]>
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 JMF |
9 | import os |
10 | import re | |
11 | import shutil | |
12 | import socket | |
13 | import sys | |
14 | import time | |
15 | import traceback | |
16 | ||
1e5b9a95 PH |
17 | if os.name == 'nt': |
18 | import ctypes | |
19 | ||
ce02ed60 PH |
20 | from .utils import ( |
21 | compat_http_client, | |
22 | compat_print, | |
23 | compat_str, | |
24 | compat_urllib_error, | |
25 | compat_urllib_request, | |
26 | ContentTooShortError, | |
27 | date_from_str, | |
28 | DateRange, | |
29 | determine_ext, | |
30 | DownloadError, | |
31 | encodeFilename, | |
32 | ExtractorError, | |
33 | locked_file, | |
34 | MaxDownloadsReached, | |
35 | PostProcessingError, | |
36 | preferredencoding, | |
37 | SameFileError, | |
38 | sanitize_filename, | |
39 | subtitles_filename, | |
40 | takewhile_inclusive, | |
41 | UnavailableVideoError, | |
42 | write_json_file, | |
43 | write_string, | |
44 | ) | |
023fa8c4 | 45 | from .extractor import get_info_extractor, gen_extractors |
8222d8de JMF |
46 | from .FileDownloader import FileDownloader |
47 | ||
48 | ||
49 | class YoutubeDL(object): | |
50 | """YoutubeDL class. | |
51 | ||
52 | YoutubeDL objects are the ones responsible of downloading the | |
53 | actual video file and writing it to disk if the user has requested | |
54 | it, among some other tasks. In most cases there should be one per | |
55 | program. As, given a video URL, the downloader doesn't know how to | |
56 | extract all the needed information, task that InfoExtractors do, it | |
57 | has to pass the URL to one of them. | |
58 | ||
59 | For this, YoutubeDL objects have a method that allows | |
60 | InfoExtractors to be registered in a given order. When it is passed | |
61 | a URL, the YoutubeDL object handles it to the first InfoExtractor it | |
62 | finds that reports being able to handle it. The InfoExtractor extracts | |
63 | all the information about the video or videos the URL refers to, and | |
64 | YoutubeDL process the extracted information, possibly using a File | |
65 | Downloader to download the video. | |
66 | ||
67 | YoutubeDL objects accept a lot of parameters. In order not to saturate | |
68 | the object constructor with arguments, it receives a dictionary of | |
69 | options instead. These options are available through the params | |
70 | attribute for the InfoExtractors to use. The YoutubeDL also | |
71 | registers itself as the downloader in charge for the InfoExtractors | |
72 | that are added to it, so this is a "mutual registration". | |
73 | ||
74 | Available options: | |
75 | ||
76 | username: Username for authentication purposes. | |
77 | password: Password for authentication purposes. | |
c6c19746 | 78 | videopassword: Password for acces a video. |
8222d8de JMF |
79 | usenetrc: Use netrc for authentication instead. |
80 | verbose: Print additional info to stdout. | |
81 | quiet: Do not print messages to stdout. | |
82 | forceurl: Force printing final URL. | |
83 | forcetitle: Force printing title. | |
84 | forceid: Force printing ID. | |
85 | forcethumbnail: Force printing thumbnail URL. | |
86 | forcedescription: Force printing description. | |
87 | forcefilename: Force printing final filename. | |
8694c600 | 88 | forcejson: Force printing info_dict as JSON. |
8222d8de JMF |
89 | simulate: Do not download the video files. |
90 | format: Video format code. | |
91 | format_limit: Highest quality format to try. | |
92 | outtmpl: Template for output names. | |
93 | restrictfilenames: Do not allow "&" and spaces in file names | |
94 | ignoreerrors: Do not stop on download errors. | |
95 | nooverwrites: Prevent overwriting files. | |
96 | playliststart: Playlist item to start at. | |
97 | playlistend: Playlist item to end at. | |
98 | matchtitle: Download only matching titles. | |
99 | rejecttitle: Reject downloads for matching titles. | |
100 | logtostderr: Log messages to stderr instead of stdout. | |
101 | writedescription: Write the video description to a .description file | |
102 | writeinfojson: Write the video description to a .info.json file | |
1fb07d10 | 103 | writeannotations: Write the video annotations to a .annotations.xml file |
8222d8de JMF |
104 | writethumbnail: Write the thumbnail image to a file |
105 | writesubtitles: Write the video subtitles to a file | |
b004821f | 106 | writeautomaticsub: Write the automatic subtitles to a file |
8222d8de | 107 | allsubtitles: Downloads all the subtitles of the video |
0b7f3118 | 108 | (requires writesubtitles or writeautomaticsub) |
8222d8de | 109 | listsubtitles: Lists all available subtitles for the video |
b98a6b2f | 110 | subtitlesformat: Subtitle format [srt/sbv/vtt] (default=srt) |
aa6a10c4 | 111 | subtitleslangs: List of languages of the subtitles to download |
8222d8de JMF |
112 | keepvideo: Keep the video file after post-processing |
113 | daterange: A DateRange object, download only if the upload_date is in the range. | |
114 | skip_download: Skip the actual download of the video file | |
