbnails):
1230 if t.get('width') and t.get('height'):
1231 t['resolution'] = '%dx%d' % (t['width'], t['height'])
1232 if t.get('id') is None:
1233 t['id'] = '%d' % i
1234
1235 if thumbnails and 'thumbnail' not in info_dict:
1236 info_dict['thumbnail'] = thumbnails[-1]['url']
1237
1238 if 'display_id' not in info_dict and 'id' in info_dict:
1239 info_dict['display_id'] = info_dict['id']
1240
1241 if info_dict.get('upload_date') is None and info_dict.get('timestamp') is not None:
1242 # Working around out-of-range timestamp values (e.g. negative ones on Windows,
1243 # see http://bugs.python.org/issue1646728)
1244 try:
1245 upload_date = datetime.datetime.utcfromtimestamp(info_dict['timestamp'])
1246 info_dict['upload_date'] = upload_date.strftime('%Y%m%d')
1247 except (ValueError, OverflowError, OSError):
1248 pass
1249
1250 # Auto generate title fields corresponding to the *_number fields when missing
1251 # in order to always have clean titles. This is very common for TV series.
1252 for field in ('chapter', 'season', 'episode'):
1253 if info_dict.get('%s_number' % field) is not None and not info_dict.get(field):
1254 info_dict[field] = '%s %d' % (field.capitalize(), info_dict['%s_number' % field])
1255
1256 subtitles = info_dict.get('subtitles')
1257 if subtitles:
1258 for _, subtitle in subtitles.items():
1259 for subtitle_format in subtitle:
1260 if 'ext' not in subtitle_format:
1261 subtitle_format['ext'] = determine_ext(subtitle_format['url']).lower()
1262
1263 if self.params.get('listsubtitles', False):
1264 if 'automatic_captions' in info_dict:
1265 self.list_subtitles(info_dict['id'], info_dict.get('automatic_captions'), 'automatic captions')
1266 self.list_subtitles(info_dict['id'], subtitles, 'subtitles')
1267 return
1268 info_dict['requested_subtitles'] = self.process_subtitles(
1269 info_dict['id'], subtitles,
1270 info_dict.get('automatic_captions'))
1271
1272 # We now pick which formats have to be downloaded
1273 if info_dict.get('formats') is None:
1274 # There's only one format available
1275 formats = [info_dict]
1276 else:
1277 formats = info_dict['formats']
1278
1279 if not formats:
1280 raise ExtractorError('No video formats found!')
1281
1282 formats_dict = {}
1283
1284 # We check that all the formats have the format and format_id fields
1285 for i, format in enumerate(formats):
1286 if 'url' not in format:
1287 raise ExtractorError('Missing "url" key in result (index %d)' % i)
1288
1289 if format.get('format_id') is None:
1290 format['format_id'] = compat_str(i)
1291 else:
1292 # Sanitize format_id from characters used in format selector expression
1293 format['format_id'] = re.sub('[\s,/+\[\]()]', '_', format['format_id'])
1294 format_id = format['format_id']
1295 if format_id not in formats_dict:
1296 formats_dict[format_id] = []
1297 formats_dict[format_id].append(format)
1298
1299 # Make sure all formats have unique format_id
1300 for format_id, ambiguous_formats in formats_dict.items():
1301 if len(ambiguous_formats) > 1:
1302 for i, format in enumerate(ambiguous_formats):
1303 format['format_id'] = '%s-%d' % (format_id, i)
1304
1305 for i, format in enumerate(formats):
1306 if format.get('format') is None:
1307 format['format'] = '{id} - {res}{note}'.format(
1308 id=format['format_id'],
1309 res=self.format_resolution(format),
1310 note=' ({0})'.format(format['format_note']) if format.get('format_note') is not None else '',
1311 )
1312 # Automatically determine file extension if missing
1313 if 'ext' not in format:
1314 format['ext'] = determine_ext(format['url']).lower()
1315 # Automatically determine protocol if missing (useful for format
1316 # selection purposes)
1317 if 'protocol' not in format:
1318 format['protocol'] = determine_protocol(format)
1319 # Add HTTP headers, so that external programs can use them from the
1320 # json output
1321 full_format_info = info_dict.copy()
1322 full_format_info.update(format)
