]>
Commit | Line | Data |
---|---|---|
8222d8de JMF |
1 | #!/usr/bin/env python |
2 | # -*- coding: utf-8 -*- | |
3 | ||
6febd1c1 | 4 | from __future__ import absolute_import, unicode_literals |
8222d8de | 5 | |
26e63931 | 6 | import collections |
9d2ecdbc | 7 | import datetime |
c1c9a79c | 8 | import errno |
8222d8de | 9 | import io |
8694c600 | 10 | import json |
62fec3b2 | 11 | import locale |
8222d8de | 12 | import os |
dca08720 | 13 | import platform |
8222d8de JMF |
14 | import re |
15 | import shutil | |
dca08720 | 16 | import subprocess |
8222d8de JMF |
17 | import socket |
18 | import sys | |
19 | import time | |
20 | import traceback | |
21 | ||
1e5b9a95 PH |
22 | if os.name == 'nt': |
23 | import ctypes | |
24 | ||
ce02ed60 | 25 | from .utils import ( |
dca08720 | 26 | compat_cookiejar, |
ce02ed60 | 27 | compat_http_client, |
ce02ed60 PH |
28 | compat_str, |
29 | compat_urllib_error, | |
30 | compat_urllib_request, | |
d05cfe06 | 31 | escape_url, |
ce02ed60 PH |
32 | ContentTooShortError, |
33 | date_from_str, | |
34 | DateRange, | |
acd69589 | 35 | DEFAULT_OUTTMPL, |
ce02ed60 PH |
36 | determine_ext, |
37 | DownloadError, | |
38 | encodeFilename, | |
39 | ExtractorError, | |
02dbf93f | 40 | format_bytes, |
525ef922 | 41 | formatSeconds, |
1c088fa8 | 42 | get_term_width, |
ce02ed60 | 43 | locked_file, |
dca08720 | 44 | make_HTTPS_handler, |
ce02ed60 | 45 | MaxDownloadsReached, |
b7ab0590 | 46 | PagedList, |
ce02ed60 | 47 | PostProcessingError, |
dca08720 | 48 | platform_name, |
ce02ed60 PH |
49 | preferredencoding, |
50 | SameFileError, | |
51 | sanitize_filename, | |
52 | subtitles_filename, | |
53 | takewhile_inclusive, | |
54 | UnavailableVideoError, | |
29eb5174 | 55 | url_basename, |
ce02ed60 PH |
56 | write_json_file, |
57 | write_string, | |
dca08720 | 58 | YoutubeDLHandler, |
6350728b | 59 | prepend_extension, |
ce02ed60 | 60 | ) |
a0e07d31 | 61 | from .cache import Cache |
023fa8c4 | 62 | from .extractor import get_info_extractor, gen_extractors |
3bc2ddcc | 63 | from .downloader import get_suitable_downloader |
56327689 | 64 | from .postprocessor import FFmpegMergerPP |
dca08720 | 65 | from .version import __version__ |
8222d8de JMF |
66 | |
67 | ||
68 | class YoutubeDL(object): | |
69 | """YoutubeDL class. | |
70 | ||
71 | YoutubeDL objects are the ones responsible of downloading the | |
72 | actual video file and writing it to disk if the user has requested | |
73 | it, among some other tasks. In most cases there should be one per | |
74 | program. As, given a video URL, the downloader doesn't know how to | |
75 | extract all the needed information, task that InfoExtractors do, it | |
76 | has to pass the URL to one of them. | |
77 | ||
78 | For this, YoutubeDL objects have a method that allows | |
79 | InfoExtractors to be registered in a given order. When it is passed | |
80 | a URL, the YoutubeDL object handles it to the first InfoExtractor it | |
81 | finds that reports being able to handle it. The InfoExtractor extracts | |
82 | all the information about the video or videos the URL refers to, and | |
83 | YoutubeDL process the extracted information, possibly using a File | |
84 | Downloader to download the video. | |
85 | ||
86 | YoutubeDL objects accept a lot of parameters. In order not to saturate | |
87 | the object constructor with arguments, it receives a dictionary of | |
88 | options instead. These options are available through the params | |
89 | attribute for the InfoExtractors to use. The YoutubeDL also | |
90 | registers itself as the downloader in charge for the InfoExtractors | |
91 | that are added to it, so this is a "mutual registration". | |
92 | ||
93 | Available options: | |
94 | ||
95 | username: Username for authentication purposes. | |
96 | password: Password for authentication purposes. | |
c6c19746 | 97 | videopassword: Password for acces a video. |
8222d8de JMF |
98 | usenetrc: Use netrc for authentication instead. |
99 | verbose: Print additional info to stdout. | |
100 | quiet: Do not print messages to stdout. | |
ad8915b7 | 101 | no_warnings: Do not print out anything for warnings. |
8222d8de JMF |
102 | forceurl: Force printing final URL. |
103 | forcetitle: Force printing title. | |
104 | forceid: Force printing ID. | |
105 | forcethumbnail: Force printing thumbnail URL. | |
106 | forcedescription: Force printing description. | |
107 | forcefilename: Force printing final filename. | |
525ef922 | 108 | forceduration: Force printing duration. |
8694c600 | 109 | forcejson: Force printing info_dict as JSON. |
8222d8de JMF |
110 | simulate: Do not download the video files. |
111 | format: Video format code. | |
112 | format_limit: Highest quality format to try. | |
113 | outtmpl: Template for output names. | |
114 | restrictfilenames: Do not allow "&" and spaces in file names | |
115 | ignoreerrors: Do not stop on download errors. | |
116 | nooverwrites: Prevent overwriting files. | |
117 | playliststart: Playlist item to start at. | |
118 | playlistend: Playlist item to end at. | |
119 | matchtitle: Download only matching titles. | |
120 | rejecttitle: Reject downloads for matching titles. | |
8bf9319e | 121 | logger: Log messages to a logging.Logger instance. |
8222d8de JMF |
122 | logtostderr: Log messages to stderr instead of stdout. |
123 | writedescription: Write the video description to a .description file | |
124 | writeinfojson: Write the video description to a .info.json file | |
1fb07d10 | 125 | writeannotations: Write the video annotations to a .annotations.xml file |
8222d8de JMF |
126 | writethumbnail: Write the thumbnail image to a file |
127 | writesubtitles: Write the video subtitles to a file | |
b004821f | 128 | writeautomaticsub: Write the automatic subtitles to a file |
8222d8de | 129 | allsubtitles: Downloads all the subtitles of the video |
0b7f3118 | 130 | (requires writesubtitles or writeautomaticsub) |
8222d8de | 131 | listsubtitles: Lists all available subtitles for the video |
b98a6b2f | 132 | subtitlesformat: Subtitle format [srt/sbv/vtt] (default=srt) |
aa6a10c4 | 133 | subtitleslangs: List of languages of the subtitles to download |
8222d8de JMF |
134 | keepvideo: Keep the video file after post-processing |
135 | daterange: A DateRange object, download only if the upload_date is in the range. | |
136 | skip_download: Skip the actual download of the video file | |
c35f9e72 | 137 | cachedir: Location of the cache files in the filesystem. |
a0e07d31 | 138 | False to disable filesystem cache. |
47192f92 | 139 | noplaylist: Download single video instead of a playlist if in doubt. |
8dbe9899 PH |
140 | age_limit: An integer representing the user's age in years. |
141 | Unsuitable videos for the given age are skipped. | |
5fe18bdb PH |
142 | min_views: An integer representing the minimum view count the video |
143 | must have in order to not be skipped. | |
144 | Videos without view count information are always | |
145 | downloaded. None for no limit. | |
146 | max_views: An integer representing the maximum view count. | |
147 | Videos that are more popular than that are not | |
148 | downloaded. | |
149 | Videos without view count information are always | |
150 | downloaded. None for no limit. | |
151 | download_archive: File name of a file where all downloads are recorded. | |
c1c9a79c PH |
152 | Videos already present in the file are not downloaded |
153 | again. | |
dca08720 | 154 | cookiefile: File name where cookies should be read from and dumped to. |
a1ee09e8 | 155 | nocheckcertificate:Do not verify SSL certificates |
7e8c0af0 PH |
156 | prefer_insecure: Use HTTP instead of HTTPS to retrieve information. |
157 | At the moment, this is only supported by YouTube. | |
a1ee09e8 | 158 | proxy: URL of the proxy server to use |
e344693b | 159 | socket_timeout: Time to wait for unresponsive hosts, in seconds |
0783b09b PH |
160 | bidi_workaround: Work around buggy terminals without bidirectional text |
161 | support, using fridibi | |
a0ddb8a2 | 162 | debug_printtraffic:Print out sent and received HTTP traffic |
