]>
Commit | Line | Data |
---|---|---|
8222d8de JMF |
1 | #!/usr/bin/env python |
2 | # -*- coding: utf-8 -*- | |
3 | ||
4 | from __future__ import absolute_import | |
5 | ||
c1c9a79c | 6 | import errno |
8222d8de | 7 | import io |
8694c600 | 8 | import json |
8222d8de | 9 | import os |
dca08720 | 10 | import platform |
8222d8de JMF |
11 | import re |
12 | import shutil | |
dca08720 | 13 | import subprocess |
8222d8de JMF |
14 | import socket |
15 | import sys | |
16 | import time | |
17 | import traceback | |
18 | ||
1e5b9a95 PH |
19 | if os.name == 'nt': |
20 | import ctypes | |
21 | ||
ce02ed60 | 22 | from .utils import ( |
dca08720 | 23 | compat_cookiejar, |
ce02ed60 PH |
24 | compat_http_client, |
25 | compat_print, | |
26 | compat_str, | |
27 | compat_urllib_error, | |
28 | compat_urllib_request, | |
29 | ContentTooShortError, | |
30 | date_from_str, | |
31 | DateRange, | |
32 | determine_ext, | |
33 | DownloadError, | |
34 | encodeFilename, | |
35 | ExtractorError, | |
36 | locked_file, | |
dca08720 | 37 | make_HTTPS_handler, |
ce02ed60 PH |
38 | MaxDownloadsReached, |
39 | PostProcessingError, | |
dca08720 | 40 | platform_name, |
ce02ed60 PH |
41 | preferredencoding, |
42 | SameFileError, | |
43 | sanitize_filename, | |
44 | subtitles_filename, | |
45 | takewhile_inclusive, | |
46 | UnavailableVideoError, | |
47 | write_json_file, | |
48 | write_string, | |
dca08720 | 49 | YoutubeDLHandler, |
ce02ed60 | 50 | ) |
023fa8c4 | 51 | from .extractor import get_info_extractor, gen_extractors |
8222d8de | 52 | from .FileDownloader import FileDownloader |
dca08720 | 53 | from .version import __version__ |
8222d8de JMF |
54 | |
55 | ||
56 | class YoutubeDL(object): | |
57 | """YoutubeDL class. | |
58 | ||
59 | YoutubeDL objects are the ones responsible of downloading the | |
60 | actual video file and writing it to disk if the user has requested | |
61 | it, among some other tasks. In most cases there should be one per | |
62 | program. As, given a video URL, the downloader doesn't know how to | |
63 | extract all the needed information, task that InfoExtractors do, it | |
64 | has to pass the URL to one of them. | |
65 | ||
66 | For this, YoutubeDL objects have a method that allows | |
67 | InfoExtractors to be registered in a given order. When it is passed | |
68 | a URL, the YoutubeDL object handles it to the first InfoExtractor it | |
69 | finds that reports being able to handle it. The InfoExtractor extracts | |
70 | all the information about the video or videos the URL refers to, and | |
71 | YoutubeDL process the extracted information, possibly using a File | |
72 | Downloader to download the video. | |
73 | ||
74 | YoutubeDL objects accept a lot of parameters. In order not to saturate | |
75 | the object constructor with arguments, it receives a dictionary of | |
76 | options instead. These options are available through the params | |
77 | attribute for the InfoExtractors to use. The YoutubeDL also | |
78 | registers itself as the downloader in charge for the InfoExtractors | |
79 | that are added to it, so this is a "mutual registration". | |
80 | ||
81 | Available options: | |
82 | ||
83 | username: Username for authentication purposes. | |
84 | password: Password for authentication purposes. | |
c6c19746 | 85 | videopassword: Password for acces a video. |
8222d8de JMF |
86 | usenetrc: Use netrc for authentication instead. |
87 | verbose: Print additional info to stdout. | |
88 | quiet: Do not print messages to stdout. | |
89 | forceurl: Force printing final URL. | |
90 | forcetitle: Force printing title. | |
91 | forceid: Force printing ID. | |
92 | forcethumbnail: Force printing thumbnail URL. | |
93 | forcedescription: Force printing description. | |
94 | forcefilename: Force printing final filename. | |
8694c600 | 95 | forcejson: Force printing info_dict as JSON. |
8222d8de JMF |
96 | simulate: Do not download the video files. |
97 | format: Video format code. | |
98 | format_limit: Highest quality format to try. | |
99 | outtmpl: Template for output names. | |
100 | restrictfilenames: Do not allow "&" and spaces in file names | |
101 | ignoreerrors: Do not stop on download errors. | |
102 | nooverwrites: Prevent overwriting files. | |
103 | playliststart: Playlist item to start at. | |
104 | playlistend: Playlist item to end at. | |
105 | matchtitle: Download only matching titles. | |
106 | rejecttitle: Reject downloads for matching titles. | |
107 | logtostderr: Log messages to stderr instead of stdout. | |
108 | writedescription: Write the video description to a .description file | |
109 | writeinfojson: Write the video description to a .info.json file | |
1fb07d10 | 110 | writeannotations: Write the video annotations to a .annotations.xml file |
8222d8de JMF |
111 | writethumbnail: Write the thumbnail image to a file |
112 | writesubtitles: Write the video subtitles to a file | |
b004821f | 113 | writeautomaticsub: Write the automatic subtitles to a file |
8222d8de | 114 | allsubtitles: Downloads all the subtitles of the video |
0b7f3118 | 115 | (requires writesubtitles or writeautomaticsub) |
8222d8de | 116 | listsubtitles: Lists all available subtitles for the video |
b98a6b2f | 117 | subtitlesformat: Subtitle format [srt/sbv/vtt] (default=srt) |
aa6a10c4 | 118 | subtitleslangs: List of languages of the subtitles to download |
8222d8de JMF |
119 | keepvideo: Keep the video file after post-processing |
120 | daterange: A DateRange object, download only if the upload_date is in the range. | |