c35f9e72 | 115 | cachedir: Location of the cache files in the filesystem. |
c3c88a26 | 116 | None to disable filesystem cache. |
47192f92 | 117 | noplaylist: Download single video instead of a playlist if in doubt. |
8dbe9899 PH |
118 | age_limit: An integer representing the user's age in years. |
119 | Unsuitable videos for the given age are skipped. | |
c1c9a79c PH |
120 | downloadarchive: File name of a file where all downloads are recorded. |
121 | Videos already present in the file are not downloaded | |
122 | again. | |
fe7e0c98 | 123 | |
8222d8de JMF |
124 | The following parameters are not used by YoutubeDL itself, they are used by |
125 | the FileDownloader: | |
126 | nopart, updatetime, buffersize, ratelimit, min_filesize, max_filesize, test, | |
127 | noresizebuffer, retries, continuedl, noprogress, consoletitle | |
128 | """ | |
129 | ||
130 | params = None | |
131 | _ies = [] | |
132 | _pps = [] | |
133 | _download_retcode = None | |
134 | _num_downloads = None | |
135 | _screen_file = None | |
136 | ||
137 | def __init__(self, params): | |
138 | """Create a FileDownloader object with the given options.""" | |
139 | self._ies = [] | |
56c73665 | 140 | self._ies_instances = {} |
8222d8de JMF |
141 | self._pps = [] |
142 | self._progress_hooks = [] | |
143 | self._download_retcode = 0 | |
144 | self._num_downloads = 0 | |
145 | self._screen_file = [sys.stdout, sys.stderr][params.get('logtostderr', False)] | |
34308b30 PH |
146 | |
147 | if (sys.version_info >= (3,) and sys.platform != 'win32' and | |
148 | sys.getfilesystemencoding() in ['ascii', 'ANSI_X3.4-1968'] | |
149 | and not params['restrictfilenames']): | |
150 | # On Python 3, the Unicode filesystem API will throw errors (#1474) | |
151 | self.report_warning( | |
1d368c75 | 152 | u'Assuming --restrict-filenames since file system encoding ' |
34308b30 PH |
153 | u'cannot encode all charactes. ' |
154 | u'Set the LC_ALL environment variable to fix this.') | |
155 | params['restrictfilenames'] = True | |
156 | ||
8222d8de JMF |
157 | self.params = params |
158 | self.fd = FileDownloader(self, self.params) | |
159 | ||
160 | if '%(stitle)s' in self.params['outtmpl']: | |
161 | 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.') | |
162 | ||
163 | def add_info_extractor(self, ie): | |
164 | """Add an InfoExtractor object to the end of the list.""" | |
165 | self._ies.append(ie) | |
56c73665 | 166 | self._ies_instances[ie.ie_key()] = ie |
8222d8de JMF |
167 | ie.set_downloader(self) |
168 | ||
56c73665 JMF |
169 | def get_info_extractor(self, ie_key): |
170 | """ | |
171 | Get an instance of an IE with name ie_key, it will try to get one from | |
172 | the _ies list, if there's no instance it will create a new one and add | |
173 | it to the extractor list. | |
174 | """ | |
175 | ie = self._ies_instances.get(ie_key) | |
176 | if ie is None: | |
177 | ie = get_info_extractor(ie_key)() | |
178 | self.add_info_extractor(ie) | |
179 | return ie | |
180 | ||
023fa8c4 JMF |
181 | def add_default_info_extractors(self): |
182 | """ | |
183 | Add the InfoExtractors returned by gen_extractors to the end of the list | |
184 | """ | |
185 | for ie in gen_extractors(): | |
186 | self.add_info_extractor(ie) | |
187 | ||
8222d8de JMF |
188 | def add_post_processor(self, pp): |
189 | """Add a PostProcessor object to the end of the chain.""" | |
190 | self._pps.append(pp) | |
191 | pp.set_downloader(self) | |
192 | ||
193 | def to_screen(self, message, skip_eol=False): | |
194 | """Print message to stdout if not in quiet mode.""" | |
8222d8de JMF |
195 | if not self.params.get('quiet', False): |
196 | terminator = [u'\n', u''][skip_eol] | |
197 | output = message + terminator | |
7459e3a2 | 198 | write_string(output, self._screen_file) |
8222d8de JMF |
199 | |
200 | def to_stderr(self, message): | |
201 | """Print message to stderr.""" | |
202 | assert type(message) == type(u'') | |
203 | output = message + u'\n' | |
204 | 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 | |
205 | output = output.encode(preferredencoding()) | |
206 | sys.stderr.write(output) | |
207 | ||
1e5b9a95 PH |
208 | def to_console_title(self, message): |
209 | if not self.params.get('consoletitle', False): | |
210 | return | |
211 | if os.name == 'nt' and ctypes.windll.kernel32.GetConsoleWindow(): | |
212 | # c_wchar_p() might not be necessary if `message` is | |
213 | # already of type unicode() | |
214 | ctypes.windll.kernel32.SetConsoleTitleW(ctypes.c_wchar_p(message)) | |
215 | elif 'TERM' in os.environ: | |
749febf4 | 216 | write_string(u'\033]0;%s\007' % message, self._screen_file) |
1e5b9a95 | 217 | |
bdde425c PH |
218 | def save_console_title(self): |
219 | if not self.params.get('consoletitle', False): | |
220 | return | |
221 | if 'TERM' in os.environ: | |
efd6c574 JMF |
222 | # Save the title on stack |
223 | write_string(u'\033[22;0t', self._screen_file) | |
bdde425c PH |
224 | |
225 | def restore_console_title(self): | |
226 | if not self.params.get('consoletitle', False): | |
227 | return | |
228 | if 'TERM' in os.environ: | |
efd6c574 JMF |
