You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.

1885 lines
83 KiB

11 years ago
11 years ago
11 years ago
11 years ago
11 years ago
11 years ago
11 years ago
11 years ago
11 years ago
11 years ago
11 years ago
11 years ago
11 years ago
11 years ago
11 years ago
10 years ago
10 years ago
11 years ago
10 years ago
10 years ago
10 years ago
10 years ago
10 years ago
  1. #!/usr/bin/env python
  2. # -*- coding: utf-8 -*-
  3. from __future__ import absolute_import, unicode_literals
  4. import collections
  5. import contextlib
  6. import datetime
  7. import errno
  8. import fileinput
  9. import io
  10. import itertools
  11. import json
  12. import locale
  13. import operator
  14. import os
  15. import platform
  16. import re
  17. import shutil
  18. import subprocess
  19. import socket
  20. import sys
  21. import time
  22. import traceback
  23. if os.name == 'nt':
  24. import ctypes
  25. from .compat import (
  26. compat_basestring,
  27. compat_cookiejar,
  28. compat_expanduser,
  29. compat_get_terminal_size,
  30. compat_http_client,
  31. compat_kwargs,
  32. compat_str,
  33. compat_urllib_error,
  34. compat_urllib_request,
  35. )
  36. from .utils import (
  37. escape_url,
  38. ContentTooShortError,
  39. date_from_str,
  40. DateRange,
  41. DEFAULT_OUTTMPL,
  42. determine_ext,
  43. DownloadError,
  44. encodeFilename,
  45. ExtractorError,
  46. format_bytes,
  47. formatSeconds,
  48. locked_file,
  49. make_HTTPS_handler,
  50. MaxDownloadsReached,
  51. PagedList,
  52. parse_filesize,
  53. PerRequestProxyHandler,
  54. PostProcessingError,
  55. platform_name,
  56. preferredencoding,
  57. render_table,
  58. SameFileError,
  59. sanitize_filename,
  60. sanitize_path,
  61. std_headers,
  62. subtitles_filename,
  63. UnavailableVideoError,
  64. url_basename,
  65. version_tuple,
  66. write_json_file,
  67. write_string,
  68. YoutubeDLHandler,
  69. prepend_extension,
  70. replace_extension,
  71. args_to_str,
  72. age_restricted,
  73. )
  74. from .cache import Cache
  75. from .extractor import get_info_extractor, gen_extractors
  76. from .downloader import get_suitable_downloader
  77. from .downloader.rtmp import rtmpdump_version
  78. from .postprocessor import (
  79. FFmpegFixupM4aPP,
  80. FFmpegFixupStretchedPP,
  81. FFmpegMergerPP,
  82. FFmpegPostProcessor,
  83. get_postprocessor,
  84. )
  85. from .version import __version__
  86. class YoutubeDL(object):
  87. """YoutubeDL class.
  88. YoutubeDL objects are the ones responsible of downloading the
  89. actual video file and writing it to disk if the user has requested
  90. it, among some other tasks. In most cases there should be one per
  91. program. As, given a video URL, the downloader doesn't know how to
  92. extract all the needed information, task that InfoExtractors do, it
  93. has to pass the URL to one of them.
  94. For this, YoutubeDL objects have a method that allows
  95. InfoExtractors to be registered in a given order. When it is passed
  96. a URL, the YoutubeDL object handles it to the first InfoExtractor it
  97. finds that reports being able to handle it. The InfoExtractor extracts
  98. all the information about the video or videos the URL refers to, and
  99. YoutubeDL process the extracted information, possibly using a File
  100. Downloader to download the video.
  101. YoutubeDL objects accept a lot of parameters. In order not to saturate
  102. the object constructor with arguments, it receives a dictionary of
  103. options instead. These options are available through the params
  104. attribute for the InfoExtractors to use. The YoutubeDL also
  105. registers itself as the downloader in charge for the InfoExtractors
  106. that are added to it, so this is a "mutual registration".
  107. Available options:
  108. username: Username for authentication purposes.
  109. password: Password for authentication purposes.
  110. videopassword: Password for acces a video.
  111. usenetrc: Use netrc for authentication instead.
  112. verbose: Print additional info to stdout.
  113. quiet: Do not print messages to stdout.
  114. no_warnings: Do not print out anything for warnings.
  115. forceurl: Force printing final URL.
  116. forcetitle: Force printing title.
  117. forceid: Force printing ID.
  118. forcethumbnail: Force printing thumbnail URL.
  119. forcedescription: Force printing description.
  120. forcefilename: Force printing final filename.
  121. forceduration: Force printing duration.
  122. forcejson: Force printing info_dict as JSON.
  123. dump_single_json: Force printing the info_dict of the whole playlist
  124. (or video) as a single JSON line.
  125. simulate: Do not download the video files.
  126. format: Video format code. See options.py for more information.
  127. outtmpl: Template for output names.
  128. restrictfilenames: Do not allow "&" and spaces in file names
  129. ignoreerrors: Do not stop on download errors.
  130. nooverwrites: Prevent overwriting files.
  131. playliststart: Playlist item to start at.
  132. playlistend: Playlist item to end at.
  133. playlist_items: Specific indices of playlist to download.
  134. playlistreverse: Download playlist items in reverse order.
  135. matchtitle: Download only matching titles.
  136. rejecttitle: Reject downloads for matching titles.
  137. logger: Log messages to a logging.Logger instance.
  138. logtostderr: Log messages to stderr instead of stdout.
  139. writedescription: Write the video description to a .description file
  140. writeinfojson: Write the video description to a .info.json file
  141. writeannotations: Write the video annotations to a .annotations.xml file
  142. writethumbnail: Write the thumbnail image to a file
  143. write_all_thumbnails: Write all thumbnail formats to files
  144. writesubtitles: Write the video subtitles to a file
  145. writeautomaticsub: Write the automatic subtitles to a file
  146. allsubtitles: Downloads all the subtitles of the video
  147. (requires writesubtitles or writeautomaticsub)
  148. listsubtitles: Lists all available subtitles for the video
  149. subtitlesformat: The format code for subtitles
  150. subtitleslangs: List of languages of the subtitles to download
  151. keepvideo: Keep the video file after post-processing
  152. daterange: A DateRange object, download only if the upload_date is in the range.
  153. skip_download: Skip the actual download of the video file
  154. cachedir: Location of the cache files in the filesystem.
  155. False to disable filesystem cache.
  156. noplaylist: Download single video instead of a playlist if in doubt.
  157. age_limit: An integer representing the user's age in years.
  158. Unsuitable videos for the given age are skipped.
  159. min_views: An integer representing the minimum view count the video
  160. must have in order to not be skipped.
  161. Videos without view count information are always
  162. downloaded. None for no limit.
  163. max_views: An integer representing the maximum view count.
  164. Videos that are more popular than that are not
  165. downloaded.
  166. Videos without view count information are always
  167. downloaded. None for no limit.
  168. download_archive: File name of a file where all downloads are recorded.
  169. Videos already present in the file are not downloaded
  170. again.
  171. cookiefile: File name where cookies should be read from and dumped to.
  172. nocheckcertificate:Do not verify SSL certificates
  173. prefer_insecure: Use HTTP instead of HTTPS to retrieve information.
  174. At the moment, this is only supported by YouTube.
  175. proxy: URL of the proxy server to use
  176. cn_verification_proxy: URL of the proxy to use for IP address verification
  177. on Chinese sites. (Experimental)
  178. socket_timeout: Time to wait for unresponsive hosts, in seconds
  179. bidi_workaround: Work around buggy terminals without bidirectional text
  180. support, using fridibi
  181. debug_printtraffic:Print out sent and received HTTP traffic
  182. include_ads: Download ads as well
  183. default_search: Prepend this string if an input url is not valid.
  184. 'auto' for elaborate guessing
  185. encoding: Use this encoding instead of the system-specified.
  186. extract_flat: Do not resolve URLs, return the immediate result.
  187. Pass in 'in_playlist' to only show this behavior for
  188. playlist items.
  189. postprocessors: A list of dictionaries, each with an entry
  190. * key: The name of the postprocessor. See
  191. youtube_dl/postprocessor/__init__.py for a list.
  192. as well as any further keyword arguments for the
  193. postprocessor.
  194. progress_hooks: A list of functions that get called on download
  195. progress, with a dictionary with the entries
  196. * status: One of "downloading", "error", or "finished".
  197. Check this first and ignore unknown values.
  198. If status is one of "downloading", or "finished", the
  199. following properties may also be present:
  200. * filename: The final filename (always present)
  201. * tmpfilename: The filename we're currently writing to
  202. * downloaded_bytes: Bytes on disk
  203. * total_bytes: Size of the whole file, None if unknown
  204. * total_bytes_estimate: Guess of the eventual file size,
  205. None if unavailable.
  206. * elapsed: The number of seconds since download started.
  207. * eta: The estimated time in seconds, None if unknown
  208. * speed: The download speed in bytes/second, None if
  209. unknown
  210. * fragment_index: The counter of the currently
  211. downloaded video fragment.
  212. * fragment_count: The number of fragments (= individual
  213. files that will be merged)
  214. Progress hooks are guaranteed to be called at least once
  215. (with status "finished") if the download is successful.
  216. merge_output_format: Extension to use when merging formats.
  217. fixup: Automatically correct known faults of the file.
  218. One of:
  219. - "never": do nothing
  220. - "warn": only emit a warning
  221. - "detect_or_warn": check whether we can do anything
  222. about it, warn otherwise (default)
  223. source_address: (Experimental) Client-side IP address to bind to.
  224. call_home: Boolean, true iff we are allowed to contact the
  225. youtube-dl servers for debugging.
  226. sleep_interval: Number of seconds to sleep before each download.