1323 format['http_headers'] = self._calc_headers(full_format_info)
1324
1325 # TODO Central sorting goes here
1326
1327 if formats[0] is not info_dict:
1328 # only set the 'formats' fields if the original info_dict list them
1329 # otherwise we end up with a circular reference, the first (and unique)
1330 # element in the 'formats' field in info_dict is info_dict itself,
1331 # which can't be exported to json
1332 info_dict['formats'] = formats
1333 if self.params.get('listformats'):
1334 self.list_formats(info_dict)
1335 return
1336 if self.params.get('list_thumbnails'):
1337 self.list_thumbnails(info_dict)
1338 return
1339
1340 req_format = self.params.get('format')
1341 if req_format is None:
1342 req_format_list = []
1343 if (self.params.get('outtmpl', DEFAULT_OUTTMPL) != '-' and
1344 not info_dict.get('is_live')):
1345 merger = FFmpegMergerPP(self)
1346 if merger.available and merger.can_merge():
1347 req_format_list.append('bestvideo+bestaudio')
1348 req_format_list.append('best')
1349 req_format = '/'.join(req_format_list)
1350 format_selector = self.build_format_selector(req_format)
1351 formats_to_download = list(format_selector(formats))
1352 if not formats_to_download:
1353 raise ExtractorError('requested format not available',
1354 expected=True)
1355
1356 if download:
1357 if len(formats_to_download) > 1:
1358 self.to_screen('[info] %s: downloading video in %s formats' % (info_dict['id'], len(formats_to_download)))
1359 for format in formats_to_download:
1360 new_info = dict(info_dict)
1361 new_info.update(format)
1362 self.process_info(new_info)
1363 # We update the info dict with the best quality format (backwards compatibility)
1364 info_dict.update(formats_to_download[-1])
1365 return info_dict
1366
1367 def process_subtitles(self, video_id, normal_subtitles, automatic_captions):
1368 """Select the requested subtitles and their format"""
1369 available_subs = {}
1370 if normal_subtitles and self.params.get('writesubtitles'):
1371 available_subs.update(normal_subtitles)
1372 if automatic_captions and self.params.get('writeautomaticsub'):
1373 for lang, cap_info in automatic_captions.items():
1374 if lang not in available_subs:
1375 available_subs[lang] = cap_info
1376
1377 if (not self.params.get('writesubtitles') and not
1378 self.params.get('writeautomaticsub') or not
1379 available_subs):
1380 return None
1381
1382 if self.params.get('allsubtitles', False):
1383 requested_langs = available_subs.keys()
1384 else:
1385 if self.params.get('subtitleslangs', False):
1386 requested_langs = self.params.get('subtitleslangs')
1387 elif 'en' in available_subs:
1388 requested_langs = ['en']
1389 else:
1390 requested_langs = [list(available_subs.keys())[0]]
1391
1392 formats_query = self.params.get('subtitlesformat', 'best')
1393 formats_preference = formats_query.split('/') if formats_query else []
1394 subs = {}
1395 for lang in requested_langs:
1396 formats = available_subs.get(lang)
1397 if formats is None:
1398 self.report_warning('%s subtitles not available for %s' % (lang, video_id))
1399 continue
1400 for ext in formats_preference:
1401 if ext == 'best':
1402 f = formats[-1]
1403 break
1404 matches = list(filter(lambda f: f['ext'] == ext, formats))
1405 if matches:
1406 f = matches[-1]
1407 break
1408 else:
1409 f = formats[-1]
1410 self.report_warning(
1411 'No subtitle format found matching "%s" for language %s, '
1412 'using %s' % (formats_query, lang, f['ext']))
1413 subs[lang] = f
1414 return subs
1415
1416 def process_info(self, info_dict):
1417 """Process a single resolved IE result."""
1418
1419 assert info_dict.get('_type', 'video') == 'video'
1420
1421 max_downloads = self.params.get('max_downloads')
1422 if max_downloads is not None:
1423 if self._num_downloads >= int(max_downloads):
1424 raise MaxDownloadsReached()
1425
1426 info_dict['fulltitle'] = info_dict['title']
1427 if len(info_dict['title']) > 200:
1428 info_dict['title'] = info_dict['title'][:197] + '...'