7b0817e8 | 163 | include_ads: Download ads as well |
04b4d394 PH |
164 | default_search: Prepend this string if an input url is not valid. |
165 | 'auto' for elaborate guessing | |
62fec3b2 | 166 | encoding: Use this encoding instead of the system-specified. |
e8ee972c | 167 | extract_flat: Do not resolve URLs, return the immediate result. |
fe7e0c98 | 168 | |
8222d8de JMF |
169 | The following parameters are not used by YoutubeDL itself, they are used by |
170 | the FileDownloader: | |
171 | nopart, updatetime, buffersize, ratelimit, min_filesize, max_filesize, test, | |
172 | noresizebuffer, retries, continuedl, noprogress, consoletitle | |
76b1bd67 JMF |
173 | |
174 | The following options are used by the post processors: | |
175 | prefer_ffmpeg: If True, use ffmpeg instead of avconv if both are available, | |
176 | otherwise prefer avconv. | |
8d31fa3c | 177 | exec_cmd: Arbitrary command to run after downloading |
8222d8de JMF |
178 | """ |
179 | ||
180 | params = None | |
181 | _ies = [] | |
182 | _pps = [] | |
183 | _download_retcode = None | |
184 | _num_downloads = None | |
185 | _screen_file = None | |
186 | ||
a3fb4675 | 187 | def __init__(self, params=None): |
8222d8de | 188 | """Create a FileDownloader object with the given options.""" |
e9f9a10f JMF |
189 | if params is None: |
190 | params = {} | |
8222d8de | 191 | self._ies = [] |
56c73665 | 192 | self._ies_instances = {} |
8222d8de | 193 | self._pps = [] |
933605d7 | 194 | self._progress_hooks = [] |
8222d8de JMF |
195 | self._download_retcode = 0 |
196 | self._num_downloads = 0 | |
197 | self._screen_file = [sys.stdout, sys.stderr][params.get('logtostderr', False)] | |
0783b09b | 198 | self._err_file = sys.stderr |
e9f9a10f | 199 | self.params = params |
a0e07d31 | 200 | self.cache = Cache(self) |
34308b30 | 201 | |
0783b09b | 202 | if params.get('bidi_workaround', False): |
1c088fa8 PH |
203 | try: |
204 | import pty | |
205 | master, slave = pty.openpty() | |
206 | width = get_term_width() | |
207 | if width is None: | |
208 | width_args = [] | |
209 | else: | |
210 | width_args = ['-w', str(width)] | |
5d681e96 | 211 | sp_kwargs = dict( |
1c088fa8 PH |
212 | stdin=subprocess.PIPE, |
213 | stdout=slave, | |
214 | stderr=self._err_file) | |
5d681e96 PH |
215 | try: |
216 | self._output_process = subprocess.Popen( | |
217 | ['bidiv'] + width_args, **sp_kwargs | |
218 | ) | |
219 | except OSError: | |
5d681e96 PH |
220 | self._output_process = subprocess.Popen( |
221 | ['fribidi', '-c', 'UTF-8'] + width_args, **sp_kwargs) | |
222 | self._output_channel = os.fdopen(master, 'rb') | |
1c088fa8 PH |
223 | except OSError as ose: |
224 | if ose.errno == 2: | |
6febd1c1 | 225 | self.report_warning('Could not find fribidi executable, ignoring --bidi-workaround . Make sure that fribidi is an executable file in one of the directories in your $PATH.') |
1c088fa8 PH |
226 | else: |
227 | raise | |
0783b09b | 228 | |
34308b30 PH |
229 | if (sys.version_info >= (3,) and sys.platform != 'win32' and |
230 | sys.getfilesystemencoding() in ['ascii', 'ANSI_X3.4-1968'] | |
231 | and not params['restrictfilenames']): | |
232 | # On Python 3, the Unicode filesystem API will throw errors (#1474) | |
233 | self.report_warning( | |
6febd1c1 PH |
234 | 'Assuming --restrict-filenames since file system encoding ' |
235 | 'cannot encode all charactes. ' | |
236 | 'Set the LC_ALL environment variable to fix this.') | |
4a98cdbf | 237 | self.params['restrictfilenames'] = True |
34308b30 | 238 | |
a3927cf7 | 239 | if '%(stitle)s' in self.params.get('outtmpl', ''): |
6febd1c1 | 240 | self.report_warning('%(stitle)s is deprecated. Use the %(title)s and the --restrict-filenames flag(which also secures %(uploader)s et al) instead.') |
8222d8de | 241 | |
dca08720 PH |
242 | self._setup_opener() |
243 | ||
8222d8de JMF |
244 | def add_info_extractor(self, ie): |
245 | """Add an InfoExtractor object to the end of the list.""" | |
246 | self._ies.append(ie) | |
56c73665 | 247 | self._ies_instances[ie.ie_key()] = ie |
8222d8de JMF |
248 | ie.set_downloader(self) |
249 | ||
56c73665 JMF |
250 | def get_info_extractor(self, ie_key): |
251 | """ | |
252 | Get an instance of an IE with name ie_key, it will try to get one from | |
253 | the _ies list, if there's no instance it will create a new one and add | |
254 | it to the extractor list. | |
255 | """ | |
256 | ie = self._ies_instances.get(ie_key) | |
257 | if ie is None: | |
258 | ie = get_info_extractor(ie_key)() | |
259 | self.add_info_extractor(ie) | |
260 | return ie | |
261 | ||
023fa8c4 JMF |
262 | def add_default_info_extractors(self): |
263 | """ | |
264 | Add the InfoExtractors returned by gen_extractors to the end of the list | |
265 | """ | |
266 | for ie in gen_extractors(): | |
267 | self.add_info_extractor(ie) | |
268 | ||
8222d8de JMF |
269 | def add_post_processor(self, pp): |
270 | """Add a PostProcessor object to the end of the chain.""" | |
271 | self._pps.append(pp) | |
272 | pp.set_downloader(self) | |
273 | ||
933605d7 JMF |
274 | def add_progress_hook(self, ph): |
275 | """Add the progress hook (currently only for the file downloader)""" | |
276 | self._progress_hooks.append(ph) | |
8ab470f1 | 277 | |
1c088fa8 | 278 | def _bidi_workaround(self, message): |
5d681e96 | 279 | if not hasattr(self, '_output_channel'): |
1c088fa8 PH |
280 | return message |
281 | ||
5d681e96 | 282 | assert hasattr(self, '_output_process') |
11b85ce6 | 283 | assert isinstance(message, compat_str) |
6febd1c1 PH |
284 | line_count = message.count('\n') + 1 |
285 | self._output_process.stdin.write((message + '\n').encode('utf-8')) | |
5d681e96 | 286 | self._output_process.stdin.flush() |
6febd1c1 | 287 | res = ''.join(self._output_channel.readline().decode('utf-8') |
1c088fa8 | 288 | for _ in range(line_count)) |
6febd1c1 | 289 | return res[:-len('\n')] |
1c088fa8 | 290 | |
8222d8de | 291 | def to_screen(self, message, skip_eol=False): |
0783b09b PH |
292 | """Print message to stdout if not in quiet mode.""" |
293 | return self.to_stdout(message, skip_eol, check_quiet=True) | |
294 | ||
734f90bb | 295 | def _write_string(self, s, out=None): |
b58ddb32 | 296 | write_string(s, out=out, encoding=self.params.get('encoding')) |
734f90bb | 297 | |
0783b09b | 298 | def to_stdout(self, message, skip_eol=False, check_quiet=False): |
8222d8de | 299 | """Print message to stdout if not in quiet mode.""" |
8bf9319e | 300 | if self.params.get('logger'): |
43afe285 | 301 | self.params['logger'].debug(message) |
0783b09b | 302 | elif not check_quiet or not self.params.get('quiet', False): |
1c088fa8 | 303 | message = self._bidi_workaround(message) |
6febd1c1 | 304 | terminator = ['\n', ''][skip_eol] |
8222d8de | 305 | output = message + terminator |
1c088fa8 | 306 | |
734f90bb | 307 | self._write_string(output, self._screen_file) |
8222d8de JMF |
308 | |
309 | def to_stderr(self, message): | |
310 | """Print message to stderr.""" | |
11b85ce6 | 311 | assert isinstance(message, compat_str) |
8bf9319e | 312 | if self.params.get('logger'): |
43afe285 IB |
313 | self.params['logger'].error(message) |
314 | else: | |
1c088fa8 | 315 | message = self._bidi_workaround(message) |
6febd1c1 | 316 | output = message + '\n' |
734f90bb | 317 | self._write_string(output, self._err_file) |
8222d8de | 318 | |
1e5b9a95 PH |
319 | def to_console_title(self, message): |
320 | if not self.params.get('consoletitle', False): | |
321 | return | |
322 | if os.name == 'nt' and ctypes.windll.kernel32.GetConsoleWindow(): | |
323 | # c_wchar_p() might not be necessary if `message` is | |
324 | # already of type unicode() | |
325 | ctypes.windll.kernel32.SetConsoleTitleW(ctypes.c_wchar_p(message)) | |
326 | elif 'TERM' in os.environ: | |
734f90bb | 327 | self._write_string('\033]0;%s\007' % message, self._screen_file) |
1e5b9a95 | 328 | |
bdde425c PH |
329 | def save_console_title(self): |
330 | if not self.params.get('consoletitle', False): | |
331 | return | |
332 | if 'TERM' in os.environ: | |
efd6c574 | 333 | # Save the title on stack |
734f90bb | 334 | self._write_string('\033[22;0t', self._screen_file) |