  227. listformats: Print an overview of available video formats and exit.
  228. list_thumbnails: Print a table of all thumbnails and exit.
  229. match_filter: A function that gets called with the info_dict of
  230. every video.
  231. If it returns a message, the video is ignored.
  232. If it returns None, the video is downloaded.
  233. match_filter_func in utils.py is one example for this.
  234. no_color: Do not emit color codes in output.
  235. The following options determine which downloader is picked:
  236. external_downloader: Executable of the external downloader to call.
  237. None or unset for standard (built-in) downloader.
  238. hls_prefer_native: Use the native HLS downloader instead of ffmpeg/avconv.
  239. The following parameters are not used by YoutubeDL itself, they are used by
  240. the downloader (see youtube_dl/downloader/common.py):
  241. nopart, updatetime, buffersize, ratelimit, min_filesize, max_filesize, test,
  242. noresizebuffer, retries, continuedl, noprogress, consoletitle,
  243. xattr_set_filesize, external_downloader_args.
  244. The following options are used by the post processors:
  245. prefer_ffmpeg: If True, use ffmpeg instead of avconv if both are available,
  246. otherwise prefer avconv.
  247. """
  248. params = None
  249. _ies = []
  250. _pps = []
  251. _download_retcode = None
  252. _num_downloads = None
  253. _screen_file = None
  254. def __init__(self, params=None, auto_init=True):
  255. """Create a FileDownloader object with the given options."""
  256. if params is None:
  257. params = {}
  258. self._ies = []
  259. self._ies_instances = {}
  260. self._pps = []
  261. self._progress_hooks = []
  262. self._download_retcode = 0
  263. self._num_downloads = 0
  264. self._screen_file = [sys.stdout, sys.stderr][params.get('logtostderr', False)]
  265. self._err_file = sys.stderr
  266. self.params = params
  267. self.cache = Cache(self)
  268. if params.get('bidi_workaround', False):
  269. try:
  270. import pty
  271. master, slave = pty.openpty()
  272. width = compat_get_terminal_size().columns
  273. if width is None:
  274. width_args = []
  275. else:
  276. width_args = ['-w', str(width)]
  277. sp_kwargs = dict(
  278. stdin=subprocess.PIPE,
  279. stdout=slave,
  280. stderr=self._err_file)
  281. try:
  282. self._output_process = subprocess.Popen(
  283. ['bidiv'] + width_args, **sp_kwargs
  284. )
  285. except OSError:
  286. self._output_process = subprocess.Popen(
  287. ['fribidi', '-c', 'UTF-8'] + width_args, **sp_kwargs)
  288. self._output_channel = os.fdopen(master, 'rb')
  289. except OSError as ose:
  290. if ose.errno == 2:
  291. 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 $PATH.')
  292. else:
  293. raise
  294. if (sys.version_info >= (3,) and sys.platform != 'win32' and
  295. sys.getfilesystemencoding() in ['ascii', 'ANSI_X3.4-1968'] and
  296. not params.get('restrictfilenames', False)):
  297. # On Python 3, the Unicode filesystem API will throw errors (#1474)
  298. self.report_warning(
  299. 'Assuming --restrict-filenames since file system encoding '
  300. 'cannot encode all characters. '
  301. 'Set the LC_ALL environment variable to fix this.')
  302. self.params['restrictfilenames'] = True
  303. if isinstance(params.get('outtmpl'), bytes):
  304. self.report_warning(
  305. 'Parameter outtmpl is bytes, but should be a unicode string. '
  306. 'Put from __future__ import unicode_literals at the top of your code file or consider switching to Python 3.x.')
  307. self._setup_opener()
  308. if auto_init:
  309. self.print_debug_header()
  310. self.add_default_info_extractors()
  311. for pp_def_raw in self.params.get('postprocessors', []):
  312. pp_class = get_postprocessor(pp_def_raw['key'])
  313. pp_def = dict(pp_def_raw)
  314. del pp_def['key']
  315. pp = pp_class(self, **compat_kwargs(pp_def))
  316. self.add_post_processor(pp)
  317. for ph in self.params.get('progress_hooks', []):
  318. self.add_progress_hook(ph)
  319. def warn_if_short_id(self, argv):
  320. # short YouTube ID starting with dash?
  321. idxs = [
  322. i for i, a in enumerate(argv)
  323. if re.match(r'^-[0-9A-Za-z_-]{10}$', a)]
  324. if idxs:
  325. correct_argv = (
  326. ['youtube-dl'] +
  327. [a for i, a in enumerate(argv) if i not in idxs] +
  328. ['--'] + [argv[i] for i in idxs]
  329. )
  330. self.report_warning(
  331. 'Long argument string detected. '
  332. 'Use -- to separate parameters and URLs, like this:\n%s\n' %
  333. args_to_str(correct_argv))
  334. def add_info_extractor(self, ie):
  335. """Add an InfoExtractor object to the end of the list."""
  336. self._ies.append(ie)
  337. self._ies_instances[ie.ie_key()] = ie
  338. ie.set_downloader(self)
  339. def get_info_extractor(self, ie_key):
  340. """
  341. Get an instance of an IE with name ie_key, it will try to get one from
  342. the _ies list, if there's no instance it will create a new one and add
  343. it to the extractor list.
  344. """
  345. ie = self._ies_instances.get(ie_key)
  346. if ie is None:
  347. ie = get_info_extractor(ie_key)()
  348. self.add_info_extractor(ie)
  349. return ie
  350. def add_default_info_extractors(self):
  351. """
  352. Add the InfoExtractors returned by gen_extractors to the end of the list
  353. """
  354. for ie in gen_extractors():
  355. self.add_info_extractor(ie)
  356. def add_post_processor(self, pp):
  357. """Add a PostProcessor object to the end of the chain."""
  358. self._pps.append(pp)
  359. pp.set_downloader(self)
  360. def add_progress_hook(self, ph):
  361. """Add the progress hook (currently only for the file downloader)"""
  362. self._progress_hooks.append(ph)
  363. def _bidi_workaround(self, message):
  364. if not hasattr(self, '_output_channel'):
  365. return message
  366. assert hasattr(self, '_output_process')
  367. assert isinstance(message, compat_str)
  368. line_count = message.count('\n') + 1
  369. self._output_process.stdin.write((message + '\n').encode('utf-8'))
  370. self._output_process.stdin.flush()
  371. res = ''.join(self._output_channel.readline().decode('utf-8')
  372. for _ in range(line_count))
  373. return res[:-len('\n')]
  374. def to_screen(self, message, skip_eol=False):
  375. """Print message to stdout if not in quiet mode."""
  376. return self.to_stdout(message, skip_eol, check_quiet=True)
  377. def _write_string(self, s, out=None):
  378. write_string(s, out=out, encoding=self.params.get('encoding'))
  379. def to_stdout(self, message, skip_eol=False, check_quiet=False):
  380. """Print message to stdout if not in quiet mode."""
  381. if self.params.get('logger'):
  382. self.params['logger'].debug(message)
  383. elif not check_quiet or not self.params.get('quiet', False):
  384. message = self._bidi_workaround(message)
  385. terminator = ['\n', ''][skip_eol]
  386. output = message + terminator
  387. self._write_string(output, self._screen_file)
  388. def to_stderr(self, message):
  389. """Print message to stderr."""
  390. assert isinstance(message, compat_str)
  391. if self.params.get('logger'):
  392. self.params['logger'].error(message)
  393. else:
  394. message = self._bidi_workaround(message)
  395. output = message + '\n'
  396. self._write_string(output, self._err_file)
  397. def to_console_title(self, message):
  398. if not self.params.get('consoletitle', False):
  399. return
  400. if os.name == 'nt' and ctypes.windll.kernel32.GetConsoleWindow():
  401. # c_wchar_p() might not be necessary if `message` is
  402. # already of type unicode()
  403. ctypes.windll.kernel32.SetConsoleTitleW(ctypes.c_wchar_p(message))
  404. elif 'TERM' in os.environ:
  405. self._write_string('\033]0;%s\007' % message, self._screen_file)
  406. def save_console_title(self):
  407. if not self.params.get('consoletitle', False):
  408. return
  409. if 'TERM' in os.environ:
  410. # Save the title on stack
  411. self._write_string('\033[22;0t', self._screen_file)
  412. def restore_console_title(self):
  413. if not self.params.get('consoletitle', False):
  414. return
  415. if 'TERM' in os.environ:
  416. # Restore the title from stack
  417. self._write_string('\033[23;0t', self._screen_file)
  418. def __enter__(self):
  419. self.save_console_title()
  420. return self
  421. def __exit__(self, *args):
  422. self.restore_console_title()
  423. if self.params.get('cookiefile') is not None:
  424. self.cookiejar.save()
  425. def trouble(self, message=None, tb=None):
  426. """Determine action to take when a download problem appears.
  427. Depending on if the downloader has been configured to ignore
  428. download errors or not, this method may throw an exception or
  429. not when errors are found, after printing the message.
  430. tb, if given, is additional traceback information.
