]>
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 JMF |
7 | import io |
8 | import os | |
9 | import re | |
10 | import shutil | |
11 | import socket | |
12 | import sys | |
13 | import time | |
14 | import traceback | |
15 | ||
16 | from .utils import * | |
023fa8c4 | 17 | from .extractor import get_info_extractor, gen_extractors |
8222d8de JMF |
18 | from .FileDownloader import FileDownloader |
19 | ||
20 | ||
21 | class YoutubeDL(object): | |
22 | """YoutubeDL class. | |
23 | ||
24 | YoutubeDL objects are the ones responsible of downloading the | |
25 | actual video file and writing it to disk if the user has requested | |
26 | it, among some other tasks. In most cases there should be one per | |
27 | program. As, given a video URL, the downloader doesn't know how to | |
28 | extract all the needed information, task that InfoExtractors do, it | |
29 | has to pass the URL to one of them. | |
30 | ||
31 | For this, YoutubeDL objects have a method that allows | |
32 | InfoExtractors to be registered in a given order. When it is passed | |
33 | a URL, the YoutubeDL object handles it to the first InfoExtractor it | |
34 | finds that reports being able to handle it. The InfoExtractor extracts | |
35 | all the information about the video or videos the URL refers to, and | |
36 | YoutubeDL process the extracted information, possibly using a File | |
37 | Downloader to download the video. | |
38 | ||
39 | YoutubeDL objects accept a lot of parameters. In order not to saturate | |
40 | the object constructor with arguments, it receives a dictionary of | |
41 | options instead. These options are available through the params | |
42 | attribute for the InfoExtractors to use. The YoutubeDL also | |
43 | registers itself as the downloader in charge for the InfoExtractors | |
44 | that are added to it, so this is a "mutual registration". | |
45 | ||
46 | Available options: | |
47 | ||
48 | username: Username for authentication purposes. | |
49 | password: Password for authentication purposes. | |
c6c19746 | 50 | videopassword: Password for acces a video. |
8222d8de JMF |
51 | usenetrc: Use netrc for authentication instead. |
52 | verbose: Print additional info to stdout. | |
53 | quiet: Do not print messages to stdout. | |
54 | forceurl: Force printing final URL. | |
55 | forcetitle: Force printing title. | |
56 | forceid: Force printing ID. | |
57 | forcethumbnail: Force printing thumbnail URL. | |
58 | forcedescription: Force printing description. | |
59 | forcefilename: Force printing final filename. | |
60 | simulate: Do not download the video files. | |
61 | format: Video format code. | |
62 | format_limit: Highest quality format to try. | |
63 | outtmpl: Template for output names. | |
64 | restrictfilenames: Do not allow "&" and spaces in file names | |
65 | ignoreerrors: Do not stop on download errors. | |
66 | nooverwrites: Prevent overwriting files. | |
67 | playliststart: Playlist item to start at. | |
68 | playlistend: Playlist item to end at. | |
69 | matchtitle: Download only matching titles. | |
70 | rejecttitle: Reject downloads for matching titles. | |
71 | logtostderr: Log messages to stderr instead of stdout. | |
72 | writedescription: Write the video description to a .description file | |
73 | writeinfojson: Write the video description to a .info.json file | |
74 | writethumbnail: Write the thumbnail image to a file | |
75 | writesubtitles: Write the video subtitles to a file | |
b004821f | 76 | writeautomaticsub: Write the automatic subtitles to a file |
8222d8de | 77 | allsubtitles: Downloads all the subtitles of the video |
0b7f3118 | 78 | (requires writesubtitles or writeautomaticsub) |