229 | # Restore the title from stack |
230 | write_string(u'\033[23;0t', self._screen_file) | |
bdde425c PH |
231 | |
232 | def __enter__(self): | |
233 | self.save_console_title() | |
234 | return self | |
235 | ||
236 | def __exit__(self, *args): | |
237 | self.restore_console_title() | |
238 | ||
8222d8de JMF |
239 | def fixed_template(self): |
240 | """Checks if the output template is fixed.""" | |
241 | return (re.search(u'(?u)%\\(.+?\\)s', self.params['outtmpl']) is None) | |
242 | ||
243 | def trouble(self, message=None, tb=None): | |
244 | """Determine action to take when a download problem appears. | |
245 | ||
246 | Depending on if the downloader has been configured to ignore | |
247 | download errors or not, this method may throw an exception or | |
248 | not when errors are found, after printing the message. | |
249 | ||
250 | tb, if given, is additional traceback information. | |
251 | """ | |
252 | if message is not None: | |
253 | self.to_stderr(message) | |
254 | if self.params.get('verbose'): | |
255 | if tb is None: | |
256 | if sys.exc_info()[0]: # if .trouble has been called from an except block | |
257 | tb = u'' | |
258 | if hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]: | |
259 | tb += u''.join(traceback.format_exception(*sys.exc_info()[1].exc_info)) | |
260 | tb += compat_str(traceback.format_exc()) | |
261 | else: | |
262 | tb_data = traceback.format_list(traceback.extract_stack()) | |
263 | tb = u''.join(tb_data) | |
264 | self.to_stderr(tb) | |
265 | if not self.params.get('ignoreerrors', False): | |
266 | if sys.exc_info()[0] and hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]: | |
267 | exc_info = sys.exc_info()[1].exc_info | |
268 | else: | |
269 | exc_info = sys.exc_info() | |
270 | raise DownloadError(message, exc_info) | |
271 | self._download_retcode = 1 | |
272 | ||
273 | def report_warning(self, message): | |
274 | ''' | |
275 | Print the message to stderr, it will be prefixed with 'WARNING:' | |
276 | If stderr is a tty file the 'WARNING:' will be colored | |
277 | ''' | |
278 | if sys.stderr.isatty() and os.name != 'nt': | |
fe7e0c98 | 279 | _msg_header = u'\033[0;33mWARNING:\033[0m' |
8222d8de | 280 | else: |
fe7e0c98 JMF |
281 | _msg_header = u'WARNING:' |
282 | warning_message = u'%s %s' % (_msg_header, message) | |
8222d8de JMF |
283 | self.to_stderr(warning_message) |
284 | ||
285 | def report_error(self, message, tb=None): | |
286 | ''' | |
287 | Do the same as trouble, but prefixes the message with 'ERROR:', colored | |
288 | in red if stderr is a tty file. | |
289 | ''' | |
290 | if sys.stderr.isatty() and os.name != 'nt': | |
291 | _msg_header = u'\033[0;31mERROR:\033[0m' | |
292 | else: | |
293 | _msg_header = u'ERROR:' | |
294 | error_message = u'%s %s' % (_msg_header, message) | |
295 | self.trouble(error_message, tb) | |
296 | ||
8222d8de JMF |
297 | def report_writedescription(self, descfn): |
298 | """ Report that the description file is being written """ | |
299 | self.to_screen(u'[info] Writing video description to: ' + descfn) | |
300 | ||
301 | def report_writesubtitles(self, sub_filename): | |
302 | """ Report that the subtitles file is being written """ | |
303 | self.to_screen(u'[info] Writing video subtitles to: ' + sub_filename) | |
304 | ||
305 | def report_writeinfojson(self, infofn): | |
306 | """ Report that the metadata file has been written """ | |
307 | self.to_screen(u'[info] Video description metadata as JSON to: ' + infofn) | |
308 | ||
1fb07d10 JG |
309 | def report_writeannotations(self, annofn): |
310 | """ Report that the annotations file has been written. """ | |
311 | self.to_screen(u'[info] Writing video annotations to: ' + annofn) | |
312 | ||
8222d8de JMF |
313 | def report_file_already_downloaded(self, file_name): |
314 | """Report file has already been fully downloaded.""" | |
315 | try: | |
316 | self.to_screen(u'[download] %s has already been downloaded' % file_name) | |
ce02ed60 | 317 | except UnicodeEncodeError: |
8222d8de JMF |
318 | self.to_screen(u'[download] The file has already been downloaded') |
319 | ||
320 | def increment_downloads(self): | |
321 | """Increment the ordinal that assigns a number to each file.""" | |
322 | self._num_downloads += 1 | |
323 | ||
324 | def prepare_filename(self, info_dict): | |
325 | """Generate the output filename.""" | |
326 | try: | |
327 | template_dict = dict(info_dict) | |
328 | ||
329 | template_dict['epoch'] = int(time.time()) | |
330 | autonumber_size = self.params.get('autonumber_size') | |
331 | if autonumber_size is None: | |
332 | autonumber_size = 5 | |
333 | autonumber_templ = u'%0' + str(autonumber_size) + u'd' | |
334 | template_dict['autonumber'] = autonumber_templ % self._num_downloads | |
702665c0 | 335 | if template_dict.get('playlist_index') is not None: |
8222d8de JMF |
336 | template_dict['playlist_index'] = u'%05d' % template_dict['playlist_index'] |
337 | ||
586a91b6 | 338 | sanitize = lambda k, v: sanitize_filename( |
8222d8de JMF |
339 | u'NA' if v is None else compat_str(v), |
340 | restricted=self.params.get('restrictfilenames'), | |