  431. """
  432. if message is not None:
  433. self.to_stderr(message)
  434. if self.params.get('verbose'):
  435. if tb is None:
  436. if sys.exc_info()[0]: # if .trouble has been called from an except block
  437. tb = ''
  438. if hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
  439. tb += ''.join(traceback.format_exception(*sys.exc_info()[1].exc_info))
  440. tb += compat_str(traceback.format_exc())
  441. else:
  442. tb_data = traceback.format_list(traceback.extract_stack())
  443. tb = ''.join(tb_data)
  444. self.to_stderr(tb)
  445. if not self.params.get('ignoreerrors', False):
  446. if sys.exc_info()[0] and hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
  447. exc_info = sys.exc_info()[1].exc_info
  448. else:
  449. exc_info = sys.exc_info()
  450. raise DownloadError(message, exc_info)
  451. self._download_retcode = 1
  452. def report_warning(self, message):
  453. '''
  454. Print the message to stderr, it will be prefixed with 'WARNING:'
  455. If stderr is a tty file the 'WARNING:' will be colored
  456. '''
  457. if self.params.get('logger') is not None:
  458. self.params['logger'].warning(message)
  459. else:
  460. if self.params.get('no_warnings'):
  461. return
  462. if not self.params.get('no_color') and self._err_file.isatty() and os.name != 'nt':
  463. _msg_header = '\033[0;33mWARNING:\033[0m'
  464. else:
  465. _msg_header = 'WARNING:'
  466. warning_message = '%s %s' % (_msg_header, message)
  467. self.to_stderr(warning_message)
  468. def report_error(self, message, tb=None):
  469. '''
  470. Do the same as trouble, but prefixes the message with 'ERROR:', colored
  471. in red if stderr is a tty file.
  472. '''
  473. if not self.params.get('no_color') and self._err_file.isatty() and os.name != 'nt':
  474. _msg_header = '\033[0;31mERROR:\033[0m'
  475. else:
  476. _msg_header = 'ERROR:'
  477. error_message = '%s %s' % (_msg_header, message)
  478. self.trouble(error_message, tb)
  479. def report_file_already_downloaded(self, file_name):
  480. """Report file has already been fully downloaded."""
  481. try:
  482. self.to_screen('[download] %s has already been downloaded' % file_name)
  483. except UnicodeEncodeError:
  484. self.to_screen('[download] The file has already been downloaded')
  485. def prepare_filename(self, info_dict):
  486. """Generate the output filename."""
  487. try:
  488. template_dict = dict(info_dict)
  489. template_dict['epoch'] = int(time.time())
  490. autonumber_size = self.params.get('autonumber_size')
  491. if autonumber_size is None:
  492. autonumber_size = 5
  493. autonumber_templ = '%0' + str(autonumber_size) + 'd'
  494. template_dict['autonumber'] = autonumber_templ % self._num_downloads
  495. if template_dict.get('playlist_index') is not None:
  496. template_dict['playlist_index'] = '%0*d' % (len(str(template_dict['n_entries'])), template_dict['playlist_index'])
  497. if template_dict.get('resolution') is None:
  498. if template_dict.get('width') and template_dict.get('height'):
  499. template_dict['resolution'] = '%dx%d' % (template_dict['width'], template_dict['height'])
  500. elif template_dict.get('height'):
  501. template_dict['resolution'] = '%sp' % template_dict['height']
  502. elif template_dict.get('width'):
  503. template_dict['resolution'] = '?x%d' % template_dict['width']
  504. sanitize = lambda k, v: sanitize_filename(
  505. compat_str(v),
  506. restricted=self.params.get('restrictfilenames'),
  507. is_id=(k == 'id'))
  508. template_dict = dict((k, sanitize(k, v))
  509. for k, v in template_dict.items()
  510. if v is not None)
  511. template_dict = collections.defaultdict(lambda: 'NA', template_dict)
  512. outtmpl = sanitize_path(self.params.get('outtmpl', DEFAULT_OUTTMPL))
  513. tmpl = compat_expanduser(outtmpl)
  514. filename = tmpl % template_dict
  515. # Temporary fix for #4787
  516. # 'Treat' all problem characters by passing filename through preferredencoding
  517. # to workaround encoding issues with subprocess on python2 @ Windows
  518. if sys.version_info < (3, 0) and sys.platform == 'win32':
  519. filename = encodeFilename(filename, True).decode(preferredencoding())
  520. return filename
  521. except ValueError as err:
  522. self.report_error('Error in output template: ' + str(err) + ' (encoding: ' + repr(preferredencoding()) + ')')
  523. return None
  524. def _match_entry(self, info_dict, incomplete):
  525. """ Returns None iff the file should be downloaded """
  526. video_title = info_dict.get('title', info_dict.get('id', 'video'))
  527. if 'title' in info_dict:
  528. # This can happen when we're just evaluating the playlist
  529. title = info_dict['title']
  530. matchtitle = self.params.get('matchtitle', False)
  531. if matchtitle:
  532. if not re.search(matchtitle, title, re.IGNORECASE):
  533. return '"' + title + '" title did not match pattern "' + matchtitle + '"'
  534. rejecttitle = self.params.get('rejecttitle', False)
  535. if rejecttitle:
  536. if re.search(rejecttitle, title, re.IGNORECASE):
  537. return '"' + title + '" title matched reject pattern "' + rejecttitle + '"'
  538. date = info_dict.get('upload_date', None)
  539. if date is not None:
  540. dateRange = self.params.get('daterange', DateRange())
  541. if date not in dateRange:
  542. return '%s upload date is not in range %s' % (date_from_str(date).isoformat(), dateRange)
  543. view_count = info_dict.get('view_count', None)
  544. if view_count is not None:
  545. min_views = self.params.get('min_views')
  546. if min_views is not None and view_count < min_views:
  547. return 'Skipping %s, because it has not reached minimum view count (%d/%d)' % (video_title, view_count, min_views)
  548. max_views = self.params.get('max_views')
  549. if max_views is not None and view_count > max_views:
  550. return 'Skipping %s, because it has exceeded the maximum view count (%d/%d)' % (video_title, view_count, max_views)
  551. if age_restricted(info_dict.get('age_limit'), self.params.get('age_limit')):
  552. return 'Skipping "%s" because it is age restricted' % video_title
  553. if self.in_download_archive(info_dict):
  554. return '%s has already been recorded in archive' % video_title
  555. if not incomplete:
  556. match_filter = self.params.get('match_filter')
  557. if match_filter is not None:
  558. ret = match_filter(info_dict)
  559. if ret is not None:
  560. return ret
  561. return None
  562. @staticmethod
  563. def add_extra_info(info_dict, extra_info):
  564. '''Set the keys from extra_info in info dict if they are missing'''
  565. for key, value in extra_info.items():
  566. info_dict.setdefault(key, value)
  567. def extract_info(self, url, download=True, ie_key=None, extra_info={},
  568. process=True):
  569. '''
  570. Returns a list with a dictionary for each video we find.
  571. If 'download', also downloads the videos.
  572. extra_info is a dict containing the extra values to add to each result
  573. '''
  574. if ie_key:
  575. ies = [self.get_info_extractor(ie_key)]
  576. else:
  577. ies = self._ies
  578. for ie in ies:
  579. if not ie.suitable(url):
  580. continue
  581. if not ie.working():
  582. self.report_warning('The program functionality for this site has been marked as broken, '
  583. 'and will probably not work.')
  584. try:
  585. ie_result = ie.extract(url)
  586. if ie_result is None: # Finished already (backwards compatibility; listformats and friends should be moved here)
  587. break
  588. if isinstance(ie_result, list):
  589. # Backwards compatibility: old IE result format
  590. ie_result = {
  591. '_type': 'compat_list',
  592. 'entries': ie_result,
  593. }
  594. self.add_default_extra_info(ie_result, ie, url)
  595. if process:
  596. return self.process_ie_result(ie_result, download, extra_info)
  597. else:
  598. return ie_result
  599. except ExtractorError as de: # An error we somewhat expected
  600. self.report_error(compat_str(de), de.format_traceback())
  601. break
  602. except MaxDownloadsReached:
  603. raise
  604. except Exception as e:
  605. if self.params.get('ignoreerrors', False):
  606. self.report_error(compat_str(e), tb=compat_str(traceback.format_exc()))
  607. break
  608. else:
  609. raise
  610. else:
  611. self.report_error('no suitable InfoExtractor for URL %s' % url)
  612. def add_default_extra_info(self, ie_result, ie, url):
  613. self.add_extra_info(ie_result, {
  614. 'extractor': ie.IE_NAME,
  615. 'webpage_url': url,
  616. 'webpage_url_basename': url_basename(url),
  617. 'extractor_key': ie.ie_key(),
  618. })
  619. def process_ie_result(self, ie_result, download=True, extra_info={}):
  620. """
  621. Take the result of the ie(may be modified) and resolve all unresolved
  622. references (URLs, playlist items).