8222d8de | 79 | listsubtitles: Lists all available subtitles for the video |
b98a6b2f | 80 | subtitlesformat: Subtitle format [srt/sbv/vtt] (default=srt) |
aa6a10c4 | 81 | subtitleslangs: List of languages of the subtitles to download |
8222d8de JMF |
82 | keepvideo: Keep the video file after post-processing |
83 | daterange: A DateRange object, download only if the upload_date is in the range. | |
84 | skip_download: Skip the actual download of the video file | |
c35f9e72 | 85 | cachedir: Location of the cache files in the filesystem. |
c3c88a26 | 86 | None to disable filesystem cache. |
47192f92 | 87 | noplaylist: Download single video instead of a playlist if in doubt. |
8dbe9899 PH |
88 | age_limit: An integer representing the user's age in years. |
89 | Unsuitable videos for the given age are skipped. | |
c1c9a79c PH |
90 | downloadarchive: File name of a file where all downloads are recorded. |
91 | Videos already present in the file are not downloaded | |
92 | again. | |
8222d8de JMF |
93 | |
94 | The following parameters are not used by YoutubeDL itself, they are used by | |
95 | the FileDownloader: | |
96 | nopart, updatetime, buffersize, ratelimit, min_filesize, max_filesize, test, | |
97 | noresizebuffer, retries, continuedl, noprogress, consoletitle | |
98 | """ | |
99 | ||
100 | params = None | |
101 | _ies = [] | |
102 | _pps = [] | |
103 | _download_retcode = None | |
104 | _num_downloads = None | |
105 | _screen_file = None | |
106 | ||
107 | def __init__(self, params): | |
108 | """Create a FileDownloader object with the given options.""" | |
109 | self._ies = [] | |
56c73665 | 110 | self._ies_instances = {} |
8222d8de JMF |
111 | self._pps = [] |
112 | self._progress_hooks = [] | |
113 | self._download_retcode = 0 | |
114 | self._num_downloads = 0 | |
115 | self._screen_file = [sys.stdout, sys.stderr][params.get('logtostderr', False)] | |
34308b30 PH |
116 | |
117 | if (sys.version_info >= (3,) and sys.platform != 'win32' and | |
118 | sys.getfilesystemencoding() in ['ascii', 'ANSI_X3.4-1968'] | |
119 | and not params['restrictfilenames']): | |
120 | # On Python 3, the Unicode filesystem API will throw errors (#1474) | |
121 | self.report_warning( | |
1d368c75 | 122 | u'Assuming --restrict-filenames since file system encoding ' |
34308b30 PH |
123 | u'cannot encode all charactes. ' |
124 | u'Set the LC_ALL environment variable to fix this.') | |
125 | params['restrictfilenames'] = True | |
126 | ||
8222d8de JMF |
127 | self.params = params |
128 | self.fd = FileDownloader(self, self.params) | |
129 | ||
130 | if '%(stitle)s' in self.params['outtmpl']: | |
131 | 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.') | |
132 | ||
133 | def add_info_extractor(self, ie): | |
134 | """Add an InfoExtractor object to the end of the list.""" | |
135 | self._ies.append(ie) | |
56c73665 | 136 | self._ies_instances[ie.ie_key()] = ie |
8222d8de JMF |
137 | ie.set_downloader(self) |
138 | ||
56c73665 JMF |
139 | def get_info_extractor(self, ie_key): |
140 | """ | |
141 | Get an instance of an IE with name ie_key, it will try to get one from | |
142 | the _ies list, if there's no instance it will create a new one and add | |
143 | it to the extractor list. | |
144 | """ | |
145 | ie = self._ies_instances.get(ie_key) | |
146 | if ie is None: | |
147 | ie = get_info_extractor(ie_key)() | |
148 | self.add_info_extractor(ie) | |
149 | return ie | |
150 | ||
023fa8c4 JMF |
151 | def add_default_info_extractors(self): |
152 | """ | |
153 | Add the InfoExtractors returned by gen_extractors to the end of the list | |
154 | """ | |
155 | for ie in gen_extractors(): | |
156 | self.add_info_extractor(ie) | |