121 | skip_download: Skip the actual download of the video file | |
c35f9e72 | 122 | cachedir: Location of the cache files in the filesystem. |
c3c88a26 | 123 | None to disable filesystem cache. |
47192f92 | 124 | noplaylist: Download single video instead of a playlist if in doubt. |
8dbe9899 PH |
125 | age_limit: An integer representing the user's age in years. |
126 | Unsuitable videos for the given age are skipped. | |
c1c9a79c PH |
127 | downloadarchive: File name of a file where all downloads are recorded. |
128 | Videos already present in the file are not downloaded | |
129 | again. | |
dca08720 PH |
130 | cookiefile: File name where cookies should be read from and dumped to. |
131 | nocheckcertificate Do not verify SSL certificates | |
fe7e0c98 | 132 | |
8222d8de JMF |
133 | The following parameters are not used by YoutubeDL itself, they are used by |
134 | the FileDownloader: | |
135 | nopart, updatetime, buffersize, ratelimit, min_filesize, max_filesize, test, | |
136 | noresizebuffer, retries, continuedl, noprogress, consoletitle | |
137 | """ | |
138 | ||
139 | params = None | |
140 | _ies = [] | |
141 | _pps = [] | |
142 | _download_retcode = None | |
143 | _num_downloads = None | |
144 | _screen_file = None | |
145 | ||
146 | def __init__(self, params): | |
147 | """Create a FileDownloader object with the given options.""" | |
148 | self._ies = [] | |
56c73665 | 149 | self._ies_instances = {} |
8222d8de JMF |
150 | self._pps = [] |
151 | self._progress_hooks = [] | |
152 | self._download_retcode = 0 | |
153 | self._num_downloads = 0 | |
154 | self._screen_file = [sys.stdout, sys.stderr][params.get('logtostderr', False)] | |
34308b30 PH |
155 | |
156 | if (sys.version_info >= (3,) and sys.platform != 'win32' and | |
157 | sys.getfilesystemencoding() in ['ascii', 'ANSI_X3.4-1968'] | |
158 | and not params['restrictfilenames']): | |
159 | # On Python 3, the Unicode filesystem API will throw errors (#1474) | |
160 | self.report_warning( | |
1d368c75 | 161 | u'Assuming --restrict-filenames since file system encoding ' |
34308b30 PH |
162 | u'cannot encode all charactes. ' |
163 | u'Set the LC_ALL environment variable to fix this.') | |
164 | params['restrictfilenames'] = True | |
165 | ||
8222d8de JMF |
166 | self.params = params |
167 | self.fd = FileDownloader(self, self.params) | |
168 | ||
169 | if '%(stitle)s' in self.params['outtmpl']: | |
170 | self.report_warning(u'%(stitle)s is deprecated. Use the %(title)s and the --restrict-filenames flag(which also secures %(uploader)s et al) instead.') | |
171 | ||
dca08720 PH |
172 | self._setup_opener() |
173 | ||
8222d8de JMF |
174 | def add_info_extractor(self, ie): |
175 | """Add an InfoExtractor object to the end of the list.""" | |
176 | self._ies.append(ie) | |
56c73665 | 177 | self._ies_instances[ie.ie_key()] = ie |
8222d8de JMF |
178 | ie.set_downloader(self) |
179 | ||
56c73665 JMF |
180 | def get_info_extractor(self, ie_key): |
181 | """ | |
182 | Get an instance of an IE with name ie_key, it will try to get one from | |
183 | the _ies list, if there's no instance it will create a new one and add | |
184 | it to the extractor list. | |
185 | """ | |
186 | ie = self._ies_instances.get(ie_key) | |
187 | if ie is None: | |
188 | ie = get_info_extractor(ie_key)() | |
189 | self.add_info_extractor(ie) | |
190 | return ie | |
191 | ||
023fa8c4 JMF |
192 | def add_default_info_extractors(self): |
193 | """ | |
194 | Add the InfoExtractors returned by gen_extractors to the end of the list | |
195 | """ | |
196 | for ie in gen_extractors(): | |
197 | self.add_info_extractor(ie) | |
198 | ||
8222d8de JMF |
199 | def add_post_processor(self, pp): |
200 | """Add a PostProcessor object to the end of the chain.""" | |
201 | self._pps.append(pp) | |
202 | pp.set_downloader(self) | |
203 | ||
204 | def to_screen(self, message, skip_eol=False): | |
205 | """Print message to stdout if not in quiet mode.""" | |
8222d8de JMF |
206 | if not self.params.get('quiet', False): |
207 | terminator = [u'\n', u''][skip_eol] | |
208 | output = message + terminator | |
7459e3a2 | 209 | write_string(output, self._screen_file) |
8222d8de JMF |
210 | |
211 | def to_stderr(self, message): | |
212 | """Print message to stderr.""" | |
213 | assert type(message) == type(u'') | |
214 | output = message + u'\n' | |
215 | if 'b' in getattr(self._screen_file, 'mode', '') or sys.version_info[0] < 3: # Python 2 lies about the mode of sys.stdout/sys.stderr | |
216 | output = output.encode(preferredencoding()) | |
217 | sys.stderr.write(output) | |
218 | ||
1e5b9a95 PH |
219 | def to_console_title(self, message): |
220 | if not self.params.get('consoletitle', False): | |
221 | return | |
222 | if os.name == 'nt' and ctypes.windll.kernel32.GetConsoleWindow(): | |
223 | # c_wchar_p() might not be necessary if `message` is | |
224 | # already of type unicode() | |
225 | ctypes.windll.kernel32.SetConsoleTitleW(ctypes.c_wchar_p(message)) | |
226 | elif 'TERM' in os.environ: | |
749febf4 | 227 | write_string(u'\033]0;%s\007' % message, self._screen_file) |
1e5b9a95 | 228 | |
bdde425c PH |
229 | def save_console_title(self): |
230 | if not self.params.get('consoletitle', False): | |
231 | return | |
232 | if 'TERM' in os.environ: | |
efd6c574 JMF |
233 | # Save the title on stack |
234 | write_string(u'\033[22;0t', self._screen_file) | |
bdde425c PH |
235 | |
236 | def restore_console_title(self): | |