bdde425c PH |
335 | |
336 | def restore_console_title(self): | |
337 | if not self.params.get('consoletitle', False): | |
338 | return | |
339 | if 'TERM' in os.environ: | |
efd6c574 | 340 | # Restore the title from stack |
734f90bb | 341 | self._write_string('\033[23;0t', self._screen_file) |
bdde425c PH |
342 | |
343 | def __enter__(self): | |
344 | self.save_console_title() | |
345 | return self | |
346 | ||
347 | def __exit__(self, *args): | |
348 | self.restore_console_title() | |
f89197d7 | 349 | |
dca08720 PH |
350 | if self.params.get('cookiefile') is not None: |
351 | self.cookiejar.save() | |
bdde425c | 352 | |
8222d8de JMF |
353 | def trouble(self, message=None, tb=None): |
354 | """Determine action to take when a download problem appears. | |
355 | ||
356 | Depending on if the downloader has been configured to ignore | |
357 | download errors or not, this method may throw an exception or | |
358 | not when errors are found, after printing the message. | |
359 | ||
360 | tb, if given, is additional traceback information. | |
361 | """ | |
362 | if message is not None: | |
363 | self.to_stderr(message) | |
364 | if self.params.get('verbose'): | |
365 | if tb is None: | |
366 | if sys.exc_info()[0]: # if .trouble has been called from an except block | |
6febd1c1 | 367 | tb = '' |
8222d8de | 368 | if hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]: |
6febd1c1 | 369 | tb += ''.join(traceback.format_exception(*sys.exc_info()[1].exc_info)) |
8222d8de JMF |
370 | tb += compat_str(traceback.format_exc()) |
371 | else: | |
372 | tb_data = traceback.format_list(traceback.extract_stack()) | |
6febd1c1 | 373 | tb = ''.join(tb_data) |
8222d8de JMF |
374 | self.to_stderr(tb) |
375 | if not self.params.get('ignoreerrors', False): | |
376 | if sys.exc_info()[0] and hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]: | |
377 | exc_info = sys.exc_info()[1].exc_info | |
378 | else: | |
379 | exc_info = sys.exc_info() | |
380 | raise DownloadError(message, exc_info) | |
381 | self._download_retcode = 1 | |
382 | ||
383 | def report_warning(self, message): | |
384 | ''' | |
385 | Print the message to stderr, it will be prefixed with 'WARNING:' | |
386 | If stderr is a tty file the 'WARNING:' will be colored | |
387 | ''' | |
6d07ce01 JMF |
388 | if self.params.get('logger') is not None: |
389 | self.params['logger'].warning(message) | |
8222d8de | 390 | else: |
ad8915b7 PH |
391 | if self.params.get('no_warnings'): |
392 | return | |
6d07ce01 JMF |
393 | if self._err_file.isatty() and os.name != 'nt': |
394 | _msg_header = '\033[0;33mWARNING:\033[0m' | |
395 | else: | |
396 | _msg_header = 'WARNING:' | |
397 | warning_message = '%s %s' % (_msg_header, message) | |
398 | self.to_stderr(warning_message) | |
8222d8de JMF |
399 | |
400 | def report_error(self, message, tb=None): | |
401 | ''' | |
402 | Do the same as trouble, but prefixes the message with 'ERROR:', colored | |
403 | in red if stderr is a tty file. | |
404 | ''' | |
0783b09b | 405 | if self._err_file.isatty() and os.name != 'nt': |
6febd1c1 | 406 | _msg_header = '\033[0;31mERROR:\033[0m' |
8222d8de | 407 | else: |
6febd1c1 PH |
408 | _msg_header = 'ERROR:' |
409 | error_message = '%s %s' % (_msg_header, message) | |
8222d8de JMF |
410 | self.trouble(error_message, tb) |
411 | ||
8222d8de JMF |
412 | def report_file_already_downloaded(self, file_name): |
413 | """Report file has already been fully downloaded.""" | |
414 | try: | |
6febd1c1 | 415 | self.to_screen('[download] %s has already been downloaded' % file_name) |
ce02ed60 | 416 | except UnicodeEncodeError: |
6febd1c1 | 417 | self.to_screen('[download] The file has already been downloaded') |
8222d8de | 418 | |
8222d8de JMF |
419 | def prepare_filename(self, info_dict): |
420 | """Generate the output filename.""" | |
421 | try: | |
422 | template_dict = dict(info_dict) | |
423 | ||
424 | template_dict['epoch'] = int(time.time()) | |
425 | autonumber_size = self.params.get('autonumber_size') | |
426 | if autonumber_size is None: | |
427 | autonumber_size = 5 | |
6febd1c1 | 428 | autonumber_templ = '%0' + str(autonumber_size) + 'd' |
8222d8de | 429 | template_dict['autonumber'] = autonumber_templ % self._num_downloads |
702665c0 | 430 | if template_dict.get('playlist_index') is not None: |
c6b4132a | 431 | template_dict['playlist_index'] = '%0*d' % (len(str(template_dict['n_entries'])), template_dict['playlist_index']) |
17b75c0d PH |
432 | if template_dict.get('resolution') is None: |
433 | if template_dict.get('width') and template_dict.get('height'): | |
434 | template_dict['resolution'] = '%dx%d' % (template_dict['width'], template_dict['height']) | |
435 | elif template_dict.get('height'): | |
805ef3c6 | 436 | template_dict['resolution'] = '%sp' % template_dict['height'] |
17b75c0d | 437 | elif template_dict.get('width'): |
805ef3c6 | 438 | template_dict['resolution'] = '?x%d' % template_dict['width'] |
8222d8de | 439 | |
586a91b6 | 440 | sanitize = lambda k, v: sanitize_filename( |
45598aab | 441 | compat_str(v), |
8222d8de | 442 | restricted=self.params.get('restrictfilenames'), |
6febd1c1 | 443 | is_id=(k == 'id')) |
586a91b6 | 444 | template_dict = dict((k, sanitize(k, v)) |
45598aab PH |
445 | for k, v in template_dict.items() |
446 | if v is not None) | |
6febd1c1 | 447 | template_dict = collections.defaultdict(lambda: 'NA', template_dict) |
8222d8de | 448 | |
acd69589 PH |
449 | outtmpl = self.params.get('outtmpl', DEFAULT_OUTTMPL) |
450 | tmpl = os.path.expanduser(outtmpl) | |
586a91b6 | 451 | filename = tmpl % template_dict |
8222d8de | 452 | return filename |
8222d8de | 453 | except ValueError as err: |
6febd1c1 | 454 | self.report_error('Error in output template: ' + str(err) + ' (encoding: ' + repr(preferredencoding()) + ')') |
8222d8de JMF |
455 | return None |
456 | ||
457 | def _match_entry(self, info_dict): | |
458 | """ Returns None iff the file should be downloaded """ | |
459 | ||
6febd1c1 | 460 | video_title = info_dict.get('title', info_dict.get('id', 'video')) |
7012b23c PH |
461 | if 'title' in info_dict: |
462 | # This can happen when we're just evaluating the playlist | |
463 | title = info_dict['title'] | |
464 | matchtitle = self.params.get('matchtitle', False) | |
465 | if matchtitle: | |
466 | if not re.search(matchtitle, title, re.IGNORECASE): | |
6febd1c1 | 467 | return '"' + title + '" title did not match pattern "' + matchtitle + '"' |
7012b23c PH |
468 | rejecttitle = self.params.get('rejecttitle', False) |
469 | if rejecttitle: | |
470 | if re.search(rejecttitle, title, re.IGNORECASE): | |
6febd1c1 | 471 | return '"' + title + '" title matched reject pattern "' + rejecttitle + '"' |
8222d8de JMF |
472 | date = info_dict.get('upload_date', None) |
473 | if date is not None: | |
474 | dateRange = self.params.get('daterange', DateRange()) | |
475 | if date not in dateRange: | |
6febd1c1 | 476 | return '%s upload date is not in range %s' % (date_from_str(date).isoformat(), dateRange) |
5fe18bdb PH |
477 | view_count = info_dict.get('view_count', None) |
478 | if view_count is not None: | |
479 | min_views = self.params.get('min_views') | |
480 | if min_views is not None and view_count < min_views: | |
6febd1c1 | 481 | return 'Skipping %s, because it has not reached minimum view count (%d/%d)' % (video_title, view_count, min_views) |
5fe18bdb PH |
482 | max_views = self.params.get('max_views') |
483 | if max_views is not None and view_count > max_views: | |
6febd1c1 | 484 | return 'Skipping %s, because it has exceeded the maximum view count (%d/%d)' % (video_title, view_count, max_views) |
8dbe9899 PH |
485 | age_limit = self.params.get('age_limit') |
486 | if age_limit is not None: | |
be843678 PH |
487 | actual_age_limit = info_dict.get('age_limit') |
488 | if actual_age_limit is None: | |
489 | actual_age_limit = 0 | |
490 | if age_limit < actual_age_limit: | |
6febd1c1 | 491 | return 'Skipping "' + title + '" because it is age restricted' |
c1c9a79c | 492 | if self.in_download_archive(info_dict): |