157 | ||
8222d8de JMF |
158 | def add_post_processor(self, pp): |
159 | """Add a PostProcessor object to the end of the chain.""" | |
160 | self._pps.append(pp) | |
161 | pp.set_downloader(self) | |
162 | ||
163 | def to_screen(self, message, skip_eol=False): | |
164 | """Print message to stdout if not in quiet mode.""" | |
8222d8de JMF |
165 | if not self.params.get('quiet', False): |
166 | terminator = [u'\n', u''][skip_eol] | |
167 | output = message + terminator | |
7459e3a2 | 168 | write_string(output, self._screen_file) |
8222d8de JMF |
169 | |
170 | def to_stderr(self, message): | |
171 | """Print message to stderr.""" | |
172 | assert type(message) == type(u'') | |
173 | output = message + u'\n' | |
174 | 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 | |
175 | output = output.encode(preferredencoding()) | |
176 | sys.stderr.write(output) | |
177 | ||
178 | def fixed_template(self): | |
179 | """Checks if the output template is fixed.""" | |
180 | return (re.search(u'(?u)%\\(.+?\\)s', self.params['outtmpl']) is None) | |
181 | ||
182 | def trouble(self, message=None, tb=None): | |
183 | """Determine action to take when a download problem appears. | |
184 | ||
185 | Depending on if the downloader has been configured to ignore | |
186 | download errors or not, this method may throw an exception or | |
187 | not when errors are found, after printing the message. | |
188 | ||
189 | tb, if given, is additional traceback information. | |
190 | """ | |
191 | if message is not None: | |
192 | self.to_stderr(message) | |
193 | if self.params.get('verbose'): | |
194 | if tb is None: | |
195 | if sys.exc_info()[0]: # if .trouble has been called from an except block | |
196 | tb = u'' | |
197 | if hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]: | |
198 | tb += u''.join(traceback.format_exception(*sys.exc_info()[1].exc_info)) | |
199 | tb += compat_str(traceback.format_exc()) | |
200 | else: | |
201 | tb_data = traceback.format_list(traceback.extract_stack()) | |
202 | tb = u''.join(tb_data) | |
203 | self.to_stderr(tb) | |
204 | if not self.params.get('ignoreerrors', False): | |
205 | if sys.exc_info()[0] and hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]: | |
206 | exc_info = sys.exc_info()[1].exc_info | |
207 | else: | |
208 | exc_info = sys.exc_info() | |
209 | raise DownloadError(message, exc_info) | |
210 | self._download_retcode = 1 | |
211 | ||
212 | def report_warning(self, message): | |
213 | ''' | |
214 | Print the message to stderr, it will be prefixed with 'WARNING:' | |
215 | If stderr is a tty file the 'WARNING:' will be colored | |
216 | ''' | |
217 | if sys.stderr.isatty() and os.name != 'nt': | |
218 | _msg_header=u'\033[0;33mWARNING:\033[0m' | |
219 | else: | |
220 | _msg_header=u'WARNING:' | |
221 | warning_message=u'%s %s' % (_msg_header,message) | |
222 | self.to_stderr(warning_message) | |
223 | ||
224 | def report_error(self, message, tb=None): | |
225 | ''' | |
226 | Do the same as trouble, but prefixes the message with 'ERROR:', colored | |
227 | in red if stderr is a tty file. | |
228 | ''' | |
229 | if sys.stderr.isatty() and os.name != 'nt': | |
230 | _msg_header = u'\033[0;31mERROR:\033[0m' | |
231 | else: | |
232 | _msg_header = u'ERROR:' | |
233 | error_message = u'%s %s' % (_msg_header, message) | |
234 | self.trouble(error_message, tb) | |
235 | ||
236 | def slow_down(self, start_time, byte_counter): | |
237 | """Sleep if the download speed is over the rate limit.""" | |
238 | rate_limit = self.params.get('ratelimit', None) | |
239 | if rate_limit is None or byte_counter == 0: | |
240 | return | |
241 | now = time.time() | |