1429
1430 if 'format' not in info_dict:
1431 info_dict['format'] = info_dict['ext']
1432
1433 reason = self._match_entry(info_dict, incomplete=False)
1434 if reason is not None:
1435 self.to_screen('[download] ' + reason)
1436 return
1437
1438 self._num_downloads += 1
1439
1440 info_dict['_filename'] = filename = self.prepare_filename(info_dict)
1441
1442 # Forced printings
1443 if self.params.get('forcetitle', False):
1444 self.to_stdout(info_dict['fulltitle'])
1445 if self.params.get('forceid', False):
1446 self.to_stdout(info_dict['id'])
1447 if self.params.get('forceurl', False):
1448 if info_dict.get('requested_formats') is not None:
1449 for f in info_dict['requested_formats']:
1450 self.to_stdout(f['url'] + f.get('play_path', ''))
1451 else:
1452 # For RTMP URLs, also include the playpath
1453 self.to_stdout(info_dict['url'] + info_dict.get('play_path', ''))
1454 if self.params.get('forcethumbnail', False) and info_dict.get('thumbnail') is not None:
1455 self.to_stdout(info_dict['thumbnail'])
1456 if self.params.get('forcedescription', False) and info_dict.get('description') is not None:
1457 self.to_stdout(info_dict['description'])
1458 if self.params.get('forcefilename', False) and filename is not None:
1459 self.to_stdout(filename)
1460 if self.params.get('forceduration', False) and info_dict.get('duration') is not None:
1461 self.to_stdout(formatSeconds(info_dict['duration']))
1462 if self.params.get('forceformat', False):
1463 self.to_stdout(info_dict['format'])
1464 if self.params.get('forcejson', False):
1465 self.to_stdout(json.dumps(info_dict))
1466
1467 # Do nothing else if in simulate mode
1468 if self.params.get('simulate', False):
1469 return
1470
1471 if filename is None:
1472 return
1473
1474 try:
1475 dn = os.path.dirname(sanitize_path(encodeFilename(filename)))
1476 if dn and not os.path.exists(dn):
1477 os.makedirs(dn)
1478 except (OSError, IOError) as err:
1479 self.report_error('unable to create directory ' + error_to_compat_str(err))
1480 return
1481
1482 if self.params.get('writedescription', False):
1483 descfn = replace_extension(filename, 'description', info_dict.get('ext'))
1484 if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(descfn)):
1485 self.to_screen('[info] Video description is already present')
1486 elif info_dict.get('description') is None:
1487 self.report_warning('There\'s no description to write.')
1488 else:
1489 try:
1490 self.to_screen('[info] Writing video description to: ' + descfn)
1491 with io.open(encodeFilename(descfn), 'w', encoding='utf-8') as descfile:
1492 descfile.write(info_dict['description'])
1493 except (OSError, IOError):
1494 self.report_error('Cannot write description file ' + descfn)
1495 return
1496
1497 if self.params.get('writeannotations', False):
1498 annofn = replace_extension(filename, 'annotations.xml', info_dict.get('ext'))
1499 if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(annofn)):
1500 self.to_screen('[info] Video annotations are already present')
1501 else:
1502 try:
1503 self.to_screen('[info] Writing video annotations to: ' + annofn)
1504 with io.open(encodeFilename(annofn), 'w', encoding='utf-8') as annofile:
1505 annofile.write(info_dict['annotations'])
1506 except (KeyError, TypeError):
1507 self.report_warning('There are no annotations to write.')
1508 except (OSError, IOError):
1509 self.report_error('Cannot write annotations file: ' + annofn)
1510 return
1511
1512 subtitles_are_requested = any([self.params.get('writesubtitles', False),
1513 self.params.get('writeautomaticsub')])
1514
1515 if subtitles_are_requested and info_dict.get('requested_subtitles'):
1516 # subtitles download errors are already managed as troubles in relevant IE
1517 # that way it will silently go on when used with unsupporting IE
1518 subtitles = info_dict['requested_subtitles']
1519 ie = self.get_info_extractor(info_dict['extractor_key'])
1520 for sub_lang, sub_info in subtitles.items():
1521 sub_format = sub_info['ext']
1522 if sub_info.get('data') is not None:
1523 sub_data = sub_info['data']
1524 else:
1525 try:
1526 sub_data = ie._download_webpage(
1527 sub_info['url'], info_dict['id'], note=False)
1528 except ExtractorError as err:
1529 self.report_warning('Unable to download subtitle for "%s": %s' %
1530 (sub_lang, error_to_compat_str(err.cause)))
1531 continue
1532 try:
1533 sub_filename = subtitles_filename(filename, sub_lang, sub_format)
1534 if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(sub_filename)):
1535 self.to_screen('[info] Video subtitle %s.%s is already_present' % (sub_lang, sub_format))
1536 else:
1537 self.to_screen('[info] Writing video subtitles to: ' + sub_filename)
1538 with io.open(encodeFilename(sub_filename), 'w', encoding='utf-8') as subfile:
1539 subfile.write(sub_data)
1540 except (OSError, IOError):
1541 self.report_error('Cannot write subtitles file ' + sub_filename)
1542 return
1543
1544 if self.params.get('writeinfojson', False):
1545 infofn = replace_extension(filename, 'info.json', info_dict.get('ext'))
1546 if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(infofn)):
1547 self.to_screen('[info] Video description metadata is already present')
1548 else:
1549 self.to_screen('[info] Writing video description metadata as JSON to: ' + infofn)
1550 try:
1551 write_json_file(self.filter_requested_info(info_dict), infofn)
1552 except (OSError, IOError):
1553 self.report_error('Cannot write metadata to JSON file ' + infofn)
1554 return
1555
1556 self._write_thumbnails(info_dict, filename)
1557
1558 if not self.params.get('skip_download', False):
1559 try:
1560 def dl(name, info):
1561 fd = get_suitable_downloader(info, self.params)(self, self.params)
1562 for ph in self._progress_hooks:
1563 fd.add_progress_hook(ph)
1564 if self.params.get('verbose'):
1565 self.to_stdout('[debug] Invoking downloader on %r' % info.get('url'))
1566 return fd.download(name, info)
1567
1568 if info_dict.get('requested_formats') is not None:
1569 downloaded = []
1570 success = True
1571 merger = FFmpegMergerPP(self)
1572 if not merger.available:
1573 postprocessors = []
1574 self.report_warning('You have requested multiple '
1575 'formats but ffmpeg or avconv are not installed.'
1576 ' The formats won\'t be merged.')
1577 else:
1578 postprocessors = [merger]
1579
1580 def compatible_formats(formats):
1581 video, audio = formats
1582 # Check extension
1583 video_ext, audio_ext = audio.get('ext'), video.get('ext')
1584 if video_ext and audio_ext:
1585 COMPATIBLE_EXTS = (
1586 ('mp3', 'mp4', 'm4a', 'm4p', 'm4b', 'm4r', 'm4v'),
1587 ('webm')
1588 )
1589 for exts in COMPATIBLE_EXTS:
1590 if video_ext in exts and audio_ext in exts:
1591 return True
1592 # TODO: Check acodec/vcodec
1593 return False
1594
1595 filename_real_ext = os.path.splitext(filename)[1][1:]
1596 filename_wo_ext = (
1597 os.path.splitext(filename)[0]
1598 if filename_real_ext == info_dict['ext']
1599 else filename)
1600 requested_formats = info_dict['requested_formats']
1601 if self.params.get('merge_output_format') is None and not compatible_formats(requested_formats):
1602 info_dict['ext'] = 'mkv'
1603 self.report_warning(
1604 'Requested formats are incompatible for merge and will be merged into mkv.')
1605 # Ensure filename always has a correct extension for successful merge
1606 filename = '%s.%s' % (filename_wo_ext, info_dict['ext'])
1607 if os.path.exists(encodeFilename(filename)):
1608 self.to_screen(
1609 '[download] %s has already been downloaded and '
1610 'merged' % filename)
1611 else:
1612 for f in requested_formats:
1613 new_info = dict(info_dict)
1614 new_info.update(f)
1615 fname = self.prepare_filename(new_info)
1616 fname = prepend_extension(fname, 'f%s' % f['format_id'], new_info['ext'])
1617 downloaded.append(fname)
1618 partial_success = dl(fname, new_info)
1619 success = success and partial_success
1620 info_dict['__postprocessors'] = postprocessors
1621 info_dict['__files_to_merge'] = downloaded
1622 else:
1623 # Just a single file
1624 success = dl(filename, info_dict)