6febd1c1 | 493 | return '%s has already been recorded in archive' % video_title |
8222d8de | 494 | return None |
fe7e0c98 | 495 | |
b6c45014 JMF |
496 | @staticmethod |
497 | def add_extra_info(info_dict, extra_info): | |
498 | '''Set the keys from extra_info in info dict if they are missing''' | |
499 | for key, value in extra_info.items(): | |
500 | info_dict.setdefault(key, value) | |
501 | ||
7fc3fa05 PH |
502 | def extract_info(self, url, download=True, ie_key=None, extra_info={}, |
503 | process=True): | |
8222d8de JMF |
504 | ''' |
505 | Returns a list with a dictionary for each video we find. | |
506 | If 'download', also downloads the videos. | |
507 | extra_info is a dict containing the extra values to add to each result | |
508 | ''' | |
fe7e0c98 | 509 | |
8222d8de | 510 | if ie_key: |
56c73665 | 511 | ies = [self.get_info_extractor(ie_key)] |
8222d8de JMF |
512 | else: |
513 | ies = self._ies | |
514 | ||
515 | for ie in ies: | |
516 | if not ie.suitable(url): | |
517 | continue | |
518 | ||
519 | if not ie.working(): | |
6febd1c1 PH |
520 | self.report_warning('The program functionality for this site has been marked as broken, ' |
521 | 'and will probably not work.') | |
8222d8de JMF |
522 | |
523 | try: | |
524 | ie_result = ie.extract(url) | |
525 | if ie_result is None: # Finished already (backwards compatibility; listformats and friends should be moved here) | |
526 | break | |
527 | if isinstance(ie_result, list): | |
528 | # Backwards compatibility: old IE result format | |
8222d8de JMF |
529 | ie_result = { |
530 | '_type': 'compat_list', | |
531 | 'entries': ie_result, | |
532 | } | |
ea38e55f | 533 | self.add_default_extra_info(ie_result, ie, url) |
7fc3fa05 PH |
534 | if process: |
535 | return self.process_ie_result(ie_result, download, extra_info) | |
536 | else: | |
537 | return ie_result | |
8222d8de JMF |
538 | except ExtractorError as de: # An error we somewhat expected |
539 | self.report_error(compat_str(de), de.format_traceback()) | |
540 | break | |
d3e5bbf4 PH |
541 | except MaxDownloadsReached: |
542 | raise | |
8222d8de JMF |
543 | except Exception as e: |
544 | if self.params.get('ignoreerrors', False): | |
545 | self.report_error(compat_str(e), tb=compat_str(traceback.format_exc())) | |
546 | break | |
547 | else: | |
548 | raise | |
549 | else: | |
1a489545 | 550 | self.report_error('no suitable InfoExtractor for URL %s' % url) |
fe7e0c98 | 551 | |
ea38e55f PH |
552 | def add_default_extra_info(self, ie_result, ie, url): |
553 | self.add_extra_info(ie_result, { | |
554 | 'extractor': ie.IE_NAME, | |
555 | 'webpage_url': url, | |
556 | 'webpage_url_basename': url_basename(url), | |
557 | 'extractor_key': ie.ie_key(), | |
558 | }) | |
559 | ||
8222d8de JMF |
560 | def process_ie_result(self, ie_result, download=True, extra_info={}): |
561 | """ | |
562 | Take the result of the ie(may be modified) and resolve all unresolved | |
563 | references (URLs, playlist items). | |
564 | ||
565 | It will also download the videos if 'download'. | |
566 | Returns the resolved ie_result. | |
567 | """ | |
568 | ||
e8ee972c PH |
569 | result_type = ie_result.get('_type', 'video') |
570 | ||
571 | if self.params.get('extract_flat', False): | |
572 | if result_type in ('url', 'url_transparent'): | |
573 | return ie_result | |
574 | ||
8222d8de | 575 | if result_type == 'video': |
b6c45014 | 576 | self.add_extra_info(ie_result, extra_info) |
feee2ecf | 577 | return self.process_video_result(ie_result, download=download) |
8222d8de JMF |
578 | elif result_type == 'url': |
579 | # We have to add extra_info to the results because it may be | |
580 | # contained in a playlist | |
581 | return self.extract_info(ie_result['url'], | |
582 | download, | |
583 | ie_key=ie_result.get('ie_key'), | |
584 | extra_info=extra_info) | |
7fc3fa05 PH |
585 | elif result_type == 'url_transparent': |
586 | # Use the information from the embedding page | |
587 | info = self.extract_info( | |
588 | ie_result['url'], ie_key=ie_result.get('ie_key'), | |
589 | extra_info=extra_info, download=False, process=False) | |
590 | ||
591 | def make_result(embedded_info): | |
592 | new_result = ie_result.copy() | |
593 | for f in ('_type', 'url', 'ext', 'player_url', 'formats', | |
1538eff6 | 594 | 'entries', 'ie_key', 'duration', |
ef4fd848 PH |
595 | 'subtitles', 'annotations', 'format', |
596 | 'thumbnail', 'thumbnails'): | |
7fc3fa05 PH |
597 | if f in new_result: |
598 | del new_result[f] | |
599 | if f in embedded_info: | |
600 | new_result[f] = embedded_info[f] | |
601 | return new_result | |
602 | new_result = make_result(info) | |
603 | ||
604 | assert new_result.get('_type') != 'url_transparent' | |
605 | if new_result.get('_type') == 'compat_list': | |
606 | new_result['entries'] = [ | |
607 | make_result(e) for e in new_result['entries']] | |
608 | ||
609 | return self.process_ie_result( | |
610 | new_result, download=download, extra_info=extra_info) | |
8222d8de JMF |
611 | elif result_type == 'playlist': |
612 | # We process each entry in the playlist | |
613 | playlist = ie_result.get('title', None) or ie_result.get('id', None) | |
6febd1c1 | 614 | self.to_screen('[download] Downloading playlist: %s' % playlist) |
8222d8de JMF |
615 | |
616 | playlist_results = [] | |
617 | ||
8222d8de | 618 | playliststart = self.params.get('playliststart', 1) - 1 |
a19fd00c PH |
619 | playlistend = self.params.get('playlistend', None) |
620 | # For backwards compatibility, interpret -1 as whole list | |
8222d8de | 621 | if playlistend == -1: |
a19fd00c | 622 | playlistend = None |
8222d8de | 623 | |
b7ab0590 PH |
624 | if isinstance(ie_result['entries'], list): |
625 | n_all_entries = len(ie_result['entries']) | |
626 | entries = ie_result['entries'][playliststart:playlistend] | |
627 | n_entries = len(entries) | |
628 | self.to_screen( | |
629 | "[%s] playlist %s: Collected %d video ids (downloading %d of them)" % | |
630 | (ie_result['extractor'], playlist, n_all_entries, n_entries)) | |
631 | else: | |
632 | assert isinstance(ie_result['entries'], PagedList) | |
633 | entries = ie_result['entries'].getslice( | |
634 | playliststart, playlistend) | |
635 | n_entries = len(entries) | |
636 | self.to_screen( | |
637 | "[%s] playlist %s: Downloading %d videos" % | |
638 | (ie_result['extractor'], playlist, n_entries)) | |
8222d8de | 639 | |
fe7e0c98 | 640 | for i, entry in enumerate(entries, 1): |
6febd1c1 | 641 | self.to_screen('[download] Downloading video #%s of %s' % (i, n_entries)) |
8222d8de | 642 | extra = { |
c6b4132a | 643 | 'n_entries': n_entries, |
fe7e0c98 JMF |
644 | 'playlist': playlist, |
645 | 'playlist_index': i + playliststart, | |
b6c45014 | 646 | 'extractor': ie_result['extractor'], |
9103bbc5 | 647 | 'webpage_url': ie_result['webpage_url'], |
29eb5174 | 648 | 'webpage_url_basename': url_basename(ie_result['webpage_url']), |
be97abc2 | 649 | 'extractor_key': ie_result['extractor_key'], |
fe7e0c98 | 650 | } |
7012b23c PH |
651 | |
652 | reason = self._match_entry(entry) | |
653 | if reason is not None: | |
6febd1c1 | 654 | self.to_screen('[download] ' + reason) |
7012b23c PH |
655 | continue |
656 | ||
8222d8de JMF |
657 | entry_result = self.process_ie_result(entry, |
658 | download=download, | |
659 | extra_info=extra) | |
660 | playlist_results.append(entry_result) | |
661 | ie_result['entries'] = playlist_results | |
662 | return ie_result | |
663 | elif result_type == 'compat_list': | |
664 | def _fixup(r): | |
b6c45014 | 665 | self.add_extra_info(r, |
9103bbc5 JMF |
666 | { |
667 | 'extractor': ie_result['extractor'], | |
668 | 'webpage_url': ie_result['webpage_url'], | |
29eb5174 | 669 | 'webpage_url_basename': url_basename(ie_result['webpage_url']), |
be97abc2 | 670 | 'extractor_key': ie_result['extractor_key'], |
9103bbc5 | 671 | }) |
8222d8de JMF |
672 | return r |
673 | ie_result['entries'] = [ | |
b6c45014 | 674 | self.process_ie_result(_fixup(r), download, extra_info) |
8222d8de JMF |
675 | for r in ie_result['entries'] |
676 | ] | |
677 | return ie_result | |
678 | else: | |