242 | elapsed = now - start_time | |
243 | if elapsed <= 0.0: | |
244 | return | |
245 | speed = float(byte_counter) / elapsed | |
246 | if speed > rate_limit: | |
247 | time.sleep((byte_counter - rate_limit * (now - start_time)) / rate_limit) | |
248 | ||
249 | def report_writedescription(self, descfn): | |
250 | """ Report that the description file is being written """ | |
251 | self.to_screen(u'[info] Writing video description to: ' + descfn) | |
252 | ||
253 | def report_writesubtitles(self, sub_filename): | |
254 | """ Report that the subtitles file is being written """ | |
255 | self.to_screen(u'[info] Writing video subtitles to: ' + sub_filename) | |
256 | ||
257 | def report_writeinfojson(self, infofn): | |
258 | """ Report that the metadata file has been written """ | |
259 | self.to_screen(u'[info] Video description metadata as JSON to: ' + infofn) | |
260 | ||
261 | def report_file_already_downloaded(self, file_name): | |
262 | """Report file has already been fully downloaded.""" | |
263 | try: | |
264 | self.to_screen(u'[download] %s has already been downloaded' % file_name) | |
265 | except (UnicodeEncodeError) as err: | |
266 | self.to_screen(u'[download] The file has already been downloaded') | |
267 | ||
268 | def increment_downloads(self): | |
269 | """Increment the ordinal that assigns a number to each file.""" | |
270 | self._num_downloads += 1 | |
271 | ||
272 | def prepare_filename(self, info_dict): | |
273 | """Generate the output filename.""" | |
274 | try: | |
275 | template_dict = dict(info_dict) | |
276 | ||
277 | template_dict['epoch'] = int(time.time()) | |
278 | autonumber_size = self.params.get('autonumber_size') | |
279 | if autonumber_size is None: | |
280 | autonumber_size = 5 | |
281 | autonumber_templ = u'%0' + str(autonumber_size) + u'd' | |
282 | template_dict['autonumber'] = autonumber_templ % self._num_downloads | |
283 | if template_dict['playlist_index'] is not None: | |
284 | template_dict['playlist_index'] = u'%05d' % template_dict['playlist_index'] | |
285 | ||
286 | sanitize = lambda k,v: sanitize_filename( | |
287 | u'NA' if v is None else compat_str(v), | |
288 | restricted=self.params.get('restrictfilenames'), | |
289 | is_id=(k==u'id')) | |
290 | template_dict = dict((k, sanitize(k, v)) for k,v in template_dict.items()) | |
291 | ||
292 | filename = self.params['outtmpl'] % template_dict | |
293 | return filename | |
294 | except KeyError as err: | |
295 | self.report_error(u'Erroneous output template') | |
296 | return None | |
297 | except ValueError as err: | |
4efba05c | 298 | self.report_error(u'Error in output template: ' + str(err) + u' (encoding: ' + repr(preferredencoding()) + ')') |
8222d8de JMF |
299 | return None |
300 | ||
301 | def _match_entry(self, info_dict): | |
302 | """ Returns None iff the file should be downloaded """ | |
303 | ||
304 | title = info_dict['title'] | |
305 | matchtitle = self.params.get('matchtitle', False) | |
306 | if matchtitle: | |
307 | if not re.search(matchtitle, title, re.IGNORECASE): | |
308 | return u'[download] "' + title + '" title did not match pattern "' + matchtitle + '"' | |
309 | rejecttitle = self.params.get('rejecttitle', False) | |
310 | if rejecttitle: | |
311 | if re.search(rejecttitle, title, re.IGNORECASE): | |
312 | return u'"' + title + '" title matched reject pattern "' + rejecttitle + '"' | |
313 | date = info_dict.get('upload_date', None) | |
314 | if date is not None: | |
315 | dateRange = self.params.get('daterange', DateRange()) | |
316 | if date not in dateRange: | |
317 | return u'[download] %s upload date is not in range %s' % (date_from_str(date).isoformat(), dateRange) | |
8dbe9899 PH |