1625 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
1626 self.report_error('unable to download video data: %s' % str(err))
1627 return
1628 except (OSError, IOError) as err:
1629 raise UnavailableVideoError(err)
1630 except (ContentTooShortError, ) as err:
1631 self.report_error('content too short (expected %s bytes and served %s)' % (err.expected, err.downloaded))
1632 return
1633
1634 if success:
1635 # Fixup content
1636 fixup_policy = self.params.get('fixup')
1637 if fixup_policy is None:
1638 fixup_policy = 'detect_or_warn'
1639
1640 stretched_ratio = info_dict.get('stretched_ratio')
1641 if stretched_ratio is not None and stretched_ratio != 1:
1642 if fixup_policy == 'warn':
1643 self.report_warning('%s: Non-uniform pixel ratio (%s)' % (
1644 info_dict['id'], stretched_ratio))
1645 elif fixup_policy == 'detect_or_warn':
1646 stretched_pp = FFmpegFixupStretchedPP(self)
1647 if stretched_pp.available:
1648 info_dict.setdefault('__postprocessors', [])
1649 info_dict['__postprocessors'].append(stretched_pp)
1650 else:
1651 self.report_warning(
1652 '%s: Non-uniform pixel ratio (%s). Install ffmpeg or avconv to fix this automatically.' % (
1653 info_dict['id'], stretched_ratio))
1654 else:
1655 assert fixup_policy in ('ignore', 'never')
1656
1657 if info_dict.get('requested_formats') is None and info_dict.get('container') == 'm4a_dash':
1658 if fixup_policy == 'warn':
1659 self.report_warning('%s: writing DASH m4a. Only some players support this container.' % (
1660 info_dict['id']))
1661 elif fixup_policy == 'detect_or_warn':
1662 fixup_pp = FFmpegFixupM4aPP(self)
1663 if fixup_pp.available:
1664 info_dict.setdefault('__postprocessors', [])
1665 info_dict['__postprocessors'].append(fixup_pp)
1666 else:
1667 self.report_warning(
1668 '%s: writing DASH m4a. Only some players support this container. Install ffmpeg or avconv to fix this automatically.' % (
1669 info_dict['id']))
1670 else:
1671 assert fixup_policy in ('ignore', 'never')
1672
1673 try:
1674 self.post_process(filename, info_dict)
1675 except (PostProcessingError) as err:
1676 self.report_error('postprocessing: %s' % str(err))
1677 return
1678 self.record_download_archive(info_dict)
1679
1680 def download(self, url_list):
1681 """Download a given list of URLs."""
1682 outtmpl = self.params.get('outtmpl', DEFAULT_OUTTMPL)
1683 if (len(url_list) > 1 and
1684 '%' not in outtmpl and
1685 self.params.get('max_downloads') != 1):
1686 raise SameFileError(outtmpl)
1687
1688 for url in url_list:
1689 try:
1690 # It also downloads the videos
1691 res = self.extract_info(
1692 url, force_generic_extractor=self.params.get('force_generic_extractor', False))
1693 except UnavailableVideoError:
1694 self.report_error('unable to download video')
1695 except MaxDownloadsReached:
1696 self.to_screen('[info] Maximum number of downloaded files reached.')
1697 raise
1698 else:
1699 if self.params.get('dump_single_json', False):
1700 self.to_stdout(json.dumps(res))
1701
1702 return self._download_retcode
1703
1704 def download_with_info_file(self, info_filename):
1705 with contextlib.closing(fileinput.FileInput(
1706 [info_filename], mode='r',
1707 openhook=fileinput.hook_encoded('utf-8'))) as f:
1708 # FileInput doesn't have a read method, we can't call json.load
1709 info = self.filter_requested_info(json.loads('\n'.join(f)))
1710 try:
1711 self.process_ie_result(info, download=True)
1712 except DownloadError:
1713 webpage_url = info.get('webpage_url')
1714 if webpage_url is not None:
1715 self.report_warning('The info failed to download, trying with "%s"' % webpage_url)
1716 return self.download([webpage_url])
1717 else:
1718 raise
1719 return self._download_retcode
1720
1721 @staticmethod
1722 def filter_requested_info(info_dict):
1723 return dict(
1724 (k, v) for k, v in info_dict.items()
1725 if k not in ['requested_formats', 'requested_subtitles'])
1726
1727 def post_process(self, filename, ie_info):
1728 """Run all the postprocessors on the given file."""