  623. It will also download the videos if 'download'.
  624. Returns the resolved ie_result.
  625. """
  626. result_type = ie_result.get('_type', 'video')
  627. if result_type in ('url', 'url_transparent'):
  628. extract_flat = self.params.get('extract_flat', False)
  629. if ((extract_flat == 'in_playlist' and 'playlist' in extra_info) or
  630. extract_flat is True):
  631. if self.params.get('forcejson', False):
  632. self.to_stdout(json.dumps(ie_result))
  633. return ie_result
  634. if result_type == 'video':
  635. self.add_extra_info(ie_result, extra_info)
  636. return self.process_video_result(ie_result, download=download)
  637. elif result_type == 'url':
  638. # We have to add extra_info to the results because it may be
  639. # contained in a playlist
  640. return self.extract_info(ie_result['url'],
  641. download,
  642. ie_key=ie_result.get('ie_key'),
  643. extra_info=extra_info)
  644. elif result_type == 'url_transparent':
  645. # Use the information from the embedding page
  646. info = self.extract_info(
  647. ie_result['url'], ie_key=ie_result.get('ie_key'),
  648. extra_info=extra_info, download=False, process=False)
  649. force_properties = dict(
  650. (k, v) for k, v in ie_result.items() if v is not None)
  651. for f in ('_type', 'url'):
  652. if f in force_properties:
  653. del force_properties[f]
  654. new_result = info.copy()
  655. new_result.update(force_properties)
  656. assert new_result.get('_type') != 'url_transparent'
  657. return self.process_ie_result(
  658. new_result, download=download, extra_info=extra_info)
  659. elif result_type == 'playlist' or result_type == 'multi_video':
  660. # We process each entry in the playlist
  661. playlist = ie_result.get('title', None) or ie_result.get('id', None)
  662. self.to_screen('[download] Downloading playlist: %s' % playlist)
  663. playlist_results = []
  664. playliststart = self.params.get('playliststart', 1) - 1
  665. playlistend = self.params.get('playlistend', None)
  666. # For backwards compatibility, interpret -1 as whole list
  667. if playlistend == -1:
  668. playlistend = None
  669. playlistitems_str = self.params.get('playlist_items', None)
  670. playlistitems = None
  671. if playlistitems_str is not None:
  672. def iter_playlistitems(format):
  673. for string_segment in format.split(','):
  674. if '-' in string_segment:
  675. start, end = string_segment.split('-')
  676. for item in range(int(start), int(end) + 1):
  677. yield int(item)
  678. else:
  679. yield int(string_segment)
  680. playlistitems = iter_playlistitems(playlistitems_str)
  681. ie_entries = ie_result['entries']
  682. if isinstance(ie_entries, list):
  683. n_all_entries = len(ie_entries)
  684. if playlistitems:
  685. entries = [
  686. ie_entries[i - 1] for i in playlistitems
  687. if -n_all_entries <= i - 1 < n_all_entries]
  688. else:
  689. entries = ie_entries[playliststart:playlistend]
  690. n_entries = len(entries)
  691. self.to_screen(
  692. "[%s] playlist %s: Collected %d video ids (downloading %d of them)" %
  693. (ie_result['extractor'], playlist, n_all_entries, n_entries))
  694. elif isinstance(ie_entries, PagedList):
  695. if playlistitems:
  696. entries = []
  697. for item in playlistitems:
  698. entries.extend(ie_entries.getslice(
  699. item - 1, item
  700. ))
  701. else:
  702. entries = ie_entries.getslice(
  703. playliststart, playlistend)
  704. n_entries = len(entries)
  705. self.to_screen(
  706. "[%s] playlist %s: Downloading %d videos" %
  707. (ie_result['extractor'], playlist, n_entries))
  708. else: # iterable
  709. if playlistitems:
  710. entry_list = list(ie_entries)
  711. entries = [entry_list[i - 1] for i in playlistitems]
  712. else:
  713. entries = list(itertools.islice(
  714. ie_entries, playliststart, playlistend))
  715. n_entries = len(entries)
  716. self.to_screen(
  717. "[%s] playlist %s: Downloading %d videos" %
  718. (ie_result['extractor'], playlist, n_entries))
  719. if self.params.get('playlistreverse', False):
  720. entries = entries[::-1]
  721. for i, entry in enumerate(entries, 1):
  722. self.to_screen('[download] Downloading video %s of %s' % (i, n_entries))
  723. extra = {
  724. 'n_entries': n_entries,
  725. 'playlist': playlist,
  726. 'playlist_id': ie_result.get('id'),
  727. 'playlist_title': ie_result.get('title'),
  728. 'playlist_index': i + playliststart,
  729. 'extractor': ie_result['extractor'],
  730. 'webpage_url': ie_result['webpage_url'],
  731. 'webpage_url_basename': url_basename(ie_result['webpage_url']),
  732. 'extractor_key': ie_result['extractor_key'],
  733. }
  734. reason = self._match_entry(entry, incomplete=True)
  735. if reason is not None:
  736. self.to_screen('[download] ' + reason)
  737. continue
  738. entry_result = self.process_ie_result(entry,
  739. download=download,
  740. extra_info=extra)
  741. playlist_results.append(entry_result)
  742. ie_result['entries'] = playlist_results
  743. return ie_result
  744. elif result_type == 'compat_list':
  745. self.report_warning(
  746. 'Extractor %s returned a compat_list result. '
  747. 'It needs to be updated.' % ie_result.get('extractor'))
  748. def _fixup(r):
  749. self.add_extra_info(
  750. r,
  751. {
  752. 'extractor': ie_result['extractor'],
  753. 'webpage_url': ie_result['webpage_url'],
  754. 'webpage_url_basename': url_basename(ie_result['webpage_url']),
  755. 'extractor_key': ie_result['extractor_key'],
  756. }
  757. )
  758. return r
  759. ie_result['entries'] = [
  760. self.process_ie_result(_fixup(r), download, extra_info)
  761. for r in ie_result['entries']
  762. ]
  763. return ie_result
  764. else:
  765. raise Exception('Invalid result type: %s' % result_type)
  766. def _apply_format_filter(self, format_spec, available_formats):
  767. " Returns a tuple of the remaining format_spec and filtered formats "
  768. OPERATORS = {
  769. '<': operator.lt,
  770. '<=': operator.le,
  771. '>': operator.gt,
  772. '>=': operator.ge,
  773. '=': operator.eq,
  774. '!=': operator.ne,
  775. }
  776. operator_rex = re.compile(r'''(?x)\s*\[
  777. (?P<key>width|height|tbr|abr|vbr|asr|filesize|fps)
  778. \s*(?P<op>%s)(?P<none_inclusive>\s*\?)?\s*
  779. (?P<value>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?)
  780. \]$
  781. ''' % '|'.join(map(re.escape, OPERATORS.keys())))
  782. m = operator_rex.search(format_spec)
  783. if m:
  784. try:
  785. comparison_value = int(m.group('value'))
  786. except ValueError:
  787. comparison_value = parse_filesize(m.group('value'))
  788. if comparison_value is None:
  789. comparison_value = parse_filesize(m.group('value') + 'B')
  790. if comparison_value is None:
  791. raise ValueError(
  792. 'Invalid value %r in format specification %r' % (
  793. m.group('value'), format_spec))
  794. op = OPERATORS[m.group('op')]
  795. if not m:
  796. STR_OPERATORS = {
  797. '=': operator.eq,
  798. '!=': operator.ne,
  799. }
  800. str_operator_rex = re.compile(r'''(?x)\s*\[
  801. \s*(?P<key>ext|acodec|vcodec|container|protocol)
  802. \s*(?P<op>%s)(?P<none_inclusive>\s*\?)?
  803. \s*(?P<value>[a-zA-Z0-9_-]+)
  804. \s*\]$
  805. ''' % '|'.join(map(re.escape, STR_OPERATORS.keys())))
  806. m = str_operator_rex.search(format_spec)
  807. if m:
  808. comparison_value = m.group('value')
  809. op = STR_OPERATORS[m.group('op')]
  810. if not m:
  811. raise ValueError('Invalid format specification %r' % format_spec)
  812. def _filter(f):
  813. actual_value = f.get(m.group('key'))
  814. if actual_value is None:
  815. return m.group('none_inclusive')
  816. return op(actual_value, comparison_value)
  817. new_formats = [f for f in available_formats if _filter(f)]
  818. new_format_spec = format_spec[:-len(m.group(0))]
  819. if not new_format_spec:
  820. new_format_spec = 'best'
  821. return (new_format_spec, new_formats)
  822. def select_format(self, format_spec, available_formats):
  823. while format_spec.endswith(']'):
  824. format_spec, available_formats = self._apply_format_filter(
  825. format_spec, available_formats)
  826. if not available_formats:
  827. return None
  828. if format_spec in ['best', 'worst', None]:
  829. format_idx = 0 if format_spec == 'worst' else -1
  830. audiovideo_formats = [
  831. f for f in available_formats
  832. if f.get('vcodec') != 'none' and f.get('acodec') != 'none']
  833. if audiovideo_formats:
  834. return audiovideo_formats[format_idx]
  835. # for audio only urls, select the best/worst audio format
  836. elif all(f.get('acodec') != 'none' for f in available_formats):
  837. return available_formats[format_idx]
  838. elif format_spec == 'bestaudio':
  839. audio_formats = [
  840. f for f in available_formats
  841. if f.get('vcodec') == 'none']
  842. if audio_formats:
  843. return audio_formats[-1]
  844. elif format_spec == 'worstaudio':
  845. audio_formats = [
  846. f for f in available_formats
  847. if f.get('vcodec') == 'none']
  848. if audio_formats:
  849. return audio_formats[0]
  850. elif format_spec == 'bestvideo':
  851. video_formats = [
  852. f for f in available_formats
  853. if f.get('acodec') == 'none']
  854. if video_formats:
  855. return video_formats[-1]
  856. elif format_spec == 'worstvideo':
  857. video_formats = [
  858. f for f in available_formats
  859. if f.get('acodec') == 'none']
  860. if video_formats:
  861. return video_formats[0]
  862. else:
  863. extensions = ['mp4', 'flv', 'webm', '3gp', 'm4a', 'mp3', 'ogg', 'aac', 'wav']
  864. if format_spec in extensions:
  865. filter_f = lambda f: f['ext'] == format_spec
  866. else:
  867. filter_f = lambda f: f['format_id'] == format_spec
  868. matches = list(filter(filter_f, available_formats))
  869. if matches:
  870. return matches[-1]
  871. return None
  872. def _calc_headers(self, info_dict):
  873. res = std_headers.copy()
  874. add_headers = info_dict.get('http_headers')
  875. if add_headers:
  876. res.update(add_headers)
  877. cookies = self._calc_cookies(info_dict)
  878. if cookies:
  879. res['Cookie'] = cookies
  880. return res
  881. def _calc_cookies(self, info_dict):
  882. pr = compat_urllib_request.Request(info_dict['url'])
  883. self.cookiejar.add_cookie_header(pr)
  884. return pr.get_header('Cookie')
  885. def process_video_result(self, info_dict, download=True):
  886. assert info_dict.get('_type', 'video') == 'video'
  887. if 'id' not in info_dict:
  888. raise ExtractorError('Missing "id" field in extractor result')
  889. if 'title' not in info_dict:
  890. raise ExtractorError('Missing "title" field in extractor result')
  891. if 'playlist' not in info_dict:
  892. # It isn't part of a playlist
  893. info_dict['playlist'] = None
  894. info_dict['playlist_index'] = None
  895. thumbnails = info_dict.get('thumbnails')
  896. if thumbnails is None:
  897. thumbnail = info_dict.get('thumbnail')
  898. if thumbnail:
  899. info_dict['thumbnails'] = thumbnails = [{'url': thumbnail}]
  900. if thumbnails:
  901. thumbnails.sort(key=lambda t: (
  902. t.get('preference'), t.get('width'), t.get('height'),
  903. t.get('id'), t.get('url')))
  904. for i, t in enumerate(thumbnails):
  905. if 'width' in t and 'height' in t:
  906. t['resolution'] = '%dx%d' % (t['width'], t['height'])
  907. if t.get('id') is None:
  908. t['id'] = '%d' % i
  909. if thumbnails and 'thumbnail' not in info_dict:
  910. info_dict['thumbnail'] = thumbnails[-1]['url']
  911. if 'display_id' not in info_dict and 'id' in info_dict:
  912. info_dict['display_id'] = info_dict['id']
  913. if info_dict.get('upload_date') is None and info_dict.get('timestamp') is not None:
  914. # Working around negative timestamps in Windows
  915. # (see http://bugs.python.org/issue1646728)
  916. if info_dict['timestamp'] < 0 and os.name == 'nt':
  917. info_dict['timestamp'] = 0
  918. upload_date = datetime.datetime.utcfromtimestamp(
  919. info_dict['timestamp'])
  920. info_dict['upload_date'] = upload_date.strftime('%Y%m%d')
  921. if self.params.get('listsubtitles', False):
  922. if 'automatic_captions' in info_dict:
  923. self.list_subtitles(info_dict['id'], info_dict.get('automatic_captions'), 'automatic captions')
  924. self.list_subtitles(info_dict['id'], info_dict.get('subtitles'), 'subtitles')
  925. return
  926. info_dict['requested_subtitles'] = self.process_subtitles(
  927. info_dict['id'], info_dict.get('subtitles'),
  928. info_dict.get('automatic_captions'))
  929. # This extractors handle format selection themselves
  930. if info_dict['extractor'] in ['Youku']:
  931. if download:
  932. self.process_info(info_dict)
  933. return info_dict
  934. # We now pick which formats have to be downloaded
  935. if info_dict.get('formats') is None:
  936. # There's only one format available
  937. formats = [info_dict]
  938. else:
  939. formats = info_dict['formats']
  940. if not formats:
  941. raise ExtractorError('No video formats found!')
