]>
Commit | Line | Data |
---|---|---|
1 | #!/usr/bin/env python | |
2 | # coding: utf-8 | |
3 | ||
4 | from __future__ import absolute_import, unicode_literals | |
5 | ||
6 | import collections | |
7 | import contextlib | |
8 | import copy | |
9 | import datetime | |
10 | import errno | |
11 | import fileinput | |
12 | import io | |
13 | import itertools | |
14 | import json | |
15 | import locale | |
16 | import operator | |
17 | import os | |
18 | import platform | |
19 | import re | |
20 | import shutil | |
21 | import subprocess | |
22 | import sys | |
23 | import time | |
24 | import tokenize | |
25 | import traceback | |
26 | import random | |
27 | ||
28 | from string import ascii_letters | |
29 | from zipimport import zipimporter | |
30 | ||
31 | from .compat import ( | |
32 | compat_basestring, | |
33 | compat_cookiejar, | |
34 | compat_get_terminal_size, | |
35 | compat_kwargs, | |
36 | compat_numeric_types, | |
37 | compat_os_name, | |
38 | compat_str, | |
39 | compat_tokenize_tokenize, | |
40 | compat_urllib_error, | |
41 | compat_urllib_request, | |
42 | compat_urllib_request_DataHandler, | |
43 | ) | |
44 | from .utils import ( | |
45 | age_restricted, | |
46 | args_to_str, | |
47 | ContentTooShortError, | |
48 | date_from_str, | |
49 | DateRange, | |
50 | DEFAULT_OUTTMPL, | |
51 | determine_ext, | |
52 | determine_protocol, | |
53 | DOT_DESKTOP_LINK_TEMPLATE, | |
54 | DOT_URL_LINK_TEMPLATE, | |
55 | DOT_WEBLOC_LINK_TEMPLATE, | |
56 | DownloadError, | |
57 | encode_compat_str, | |
58 | encodeFilename, | |
59 | EntryNotInPlaylist, | |
60 | error_to_compat_str, | |
61 | ExistingVideoReached, | |
62 | expand_path, | |
63 | ExtractorError, | |
64 | float_or_none, | |
65 | format_bytes, | |
66 | format_field, | |
67 | FORMAT_RE, | |
68 | formatSeconds, | |
69 | GeoRestrictedError, | |
70 | int_or_none, | |
71 | iri_to_uri, | |
72 | ISO3166Utils, | |
73 | LazyList, | |
74 | locked_file, | |
75 | make_dir, | |
76 | make_HTTPS_handler, | |
77 | MaxDownloadsReached, | |
78 | network_exceptions, | |
79 | orderedSet, | |
80 | OUTTMPL_TYPES, | |
81 | PagedList, | |
82 | parse_filesize, | |
83 | PerRequestProxyHandler, | |
84 | platform_name, | |
85 | PostProcessingError, | |
86 | preferredencoding, | |
87 | prepend_extension, | |
88 | process_communicate_or_kill, | |
89 | random_uuidv4, | |
90 | register_socks_protocols, | |
91 | RejectedVideoReached, | |
92 | render_table, | |
93 | replace_extension, | |
94 | SameFileError, | |
95 | sanitize_filename, | |
96 | sanitize_path, | |
97 | sanitize_url, | |
98 | sanitized_Request, | |
99 | std_headers, | |
100 | str_or_none, | |
101 | strftime_or_none, | |
102 | subtitles_filename, | |
103 | to_high_limit_path, | |
104 | traverse_dict, | |
105 | UnavailableVideoError, | |
106 | url_basename, | |
107 | version_tuple, | |
108 | write_json_file, | |
109 | write_string, | |
110 | YoutubeDLCookieJar, | |
111 | YoutubeDLCookieProcessor, | |
112 | YoutubeDLHandler, | |
113 | YoutubeDLRedirectHandler, | |
114 | ) | |
115 | from .cache import Cache | |
116 | from .extractor import ( | |
117 | gen_extractor_classes, | |
118 | get_info_extractor, | |
119 | _LAZY_LOADER, | |
120 | _PLUGIN_CLASSES | |
121 | ) | |
122 | from .extractor.openload import PhantomJSwrapper | |
123 | from .downloader import ( | |
124 | get_suitable_downloader, | |
125 | shorten_protocol_name | |
126 | ) | |
127 | from .downloader.rtmp import rtmpdump_version | |
128 | from .postprocessor import ( | |
129 | FFmpegFixupM3u8PP, | |
130 | FFmpegFixupM4aPP, | |
131 | FFmpegFixupStretchedPP, | |
132 | FFmpegMergerPP, | |
133 | FFmpegPostProcessor, | |
134 | # FFmpegSubtitlesConvertorPP, | |
135 | get_postprocessor, | |
136 | MoveFilesAfterDownloadPP, | |
137 | ) | |
138 | from .version import __version__ | |
139 | ||
140 | if compat_os_name == 'nt': | |
141 | import ctypes | |
142 | ||
143 | ||
144 | class YoutubeDL(object): | |
145 | """YoutubeDL class. | |
146 | ||
147 | YoutubeDL objects are the ones responsible of downloading the | |
148 | actual video file and writing it to disk if the user has requested | |
149 | it, among some other tasks. In most cases there should be one per | |
150 | program. As, given a video URL, the downloader doesn't know how to | |
151 | extract all the needed information, task that InfoExtractors do, it | |
152 | has to pass the URL to one of them. | |
153 | ||
154 | For this, YoutubeDL objects have a method that allows | |
155 | InfoExtractors to be registered in a given order. When it is passed | |
156 | a URL, the YoutubeDL object handles it to the first InfoExtractor it | |
157 | finds that reports being able to handle it. The InfoExtractor extracts | |
158 | all the information about the video or videos the URL refers to, and | |
159 | YoutubeDL process the extracted information, possibly using a File | |
160 | Downloader to download the video. | |
161 | ||
162 | YoutubeDL objects accept a lot of parameters. In order not to saturate | |
163 | the object constructor with arguments, it receives a dictionary of | |
164 | options instead. These options are available through the params | |
165 | attribute for the InfoExtractors to use. The YoutubeDL also | |
166 | registers itself as the downloader in charge for the InfoExtractors | |
167 | that are added to it, so this is a "mutual registration". | |
168 | ||
169 | Available options: | |
170 | ||
171 | username: Username for authentication purposes. | |
172 | password: Password for authentication purposes. | |
173 | videopassword: Password for accessing a video. | |
174 | ap_mso: Adobe Pass multiple-system operator identifier. | |
175 | ap_username: Multiple-system operator account username. | |
176 | ap_password: Multiple-system operator account password. | |
177 | usenetrc: Use netrc for authentication instead. | |
178 | verbose: Print additional info to stdout. | |
179 | quiet: Do not print messages to stdout. | |
180 | no_warnings: Do not print out anything for warnings. | |
181 | forceprint: A list of templates to force print | |
182 | forceurl: Force printing final URL. (Deprecated) | |
183 | forcetitle: Force printing title. (Deprecated) | |
184 | forceid: Force printing ID. (Deprecated) | |
185 | forcethumbnail: Force printing thumbnail URL. (Deprecated) | |
186 | forcedescription: Force printing description. (Deprecated) | |
187 | forcefilename: Force printing final filename. (Deprecated) | |
188 | forceduration: Force printing duration. (Deprecated) | |
189 | forcejson: Force printing info_dict as JSON. | |
190 | dump_single_json: Force printing the info_dict of the whole playlist | |
191 | (or video) as a single JSON line. | |
192 | force_write_download_archive: Force writing download archive regardless | |
193 | of 'skip_download' or 'simulate'. | |
194 | simulate: Do not download the video files. | |
195 | format: Video format code. see "FORMAT SELECTION" for more details. | |
196 | allow_unplayable_formats: Allow unplayable formats to be extracted and downloaded. | |
197 | ignore_no_formats_error: Ignore "No video formats" error. Usefull for | |
198 | extracting metadata even if the video is not actually | |
199 | available for download (experimental) | |
200 | format_sort: How to sort the video formats. see "Sorting Formats" | |
201 | for more details. | |
202 | format_sort_force: Force the given format_sort. see "Sorting Formats" | |
203 | for more details. | |
204 | allow_multiple_video_streams: Allow multiple video streams to be merged | |
205 | into a single file | |
206 | allow_multiple_audio_streams: Allow multiple audio streams to be merged | |
207 | into a single file | |
208 | paths: Dictionary of output paths. The allowed keys are 'home' | |
209 | 'temp' and the keys of OUTTMPL_TYPES (in utils.py) | |
210 | outtmpl: Dictionary of templates for output names. Allowed keys | |
211 | are 'default' and the keys of OUTTMPL_TYPES (in utils.py). | |
212 | A string a also accepted for backward compatibility | |
213 | outtmpl_na_placeholder: Placeholder for unavailable meta fields. | |
214 | restrictfilenames: Do not allow "&" and spaces in file names | |
215 | trim_file_name: Limit length of filename (extension excluded) | |
216 | windowsfilenames: Force the filenames to be windows compatible | |
217 | ignoreerrors: Do not stop on download errors | |
218 | (Default True when running yt-dlp, | |
219 | but False when directly accessing YoutubeDL class) | |
220 | skip_playlist_after_errors: Number of allowed failures until the rest of | |
221 | the playlist is skipped | |
222 | force_generic_extractor: Force downloader to use the generic extractor | |
223 | overwrites: Overwrite all video and metadata files if True, | |
224 | overwrite only non-video files if None | |
225 | and don't overwrite any file if False | |
226 | playliststart: Playlist item to start at. | |
227 | playlistend: Playlist item to end at. | |
228 | playlist_items: Specific indices of playlist to download. | |
229 | playlistreverse: Download playlist items in reverse order. | |
230 | playlistrandom: Download playlist items in random order. | |
231 | matchtitle: Download only matching titles. | |
232 | rejecttitle: Reject downloads for matching titles. | |
233 | logger: Log messages to a logging.Logger instance. | |
234 | logtostderr: Log messages to stderr instead of stdout. | |
235 | writedescription: Write the video description to a .description file | |
236 | writeinfojson: Write the video description to a .info.json file | |
237 | clean_infojson: Remove private fields from the infojson | |
238 | writecomments: Extract video comments. This will not be written to disk | |
239 | unless writeinfojson is also given | |
240 | writeannotations: Write the video annotations to a .annotations.xml file | |
241 | writethumbnail: Write the thumbnail image to a file | |
242 | allow_playlist_files: Whether to write playlists' description, infojson etc | |
243 | also to disk when using the 'write*' options | |
244 | write_all_thumbnails: Write all thumbnail formats to files | |
245 | writelink: Write an internet shortcut file, depending on the | |
246 | current platform (.url/.webloc/.desktop) | |
247 | writeurllink: Write a Windows internet shortcut file (.url) | |
248 | writewebloclink: Write a macOS internet shortcut file (.webloc) | |
249 | writedesktoplink: Write a Linux internet shortcut file (.desktop) | |
250 | writesubtitles: Write the video subtitles to a file | |
251 | writeautomaticsub: Write the automatically generated subtitles to a file | |
252 | allsubtitles: Deprecated - Use subtitlelangs = ['all'] | |
253 | Downloads all the subtitles of the video | |
254 | (requires writesubtitles or writeautomaticsub) | |
255 | listsubtitles: Lists all available subtitles for the video | |
256 | subtitlesformat: The format code for subtitles | |
257 | subtitleslangs: List of languages of the subtitles to download (can be regex). | |
258 | The list may contain "all" to refer to all the available | |
259 | subtitles. The language can be prefixed with a "-" to | |
260 | exclude it from the requested languages. Eg: ['all', '-live_chat'] | |
261 | keepvideo: Keep the video file after post-processing | |
262 | daterange: A DateRange object, download only if the upload_date is in the range. | |
263 | skip_download: Skip the actual download of the video file | |
264 | cachedir: Location of the cache files in the filesystem. | |
265 | False to disable filesystem cache. | |
266 | noplaylist: Download single video instead of a playlist if in doubt. | |
267 | age_limit: An integer representing the user's age in years. | |
268 | Unsuitable videos for the given age are skipped. | |
269 | min_views: An integer representing the minimum view count the video | |
270 | must have in order to not be skipped. | |
271 | Videos without view count information are always | |
272 | downloaded. None for no limit. | |
273 | max_views: An integer representing the maximum view count. | |
274 | Videos that are more popular than that are not | |
275 | downloaded. | |
276 | Videos without view count information are always | |
277 | downloaded. None for no limit. | |
278 | download_archive: File name of a file where all downloads are recorded. | |
279 | Videos already present in the file are not downloaded | |
280 | again. | |
281 | break_on_existing: Stop the download process after attempting to download a | |
282 | file that is in the archive. | |
283 | break_on_reject: Stop the download process when encountering a video that | |
284 | has been filtered out. | |
285 | cookiefile: File name where cookies should be read from and dumped to | |
286 | nocheckcertificate:Do not verify SSL certificates | |
287 | prefer_insecure: Use HTTP instead of HTTPS to retrieve information. | |
288 | At the moment, this is only supported by YouTube. | |
289 | proxy: URL of the proxy server to use | |
290 | geo_verification_proxy: URL of the proxy to use for IP address verification | |
291 | on geo-restricted sites. | |
292 | socket_timeout: Time to wait for unresponsive hosts, in seconds | |
293 | bidi_workaround: Work around buggy terminals without bidirectional text | |
294 | support, using fridibi | |
295 | debug_printtraffic:Print out sent and received HTTP traffic | |
296 | include_ads: Download ads as well | |
297 | default_search: Prepend this string if an input url is not valid. | |
298 | 'auto' for elaborate guessing | |
299 | encoding: Use this encoding instead of the system-specified. | |
300 | extract_flat: Do not resolve URLs, return the immediate result. | |
301 | Pass in 'in_playlist' to only show this behavior for | |
302 | playlist items. | |
303 | postprocessors: A list of dictionaries, each with an entry | |
304 | * key: The name of the postprocessor. See | |
305 | yt_dlp/postprocessor/__init__.py for a list. | |
306 | * when: When to run the postprocessor. Can be one of | |
307 | pre_process|before_dl|post_process|after_move. | |
308 | Assumed to be 'post_process' if not given | |
309 | post_hooks: A list of functions that get called as the final step | |
310 | for each video file, after all postprocessors have been | |
311 | called. The filename will be passed as the only argument. | |
312 | progress_hooks: A list of functions that get called on download | |
313 | progress, with a dictionary with the entries | |
314 | * status: One of "downloading", "error", or "finished". | |
315 | Check this first and ignore unknown values. | |
316 | ||
317 | If status is one of "downloading", or "finished", the | |
318 | following properties may also be present: | |
319 | * filename: The final filename (always present) | |
320 | * tmpfilename: The filename we're currently writing to | |
321 | * downloaded_bytes: Bytes on disk | |
322 | * total_bytes: Size of the whole file, None if unknown | |
323 | * total_bytes_estimate: Guess of the eventual file size, | |
324 | None if unavailable. | |
325 | * elapsed: The number of seconds since download started. | |
326 | * eta: The estimated time in seconds, None if unknown | |
327 | * speed: The download speed in bytes/second, None if | |
328 | unknown | |
329 | * fragment_index: The counter of the currently | |
330 | downloaded video fragment. | |
331 | * fragment_count: The number of fragments (= individual | |
332 | files that will be merged) | |
333 | ||
334 | Progress hooks are guaranteed to be called at least once | |
335 | (with status "finished") if the download is successful. | |