679 | raise Exception('Invalid result type: %s' % result_type) | |
680 | ||
a9c58ad9 JMF |
681 | def select_format(self, format_spec, available_formats): |
682 | if format_spec == 'best' or format_spec is None: | |
683 | return available_formats[-1] | |
684 | elif format_spec == 'worst': | |
685 | return available_formats[0] | |
ba7678f9 PH |
686 | elif format_spec == 'bestaudio': |
687 | audio_formats = [ | |
688 | f for f in available_formats | |
689 | if f.get('vcodec') == 'none'] | |
690 | if audio_formats: | |
691 | return audio_formats[-1] | |
692 | elif format_spec == 'worstaudio': | |
693 | audio_formats = [ | |
694 | f for f in available_formats | |
695 | if f.get('vcodec') == 'none'] | |
696 | if audio_formats: | |
697 | return audio_formats[0] | |
bc6d5978 JMF |
698 | elif format_spec == 'bestvideo': |
699 | video_formats = [ | |
700 | f for f in available_formats | |
701 | if f.get('acodec') == 'none'] | |
702 | if video_formats: | |
703 | return video_formats[-1] | |
704 | elif format_spec == 'worstvideo': | |
705 | video_formats = [ | |
706 | f for f in available_formats | |
707 | if f.get('acodec') == 'none'] | |
708 | if video_formats: | |
709 | return video_formats[0] | |
a9c58ad9 | 710 | else: |
6febd1c1 | 711 | extensions = ['mp4', 'flv', 'webm', '3gp'] |
49e86983 JMF |
712 | if format_spec in extensions: |
713 | filter_f = lambda f: f['ext'] == format_spec | |
714 | else: | |
715 | filter_f = lambda f: f['format_id'] == format_spec | |
fe7e0c98 | 716 | matches = list(filter(filter_f, available_formats)) |
a9c58ad9 JMF |
717 | if matches: |
718 | return matches[-1] | |
719 | return None | |
720 | ||
dd82ffea JMF |
721 | def process_video_result(self, info_dict, download=True): |
722 | assert info_dict.get('_type', 'video') == 'video' | |
723 | ||
bec1fad2 PH |
724 | if 'id' not in info_dict: |
725 | raise ExtractorError('Missing "id" field in extractor result') | |
726 | if 'title' not in info_dict: | |
727 | raise ExtractorError('Missing "title" field in extractor result') | |
728 | ||
dd82ffea JMF |
729 | if 'playlist' not in info_dict: |
730 | # It isn't part of a playlist | |
731 | info_dict['playlist'] = None | |
732 | info_dict['playlist_index'] = None | |
733 | ||
d5519808 PH |
734 | thumbnails = info_dict.get('thumbnails') |
735 | if thumbnails: | |
be6d7229 PH |
736 | thumbnails.sort(key=lambda t: ( |
737 | t.get('width'), t.get('height'), t.get('url'))) | |
d5519808 PH |
738 | for t in thumbnails: |
739 | if 'width' in t and 'height' in t: | |
740 | t['resolution'] = '%dx%d' % (t['width'], t['height']) | |
741 | ||
742 | if thumbnails and 'thumbnail' not in info_dict: | |
743 | info_dict['thumbnail'] = thumbnails[-1]['url'] | |
744 | ||
c9ae7b95 | 745 | if 'display_id' not in info_dict and 'id' in info_dict: |
0afef30b PH |
746 | info_dict['display_id'] = info_dict['id'] |
747 | ||
955c4514 | 748 | if info_dict.get('upload_date') is None and info_dict.get('timestamp') is not None: |
9d2ecdbc | 749 | upload_date = datetime.datetime.utcfromtimestamp( |
955c4514 | 750 | info_dict['timestamp']) |
9d2ecdbc PH |
751 | info_dict['upload_date'] = upload_date.strftime('%Y%m%d') |
752 | ||
6ff000b8 | 753 | # This extractors handle format selection themselves |
6febd1c1 | 754 | if info_dict['extractor'] in ['Youku']: |
12893efe JMF |
755 | if download: |
756 | self.process_info(info_dict) | |
6ff000b8 JMF |
757 | return info_dict |
758 | ||
dd82ffea JMF |
759 | # We now pick which formats have to be downloaded |
760 | if info_dict.get('formats') is None: | |
761 | # There's only one format available | |
762 | formats = [info_dict] | |
763 | else: | |
764 | formats = info_dict['formats'] | |
765 | ||
db95dc13 PH |
766 | if not formats: |
767 | raise ExtractorError('No video formats found!') | |
768 | ||
dd82ffea | 769 | # We check that all the formats have the format and format_id fields |
db95dc13 | 770 | for i, format in enumerate(formats): |
bec1fad2 PH |
771 | if 'url' not in format: |
772 | raise ExtractorError('Missing "url" key in result (index %d)' % i) | |
773 | ||
dd82ffea | 774 | if format.get('format_id') is None: |
8016c922 | 775 | format['format_id'] = compat_str(i) |
8c51aa65 | 776 | if format.get('format') is None: |
6febd1c1 | 777 | format['format'] = '{id} - {res}{note}'.format( |
8c51aa65 JMF |
778 | id=format['format_id'], |
779 | res=self.format_resolution(format), | |
6febd1c1 | 780 | note=' ({0})'.format(format['format_note']) if format.get('format_note') is not None else '', |
8c51aa65 | 781 | ) |
c1002e96 PH |
782 | # Automatically determine file extension if missing |
783 | if 'ext' not in format: | |
cce929ea | 784 | format['ext'] = determine_ext(format['url']).lower() |
dd82ffea | 785 | |
99e206d5 JMF |
786 | format_limit = self.params.get('format_limit', None) |
787 | if format_limit: | |
f4d96df0 PH |
788 | formats = list(takewhile_inclusive( |
789 | lambda f: f['format_id'] != format_limit, formats | |
790 | )) | |
4bcc7bd1 PH |
791 | |
792 | # TODO Central sorting goes here | |
99e206d5 | 793 | |
f89197d7 | 794 | if formats[0] is not info_dict: |
b3d9ef88 JMF |
795 | # only set the 'formats' fields if the original info_dict list them |
796 | # otherwise we end up with a circular reference, the first (and unique) | |
f89197d7 | 797 | # element in the 'formats' field in info_dict is info_dict itself, |
b3d9ef88 JMF |
798 | # wich can't be exported to json |
799 | info_dict['formats'] = formats | |
bfaae0a7 | 800 | if self.params.get('listformats', None): |
801 | self.list_formats(info_dict) | |
802 | return | |
803 | ||
de3ef3ed | 804 | req_format = self.params.get('format') |
a9c58ad9 JMF |
805 | if req_format is None: |
806 | req_format = 'best' | |
dd82ffea | 807 | formats_to_download = [] |
dd82ffea | 808 | # The -1 is for supporting YoutubeIE |
a9c58ad9 | 809 | if req_format in ('-1', 'all'): |
dd82ffea JMF |
810 | formats_to_download = formats |
811 | else: | |
a9c5e5ca | 812 | # We can accept formats requested in the format: 34/5/best, we pick |
416a5efc | 813 | # the first that is available, starting from left |
dd82ffea JMF |
814 | req_formats = req_format.split('/') |
815 | for rf in req_formats: | |
6350728b JMF |
816 | if re.match(r'.+?\+.+?', rf) is not None: |
817 | # Two formats have been requested like '137+139' | |
818 | format_1, format_2 = rf.split('+') | |
819 | formats_info = (self.select_format(format_1, formats), | |
820 | self.select_format(format_2, formats)) | |
821 | if all(formats_info): | |
a9c5e5ca PH |
822 | selected_format = { |
823 | 'requested_formats': formats_info, | |
824 | 'format': rf, | |
825 | 'ext': formats_info[0]['ext'], | |
826 | } | |
6350728b JMF |
827 | else: |
828 | selected_format = None | |
829 | else: | |
830 | selected_format = self.select_format(rf, formats) | |
a9c58ad9 JMF |
831 | if selected_format is not None: |
832 | formats_to_download = [selected_format] | |
dd82ffea JMF |
833 | break |
834 | if not formats_to_download: | |
6febd1c1 | 835 | raise ExtractorError('requested format not available', |
78a3a9f8 | 836 | expected=True) |
dd82ffea JMF |
837 | |
838 | if download: | |
839 | if len(formats_to_download) > 1: | |
6febd1c1 | 840 | self.to_screen('[info] %s: downloading video in %s formats' % (info_dict['id'], len(formats_to_download))) |
dd82ffea JMF |
841 | for format in formats_to_download: |
842 | new_info = dict(info_dict) | |
843 | new_info.update(format) | |
844 | self.process_info(new_info) | |
845 | # We update the info dict with the best quality format (backwards compatibility) | |
846 | info_dict.update(formats_to_download[-1]) | |
847 | return info_dict | |
848 | ||
8222d8de JMF |
849 | def process_info(self, info_dict): |
850 | """Process a single resolved IE result.""" | |
851 | ||
852 | assert info_dict.get('_type', 'video') == 'video' | |
fd288278 PH |
853 | |
854 | max_downloads = self.params.get('max_downloads') | |
855 | if max_downloads is not None: | |
856 | if self._num_downloads >= int(max_downloads): | |
857 | raise MaxDownloadsReached() | |
8222d8de JMF |
858 | |