318 | age_limit = self.params.get('age_limit') |
319 | if age_limit is not None: | |
cfadd183 | 320 | if age_limit < info_dict.get('age_limit', 0): |
8dbe9899 | 321 | return u'Skipping "' + title + '" because it is age restricted' |
c1c9a79c | 322 | if self.in_download_archive(info_dict): |
ee6c9f95 | 323 | return (u'%(title)s has already been recorded in archive' |
c1c9a79c | 324 | % info_dict) |
8222d8de JMF |
325 | return None |
326 | ||
327 | def extract_info(self, url, download=True, ie_key=None, extra_info={}): | |
328 | ''' | |
329 | Returns a list with a dictionary for each video we find. | |
330 | If 'download', also downloads the videos. | |
331 | extra_info is a dict containing the extra values to add to each result | |
332 | ''' | |
333 | ||
334 | if ie_key: | |
56c73665 | 335 | ies = [self.get_info_extractor(ie_key)] |
8222d8de JMF |
336 | else: |
337 | ies = self._ies | |
338 | ||
339 | for ie in ies: | |
340 | if not ie.suitable(url): | |
341 | continue | |
342 | ||
343 | if not ie.working(): | |
344 | self.report_warning(u'The program functionality for this site has been marked as broken, ' | |
345 | u'and will probably not work.') | |
346 | ||
347 | try: | |
348 | ie_result = ie.extract(url) | |
349 | if ie_result is None: # Finished already (backwards compatibility; listformats and friends should be moved here) | |
350 | break | |
351 | if isinstance(ie_result, list): | |
352 | # Backwards compatibility: old IE result format | |
353 | for result in ie_result: | |
354 | result.update(extra_info) | |
355 | ie_result = { | |
356 | '_type': 'compat_list', | |
357 | 'entries': ie_result, | |
358 | } | |
359 | else: | |
360 | ie_result.update(extra_info) | |
361 | if 'extractor' not in ie_result: | |
362 | ie_result['extractor'] = ie.IE_NAME | |
363 | return self.process_ie_result(ie_result, download=download) | |
364 | except ExtractorError as de: # An error we somewhat expected | |
365 | self.report_error(compat_str(de), de.format_traceback()) | |
366 | break | |
367 | except Exception as e: | |
368 | if self.params.get('ignoreerrors', False): | |
369 | self.report_error(compat_str(e), tb=compat_str(traceback.format_exc())) | |
370 | break | |
371 | else: | |
372 | raise | |
373 | else: | |
374 | self.report_error(u'no suitable InfoExtractor: %s' % url) | |
375 | ||
376 | def process_ie_result(self, ie_result, download=True, extra_info={}): | |
377 | """ | |
378 | Take the result of the ie(may be modified) and resolve all unresolved | |
379 | references (URLs, playlist items). | |
380 | ||
381 | It will also download the videos if 'download'. | |
382 | Returns the resolved ie_result. | |
383 | """ | |
384 | ||
385 | result_type = ie_result.get('_type', 'video') # If not given we suppose it's a video, support the default old system | |
386 | if result_type == 'video': | |
a4311547 | 387 | ie_result.update(extra_info) |
8222d8de JMF |
388 | if 'playlist' not in ie_result: |
389 | # It isn't part of a playlist | |
390 | ie_result['playlist'] = None | |
391 | ie_result['playlist_index'] = None | |
392 | if download: | |
393 | self.process_info(ie_result) | |
394 | return ie_result | |
395 | elif result_type == 'url': | |
396 | # We have to add extra_info to the results because it may be | |
397 | # contained in a playlist | |
398 | return self.extract_info(ie_result['url'], | |
399 | download, | |
400 | ie_key=ie_result.get('ie_key'), | |
401 | extra_info=extra_info) | |
402 | elif result_type == 'playlist': | |
403 | # We process each entry in the playlist | |
404 | playlist = ie_result.get('title', None) or ie_result.get('id', None) | |
405 | self.to_screen(u'[download] Downloading playlist: %s' % playlist) | |