336 | merge_output_format: Extension to use when merging formats. | |
337 | final_ext: Expected final extension; used to detect when the file was | |
338 | already downloaded and converted. "merge_output_format" is | |
339 | replaced by this extension when given | |
340 | fixup: Automatically correct known faults of the file. | |
341 | One of: | |
342 | - "never": do nothing | |
343 | - "warn": only emit a warning | |
344 | - "detect_or_warn": check whether we can do anything | |
345 | about it, warn otherwise (default) | |
346 | source_address: Client-side IP address to bind to. | |
347 | call_home: Boolean, true iff we are allowed to contact the | |
348 | yt-dlp servers for debugging. (BROKEN) | |
349 | sleep_interval_requests: Number of seconds to sleep between requests | |
350 | during extraction | |
351 | sleep_interval: Number of seconds to sleep before each download when | |
352 | used alone or a lower bound of a range for randomized | |
353 | sleep before each download (minimum possible number | |
354 | of seconds to sleep) when used along with | |
355 | max_sleep_interval. | |
356 | max_sleep_interval:Upper bound of a range for randomized sleep before each | |
357 | download (maximum possible number of seconds to sleep). | |
358 | Must only be used along with sleep_interval. | |
359 | Actual sleep time will be a random float from range | |
360 | [sleep_interval; max_sleep_interval]. | |
361 | sleep_interval_subtitles: Number of seconds to sleep before each subtitle download | |
362 | listformats: Print an overview of available video formats and exit. | |
363 | list_thumbnails: Print a table of all thumbnails and exit. | |
364 | match_filter: A function that gets called with the info_dict of | |
365 | every video. | |
366 | If it returns a message, the video is ignored. | |
367 | If it returns None, the video is downloaded. | |
368 | match_filter_func in utils.py is one example for this. | |
369 | no_color: Do not emit color codes in output. | |
370 | geo_bypass: Bypass geographic restriction via faking X-Forwarded-For | |
371 | HTTP header | |
372 | geo_bypass_country: | |
373 | Two-letter ISO 3166-2 country code that will be used for | |
374 | explicit geographic restriction bypassing via faking | |
375 | X-Forwarded-For HTTP header | |
376 | geo_bypass_ip_block: | |
377 | IP range in CIDR notation that will be used similarly to | |
378 | geo_bypass_country | |
379 | ||
380 | The following options determine which downloader is picked: | |
381 | external_downloader: A dictionary of protocol keys and the executable of the | |
382 | external downloader to use for it. The allowed protocols | |
383 | are default|http|ftp|m3u8|dash|rtsp|rtmp|mms. | |
384 | Set the value to 'native' to use the native downloader | |
385 | hls_prefer_native: Deprecated - Use external_downloader = {'m3u8': 'native'} | |
386 | or {'m3u8': 'ffmpeg'} instead. | |
387 | Use the native HLS downloader instead of ffmpeg/avconv | |
388 | if True, otherwise use ffmpeg/avconv if False, otherwise | |
389 | use downloader suggested by extractor if None. | |
390 | compat_opts: Compatibility options. See "Differences in default behavior". | |
391 | Note that only format-sort, format-spec, no-live-chat, | |
392 | no-attach-info-json, playlist-index, list-formats, | |
393 | no-direct-merge, no-youtube-channel-redirect, | |
394 | and no-youtube-unavailable-videos works when used via the API | |
395 | ||
396 | The following parameters are not used by YoutubeDL itself, they are used by | |
397 | the downloader (see yt_dlp/downloader/common.py): | |
398 | nopart, updatetime, buffersize, ratelimit, min_filesize, max_filesize, test, | |
399 | noresizebuffer, retries, continuedl, noprogress, consoletitle, | |
400 | xattr_set_filesize, external_downloader_args, hls_use_mpegts, | |
401 | http_chunk_size. | |
402 | ||
403 | The following options are used by the post processors: | |
404 | prefer_ffmpeg: If False, use avconv instead of ffmpeg if both are available, | |
405 | otherwise prefer ffmpeg. (avconv support is deprecated) | |
406 | ffmpeg_location: Location of the ffmpeg/avconv binary; either the path | |
407 | to the binary or its containing directory. | |
408 | postprocessor_args: A dictionary of postprocessor/executable keys (in lower case) | |
409 | and a list of additional command-line arguments for the | |
410 | postprocessor/executable. The dict can also have "PP+EXE" keys | |
411 | which are used when the given exe is used by the given PP. | |
412 | Use 'default' as the name for arguments to passed to all PP | |
413 | ||
414 | The following options are used by the extractors: | |
415 | extractor_retries: Number of times to retry for known errors | |
416 | dynamic_mpd: Whether to process dynamic DASH manifests (default: True) | |
417 | hls_split_discontinuity: Split HLS playlists to different formats at | |
418 | discontinuities such as ad breaks (default: False) | |
419 | youtube_include_dash_manifest: If True (default), DASH manifests and related | |
420 | data will be downloaded and processed by extractor. | |
421 | You can reduce network I/O by disabling it if you don't | |
422 | care about DASH. (only for youtube) | |
423 | youtube_include_hls_manifest: If True (default), HLS manifests and related | |
424 | data will be downloaded and processed by extractor. | |
425 | You can reduce network I/O by disabling it if you don't | |
426 | care about HLS. (only for youtube) | |
427 | """ | |
428 | ||
429 | _NUMERIC_FIELDS = set(( | |
430 | 'width', 'height', 'tbr', 'abr', 'asr', 'vbr', 'fps', 'filesize', 'filesize_approx', | |
431 | 'timestamp', 'upload_year', 'upload_month', 'upload_day', | |
432 | 'duration', 'view_count', 'like_count', 'dislike_count', 'repost_count', | |
433 | 'average_rating', 'comment_count', 'age_limit', | |
434 | 'start_time', 'end_time', | |
435 | 'chapter_number', 'season_number', 'episode_number', | |
436 | 'track_number', 'disc_number', 'release_year', | |
437 | 'playlist_index', | |
438 | )) | |
439 | ||
440 | params = None | |
441 | _ies = [] | |
442 | _pps = {'pre_process': [], 'before_dl': [], 'after_move': [], 'post_process': []} | |
443 | __prepare_filename_warned = False | |
444 | _first_webpage_request = True | |
445 | _download_retcode = None | |
446 | _num_downloads = None | |
447 | _playlist_level = 0 | |
448 | _playlist_urls = set() | |
449 | _screen_file = None | |
450 | ||
451 | def __init__(self, params=None, auto_init=True): | |
452 | """Create a FileDownloader object with the given options.""" | |
453 | if params is None: | |
454 | params = {} | |
455 | self._ies = [] | |
456 | self._ies_instances = {} | |
457 | self._pps = {'pre_process': [], 'before_dl': [], 'after_move': [], 'post_process': []} | |
458 | self.__prepare_filename_warned = False | |
459 | self._first_webpage_request = True | |
460 | self._post_hooks = [] | |
461 | self._progress_hooks = [] | |
462 | self._download_retcode = 0 | |
463 | self._num_downloads = 0 | |
464 | self._screen_file = [sys.stdout, sys.stderr][params.get('logtostderr', False)] | |
465 | self._err_file = sys.stderr | |
466 | self.params = { | |
467 | # Default parameters | |
468 | 'nocheckcertificate': False, | |
469 | } | |
470 | self.params.update(params) | |
471 | self.cache = Cache(self) | |
472 | ||
473 | if sys.version_info < (3, 6): | |
474 | self.report_warning( | |
475 | 'Support for Python version %d.%d have been deprecated and will break in future versions of yt-dlp! ' | |
476 | 'Update to Python 3.6 or above' % sys.version_info[:2]) | |
477 | ||
478 | def check_deprecated(param, option, suggestion): | |
479 | if self.params.get(param) is not None: | |
480 | self.report_warning('%s is deprecated. Use %s instead' % (option, suggestion)) | |
481 | return True | |
482 | return False | |
483 | ||
484 | if check_deprecated('cn_verification_proxy', '--cn-verification-proxy', '--geo-verification-proxy'): | |
485 | if self.params.get('geo_verification_proxy') is None: | |
486 | self.params['geo_verification_proxy'] = self.params['cn_verification_proxy'] | |
487 | ||
488 | check_deprecated('autonumber', '--auto-number', '-o "%(autonumber)s-%(title)s.%(ext)s"') | |
489 | check_deprecated('usetitle', '--title', '-o "%(title)s-%(id)s.%(ext)s"') | |
490 | check_deprecated('useid', '--id', '-o "%(id)s.%(ext)s"') | |
491 | ||
492 | for msg in self.params.get('warnings', []): | |
493 | self.report_warning(msg) | |
494 | ||
495 | if self.params.get('final_ext'): | |
496 | if self.params.get('merge_output_format'): | |
497 | self.report_warning('--merge-output-format will be ignored since --remux-video or --recode-video is given') | |
498 | self.params['merge_output_format'] = self.params['final_ext'] | |
499 | ||
500 | if 'overwrites' in self.params and self.params['overwrites'] is None: | |
501 | del self.params['overwrites'] | |
502 | ||
503 | if params.get('bidi_workaround', False): | |
504 | try: | |
505 | import pty | |
506 | master, slave = pty.openpty() | |
507 | width = compat_get_terminal_size().columns | |
508 | if width is None: | |
509 | width_args = [] | |
510 | else: | |
511 | width_args = ['-w', str(width)] | |
512 | sp_kwargs = dict( | |
513 | stdin=subprocess.PIPE, | |
514 | stdout=slave, | |
515 | stderr=self._err_file) | |
516 | try: | |
517 | self._output_process = subprocess.Popen( | |
518 | ['bidiv'] + width_args, **sp_kwargs | |
519 | ) | |
520 | except OSError: | |
521 | self._output_process = subprocess.Popen( | |
522 | ['fribidi', '-c', 'UTF-8'] + width_args, **sp_kwargs) | |
523 | self._output_channel = os.fdopen(master, 'rb') | |
524 | except OSError as ose: | |
525 | if ose.errno == errno.ENOENT: | |
526 | self.report_warning('Could not find fribidi executable, ignoring --bidi-workaround . Make sure that fribidi is an executable file in one of the directories in your $PATH.') | |
527 | else: | |
528 | raise | |
529 | ||
530 | if (sys.platform != 'win32' | |
531 | and sys.getfilesystemencoding() in ['ascii', 'ANSI_X3.4-1968'] | |
532 | and not params.get('restrictfilenames', False)): | |
533 | # Unicode filesystem API will throw errors (#1474, #13027) | |
534 | self.report_warning( | |
535 | 'Assuming --restrict-filenames since file system encoding ' | |
536 | 'cannot encode all characters. ' | |
537 | 'Set the LC_ALL environment variable to fix this.') | |
538 | self.params['restrictfilenames'] = True | |
539 | ||
540 | self.outtmpl_dict = self.parse_outtmpl() | |
541 | ||
542 | self._setup_opener() | |
543 | ||
544 | """Preload the archive, if any is specified""" | |
545 | def preload_download_archive(fn): | |
546 | if fn is None: | |
547 | return False | |
548 | self.write_debug('Loading archive file %r\n' % fn) | |
549 | try: | |
550 | with locked_file(fn, 'r', encoding='utf-8') as archive_file: | |
551 | for line in archive_file: | |
552 | self.archive.add(line.strip()) | |
553 | except IOError as ioe: | |
554 | if ioe.errno != errno.ENOENT: | |
555 | raise | |
556 | return False | |
557 | return True | |
558 | ||
559 | self.archive = set() | |
560 | preload_download_archive(self.params.get('download_archive')) | |
561 | ||
562 | if auto_init: | |
563 | self.print_debug_header() | |
564 | self.add_default_info_extractors() | |
565 | ||
566 | for pp_def_raw in self.params.get('postprocessors', []): | |
567 | pp_class = get_postprocessor(pp_def_raw['key']) | |
568 | pp_def = dict(pp_def_raw) | |
569 | del pp_def['key'] | |
570 | if 'when' in pp_def: | |
571 | when = pp_def['when'] | |
572 | del pp_def['when'] | |
573 | else: | |
574 | when = 'post_process' | |
575 | pp = pp_class(self, **compat_kwargs(pp_def)) | |
576 | self.add_post_processor(pp, when=when) | |
577 | ||
578 | for ph in self.params.get('post_hooks', []): | |
579 | self.add_post_hook(ph) | |
580 | ||
581 | for ph in self.params.get('progress_hooks', []): | |
582 | self.add_progress_hook(ph) | |
583 | ||
584 | register_socks_protocols() | |
585 | ||
586 | def warn_if_short_id(self, argv): | |
587 | # short YouTube ID starting with dash? | |
588 | idxs = [ | |
589 | i for i, a in enumerate(argv) | |
590 | if re.match(r'^-[0-9A-Za-z_-]{10}$', a)] | |
591 | if idxs: | |
592 | correct_argv = ( | |
593 | ['yt-dlp'] | |
594 | + [a for i, a in enumerate(argv) if i not in idxs] | |
595 | + ['--'] + [argv[i] for i in idxs] | |
596 | ) | |
597 | self.report_warning( | |
598 | 'Long argument string detected. ' | |
599 | 'Use -- to separate parameters and URLs, like this:\n%s\n' % | |
600 | args_to_str(correct_argv)) | |
601 | ||
602 | def add_info_extractor(self, ie): | |
603 | """Add an InfoExtractor object to the end of the list.""" | |
604 | self._ies.append(ie) | |
605 | if not isinstance(ie, type): | |
606 | self._ies_instances[ie.ie_key()] = ie | |
607 | ie.set_downloader(self) | |
608 | ||
609 | def get_info_extractor(self, ie_key): | |
610 | """ | |
611 | Get an instance of an IE with name ie_key, it will try to get one from | |
612 | the _ies list, if there's no instance it will create a new one and add | |
613 | it to the extractor list. | |
614 | """ | |
615 | ie = self._ies_instances.get(ie_key) | |
616 | if ie is None: | |
617 | ie = get_info_extractor(ie_key)() | |
618 | self.add_info_extractor(ie) | |
619 | return ie | |
620 | ||
621 | def add_default_info_extractors(self): | |
622 | """ | |
623 | Add the InfoExtractors returned by gen_extractors to the end of the list | |
624 | """ | |
625 | for ie in gen_extractor_classes(): | |
626 | self.add_info_extractor(ie) | |
627 | ||
628 | def add_post_processor(self, pp, when='post_process'): | |
629 | """Add a PostProcessor object to the end of the chain.""" | |
630 | self._pps[when].append(pp) | |
631 | pp.set_downloader(self) | |
632 | ||
633 | def add_post_hook(self, ph): | |
634 | """Add the post hook""" | |
635 | self._post_hooks.append(ph) | |
636 | ||
637 | def add_progress_hook(self, ph): | |
638 | """Add the progress hook (currently only for the file downloader)""" | |
639 | self._progress_hooks.append(ph) | |
640 | ||
641 | def _bidi_workaround(self, message): | |
642 | if not hasattr(self, '_output_channel'): | |
643 | return message | |
644 | ||
645 | assert hasattr(self, '_output_process') | |
646 | assert isinstance(message, compat_str) | |
647 | line_count = message.count('\n') + 1 | |
648 | self._output_process.stdin.write((message + '\n').encode('utf-8')) | |
649 | self._output_process.stdin.flush() | |
650 | res = ''.join(self._output_channel.readline().decode('utf-8') | |
651 | for _ in range(line_count)) | |
652 | return res[:-len('\n')] | |
653 | ||
654 | def _write_string(self, s, out=None): | |
655 | write_string(s, out=out, encoding=self.params.get('encoding')) | |
656 | ||
657 | def to_stdout(self, message, skip_eol=False, quiet=False): | |
658 | """Print message to stdout""" | |
659 | if self.params.get('logger'): | |
660 | self.params['logger'].debug(message) | |
661 | elif not quiet or self.params.get('verbose'): | |
662 | self._write_string( | |
663 | '%s%s' % (self._bidi_workaround(message), ('' if skip_eol else '\n')), | |
664 | self._err_file if quiet else self._screen_file) | |
665 | ||
666 | def to_stderr(self, message): | |
667 | """Print message to stderr""" | |
668 | assert isinstance(message, compat_str) | |
669 | if self.params.get('logger'): | |
670 | self.params['logger'].error(message) | |
671 | else: | |
672 | self._write_string('%s\n' % self._bidi_workaround(message), self._err_file) | |
673 | ||