859 | info_dict['fulltitle'] = info_dict['title'] | |
860 | if len(info_dict['title']) > 200: | |
6febd1c1 | 861 | info_dict['title'] = info_dict['title'][:197] + '...' |
8222d8de JMF |
862 | |
863 | # Keep for backwards compatibility | |
864 | info_dict['stitle'] = info_dict['title'] | |
865 | ||
11b85ce6 | 866 | if 'format' not in info_dict: |
8222d8de JMF |
867 | info_dict['format'] = info_dict['ext'] |
868 | ||
869 | reason = self._match_entry(info_dict) | |
870 | if reason is not None: | |
6febd1c1 | 871 | self.to_screen('[download] ' + reason) |
8222d8de JMF |
872 | return |
873 | ||
fd288278 | 874 | self._num_downloads += 1 |
8222d8de JMF |
875 | |
876 | filename = self.prepare_filename(info_dict) | |
877 | ||
878 | # Forced printings | |
879 | if self.params.get('forcetitle', False): | |
0783b09b | 880 | self.to_stdout(info_dict['fulltitle']) |
8222d8de | 881 | if self.params.get('forceid', False): |
0783b09b | 882 | self.to_stdout(info_dict['id']) |
8222d8de | 883 | if self.params.get('forceurl', False): |
edde6c56 | 884 | # For RTMP URLs, also include the playpath |
6febd1c1 | 885 | self.to_stdout(info_dict['url'] + info_dict.get('play_path', '')) |
216d71d0 | 886 | if self.params.get('forcethumbnail', False) and info_dict.get('thumbnail') is not None: |
0783b09b | 887 | self.to_stdout(info_dict['thumbnail']) |
216d71d0 | 888 | if self.params.get('forcedescription', False) and info_dict.get('description') is not None: |
0783b09b | 889 | self.to_stdout(info_dict['description']) |
8222d8de | 890 | if self.params.get('forcefilename', False) and filename is not None: |
0783b09b | 891 | self.to_stdout(filename) |
525ef922 PH |
892 | if self.params.get('forceduration', False) and info_dict.get('duration') is not None: |
893 | self.to_stdout(formatSeconds(info_dict['duration'])) | |
8222d8de | 894 | if self.params.get('forceformat', False): |
0783b09b | 895 | self.to_stdout(info_dict['format']) |
9d153818 | 896 | if self.params.get('forcejson', False): |
a0d96c98 | 897 | info_dict['_filename'] = filename |
0783b09b | 898 | self.to_stdout(json.dumps(info_dict)) |
8222d8de JMF |
899 | |
900 | # Do nothing else if in simulate mode | |
901 | if self.params.get('simulate', False): | |
902 | return | |
903 | ||
904 | if filename is None: | |
905 | return | |
906 | ||
907 | try: | |
908 | dn = os.path.dirname(encodeFilename(filename)) | |
d26e981d | 909 | if dn and not os.path.exists(dn): |
8222d8de JMF |
910 | os.makedirs(dn) |
911 | except (OSError, IOError) as err: | |
6febd1c1 | 912 | self.report_error('unable to create directory ' + compat_str(err)) |
8222d8de JMF |
913 | return |
914 | ||
915 | if self.params.get('writedescription', False): | |
6febd1c1 | 916 | descfn = filename + '.description' |
7b6fefc9 | 917 | if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(descfn)): |
6febd1c1 | 918 | self.to_screen('[info] Video description is already present') |
7b6fefc9 PH |
919 | else: |
920 | try: | |
6febd1c1 | 921 | self.to_screen('[info] Writing video description to: ' + descfn) |
7b6fefc9 PH |
922 | with io.open(encodeFilename(descfn), 'w', encoding='utf-8') as descfile: |
923 | descfile.write(info_dict['description']) | |
924 | except (KeyError, TypeError): | |
6febd1c1 | 925 | self.report_warning('There\'s no description to write.') |
7b6fefc9 | 926 | except (OSError, IOError): |
6febd1c1 | 927 | self.report_error('Cannot write description file ' + descfn) |
7b6fefc9 | 928 | return |
8222d8de | 929 | |
1fb07d10 | 930 | if self.params.get('writeannotations', False): |
6febd1c1 | 931 | annofn = filename + '.annotations.xml' |
7b6fefc9 | 932 | if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(annofn)): |
6febd1c1 | 933 | self.to_screen('[info] Video annotations are already present') |
7b6fefc9 PH |
934 | else: |
935 | try: | |
6febd1c1 | 936 | self.to_screen('[info] Writing video annotations to: ' + annofn) |
7b6fefc9 PH |
937 | with io.open(encodeFilename(annofn), 'w', encoding='utf-8') as annofile: |
938 | annofile.write(info_dict['annotations']) | |
939 | except (KeyError, TypeError): | |
6febd1c1 | 940 | self.report_warning('There are no annotations to write.') |
7b6fefc9 | 941 | except (OSError, IOError): |
6febd1c1 | 942 | self.report_error('Cannot write annotations file: ' + annofn) |
7b6fefc9 | 943 | return |
1fb07d10 | 944 | |
c4a91be7 | 945 | subtitles_are_requested = any([self.params.get('writesubtitles', False), |
0b7f3118 | 946 | self.params.get('writeautomaticsub')]) |
c4a91be7 | 947 | |
fe7e0c98 | 948 | if subtitles_are_requested and 'subtitles' in info_dict and info_dict['subtitles']: |
8222d8de JMF |
949 | # subtitles download errors are already managed as troubles in relevant IE |
950 | # that way it will silently go on when used with unsupporting IE | |
8222d8de | 951 | subtitles = info_dict['subtitles'] |
ca715127 | 952 | sub_format = self.params.get('subtitlesformat', 'srt') |
5d51a883 JMF |
953 | for sub_lang in subtitles.keys(): |
954 | sub = subtitles[sub_lang] | |
6804038d JMF |
955 | if sub is None: |
956 | continue | |
8222d8de | 957 | try: |
d4051a8e | 958 | sub_filename = subtitles_filename(filename, sub_lang, sub_format) |
7b6fefc9 | 959 | if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(sub_filename)): |
6febd1c1 | 960 | self.to_screen('[info] Video subtitle %s.%s is already_present' % (sub_lang, sub_format)) |
7b6fefc9 | 961 | else: |
6febd1c1 | 962 | self.to_screen('[info] Writing video subtitles to: ' + sub_filename) |
7b6fefc9 PH |
963 | with io.open(encodeFilename(sub_filename), 'w', encoding='utf-8') as subfile: |
964 | subfile.write(sub) | |
8222d8de | 965 | except (OSError, IOError): |
e4db1951 | 966 | self.report_error('Cannot write subtitles file ' + sub_filename) |
8222d8de JMF |
967 | return |
968 | ||
8222d8de | 969 | if self.params.get('writeinfojson', False): |
6febd1c1 | 970 | infofn = os.path.splitext(filename)[0] + '.info.json' |
7b6fefc9 | 971 | if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(infofn)): |
6febd1c1 | 972 | self.to_screen('[info] Video description metadata is already present') |
7b6fefc9 | 973 | else: |
6febd1c1 | 974 | self.to_screen('[info] Writing video description metadata as JSON to: ' + infofn) |
7b6fefc9 | 975 | try: |
1538eff6 | 976 | write_json_file(info_dict, encodeFilename(infofn)) |
7b6fefc9 | 977 | except (OSError, IOError): |
6febd1c1 | 978 | self.report_error('Cannot write metadata to JSON file ' + infofn) |
7b6fefc9 | 979 | return |
8222d8de JMF |
980 | |
981 | if self.params.get('writethumbnail', False): | |
d8269e1d | 982 | if info_dict.get('thumbnail') is not None: |
6febd1c1 PH |
983 | thumb_format = determine_ext(info_dict['thumbnail'], 'jpg') |
984 | thumb_filename = os.path.splitext(filename)[0] + '.' + thumb_format | |
0a9ce268 | 985 | if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(thumb_filename)): |
6febd1c1 | 986 | self.to_screen('[%s] %s: Thumbnail is already present' % |
7b6fefc9 PH |
987 | (info_dict['extractor'], info_dict['id'])) |
988 | else: | |
6febd1c1 | 989 | self.to_screen('[%s] %s: Downloading thumbnail ...' % |
7b6fefc9 PH |
990 | (info_dict['extractor'], info_dict['id'])) |
991 | try: | |
e9c092f1 | 992 | uf = self.urlopen(info_dict['thumbnail']) |
7b6fefc9 PH |
993 | with open(thumb_filename, 'wb') as thumbf: |
994 | shutil.copyfileobj(uf, thumbf) | |
6febd1c1 | 995 | self.to_screen('[%s] %s: Writing thumbnail to: %s' % |
7b6fefc9 PH |
996 | (info_dict['extractor'], info_dict['id'], thumb_filename)) |
997 | except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: | |
6febd1c1 | 998 | self.report_warning('Unable to download thumbnail "%s": %s' % |
7b6fefc9 | 999 | (info_dict['thumbnail'], compat_str(err))) |
8222d8de JMF |
1000 | |
1001 | if not self.params.get('skip_download', False): | |
1002 | if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(filename)): | |
1003 | success = True | |
1004 | else: | |
1005 | try: | |
6350728b JMF |
1006 | def dl(name, info): |
1007 | fd = get_suitable_downloader(info)(self, self.params) | |
1008 | for ph in self._progress_hooks: | |