1729 info = dict(ie_info)
1730 info['filepath'] = filename
1731 pps_chain = []
1732 if ie_info.get('__postprocessors') is not None:
1733 pps_chain.extend(ie_info['__postprocessors'])
1734 pps_chain.extend(self._pps)
1735 for pp in pps_chain:
1736 files_to_delete = []
1737 try:
1738 files_to_delete, info = pp.run(info)
1739 except PostProcessingError as e:
1740 self.report_error(e.msg)
1741 if files_to_delete and not self.params.get('keepvideo', False):
1742 for old_filename in files_to_delete:
1743 self.to_screen('Deleting original file %s (pass -k to keep)' % old_filename)
1744 try:
1745 os.remove(encodeFilename(old_filename))
1746 except (IOError, OSError):
1747 self.report_warning('Unable to remove downloaded original file')
1748
1749 def _make_archive_id(self, info_dict):
1750 # Future-proof against any change in case
1751 # and backwards compatibility with prior versions
1752 extractor = info_dict.get('extractor_key')
1753 if extractor is None:
1754 if 'id' in info_dict:
1755 extractor = info_dict.get('ie_key') # key in a playlist
1756 if extractor is None:
1757 return None # Incomplete video information
1758 return extractor.lower() + ' ' + info_dict['id']
1759
1760 def in_download_archive(self, info_dict):
1761 fn = self.params.get('download_archive')
1762 if fn is None:
1763 return False
1764
1765 vid_id = self._make_archive_id(info_dict)
1766 if vid_id is None:
1767 return False # Incomplete video information
1768
1769 try:
1770 with locked_file(fn, 'r', encoding='utf-8') as archive_file:
1771 for line in archive_file:
1772 if line.strip() == vid_id:
1773 return True
1774 except IOError as ioe:
1775 if ioe.errno != errno.ENOENT:
1776 raise
1777 return False
1778
1779 def record_download_archive(self, info_dict):
1780 fn = self.params.get('download_archive')
1781 if fn is None:
1782 return
1783 vid_id = self._make_archive_id(info_dict)
1784 assert vid_id
1785 with locked_file(fn, 'a', encoding='utf-8') as archive_file:
1786 archive_file.write(vid_id + '\n')
1787
1788 @staticmethod
1789 def format_resolution(format, default='unknown'):
1790 if format.get('vcodec') == 'none':
1791 return 'audio only'
1792 if format.get('resolution') is not None:
1793 return format['resolution']
1794 if format.get('height') is not None:
1795 if format.get('width') is not None:
1796 res = '%sx%s' % (format['width'], format['height'])
1797 else:
1798 res = '%sp' % format['height']
1799 elif format.get('width') is not None:
1800 res = '%dx?' % format['width']
1801 else:
1802 res = default
1803 return res
1804
1805 def _format_note(self, fdict):
1806 res = ''
1807 if fdict.get('ext') in ['f4f', 'f4m']:
1808 res += '(unsupported) '
1809 if fdict.get('language'):
1810 if res:
1811 res += ' '
1812 res += '[%s]' % fdict['language']
1813 if fdict.get('format_note') is not None:
1814 res += fdict['format_note'] + ' '
1815 if fdict.get('tbr') is not None:
1816 res += '%4dk ' % fdict['tbr']
1817 if fdict.get('container') is not None:
1818 if res:
1819 res += ', '
1820 res += '%s container' % fdict['container']
1821 if (fdict.get('vcodec') is not None and
1822 fdict.get('vcodec') != 'none'):
1823 if res:
1824 res += ', '
1825 res += fdict['vcodec']
1826 if fdict.get('vbr') is not None:
1827 res += '@'
1828 elif fdict.get('vbr') is not None and fdict.get('abr') is not None:
1829 res += 'video@'
1830 if fdict.get('vbr') is not None:
1831 res += '%4dk' % fdict['vbr']
1832 if fdict.get('fps') is not None:
1833 res += ', %sfps' % fdict['fps']
1834 if fdict.get('acodec') is not None:
1835 if res:
1836 res += ', '
1837 if fdict['acodec'] == 'none':
1838 res += 'video only'
1839 else:
1840 res += '%-5s' % fdict['acodec']
1841 elif fdict.get('abr') is not None:
1842 if res:
1843 res += ', '
1844 res += 'audio'
1845 if fdict.get('abr') is not None:
1846 res += '@%3dk' % fdict['abr']
1847 if fdict.get('asr') is not None:
1848 res += ' (%5dHz)' % fdict['asr']
1849 if fdict.get('filesize') is not None:
1850 if res:
1851 res += ', '
1852 res += format_bytes(fdict['filesize'])
1853 elif fdict.get('filesize_approx') is not None:
1854 if res:
1855 res += ', '
1856 res += '~' + format_bytes(fdict['filesize_approx'])
1857 return res
1858
1859 def list_formats(self, info_dict):