237 | if not self.params.get('consoletitle', False): | |
238 | return | |
239 | if 'TERM' in os.environ: | |
efd6c574 JMF |
240 | # Restore the title from stack |
241 | write_string(u'\033[23;0t', self._screen_file) | |
bdde425c PH |
242 | |
243 | def __enter__(self): | |
244 | self.save_console_title() | |
245 | return self | |
246 | ||
247 | def __exit__(self, *args): | |
248 | self.restore_console_title() | |
dca08720 PH |
249 | |
250 | if self.params.get('cookiefile') is not None: | |
251 | self.cookiejar.save() | |
bdde425c | 252 | |
8222d8de JMF |
253 | def fixed_template(self): |
254 | """Checks if the output template is fixed.""" | |
255 | return (re.search(u'(?u)%\\(.+?\\)s', self.params['outtmpl']) is None) | |
256 | ||
257 | def trouble(self, message=None, tb=None): | |
258 | """Determine action to take when a download problem appears. | |
259 | ||
260 | Depending on if the downloader has been configured to ignore | |
261 | download errors or not, this method may throw an exception or | |
262 | not when errors are found, after printing the message. | |
263 | ||
264 | tb, if given, is additional traceback information. | |
265 | """ | |
266 | if message is not None: | |
267 | self.to_stderr(message) | |
268 | if self.params.get('verbose'): | |
269 | if tb is None: | |
270 | if sys.exc_info()[0]: # if .trouble has been called from an except block | |
271 | tb = u'' | |
272 | if hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]: | |
273 | tb += u''.join(traceback.format_exception(*sys.exc_info()[1].exc_info)) | |
274 | tb += compat_str(traceback.format_exc()) | |
275 | else: | |
276 | tb_data = traceback.format_list(traceback.extract_stack()) | |
277 | tb = u''.join(tb_data) | |
278 | self.to_stderr(tb) | |
279 | if not self.params.get('ignoreerrors', False): | |
280 | if sys.exc_info()[0] and hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]: | |
281 | exc_info = sys.exc_info()[1].exc_info | |
282 | else: | |
283 | exc_info = sys.exc_info() | |
284 | raise DownloadError(message, exc_info) | |
285 | self._download_retcode = 1 | |
286 | ||
287 | def report_warning(self, message): | |
288 | ''' | |
289 | Print the message to stderr, it will be prefixed with 'WARNING:' | |
290 | If stderr is a tty file the 'WARNING:' will be colored | |
291 | ''' | |
292 | if sys.stderr.isatty() and os.name != 'nt': | |
fe7e0c98 | 293 | _msg_header = u'\033[0;33mWARNING:\033[0m' |
8222d8de | 294 | else: |
fe7e0c98 JMF |
295 | _msg_header = u'WARNING:' |
296 | warning_message = u'%s %s' % (_msg_header, message) | |
8222d8de JMF |
297 | self.to_stderr(warning_message) |
298 | ||
299 | def report_error(self, message, tb=None): | |
300 | ''' | |
301 | Do the same as trouble, but prefixes the message with 'ERROR:', colored | |
302 | in red if stderr is a tty file. | |
303 | ''' | |
304 | if sys.stderr.isatty() and os.name != 'nt': | |
305 | _msg_header = u'\033[0;31mERROR:\033[0m' | |
306 | else: | |
307 | _msg_header = u'ERROR:' | |
308 | error_message = u'%s %s' % (_msg_header, message) | |
309 | self.trouble(error_message, tb) | |
310 | ||
8222d8de JMF |
311 | def report_writedescription(self, descfn): |
312 | """ Report that the description file is being written """ | |
313 | self.to_screen(u'[info] Writing video description to: ' + descfn) | |
314 | ||
315 | def report_writesubtitles(self, sub_filename): | |
316 | """ Report that the subtitles file is being written """ | |
317 | self.to_screen(u'[info] Writing video subtitles to: ' + sub_filename) | |
318 | ||
319 | def report_writeinfojson(self, infofn): | |
320 | """ Report that the metadata file has been written """ | |
321 | self.to_screen(u'[info] Video description metadata as JSON to: ' + infofn) | |
322 | ||
1fb07d10 JG |
323 | def report_writeannotations(self, annofn): |
324 | """ Report that the annotations file has been written. """ | |
325 | self.to_screen(u'[info] Writing video annotations to: ' + annofn) | |
326 | ||
8222d8de JMF |
327 | def report_file_already_downloaded(self, file_name): |
328 | """Report file has already been fully downloaded.""" | |
329 | try: | |
330 | self.to_screen(u'[download] %s has already been downloaded' % file_name) | |
ce02ed60 | 331 | except UnicodeEncodeError: |
8222d8de JMF |
332 | self.to_screen(u'[download] The file has already been downloaded') |
333 | ||
334 | def increment_downloads(self): | |
335 | """Increment the ordinal that assigns a number to each file.""" | |
336 | self._num_downloads += 1 | |
337 | ||
338 | def prepare_filename(self, info_dict): | |
339 | """Generate the output filename.""" | |
340 | try: | |
341 | template_dict = dict(info_dict) | |
342 | ||
343 | template_dict['epoch'] = int(time.time()) | |
344 | autonumber_size = self.params.get('autonumber_size') | |
345 | if autonumber_size is None: | |
346 | autonumber_size = 5 | |
347 | autonumber_templ = u'%0' + str(autonumber_size) + u'd' | |
348 | template_dict['autonumber'] = autonumber_templ % self._num_downloads | |
702665c0 | 349 | if template_dict.get('playlist_index') is not None: |
8222d8de JMF |
350 | template_dict['playlist_index'] = u'%05d' % template_dict['playlist_index'] |
351 | ||
586a91b6 | 352 | sanitize = lambda k, v: sanitize_filename( |
8222d8de JMF |
353 | u'NA' if v is None else compat_str(v), |
354 | restricted=self.params.get('restrictfilenames'), | |
586a91b6 PH |
355 | is_id=(k == u'id')) |
356 | template_dict = dict((k, sanitize(k, v)) | |
357 | for k, v in template_dict.items()) | |