586a91b6 PH |
341 | is_id=(k == u'id')) |
342 | template_dict = dict((k, sanitize(k, v)) | |
343 | for k, v in template_dict.items()) | |
8222d8de | 344 | |
586a91b6 PH |
345 | tmpl = os.path.expanduser(self.params['outtmpl']) |
346 | filename = tmpl % template_dict | |
8222d8de JMF |
347 | return filename |
348 | except KeyError as err: | |
349 | self.report_error(u'Erroneous output template') | |
350 | return None | |
351 | except ValueError as err: | |
4efba05c | 352 | self.report_error(u'Error in output template: ' + str(err) + u' (encoding: ' + repr(preferredencoding()) + ')') |
8222d8de JMF |
353 | return None |
354 | ||
355 | def _match_entry(self, info_dict): | |
356 | """ Returns None iff the file should be downloaded """ | |
357 | ||
7012b23c PH |
358 | if 'title' in info_dict: |
359 | # This can happen when we're just evaluating the playlist | |
360 | title = info_dict['title'] | |
361 | matchtitle = self.params.get('matchtitle', False) | |
362 | if matchtitle: | |
363 | if not re.search(matchtitle, title, re.IGNORECASE): | |
364 | return u'[download] "' + title + '" title did not match pattern "' + matchtitle + '"' | |
365 | rejecttitle = self.params.get('rejecttitle', False) | |
366 | if rejecttitle: | |
367 | if re.search(rejecttitle, title, re.IGNORECASE): | |
368 | return u'"' + title + '" title matched reject pattern "' + rejecttitle + '"' | |
8222d8de JMF |
369 | date = info_dict.get('upload_date', None) |
370 | if date is not None: | |
371 | dateRange = self.params.get('daterange', DateRange()) | |
372 | if date not in dateRange: | |
373 | return u'[download] %s upload date is not in range %s' % (date_from_str(date).isoformat(), dateRange) | |
8dbe9899 PH |
374 | age_limit = self.params.get('age_limit') |
375 | if age_limit is not None: | |
cfadd183 | 376 | if age_limit < info_dict.get('age_limit', 0): |
8dbe9899 | 377 | return u'Skipping "' + title + '" because it is age restricted' |
c1c9a79c | 378 | if self.in_download_archive(info_dict): |
7012b23c PH |
379 | return (u'%s has already been recorded in archive' |
380 | % info_dict.get('title', info_dict.get('id', u'video'))) | |
8222d8de | 381 | return None |
fe7e0c98 | 382 | |
b6c45014 JMF |
383 | @staticmethod |
384 | def add_extra_info(info_dict, extra_info): | |
385 | '''Set the keys from extra_info in info dict if they are missing''' | |
386 | for key, value in extra_info.items(): | |
387 | info_dict.setdefault(key, value) | |
388 | ||
8222d8de JMF |
389 | def extract_info(self, url, download=True, ie_key=None, extra_info={}): |
390 | ''' | |
391 | Returns a list with a dictionary for each video we find. | |
392 | If 'download', also downloads the videos. | |
393 | extra_info is a dict containing the extra values to add to each result | |
394 | ''' | |
fe7e0c98 | 395 | |
8222d8de | 396 | if ie_key: |
56c73665 | 397 | ies = [self.get_info_extractor(ie_key)] |
8222d8de JMF |
398 | else: |
399 | ies = self._ies | |
400 | ||
401 | for ie in ies: | |
402 | if not ie.suitable(url): | |
403 | continue | |
404 | ||
405 | if not ie.working(): | |
406 | self.report_warning(u'The program functionality for this site has been marked as broken, ' | |
407 | u'and will probably not work.') | |
408 | ||
409 | try: | |
410 | ie_result = ie.extract(url) | |
411 | if ie_result is None: # Finished already (backwards compatibility; listformats and friends should be moved here) | |
412 | break | |
413 | if isinstance(ie_result, list): | |
414 | # Backwards compatibility: old IE result format | |
8222d8de JMF |
415 | ie_result = { |
416 | '_type': 'compat_list', | |
417 | 'entries': ie_result, | |
418 | } | |
9103bbc5 JMF |
419 | self.add_extra_info(ie_result, |
420 | { | |
421 | 'extractor': ie.IE_NAME, | |
be97abc2 JMF |
422 | 'webpage_url': url, |
423 | 'extractor_key': ie.ie_key(), | |
9103bbc5 | 424 | }) |
b6c45014 | 425 | return self.process_ie_result(ie_result, download, extra_info) |
8222d8de JMF |
426 | except ExtractorError as de: # An error we somewhat expected |
427 | self.report_error(compat_str(de), de.format_traceback()) | |
428 | break | |
429 | except Exception as e: | |
430 | if self.params.get('ignoreerrors', False): | |
431 | self.report_error(compat_str(e), tb=compat_str(traceback.format_exc())) | |
432 | break | |
433 | else: | |
434 | raise | |
435 | else: | |
436 | self.report_error(u'no suitable InfoExtractor: %s' % url) | |
fe7e0c98 | 437 | |
8222d8de JMF |
438 | def process_ie_result(self, ie_result, download=True, extra_info={}): |
439 | """ | |
440 | Take the result of the ie(may be modified) and resolve all unresolved | |
441 | references (URLs, playlist items). | |
442 | ||
443 | It will also download the videos if 'download'. | |
444 | Returns the resolved ie_result. | |
445 | """ | |
446 | ||
447 | result_type = ie_result.get('_type', 'video') # If not given we suppose it's a video, support the default old system | |
448 | if result_type == 'video': | |
b6c45014 | 449 | self.add_extra_info(ie_result, extra_info) |