406 | ||
407 | playlist_results = [] | |
408 | ||
409 | n_all_entries = len(ie_result['entries']) | |
410 | playliststart = self.params.get('playliststart', 1) - 1 | |
411 | playlistend = self.params.get('playlistend', -1) | |
412 | ||
413 | if playlistend == -1: | |
414 | entries = ie_result['entries'][playliststart:] | |
415 | else: | |
416 | entries = ie_result['entries'][playliststart:playlistend] | |
417 | ||
418 | n_entries = len(entries) | |
419 | ||
420 | self.to_screen(u"[%s] playlist '%s': Collected %d video ids (downloading %d of them)" % | |
421 | (ie_result['extractor'], playlist, n_all_entries, n_entries)) | |
422 | ||
423 | for i,entry in enumerate(entries,1): | |
424 | self.to_screen(u'[download] Downloading video #%s of %s' %(i, n_entries)) | |
425 | extra = { | |
426 | 'playlist': playlist, | |
427 | 'playlist_index': i + playliststart, | |
428 | } | |
429 | if not 'extractor' in entry: | |
430 | # We set the extractor, if it's an url it will be set then to | |
431 | # the new extractor, but if it's already a video we must make | |
432 | # sure it's present: see issue #877 | |
433 | entry['extractor'] = ie_result['extractor'] | |
434 | entry_result = self.process_ie_result(entry, | |
435 | download=download, | |
436 | extra_info=extra) | |
437 | playlist_results.append(entry_result) | |
438 | ie_result['entries'] = playlist_results | |
439 | return ie_result | |
440 | elif result_type == 'compat_list': | |
441 | def _fixup(r): | |
442 | r.setdefault('extractor', ie_result['extractor']) | |
443 | return r | |
444 | ie_result['entries'] = [ | |
445 | self.process_ie_result(_fixup(r), download=download) | |
446 | for r in ie_result['entries'] | |
447 | ] | |
448 | return ie_result | |
449 | else: | |
450 | raise Exception('Invalid result type: %s' % result_type) | |
451 | ||
452 | def process_info(self, info_dict): | |
453 | """Process a single resolved IE result.""" | |
454 | ||
455 | assert info_dict.get('_type', 'video') == 'video' | |
456 | #We increment the download the download count here to match the previous behaviour. | |
457 | self.increment_downloads() | |
458 | ||
459 | info_dict['fulltitle'] = info_dict['title'] | |
460 | if len(info_dict['title']) > 200: | |
461 | info_dict['title'] = info_dict['title'][:197] + u'...' | |
462 | ||
463 | # Keep for backwards compatibility | |
464 | info_dict['stitle'] = info_dict['title'] | |
465 | ||
466 | if not 'format' in info_dict: | |
467 | info_dict['format'] = info_dict['ext'] | |
468 | ||
469 | reason = self._match_entry(info_dict) | |
470 | if reason is not None: | |
471 | self.to_screen(u'[download] ' + reason) | |
472 | return | |
473 | ||
474 | max_downloads = self.params.get('max_downloads') | |
475 | if max_downloads is not None: | |
476 | if self._num_downloads > int(max_downloads): | |
477 | raise MaxDownloadsReached() | |
478 | ||
479 | filename = self.prepare_filename(info_dict) | |
480 | ||
481 | # Forced printings | |
482 | if self.params.get('forcetitle', False): | |
483 | compat_print(info_dict['title']) | |
484 | if self.params.get('forceid', False): | |
485 | compat_print(info_dict['id']) | |
486 | if self.params.get('forceurl', False): | |
edde6c56 PH |
487 | # For RTMP URLs, also include the playpath |
488 | compat_print(info_dict['url'] + info_dict.get('play_path', u'')) | |
8222d8de JMF |
489 | if self.params.get('forcethumbnail', False) and 'thumbnail' in info_dict: |
490 | compat_print(info_dict['thumbnail']) | |
491 | if self.params.get('forcedescription', False) and 'description' in info_dict: | |
492 | compat_print(info_dict['description']) | |