8222d8de | 358 | |
586a91b6 PH |
359 | tmpl = os.path.expanduser(self.params['outtmpl']) |
360 | filename = tmpl % template_dict | |
8222d8de JMF |
361 | return filename |
362 | except KeyError as err: | |
363 | self.report_error(u'Erroneous output template') | |
364 | return None | |
365 | except ValueError as err: | |
4efba05c | 366 | self.report_error(u'Error in output template: ' + str(err) + u' (encoding: ' + repr(preferredencoding()) + ')') |
8222d8de JMF |
367 | return None |
368 | ||
369 | def _match_entry(self, info_dict): | |
370 | """ Returns None iff the file should be downloaded """ | |
371 | ||
372 | title = info_dict['title'] | |
373 | matchtitle = self.params.get('matchtitle', False) | |
374 | if matchtitle: | |
375 | if not re.search(matchtitle, title, re.IGNORECASE): | |
376 | return u'[download] "' + title + '" title did not match pattern "' + matchtitle + '"' | |
377 | rejecttitle = self.params.get('rejecttitle', False) | |
378 | if rejecttitle: | |
379 | if re.search(rejecttitle, title, re.IGNORECASE): | |
380 | return u'"' + title + '" title matched reject pattern "' + rejecttitle + '"' | |
381 | date = info_dict.get('upload_date', None) | |
382 | if date is not None: | |
383 | dateRange = self.params.get('daterange', DateRange()) | |
384 | if date not in dateRange: | |
385 | return u'[download] %s upload date is not in range %s' % (date_from_str(date).isoformat(), dateRange) | |
8dbe9899 PH |
386 | age_limit = self.params.get('age_limit') |
387 | if age_limit is not None: | |
cfadd183 | 388 | if age_limit < info_dict.get('age_limit', 0): |
8dbe9899 | 389 | return u'Skipping "' + title + '" because it is age restricted' |
c1c9a79c | 390 | if self.in_download_archive(info_dict): |
ee6c9f95 | 391 | return (u'%(title)s has already been recorded in archive' |
c1c9a79c | 392 | % info_dict) |
8222d8de | 393 | return None |
fe7e0c98 | 394 | |
b6c45014 JMF |
395 | @staticmethod |
396 | def add_extra_info(info_dict, extra_info): | |
397 | '''Set the keys from extra_info in info dict if they are missing''' | |
398 | for key, value in extra_info.items(): | |
399 | info_dict.setdefault(key, value) | |
400 | ||
8222d8de JMF |
401 | def extract_info(self, url, download=True, ie_key=None, extra_info={}): |
402 | ''' | |
403 | Returns a list with a dictionary for each video we find. | |
404 | If 'download', also downloads the videos. | |
405 | extra_info is a dict containing the extra values to add to each result | |
406 | ''' | |
fe7e0c98 | 407 | |
8222d8de | 408 | if ie_key: |
56c73665 | 409 | ies = [self.get_info_extractor(ie_key)] |
8222d8de JMF |
410 | else: |
411 | ies = self._ies | |
412 | ||
413 | for ie in ies: | |
414 | if not ie.suitable(url): | |
415 | continue | |
416 | ||
417 | if not ie.working(): | |
418 | self.report_warning(u'The program functionality for this site has been marked as broken, ' | |
419 | u'and will probably not work.') | |
420 | ||
421 | try: | |
422 | ie_result = ie.extract(url) | |
423 | if ie_result is None: # Finished already (backwards compatibility; listformats and friends should be moved here) | |
424 | break | |
425 | if isinstance(ie_result, list): | |
426 | # Backwards compatibility: old IE result format | |
8222d8de JMF |
427 | ie_result = { |
428 | '_type': 'compat_list', | |
429 | 'entries': ie_result, | |
430 | } | |
9103bbc5 JMF |
431 | self.add_extra_info(ie_result, |
432 | { | |
433 | 'extractor': ie.IE_NAME, | |
be97abc2 JMF |
434 | 'webpage_url': url, |
435 | 'extractor_key': ie.ie_key(), | |
9103bbc5 | 436 | }) |
b6c45014 | 437 | return self.process_ie_result(ie_result, download, extra_info) |
8222d8de JMF |
438 | except ExtractorError as de: # An error we somewhat expected |
439 | self.report_error(compat_str(de), de.format_traceback()) | |
440 | break | |
441 | except Exception as e: | |
442 | if self.params.get('ignoreerrors', False): | |
443 | self.report_error(compat_str(e), tb=compat_str(traceback.format_exc())) | |
444 | break | |
445 | else: | |
446 | raise | |
447 | else: | |
448 | self.report_error(u'no suitable InfoExtractor: %s' % url) | |
fe7e0c98 | 449 | |
8222d8de JMF |
450 | def process_ie_result(self, ie_result, download=True, extra_info={}): |
451 | """ | |
452 | Take the result of the ie(may be modified) and resolve all unresolved | |
453 | references (URLs, playlist items). | |
454 | ||
455 | It will also download the videos if 'download'. | |
456 | Returns the resolved ie_result. | |
457 | """ | |
458 | ||
459 | result_type = ie_result.get('_type', 'video') # If not given we suppose it's a video, support the default old system | |
460 | if result_type == 'video': | |
b6c45014 | 461 | self.add_extra_info(ie_result, extra_info) |
feee2ecf | 462 | return self.process_video_result(ie_result, download=download) |
8222d8de JMF |
463 | elif result_type == 'url': |
464 | # We have to add extra_info to the results because it may be | |
465 | # contained in a playlist | |
466 | return self.extract_info(ie_result['url'], | |
467 | download, | |
468 | ie_key=ie_result.get('ie_key'), | |
469 | extra_info=extra_info) | |
470 | elif result_type == 'playlist': | |
b6c45014 | 471 | self.add_extra_info(ie_result, extra_info) |
8222d8de JMF |
472 | # We process each entry in the playlist |
473 | playlist = ie_result.get('title', None) or ie_result.get('id', None) | |
fe7e0c98 | 474 | self.to_screen(u'[download] Downloading playlist: %s' % playlist) |
8222d8de JMF |
475 | |
476 | playlist_results = [] | |