feee2ecf | 450 | return self.process_video_result(ie_result, download=download) |
8222d8de JMF |
451 | elif result_type == 'url': |
452 | # We have to add extra_info to the results because it may be | |
453 | # contained in a playlist | |
454 | return self.extract_info(ie_result['url'], | |
455 | download, | |
456 | ie_key=ie_result.get('ie_key'), | |
457 | extra_info=extra_info) | |
458 | elif result_type == 'playlist': | |
7012b23c | 459 | |
8222d8de JMF |
460 | # We process each entry in the playlist |
461 | playlist = ie_result.get('title', None) or ie_result.get('id', None) | |
fe7e0c98 | 462 | self.to_screen(u'[download] Downloading playlist: %s' % playlist) |
8222d8de JMF |
463 | |
464 | playlist_results = [] | |
465 | ||
466 | n_all_entries = len(ie_result['entries']) | |
467 | playliststart = self.params.get('playliststart', 1) - 1 | |
468 | playlistend = self.params.get('playlistend', -1) | |
469 | ||
470 | if playlistend == -1: | |
471 | entries = ie_result['entries'][playliststart:] | |
472 | else: | |
473 | entries = ie_result['entries'][playliststart:playlistend] | |
474 | ||
475 | n_entries = len(entries) | |
476 | ||
477 | self.to_screen(u"[%s] playlist '%s': Collected %d video ids (downloading %d of them)" % | |
478 | (ie_result['extractor'], playlist, n_all_entries, n_entries)) | |
479 | ||
fe7e0c98 JMF |
480 | for i, entry in enumerate(entries, 1): |
481 | self.to_screen(u'[download] Downloading video #%s of %s' % (i, n_entries)) | |
8222d8de | 482 | extra = { |
fe7e0c98 JMF |
483 | 'playlist': playlist, |
484 | 'playlist_index': i + playliststart, | |
b6c45014 | 485 | 'extractor': ie_result['extractor'], |
9103bbc5 | 486 | 'webpage_url': ie_result['webpage_url'], |
be97abc2 | 487 | 'extractor_key': ie_result['extractor_key'], |
fe7e0c98 | 488 | } |
7012b23c PH |
489 | |
490 | reason = self._match_entry(entry) | |
491 | if reason is not None: | |
492 | self.to_screen(u'[download] ' + reason) | |
493 | continue | |
494 | ||
8222d8de JMF |
495 | entry_result = self.process_ie_result(entry, |
496 | download=download, | |
497 | extra_info=extra) | |
498 | playlist_results.append(entry_result) | |
499 | ie_result['entries'] = playlist_results | |
500 | return ie_result | |
501 | elif result_type == 'compat_list': | |
502 | def _fixup(r): | |
b6c45014 | 503 | self.add_extra_info(r, |
9103bbc5 JMF |
504 | { |
505 | 'extractor': ie_result['extractor'], | |
506 | 'webpage_url': ie_result['webpage_url'], | |
be97abc2 | 507 | 'extractor_key': ie_result['extractor_key'], |
9103bbc5 | 508 | }) |
8222d8de JMF |
509 | return r |
510 | ie_result['entries'] = [ | |
b6c45014 | 511 | self.process_ie_result(_fixup(r), download, extra_info) |
8222d8de JMF |
512 | for r in ie_result['entries'] |
513 | ] | |
514 | return ie_result | |
515 | else: | |
516 | raise Exception('Invalid result type: %s' % result_type) | |
517 | ||
a9c58ad9 JMF |
518 | def select_format(self, format_spec, available_formats): |
519 | if format_spec == 'best' or format_spec is None: | |
520 | return available_formats[-1] | |
521 | elif format_spec == 'worst': | |
522 | return available_formats[0] | |
523 | else: | |
49e86983 JMF |
524 | extensions = [u'mp4', u'flv', u'webm', u'3gp'] |
525 | if format_spec in extensions: | |
526 | filter_f = lambda f: f['ext'] == format_spec | |
527 | else: | |
528 | filter_f = lambda f: f['format_id'] == format_spec | |
fe7e0c98 | 529 | matches = list(filter(filter_f, available_formats)) |
a9c58ad9 JMF |
530 | if matches: |
531 | return matches[-1] | |
532 | return None | |
533 | ||
dd82ffea JMF |
534 | def process_video_result(self, info_dict, download=True): |
535 | assert info_dict.get('_type', 'video') == 'video' | |
536 | ||
537 | if 'playlist' not in info_dict: | |
538 | # It isn't part of a playlist | |
539 | info_dict['playlist'] = None | |
540 | info_dict['playlist_index'] = None | |
541 | ||
6ff000b8 | 542 | # This extractors handle format selection themselves |
a7685f3b | 543 | if info_dict['extractor'] in [u'youtube', u'Youku']: |
12893efe JMF |
544 | if download: |
545 | self.process_info(info_dict) | |
6ff000b8 JMF |
546 | return info_dict |
547 | ||
dd82ffea JMF |
548 | # We now pick which formats have to be downloaded |
549 | if info_dict.get('formats') is None: | |
550 | # There's only one format available | |
551 | formats = [info_dict] | |
552 | else: | |
553 | formats = info_dict['formats'] | |
554 | ||
555 | # We check that all the formats have the format and format_id fields | |
556 | for (i, format) in enumerate(formats): | |
dd82ffea | 557 | if format.get('format_id') is None: |
8016c922 | 558 | format['format_id'] = compat_str(i) |
8c51aa65 JMF |
559 | if format.get('format') is None: |
560 | format['format'] = u'{id} - {res}{note}'.format( | |
561 | id=format['format_id'], | |
562 | res=self.format_resolution(format), | |
71934988 | 563 | note=u' ({0})'.format(format['format_note']) if format.get('format_note') is not None else '', |
8c51aa65 | 564 | ) |
c1002e96 PH |
565 | # Automatically determine file extension if missing |
566 | if 'ext' not in format: | |