1860 formats = info_dict.get('formats', [info_dict])
1861 table = [
1862 [f['format_id'], f['ext'], self.format_resolution(f), self._format_note(f)]
1863 for f in formats
1864 if f.get('preference') is None or f['preference'] >= -1000]
1865 if len(formats) > 1:
1866 table[-1][-1] += (' ' if table[-1][-1] else '') + '(best)'
1867
1868 header_line = ['format code', 'extension', 'resolution', 'note']
1869 self.to_screen(
1870 '[info] Available formats for %s:\n%s' %
1871 (info_dict['id'], render_table(header_line, table)))
1872
1873 def list_thumbnails(self, info_dict):
1874 thumbnails = info_dict.get('thumbnails')
1875 if not thumbnails:
1876 tn_url = info_dict.get('thumbnail')
1877 if tn_url:
1878 thumbnails = [{'id': '0', 'url': tn_url}]
1879 else:
1880 self.to_screen(
1881 '[info] No thumbnails present for %s' % info_dict['id'])
1882 return
1883
1884 self.to_screen(
1885 '[info] Thumbnails for %s:' % info_dict['id'])
1886 self.to_screen(render_table(
1887 ['ID', 'width', 'height', 'URL'],
1888 [[t['id'], t.get('width', 'unknown'), t.get('height', 'unknown'), t['url']] for t in thumbnails]))
1889
1890 def list_subtitles(self, video_id, subtitles, name='subtitles'):
1891 if not subtitles:
1892 self.to_screen('%s has no %s' % (video_id, name))
1893 return
1894 self.to_screen(
1895 'Available %s for %s:' % (name, video_id))
1896 self.to_screen(render_table(
1897 ['Language', 'formats'],
1898 [[lang, ', '.join(f['ext'] for f in reversed(formats))]
1899 for lang, formats in subtitles.items()]))
1900
1901 def urlopen(self, req):
1902 """ Start an HTTP download """
1903 if isinstance(req, compat_basestring):
1904 req = sanitized_Request(req)
1905 return self._opener.open(req, timeout=self._socket_timeout)
1906
1907 def print_debug_header(self):
1908 if not self.params.get('verbose'):
1909 return
1910
1911 if type('') is not compat_str:
1912 # Python 2.6 on SLES11 SP1 (https://github.com/rg3/youtube-dl/issues/3326)
1913 self.report_warning(
1914 'Your Python is broken! Update to a newer and supported version')
1915
1916 stdout_encoding = getattr(
1917 sys.stdout, 'encoding', 'missing (%s)' % type(sys.stdout).__name__)
1918 encoding_str = (
1919 '[debug] Encodings: locale %s, fs %s, out %s, pref %s\n' % (
1920 locale.getpreferredencoding(),
1921 sys.getfilesystemencoding(),
1922 stdout_encoding,
1923 self.get_encoding()))
1924 write_string(encoding_str, encoding=None)
1925
1926 self._write_string('[debug] youtube-dl version ' + __version__ + '\n')
1927 try:
1928 sp = subprocess.Popen(
1929 ['git', 'rev-parse', '--short', 'HEAD'],
1930 stdout=subprocess.PIPE, stderr=subprocess.PIPE,
1931 cwd=os.path.dirname(os.path.abspath(__file__)))
1932 out, err = sp.communicate()
1933 out = out.decode().strip()
1934 if re.match('[0-9a-f]+', out):
1935 self._write_string('[debug] Git HEAD: ' + out + '\n')
1936 except Exception:
1937 try:
1938 sys.exc_clear()
1939 except Exception:
1940 pass
1941 self._write_string('[debug] Python version %s - %s\n' % (
1942 platform.python_version(), platform_name()))
1943
1944 exe_versions = FFmpegPostProcessor.get_versions(self)
1945 exe_versions['rtmpdump'] = rtmpdump_version()
1946 exe_str = ', '.join(
1947 '%s %s' % (exe, v)
1948 for exe, v in sorted(exe_versions.items())
1949 if v
1950 )
1951 if not exe_str:
1952 exe_str = 'none'
1953 self._write_string('[debug] exe versions: %s\n' % exe_str)
1954
1955 proxy_map = {}
1956 for handler in self._opener.handlers:
1957 if hasattr(handler, 'proxies'):
1958 proxy_map.update(handler.proxies)
1959 self._write_string('[debug] Proxy map: ' + compat_str(proxy_map) + '\n')
1960
1961 if self.params.get('call_home', False):
1962 ipaddr = self.urlopen('https://yt-dl.org/ip').read().decode('utf-8')
1963 self._write_string('[debug] Public IP address: %s\n' % ipaddr)
1964 latest_version = self.urlopen(
1965 'https://yt-dl.org/latest/version').read().decode('utf-8')
1966 if version_tuple(latest_version) > version_tuple(__version__):
1967 self.report_warning(
1968 'You are using an outdated version (newest version: %s)! '
1969 'See https://yt-dl.org/update if you need help updating.' %
1970 latest_version)
1971
1972 def _setup_opener(self):