674 | def to_console_title(self, message): | |
675 | if not self.params.get('consoletitle', False): | |
676 | return | |
677 | if compat_os_name == 'nt': | |
678 | if ctypes.windll.kernel32.GetConsoleWindow(): | |
679 | # c_wchar_p() might not be necessary if `message` is | |
680 | # already of type unicode() | |
681 | ctypes.windll.kernel32.SetConsoleTitleW(ctypes.c_wchar_p(message)) | |
682 | elif 'TERM' in os.environ: | |
683 | self._write_string('\033]0;%s\007' % message, self._screen_file) | |
684 | ||
685 | def save_console_title(self): | |
686 | if not self.params.get('consoletitle', False): | |
687 | return | |
688 | if self.params.get('simulate', False): | |
689 | return | |
690 | if compat_os_name != 'nt' and 'TERM' in os.environ: | |
691 | # Save the title on stack | |
692 | self._write_string('\033[22;0t', self._screen_file) | |
693 | ||
694 | def restore_console_title(self): | |
695 | if not self.params.get('consoletitle', False): | |
696 | return | |
697 | if self.params.get('simulate', False): | |
698 | return | |
699 | if compat_os_name != 'nt' and 'TERM' in os.environ: | |
700 | # Restore the title from stack | |
701 | self._write_string('\033[23;0t', self._screen_file) | |
702 | ||
703 | def __enter__(self): | |
704 | self.save_console_title() | |
705 | return self | |
706 | ||
707 | def __exit__(self, *args): | |
708 | self.restore_console_title() | |
709 | ||
710 | if self.params.get('cookiefile') is not None: | |
711 | self.cookiejar.save(ignore_discard=True, ignore_expires=True) | |
712 | ||
713 | def trouble(self, message=None, tb=None): | |
714 | """Determine action to take when a download problem appears. | |
715 | ||
716 | Depending on if the downloader has been configured to ignore | |
717 | download errors or not, this method may throw an exception or | |
718 | not when errors are found, after printing the message. | |
719 | ||
720 | tb, if given, is additional traceback information. | |
721 | """ | |
722 | if message is not None: | |
723 | self.to_stderr(message) | |
724 | if self.params.get('verbose'): | |
725 | if tb is None: | |
726 | if sys.exc_info()[0]: # if .trouble has been called from an except block | |
727 | tb = '' | |
728 | if hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]: | |
729 | tb += ''.join(traceback.format_exception(*sys.exc_info()[1].exc_info)) | |
730 | tb += encode_compat_str(traceback.format_exc()) | |
731 | else: | |
732 | tb_data = traceback.format_list(traceback.extract_stack()) | |
733 | tb = ''.join(tb_data) | |
734 | if tb: | |
735 | self.to_stderr(tb) | |
736 | if not self.params.get('ignoreerrors', False): | |
737 | if sys.exc_info()[0] and hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]: | |
738 | exc_info = sys.exc_info()[1].exc_info | |
739 | else: | |
740 | exc_info = sys.exc_info() | |
741 | raise DownloadError(message, exc_info) | |
742 | self._download_retcode = 1 | |
743 | ||
744 | def to_screen(self, message, skip_eol=False): | |
745 | """Print message to stdout if not in quiet mode""" | |
746 | self.to_stdout( | |
747 | message, skip_eol, quiet=self.params.get('quiet', False)) | |
748 | ||
749 | def report_warning(self, message): | |
750 | ''' | |
751 | Print the message to stderr, it will be prefixed with 'WARNING:' | |
752 | If stderr is a tty file the 'WARNING:' will be colored | |
753 | ''' | |
754 | if self.params.get('logger') is not None: | |
755 | self.params['logger'].warning(message) | |
756 | else: | |
757 | if self.params.get('no_warnings'): | |
758 | return | |
759 | if not self.params.get('no_color') and self._err_file.isatty() and compat_os_name != 'nt': | |
760 | _msg_header = '\033[0;33mWARNING:\033[0m' | |
761 | else: | |
762 | _msg_header = 'WARNING:' | |
763 | warning_message = '%s %s' % (_msg_header, message) | |
764 | self.to_stderr(warning_message) | |
765 | ||
766 | def report_error(self, message, tb=None): | |
767 | ''' | |
768 | Do the same as trouble, but prefixes the message with 'ERROR:', colored | |
769 | in red if stderr is a tty file. | |
770 | ''' | |
771 | if not self.params.get('no_color') and self._err_file.isatty() and compat_os_name != 'nt': | |
772 | _msg_header = '\033[0;31mERROR:\033[0m' | |
773 | else: | |
774 | _msg_header = 'ERROR:' | |
775 | error_message = '%s %s' % (_msg_header, message) | |
776 | self.trouble(error_message, tb) | |
777 | ||
778 | def write_debug(self, message): | |
779 | '''Log debug message or Print message to stderr''' | |
780 | if not self.params.get('verbose', False): | |
781 | return | |
782 | message = '[debug] %s' % message | |
783 | if self.params.get('logger'): | |
784 | self.params['logger'].debug(message) | |
785 | else: | |
786 | self._write_string('%s\n' % message) | |
787 | ||
788 | def report_file_already_downloaded(self, file_name): | |
789 | """Report file has already been fully downloaded.""" | |
790 | try: | |
791 | self.to_screen('[download] %s has already been downloaded' % file_name) | |
792 | except UnicodeEncodeError: | |
793 | self.to_screen('[download] The file has already been downloaded') | |
794 | ||
795 | def report_file_delete(self, file_name): | |
796 | """Report that existing file will be deleted.""" | |
797 | try: | |
798 | self.to_screen('Deleting existing file %s' % file_name) | |
799 | except UnicodeEncodeError: | |
800 | self.to_screen('Deleting existing file') | |
801 | ||
802 | def parse_outtmpl(self): | |
803 | outtmpl_dict = self.params.get('outtmpl', {}) | |
804 | if not isinstance(outtmpl_dict, dict): | |
805 | outtmpl_dict = {'default': outtmpl_dict} | |
806 | outtmpl_dict.update({ | |
807 | k: v for k, v in DEFAULT_OUTTMPL.items() | |
808 | if not outtmpl_dict.get(k)}) | |
809 | for key, val in outtmpl_dict.items(): | |
810 | if isinstance(val, bytes): | |
811 | self.report_warning( | |
812 | 'Parameter outtmpl is bytes, but should be a unicode string. ' | |
813 | 'Put from __future__ import unicode_literals at the top of your code file or consider switching to Python 3.x.') | |
814 | return outtmpl_dict | |
815 | ||
816 | def prepare_outtmpl(self, outtmpl, info_dict, sanitize=None): | |
817 | """ Make the template and info_dict suitable for substitution (outtmpl % info_dict)""" | |
818 | template_dict = dict(info_dict) | |
819 | na = self.params.get('outtmpl_na_placeholder', 'NA') | |
820 | ||
821 | # duration_string | |
822 | template_dict['duration_string'] = ( # %(duration>%H-%M-%S)s is wrong if duration > 24hrs | |
823 | formatSeconds(info_dict['duration'], '-' if sanitize else ':') | |
824 | if info_dict.get('duration', None) is not None | |
825 | else None) | |
826 | ||
827 | # epoch | |
828 | template_dict['epoch'] = int(time.time()) | |
829 | ||
830 | # autonumber | |
831 | autonumber_size = self.params.get('autonumber_size') | |
832 | if autonumber_size is None: | |
833 | autonumber_size = 5 | |
834 | template_dict['autonumber'] = self.params.get('autonumber_start', 1) - 1 + self._num_downloads | |
835 | ||
836 | # resolution if not defined | |
837 | if template_dict.get('resolution') is None: | |
838 | if template_dict.get('width') and template_dict.get('height'): | |
839 | template_dict['resolution'] = '%dx%d' % (template_dict['width'], template_dict['height']) | |
840 | elif template_dict.get('height'): | |
841 | template_dict['resolution'] = '%sp' % template_dict['height'] | |
842 | elif template_dict.get('width'): | |
843 | template_dict['resolution'] = '%dx?' % template_dict['width'] | |
844 | ||
845 | # For fields playlist_index and autonumber convert all occurrences | |
846 | # of %(field)s to %(field)0Nd for backward compatibility | |
847 | field_size_compat_map = { | |
848 | 'playlist_index': len(str(template_dict.get('_last_playlist_index') or '')), | |
849 | 'autonumber': autonumber_size, | |
850 | } | |
851 | FIELD_SIZE_COMPAT_RE = r'(?<!%)%\((?P<field>autonumber|playlist_index)\)s' | |
852 | mobj = re.search(FIELD_SIZE_COMPAT_RE, outtmpl) | |
853 | if mobj: | |
854 | outtmpl = re.sub( | |
855 | FIELD_SIZE_COMPAT_RE, | |
856 | r'%%(\1)0%dd' % field_size_compat_map[mobj.group('field')], | |
857 | outtmpl) | |
858 | ||
859 | numeric_fields = list(self._NUMERIC_FIELDS) | |
860 | if sanitize is None: | |
861 | sanitize = lambda k, v: v | |
862 | ||
863 | EXTERNAL_FORMAT_RE = FORMAT_RE.format('(?P<key>[^)]*)') | |
864 | # Field is of the form key1.key2... | |
865 | # where keys (except first) can be string, int or slice | |
866 | FIELD_RE = r'\w+(?:\.(?:\w+|[-\d]*(?::[-\d]*){0,2}))*' | |
867 | INTERNAL_FORMAT_RE = re.compile(r'''(?x) | |
868 | (?P<negate>-)? | |
869 | (?P<fields>{0}) | |
870 | (?P<maths>(?:[-+]-?(?:\d+(?:\.\d+)?|{0}))*) | |
871 | (?:>(?P<strf_format>.+?))? | |
872 | (?:\|(?P<default>.*?))? | |
873 | $'''.format(FIELD_RE)) | |
874 | MATH_OPERATORS_RE = re.compile(r'(?<![-+])([-+])') | |
875 | MATH_FUNCTIONS = { | |
876 | '+': float.__add__, | |
877 | '-': float.__sub__, | |
878 | } | |
879 | for outer_mobj in re.finditer(EXTERNAL_FORMAT_RE, outtmpl): | |
880 | final_key = outer_mobj.group('key') | |
881 | str_type = outer_mobj.group('type') | |
882 | value = None | |
883 | mobj = re.match(INTERNAL_FORMAT_RE, final_key) | |
884 | if mobj is not None: | |
885 | mobj = mobj.groupdict() | |
886 | # Object traversal | |
887 | fields = mobj['fields'].split('.') | |
888 | value = traverse_dict(template_dict, fields) | |
889 | # Negative | |
890 | if mobj['negate']: | |
891 | value = float_or_none(value) | |
892 | if value is not None: | |
893 | value *= -1 | |
894 | # Do maths | |
895 | if mobj['maths']: | |
896 | value = float_or_none(value) | |
897 | operator = None | |
898 | for item in MATH_OPERATORS_RE.split(mobj['maths'])[1:]: | |
899 | if item == '': | |
900 | value = None | |
901 | if value is None: | |
902 | break | |
903 | if operator: | |
904 | item, multiplier = (item[1:], -1) if item[0] == '-' else (item, 1) | |
905 | offset = float_or_none(item) | |
906 | if offset is None: | |
907 | offset = float_or_none(traverse_dict(template_dict, item.split('.'))) | |
908 | try: | |
909 | value = operator(value, multiplier * offset) | |
910 | except (TypeError, ZeroDivisionError): | |
911 | value = None | |
912 | operator = None | |
913 | else: | |
914 | operator = MATH_FUNCTIONS[item] | |
915 | # Datetime formatting | |
916 | if mobj['strf_format']: | |
917 | value = strftime_or_none(value, mobj['strf_format']) | |
918 | # Set default | |
919 | if value is None and mobj['default'] is not None: | |
920 | value = mobj['default'] | |
921 | # Sanitize | |
922 | if str_type in 'crs' and value is not None: # string | |
923 | value = sanitize('%{}'.format(str_type) % fields[-1], value) | |
924 | else: # numeric | |
925 | numeric_fields.append(final_key) | |
926 | value = float_or_none(value) | |
927 | if value is not None: | |
928 | template_dict[final_key] = value | |
929 | ||
930 | # Missing numeric fields used together with integer presentation types | |
931 | # in format specification will break the argument substitution since | |
932 | # string NA placeholder is returned for missing fields. We will patch | |
933 | # output template for missing fields to meet string presentation type. | |
934 | for numeric_field in numeric_fields: | |
935 | if template_dict.get(numeric_field) is None: | |
936 | outtmpl = re.sub( | |
937 | FORMAT_RE.format(re.escape(numeric_field)), | |
938 | r'%({0})s'.format(numeric_field), outtmpl) | |
939 | ||
940 | template_dict = collections.defaultdict(lambda: na, ( | |
941 | (k, v if isinstance(v, compat_numeric_types) else sanitize(k, v)) | |
942 | for k, v in template_dict.items() if v is not None)) | |
943 | return outtmpl, template_dict | |
944 | ||
945 | def _prepare_filename(self, info_dict, tmpl_type='default'): | |
946 | try: | |
947 | sanitize = lambda k, v: sanitize_filename( | |
948 | compat_str(v), | |
949 | restricted=self.params.get('restrictfilenames'), | |
950 | is_id=(k == 'id' or k.endswith('_id'))) | |
951 | outtmpl = self.outtmpl_dict.get(tmpl_type, self.outtmpl_dict['default']) | |
952 | outtmpl, template_dict = self.prepare_outtmpl(outtmpl, info_dict, sanitize) | |
953 | ||
954 | # expand_path translates '%%' into '%' and '$$' into '$' | |
955 | # correspondingly that is not what we want since we need to keep | |
956 | # '%%' intact for template dict substitution step. Working around | |
957 | # with boundary-alike separator hack. | |
958 | sep = ''.join([random.choice(ascii_letters) for _ in range(32)]) | |
959 | outtmpl = outtmpl.replace('%%', '%{0}%'.format(sep)).replace('$$', '${0}$'.format(sep)) | |
960 | ||
961 | # outtmpl should be expand_path'ed before template dict substitution | |
962 | # because meta fields may contain env variables we don't want to | |
963 | # be expanded. For example, for outtmpl "%(title)s.%(ext)s" and | |
964 | # title "Hello $PATH", we don't want `$PATH` to be expanded. | |
965 | filename = expand_path(outtmpl).replace(sep, '') % template_dict | |
966 | ||
967 | force_ext = OUTTMPL_TYPES.get(tmpl_type) | |
968 | if force_ext is not None: | |
969 | filename = replace_extension(filename, force_ext, template_dict.get('ext')) | |
970 | ||
971 | # https://github.com/blackjack4494/youtube-dlc/issues/85 | |
972 | trim_file_name = self.params.get('trim_file_name', False) | |
973 | if trim_file_name: | |
974 | fn_groups = filename.rsplit('.') | |
975 | ext = fn_groups[-1] | |
976 | sub_ext = '' | |
977 | if len(fn_groups) > 2: | |
978 | sub_ext = fn_groups[-2] | |
979 | filename = '.'.join(filter(None, [fn_groups[0][:trim_file_name], sub_ext, ext])) | |
980 | ||
981 | return filename | |
982 | except ValueError as err: | |
983 | self.report_error('Error in output template: ' + str(err) + ' (encoding: ' + repr(preferredencoding()) + ')') | |
984 | return None | |
985 | ||
986 | def prepare_filename(self, info_dict, dir_type='', warn=False): | |
987 | """Generate the output filename.""" | |
988 | paths = self.params.get('paths', {}) | |
989 | assert isinstance(paths, dict) | |
990 | filename = self._prepare_filename(info_dict, dir_type or 'default') | |
991 | ||
992 | if warn and not self.__prepare_filename_warned: | |
993 | if not paths: | |
994 | pass | |
995 | elif filename == '-': | |
996 | self.report_warning('--paths is ignored when an outputting to stdout') | |
997 | elif os.path.isabs(filename): | |
998 | self.report_warning('--paths is ignored since an absolute path is given in output template') | |
999 | self.__prepare_filename_warned = True | |
1000 | if filename == '-' or not filename: | |
1001 | return filename | |
1002 | ||
1003 | homepath = expand_path(paths.get('home', '').strip()) | |
1004 | assert isinstance(homepath, compat_str) | |
1005 | subdir = expand_path(paths.get(dir_type, '').strip()) if dir_type else '' | |
1006 | assert isinstance(subdir, compat_str) | |
1007 | path = os.path.join(homepath, subdir, filename) | |
1008 | ||
1009 | # Temporary fix for #4787 | |
1010 | # 'Treat' all problem characters by passing filename through preferredencoding | |
1011 | # to workaround encoding issues with subprocess on python2 @ Windows | |
1012 | if sys.version_info < (3, 0) and sys.platform == 'win32': | |
1013 | path = encodeFilename(path, True).decode(preferredencoding()) | |
1014 | return sanitize_path(path, force=self.params.get('windowsfilenames')) | |
1015 | ||
1016 | def _match_entry(self, info_dict, incomplete=False, silent=False): | |
1017 | """ Returns None if the file should be downloaded """ | |
1018 | ||