567 | format['ext'] = determine_ext(format['url']) | |
dd82ffea JMF |
568 | |
569 | if self.params.get('listformats', None): | |
570 | self.list_formats(info_dict) | |
571 | return | |
572 | ||
99e206d5 JMF |
573 | format_limit = self.params.get('format_limit', None) |
574 | if format_limit: | |
f4d96df0 PH |
575 | formats = list(takewhile_inclusive( |
576 | lambda f: f['format_id'] != format_limit, formats | |
577 | )) | |
e028d0d1 JMF |
578 | if self.params.get('prefer_free_formats'): |
579 | def _free_formats_key(f): | |
580 | try: | |
581 | ext_ord = [u'flv', u'mp4', u'webm'].index(f['ext']) | |
582 | except ValueError: | |
583 | ext_ord = -1 | |
584 | # We only compare the extension if they have the same height and width | |
585 | return (f.get('height'), f.get('width'), ext_ord) | |
586 | formats = sorted(formats, key=_free_formats_key) | |
99e206d5 | 587 | |
dd82ffea | 588 | req_format = self.params.get('format', 'best') |
a9c58ad9 JMF |
589 | if req_format is None: |
590 | req_format = 'best' | |
dd82ffea | 591 | formats_to_download = [] |
dd82ffea | 592 | # The -1 is for supporting YoutubeIE |
a9c58ad9 | 593 | if req_format in ('-1', 'all'): |
dd82ffea JMF |
594 | formats_to_download = formats |
595 | else: | |
a9c58ad9 | 596 | # We can accept formats requestd in the format: 34/5/best, we pick |
416a5efc | 597 | # the first that is available, starting from left |
dd82ffea JMF |
598 | req_formats = req_format.split('/') |
599 | for rf in req_formats: | |
a9c58ad9 JMF |
600 | selected_format = self.select_format(rf, formats) |
601 | if selected_format is not None: | |
602 | formats_to_download = [selected_format] | |
dd82ffea JMF |
603 | break |
604 | if not formats_to_download: | |
78a3a9f8 PH |
605 | raise ExtractorError(u'requested format not available', |
606 | expected=True) | |
dd82ffea JMF |
607 | |
608 | if download: | |
609 | if len(formats_to_download) > 1: | |
610 | self.to_screen(u'[info] %s: downloading video in %s formats' % (info_dict['id'], len(formats_to_download))) | |
611 | for format in formats_to_download: | |
612 | new_info = dict(info_dict) | |
613 | new_info.update(format) | |
614 | self.process_info(new_info) | |
615 | # We update the info dict with the best quality format (backwards compatibility) | |
616 | info_dict.update(formats_to_download[-1]) | |
617 | return info_dict | |
618 | ||
8222d8de JMF |
619 | def process_info(self, info_dict): |
620 | """Process a single resolved IE result.""" | |
621 | ||
622 | assert info_dict.get('_type', 'video') == 'video' | |
623 | #We increment the download the download count here to match the previous behaviour. | |
624 | self.increment_downloads() | |
625 | ||
626 | info_dict['fulltitle'] = info_dict['title'] | |
627 | if len(info_dict['title']) > 200: | |
628 | info_dict['title'] = info_dict['title'][:197] + u'...' | |
629 | ||
630 | # Keep for backwards compatibility | |
631 | info_dict['stitle'] = info_dict['title'] | |
632 | ||
633 | if not 'format' in info_dict: | |
634 | info_dict['format'] = info_dict['ext'] | |
635 | ||
636 | reason = self._match_entry(info_dict) | |
637 | if reason is not None: | |
638 | self.to_screen(u'[download] ' + reason) | |
639 | return | |
640 | ||
641 | max_downloads = self.params.get('max_downloads') | |
642 | if max_downloads is not None: | |
643 | if self._num_downloads > int(max_downloads): | |
644 | raise MaxDownloadsReached() | |
645 | ||
646 | filename = self.prepare_filename(info_dict) | |
647 | ||
648 | # Forced printings | |
649 | if self.params.get('forcetitle', False): | |
00ea0f11 | 650 | compat_print(info_dict['fulltitle']) |
8222d8de JMF |
651 | if self.params.get('forceid', False): |
652 | compat_print(info_dict['id']) | |
653 | if self.params.get('forceurl', False): | |
edde6c56 PH |
654 | # For RTMP URLs, also include the playpath |
655 | compat_print(info_dict['url'] + info_dict.get('play_path', u'')) | |
216d71d0 | 656 | if self.params.get('forcethumbnail', False) and info_dict.get('thumbnail') is not None: |
8222d8de | 657 | compat_print(info_dict['thumbnail']) |
216d71d0 | 658 | if self.params.get('forcedescription', False) and info_dict.get('description') is not None: |
8222d8de JMF |
659 | compat_print(info_dict['description']) |
660 | if self.params.get('forcefilename', False) and filename is not None: | |
661 | compat_print(filename) | |
662 | if self.params.get('forceformat', False): | |
663 | compat_print(info_dict['format']) | |
9d153818 MF |
664 | if self.params.get('forcejson', False): |
665 | compat_print(json.dumps(info_dict)) | |
8222d8de JMF |
666 | |
667 | # Do nothing else if in simulate mode | |
668 | if self.params.get('simulate', False): | |
669 | return | |
670 | ||
671 | if filename is None: | |
672 | return | |
673 | ||
674 | try: | |
675 | dn = os.path.dirname(encodeFilename(filename)) | |
676 | if dn != '' and not os.path.exists(dn): | |
677 | os.makedirs(dn) | |
678 | except (OSError, IOError) as err: | |
679 | self.report_error(u'unable to create directory ' + compat_str(err)) | |
680 | return | |
681 | ||