1973 timeout_val = self.params.get('socket_timeout')
1974 self._socket_timeout = 600 if timeout_val is None else float(timeout_val)
1975
1976 opts_cookiefile = self.params.get('cookiefile')
1977 opts_proxy = self.params.get('proxy')
1978
1979 if opts_cookiefile is None:
1980 self.cookiejar = compat_cookiejar.CookieJar()
1981 else:
1982 self.cookiejar = compat_cookiejar.MozillaCookieJar(
1983 opts_cookiefile)
1984 if os.access(opts_cookiefile, os.R_OK):
1985 self.cookiejar.load()
1986
1987 cookie_processor = YoutubeDLCookieProcessor(self.cookiejar)
1988 if opts_proxy is not None:
1989 if opts_proxy == '':
1990 proxies = {}
1991 else:
1992 proxies = {'http': opts_proxy, 'https': opts_proxy}
1993 else:
1994 proxies = compat_urllib_request.getproxies()
1995 # Set HTTPS proxy to HTTP one if given (https://github.com/rg3/youtube-dl/issues/805)
1996 if 'http' in proxies and 'https' not in proxies:
1997 proxies['https'] = proxies['http']
1998 proxy_handler = PerRequestProxyHandler(proxies)
1999
2000 debuglevel = 1 if self.params.get('debug_printtraffic') else 0
2001 https_handler = make_HTTPS_handler(self.params, debuglevel=debuglevel)
2002 ydlh = YoutubeDLHandler(self.params, debuglevel=debuglevel)
2003 data_handler = compat_urllib_request_DataHandler()
2004
2005 # When passing our own FileHandler instance, build_opener won't add the
2006 # default FileHandler and allows us to disable the file protocol, which
2007 # can be used for malicious purposes (see
2008 # https://github.com/rg3/youtube-dl/issues/8227)
2009 file_handler = compat_urllib_request.FileHandler()
2010
2011 def file_open(*args, **kwargs):
2012 raise compat_urllib_error.URLError('file:// scheme is explicitly disabled in youtube-dl for security reasons')
2013 file_handler.file_open = file_open
2014
2015 opener = compat_urllib_request.build_opener(
2016 proxy_handler, https_handler, cookie_processor, ydlh, data_handler, file_handler)
2017
2018 # Delete the default user-agent header, which would otherwise apply in
2019 # cases where our custom HTTP handler doesn't come into play
2020 # (See https://github.com/rg3/youtube-dl/issues/1309 for details)
2021 opener.addheaders = []
2022 self._opener = opener
2023
2024 def encode(self, s):
2025 if isinstance(s, bytes):
2026 return s # Already encoded
2027
2028 try:
2029 return s.encode(self.get_encoding())
2030 except UnicodeEncodeError as err:
2031 err.reason = err.reason + '. Check your system encoding configuration or use the --encoding option.'
2032 raise
2033
2034 def get_encoding(self):
2035 encoding = self.params.get('encoding')
2036 if encoding is None:
2037 encoding = preferredencoding()
2038 return encoding
2039
2040 def _write_thumbnails(self, info_dict, filename):
2041 if self.params.get('writethumbnail', False):
2042 thumbnails = info_dict.get('thumbnails')
2043 if thumbnails:
2044 thumbnails = [thumbnails[-1]]
2045 elif self.params.get('write_all_thumbnails', False):
2046 thumbnails = info_dict.get('thumbnails')
2047 else:
2048 return
2049
2050 if not thumbnails:
2051 # No thumbnails present, so return immediately
2052 return
2053
2054 for t in thumbnails:
2055 thumb_ext = determine_ext(t['url'], 'jpg')
2056 suffix = '_%s' % t['id'] if len(thumbnails) > 1 else ''
2057 thumb_display_id = '%s ' % t['id'] if len(thumbnails) > 1 else ''
2058 t['filename'] = thumb_filename = os.path.splitext(filename)[0] + suffix + '.' + thumb_ext
2059
2060 if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(thumb_filename)):
2061 self.to_screen('[%s] %s: Thumbnail %sis already present' %
2062 (info_dict['extractor'], info_dict['id'], thumb_display_id))
2063 else:
2064 self.to_screen('[%s] %s: Downloading thumbnail %s...' %
2065 (info_dict['extractor'], info_dict['id'], thumb_display_id))
2066 try:
2067 uf = self.urlopen(t['url'])
2068 with open(encodeFilename(thumb_filename), 'wb') as thumbf:
2069 shutil.copyfileobj(uf, thumbf)
2070 self.to_screen('[%s] %s: Writing thumbnail %sto: %s' %
2071 (info_dict['extractor'], info_dict['id'], thumb_display_id, thumb_filename))
2072 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
2073 self.report_warning('Unable to download thumbnail "%s": %s' %
2074 (t['url'], error_to_compat_str(err)))