1019 | video_title = info_dict.get('title', info_dict.get('id', 'video')) | |
1020 | ||
1021 | def check_filter(): | |
1022 | if 'title' in info_dict: | |
1023 | # This can happen when we're just evaluating the playlist | |
1024 | title = info_dict['title'] | |
1025 | matchtitle = self.params.get('matchtitle', False) | |
1026 | if matchtitle: | |
1027 | if not re.search(matchtitle, title, re.IGNORECASE): | |
1028 | return '"' + title + '" title did not match pattern "' + matchtitle + '"' | |
1029 | rejecttitle = self.params.get('rejecttitle', False) | |
1030 | if rejecttitle: | |
1031 | if re.search(rejecttitle, title, re.IGNORECASE): | |
1032 | return '"' + title + '" title matched reject pattern "' + rejecttitle + '"' | |
1033 | date = info_dict.get('upload_date') | |
1034 | if date is not None: | |
1035 | dateRange = self.params.get('daterange', DateRange()) | |
1036 | if date not in dateRange: | |
1037 | return '%s upload date is not in range %s' % (date_from_str(date).isoformat(), dateRange) | |
1038 | view_count = info_dict.get('view_count') | |
1039 | if view_count is not None: | |
1040 | min_views = self.params.get('min_views') | |
1041 | if min_views is not None and view_count < min_views: | |
1042 | return 'Skipping %s, because it has not reached minimum view count (%d/%d)' % (video_title, view_count, min_views) | |
1043 | max_views = self.params.get('max_views') | |
1044 | if max_views is not None and view_count > max_views: | |
1045 | return 'Skipping %s, because it has exceeded the maximum view count (%d/%d)' % (video_title, view_count, max_views) | |
1046 | if age_restricted(info_dict.get('age_limit'), self.params.get('age_limit')): | |
1047 | return 'Skipping "%s" because it is age restricted' % video_title | |
1048 | ||
1049 | if not incomplete: | |
1050 | match_filter = self.params.get('match_filter') | |
1051 | if match_filter is not None: | |
1052 | ret = match_filter(info_dict) | |
1053 | if ret is not None: | |
1054 | return ret | |
1055 | return None | |
1056 | ||
1057 | if self.in_download_archive(info_dict): | |
1058 | reason = '%s has already been recorded in the archive' % video_title | |
1059 | break_opt, break_err = 'break_on_existing', ExistingVideoReached | |
1060 | else: | |
1061 | reason = check_filter() | |
1062 | break_opt, break_err = 'break_on_reject', RejectedVideoReached | |
1063 | if reason is not None: | |
1064 | if not silent: | |
1065 | self.to_screen('[download] ' + reason) | |
1066 | if self.params.get(break_opt, False): | |
1067 | raise break_err() | |
1068 | return reason | |
1069 | ||
1070 | @staticmethod | |
1071 | def add_extra_info(info_dict, extra_info): | |
1072 | '''Set the keys from extra_info in info dict if they are missing''' | |
1073 | for key, value in extra_info.items(): | |
1074 | info_dict.setdefault(key, value) | |
1075 | ||
1076 | def extract_info(self, url, download=True, ie_key=None, extra_info={}, | |
1077 | process=True, force_generic_extractor=False): | |
1078 | """ | |
1079 | Return a list with a dictionary for each video extracted. | |
1080 | ||
1081 | Arguments: | |
1082 | url -- URL to extract | |
1083 | ||
1084 | Keyword arguments: | |
1085 | download -- whether to download videos during extraction | |
1086 | ie_key -- extractor key hint | |
1087 | extra_info -- dictionary containing the extra values to add to each result | |
1088 | process -- whether to resolve all unresolved references (URLs, playlist items), | |
1089 | must be True for download to work. | |
1090 | force_generic_extractor -- force using the generic extractor | |
1091 | """ | |
1092 | ||
1093 | if not ie_key and force_generic_extractor: | |
1094 | ie_key = 'Generic' | |
1095 | ||
1096 | if ie_key: | |
1097 | ies = [self.get_info_extractor(ie_key)] | |
1098 | else: | |
1099 | ies = self._ies | |
1100 | ||
1101 | for ie in ies: | |
1102 | if not ie.suitable(url): | |
1103 | continue | |
1104 | ||
1105 | ie_key = ie.ie_key() | |
1106 | ie = self.get_info_extractor(ie_key) | |
1107 | if not ie.working(): | |
1108 | self.report_warning('The program functionality for this site has been marked as broken, ' | |
1109 | 'and will probably not work.') | |
1110 | ||
1111 | try: | |
1112 | temp_id = str_or_none( | |
1113 | ie.extract_id(url) if callable(getattr(ie, 'extract_id', None)) | |
1114 | else ie._match_id(url)) | |
1115 | except (AssertionError, IndexError, AttributeError): | |
1116 | temp_id = None | |
1117 | if temp_id is not None and self.in_download_archive({'id': temp_id, 'ie_key': ie_key}): | |
1118 | self.to_screen("[%s] %s: has already been recorded in archive" % ( | |
1119 | ie_key, temp_id)) | |
1120 | break | |
1121 | return self.__extract_info(url, ie, download, extra_info, process) | |
1122 | else: | |
1123 | self.report_error('no suitable InfoExtractor for URL %s' % url) | |
1124 | ||
1125 | def __handle_extraction_exceptions(func): | |
1126 | def wrapper(self, *args, **kwargs): | |
1127 | try: | |
1128 | return func(self, *args, **kwargs) | |
1129 | except GeoRestrictedError as e: | |
1130 | msg = e.msg | |
1131 | if e.countries: | |
1132 | msg += '\nThis video is available in %s.' % ', '.join( | |
1133 | map(ISO3166Utils.short2full, e.countries)) | |
1134 | msg += '\nYou might want to use a VPN or a proxy server (with --proxy) to workaround.' | |
1135 | self.report_error(msg) | |
1136 | except ExtractorError as e: # An error we somewhat expected | |
1137 | self.report_error(compat_str(e), e.format_traceback()) | |
1138 | except (MaxDownloadsReached, ExistingVideoReached, RejectedVideoReached): | |
1139 | raise | |
1140 | except Exception as e: | |
1141 | if self.params.get('ignoreerrors', False): | |
1142 | self.report_error(error_to_compat_str(e), tb=encode_compat_str(traceback.format_exc())) | |
1143 | else: | |
1144 | raise | |
1145 | return wrapper | |
1146 | ||
1147 | @__handle_extraction_exceptions | |
1148 | def __extract_info(self, url, ie, download, extra_info, process): | |
1149 | ie_result = ie.extract(url) | |
1150 | if ie_result is None: # Finished already (backwards compatibility; listformats and friends should be moved here) | |
1151 | return | |
1152 | if isinstance(ie_result, list): | |
1153 | # Backwards compatibility: old IE result format | |
1154 | ie_result = { | |
1155 | '_type': 'compat_list', | |
1156 | 'entries': ie_result, | |
1157 | } | |
1158 | self.add_default_extra_info(ie_result, ie, url) | |
1159 | if process: | |
1160 | return self.process_ie_result(ie_result, download, extra_info) | |
1161 | else: | |
1162 | return ie_result | |
1163 | ||
1164 | def add_default_extra_info(self, ie_result, ie, url): | |
1165 | self.add_extra_info(ie_result, { | |
1166 | 'extractor': ie.IE_NAME, | |
1167 | 'webpage_url': url, | |
1168 | 'webpage_url_basename': url_basename(url), | |
1169 | 'extractor_key': ie.ie_key(), | |
1170 | }) | |
1171 | ||
1172 | def process_ie_result(self, ie_result, download=True, extra_info={}): | |
1173 | """ | |
1174 | Take the result of the ie(may be modified) and resolve all unresolved | |
1175 | references (URLs, playlist items). | |
1176 | ||
1177 | It will also download the videos if 'download'. | |
1178 | Returns the resolved ie_result. | |
1179 | """ | |
1180 | result_type = ie_result.get('_type', 'video') | |
1181 | ||
1182 | if result_type in ('url', 'url_transparent'): | |
1183 | ie_result['url'] = sanitize_url(ie_result['url']) | |
1184 | extract_flat = self.params.get('extract_flat', False) | |
1185 | if ((extract_flat == 'in_playlist' and 'playlist' in extra_info) | |
1186 | or extract_flat is True): | |
1187 | self.__forced_printings(ie_result, self.prepare_filename(ie_result), incomplete=True) | |
1188 | return ie_result | |
1189 | ||
1190 | if result_type == 'video': | |
1191 | self.add_extra_info(ie_result, extra_info) | |
1192 | ie_result = self.process_video_result(ie_result, download=download) | |
1193 | additional_urls = (ie_result or {}).get('additional_urls') | |
1194 | if additional_urls: | |
1195 | # TODO: Improve MetadataFromFieldPP to allow setting a list | |
1196 | if isinstance(additional_urls, compat_str): | |
1197 | additional_urls = [additional_urls] | |
1198 | self.to_screen( | |
1199 | '[info] %s: %d additional URL(s) requested' % (ie_result['id'], len(additional_urls))) | |
1200 | self.write_debug('Additional URLs: "%s"' % '", "'.join(additional_urls)) | |
1201 | ie_result['additional_entries'] = [ | |
1202 | self.extract_info( | |
1203 | url, download, extra_info, | |
1204 | force_generic_extractor=self.params.get('force_generic_extractor')) | |
1205 | for url in additional_urls | |
1206 | ] | |
1207 | return ie_result | |
1208 | elif result_type == 'url': | |
1209 | # We have to add extra_info to the results because it may be | |
1210 | # contained in a playlist | |
1211 | return self.extract_info( | |
1212 | ie_result['url'], download, | |
1213 | ie_key=ie_result.get('ie_key'), | |
1214 | extra_info=extra_info) | |
1215 | elif result_type == 'url_transparent': | |
1216 | # Use the information from the embedding page | |
1217 | info = self.extract_info( | |
1218 | ie_result['url'], ie_key=ie_result.get('ie_key'), | |
1219 | extra_info=extra_info, download=False, process=False) | |
1220 | ||
1221 | # extract_info may return None when ignoreerrors is enabled and | |
1222 | # extraction failed with an error, don't crash and return early | |
1223 | # in this case | |
1224 | if not info: | |
1225 | return info | |
1226 | ||
1227 | force_properties = dict( | |
1228 | (k, v) for k, v in ie_result.items() if v is not None) | |
1229 | for f in ('_type', 'url', 'id', 'extractor', 'extractor_key', 'ie_key'): | |
1230 | if f in force_properties: | |
1231 | del force_properties[f] | |
1232 | new_result = info.copy() | |
1233 | new_result.update(force_properties) | |
1234 | ||
1235 | # Extracted info may not be a video result (i.e. | |
1236 | # info.get('_type', 'video') != video) but rather an url or | |
1237 | # url_transparent. In such cases outer metadata (from ie_result) | |
1238 | # should be propagated to inner one (info). For this to happen | |
1239 | # _type of info should be overridden with url_transparent. This | |
1240 | # fixes issue from https://github.com/ytdl-org/youtube-dl/pull/11163. | |
1241 | if new_result.get('_type') == 'url': | |
1242 | new_result['_type'] = 'url_transparent' | |
1243 | ||
1244 | return self.process_ie_result( | |
1245 | new_result, download=download, extra_info=extra_info) | |
1246 | elif result_type in ('playlist', 'multi_video'): | |
1247 | # Protect from infinite recursion due to recursively nested playlists | |
1248 | # (see https://github.com/ytdl-org/youtube-dl/issues/27833) | |
1249 | webpage_url = ie_result['webpage_url'] | |
1250 | if webpage_url in self._playlist_urls: | |
1251 | self.to_screen( | |
1252 | '[download] Skipping already downloaded playlist: %s' | |
1253 | % ie_result.get('title') or ie_result.get('id')) | |
1254 | return | |
1255 | ||
1256 | self._playlist_level += 1 | |
1257 | self._playlist_urls.add(webpage_url) | |
1258 | self._sanitize_thumbnails(ie_result) | |
1259 | try: | |
1260 | return self.__process_playlist(ie_result, download) | |
1261 | finally: | |
1262 | self._playlist_level -= 1 | |
1263 | if not self._playlist_level: | |
1264 | self._playlist_urls.clear() | |
1265 | elif result_type == 'compat_list': | |
1266 | self.report_warning( | |
1267 | 'Extractor %s returned a compat_list result. ' | |
1268 | 'It needs to be updated.' % ie_result.get('extractor')) | |
1269 | ||
1270 | def _fixup(r): | |
1271 | self.add_extra_info( | |
1272 | r, | |
1273 | { | |
1274 | 'extractor': ie_result['extractor'], | |
1275 | 'webpage_url': ie_result['webpage_url'], | |
1276 | 'webpage_url_basename': url_basename(ie_result['webpage_url']), | |
1277 | 'extractor_key': ie_result['extractor_key'], | |
1278 | } | |
1279 | ) | |
1280 | return r | |
1281 | ie_result['entries'] = [ | |
1282 | self.process_ie_result(_fixup(r), download, extra_info) | |
1283 | for r in ie_result['entries'] | |
1284 | ] | |
1285 | return ie_result | |
1286 | else: | |
1287 | raise Exception('Invalid result type: %s' % result_type) | |
1288 | ||
1289 | def _ensure_dir_exists(self, path): | |
1290 | return make_dir(path, self.report_error) | |
1291 | ||
1292 | def __process_playlist(self, ie_result, download): | |
1293 | # We process each entry in the playlist | |
1294 | playlist = ie_result.get('title') or ie_result.get('id') | |
1295 | self.to_screen('[download] Downloading playlist: %s' % playlist) | |
1296 | ||
1297 | if 'entries' not in ie_result: | |
1298 | raise EntryNotInPlaylist() | |
1299 | incomplete_entries = bool(ie_result.get('requested_entries')) | |
1300 | if incomplete_entries: | |
1301 | def fill_missing_entries(entries, indexes): | |
1302 | ret = [None] * max(*indexes) | |
1303 | for i, entry in zip(indexes, entries): | |
1304 | ret[i - 1] = entry | |
1305 | return ret | |
1306 | ie_result['entries'] = fill_missing_entries(ie_result['entries'], ie_result['requested_entries']) | |
1307 | ||
1308 | playlist_results = [] | |
1309 | ||
1310 | playliststart = self.params.get('playliststart', 1) | |
1311 | playlistend = self.params.get('playlistend') | |
1312 | # For backwards compatibility, interpret -1 as whole list | |
1313 | if playlistend == -1: | |
1314 | playlistend = None | |
1315 | ||
1316 | playlistitems_str = self.params.get('playlist_items') | |
1317 | playlistitems = None | |
1318 | if playlistitems_str is not None: | |
1319 | def iter_playlistitems(format): | |
1320 | for string_segment in format.split(','): | |
1321 | if '-' in string_segment: | |
1322 | start, end = string_segment.split('-') | |
1323 | for item in range(int(start), int(end) + 1): | |
1324 | yield int(item) | |
1325 | else: | |
1326 | yield int(string_segment) | |
1327 | playlistitems = orderedSet(iter_playlistitems(playlistitems_str)) | |
1328 | ||
1329 | ie_entries = ie_result['entries'] | |
1330 | msg = ( | |
1331 | 'Downloading %d videos' if not isinstance(ie_entries, list) | |
1332 | else 'Collected %d videos; downloading %%d of them' % len(ie_entries)) | |
1333 | if not isinstance(ie_entries, (list, PagedList)): | |
1334 | ie_entries = LazyList(ie_entries) | |
1335 | ||
1336 | entries = [] | |
1337 | for i in playlistitems or itertools.count(playliststart): | |
1338 | if playlistitems is None and playlistend is not None and playlistend < i: | |
1339 | break | |
1340 | entry = None | |
1341 | try: | |
1342 | entry = ie_entries[i - 1] | |
1343 | if entry is None: | |
1344 | raise EntryNotInPlaylist() | |
1345 | except (IndexError, EntryNotInPlaylist): | |
1346 | if incomplete_entries: | |
1347 | raise EntryNotInPlaylist() | |
1348 | elif not playlistitems: | |
1349 | break | |
1350 | entries.append(entry) | |
1351 | try: | |
1352 | if entry is not None: | |
1353 | self._match_entry(entry, incomplete=True, silent=True) | |
1354 | except (ExistingVideoReached, RejectedVideoReached): | |
1355 | break | |
1356 | ie_result['entries'] = entries | |
1357 | ||
1358 | # Save playlist_index before re-ordering | |
1359 | entries = [ | |
1360 | ((playlistitems[i - 1] if playlistitems else i), entry) | |
1361 | for i, entry in enumerate(entries, 1) | |
1362 | if entry is not None] | |
1363 | n_entries = len(entries) | |
1364 | ||
1365 | if not playlistitems and (playliststart or playlistend): | |
1366 | playlistitems = list(range(playliststart, playliststart + n_entries)) | |
1367 | ie_result['requested_entries'] = playlistitems | |