682 | if self.params.get('writedescription', False): | |
683 | try: | |
684 | descfn = filename + u'.description' | |
685 | self.report_writedescription(descfn) | |
686 | with io.open(encodeFilename(descfn), 'w', encoding='utf-8') as descfile: | |
687 | descfile.write(info_dict['description']) | |
b3f0e530 | 688 | except (KeyError, TypeError): |
535f59bb | 689 | self.report_warning(u'There\'s no description to write.') |
8222d8de JMF |
690 | except (OSError, IOError): |
691 | self.report_error(u'Cannot write description file ' + descfn) | |
692 | return | |
693 | ||
1fb07d10 JG |
694 | if self.params.get('writeannotations', False): |
695 | try: | |
fe7e0c98 JMF |
696 | annofn = filename + u'.annotations.xml' |
697 | self.report_writeannotations(annofn) | |
698 | with io.open(encodeFilename(annofn), 'w', encoding='utf-8') as annofile: | |
699 | annofile.write(info_dict['annotations']) | |
1fb07d10 JG |
700 | except (KeyError, TypeError): |
701 | self.report_warning(u'There are no annotations to write.') | |
702 | except (OSError, IOError): | |
fe7e0c98 JMF |
703 | self.report_error(u'Cannot write annotations file: ' + annofn) |
704 | return | |
1fb07d10 | 705 | |
c4a91be7 | 706 | subtitles_are_requested = any([self.params.get('writesubtitles', False), |
0b7f3118 | 707 | self.params.get('writeautomaticsub')]) |
c4a91be7 | 708 | |
fe7e0c98 | 709 | if subtitles_are_requested and 'subtitles' in info_dict and info_dict['subtitles']: |
8222d8de JMF |
710 | # subtitles download errors are already managed as troubles in relevant IE |
711 | # that way it will silently go on when used with unsupporting IE | |
8222d8de | 712 | subtitles = info_dict['subtitles'] |
ca715127 | 713 | sub_format = self.params.get('subtitlesformat', 'srt') |
5d51a883 JMF |
714 | for sub_lang in subtitles.keys(): |
715 | sub = subtitles[sub_lang] | |
6804038d JMF |
716 | if sub is None: |
717 | continue | |
8222d8de | 718 | try: |
d4051a8e | 719 | sub_filename = subtitles_filename(filename, sub_lang, sub_format) |
8222d8de JMF |
720 | self.report_writesubtitles(sub_filename) |
721 | with io.open(encodeFilename(sub_filename), 'w', encoding='utf-8') as subfile: | |
5d51a883 | 722 | subfile.write(sub) |
8222d8de JMF |
723 | except (OSError, IOError): |
724 | self.report_error(u'Cannot write subtitles file ' + descfn) | |
725 | return | |
726 | ||
8222d8de | 727 | if self.params.get('writeinfojson', False): |
9771cceb | 728 | infofn = os.path.splitext(filename)[0] + u'.info.json' |
8222d8de JMF |
729 | self.report_writeinfojson(infofn) |
730 | try: | |
fe7e0c98 | 731 | json_info_dict = dict((k, v) for k, v in info_dict.items() if not k in ['urlhandle']) |
8222d8de JMF |
732 | write_json_file(json_info_dict, encodeFilename(infofn)) |
733 | except (OSError, IOError): | |
734 | self.report_error(u'Cannot write metadata to JSON file ' + infofn) | |
735 | return | |
736 | ||
737 | if self.params.get('writethumbnail', False): | |
d8269e1d | 738 | if info_dict.get('thumbnail') is not None: |
cbdbb766 | 739 | thumb_format = determine_ext(info_dict['thumbnail'], u'jpg') |
8222d8de JMF |
740 | thumb_filename = filename.rpartition('.')[0] + u'.' + thumb_format |
741 | self.to_screen(u'[%s] %s: Downloading thumbnail ...' % | |
742 | (info_dict['extractor'], info_dict['id'])) | |
0a60edcf JMF |
743 | try: |
744 | uf = compat_urllib_request.urlopen(info_dict['thumbnail']) | |
745 | with open(thumb_filename, 'wb') as thumbf: | |
746 | shutil.copyfileobj(uf, thumbf) | |
747 | self.to_screen(u'[%s] %s: Writing thumbnail to: %s' % | |
748 | (info_dict['extractor'], info_dict['id'], thumb_filename)) | |
749 | except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: | |
750 | self.report_warning(u'Unable to download thumbnail "%s": %s' % | |
751 | (info_dict['thumbnail'], compat_str(err))) | |
8222d8de JMF |
752 | |
753 | if not self.params.get('skip_download', False): | |
754 | if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(filename)): | |
755 | success = True | |
756 | else: | |
757 | try: | |
758 | success = self.fd._do_download(filename, info_dict) | |
8222d8de JMF |
759 | except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: |
760 | self.report_error(u'unable to download video data: %s' % str(err)) | |
761 | return | |
c40c6aaa JMF |
762 | except (OSError, IOError) as err: |
763 | raise UnavailableVideoError(err) | |
8222d8de JMF |
764 | except (ContentTooShortError, ) as err: |
765 | self.report_error(u'content too short (expected %s bytes and served %s)' % (err.expected, err.downloaded)) | |
766 | return | |
767 | ||
768 | if success: | |
769 | try: | |
770 | self.post_process(filename, info_dict) | |
771 | except (PostProcessingError) as err: | |
772 | self.report_error(u'postprocessing: %s' % str(err)) | |
773 | return | |
774 | ||
c1c9a79c PH |
775 | self.record_download_archive(info_dict) |
776 | ||
8222d8de JMF |
777 | def download(self, url_list): |
778 | """Download a given list of URLs.""" | |
779 | if len(url_list) > 1 and self.fixed_template(): | |
780 | raise SameFileError(self.params['outtmpl']) | |