1009 | fd.add_progress_hook(ph) | |
8d5797b0 PH |
1010 | if self.params.get('verbose'): |
1011 | self.to_stdout('[debug] Invoking downloader on %r' % info.get('url')) | |
6350728b JMF |
1012 | return fd.download(name, info) |
1013 | if info_dict.get('requested_formats') is not None: | |
1014 | downloaded = [] | |
1015 | success = True | |
b7f81164 | 1016 | merger = FFmpegMergerPP(self, not self.params.get('keepvideo')) |
58c3c7ae JMF |
1017 | if not merger._get_executable(): |
1018 | postprocessors = [] | |
1019 | self.report_warning('You have requested multiple ' | |
1020 | 'formats but ffmpeg or avconv are not installed.' | |
1021 | ' The formats won\'t be merged') | |
1022 | else: | |
1023 | postprocessors = [merger] | |
6350728b JMF |
1024 | for f in info_dict['requested_formats']: |
1025 | new_info = dict(info_dict) | |
1026 | new_info.update(f) | |
1027 | fname = self.prepare_filename(new_info) | |
1028 | fname = prepend_extension(fname, 'f%s' % f['format_id']) | |
1029 | downloaded.append(fname) | |
1030 | partial_success = dl(fname, new_info) | |
1031 | success = success and partial_success | |
58c3c7ae | 1032 | info_dict['__postprocessors'] = postprocessors |
6350728b JMF |
1033 | info_dict['__files_to_merge'] = downloaded |
1034 | else: | |
1035 | # Just a single file | |
1036 | success = dl(filename, info_dict) | |
8222d8de | 1037 | except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: |
6febd1c1 | 1038 | self.report_error('unable to download video data: %s' % str(err)) |
8222d8de | 1039 | return |
c40c6aaa JMF |
1040 | except (OSError, IOError) as err: |
1041 | raise UnavailableVideoError(err) | |
8222d8de | 1042 | except (ContentTooShortError, ) as err: |
6febd1c1 | 1043 | self.report_error('content too short (expected %s bytes and served %s)' % (err.expected, err.downloaded)) |
8222d8de JMF |
1044 | return |
1045 | ||
1046 | if success: | |
1047 | try: | |
1048 | self.post_process(filename, info_dict) | |
1049 | except (PostProcessingError) as err: | |
6febd1c1 | 1050 | self.report_error('postprocessing: %s' % str(err)) |
8222d8de JMF |
1051 | return |
1052 | ||
c1c9a79c PH |
1053 | self.record_download_archive(info_dict) |
1054 | ||
8222d8de JMF |
1055 | def download(self, url_list): |
1056 | """Download a given list of URLs.""" | |
acd69589 | 1057 | outtmpl = self.params.get('outtmpl', DEFAULT_OUTTMPL) |
0c75c3fa | 1058 | if (len(url_list) > 1 and |
acd69589 | 1059 | '%' not in outtmpl |
0c75c3fa | 1060 | and self.params.get('max_downloads') != 1): |
acd69589 | 1061 | raise SameFileError(outtmpl) |
8222d8de JMF |
1062 | |
1063 | for url in url_list: | |
1064 | try: | |
1065 | #It also downloads the videos | |
dca08720 | 1066 | self.extract_info(url) |
8222d8de | 1067 | except UnavailableVideoError: |
6febd1c1 | 1068 | self.report_error('unable to download video') |
8222d8de | 1069 | except MaxDownloadsReached: |
6febd1c1 | 1070 | self.to_screen('[info] Maximum number of downloaded files reached.') |
8222d8de JMF |
1071 | raise |
1072 | ||
1073 | return self._download_retcode | |
1074 | ||
1dcc4c0c | 1075 | def download_with_info_file(self, info_filename): |
395293a8 | 1076 | with io.open(info_filename, 'r', encoding='utf-8') as f: |
1dcc4c0c | 1077 | info = json.load(f) |
d4943898 JMF |
1078 | try: |
1079 | self.process_ie_result(info, download=True) | |
1080 | except DownloadError: | |
1081 | webpage_url = info.get('webpage_url') | |
1082 | if webpage_url is not None: | |
6febd1c1 | 1083 | self.report_warning('The info failed to download, trying with "%s"' % webpage_url) |
d4943898 JMF |
1084 | return self.download([webpage_url]) |
1085 | else: | |
1086 | raise | |
1087 | return self._download_retcode | |
1dcc4c0c | 1088 | |
8222d8de JMF |
1089 | def post_process(self, filename, ie_info): |
1090 | """Run all the postprocessors on the given file.""" | |
1091 | info = dict(ie_info) | |
1092 | info['filepath'] = filename | |
1093 | keep_video = None | |
6350728b JMF |
1094 | pps_chain = [] |
1095 | if ie_info.get('__postprocessors') is not None: | |
1096 | pps_chain.extend(ie_info['__postprocessors']) | |
1097 | pps_chain.extend(self._pps) | |
1098 | for pp in pps_chain: | |
8222d8de | 1099 | try: |
fe7e0c98 | 1100 | keep_video_wish, new_info = pp.run(info) |
8222d8de JMF |
1101 | if keep_video_wish is not None: |
1102 | if keep_video_wish: | |
1103 | keep_video = keep_video_wish | |
1104 | elif keep_video is None: | |
1105 | # No clear decision yet, let IE decide | |
1106 | keep_video = keep_video_wish | |
1107 | except PostProcessingError as e: | |
bbcbf4d4 | 1108 | self.report_error(e.msg) |
8222d8de JMF |
1109 | if keep_video is False and not self.params.get('keepvideo', False): |
1110 | try: | |
6febd1c1 | 1111 | self.to_screen('Deleting original file %s (pass -k to keep)' % filename) |
8222d8de JMF |
1112 | os.remove(encodeFilename(filename)) |
1113 | except (IOError, OSError): | |
6febd1c1 | 1114 | self.report_warning('Unable to remove downloaded video file') |
c1c9a79c | 1115 | |
5db07df6 PH |
1116 | def _make_archive_id(self, info_dict): |
1117 | # Future-proof against any change in case | |
1118 | # and backwards compatibility with prior versions | |
d31209a1 | 1119 | extractor = info_dict.get('extractor_key') |
7012b23c PH |
1120 | if extractor is None: |
1121 | if 'id' in info_dict: | |
1122 | extractor = info_dict.get('ie_key') # key in a playlist | |
1123 | if extractor is None: | |
5db07df6 | 1124 | return None # Incomplete video information |
6febd1c1 | 1125 | return extractor.lower() + ' ' + info_dict['id'] |
5db07df6 PH |
1126 | |
1127 | def in_download_archive(self, info_dict): | |
1128 | fn = self.params.get('download_archive') | |
1129 | if fn is None: | |
1130 | return False | |
1131 | ||
1132 | vid_id = self._make_archive_id(info_dict) | |
1133 | if vid_id is None: | |
7012b23c | 1134 | return False # Incomplete video information |
5db07df6 | 1135 | |
c1c9a79c PH |
1136 | try: |
1137 | with locked_file(fn, 'r', encoding='utf-8') as archive_file: | |
1138 | for line in archive_file: | |
1139 | if line.strip() == vid_id: | |
1140 | return True | |
1141 | except IOError as ioe: | |
1142 | if ioe.errno != errno.ENOENT: | |
1143 | raise | |
1144 | return False | |
1145 | ||
1146 | def record_download_archive(self, info_dict): | |
1147 | fn = self.params.get('download_archive') | |
1148 | if fn is None: | |
1149 | return | |
5db07df6 PH |
1150 | vid_id = self._make_archive_id(info_dict) |
1151 | assert vid_id | |
c1c9a79c | 1152 | with locked_file(fn, 'a', encoding='utf-8') as archive_file: |
6febd1c1 | 1153 | archive_file.write(vid_id + '\n') |
dd82ffea | 1154 | |
8c51aa65 | 1155 | @staticmethod |
8abeeb94 | 1156 | def format_resolution(format, default='unknown'): |
fb04e403 PH |
1157 | if format.get('vcodec') == 'none': |
1158 | return 'audio only' | |
f49d89ee PH |
1159 | if format.get('resolution') is not None: |
1160 | return format['resolution'] | |
8c51aa65 JMF |
1161 | if format.get('height') is not None: |
1162 | if format.get('width') is not None: | |
6febd1c1 | 1163 | res = '%sx%s' % (format['width'], format['height']) |
8c51aa65 | 1164 | else: |
6febd1c1 | 1165 | res = '%sp' % format['height'] |
f49d89ee | 1166 | elif format.get('width') is not None: |
6febd1c1 | 1167 | res = '?x%d' % format['width'] |
8c51aa65 | 1168 | else: |
8abeeb94 | 1169 | res = default |
8c51aa65 JMF |
1170 | return res |
1171 | ||
c57f7757 PH |
1172 | def _format_note(self, fdict): |
1173 | res = '' | |
1174 | if fdict.get('ext') in ['f4f', 'f4m']: | |
1175 | res += '(unsupported) ' | |
1176 | if fdict.get('format_note') is not None: | |
1177 | res += fdict['format_note'] + ' ' | |
1178 | if fdict.get('tbr') is not None: | |
1179 | res += '%4dk ' % fdict['tbr'] | |
1180 | if fdict.get('container') is not None: | |
1181 | if res: | |
1182 | res += ', ' | |
1183 | res += '%s container' % fdict['container'] | |
1184 | if (fdict.get('vcodec') is not None and | |
1185 | fdict.get('vcodec') != 'none'): | |
1186 | if res: | |
1187 | res += ', ' | |
1188 | res += fdict['vcodec'] | |