1368 | ||
1369 | if self.params.get('allow_playlist_files', True): | |
1370 | ie_copy = { | |
1371 | 'playlist': playlist, | |
1372 | 'playlist_id': ie_result.get('id'), | |
1373 | 'playlist_title': ie_result.get('title'), | |
1374 | 'playlist_uploader': ie_result.get('uploader'), | |
1375 | 'playlist_uploader_id': ie_result.get('uploader_id'), | |
1376 | 'playlist_index': 0, | |
1377 | } | |
1378 | ie_copy.update(dict(ie_result)) | |
1379 | ||
1380 | if self.params.get('writeinfojson', False): | |
1381 | infofn = self.prepare_filename(ie_copy, 'pl_infojson') | |
1382 | if not self._ensure_dir_exists(encodeFilename(infofn)): | |
1383 | return | |
1384 | if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(infofn)): | |
1385 | self.to_screen('[info] Playlist metadata is already present') | |
1386 | else: | |
1387 | self.to_screen('[info] Writing playlist metadata as JSON to: ' + infofn) | |
1388 | try: | |
1389 | write_json_file(self.filter_requested_info(ie_result, self.params.get('clean_infojson', True)), infofn) | |
1390 | except (OSError, IOError): | |
1391 | self.report_error('Cannot write playlist metadata to JSON file ' + infofn) | |
1392 | ||
1393 | # TODO: This should be passed to ThumbnailsConvertor if necessary | |
1394 | self._write_thumbnails(ie_copy, self.prepare_filename(ie_copy, 'pl_thumbnail')) | |
1395 | ||
1396 | if self.params.get('writedescription', False): | |
1397 | descfn = self.prepare_filename(ie_copy, 'pl_description') | |
1398 | if not self._ensure_dir_exists(encodeFilename(descfn)): | |
1399 | return | |
1400 | if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(descfn)): | |
1401 | self.to_screen('[info] Playlist description is already present') | |
1402 | elif ie_result.get('description') is None: | |
1403 | self.report_warning('There\'s no playlist description to write.') | |
1404 | else: | |
1405 | try: | |
1406 | self.to_screen('[info] Writing playlist description to: ' + descfn) | |
1407 | with io.open(encodeFilename(descfn), 'w', encoding='utf-8') as descfile: | |
1408 | descfile.write(ie_result['description']) | |
1409 | except (OSError, IOError): | |
1410 | self.report_error('Cannot write playlist description file ' + descfn) | |
1411 | return | |
1412 | ||
1413 | if self.params.get('playlistreverse', False): | |
1414 | entries = entries[::-1] | |
1415 | if self.params.get('playlistrandom', False): | |
1416 | random.shuffle(entries) | |
1417 | ||
1418 | x_forwarded_for = ie_result.get('__x_forwarded_for_ip') | |
1419 | ||
1420 | self.to_screen('[%s] playlist %s: %s' % (ie_result['extractor'], playlist, msg % n_entries)) | |
1421 | failures = 0 | |
1422 | max_failures = self.params.get('skip_playlist_after_errors') or float('inf') | |
1423 | for i, entry_tuple in enumerate(entries, 1): | |
1424 | playlist_index, entry = entry_tuple | |
1425 | if 'playlist_index' in self.params.get('compat_options', []): | |
1426 | playlist_index = playlistitems[i - 1] if playlistitems else i | |
1427 | self.to_screen('[download] Downloading video %s of %s' % (i, n_entries)) | |
1428 | # This __x_forwarded_for_ip thing is a bit ugly but requires | |
1429 | # minimal changes | |
1430 | if x_forwarded_for: | |
1431 | entry['__x_forwarded_for_ip'] = x_forwarded_for | |
1432 | extra = { | |
1433 | 'n_entries': n_entries, | |
1434 | '_last_playlist_index': max(playlistitems) if playlistitems else (playlistend or n_entries), | |
1435 | 'playlist_index': playlist_index, | |
1436 | 'playlist_autonumber': i, | |
1437 | 'playlist': playlist, | |
1438 | 'playlist_id': ie_result.get('id'), | |
1439 | 'playlist_title': ie_result.get('title'), | |
1440 | 'playlist_uploader': ie_result.get('uploader'), | |
1441 | 'playlist_uploader_id': ie_result.get('uploader_id'), | |
1442 | 'extractor': ie_result['extractor'], | |
1443 | 'webpage_url': ie_result['webpage_url'], | |
1444 | 'webpage_url_basename': url_basename(ie_result['webpage_url']), | |
1445 | 'extractor_key': ie_result['extractor_key'], | |
1446 | } | |
1447 | ||
1448 | if self._match_entry(entry, incomplete=True) is not None: | |
1449 | continue | |
1450 | ||
1451 | entry_result = self.__process_iterable_entry(entry, download, extra) | |
1452 | if not entry_result: | |
1453 | failures += 1 | |
1454 | if failures >= max_failures: | |
1455 | self.report_error( | |
1456 | 'Skipping the remaining entries in playlist "%s" since %d items failed extraction' % (playlist, failures)) | |
1457 | break | |
1458 | # TODO: skip failed (empty) entries? | |
1459 | playlist_results.append(entry_result) | |
1460 | ie_result['entries'] = playlist_results | |
1461 | self.to_screen('[download] Finished downloading playlist: %s' % playlist) | |
1462 | return ie_result | |
1463 | ||
1464 | @__handle_extraction_exceptions | |
1465 | def __process_iterable_entry(self, entry, download, extra_info): | |
1466 | return self.process_ie_result( | |
1467 | entry, download=download, extra_info=extra_info) | |
1468 | ||
1469 | def _build_format_filter(self, filter_spec): | |
1470 | " Returns a function to filter the formats according to the filter_spec " | |
1471 | ||
1472 | OPERATORS = { | |
1473 | '<': operator.lt, | |
1474 | '<=': operator.le, | |
1475 | '>': operator.gt, | |
1476 | '>=': operator.ge, | |
1477 | '=': operator.eq, | |
1478 | '!=': operator.ne, | |
1479 | } | |
1480 | operator_rex = re.compile(r'''(?x)\s* | |
1481 | (?P<key>width|height|tbr|abr|vbr|asr|filesize|filesize_approx|fps) | |
1482 | \s*(?P<op>%s)(?P<none_inclusive>\s*\?)?\s* | |
1483 | (?P<value>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?) | |
1484 | $ | |
1485 | ''' % '|'.join(map(re.escape, OPERATORS.keys()))) | |
1486 | m = operator_rex.search(filter_spec) | |
1487 | if m: | |
1488 | try: | |
1489 | comparison_value = int(m.group('value')) | |
1490 | except ValueError: | |
1491 | comparison_value = parse_filesize(m.group('value')) | |
1492 | if comparison_value is None: | |
1493 | comparison_value = parse_filesize(m.group('value') + 'B') | |
1494 | if comparison_value is None: | |
1495 | raise ValueError( | |
1496 | 'Invalid value %r in format specification %r' % ( | |
1497 | m.group('value'), filter_spec)) | |
1498 | op = OPERATORS[m.group('op')] | |
1499 | ||
1500 | if not m: | |
1501 | STR_OPERATORS = { | |
1502 | '=': operator.eq, | |
1503 | '^=': lambda attr, value: attr.startswith(value), | |
1504 | '$=': lambda attr, value: attr.endswith(value), | |
1505 | '*=': lambda attr, value: value in attr, | |
1506 | } | |
1507 | str_operator_rex = re.compile(r'''(?x) | |
1508 | \s*(?P<key>[a-zA-Z0-9._-]+) | |
1509 | \s*(?P<negation>!\s*)?(?P<op>%s)(?P<none_inclusive>\s*\?)? | |
1510 | \s*(?P<value>[a-zA-Z0-9._-]+) | |
1511 | \s*$ | |
1512 | ''' % '|'.join(map(re.escape, STR_OPERATORS.keys()))) | |
1513 | m = str_operator_rex.search(filter_spec) | |
1514 | if m: | |
1515 | comparison_value = m.group('value') | |
1516 | str_op = STR_OPERATORS[m.group('op')] | |
1517 | if m.group('negation'): | |
1518 | op = lambda attr, value: not str_op(attr, value) | |
1519 | else: | |
1520 | op = str_op | |
1521 | ||
1522 | if not m: | |
1523 | raise ValueError('Invalid filter specification %r' % filter_spec) | |
1524 | ||
1525 | def _filter(f): | |
1526 | actual_value = f.get(m.group('key')) | |
1527 | if actual_value is None: | |
1528 | return m.group('none_inclusive') | |
1529 | return op(actual_value, comparison_value) | |
1530 | return _filter | |
1531 | ||
1532 | def _default_format_spec(self, info_dict, download=True): | |
1533 | ||
1534 | def can_merge(): | |
1535 | merger = FFmpegMergerPP(self) | |
1536 | return merger.available and merger.can_merge() | |
1537 | ||
1538 | prefer_best = ( | |
1539 | not self.params.get('simulate', False) | |
1540 | and download | |
1541 | and ( | |
1542 | not can_merge() | |
1543 | or info_dict.get('is_live', False) | |
1544 | or self.outtmpl_dict['default'] == '-')) | |
1545 | compat = ( | |
1546 | prefer_best | |
1547 | or self.params.get('allow_multiple_audio_streams', False) | |
1548 | or 'format-spec' in self.params.get('compat_opts', [])) | |
1549 | ||
1550 | return ( | |
1551 | 'best/bestvideo+bestaudio' if prefer_best | |
1552 | else 'bestvideo*+bestaudio/best' if not compat | |
1553 | else 'bestvideo+bestaudio/best') | |
1554 | ||
1555 | def build_format_selector(self, format_spec): | |
1556 | def syntax_error(note, start): | |
1557 | message = ( | |
1558 | 'Invalid format specification: ' | |
1559 | '{0}\n\t{1}\n\t{2}^'.format(note, format_spec, ' ' * start[1])) | |
1560 | return SyntaxError(message) | |
1561 | ||
1562 | PICKFIRST = 'PICKFIRST' | |
1563 | MERGE = 'MERGE' | |
1564 | SINGLE = 'SINGLE' | |
1565 | GROUP = 'GROUP' | |
1566 | FormatSelector = collections.namedtuple('FormatSelector', ['type', 'selector', 'filters']) | |
1567 | ||
1568 | allow_multiple_streams = {'audio': self.params.get('allow_multiple_audio_streams', False), | |
1569 | 'video': self.params.get('allow_multiple_video_streams', False)} | |
1570 | ||
1571 | check_formats = self.params.get('check_formats') | |
1572 | ||
1573 | def _parse_filter(tokens): | |
1574 | filter_parts = [] | |
1575 | for type, string, start, _, _ in tokens: | |
1576 | if type == tokenize.OP and string == ']': | |
1577 | return ''.join(filter_parts) | |
1578 | else: | |
1579 | filter_parts.append(string) | |
1580 | ||
1581 | def _remove_unused_ops(tokens): | |
1582 | # Remove operators that we don't use and join them with the surrounding strings | |
1583 | # for example: 'mp4' '-' 'baseline' '-' '16x9' is converted to 'mp4-baseline-16x9' | |
1584 | ALLOWED_OPS = ('/', '+', ',', '(', ')') | |
1585 | last_string, last_start, last_end, last_line = None, None, None, None | |
1586 | for type, string, start, end, line in tokens: | |
1587 | if type == tokenize.OP and string == '[': | |
1588 | if last_string: | |
1589 | yield tokenize.NAME, last_string, last_start, last_end, last_line | |
1590 | last_string = None | |
1591 | yield type, string, start, end, line | |
1592 | # everything inside brackets will be handled by _parse_filter | |
1593 | for type, string, start, end, line in tokens: | |
1594 | yield type, string, start, end, line | |
1595 | if type == tokenize.OP and string == ']': | |
1596 | break | |
1597 | elif type == tokenize.OP and string in ALLOWED_OPS: | |
1598 | if last_string: | |
1599 | yield tokenize.NAME, last_string, last_start, last_end, last_line | |
1600 | last_string = None | |
1601 | yield type, string, start, end, line | |
1602 | elif type in [tokenize.NAME, tokenize.NUMBER, tokenize.OP]: | |
1603 | if not last_string: | |
1604 | last_string = string | |
1605 | last_start = start | |
1606 | last_end = end | |
1607 | else: | |
1608 | last_string += string | |
1609 | if last_string: | |
1610 | yield tokenize.NAME, last_string, last_start, last_end, last_line | |
1611 | ||
1612 | def _parse_format_selection(tokens, inside_merge=False, inside_choice=False, inside_group=False): | |
1613 | selectors = [] | |
1614 | current_selector = None | |
1615 | for type, string, start, _, _ in tokens: | |
1616 | # ENCODING is only defined in python 3.x | |
1617 | if type == getattr(tokenize, 'ENCODING', None): | |
1618 | continue | |
1619 | elif type in [tokenize.NAME, tokenize.NUMBER]: | |
1620 | current_selector = FormatSelector(SINGLE, string, []) | |
1621 | elif type == tokenize.OP: | |
1622 | if string == ')': | |
1623 | if not inside_group: | |
1624 | # ')' will be handled by the parentheses group | |
1625 | tokens.restore_last_token() | |
1626 | break | |
1627 | elif inside_merge and string in ['/', ',']: | |
1628 | tokens.restore_last_token() | |
1629 | break | |
1630 | elif inside_choice and string == ',': | |
1631 | tokens.restore_last_token() | |
1632 | break | |
1633 | elif string == ',': | |
1634 | if not current_selector: | |
1635 | raise syntax_error('"," must follow a format selector', start) | |
1636 | selectors.append(current_selector) | |
1637 | current_selector = None | |
1638 | elif string == '/': | |
1639 | if not current_selector: | |
1640 | raise syntax_error('"/" must follow a format selector', start) | |
1641 | first_choice = current_selector | |
1642 | second_choice = _parse_format_selection(tokens, inside_choice=True) | |
1643 | current_selector = FormatSelector(PICKFIRST, (first_choice, second_choice), []) | |
1644 | elif string == '[': | |
1645 | if not current_selector: | |
1646 | current_selector = FormatSelector(SINGLE, 'best', []) | |
1647 | format_filter = _parse_filter(tokens) | |
1648 | current_selector.filters.append(format_filter) | |
1649 | elif string == '(': | |
1650 | if current_selector: | |
1651 | raise syntax_error('Unexpected "("', start) | |
1652 | group = _parse_format_selection(tokens, inside_group=True) | |
1653 | current_selector = FormatSelector(GROUP, group, []) | |
1654 | elif string == '+': | |
1655 | if not current_selector: | |
1656 | raise syntax_error('Unexpected "+"', start) | |
1657 | selector_1 = current_selector | |
1658 | selector_2 = _parse_format_selection(tokens, inside_merge=True) | |
1659 | if not selector_2: | |
1660 | raise syntax_error('Expected a selector', start) | |
1661 | current_selector = FormatSelector(MERGE, (selector_1, selector_2), []) | |
1662 | else: | |
1663 | raise syntax_error('Operator not recognized: "{0}"'.format(string), start) | |
1664 | elif type == tokenize.ENDMARKER: | |
1665 | break | |
1666 | if current_selector: | |
1667 | selectors.append(current_selector) | |
1668 | return selectors | |
1669 | ||
1670 | def _merge(formats_pair): | |
1671 | format_1, format_2 = formats_pair | |
1672 | ||
1673 | formats_info = [] | |
1674 | formats_info.extend(format_1.get('requested_formats', (format_1,))) | |
1675 | formats_info.extend(format_2.get('requested_formats', (format_2,))) | |
1676 | ||
1677 | if not allow_multiple_streams['video'] or not allow_multiple_streams['audio']: | |
1678 | get_no_more = {"video": False, "audio": False} | |
1679 | for (i, fmt_info) in enumerate(formats_info): | |
1680 | for aud_vid in ["audio", "video"]: | |
1681 | if not allow_multiple_streams[aud_vid] and fmt_info.get(aud_vid[0] + 'codec') != 'none': | |
1682 | if get_no_more[aud_vid]: | |
1683 | formats_info.pop(i) | |
1684 | get_no_more[aud_vid] = True | |
1685 | ||
1686 | if len(formats_info) == 1: | |
1687 | return formats_info[0] | |
1688 | ||
1689 | video_fmts = [fmt_info for fmt_info in formats_info if fmt_info.get('vcodec') != 'none'] | |
1690 | audio_fmts = [fmt_info for fmt_info in formats_info if fmt_info.get('acodec') != 'none'] | |
1691 | ||
1692 | the_only_video = video_fmts[0] if len(video_fmts) == 1 else None | |
1693 | the_only_audio = audio_fmts[0] if len(audio_fmts) == 1 else None | |
1694 | ||
1695 | output_ext = self.params.get('merge_output_format') | |
1696 | if not output_ext: | |
1697 | if the_only_video: | |
1698 | output_ext = the_only_video['ext'] | |
1699 | elif the_only_audio and not video_fmts: | |
1700 | output_ext = the_only_audio['ext'] | |
1701 | else: | |
1702 | output_ext = 'mkv' | |
1703 | ||
1704 | new_dict = { | |
1705 | 'requested_formats': formats_info, | |
1706 | 'format': '+'.join(fmt_info.get('format') for fmt_info in formats_info), | |
1707 | 'format_id': '+'.join(fmt_info.get('format_id') for fmt_info in formats_info), | |
1708 | 'ext': output_ext, | |
1709 | } | |
1710 | ||
1711 | if the_only_video: | |
1712 | new_dict.update({ | |
1713 | 'width': the_only_video.get('width'), | |
