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