91c7271a | 1189 | if fdict.get('vbr') is not None: |
c57f7757 PH |
1190 | res += '@' |
1191 | elif fdict.get('vbr') is not None and fdict.get('abr') is not None: | |
1192 | res += 'video@' | |
1193 | if fdict.get('vbr') is not None: | |
1194 | res += '%4dk' % fdict['vbr'] | |
1195 | if fdict.get('acodec') is not None: | |
1196 | if res: | |
1197 | res += ', ' | |
1198 | if fdict['acodec'] == 'none': | |
1199 | res += 'video only' | |
1200 | else: | |
1201 | res += '%-5s' % fdict['acodec'] | |
1202 | elif fdict.get('abr') is not None: | |
1203 | if res: | |
1204 | res += ', ' | |
1205 | res += 'audio' | |
1206 | if fdict.get('abr') is not None: | |
1207 | res += '@%3dk' % fdict['abr'] | |
1208 | if fdict.get('asr') is not None: | |
1209 | res += ' (%5dHz)' % fdict['asr'] | |
1210 | if fdict.get('filesize') is not None: | |
1211 | if res: | |
1212 | res += ', ' | |
1213 | res += format_bytes(fdict['filesize']) | |
9732d77e PH |
1214 | elif fdict.get('filesize_approx') is not None: |
1215 | if res: | |
1216 | res += ', ' | |
1217 | res += '~' + format_bytes(fdict['filesize_approx']) | |
c57f7757 | 1218 | return res |
91c7271a | 1219 | |
c57f7757 | 1220 | def list_formats(self, info_dict): |
02dbf93f | 1221 | def line(format, idlen=20): |
6febd1c1 | 1222 | return (('%-' + compat_str(idlen + 1) + 's%-10s%-12s%s') % ( |
8c51aa65 JMF |
1223 | format['format_id'], |
1224 | format['ext'], | |
8c51aa65 | 1225 | self.format_resolution(format), |
c57f7757 | 1226 | self._format_note(format), |
02dbf93f | 1227 | )) |
57dd9a8f | 1228 | |
94badb25 | 1229 | formats = info_dict.get('formats', [info_dict]) |
6febd1c1 | 1230 | idlen = max(len('format code'), |
02dbf93f PH |
1231 | max(len(f['format_id']) for f in formats)) |
1232 | formats_s = [line(f, idlen) for f in formats] | |
94badb25 | 1233 | if len(formats) > 1: |
c57f7757 PH |
1234 | formats_s[0] += (' ' if self._format_note(formats[0]) else '') + '(worst)' |
1235 | formats_s[-1] += (' ' if self._format_note(formats[-1]) else '') + '(best)' | |
57dd9a8f PH |
1236 | |
1237 | header_line = line({ | |
6febd1c1 PH |
1238 | 'format_id': 'format code', 'ext': 'extension', |
1239 | 'resolution': 'resolution', 'format_note': 'note'}, idlen=idlen) | |
1240 | self.to_screen('[info] Available formats for %s:\n%s\n%s' % | |
1241 | (info_dict['id'], header_line, '\n'.join(formats_s))) | |
dca08720 PH |
1242 | |
1243 | def urlopen(self, req): | |
1244 | """ Start an HTTP download """ | |
37419b4f | 1245 | |
d05cfe06 S |
1246 | # According to RFC 3986, URLs can not contain non-ASCII characters, however this is not |
1247 | # always respected by websites, some tend to give out URLs with non percent-encoded | |
1248 | # non-ASCII characters (see telemb.py, ard.py [#3412]) | |
37419b4f | 1249 | # urllib chokes on URLs with non-ASCII characters (see http://bugs.python.org/issue3991) |
d05cfe06 S |
1250 | # To work around aforementioned issue we will replace request's original URL with |
1251 | # percent-encoded one | |
37419b4f | 1252 | url = req if isinstance(req, compat_str) else req.get_full_url() |
d05cfe06 | 1253 | url_escaped = escape_url(url) |
37419b4f S |
1254 | |
1255 | # Substitute URL if any change after escaping | |
1256 | if url != url_escaped: | |
1257 | if isinstance(req, compat_str): | |
1258 | req = url_escaped | |
1259 | else: | |
1260 | req = compat_urllib_request.Request( | |
1261 | url_escaped, data=req.data, headers=req.headers, | |
1262 | origin_req_host=req.origin_req_host, unverifiable=req.unverifiable) | |
1263 | ||
19a41fc6 | 1264 | return self._opener.open(req, timeout=self._socket_timeout) |
dca08720 PH |
1265 | |
1266 | def print_debug_header(self): | |
1267 | if not self.params.get('verbose'): | |
1268 | return | |
62fec3b2 | 1269 | |
4192b51c PH |
1270 | if type('') is not compat_str: |
1271 | # Python 2.6 on SLES11 SP1 (https://github.com/rg3/youtube-dl/issues/3326) | |
1272 | self.report_warning( | |
1273 | 'Your Python is broken! Update to a newer and supported version') | |
1274 | ||
b0472057 | 1275 | encoding_str = ( |
734f90bb PH |
1276 | '[debug] Encodings: locale %s, fs %s, out %s, pref %s\n' % ( |
1277 | locale.getpreferredencoding(), | |
1278 | sys.getfilesystemencoding(), | |
1279 | sys.stdout.encoding, | |
b0472057 | 1280 | self.get_encoding())) |
4192b51c | 1281 | write_string(encoding_str, encoding=None) |
734f90bb PH |
1282 | |
1283 | self._write_string('[debug] youtube-dl version ' + __version__ + '\n') | |
dca08720 PH |
1284 | try: |
1285 | sp = subprocess.Popen( | |
1286 | ['git', 'rev-parse', '--short', 'HEAD'], | |
1287 | stdout=subprocess.PIPE, stderr=subprocess.PIPE, | |
1288 | cwd=os.path.dirname(os.path.abspath(__file__))) | |
1289 | out, err = sp.communicate() | |
1290 | out = out.decode().strip() | |
1291 | if re.match('[0-9a-f]+', out): | |
734f90bb | 1292 | self._write_string('[debug] Git HEAD: ' + out + '\n') |
dca08720 PH |
1293 | except: |
1294 | try: | |
1295 | sys.exc_clear() | |
1296 | except: | |
1297 | pass | |
734f90bb | 1298 | self._write_string('[debug] Python version %s - %s' % |
6febd1c1 | 1299 | (platform.python_version(), platform_name()) + '\n') |
dca08720 PH |
1300 | |
1301 | proxy_map = {} | |
1302 | for handler in self._opener.handlers: | |
1303 | if hasattr(handler, 'proxies'): | |
1304 | proxy_map.update(handler.proxies) | |
734f90bb | 1305 | self._write_string('[debug] Proxy map: ' + compat_str(proxy_map) + '\n') |
dca08720 | 1306 | |
e344693b | 1307 | def _setup_opener(self): |
6ad14cab | 1308 | timeout_val = self.params.get('socket_timeout') |
19a41fc6 | 1309 | self._socket_timeout = 600 if timeout_val is None else float(timeout_val) |
6ad14cab | 1310 | |
dca08720 PH |
1311 | opts_cookiefile = self.params.get('cookiefile') |
1312 | opts_proxy = self.params.get('proxy') | |
1313 | ||
1314 | if opts_cookiefile is None: | |
1315 | self.cookiejar = compat_cookiejar.CookieJar() | |
1316 | else: | |
1317 | self.cookiejar = compat_cookiejar.MozillaCookieJar( | |
1318 | opts_cookiefile) | |
1319 | if os.access(opts_cookiefile, os.R_OK): | |
1320 | self.cookiejar.load() | |
1321 | ||
1322 | cookie_processor = compat_urllib_request.HTTPCookieProcessor( | |
1323 | self.cookiejar) | |
1324 | if opts_proxy is not None: | |
1325 | if opts_proxy == '': | |
1326 | proxies = {} | |
1327 | else: | |
1328 | proxies = {'http': opts_proxy, 'https': opts_proxy} | |
1329 | else: | |
1330 | proxies = compat_urllib_request.getproxies() | |
1331 | # Set HTTPS proxy to HTTP one if given (https://github.com/rg3/youtube-dl/issues/805) | |
1332 | if 'http' in proxies and 'https' not in proxies: | |
1333 | proxies['https'] = proxies['http'] | |
1334 | proxy_handler = compat_urllib_request.ProxyHandler(proxies) | |
a0ddb8a2 PH |
1335 | |
1336 | debuglevel = 1 if self.params.get('debug_printtraffic') else 0 | |
dca08720 | 1337 | https_handler = make_HTTPS_handler( |
a0ddb8a2 PH |
1338 | self.params.get('nocheckcertificate', False), debuglevel=debuglevel) |
1339 | ydlh = YoutubeDLHandler(debuglevel=debuglevel) | |
dca08720 | 1340 | opener = compat_urllib_request.build_opener( |
a0ddb8a2 | 1341 | https_handler, proxy_handler, cookie_processor, ydlh) |
dca08720 PH |
1342 | # Delete the default user-agent header, which would otherwise apply in |
1343 | # cases where our custom HTTP handler doesn't come into play | |
1344 | # (See https://github.com/rg3/youtube-dl/issues/1309 for details) | |
1345 | opener.addheaders = [] | |
1346 | self._opener = opener | |
62fec3b2 PH |
1347 | |
1348 | def encode(self, s): | |
1349 | if isinstance(s, bytes): | |
1350 | return s # Already encoded | |
1351 | ||
1352 | try: | |
1353 | return s.encode(self.get_encoding()) | |
1354 | except UnicodeEncodeError as err: | |
1355 | err.reason = err.reason + '. Check your system encoding configuration or use the --encoding option.' | |
1356 | raise | |
1357 | ||
1358 | def get_encoding(self): | |
1359 | encoding = self.params.get('encoding') | |
1360 | if encoding is None: | |
1361 | encoding = preferredencoding() | |
1362 | return encoding |