477 | ||
478 | n_all_entries = len(ie_result['entries']) | |
479 | playliststart = self.params.get('playliststart', 1) - 1 | |
480 | playlistend = self.params.get('playlistend', -1) | |
481 | ||
482 | if playlistend == -1: | |
483 | entries = ie_result['entries'][playliststart:] | |
484 | else: | |
485 | entries = ie_result['entries'][playliststart:playlistend] | |
486 | ||
487 | n_entries = len(entries) | |
488 | ||
489 | self.to_screen(u"[%s] playlist '%s': Collected %d video ids (downloading %d of them)" % | |
490 | (ie_result['extractor'], playlist, n_all_entries, n_entries)) | |
491 | ||
fe7e0c98 JMF |
492 | for i, entry in enumerate(entries, 1): |
493 | self.to_screen(u'[download] Downloading video #%s of %s' % (i, n_entries)) | |
8222d8de | 494 | extra = { |
fe7e0c98 JMF |
495 | 'playlist': playlist, |
496 | 'playlist_index': i + playliststart, | |
b6c45014 | 497 | 'extractor': ie_result['extractor'], |
9103bbc5 | 498 | 'webpage_url': ie_result['webpage_url'], |
be97abc2 | 499 | 'extractor_key': ie_result['extractor_key'], |
fe7e0c98 | 500 | } |
8222d8de JMF |
501 | entry_result = self.process_ie_result(entry, |
502 | download=download, | |
503 | extra_info=extra) | |
504 | playlist_results.append(entry_result) | |
505 | ie_result['entries'] = playlist_results | |
506 | return ie_result | |
507 | elif result_type == 'compat_list': | |
508 | def _fixup(r): | |
b6c45014 | 509 | self.add_extra_info(r, |
9103bbc5 JMF |
510 | { |
511 | 'extractor': ie_result['extractor'], | |
512 | 'webpage_url': ie_result['webpage_url'], | |
be97abc2 | 513 | 'extractor_key': ie_result['extractor_key'], |
9103bbc5 | 514 | }) |
8222d8de JMF |
515 | return r |
516 | ie_result['entries'] = [ | |
b6c45014 | 517 | self.process_ie_result(_fixup(r), download, extra_info) |
8222d8de JMF |
518 | for r in ie_result['entries'] |
519 | ] | |
520 | return ie_result | |
521 | else: | |
522 | raise Exception('Invalid result type: %s' % result_type) | |
523 | ||
a9c58ad9 JMF |
524 | def select_format(self, format_spec, available_formats): |
525 | if format_spec == 'best' or format_spec is None: | |
526 | return available_formats[-1] | |
527 | elif format_spec == 'worst': | |
528 | return available_formats[0] | |
529 | else: | |
49e86983 JMF |
530 | extensions = [u'mp4', u'flv', u'webm', u'3gp'] |
531 | if format_spec in extensions: | |
532 | filter_f = lambda f: f['ext'] == format_spec | |
533 | else: | |
534 | filter_f = lambda f: f['format_id'] == format_spec | |
fe7e0c98 | 535 | matches = list(filter(filter_f, available_formats)) |
a9c58ad9 JMF |
536 | if matches: |
537 | return matches[-1] | |
538 | return None | |
539 | ||
dd82ffea JMF |
540 | def process_video_result(self, info_dict, download=True): |
541 | assert info_dict.get('_type', 'video') == 'video' | |
542 | ||
543 | if 'playlist' not in info_dict: | |
544 | # It isn't part of a playlist | |
545 | info_dict['playlist'] = None | |
546 | info_dict['playlist_index'] = None | |
547 | ||
6ff000b8 | 548 | # This extractors handle format selection themselves |
a7685f3b | 549 | if info_dict['extractor'] in [u'youtube', u'Youku']: |
12893efe JMF |
550 | if download: |
551 | self.process_info(info_dict) | |
6ff000b8 JMF |
552 | return info_dict |
553 | ||
dd82ffea JMF |
554 | # We now pick which formats have to be downloaded |
555 | if info_dict.get('formats') is None: | |
556 | # There's only one format available | |
557 | formats = [info_dict] | |
558 | else: | |
559 | formats = info_dict['formats'] | |
560 | ||
561 | # We check that all the formats have the format and format_id fields | |
562 | for (i, format) in enumerate(formats): | |
dd82ffea | 563 | if format.get('format_id') is None: |
8016c922 | 564 | format['format_id'] = compat_str(i) |
8c51aa65 JMF |
565 | if format.get('format') is None: |
566 | format['format'] = u'{id} - {res}{note}'.format( | |
567 | id=format['format_id'], | |
568 | res=self.format_resolution(format), | |
71934988 | 569 | note=u' ({0})'.format(format['format_note']) if format.get('format_note') is not None else '', |
8c51aa65 | 570 | ) |
c1002e96 PH |
571 | # Automatically determine file extension if missing |
572 | if 'ext' not in format: | |
573 | format['ext'] = determine_ext(format['url']) | |
dd82ffea JMF |
574 | |
575 | if self.params.get('listformats', None): | |
576 | self.list_formats(info_dict) | |
577 | return | |
578 | ||
99e206d5 JMF |
579 | format_limit = self.params.get('format_limit', None) |
580 | if format_limit: | |
f4d96df0 PH |
581 | formats = list(takewhile_inclusive( |
582 | lambda f: f['format_id'] != format_limit, formats | |
583 | )) | |
e028d0d1 JMF |
584 | if self.params.get('prefer_free_formats'): |
585 | def _free_formats_key(f): | |
586 | try: | |
587 | ext_ord = [u'flv', u'mp4', u'webm'].index(f['ext']) | |
588 | except ValueError: | |
589 | ext_ord = -1 | |
590 | # We only compare the extension if they have the same height and width | |
591 | return (f.get('height'), f.get('width'), ext_ord) | |
592 | formats = sorted(formats, key=_free_formats_key) | |
99e206d5 | 593 | |
dd82ffea | 594 | req_format = self.params.get('format', 'best') |
a9c58ad9 JMF |
595 | if req_format is None: |
596 | req_format = 'best' | |
dd82ffea | 597 | formats_to_download = [] |
dd82ffea | 598 | # The -1 is for supporting YoutubeIE |
a9c58ad9 | 599 | if req_format in ('-1', 'all'): |
dd82ffea JMF |
600 | formats_to_download = formats |
601 | else: | |