  942. formats_dict = {}
  943. # We check that all the formats have the format and format_id fields
  944. for i, format in enumerate(formats):
  945. if 'url' not in format:
  946. raise ExtractorError('Missing "url" key in result (index %d)' % i)
  947. if format.get('format_id') is None:
  948. format['format_id'] = compat_str(i)
  949. format_id = format['format_id']
  950. if format_id not in formats_dict:
  951. formats_dict[format_id] = []
  952. formats_dict[format_id].append(format)
  953. # Make sure all formats have unique format_id
  954. for format_id, ambiguous_formats in formats_dict.items():
  955. if len(ambiguous_formats) > 1:
  956. for i, format in enumerate(ambiguous_formats):
  957. format['format_id'] = '%s-%d' % (format_id, i)
  958. for i, format in enumerate(formats):
  959. if format.get('format') is None:
  960. format['format'] = '{id} - {res}{note}'.format(
  961. id=format['format_id'],
  962. res=self.format_resolution(format),
  963. note=' ({0})'.format(format['format_note']) if format.get('format_note') is not None else '',
  964. )
  965. # Automatically determine file extension if missing
  966. if 'ext' not in format:
  967. format['ext'] = determine_ext(format['url']).lower()
  968. # Add HTTP headers, so that external programs can use them from the
  969. # json output
  970. full_format_info = info_dict.copy()
  971. full_format_info.update(format)
  972. format['http_headers'] = self._calc_headers(full_format_info)
  973. # TODO Central sorting goes here
  974. if formats[0] is not info_dict:
  975. # only set the 'formats' fields if the original info_dict list them
  976. # otherwise we end up with a circular reference, the first (and unique)
  977. # element in the 'formats' field in info_dict is info_dict itself,
  978. # wich can't be exported to json
  979. info_dict['formats'] = formats
  980. if self.params.get('listformats'):
  981. self.list_formats(info_dict)
  982. return
  983. if self.params.get('list_thumbnails'):
  984. self.list_thumbnails(info_dict)
  985. return
  986. req_format = self.params.get('format')
  987. if req_format is None:
  988. req_format_list = []
  989. if (self.params.get('outtmpl', DEFAULT_OUTTMPL) != '-' and
  990. info_dict['extractor'] in ['youtube', 'ted']):
  991. merger = FFmpegMergerPP(self)
  992. if merger.available and merger.can_merge():
  993. req_format_list.append('bestvideo+bestaudio')
  994. req_format_list.append('best')
  995. req_format = '/'.join(req_format_list)
  996. formats_to_download = []
  997. if req_format == 'all':
  998. formats_to_download = formats
  999. else:
  1000. for rfstr in req_format.split(','):
  1001. # We can accept formats requested in the format: 34/5/best, we pick
  1002. # the first that is available, starting from left
  1003. req_formats = rfstr.split('/')
  1004. for rf in req_formats:
  1005. if re.match(r'.+?\+.+?', rf) is not None:
  1006. # Two formats have been requested like '137+139'
  1007. format_1, format_2 = rf.split('+')
  1008. formats_info = (self.select_format(format_1, formats),
  1009. self.select_format(format_2, formats))
  1010. if all(formats_info):
  1011. # The first format must contain the video and the
  1012. # second the audio
  1013. if formats_info[0].get('vcodec') == 'none':
  1014. self.report_error('The first format must '
  1015. 'contain the video, try using '
  1016. '"-f %s+%s"' % (format_2, format_1))
  1017. return
  1018. output_ext = (
  1019. formats_info[0]['ext']
  1020. if self.params.get('merge_output_format') is None
  1021. else self.params['merge_output_format'])
  1022. selected_format = {
  1023. 'requested_formats': formats_info,
  1024. 'format': '%s+%s' % (formats_info[0].get('format'),
  1025. formats_info[1].get('format')),
  1026. 'format_id': '%s+%s' % (formats_info[0].get('format_id'),
  1027. formats_info[1].get('format_id')),
  1028. 'width': formats_info[0].get('width'),
  1029. 'height': formats_info[0].get('height'),
  1030. 'resolution': formats_info[0].get('resolution'),
  1031. 'fps': formats_info[0].get('fps'),
  1032. 'vcodec': formats_info[0].get('vcodec'),
  1033. 'vbr': formats_info[0].get('vbr'),
  1034. 'stretched_ratio': formats_info[0].get('stretched_ratio'),
  1035. 'acodec': formats_info[1].get('acodec'),
  1036. 'abr': formats_info[1].get('abr'),
  1037. 'ext': output_ext,
  1038. }
  1039. else:
  1040. selected_format = None
  1041. else:
  1042. selected_format = self.select_format(rf, formats)
  1043. if selected_format is not None:
  1044. formats_to_download.append(selected_format)
  1045. break
  1046. if not formats_to_download:
  1047. raise ExtractorError('requested format not available',
  1048. expected=True)
  1049. if download:
  1050. if len(formats_to_download) > 1:
  1051. self.to_screen('[info] %s: downloading video in %s formats' % (info_dict['id'], len(formats_to_download)))
  1052. for format in formats_to_download:
  1053. new_info = dict(info_dict)
  1054. new_info.update(format)
  1055. self.process_info(new_info)
  1056. # We update the info dict with the best quality format (backwards compatibility)
  1057. info_dict.update(formats_to_download[-1])
  1058. return info_dict
  1059. def process_subtitles(self, video_id, normal_subtitles, automatic_captions):
  1060. """Select the requested subtitles and their format"""
  1061. available_subs = {}
  1062. if normal_subtitles and self.params.get('writesubtitles'):
  1063. available_subs.update(normal_subtitles)
  1064. if automatic_captions and self.params.get('writeautomaticsub'):
  1065. for lang, cap_info in automatic_captions.items():
  1066. if lang not in available_subs:
  1067. available_subs[lang] = cap_info
  1068. if (not self.params.get('writesubtitles') and not
  1069. self.params.get('writeautomaticsub') or not
  1070. available_subs):
  1071. return None
  1072. if self.params.get('allsubtitles', False):
  1073. requested_langs = available_subs.keys()
  1074. else:
  1075. if self.params.get('subtitleslangs', False):
  1076. requested_langs = self.params.get('subtitleslangs')
  1077. elif 'en' in available_subs:
  1078. requested_langs = ['en']
  1079. else:
  1080. requested_langs = [list(available_subs.keys())[0]]
  1081. formats_query = self.params.get('subtitlesformat', 'best')
  1082. formats_preference = formats_query.split('/') if formats_query else []
  1083. subs = {}
  1084. for lang in requested_langs:
  1085. formats = available_subs.get(lang)
  1086. if formats is None:
  1087. self.report_warning('%s subtitles not available for %s' % (lang, video_id))
  1088. continue
  1089. for ext in formats_preference:
  1090. if ext == 'best':
  1091. f = formats[-1]
  1092. break
  1093. matches = list(filter(lambda f: f['ext'] == ext, formats))
  1094. if matches:
  1095. f = matches[-1]
  1096. break
  1097. else:
  1098. f = formats[-1]
  1099. self.report_warning(
  1100. 'No subtitle format found matching "%s" for language %s, '
  1101. 'using %s' % (formats_query, lang, f['ext']))
  1102. subs[lang] = f
  1103. return subs
  1104. def process_info(self, info_dict):
  1105. """Process a single resolved IE result."""