1714 | 'height': the_only_video.get('height'), | |
1715 | 'resolution': the_only_video.get('resolution') or self.format_resolution(the_only_video), | |
1716 | 'fps': the_only_video.get('fps'), | |
1717 | 'vcodec': the_only_video.get('vcodec'), | |
1718 | 'vbr': the_only_video.get('vbr'), | |
1719 | 'stretched_ratio': the_only_video.get('stretched_ratio'), | |
1720 | }) | |
1721 | ||
1722 | if the_only_audio: | |
1723 | new_dict.update({ | |
1724 | 'acodec': the_only_audio.get('acodec'), | |
1725 | 'abr': the_only_audio.get('abr'), | |
1726 | }) | |
1727 | ||
1728 | return new_dict | |
1729 | ||
1730 | def _check_formats(formats): | |
1731 | for f in formats: | |
1732 | self.to_screen('[info] Testing format %s' % f['format_id']) | |
1733 | paths = self.params.get('paths', {}) | |
1734 | temp_file = os.path.join( | |
1735 | expand_path(paths.get('home', '').strip()), | |
1736 | expand_path(paths.get('temp', '').strip()), | |
1737 | 'ytdl.%s.f%s.check-format' % (random_uuidv4(), f['format_id'])) | |
1738 | try: | |
1739 | dl, _ = self.dl(temp_file, f, test=True) | |
1740 | except (ExtractorError, IOError, OSError, ValueError) + network_exceptions: | |
1741 | dl = False | |
1742 | finally: | |
1743 | if os.path.exists(temp_file): | |
1744 | os.remove(temp_file) | |
1745 | if dl: | |
1746 | yield f | |
1747 | else: | |
1748 | self.to_screen('[info] Unable to download format %s. Skipping...' % f['format_id']) | |
1749 | ||
1750 | def _build_selector_function(selector): | |
1751 | if isinstance(selector, list): # , | |
1752 | fs = [_build_selector_function(s) for s in selector] | |
1753 | ||
1754 | def selector_function(ctx): | |
1755 | for f in fs: | |
1756 | for format in f(ctx): | |
1757 | yield format | |
1758 | return selector_function | |
1759 | ||
1760 | elif selector.type == GROUP: # () | |
1761 | selector_function = _build_selector_function(selector.selector) | |
1762 | ||
1763 | elif selector.type == PICKFIRST: # / | |
1764 | fs = [_build_selector_function(s) for s in selector.selector] | |
1765 | ||
1766 | def selector_function(ctx): | |
1767 | for f in fs: | |
1768 | picked_formats = list(f(ctx)) | |
1769 | if picked_formats: | |
1770 | return picked_formats | |
1771 | return [] | |
1772 | ||
1773 | elif selector.type == SINGLE: # atom | |
1774 | format_spec = selector.selector or 'best' | |
1775 | ||
1776 | # TODO: Add allvideo, allaudio etc by generalizing the code with best/worst selector | |
1777 | if format_spec == 'all': | |
1778 | def selector_function(ctx): | |
1779 | formats = list(ctx['formats']) | |
1780 | if check_formats: | |
1781 | formats = _check_formats(formats) | |
1782 | for f in formats: | |
1783 | yield f | |
1784 | elif format_spec == 'mergeall': | |
1785 | def selector_function(ctx): | |
1786 | formats = list(_check_formats(ctx['formats'])) | |
1787 | if not formats: | |
1788 | return | |
1789 | merged_format = formats[-1] | |
1790 | for f in formats[-2::-1]: | |
1791 | merged_format = _merge((merged_format, f)) | |
1792 | yield merged_format | |
1793 | ||
1794 | else: | |
1795 | format_fallback, format_reverse, format_idx = False, True, 1 | |
1796 | mobj = re.match( | |
1797 | r'(?P<bw>best|worst|b|w)(?P<type>video|audio|v|a)?(?P<mod>\*)?(?:\.(?P<n>[1-9]\d*))?$', | |
1798 | format_spec) | |
1799 | if mobj is not None: | |
1800 | format_idx = int_or_none(mobj.group('n'), default=1) | |
1801 | format_reverse = mobj.group('bw')[0] == 'b' | |
1802 | format_type = (mobj.group('type') or [None])[0] | |
1803 | not_format_type = {'v': 'a', 'a': 'v'}.get(format_type) | |
1804 | format_modified = mobj.group('mod') is not None | |
1805 | ||
1806 | format_fallback = not format_type and not format_modified # for b, w | |
1807 | filter_f = ( | |
1808 | (lambda f: f.get('%scodec' % format_type) != 'none') | |
1809 | if format_type and format_modified # bv*, ba*, wv*, wa* | |
1810 | else (lambda f: f.get('%scodec' % not_format_type) == 'none') | |
1811 | if format_type # bv, ba, wv, wa | |
1812 | else (lambda f: f.get('vcodec') != 'none' and f.get('acodec') != 'none') | |
1813 | if not format_modified # b, w | |
1814 | else None) # b*, w* | |
1815 | else: | |
1816 | filter_f = ((lambda f: f.get('ext') == format_spec) | |
1817 | if format_spec in ['mp4', 'flv', 'webm', '3gp', 'm4a', 'mp3', 'ogg', 'aac', 'wav'] # extension | |
1818 | else (lambda f: f.get('format_id') == format_spec)) # id | |
1819 | ||
1820 | def selector_function(ctx): | |
1821 | formats = list(ctx['formats']) | |
1822 | if not formats: | |
1823 | return | |
1824 | matches = list(filter(filter_f, formats)) if filter_f is not None else formats | |
1825 | if format_fallback and ctx['incomplete_formats'] and not matches: | |
1826 | # for extractors with incomplete formats (audio only (soundcloud) | |
1827 | # or video only (imgur)) best/worst will fallback to | |
1828 | # best/worst {video,audio}-only format | |
1829 | matches = formats | |
1830 | if format_reverse: | |
1831 | matches = matches[::-1] | |
1832 | if check_formats: | |
1833 | matches = list(itertools.islice(_check_formats(matches), format_idx)) | |
1834 | n = len(matches) | |
1835 | if -n <= format_idx - 1 < n: | |
1836 | yield matches[format_idx - 1] | |
1837 | ||
1838 | elif selector.type == MERGE: # + | |
1839 | selector_1, selector_2 = map(_build_selector_function, selector.selector) | |
1840 | ||
1841 | def selector_function(ctx): | |
1842 | for pair in itertools.product( | |
1843 | selector_1(copy.deepcopy(ctx)), selector_2(copy.deepcopy(ctx))): | |
1844 | yield _merge(pair) | |
1845 | ||
1846 | filters = [self._build_format_filter(f) for f in selector.filters] | |
1847 | ||
1848 | def final_selector(ctx): | |
1849 | ctx_copy = copy.deepcopy(ctx) | |
1850 | for _filter in filters: | |
1851 | ctx_copy['formats'] = list(filter(_filter, ctx_copy['formats'])) | |
1852 | return selector_function(ctx_copy) | |
1853 | return final_selector | |
1854 | ||
1855 | stream = io.BytesIO(format_spec.encode('utf-8')) | |
1856 | try: | |
1857 | tokens = list(_remove_unused_ops(compat_tokenize_tokenize(stream.readline))) | |
1858 | except tokenize.TokenError: | |
1859 | raise syntax_error('Missing closing/opening brackets or parenthesis', (0, len(format_spec))) | |
1860 | ||
1861 | class TokenIterator(object): | |
1862 | def __init__(self, tokens): | |
1863 | self.tokens = tokens | |
1864 | self.counter = 0 | |
1865 | ||
1866 | def __iter__(self): | |
1867 | return self | |
1868 | ||
1869 | def __next__(self): | |
1870 | if self.counter >= len(self.tokens): | |
1871 | raise StopIteration() | |
1872 | value = self.tokens[self.counter] | |
1873 | self.counter += 1 | |
1874 | return value | |
1875 | ||
1876 | next = __next__ | |
1877 | ||
1878 | def restore_last_token(self): | |
1879 | self.counter -= 1 | |
1880 | ||
1881 | parsed_selector = _parse_format_selection(iter(TokenIterator(tokens))) | |
1882 | return _build_selector_function(parsed_selector) | |
1883 | ||
1884 | def _calc_headers(self, info_dict): | |
1885 | res = std_headers.copy() | |
1886 | ||
1887 | add_headers = info_dict.get('http_headers') | |
1888 | if add_headers: | |
1889 | res.update(add_headers) | |
1890 | ||
1891 | cookies = self._calc_cookies(info_dict) | |
1892 | if cookies: | |
1893 | res['Cookie'] = cookies | |
1894 | ||
1895 | if 'X-Forwarded-For' not in res: | |
1896 | x_forwarded_for_ip = info_dict.get('__x_forwarded_for_ip') | |
1897 | if x_forwarded_for_ip: | |
1898 | res['X-Forwarded-For'] = x_forwarded_for_ip | |
1899 | ||
1900 | return res | |
1901 | ||
1902 | def _calc_cookies(self, info_dict): | |
1903 | pr = sanitized_Request(info_dict['url']) | |
1904 | self.cookiejar.add_cookie_header(pr) | |
1905 | return pr.get_header('Cookie') | |
1906 | ||
1907 | @staticmethod | |
1908 | def _sanitize_thumbnails(info_dict): | |
1909 | thumbnails = info_dict.get('thumbnails') | |
1910 | if thumbnails is None: | |
1911 | thumbnail = info_dict.get('thumbnail') | |
1912 | if thumbnail: | |
1913 | info_dict['thumbnails'] = thumbnails = [{'url': thumbnail}] | |
1914 | if thumbnails: | |
1915 | thumbnails.sort(key=lambda t: ( | |
1916 | t.get('preference') if t.get('preference') is not None else -1, | |
1917 | t.get('width') if t.get('width') is not None else -1, | |
1918 | t.get('height') if t.get('height') is not None else -1, | |
1919 | t.get('id') if t.get('id') is not None else '', | |
1920 | t.get('url'))) | |
1921 | for i, t in enumerate(thumbnails): | |
1922 | t['url'] = sanitize_url(t['url']) | |
1923 | if t.get('width') and t.get('height'): | |
1924 | t['resolution'] = '%dx%d' % (t['width'], t['height']) | |
1925 | if t.get('id') is None: | |
1926 | t['id'] = '%d' % i | |
1927 | ||
1928 | def process_video_result(self, info_dict, download=True): | |
1929 | assert info_dict.get('_type', 'video') == 'video' | |
1930 | ||
1931 | if 'id' not in info_dict: | |
1932 | raise ExtractorError('Missing "id" field in extractor result') | |
1933 | if 'title' not in info_dict: | |
1934 | raise ExtractorError('Missing "title" field in extractor result') | |
1935 | ||
1936 | def report_force_conversion(field, field_not, conversion): | |
1937 | self.report_warning( | |
1938 | '"%s" field is not %s - forcing %s conversion, there is an error in extractor' | |
1939 | % (field, field_not, conversion)) | |
1940 | ||
1941 | def sanitize_string_field(info, string_field): | |
1942 | field = info.get(string_field) | |
1943 | if field is None or isinstance(field, compat_str): | |
1944 | return | |
1945 | report_force_conversion(string_field, 'a string', 'string') | |
1946 | info[string_field] = compat_str(field) | |
1947 | ||
1948 | def sanitize_numeric_fields(info): | |
1949 | for numeric_field in self._NUMERIC_FIELDS: | |
1950 | field = info.get(numeric_field) | |
1951 | if field is None or isinstance(field, compat_numeric_types): | |
1952 | continue | |
1953 | report_force_conversion(numeric_field, 'numeric', 'int') | |
1954 | info[numeric_field] = int_or_none(field) | |
1955 | ||
1956 | sanitize_string_field(info_dict, 'id') | |
1957 | sanitize_numeric_fields(info_dict) | |
1958 | ||
1959 | if 'playlist' not in info_dict: | |
1960 | # It isn't part of a playlist | |
1961 | info_dict['playlist'] = None | |
1962 | info_dict['playlist_index'] = None | |
1963 | ||
1964 | self._sanitize_thumbnails(info_dict) | |
1965 | ||
1966 | if self.params.get('list_thumbnails'): | |
1967 | self.list_thumbnails(info_dict) | |
1968 | return | |
1969 | ||
1970 | thumbnail = info_dict.get('thumbnail') | |
1971 | thumbnails = info_dict.get('thumbnails') | |
1972 | if thumbnail: | |
1973 | info_dict['thumbnail'] = sanitize_url(thumbnail) | |
1974 | elif thumbnails: | |
1975 | info_dict['thumbnail'] = thumbnails[-1]['url'] | |
1976 | ||
1977 | if 'display_id' not in info_dict and 'id' in info_dict: | |
1978 | info_dict['display_id'] = info_dict['id'] | |
1979 | ||
1980 | for ts_key, date_key in ( | |
1981 | ('timestamp', 'upload_date'), | |
1982 | ('release_timestamp', 'release_date'), | |
1983 | ): | |
1984 | if info_dict.get(date_key) is None and info_dict.get(ts_key) is not None: | |
1985 | # Working around out-of-range timestamp values (e.g. negative ones on Windows, | |
1986 | # see http://bugs.python.org/issue1646728) | |
1987 | try: | |
1988 | upload_date = datetime.datetime.utcfromtimestamp(info_dict[ts_key]) | |
1989 | info_dict[date_key] = upload_date.strftime('%Y%m%d') | |
1990 | except (ValueError, OverflowError, OSError): | |
1991 | pass | |
1992 | ||
1993 | # Auto generate title fields corresponding to the *_number fields when missing | |
1994 | # in order to always have clean titles. This is very common for TV series. | |
1995 | for field in ('chapter', 'season', 'episode'): | |
1996 | if info_dict.get('%s_number' % field) is not None and not info_dict.get(field): | |
1997 | info_dict[field] = '%s %d' % (field.capitalize(), info_dict['%s_number' % field]) | |
1998 | ||
1999 | for cc_kind in ('subtitles', 'automatic_captions'): | |
2000 | cc = info_dict.get(cc_kind) | |
2001 | if cc: | |
2002 | for _, subtitle in cc.items(): | |
2003 | for subtitle_format in subtitle: | |
2004 | if subtitle_format.get('url'): | |
2005 | subtitle_format['url'] = sanitize_url(subtitle_format['url']) | |
2006 | if subtitle_format.get('ext') is None: | |
2007 | subtitle_format['ext'] = determine_ext(subtitle_format['url']).lower() | |
2008 | ||
2009 | automatic_captions = info_dict.get('automatic_captions') | |
2010 | subtitles = info_dict.get('subtitles') | |
2011 | ||
2012 | if self.params.get('listsubtitles', False): | |
2013 | if 'automatic_captions' in info_dict: | |
2014 | self.list_subtitles( | |
2015 | info_dict['id'], automatic_captions, 'automatic captions') | |
2016 | self.list_subtitles(info_dict['id'], subtitles, 'subtitles') | |
2017 | return | |
2018 | ||
2019 | info_dict['requested_subtitles'] = self.process_subtitles( | |
2020 | info_dict['id'], subtitles, automatic_captions) | |
2021 | ||
2022 | # We now pick which formats have to be downloaded | |
2023 | if info_dict.get('formats') is None: | |
2024 | # There's only one format available | |
2025 | formats = [info_dict] | |
2026 | else: | |
2027 | formats = info_dict['formats'] | |
2028 | ||
2029 | if not formats: | |
2030 | if not self.params.get('ignore_no_formats_error'): | |
2031 | raise ExtractorError('No video formats found!') | |
2032 | else: | |
2033 | self.report_warning('No video formats found!') | |
2034 | ||
2035 | def is_wellformed(f): | |
2036 | url = f.get('url') | |
2037 | if not url: | |
2038 | self.report_warning( | |
2039 | '"url" field is missing or empty - skipping format, ' | |
2040 | 'there is an error in extractor') | |
2041 | return False | |
2042 | if isinstance(url, bytes): | |
2043 | sanitize_string_field(f, 'url') | |
2044 | return True | |
2045 | ||
2046 | # Filter out malformed formats for better extraction robustness | |
2047 | formats = list(filter(is_wellformed, formats)) | |
2048 | ||
2049 | formats_dict = {} | |
2050 | ||
2051 | # We check that all the formats have the format and format_id fields | |
2052 | for i, format in enumerate(formats): | |
2053 | sanitize_string_field(format, 'format_id') | |
2054 | sanitize_numeric_fields(format) | |
2055 | format['url'] = sanitize_url(format['url']) | |
2056 | if not format.get('format_id'): | |
2057 | format['format_id'] = compat_str(i) | |
2058 | else: | |
2059 | # Sanitize format_id from characters used in format selector expression | |
2060 | format['format_id'] = re.sub(r'[\s,/+\[\]()]', '_', format['format_id']) | |
2061 | format_id = format['format_id'] | |
2062 | if format_id not in formats_dict: | |
2063 | formats_dict[format_id] = [] | |
2064 | formats_dict[format_id].append(format) | |
2065 | ||
2066 | # Make sure all formats have unique format_id | |
2067 | for format_id, ambiguous_formats in formats_dict.items(): | |
2068 | if len(ambiguous_formats) > 1: | |
2069 | for i, format in enumerate(ambiguous_formats): | |
2070 | format['format_id'] = '%s-%d' % (format_id, i) | |
2071 | ||
2072 | for i, format in enumerate(formats): | |
2073 | if format.get('format') is None: | |
2074 | format['format'] = '{id} - {res}{note}'.format( | |
2075 | id=format['format_id'], | |
2076 | res=self.format_resolution(format), | |
2077 | note=' ({0})'.format(format['format_note']) if format.get('format_note') is not None else '', | |
2078 | ) | |
2079 | # Automatically determine file extension if missing | |
2080 | if format.get('ext') is None: | |