a9c58ad9 | 602 | # We can accept formats requestd in the format: 34/5/best, we pick |
416a5efc | 603 | # the first that is available, starting from left |
dd82ffea JMF |
604 | req_formats = req_format.split('/') |
605 | for rf in req_formats: | |
a9c58ad9 JMF |
606 | selected_format = self.select_format(rf, formats) |
607 | if selected_format is not None: | |
608 | formats_to_download = [selected_format] | |
dd82ffea JMF |
609 | break |
610 | if not formats_to_download: | |
78a3a9f8 PH |
611 | raise ExtractorError(u'requested format not available', |
612 | expected=True) | |
dd82ffea JMF |
613 | |
614 | if download: | |
615 | if len(formats_to_download) > 1: | |
616 | self.to_screen(u'[info] %s: downloading video in %s formats' % (info_dict['id'], len(formats_to_download))) | |
617 | for format in formats_to_download: | |
618 | new_info = dict(info_dict) | |
619 | new_info.update(format) | |
620 | self.process_info(new_info) | |
621 | # We update the info dict with the best quality format (backwards compatibility) | |
622 | info_dict.update(formats_to_download[-1]) | |
623 | return info_dict | |
624 | ||
8222d8de JMF |
625 | def process_info(self, info_dict): |
626 | """Process a single resolved IE result.""" | |
627 | ||
628 | assert info_dict.get('_type', 'video') == 'video' | |
629 | #We increment the download the download count here to match the previous behaviour. | |
630 | self.increment_downloads() | |
631 | ||
632 | info_dict['fulltitle'] = info_dict['title'] | |
633 | if len(info_dict['title']) > 200: | |
634 | info_dict['title'] = info_dict['title'][:197] + u'...' | |
635 | ||
636 | # Keep for backwards compatibility | |
637 | info_dict['stitle'] = info_dict['title'] | |
638 | ||
639 | if not 'format' in info_dict: | |
640 | info_dict['format'] = info_dict['ext'] | |
641 | ||
642 | reason = self._match_entry(info_dict) | |
643 | if reason is not None: | |
644 | self.to_screen(u'[download] ' + reason) | |
645 | return | |
646 | ||
647 | max_downloads = self.params.get('max_downloads') | |
648 | if max_downloads is not None: | |
649 | if self._num_downloads > int(max_downloads): | |
650 | raise MaxDownloadsReached() | |
651 | ||
652 | filename = self.prepare_filename(info_dict) | |
653 | ||
654 | # Forced printings | |
655 | if self.params.get('forcetitle', False): | |
656 | compat_print(info_dict['title']) | |
657 | if self.params.get('forceid', False): | |
658 | compat_print(info_dict['id']) | |
659 | if self.params.get('forceurl', False): | |
edde6c56 PH |
660 | # For RTMP URLs, also include the playpath |
661 | compat_print(info_dict['url'] + info_dict.get('play_path', u'')) | |
216d71d0 | 662 | if self.params.get('forcethumbnail', False) and info_dict.get('thumbnail') is not None: |
8222d8de | 663 | compat_print(info_dict['thumbnail']) |
216d71d0 | 664 | if self.params.get('forcedescription', False) and info_dict.get('description') is not None: |
8222d8de JMF |
665 | compat_print(info_dict['description']) |
666 | if self.params.get('forcefilename', False) and filename is not None: | |
667 | compat_print(filename) | |
668 | if self.params.get('forceformat', False): | |
669 | compat_print(info_dict['format']) | |
9d153818 MF |
670 | if self.params.get('forcejson', False): |
671 | compat_print(json.dumps(info_dict)) | |
8222d8de JMF |
672 | |
673 | # Do nothing else if in simulate mode | |
674 | if self.params.get('simulate', False): | |
675 | return | |
676 | ||
677 | if filename is None: | |
678 | return | |
679 | ||
680 | try: | |
681 | dn = os.path.dirname(encodeFilename(filename)) | |
682 | if dn != '' and not os.path.exists(dn): | |
683 | os.makedirs(dn) | |
684 | except (OSError, IOError) as err: | |
685 | self.report_error(u'unable to create directory ' + compat_str(err)) | |
686 | return | |
687 | ||
688 | if self.params.get('writedescription', False): | |
689 | try: | |
690 | descfn = filename + u'.description' | |
691 | self.report_writedescription(descfn) | |
692 | with io.open(encodeFilename(descfn), 'w', encoding='utf-8') as descfile: | |
693 | descfile.write(info_dict['description']) | |
b3f0e530 | 694 | except (KeyError, TypeError): |
535f59bb | 695 | self.report_warning(u'There\'s no description to write.') |
8222d8de JMF |
696 | except (OSError, IOError): |
697 | self.report_error(u'Cannot write description file ' + descfn) | |
698 | return | |
699 | ||
1fb07d10 JG |
700 | if self.params.get('writeannotations', False): |
701 | try: | |
fe7e0c98 JMF |
702 | annofn = filename + u'.annotations.xml' |
703 | self.report_writeannotations(annofn) | |
704 | with io.open(encodeFilename(annofn), 'w', encoding='utf-8') as annofile: | |
705 | annofile.write(info_dict['annotations']) | |
1fb07d10 JG |
706 | except (KeyError, TypeError): |
707 | self.report_warning(u'There are no annotations to write.') | |
708 | except (OSError, IOError): | |
fe7e0c98 JMF |
709 | self.report_error(u'Cannot write annotations file: ' + annofn) |
710 | return | |
1fb07d10 | 711 | |
c4a91be7 | 712 | subtitles_are_requested = any([self.params.get('writesubtitles', False), |
0b7f3118 | 713 | self.params.get('writeautomaticsub')]) |
c4a91be7 | 714 | |
fe7e0c98 | 715 | if subtitles_are_requested and 'subtitles' in info_dict and info_dict['subtitles']: |
8222d8de JMF |
716 | # subtitles download errors are already managed as troubles in relevant IE |
717 | # that way it will silently go on when used with unsupporting IE | |
8222d8de | 718 | subtitles = info_dict['subtitles'] |
ca715127 | 719 | sub_format = self.params.get('subtitlesformat', 'srt') |