  1106. assert info_dict.get('_type', 'video') == 'video'
  1107. max_downloads = self.params.get('max_downloads')
  1108. if max_downloads is not None:
  1109. if self._num_downloads >= int(max_downloads):
  1110. raise MaxDownloadsReached()
  1111. info_dict['fulltitle'] = info_dict['title']
  1112. if len(info_dict['title']) > 200:
  1113. info_dict['title'] = info_dict['title'][:197] + '...'
  1114. if 'format' not in info_dict:
  1115. info_dict['format'] = info_dict['ext']
  1116. reason = self._match_entry(info_dict, incomplete=False)
  1117. if reason is not None:
  1118. self.to_screen('[download] ' + reason)
  1119. return
  1120. self._num_downloads += 1
  1121. info_dict['_filename'] = filename = self.prepare_filename(info_dict)
  1122. # Forced printings
  1123. if self.params.get('forcetitle', False):
  1124. self.to_stdout(info_dict['fulltitle'])
  1125. if self.params.get('forceid', False):
  1126. self.to_stdout(info_dict['id'])
  1127. if self.params.get('forceurl', False):
  1128. if info_dict.get('requested_formats') is not None:
  1129. for f in info_dict['requested_formats']:
  1130. self.to_stdout(f['url'] + f.get('play_path', ''))
  1131. else:
  1132. # For RTMP URLs, also include the playpath
  1133. self.to_stdout(info_dict['url'] + info_dict.get('play_path', ''))
  1134. if self.params.get('forcethumbnail', False) and info_dict.get('thumbnail') is not None:
  1135. self.to_stdout(info_dict['thumbnail'])
  1136. if self.params.get('forcedescription', False) and info_dict.get('description') is not None:
  1137. self.to_stdout(info_dict['description'])
  1138. if self.params.get('forcefilename', False) and filename is not None:
  1139. self.to_stdout(filename)
  1140. if self.params.get('forceduration', False) and info_dict.get('duration') is not None:
  1141. self.to_stdout(formatSeconds(info_dict['duration']))
  1142. if self.params.get('forceformat', False):
  1143. self.to_stdout(info_dict['format'])
  1144. if self.params.get('forcejson', False):
  1145. self.to_stdout(json.dumps(info_dict))
  1146. # Do nothing else if in simulate mode
  1147. if self.params.get('simulate', False):
  1148. return
  1149. if filename is None:
  1150. return
  1151. try:
  1152. dn = os.path.dirname(sanitize_path(encodeFilename(filename)))
  1153. if dn and not os.path.exists(dn):
  1154. os.makedirs(dn)
  1155. except (OSError, IOError) as err:
  1156. self.report_error('unable to create directory ' + compat_str(err))
  1157. return
  1158. if self.params.get('writedescription', False):
  1159. descfn = replace_extension(filename, 'description', info_dict.get('ext'))
  1160. if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(descfn)):
  1161. self.to_screen('[info] Video description is already present')
  1162. elif info_dict.get('description') is None:
  1163. self.report_warning('There\'s no description to write.')
  1164. else:
  1165. try:
  1166. self.to_screen('[info] Writing video description to: ' + descfn)
  1167. with io.open(encodeFilename(descfn), 'w', encoding='utf-8') as descfile:
  1168. descfile.write(info_dict['description'])
  1169. except (OSError, IOError):
  1170. self.report_error('Cannot write description file ' + descfn)
  1171. return
  1172. if self.params.get('writeannotations', False):
  1173. annofn = replace_extension(filename, 'annotations.xml', info_dict.get('ext'))
  1174. if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(annofn)):
  1175. self.to_screen('[info] Video annotations are already present')
  1176. else:
  1177. try:
  1178. self.to_screen('[info] Writing video annotations to: ' + annofn)
  1179. with io.open(encodeFilename(annofn), 'w', encoding='utf-8') as annofile:
  1180. annofile.write(info_dict['annotations'])
  1181. except (KeyError, TypeError):
  1182. self.report_warning('There are no annotations to write.')
  1183. except (OSError, IOError):
  1184. self.report_error('Cannot write annotations file: ' + annofn)
  1185. return
  1186. subtitles_are_requested = any([self.params.get('writesubtitles', False),
  1187. self.params.get('writeautomaticsub')])
  1188. if subtitles_are_requested and info_dict.get('requested_subtitles'):
  1189. # subtitles download errors are already managed as troubles in relevant IE
  1190. # that way it will silently go on when used with unsupporting IE
  1191. subtitles = info_dict['requested_subtitles']
  1192. ie = self.get_info_extractor(info_dict['extractor_key'])
  1193. for sub_lang, sub_info in subtitles.items():
  1194. sub_format = sub_info['ext']
  1195. if sub_info.get('data') is not None:
  1196. sub_data = sub_info['data']
  1197. else:
  1198. try:
  1199. sub_data = ie._download_webpage(
  1200. sub_info['url'], info_dict['id'], note=False)
  1201. except ExtractorError as err:
  1202. self.report_warning('Unable to download subtitle for "%s": %s' %
  1203. (sub_lang, compat_str(err.cause)))
  1204. continue
  1205. try:
  1206. sub_filename = subtitles_filename(filename, sub_lang, sub_format)
  1207. if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(sub_filename)):
  1208. self.to_screen('[info] Video subtitle %s.%s is already_present' % (sub_lang, sub_format))
  1209. else:
  1210. self.to_screen('[info] Writing video subtitles to: ' + sub_filename)
  1211. with io.open(encodeFilename(sub_filename), 'w', encoding='utf-8') as subfile:
  1212. subfile.write(sub_data)
  1213. except (OSError, IOError):
  1214. self.report_error('Cannot write subtitles file ' + sub_filename)
  1215. return
  1216. if self.params.get('writeinfojson', False):
  1217. infofn = replace_extension(filename, 'info.json', info_dict.get('ext'))
  1218. if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(infofn)):
  1219. self.to_screen('[info] Video description metadata is already present')
  1220. else:
  1221. self.to_screen('[info] Writing video description metadata as JSON to: ' + infofn)
  1222. try:
  1223. write_json_file(self.filter_requested_info(info_dict), infofn)
  1224. except (OSError, IOError):
  1225. self.report_error('Cannot write metadata to JSON file ' + infofn)
  1226. return
  1227. self._write_thumbnails(info_dict, filename)
  1228. if not self.params.get('skip_download', False):
  1229. try:
  1230. def dl(name, info):
  1231. fd = get_suitable_downloader(info, self.params)(self, self.params)
  1232. for ph in self._progress_hooks:
  1233. fd.add_progress_hook(ph)
  1234. if self.params.get('verbose'):
  1235. self.to_stdout('[debug] Invoking downloader on %r' % info.get('url'))
  1236. return fd.download(name, info)
  1237. if info_dict.get('requested_formats') is not None:
  1238. downloaded = []
  1239. success = True
  1240. merger = FFmpegMergerPP(self)
  1241. if not merger.available:
  1242. postprocessors = []
  1243. self.report_warning('You have requested multiple '
  1244. 'formats but ffmpeg or avconv are not installed.'
  1245. ' The formats won\'t be merged.')
  1246. else:
  1247. postprocessors = [merger]
  1248. def compatible_formats(formats):
  1249. video, audio = formats
  1250. # Check extension
  1251. video_ext, audio_ext = audio.get('ext'), video.get('ext')
  1252. if video_ext and audio_ext:
  1253. COMPATIBLE_EXTS = (
  1254. ('mp3', 'mp4', 'm4a', 'm4p', 'm4b', 'm4r', 'm4v'),
  1255. ('webm')
  1256. )
  1257. for exts in COMPATIBLE_EXTS:
  1258. if video_ext in exts and audio_ext in exts:
  1259. return True
  1260. # TODO: Check acodec/vcodec
  1261. return False
  1262. filename_real_ext = os.path.splitext(filename)[1][1:]
  1263. filename_wo_ext = (
  1264. os.path.splitext(filename)[0]
  1265. if filename_real_ext == info_dict['ext']
  1266. else filename)
  1267. requested_formats = info_dict['requested_formats']
  1268. if self.params.get('merge_output_format') is None and not compatible_formats(requested_formats):
  1269. info_dict['ext'] = 'mkv'
  1270. self.report_warning(
  1271. 'Requested formats are incompatible for merge and will be merged into mkv.')