493 | if self.params.get('forcefilename', False) and filename is not None: | |
494 | compat_print(filename) | |
495 | if self.params.get('forceformat', False): | |
496 | compat_print(info_dict['format']) | |
497 | ||
498 | # Do nothing else if in simulate mode | |
499 | if self.params.get('simulate', False): | |
500 | return | |
501 | ||
502 | if filename is None: | |
503 | return | |
504 | ||
505 | try: | |
506 | dn = os.path.dirname(encodeFilename(filename)) | |
507 | if dn != '' and not os.path.exists(dn): | |
508 | os.makedirs(dn) | |
509 | except (OSError, IOError) as err: | |
510 | self.report_error(u'unable to create directory ' + compat_str(err)) | |
511 | return | |
512 | ||
513 | if self.params.get('writedescription', False): | |
514 | try: | |
515 | descfn = filename + u'.description' | |
516 | self.report_writedescription(descfn) | |
517 | with io.open(encodeFilename(descfn), 'w', encoding='utf-8') as descfile: | |
518 | descfile.write(info_dict['description']) | |
b3f0e530 | 519 | except (KeyError, TypeError): |
535f59bb | 520 | self.report_warning(u'There\'s no description to write.') |
8222d8de JMF |
521 | except (OSError, IOError): |
522 | self.report_error(u'Cannot write description file ' + descfn) | |
523 | return | |
524 | ||
c4a91be7 | 525 | subtitles_are_requested = any([self.params.get('writesubtitles', False), |
0b7f3118 | 526 | self.params.get('writeautomaticsub')]) |
c4a91be7 JMF |
527 | |
528 | if subtitles_are_requested and 'subtitles' in info_dict and info_dict['subtitles']: | |
8222d8de JMF |
529 | # subtitles download errors are already managed as troubles in relevant IE |
530 | # that way it will silently go on when used with unsupporting IE | |
8222d8de | 531 | subtitles = info_dict['subtitles'] |
8222d8de | 532 | sub_format = self.params.get('subtitlesformat') |
5d51a883 JMF |
533 | for sub_lang in subtitles.keys(): |
534 | sub = subtitles[sub_lang] | |
6804038d JMF |
535 | if sub is None: |
536 | continue | |
8222d8de | 537 | try: |
d4051a8e | 538 | sub_filename = subtitles_filename(filename, sub_lang, sub_format) |
8222d8de JMF |
539 | self.report_writesubtitles(sub_filename) |
540 | with io.open(encodeFilename(sub_filename), 'w', encoding='utf-8') as subfile: | |
5d51a883 | 541 | subfile.write(sub) |
8222d8de JMF |
542 | except (OSError, IOError): |
543 | self.report_error(u'Cannot write subtitles file ' + descfn) | |
544 | return | |
545 | ||
8222d8de JMF |
546 | if self.params.get('writeinfojson', False): |
547 | infofn = filename + u'.info.json' | |
548 | self.report_writeinfojson(infofn) | |
549 | try: | |
550 | json_info_dict = dict((k, v) for k,v in info_dict.items() if not k in ['urlhandle']) | |
551 | write_json_file(json_info_dict, encodeFilename(infofn)) | |
552 | except (OSError, IOError): | |
553 | self.report_error(u'Cannot write metadata to JSON file ' + infofn) | |
554 | return | |
555 | ||
556 | if self.params.get('writethumbnail', False): | |
d8269e1d | 557 | if info_dict.get('thumbnail') is not None: |
cbdbb766 | 558 | thumb_format = determine_ext(info_dict['thumbnail'], u'jpg') |
8222d8de JMF |
559 | thumb_filename = filename.rpartition('.')[0] + u'.' + thumb_format |
560 | self.to_screen(u'[%s] %s: Downloading thumbnail ...' % | |
561 | (info_dict['extractor'], info_dict['id'])) | |
0a60edcf JMF |
562 | try: |
563 | uf = compat_urllib_request.urlopen(info_dict['thumbnail']) | |
564 | with open(thumb_filename, 'wb') as thumbf: | |
565 | shutil.copyfileobj(uf, thumbf) | |
566 | self.to_screen(u'[%s] %s: Writing thumbnail to: %s' % | |
567 | (info_dict['extractor'], info_dict['id'], thumb_filename)) | |
568 | except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: | |