5d51a883 JMF |
720 | for sub_lang in subtitles.keys(): |
721 | sub = subtitles[sub_lang] | |
6804038d JMF |
722 | if sub is None: |
723 | continue | |
8222d8de | 724 | try: |
d4051a8e | 725 | sub_filename = subtitles_filename(filename, sub_lang, sub_format) |
8222d8de JMF |
726 | self.report_writesubtitles(sub_filename) |
727 | with io.open(encodeFilename(sub_filename), 'w', encoding='utf-8') as subfile: | |
5d51a883 | 728 | subfile.write(sub) |
8222d8de JMF |
729 | except (OSError, IOError): |
730 | self.report_error(u'Cannot write subtitles file ' + descfn) | |
731 | return | |
732 | ||
8222d8de | 733 | if self.params.get('writeinfojson', False): |
9771cceb | 734 | infofn = os.path.splitext(filename)[0] + u'.info.json' |
8222d8de JMF |
735 | self.report_writeinfojson(infofn) |
736 | try: | |
fe7e0c98 | 737 | json_info_dict = dict((k, v) for k, v in info_dict.items() if not k in ['urlhandle']) |
8222d8de JMF |
738 | write_json_file(json_info_dict, encodeFilename(infofn)) |
739 | except (OSError, IOError): | |
740 | self.report_error(u'Cannot write metadata to JSON file ' + infofn) | |
741 | return | |
742 | ||
743 | if self.params.get('writethumbnail', False): | |
d8269e1d | 744 | if info_dict.get('thumbnail') is not None: |
cbdbb766 | 745 | thumb_format = determine_ext(info_dict['thumbnail'], u'jpg') |
8222d8de JMF |
746 | thumb_filename = filename.rpartition('.')[0] + u'.' + thumb_format |
747 | self.to_screen(u'[%s] %s: Downloading thumbnail ...' % | |
748 | (info_dict['extractor'], info_dict['id'])) | |
0a60edcf JMF |
749 | try: |
750 | uf = compat_urllib_request.urlopen(info_dict['thumbnail']) | |
751 | with open(thumb_filename, 'wb') as thumbf: | |
752 | shutil.copyfileobj(uf, thumbf) | |
753 | self.to_screen(u'[%s] %s: Writing thumbnail to: %s' % | |
754 | (info_dict['extractor'], info_dict['id'], thumb_filename)) | |
755 | except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: | |
756 | self.report_warning(u'Unable to download thumbnail "%s": %s' % | |
757 | (info_dict['thumbnail'], compat_str(err))) | |
8222d8de JMF |
758 | |
759 | if not self.params.get('skip_download', False): | |
760 | if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(filename)): | |
761 | success = True | |
762 | else: | |
763 | try: | |
764 | success = self.fd._do_download(filename, info_dict) | |
8222d8de JMF |
765 | except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: |
766 | self.report_error(u'unable to download video data: %s' % str(err)) | |
767 | return | |
c40c6aaa JMF |
768 | except (OSError, IOError) as err: |
769 | raise UnavailableVideoError(err) | |
8222d8de JMF |
770 | except (ContentTooShortError, ) as err: |
771 | self.report_error(u'content too short (expected %s bytes and served %s)' % (err.expected, err.downloaded)) | |
772 | return | |
773 | ||
774 | if success: | |
775 | try: | |
776 | self.post_process(filename, info_dict) | |
777 | except (PostProcessingError) as err: | |
778 | self.report_error(u'postprocessing: %s' % str(err)) | |
779 | return | |
780 | ||
c1c9a79c PH |
781 | self.record_download_archive(info_dict) |
782 | ||
8222d8de JMF |
783 | def download(self, url_list): |
784 | """Download a given list of URLs.""" | |
785 | if len(url_list) > 1 and self.fixed_template(): | |
786 | raise SameFileError(self.params['outtmpl']) | |
787 | ||
788 | for url in url_list: | |
789 | try: | |
790 | #It also downloads the videos | |
dca08720 | 791 | self.extract_info(url) |
8222d8de JMF |
792 | except UnavailableVideoError: |
793 | self.report_error(u'unable to download video') | |
794 | except MaxDownloadsReached: | |
795 | self.to_screen(u'[info] Maximum number of downloaded files reached.') | |
796 | raise | |
797 | ||
798 | return self._download_retcode | |
799 | ||
800 | def post_process(self, filename, ie_info): | |
801 | """Run all the postprocessors on the given file.""" | |
802 | info = dict(ie_info) | |
803 | info['filepath'] = filename | |
804 | keep_video = None | |
805 | for pp in self._pps: | |
806 | try: | |
fe7e0c98 | 807 | keep_video_wish, new_info = pp.run(info) |
8222d8de JMF |
808 | if keep_video_wish is not None: |
809 | if keep_video_wish: | |
810 | keep_video = keep_video_wish | |
811 | elif keep_video is None: | |
812 | # No clear decision yet, let IE decide | |
813 | keep_video = keep_video_wish | |
814 | except PostProcessingError as e: | |
bbcbf4d4 | 815 | self.report_error(e.msg) |
8222d8de JMF |
816 | if keep_video is False and not self.params.get('keepvideo', False): |
817 | try: | |
818 | self.to_screen(u'Deleting original file %s (pass -k to keep)' % filename) | |
819 | os.remove(encodeFilename(filename)) | |
820 | except (IOError, OSError): | |
821 | self.report_warning(u'Unable to remove downloaded video file') | |
c1c9a79c PH |
822 | |
823 | def in_download_archive(self, info_dict): | |
824 | fn = self.params.get('download_archive') | |
825 | if fn is None: | |
826 | return False | |
827 | vid_id = info_dict['extractor'] + u' ' + info_dict['id'] | |
828 | try: | |
829 | with locked_file(fn, 'r', encoding='utf-8') as archive_file: | |
830 | for line in archive_file: | |
831 | if line.strip() == vid_id: | |
832 | return True | |
833 | except IOError as ioe: | |
834 | if ioe.errno != errno.ENOENT: | |
835 | raise | |
836 | return False | |
837 | ||
838 | def record_download_archive(self, info_dict): | |
839 | fn = self.params.get('download_archive') | |
840 | if fn is None: | |