  1272. # Ensure filename always has a correct extension for successful merge
  1273. filename = '%s.%s' % (filename_wo_ext, info_dict['ext'])
  1274. if os.path.exists(encodeFilename(filename)):
  1275. self.to_screen(
  1276. '[download] %s has already been downloaded and '
  1277. 'merged' % filename)
  1278. else:
  1279. for f in requested_formats:
  1280. new_info = dict(info_dict)
  1281. new_info.update(f)
  1282. fname = self.prepare_filename(new_info)
  1283. fname = prepend_extension(fname, 'f%s' % f['format_id'], new_info['ext'])
  1284. downloaded.append(fname)
  1285. partial_success = dl(fname, new_info)
  1286. success = success and partial_success
  1287. info_dict['__postprocessors'] = postprocessors
  1288. info_dict['__files_to_merge'] = downloaded
  1289. else:
  1290. # Just a single file
  1291. success = dl(filename, info_dict)
  1292. except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
  1293. self.report_error('unable to download video data: %s' % str(err))
  1294. return
  1295. except (OSError, IOError) as err:
  1296. raise UnavailableVideoError(err)
  1297. except (ContentTooShortError, ) as err:
  1298. self.report_error('content too short (expected %s bytes and served %s)' % (err.expected, err.downloaded))
  1299. return
  1300. if success:
  1301. # Fixup content
  1302. fixup_policy = self.params.get('fixup')
  1303. if fixup_policy is None:
  1304. fixup_policy = 'detect_or_warn'
  1305. stretched_ratio = info_dict.get('stretched_ratio')
  1306. if stretched_ratio is not None and stretched_ratio != 1:
  1307. if fixup_policy == 'warn':
  1308. self.report_warning('%s: Non-uniform pixel ratio (%s)' % (
  1309. info_dict['id'], stretched_ratio))
  1310. elif fixup_policy == 'detect_or_warn':
  1311. stretched_pp = FFmpegFixupStretchedPP(self)
  1312. if stretched_pp.available:
  1313. info_dict.setdefault('__postprocessors', [])
  1314. info_dict['__postprocessors'].append(stretched_pp)
  1315. else:
  1316. self.report_warning(
  1317. '%s: Non-uniform pixel ratio (%s). Install ffmpeg or avconv to fix this automatically.' % (
  1318. info_dict['id'], stretched_ratio))
  1319. else:
  1320. assert fixup_policy in ('ignore', 'never')
  1321. if info_dict.get('requested_formats') is None and info_dict.get('container') == 'm4a_dash':
  1322. if fixup_policy == 'warn':
  1323. self.report_warning('%s: writing DASH m4a. Only some players support this container.' % (
  1324. info_dict['id']))
  1325. elif fixup_policy == 'detect_or_warn':
  1326. fixup_pp = FFmpegFixupM4aPP(self)
  1327. if fixup_pp.available:
  1328. info_dict.setdefault('__postprocessors', [])
  1329. info_dict['__postprocessors'].append(fixup_pp)
  1330. else:
  1331. self.report_warning(
  1332. '%s: writing DASH m4a. Only some players support this container. Install ffmpeg or avconv to fix this automatically.' % (
  1333. info_dict['id']))
  1334. else:
  1335. assert fixup_policy in ('ignore', 'never')
  1336. try:
  1337. self.post_process(filename, info_dict)
  1338. except (PostProcessingError) as err:
  1339. self.report_error('postprocessing: %s' % str(err))
  1340. return
  1341. self.record_download_archive(info_dict)
  1342. def download(self, url_list):
  1343. """Download a given list of URLs."""
  1344. outtmpl = self.params.get('outtmpl', DEFAULT_OUTTMPL)
  1345. if (len(url_list) > 1 and
  1346. '%' not in outtmpl and
  1347. self.params.get('max_downloads') != 1):
  1348. raise SameFileError(outtmpl)
  1349. for url in url_list:
  1350. try:
  1351. # It also downloads the videos
  1352. res = self.extract_info(url)
  1353. except UnavailableVideoError:
  1354. self.report_error('unable to download video')
  1355. except MaxDownloadsReached:
  1356. self.to_screen('[info] Maximum number of downloaded files reached.')
  1357. raise
  1358. else:
  1359. if self.params.get('dump_single_json', False):
  1360. self.to_stdout(json.dumps(res))
  1361. return self._download_retcode
  1362. def download_with_info_file(self, info_filename):
  1363. with contextlib.closing(fileinput.FileInput(
  1364. [info_filename], mode='r',
  1365. openhook=fileinput.hook_encoded('utf-8'))) as f:
  1366. # FileInput doesn't have a read method, we can't call json.load
  1367. info = self.filter_requested_info(json.loads('\n'.join(f)))
  1368. try:
  1369. self.process_ie_result(info, download=True)
  1370. except DownloadError:
  1371. webpage_url = info.get('webpage_url')
  1372. if webpage_url is not None:
  1373. self.report_warning('The info failed to download, trying with "%s"' % webpage_url)
  1374. return self.download([webpage_url])
  1375. else:
  1376. raise
  1377. return self._download_retcode
  1378. @staticmethod
  1379. def filter_requested_info(info_dict):
  1380. return dict(
  1381. (k, v) for k, v in info_dict.items()
  1382. if k not in ['requested_formats', 'requested_subtitles'])
  1383. def post_process(self, filename, ie_info):
  1384. """Run all the postprocessors on the given file."""
  1385. info = dict(ie_info)
  1386. info['filepath'] = filename
  1387. pps_chain = []
  1388. if ie_info.get('__postprocessors') is not None:
  1389. pps_chain.extend(ie_info['__postprocessors'])
  1390. pps_chain.extend(self._pps)
  1391. for pp in pps_chain:
  1392. files_to_delete = []
  1393. try:
  1394. files_to_delete, info = pp.run(info)
  1395. except PostProcessingError as e:
  1396. self.report_error(e.msg)
  1397. if files_to_delete and not self.params.get('keepvideo', False):
  1398. for old_filename in files_to_delete:
  1399. self.to_screen('Deleting original file %s (pass -k to keep)' % old_filename)
  1400. try:
  1401. os.remove(encodeFilename(old_filename))
  1402. except (IOError, OSError):
  1403. self.report_warning('Unable to remove downloaded original file')
  1404. def _make_archive_id(self, info_dict):
  1405. # Future-proof against any change in case
  1406. # and backwards compatibility with prior versions
  1407. extractor = info_dict.get('extractor_key')
  1408. if extractor is None:
  1409. if 'id' in info_dict:
  1410. extractor = info_dict.get('ie_key') # key in a playlist
  1411. if extractor is None:
  1412. return None # Incomplete video information
  1413. return extractor.lower() + ' ' + info_dict['id']
  1414. def in_download_archive(self, info_dict):
  1415. fn = self.params.get('download_archive')
  1416. if fn is None:
  1417. return False
  1418. vid_id = self._make_archive_id(info_dict)
  1419. if vid_id is None:
  1420. return False # Incomplete video information
  1421. try:
  1422. with locked_file(fn, 'r', encoding='utf-8') as archive_file:
  1423. for line in archive_file:
  1424. if line.strip() == vid_id:
  1425. return True
  1426. except IOError as ioe:
  1427. if ioe.errno != errno.ENOENT:
  1428. raise
  1429. return False
  1430. def record_download_archive(self, info_dict):
  1431. fn = self.params.get('download_archive')
  1432. if fn is None:
  1433. return
  1434. vid_id = self._make_archive_id(info_dict)
  1435. assert vid_id
  1436. with locked_file(fn, 'a', encoding='utf-8') as archive_file:
  1437. archive_file.write(vid_id + '\n')
  1438. @staticmethod
  1439. def format_resolution(format, default='unknown'):
  1440. if format.get('vcodec') == 'none':
  1441. return 'audio only'
  1442. if format.get('resolution') is not None:
  1443. return format['resolution']
  1444. if format.get('height') is not None:
  1445. if format.get('width') is not None:
  1446. res = '%sx%s' % (format['width'], format['height'])
  1447. else:
  1448. res = '%sp' % format['height']
  1449. elif format.get('width') is not None:
  1450. res = '?x%d' % format['width']
  1451. else:
  1452. res = default
  1453. return res
  1454. def _format_note(self, fdict):
  1455. res = ''
  1456. if fdict.get('ext') in ['f4f', 'f4m']:
  1457. res += '(unsupported) '
  1458. if fdict.get('format_note') is not None:
  1459. res += fdict['format_note'] + ' '
  1460. if fdict.get('tbr') is not None:
  1461. res += '%4dk ' % fdict['tbr']
  1462. if fdict.get('container') is not None:
  1463. if res:
  1464. res += ', '
  1465. res += '%s container' % fdict['container']
  1466. if (fdict.get('vcodec') is not None and
  1467. fdict.get('vcodec') != 'none'):
  1468. if res:
  1469. res += ', '
  1470. res += fdict['vcodec']
  1471. if fdict.get('vbr') is not None:
  1472. res += '@'
  1473. elif fdict.get('vbr') is not None and fdict.get('abr') is not None:
  1474. res += 'video@'
  1475. if fdict.get('vbr') is not None:
  1476. res += '%4dk' % fdict['vbr']
  1477. if fdict.get('fps') is not None:
  1478. res += ', %sfps' % fdict['fps']
  1479. if fdict.get('acodec') is not None:
  1480. if res:
  1481. res += ', '
  1482. if fdict['acodec'] == 'none':
  1483. res += 'video only'
  1484. else:
  1485. res += '%-5s' % fdict['acodec']
  1486. elif fdict.get('abr') is not None:
  1487. if res:
  1488. res += ', '
  1489. res += 'audio'
  1490. if fdict.get('abr') is not None:
  1491. res += '@%3dk' % fdict['abr']
  1492. if fdict.get('asr') is not None:
  1493. res += ' (%5dHz)' % fdict['asr']
  1494. if fdict.get('filesize') is not None:
  1495. if res:
  1496. res += ', '
  1497. res += format_bytes(fdict['filesize'])