2081 | format['ext'] = determine_ext(format['url']).lower() | |
2082 | # Automatically determine protocol if missing (useful for format | |
2083 | # selection purposes) | |
2084 | if format.get('protocol') is None: | |
2085 | format['protocol'] = determine_protocol(format) | |
2086 | # Add HTTP headers, so that external programs can use them from the | |
2087 | # json output | |
2088 | full_format_info = info_dict.copy() | |
2089 | full_format_info.update(format) | |
2090 | format['http_headers'] = self._calc_headers(full_format_info) | |
2091 | # Remove private housekeeping stuff | |
2092 | if '__x_forwarded_for_ip' in info_dict: | |
2093 | del info_dict['__x_forwarded_for_ip'] | |
2094 | ||
2095 | # TODO Central sorting goes here | |
2096 | ||
2097 | if formats and formats[0] is not info_dict: | |
2098 | # only set the 'formats' fields if the original info_dict list them | |
2099 | # otherwise we end up with a circular reference, the first (and unique) | |
2100 | # element in the 'formats' field in info_dict is info_dict itself, | |
2101 | # which can't be exported to json | |
2102 | info_dict['formats'] = formats | |
2103 | ||
2104 | info_dict, _ = self.pre_process(info_dict) | |
2105 | ||
2106 | if self.params.get('listformats'): | |
2107 | if not info_dict.get('formats'): | |
2108 | raise ExtractorError('No video formats found', expected=True) | |
2109 | self.list_formats(info_dict) | |
2110 | return | |
2111 | ||
2112 | req_format = self.params.get('format') | |
2113 | if req_format is None: | |
2114 | req_format = self._default_format_spec(info_dict, download=download) | |
2115 | self.write_debug('Default format spec: %s' % req_format) | |
2116 | ||
2117 | format_selector = self.build_format_selector(req_format) | |
2118 | ||
2119 | # While in format selection we may need to have an access to the original | |
2120 | # format set in order to calculate some metrics or do some processing. | |
2121 | # For now we need to be able to guess whether original formats provided | |
2122 | # by extractor are incomplete or not (i.e. whether extractor provides only | |
2123 | # video-only or audio-only formats) for proper formats selection for | |
2124 | # extractors with such incomplete formats (see | |
2125 | # https://github.com/ytdl-org/youtube-dl/pull/5556). | |
2126 | # Since formats may be filtered during format selection and may not match | |
2127 | # the original formats the results may be incorrect. Thus original formats | |
2128 | # or pre-calculated metrics should be passed to format selection routines | |
2129 | # as well. | |
2130 | # We will pass a context object containing all necessary additional data | |
2131 | # instead of just formats. | |
2132 | # This fixes incorrect format selection issue (see | |
2133 | # https://github.com/ytdl-org/youtube-dl/issues/10083). | |
2134 | incomplete_formats = ( | |
2135 | # All formats are video-only or | |
2136 | all(f.get('vcodec') != 'none' and f.get('acodec') == 'none' for f in formats) | |
2137 | # all formats are audio-only | |
2138 | or all(f.get('vcodec') == 'none' and f.get('acodec') != 'none' for f in formats)) | |
2139 | ||
2140 | ctx = { | |
2141 | 'formats': formats, | |
2142 | 'incomplete_formats': incomplete_formats, | |
2143 | } | |
2144 | ||
2145 | formats_to_download = list(format_selector(ctx)) | |
2146 | if not formats_to_download: | |
2147 | if not self.params.get('ignore_no_formats_error'): | |
2148 | raise ExtractorError('Requested format is not available', expected=True) | |
2149 | else: | |
2150 | self.report_warning('Requested format is not available') | |
2151 | elif download: | |
2152 | self.to_screen( | |
2153 | '[info] %s: Downloading %d format(s): %s' % ( | |
2154 | info_dict['id'], len(formats_to_download), | |
2155 | ", ".join([f['format_id'] for f in formats_to_download]))) | |
2156 | for fmt in formats_to_download: | |
2157 | new_info = dict(info_dict) | |
2158 | # Save a reference to the original info_dict so that it can be modified in process_info if needed | |
2159 | new_info['__original_infodict'] = info_dict | |
2160 | new_info.update(fmt) | |
2161 | self.process_info(new_info) | |
2162 | # We update the info dict with the best quality format (backwards compatibility) | |
2163 | if formats_to_download: | |
2164 | info_dict.update(formats_to_download[-1]) | |
2165 | return info_dict | |
2166 | ||
2167 | def process_subtitles(self, video_id, normal_subtitles, automatic_captions): | |
2168 | """Select the requested subtitles and their format""" | |
2169 | available_subs = {} | |
2170 | if normal_subtitles and self.params.get('writesubtitles'): | |
2171 | available_subs.update(normal_subtitles) | |
2172 | if automatic_captions and self.params.get('writeautomaticsub'): | |
2173 | for lang, cap_info in automatic_captions.items(): | |
2174 | if lang not in available_subs: | |
2175 | available_subs[lang] = cap_info | |
2176 | ||
2177 | if (not self.params.get('writesubtitles') and not | |
2178 | self.params.get('writeautomaticsub') or not | |
2179 | available_subs): | |
2180 | return None | |
2181 | ||
2182 | all_sub_langs = available_subs.keys() | |
2183 | if self.params.get('allsubtitles', False): | |
2184 | requested_langs = all_sub_langs | |
2185 | elif self.params.get('subtitleslangs', False): | |
2186 | requested_langs = set() | |
2187 | for lang in self.params.get('subtitleslangs'): | |
2188 | if lang == 'all': | |
2189 | requested_langs.update(all_sub_langs) | |
2190 | continue | |
2191 | discard = lang[0] == '-' | |
2192 | if discard: | |
2193 | lang = lang[1:] | |
2194 | current_langs = filter(re.compile(lang + '$').match, all_sub_langs) | |
2195 | if discard: | |
2196 | for lang in current_langs: | |
2197 | requested_langs.discard(lang) | |
2198 | else: | |
2199 | requested_langs.update(current_langs) | |
2200 | elif 'en' in available_subs: | |
2201 | requested_langs = ['en'] | |
2202 | else: | |
2203 | requested_langs = [list(all_sub_langs)[0]] | |
2204 | self.write_debug('Downloading subtitles: %s' % ', '.join(requested_langs)) | |
2205 | ||
2206 | formats_query = self.params.get('subtitlesformat', 'best') | |
2207 | formats_preference = formats_query.split('/') if formats_query else [] | |
2208 | subs = {} | |
2209 | for lang in requested_langs: | |
2210 | formats = available_subs.get(lang) | |
2211 | if formats is None: | |
2212 | self.report_warning('%s subtitles not available for %s' % (lang, video_id)) | |
2213 | continue | |
2214 | for ext in formats_preference: | |
2215 | if ext == 'best': | |
2216 | f = formats[-1] | |
2217 | break | |
2218 | matches = list(filter(lambda f: f['ext'] == ext, formats)) | |
2219 | if matches: | |
2220 | f = matches[-1] | |
2221 | break | |
2222 | else: | |
2223 | f = formats[-1] | |
2224 | self.report_warning( | |
2225 | 'No subtitle format found matching "%s" for language %s, ' | |
2226 | 'using %s' % (formats_query, lang, f['ext'])) | |
2227 | subs[lang] = f | |
2228 | return subs | |
2229 | ||
2230 | def __forced_printings(self, info_dict, filename, incomplete): | |
2231 | def print_mandatory(field, actual_field=None): | |
2232 | if actual_field is None: | |
2233 | actual_field = field | |
2234 | if (self.params.get('force%s' % field, False) | |
2235 | and (not incomplete or info_dict.get(actual_field) is not None)): | |
2236 | self.to_stdout(info_dict[actual_field]) | |
2237 | ||
2238 | def print_optional(field): | |
2239 | if (self.params.get('force%s' % field, False) | |
2240 | and info_dict.get(field) is not None): | |
2241 | self.to_stdout(info_dict[field]) | |
2242 | ||
2243 | info_dict = info_dict.copy() | |
2244 | if filename is not None: | |
2245 | info_dict['filename'] = filename | |
2246 | if info_dict.get('requested_formats') is not None: | |
2247 | # For RTMP URLs, also include the playpath | |
2248 | info_dict['urls'] = '\n'.join(f['url'] + f.get('play_path', '') for f in info_dict['requested_formats']) | |
2249 | elif 'url' in info_dict: | |
2250 | info_dict['urls'] = info_dict['url'] + info_dict.get('play_path', '') | |
2251 | ||
2252 | for tmpl in self.params.get('forceprint', []): | |
2253 | if re.match(r'\w+$', tmpl): | |
2254 | tmpl = '%({})s'.format(tmpl) | |
2255 | tmpl, info_copy = self.prepare_outtmpl(tmpl, info_dict) | |
2256 | self.to_stdout(tmpl % info_copy) | |
2257 | ||
2258 | print_mandatory('title') | |
2259 | print_mandatory('id') | |
2260 | print_mandatory('url', 'urls') | |
2261 | print_optional('thumbnail') | |
2262 | print_optional('description') | |
2263 | print_optional('filename') | |
2264 | if self.params.get('forceduration', False) and info_dict.get('duration') is not None: | |
2265 | self.to_stdout(formatSeconds(info_dict['duration'])) | |
2266 | print_mandatory('format') | |
2267 | ||
2268 | if self.params.get('forcejson', False): | |
2269 | self.post_extract(info_dict) | |
2270 | self.to_stdout(json.dumps(info_dict, default=repr)) | |
2271 | ||
2272 | def dl(self, name, info, subtitle=False, test=False): | |
2273 | ||
2274 | if test: | |
2275 | verbose = self.params.get('verbose') | |
2276 | params = { | |
2277 | 'test': True, | |
2278 | 'quiet': not verbose, | |
2279 | 'verbose': verbose, | |
2280 | 'noprogress': not verbose, | |
2281 | 'nopart': True, | |
2282 | 'skip_unavailable_fragments': False, | |
2283 | 'keep_fragments': False, | |
2284 | 'overwrites': True, | |
2285 | '_no_ytdl_file': True, | |
2286 | } | |
2287 | else: | |
2288 | params = self.params | |
2289 | fd = get_suitable_downloader(info, params)(self, params) | |
2290 | if not test: | |
2291 | for ph in self._progress_hooks: | |
2292 | fd.add_progress_hook(ph) | |
2293 | urls = '", "'.join([f['url'] for f in info.get('requested_formats', [])] or [info['url']]) | |
2294 | self.write_debug('Invoking downloader on "%s"' % urls) | |
2295 | new_info = dict(info) | |
2296 | if new_info.get('http_headers') is None: | |
2297 | new_info['http_headers'] = self._calc_headers(new_info) | |
2298 | return fd.download(name, new_info, subtitle) | |
2299 | ||
2300 | def process_info(self, info_dict): | |
2301 | """Process a single resolved IE result.""" | |
2302 | ||
2303 | assert info_dict.get('_type', 'video') == 'video' | |
2304 | ||
2305 | info_dict.setdefault('__postprocessors', []) | |
2306 | ||
2307 | max_downloads = self.params.get('max_downloads') | |
2308 | if max_downloads is not None: | |
2309 | if self._num_downloads >= int(max_downloads): | |
2310 | raise MaxDownloadsReached() | |
2311 | ||
2312 | # TODO: backward compatibility, to be removed | |
2313 | info_dict['fulltitle'] = info_dict['title'] | |
2314 | ||
2315 | if 'format' not in info_dict: | |
2316 | info_dict['format'] = info_dict['ext'] | |
2317 | ||
2318 | if self._match_entry(info_dict) is not None: | |
2319 | return | |
2320 | ||
2321 | self.post_extract(info_dict) | |
2322 | self._num_downloads += 1 | |
2323 | ||
2324 | # info_dict['_filename'] needs to be set for backward compatibility | |
2325 | info_dict['_filename'] = full_filename = self.prepare_filename(info_dict, warn=True) | |
2326 | temp_filename = self.prepare_filename(info_dict, 'temp') | |
2327 | files_to_move = {} | |
2328 | ||
2329 | # Forced printings | |
2330 | self.__forced_printings(info_dict, full_filename, incomplete=False) | |
2331 | ||
2332 | if self.params.get('simulate', False): | |
2333 | if self.params.get('force_write_download_archive', False): | |
2334 | self.record_download_archive(info_dict) | |
2335 | ||
2336 | # Do nothing else if in simulate mode | |
2337 | return | |
2338 | ||
2339 | if full_filename is None: | |
2340 | return | |
2341 | ||
2342 | if not self._ensure_dir_exists(encodeFilename(full_filename)): | |
2343 | return | |
2344 | if not self._ensure_dir_exists(encodeFilename(temp_filename)): | |
2345 | return | |
2346 | ||
2347 | if self.params.get('writedescription', False): | |
2348 | descfn = self.prepare_filename(info_dict, 'description') | |
2349 | if not self._ensure_dir_exists(encodeFilename(descfn)): | |
2350 | return | |
2351 | if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(descfn)): | |
2352 | self.to_screen('[info] Video description is already present') | |
2353 | elif info_dict.get('description') is None: | |
2354 | self.report_warning('There\'s no description to write.') | |
2355 | else: | |
2356 | try: | |
2357 | self.to_screen('[info] Writing video description to: ' + descfn) | |
2358 | with io.open(encodeFilename(descfn), 'w', encoding='utf-8') as descfile: | |
2359 | descfile.write(info_dict['description']) | |
2360 | except (OSError, IOError): | |
2361 | self.report_error('Cannot write description file ' + descfn) | |
2362 | return | |
2363 | ||
2364 | if self.params.get('writeannotations', False): | |
2365 | annofn = self.prepare_filename(info_dict, 'annotation') | |
2366 | if not self._ensure_dir_exists(encodeFilename(annofn)): | |
2367 | return | |
2368 | if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(annofn)): | |
2369 | self.to_screen('[info] Video annotations are already present') | |
2370 | elif not info_dict.get('annotations'): | |
2371 | self.report_warning('There are no annotations to write.') | |
2372 | else: | |
2373 | try: | |
2374 | self.to_screen('[info] Writing video annotations to: ' + annofn) | |
2375 | with io.open(encodeFilename(annofn), 'w', encoding='utf-8') as annofile: | |
2376 | annofile.write(info_dict['annotations']) | |
2377 | except (KeyError, TypeError): | |
2378 | self.report_warning('There are no annotations to write.') | |
2379 | except (OSError, IOError): | |
2380 | self.report_error('Cannot write annotations file: ' + annofn) | |
2381 | return | |
2382 | ||
2383 | subtitles_are_requested = any([self.params.get('writesubtitles', False), | |
2384 | self.params.get('writeautomaticsub')]) | |
2385 | ||
2386 | if subtitles_are_requested and info_dict.get('requested_subtitles'): | |
2387 | # subtitles download errors are already managed as troubles in relevant IE | |
2388 | # that way it will silently go on when used with unsupporting IE | |
2389 | subtitles = info_dict['requested_subtitles'] | |
2390 | # ie = self.get_info_extractor(info_dict['extractor_key']) | |
2391 | for sub_lang, sub_info in subtitles.items(): | |
2392 | sub_format = sub_info['ext'] | |
2393 | sub_filename = subtitles_filename(temp_filename, sub_lang, sub_format, info_dict.get('ext')) | |
2394 | sub_filename_final = subtitles_filename( | |
2395 | self.prepare_filename(info_dict, 'subtitle'), sub_lang, sub_format, info_dict.get('ext')) | |
2396 | if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(sub_filename)): | |
2397 | self.to_screen('[info] Video subtitle %s.%s is already present' % (sub_lang, sub_format)) | |
2398 | sub_info['filepath'] = sub_filename | |
2399 | files_to_move[sub_filename] = sub_filename_final | |
2400 | else: | |
2401 | self.to_screen('[info] Writing video subtitles to: ' + sub_filename) | |
2402 | if sub_info.get('data') is not None: | |
2403 | try: | |
2404 | # Use newline='' to prevent conversion of newline characters | |
2405 | # See https://github.com/ytdl-org/youtube-dl/issues/10268 | |
2406 | with io.open(encodeFilename(sub_filename), 'w', encoding='utf-8', newline='') as subfile: | |
2407 | subfile.write(sub_info['data']) | |
2408 | sub_info['filepath'] = sub_filename | |
2409 | files_to_move[sub_filename] = sub_filename_final | |
2410 | except (OSError, IOError): | |
2411 | self.report_error('Cannot write subtitles file ' + sub_filename) | |
2412 | return | |
2413 | else: | |
2414 | try: | |