569 | self.report_warning(u'Unable to download thumbnail "%s": %s' % | |
570 | (info_dict['thumbnail'], compat_str(err))) | |
8222d8de JMF |
571 | |
572 | if not self.params.get('skip_download', False): | |
573 | if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(filename)): | |
574 | success = True | |
575 | else: | |
576 | try: | |
577 | success = self.fd._do_download(filename, info_dict) | |
8222d8de JMF |
578 | except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: |
579 | self.report_error(u'unable to download video data: %s' % str(err)) | |
580 | return | |
c40c6aaa JMF |
581 | except (OSError, IOError) as err: |
582 | raise UnavailableVideoError(err) | |
8222d8de JMF |
583 | except (ContentTooShortError, ) as err: |
584 | self.report_error(u'content too short (expected %s bytes and served %s)' % (err.expected, err.downloaded)) | |
585 | return | |
586 | ||
587 | if success: | |
588 | try: | |
589 | self.post_process(filename, info_dict) | |
590 | except (PostProcessingError) as err: | |
591 | self.report_error(u'postprocessing: %s' % str(err)) | |
592 | return | |
593 | ||
c1c9a79c PH |
594 | self.record_download_archive(info_dict) |
595 | ||
8222d8de JMF |
596 | def download(self, url_list): |
597 | """Download a given list of URLs.""" | |
598 | if len(url_list) > 1 and self.fixed_template(): | |
599 | raise SameFileError(self.params['outtmpl']) | |
600 | ||
601 | for url in url_list: | |
602 | try: | |
603 | #It also downloads the videos | |
604 | videos = self.extract_info(url) | |
605 | except UnavailableVideoError: | |
606 | self.report_error(u'unable to download video') | |
607 | except MaxDownloadsReached: | |
608 | self.to_screen(u'[info] Maximum number of downloaded files reached.') | |
609 | raise | |
610 | ||
611 | return self._download_retcode | |
612 | ||
613 | def post_process(self, filename, ie_info): | |
614 | """Run all the postprocessors on the given file.""" | |
615 | info = dict(ie_info) | |
616 | info['filepath'] = filename | |
617 | keep_video = None | |
618 | for pp in self._pps: | |
619 | try: | |
620 | keep_video_wish,new_info = pp.run(info) | |
621 | if keep_video_wish is not None: | |
622 | if keep_video_wish: | |
623 | keep_video = keep_video_wish | |
624 | elif keep_video is None: | |
625 | # No clear decision yet, let IE decide | |
626 | keep_video = keep_video_wish | |
627 | except PostProcessingError as e: | |
bbcbf4d4 | 628 | self.report_error(e.msg) |
8222d8de JMF |
629 | if keep_video is False and not self.params.get('keepvideo', False): |
630 | try: | |
631 | self.to_screen(u'Deleting original file %s (pass -k to keep)' % filename) | |
632 | os.remove(encodeFilename(filename)) | |
633 | except (IOError, OSError): | |
634 | self.report_warning(u'Unable to remove downloaded video file') | |
c1c9a79c PH |
635 | |
636 | def in_download_archive(self, info_dict): | |
637 | fn = self.params.get('download_archive') | |
638 | if fn is None: | |
639 | return False | |
640 | vid_id = info_dict['extractor'] + u' ' + info_dict['id'] | |
641 | try: | |
642 | with locked_file(fn, 'r', encoding='utf-8') as archive_file: | |
643 | for line in archive_file: | |
644 | if line.strip() == vid_id: | |
645 | return True | |
646 | except IOError as ioe: | |
647 | if ioe.errno != errno.ENOENT: | |
648 | raise | |
649 | return False | |
650 | ||
651 | def record_download_archive(self, info_dict): | |
652 | fn = self.params.get('download_archive') | |
653 | if fn is None: | |
654 | return | |
655 | vid_id = info_dict['extractor'] + u' ' + info_dict['id'] | |
656 | with locked_file(fn, 'a', encoding='utf-8') as archive_file: | |
657 | archive_file.write(vid_id + u'\n') |