  1498. elif fdict.get('filesize_approx') is not None:
  1499. if res:
  1500. res += ', '
  1501. res += '~' + format_bytes(fdict['filesize_approx'])
  1502. return res
  1503. def list_formats(self, info_dict):
  1504. formats = info_dict.get('formats', [info_dict])
  1505. table = [
  1506. [f['format_id'], f['ext'], self.format_resolution(f), self._format_note(f)]
  1507. for f in formats
  1508. if f.get('preference') is None or f['preference'] >= -1000]
  1509. if len(formats) > 1:
  1510. table[-1][-1] += (' ' if table[-1][-1] else '') + '(best)'
  1511. header_line = ['format code', 'extension', 'resolution', 'note']
  1512. self.to_screen(
  1513. '[info] Available formats for %s:\n%s' %
  1514. (info_dict['id'], render_table(header_line, table)))
  1515. def list_thumbnails(self, info_dict):
  1516. thumbnails = info_dict.get('thumbnails')
  1517. if not thumbnails:
  1518. tn_url = info_dict.get('thumbnail')
  1519. if tn_url:
  1520. thumbnails = [{'id': '0', 'url': tn_url}]
  1521. else:
  1522. self.to_screen(
  1523. '[info] No thumbnails present for %s' % info_dict['id'])
  1524. return
  1525. self.to_screen(
  1526. '[info] Thumbnails for %s:' % info_dict['id'])
  1527. self.to_screen(render_table(
  1528. ['ID', 'width', 'height', 'URL'],
  1529. [[t['id'], t.get('width', 'unknown'), t.get('height', 'unknown'), t['url']] for t in thumbnails]))
  1530. def list_subtitles(self, video_id, subtitles, name='subtitles'):
  1531. if not subtitles:
  1532. self.to_screen('%s has no %s' % (video_id, name))
  1533. return
  1534. self.to_screen(
  1535. 'Available %s for %s:' % (name, video_id))
  1536. self.to_screen(render_table(
  1537. ['Language', 'formats'],
  1538. [[lang, ', '.join(f['ext'] for f in reversed(formats))]
  1539. for lang, formats in subtitles.items()]))
  1540. def urlopen(self, req):
  1541. """ Start an HTTP download """
  1542. # According to RFC 3986, URLs can not contain non-ASCII characters, however this is not
  1543. # always respected by websites, some tend to give out URLs with non percent-encoded
  1544. # non-ASCII characters (see telemb.py, ard.py [#3412])
  1545. # urllib chokes on URLs with non-ASCII characters (see http://bugs.python.org/issue3991)
  1546. # To work around aforementioned issue we will replace request's original URL with
  1547. # percent-encoded one
  1548. req_is_string = isinstance(req, compat_basestring)
  1549. url = req if req_is_string else req.get_full_url()
  1550. url_escaped = escape_url(url)
  1551. # Substitute URL if any change after escaping
  1552. if url != url_escaped:
  1553. if req_is_string:
  1554. req = url_escaped
  1555. else:
  1556. req = compat_urllib_request.Request(
  1557. url_escaped, data=req.data, headers=req.headers,
  1558. origin_req_host=req.origin_req_host, unverifiable=req.unverifiable)
  1559. return self._opener.open(req, timeout=self._socket_timeout)
  1560. def print_debug_header(self):
  1561. if not self.params.get('verbose'):
  1562. return
  1563. if type('') is not compat_str:
  1564. # Python 2.6 on SLES11 SP1 (https://github.com/rg3/youtube-dl/issues/3326)
  1565. self.report_warning(
  1566. 'Your Python is broken! Update to a newer and supported version')
  1567. stdout_encoding = getattr(
  1568. sys.stdout, 'encoding', 'missing (%s)' % type(sys.stdout).__name__)
  1569. encoding_str = (
  1570. '[debug] Encodings: locale %s, fs %s, out %s, pref %s\n' % (
  1571. locale.getpreferredencoding(),
  1572. sys.getfilesystemencoding(),
  1573. stdout_encoding,
  1574. self.get_encoding()))
  1575. write_string(encoding_str, encoding=None)
  1576. self._write_string('[debug] youtube-dl version ' + __version__ + '\n')
  1577. try:
  1578. sp = subprocess.Popen(
  1579. ['git', 'rev-parse', '--short', 'HEAD'],
  1580. stdout=subprocess.PIPE, stderr=subprocess.PIPE,
  1581. cwd=os.path.dirname(os.path.abspath(__file__)))
  1582. out, err = sp.communicate()
  1583. out = out.decode().strip()
  1584. if re.match('[0-9a-f]+', out):
  1585. self._write_string('[debug] Git HEAD: ' + out + '\n')
  1586. except Exception:
  1587. try:
  1588. sys.exc_clear()
  1589. except Exception:
  1590. pass
  1591. self._write_string('[debug] Python version %s - %s\n' % (
  1592. platform.python_version(), platform_name()))
  1593. exe_versions = FFmpegPostProcessor.get_versions(self)
  1594. exe_versions['rtmpdump'] = rtmpdump_version()
  1595. exe_str = ', '.join(
  1596. '%s %s' % (exe, v)
  1597. for exe, v in sorted(exe_versions.items())
  1598. if v
  1599. )
  1600. if not exe_str:
  1601. exe_str = 'none'
  1602. self._write_string('[debug] exe versions: %s\n' % exe_str)
  1603. proxy_map = {}
  1604. for handler in self._opener.handlers:
  1605. if hasattr(handler, 'proxies'):
  1606. proxy_map.update(handler.proxies)
  1607. self._write_string('[debug] Proxy map: ' + compat_str(proxy_map) + '\n')
  1608. if self.params.get('call_home', False):
  1609. ipaddr = self.urlopen('https://yt-dl.org/ip').read().decode('utf-8')
  1610. self._write_string('[debug] Public IP address: %s\n' % ipaddr)
  1611. latest_version = self.urlopen(
  1612. 'https://yt-dl.org/latest/version').read().decode('utf-8')
  1613. if version_tuple(latest_version) > version_tuple(__version__):
  1614. self.report_warning(
  1615. 'You are using an outdated version (newest version: %s)! '
  1616. 'See https://yt-dl.org/update if you need help updating.' %
  1617. latest_version)
  1618. def _setup_opener(self):
  1619. timeout_val = self.params.get('socket_timeout')
  1620. self._socket_timeout = 600 if timeout_val is None else float(timeout_val)
  1621. opts_cookiefile = self.params.get('cookiefile')
  1622. opts_proxy = self.params.get('proxy')
  1623. if opts_cookiefile is None:
  1624. self.cookiejar = compat_cookiejar.CookieJar()
  1625. else:
  1626. self.cookiejar = compat_cookiejar.MozillaCookieJar(
  1627. opts_cookiefile)
  1628. if os.access(opts_cookiefile, os.R_OK):
  1629. self.cookiejar.load()
  1630. cookie_processor = compat_urllib_request.HTTPCookieProcessor(
  1631. self.cookiejar)
  1632. if opts_proxy is not None:
  1633. if opts_proxy == '':
  1634. proxies = {}
  1635. else:
  1636. proxies = {'http': opts_proxy, 'https': opts_proxy}
  1637. else:
  1638. proxies = compat_urllib_request.getproxies()
  1639. # Set HTTPS proxy to HTTP one if given (https://github.com/rg3/youtube-dl/issues/805)
  1640. if 'http' in proxies and 'https' not in proxies:
  1641. proxies['https'] = proxies['http']
  1642. proxy_handler = PerRequestProxyHandler(proxies)
  1643. debuglevel = 1 if self.params.get('debug_printtraffic') else 0
  1644. https_handler = make_HTTPS_handler(self.params, debuglevel=debuglevel)
  1645. ydlh = YoutubeDLHandler(self.params, debuglevel=debuglevel)
  1646. opener = compat_urllib_request.build_opener(
  1647. proxy_handler, https_handler, cookie_processor, ydlh)
  1648. # Delete the default user-agent header, which would otherwise apply in
  1649. # cases where our custom HTTP handler doesn't come into play
  1650. # (See https://github.com/rg3/youtube-dl/issues/1309 for details)
  1651. opener.addheaders = []
  1652. self._opener = opener
  1653. def encode(self, s):
  1654. if isinstance(s, bytes):
  1655. return s # Already encoded
  1656. try:
  1657. return s.encode(self.get_encoding())
  1658. except UnicodeEncodeError as err:
  1659. err.reason = err.reason + '. Check your system encoding configuration or use the --encoding option.'
  1660. raise
  1661. def get_encoding(self):
  1662. encoding = self.params.get('encoding')
  1663. if encoding is None:
  1664. encoding = preferredencoding()
  1665. return encoding
  1666. def _write_thumbnails(self, info_dict, filename):
  1667. if self.params.get('writethumbnail', False):
  1668. thumbnails = info_dict.get('thumbnails')
  1669. if thumbnails:
  1670. thumbnails = [thumbnails[-1]]
  1671. elif self.params.get('write_all_thumbnails', False):
  1672. thumbnails = info_dict.get('thumbnails')
  1673. else:
  1674. return
  1675. if not thumbnails:
  1676. # No thumbnails present, so return immediately
  1677. return
  1678. for t in thumbnails:
  1679. thumb_ext = determine_ext(t['url'], 'jpg')
  1680. suffix = '_%s' % t['id'] if len(thumbnails) > 1 else ''
  1681. thumb_display_id = '%s ' % t['id'] if len(thumbnails) > 1 else ''
  1682. t['filename'] = thumb_filename = os.path.splitext(filename)[0] + suffix + '.' + thumb_ext
  1683. if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(thumb_filename)):
  1684. self.to_screen('[%s] %s: Thumbnail %sis already present' %
  1685. (info_dict['extractor'], info_dict['id'], thumb_display_id))
  1686. else:
  1687. self.to_screen('[%s] %s: Downloading thumbnail %s...' %
  1688. (info_dict['extractor'], info_dict['id'], thumb_display_id))
  1689. try:
  1690. uf = self.urlopen(t['url'])
  1691. with open(thumb_filename, 'wb') as thumbf:
  1692. shutil.copyfileobj(uf, thumbf)
  1693. self.to_screen('[%s] %s: Writing thumbnail %sto: %s' %
  1694. (info_dict['extractor'], info_dict['id'], thumb_display_id, thumb_filename))
  1695. except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
  1696. self.report_warning('Unable to download thumbnail "%s": %s' %
  1697. (t['url'], compat_str(err)))