781 | ||
782 | for url in url_list: | |
783 | try: | |
784 | #It also downloads the videos | |
785 | videos = self.extract_info(url) | |
786 | except UnavailableVideoError: | |
787 | self.report_error(u'unable to download video') | |
788 | except MaxDownloadsReached: | |
789 | self.to_screen(u'[info] Maximum number of downloaded files reached.') | |
790 | raise | |
791 | ||
792 | return self._download_retcode | |
793 | ||
794 | def post_process(self, filename, ie_info): | |
795 | """Run all the postprocessors on the given file.""" | |
796 | info = dict(ie_info) | |
797 | info['filepath'] = filename | |
798 | keep_video = None | |
799 | for pp in self._pps: | |
800 | try: | |
fe7e0c98 | 801 | keep_video_wish, new_info = pp.run(info) |
8222d8de JMF |
802 | if keep_video_wish is not None: |
803 | if keep_video_wish: | |
804 | keep_video = keep_video_wish | |
805 | elif keep_video is None: | |
806 | # No clear decision yet, let IE decide | |
807 | keep_video = keep_video_wish | |
808 | except PostProcessingError as e: | |
bbcbf4d4 | 809 | self.report_error(e.msg) |
8222d8de JMF |
810 | if keep_video is False and not self.params.get('keepvideo', False): |
811 | try: | |
812 | self.to_screen(u'Deleting original file %s (pass -k to keep)' % filename) | |
813 | os.remove(encodeFilename(filename)) | |
814 | except (IOError, OSError): | |
815 | self.report_warning(u'Unable to remove downloaded video file') | |
c1c9a79c PH |
816 | |
817 | def in_download_archive(self, info_dict): | |
818 | fn = self.params.get('download_archive') | |
819 | if fn is None: | |
820 | return False | |
7012b23c PH |
821 | extractor = info_dict.get('extractor_id') |
822 | if extractor is None: | |
823 | if 'id' in info_dict: | |
824 | extractor = info_dict.get('ie_key') # key in a playlist | |
825 | if extractor is None: | |
826 | return False # Incomplete video information | |
827 | # Future-proof against any change in case | |
828 | # and backwards compatibility with prior versions | |
829 | extractor = extractor.lower() | |
830 | vid_id = extractor + u' ' + info_dict['id'] | |
c1c9a79c PH |
831 | try: |
832 | with locked_file(fn, 'r', encoding='utf-8') as archive_file: | |
833 | for line in archive_file: | |
834 | if line.strip() == vid_id: | |
835 | return True | |
836 | except IOError as ioe: | |
837 | if ioe.errno != errno.ENOENT: | |
838 | raise | |
839 | return False | |
840 | ||
841 | def record_download_archive(self, info_dict): | |
842 | fn = self.params.get('download_archive') | |
843 | if fn is None: | |
844 | return | |
845 | vid_id = info_dict['extractor'] + u' ' + info_dict['id'] | |
846 | with locked_file(fn, 'a', encoding='utf-8') as archive_file: | |
847 | archive_file.write(vid_id + u'\n') | |
dd82ffea | 848 | |
8c51aa65 | 849 | @staticmethod |
8abeeb94 | 850 | def format_resolution(format, default='unknown'): |
57dd9a8f PH |
851 | if format.get('_resolution') is not None: |
852 | return format['_resolution'] | |
8c51aa65 JMF |
853 | if format.get('height') is not None: |
854 | if format.get('width') is not None: | |
855 | res = u'%sx%s' % (format['width'], format['height']) | |
856 | else: | |
857 | res = u'%sp' % format['height'] | |
858 | else: | |
8abeeb94 | 859 | res = default |
8c51aa65 JMF |
860 | return res |
861 | ||
dd82ffea | 862 | def list_formats(self, info_dict): |
91c7271a PH |
863 | def format_note(fdict): |
864 | if fdict.get('format_note') is not None: | |
865 | return fdict['format_note'] | |
866 | res = u'' | |
867 | if fdict.get('vcodec') is not None: | |
7150858d PH |
868 | res += u'%-5s' % fdict['vcodec'] |
869 | elif fdict.get('vbr') is not None: | |
870 | res += u'video' | |
91c7271a PH |
871 | if fdict.get('vbr') is not None: |
872 | res += u'@%4dk' % fdict['vbr'] | |
873 | if fdict.get('acodec') is not None: | |
874 | if res: | |
875 | res += u', ' | |
7150858d PH |
876 | res += u'%-5s' % fdict['acodec'] |
877 | elif fdict.get('abr') is not None: | |
878 | if res: | |
879 | res += u', ' | |
880 | res += 'audio' | |
91c7271a PH |
881 | if fdict.get('abr') is not None: |
882 | res += u'@%3dk' % fdict['abr'] | |
883 | return res | |
884 | ||
57dd9a8f | 885 | def line(format): |
897d6cc4 | 886 | return (u'%-20s%-10s%-12s%s' % ( |
8c51aa65 JMF |
887 | format['format_id'], |
888 | format['ext'], | |
8c51aa65 | 889 | self.format_resolution(format), |
91c7271a | 890 | format_note(format), |
8c51aa65 JMF |
891 | ) |
892 | ) | |
57dd9a8f | 893 | |
94badb25 PH |
894 | formats = info_dict.get('formats', [info_dict]) |
895 | formats_s = list(map(line, formats)) | |
896 | if len(formats) > 1: | |
b5349e87 PH |
897 | formats_s[0] += (' ' if format_note(formats[0]) else '') + '(worst)' |
898 | formats_s[-1] += (' ' if format_note(formats[-1]) else '') + '(best)' | |
57dd9a8f PH |
899 | |
900 | header_line = line({ | |
901 | 'format_id': u'format code', 'ext': u'extension', | |
902 | '_resolution': u'resolution', 'format_note': u'note'}) | |
903 | self.to_screen(u'[info] Available formats for %s:\n%s\n%s' % | |
904 | (info_dict['id'], header_line, u"\n".join(formats_s))) |