841 | return | |
842 | vid_id = info_dict['extractor'] + u' ' + info_dict['id'] | |
843 | with locked_file(fn, 'a', encoding='utf-8') as archive_file: | |
844 | archive_file.write(vid_id + u'\n') | |
dd82ffea | 845 | |
8c51aa65 | 846 | @staticmethod |
8abeeb94 | 847 | def format_resolution(format, default='unknown'): |
57dd9a8f PH |
848 | if format.get('_resolution') is not None: |
849 | return format['_resolution'] | |
8c51aa65 JMF |
850 | if format.get('height') is not None: |
851 | if format.get('width') is not None: | |
852 | res = u'%sx%s' % (format['width'], format['height']) | |
853 | else: | |
854 | res = u'%sp' % format['height'] | |
855 | else: | |
8abeeb94 | 856 | res = default |
8c51aa65 JMF |
857 | return res |
858 | ||
dd82ffea | 859 | def list_formats(self, info_dict): |
91c7271a PH |
860 | def format_note(fdict): |
861 | if fdict.get('format_note') is not None: | |
862 | return fdict['format_note'] | |
863 | res = u'' | |
864 | if fdict.get('vcodec') is not None: | |
7150858d PH |
865 | res += u'%-5s' % fdict['vcodec'] |
866 | elif fdict.get('vbr') is not None: | |
867 | res += u'video' | |
91c7271a PH |
868 | if fdict.get('vbr') is not None: |
869 | res += u'@%4dk' % fdict['vbr'] | |
870 | if fdict.get('acodec') is not None: | |
871 | if res: | |
872 | res += u', ' | |
7150858d PH |
873 | res += u'%-5s' % fdict['acodec'] |
874 | elif fdict.get('abr') is not None: | |
875 | if res: | |
876 | res += u', ' | |
877 | res += 'audio' | |
91c7271a PH |
878 | if fdict.get('abr') is not None: |
879 | res += u'@%3dk' % fdict['abr'] | |
880 | return res | |
881 | ||
57dd9a8f | 882 | def line(format): |
897d6cc4 | 883 | return (u'%-20s%-10s%-12s%s' % ( |
8c51aa65 JMF |
884 | format['format_id'], |
885 | format['ext'], | |
8c51aa65 | 886 | self.format_resolution(format), |
91c7271a | 887 | format_note(format), |
8c51aa65 JMF |
888 | ) |
889 | ) | |
57dd9a8f | 890 | |
94badb25 PH |
891 | formats = info_dict.get('formats', [info_dict]) |
892 | formats_s = list(map(line, formats)) | |
893 | if len(formats) > 1: | |
b5349e87 PH |
894 | formats_s[0] += (' ' if format_note(formats[0]) else '') + '(worst)' |
895 | formats_s[-1] += (' ' if format_note(formats[-1]) else '') + '(best)' | |
57dd9a8f PH |
896 | |
897 | header_line = line({ | |
898 | 'format_id': u'format code', 'ext': u'extension', | |
899 | '_resolution': u'resolution', 'format_note': u'note'}) | |
900 | self.to_screen(u'[info] Available formats for %s:\n%s\n%s' % | |
901 | (info_dict['id'], header_line, u"\n".join(formats_s))) | |
dca08720 PH |
902 | |
903 | def urlopen(self, req): | |
904 | """ Start an HTTP download """ | |
905 | return self._opener.open(req) | |
906 | ||
907 | def print_debug_header(self): | |
908 | if not self.params.get('verbose'): | |
909 | return | |
910 | write_string(u'[debug] youtube-dl version ' + __version__ + u'\n') | |
911 | try: | |
912 | sp = subprocess.Popen( | |
913 | ['git', 'rev-parse', '--short', 'HEAD'], | |
914 | stdout=subprocess.PIPE, stderr=subprocess.PIPE, | |
915 | cwd=os.path.dirname(os.path.abspath(__file__))) | |
916 | out, err = sp.communicate() | |
917 | out = out.decode().strip() | |
918 | if re.match('[0-9a-f]+', out): | |
919 | write_string(u'[debug] Git HEAD: ' + out + u'\n') | |
920 | except: | |
921 | try: | |
922 | sys.exc_clear() | |
923 | except: | |
924 | pass | |
925 | write_string(u'[debug] Python version %s - %s' % | |
926 | (platform.python_version(), platform_name()) + u'\n') | |
927 | ||
928 | proxy_map = {} | |
929 | for handler in self._opener.handlers: | |
930 | if hasattr(handler, 'proxies'): | |
931 | proxy_map.update(handler.proxies) | |
932 | write_string(u'[debug] Proxy map: ' + compat_str(proxy_map) + u'\n') | |
933 | ||
934 | def _setup_opener(self, timeout=300): | |
935 | opts_cookiefile = self.params.get('cookiefile') | |
936 | opts_proxy = self.params.get('proxy') | |
937 | ||
938 | if opts_cookiefile is None: | |
939 | self.cookiejar = compat_cookiejar.CookieJar() | |
940 | else: | |
941 | self.cookiejar = compat_cookiejar.MozillaCookieJar( | |
942 | opts_cookiefile) | |
943 | if os.access(opts_cookiefile, os.R_OK): | |
944 | self.cookiejar.load() | |
945 | ||
946 | cookie_processor = compat_urllib_request.HTTPCookieProcessor( | |
947 | self.cookiejar) | |
948 | if opts_proxy is not None: | |
949 | if opts_proxy == '': | |
950 | proxies = {} | |
951 | else: | |
952 | proxies = {'http': opts_proxy, 'https': opts_proxy} | |
953 | else: | |
954 | proxies = compat_urllib_request.getproxies() | |
955 | # Set HTTPS proxy to HTTP one if given (https://github.com/rg3/youtube-dl/issues/805) | |
956 | if 'http' in proxies and 'https' not in proxies: | |
957 | proxies['https'] = proxies['http'] | |
958 | proxy_handler = compat_urllib_request.ProxyHandler(proxies) | |
959 | https_handler = make_HTTPS_handler( | |
960 | self.params.get('nocheckcertificate', False)) | |
961 | opener = compat_urllib_request.build_opener( | |
962 | https_handler, proxy_handler, cookie_processor, YoutubeDLHandler()) | |
963 | # Delete the default user-agent header, which would otherwise apply in | |
964 | # cases where our custom HTTP handler doesn't come into play | |
965 | # (See https://github.com/rg3/youtube-dl/issues/1309 for details) | |
966 | opener.addheaders = [] | |
967 | self._opener = opener | |
968 | ||
969 | # TODO remove this global modification | |
970 | compat_urllib_request.install_opener(opener) | |
971 | socket.setdefaulttimeout(timeout) |