2415 | self.dl(sub_filename, sub_info.copy(), subtitle=True) | |
2416 | sub_info['filepath'] = sub_filename | |
2417 | files_to_move[sub_filename] = sub_filename_final | |
2418 | except (ExtractorError, IOError, OSError, ValueError) + network_exceptions as err: | |
2419 | self.report_warning('Unable to download subtitle for "%s": %s' % | |
2420 | (sub_lang, error_to_compat_str(err))) | |
2421 | continue | |
2422 | ||
2423 | if self.params.get('writeinfojson', False): | |
2424 | infofn = self.prepare_filename(info_dict, 'infojson') | |
2425 | if not self._ensure_dir_exists(encodeFilename(infofn)): | |
2426 | return | |
2427 | if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(infofn)): | |
2428 | self.to_screen('[info] Video metadata is already present') | |
2429 | else: | |
2430 | self.to_screen('[info] Writing video metadata as JSON to: ' + infofn) | |
2431 | try: | |
2432 | write_json_file(self.filter_requested_info(info_dict, self.params.get('clean_infojson', True)), infofn) | |
2433 | except (OSError, IOError): | |
2434 | self.report_error('Cannot write video metadata to JSON file ' + infofn) | |
2435 | return | |
2436 | info_dict['__infojson_filename'] = infofn | |
2437 | ||
2438 | for thumb_ext in self._write_thumbnails(info_dict, temp_filename): | |
2439 | thumb_filename_temp = replace_extension(temp_filename, thumb_ext, info_dict.get('ext')) | |
2440 | thumb_filename = replace_extension( | |
2441 | self.prepare_filename(info_dict, 'thumbnail'), thumb_ext, info_dict.get('ext')) | |
2442 | files_to_move[thumb_filename_temp] = thumb_filename | |
2443 | ||
2444 | # Write internet shortcut files | |
2445 | url_link = webloc_link = desktop_link = False | |
2446 | if self.params.get('writelink', False): | |
2447 | if sys.platform == "darwin": # macOS. | |
2448 | webloc_link = True | |
2449 | elif sys.platform.startswith("linux"): | |
2450 | desktop_link = True | |
2451 | else: # if sys.platform in ['win32', 'cygwin']: | |
2452 | url_link = True | |
2453 | if self.params.get('writeurllink', False): | |
2454 | url_link = True | |
2455 | if self.params.get('writewebloclink', False): | |
2456 | webloc_link = True | |
2457 | if self.params.get('writedesktoplink', False): | |
2458 | desktop_link = True | |
2459 | ||
2460 | if url_link or webloc_link or desktop_link: | |
2461 | if 'webpage_url' not in info_dict: | |
2462 | self.report_error('Cannot write internet shortcut file because the "webpage_url" field is missing in the media information') | |
2463 | return | |
2464 | ascii_url = iri_to_uri(info_dict['webpage_url']) | |
2465 | ||
2466 | def _write_link_file(extension, template, newline, embed_filename): | |
2467 | linkfn = replace_extension(full_filename, extension, info_dict.get('ext')) | |
2468 | if self.params.get('overwrites', True) and os.path.exists(encodeFilename(linkfn)): | |
2469 | self.to_screen('[info] Internet shortcut is already present') | |
2470 | else: | |
2471 | try: | |
2472 | self.to_screen('[info] Writing internet shortcut to: ' + linkfn) | |
2473 | with io.open(encodeFilename(to_high_limit_path(linkfn)), 'w', encoding='utf-8', newline=newline) as linkfile: | |
2474 | template_vars = {'url': ascii_url} | |
2475 | if embed_filename: | |
2476 | template_vars['filename'] = linkfn[:-(len(extension) + 1)] | |
2477 | linkfile.write(template % template_vars) | |
2478 | except (OSError, IOError): | |
2479 | self.report_error('Cannot write internet shortcut ' + linkfn) | |
2480 | return False | |
2481 | return True | |
2482 | ||
2483 | if url_link: | |
2484 | if not _write_link_file('url', DOT_URL_LINK_TEMPLATE, '\r\n', embed_filename=False): | |
2485 | return | |
2486 | if webloc_link: | |
2487 | if not _write_link_file('webloc', DOT_WEBLOC_LINK_TEMPLATE, '\n', embed_filename=False): | |
2488 | return | |
2489 | if desktop_link: | |
2490 | if not _write_link_file('desktop', DOT_DESKTOP_LINK_TEMPLATE, '\n', embed_filename=True): | |
2491 | return | |
2492 | ||
2493 | try: | |
2494 | info_dict, files_to_move = self.pre_process(info_dict, 'before_dl', files_to_move) | |
2495 | except PostProcessingError as err: | |
2496 | self.report_error('Preprocessing: %s' % str(err)) | |
2497 | return | |
2498 | ||
2499 | must_record_download_archive = False | |
2500 | if self.params.get('skip_download', False): | |
2501 | info_dict['filepath'] = temp_filename | |
2502 | info_dict['__finaldir'] = os.path.dirname(os.path.abspath(encodeFilename(full_filename))) | |
2503 | info_dict['__files_to_move'] = files_to_move | |
2504 | info_dict = self.run_pp(MoveFilesAfterDownloadPP(self, False), info_dict) | |
2505 | else: | |
2506 | # Download | |
2507 | try: | |
2508 | ||
2509 | def existing_file(*filepaths): | |
2510 | ext = info_dict.get('ext') | |
2511 | final_ext = self.params.get('final_ext', ext) | |
2512 | existing_files = [] | |
2513 | for file in orderedSet(filepaths): | |
2514 | if final_ext != ext: | |
2515 | converted = replace_extension(file, final_ext, ext) | |
2516 | if os.path.exists(encodeFilename(converted)): | |
2517 | existing_files.append(converted) | |
2518 | if os.path.exists(encodeFilename(file)): | |
2519 | existing_files.append(file) | |
2520 | ||
2521 | if not existing_files or self.params.get('overwrites', False): | |
2522 | for file in orderedSet(existing_files): | |
2523 | self.report_file_delete(file) | |
2524 | os.remove(encodeFilename(file)) | |
2525 | return None | |
2526 | ||
2527 | self.report_file_already_downloaded(existing_files[0]) | |
2528 | info_dict['ext'] = os.path.splitext(existing_files[0])[1][1:] | |
2529 | return existing_files[0] | |
2530 | ||
2531 | success = True | |
2532 | if info_dict.get('requested_formats') is not None: | |
2533 | ||
2534 | def compatible_formats(formats): | |
2535 | # TODO: some formats actually allow this (mkv, webm, ogg, mp4), but not all of them. | |
2536 | video_formats = [format for format in formats if format.get('vcodec') != 'none'] | |
2537 | audio_formats = [format for format in formats if format.get('acodec') != 'none'] | |
2538 | if len(video_formats) > 2 or len(audio_formats) > 2: | |
2539 | return False | |
2540 | ||
2541 | # Check extension | |
2542 | exts = set(format.get('ext') for format in formats) | |
2543 | COMPATIBLE_EXTS = ( | |
2544 | set(('mp3', 'mp4', 'm4a', 'm4p', 'm4b', 'm4r', 'm4v', 'ismv', 'isma')), | |
2545 | set(('webm',)), | |
2546 | ) | |
2547 | for ext_sets in COMPATIBLE_EXTS: | |
2548 | if ext_sets.issuperset(exts): | |
2549 | return True | |
2550 | # TODO: Check acodec/vcodec | |
2551 | return False | |
2552 | ||
2553 | requested_formats = info_dict['requested_formats'] | |
2554 | old_ext = info_dict['ext'] | |
2555 | if self.params.get('merge_output_format') is None: | |
2556 | if not compatible_formats(requested_formats): | |
2557 | info_dict['ext'] = 'mkv' | |
2558 | self.report_warning( | |
2559 | 'Requested formats are incompatible for merge and will be merged into mkv.') | |
2560 | if (info_dict['ext'] == 'webm' | |
2561 | and self.params.get('writethumbnail', False) | |
2562 | and info_dict.get('thumbnails')): | |
2563 | info_dict['ext'] = 'mkv' | |
2564 | self.report_warning( | |
2565 | 'webm doesn\'t support embedding a thumbnail, mkv will be used.') | |
2566 | ||
2567 | def correct_ext(filename): | |
2568 | filename_real_ext = os.path.splitext(filename)[1][1:] | |
2569 | filename_wo_ext = ( | |
2570 | os.path.splitext(filename)[0] | |
2571 | if filename_real_ext == old_ext | |
2572 | else filename) | |
2573 | return '%s.%s' % (filename_wo_ext, info_dict['ext']) | |
2574 | ||
2575 | # Ensure filename always has a correct extension for successful merge | |
2576 | full_filename = correct_ext(full_filename) | |
2577 | temp_filename = correct_ext(temp_filename) | |
2578 | dl_filename = existing_file(full_filename, temp_filename) | |
2579 | info_dict['__real_download'] = False | |
2580 | ||
2581 | _protocols = set(determine_protocol(f) for f in requested_formats) | |
2582 | if len(_protocols) == 1: | |
2583 | info_dict['protocol'] = _protocols.pop() | |
2584 | directly_mergable = ( | |
2585 | 'no-direct-merge' not in self.params.get('compat_opts', []) | |
2586 | and info_dict.get('protocol') is not None # All requested formats have same protocol | |
2587 | and not self.params.get('allow_unplayable_formats') | |
2588 | and get_suitable_downloader(info_dict, self.params).__name__ == 'FFmpegFD') | |
2589 | if directly_mergable: | |
2590 | info_dict['url'] = requested_formats[0]['url'] | |
2591 | # Treat it as a single download | |
2592 | dl_filename = existing_file(full_filename, temp_filename) | |
2593 | if dl_filename is None: | |
2594 | success, real_download = self.dl(temp_filename, info_dict) | |
2595 | info_dict['__real_download'] = real_download | |
2596 | else: | |
2597 | downloaded = [] | |
2598 | merger = FFmpegMergerPP(self) | |
2599 | if self.params.get('allow_unplayable_formats'): | |
2600 | self.report_warning( | |
2601 | 'You have requested merging of multiple formats ' | |
2602 | 'while also allowing unplayable formats to be downloaded. ' | |
2603 | 'The formats won\'t be merged to prevent data corruption.') | |
2604 | elif not merger.available: | |
2605 | self.report_warning( | |
2606 | 'You have requested merging of multiple formats but ffmpeg is not installed. ' | |
2607 | 'The formats won\'t be merged.') | |
2608 | ||
2609 | if dl_filename is None: | |
2610 | for f in requested_formats: | |
2611 | new_info = dict(info_dict) | |
2612 | del new_info['requested_formats'] | |
2613 | new_info.update(f) | |
2614 | fname = prepend_extension( | |
2615 | self.prepare_filename(new_info, 'temp'), | |
2616 | 'f%s' % f['format_id'], new_info['ext']) | |
2617 | if not self._ensure_dir_exists(fname): | |
2618 | return | |
2619 | downloaded.append(fname) | |
2620 | partial_success, real_download = self.dl(fname, new_info) | |
2621 | info_dict['__real_download'] = info_dict['__real_download'] or real_download | |
2622 | success = success and partial_success | |
2623 | if merger.available and not self.params.get('allow_unplayable_formats'): | |
2624 | info_dict['__postprocessors'].append(merger) | |
2625 | info_dict['__files_to_merge'] = downloaded | |
2626 | # Even if there were no downloads, it is being merged only now | |
2627 | info_dict['__real_download'] = True | |
2628 | else: | |
2629 | for file in downloaded: | |
2630 | files_to_move[file] = None | |
2631 | else: | |
2632 | # Just a single file | |
2633 | dl_filename = existing_file(full_filename, temp_filename) | |
2634 | if dl_filename is None: | |
2635 | success, real_download = self.dl(temp_filename, info_dict) | |
2636 | info_dict['__real_download'] = real_download | |
2637 | ||
2638 | dl_filename = dl_filename or temp_filename | |
2639 | info_dict['__finaldir'] = os.path.dirname(os.path.abspath(encodeFilename(full_filename))) | |
2640 | ||
2641 | except network_exceptions as err: | |
2642 | self.report_error('unable to download video data: %s' % error_to_compat_str(err)) | |
2643 | return | |
2644 | except (OSError, IOError) as err: | |
2645 | raise UnavailableVideoError(err) | |
2646 | except (ContentTooShortError, ) as err: | |
2647 | self.report_error('content too short (expected %s bytes and served %s)' % (err.expected, err.downloaded)) | |
2648 | return | |
2649 | ||
2650 | if success and full_filename != '-': | |
2651 | # Fixup content | |
2652 | fixup_policy = self.params.get('fixup') | |
2653 | if fixup_policy is None: | |
2654 | fixup_policy = 'detect_or_warn' | |
2655 | ||
2656 | INSTALL_FFMPEG_MESSAGE = 'Install ffmpeg to fix this automatically.' | |
2657 | ||
2658 | stretched_ratio = info_dict.get('stretched_ratio') | |
2659 | if stretched_ratio is not None and stretched_ratio != 1: | |
2660 | if fixup_policy == 'warn': | |
2661 | self.report_warning('%s: Non-uniform pixel ratio (%s)' % ( | |
2662 | info_dict['id'], stretched_ratio)) | |
2663 | elif fixup_policy == 'detect_or_warn': | |
2664 | stretched_pp = FFmpegFixupStretchedPP(self) | |
2665 | if stretched_pp.available: | |
2666 | info_dict['__postprocessors'].append(stretched_pp) | |
2667 | else: | |
2668 | self.report_warning( | |
2669 | '%s: Non-uniform pixel ratio (%s). %s' | |
2670 | % (info_dict['id'], stretched_ratio, INSTALL_FFMPEG_MESSAGE)) | |
2671 | else: | |
2672 | assert fixup_policy in ('ignore', 'never') | |
2673 | ||
2674 | if (info_dict.get('requested_formats') is None | |
2675 | and info_dict.get('container') == 'm4a_dash' | |
2676 | and info_dict.get('ext') == 'm4a'): | |
2677 | if fixup_policy == 'warn': | |
2678 | self.report_warning( | |
2679 | '%s: writing DASH m4a. ' | |
2680 | 'Only some players support this container.' | |
2681 | % info_dict['id']) | |
2682 | elif fixup_policy == 'detect_or_warn': | |
2683 | fixup_pp = FFmpegFixupM4aPP(self) | |
2684 | if fixup_pp.available: | |
2685 | info_dict['__postprocessors'].append(fixup_pp) | |
2686 | else: | |
2687 | self.report_warning( | |
2688 | '%s: writing DASH m4a. ' | |
2689 | 'Only some players support this container. %s' | |
2690 | % (info_dict['id'], INSTALL_FFMPEG_MESSAGE)) | |
2691 | else: | |
2692 | assert fixup_policy in ('ignore', 'never') | |
2693 | ||
2694 | if ('protocol' in info_dict | |
2695 | and get_suitable_downloader(info_dict, self.params).__name__ == 'HlsFD'): | |
2696 | if fixup_policy == 'warn': | |
2697 | self.report_warning('%s: malformed AAC bitstream detected.' % ( | |
2698 | info_dict['id'])) | |
2699 | elif fixup_policy == 'detect_or_warn': | |
2700 | fixup_pp = FFmpegFixupM3u8PP(self) | |
2701 | if fixup_pp.available: | |
2702 | info_dict['__postprocessors'].append(fixup_pp) | |
2703 | else: | |
2704 | self.report_warning( | |
2705 | '%s: malformed AAC bitstream detected. %s' | |
2706 | % (info_dict['id'], INSTALL_FFMPEG_MESSAGE)) | |
2707 | else: | |
2708 | assert fixup_policy in ('ignore', 'never') | |
2709 | ||
2710 | try: | |
2711 | info_dict = self.post_process(dl_filename, info_dict, files_to_move) | |
2712 | except PostProcessingError as err: | |
2713 | self.report_error('Postprocessing: %s' % str(err)) | |
2714 | return | |
2715 | try: | |
2716 | for ph in self._post_hooks: | |
2717 | ph(info_dict['filepath']) | |
2718 | except Exception as err: | |
2719 | self.report_error('post hooks: %s' % str(err)) | |
2720 | return | |
2721 | must_record_download_archive = True | |
2722 | ||
2723 | if must_record_download_archive or self.params.get('force_write_download_archive', False): | |
2724 | self.record_download_archive(info_dict) | |
2725 | max_downloads = self.params.get('max_downloads') | |
2726 | if max_downloads is not None and self._num_downloads >= int(max_downloads): | |
2727 | raise MaxDownloadsReached() | |
2728 | ||
2729 | def download(self, url_list): | |
2730 | """Download a given list of URLs.""" | |
2731 | outtmpl = self.outtmpl_dict['default'] | |
2732 | if (len(url_list) > 1 | |
2733 | and outtmpl != '-' | |
2734 | and '%' not in outtmpl | |
2735 | and self.params.get('max_downloads') != 1): | |
2736 | raise SameFileError(outtmpl) | |
2737 | ||
2738 | for url in url_list: | |
2739 | try: | |
2740 | # It also downloads the videos | |
2741 | res = self.extract_info( | |
2742 | url, force_generic_extractor=self.params.get('force_generic_extractor', False)) | |
2743 | except UnavailableVideoError: | |
2744 | self.report_error('unable to download video') | |
2745 | except MaxDownloadsReached: | |
2746 | self.to_screen('[info] Maximum number of downloaded files reached') | |
2747 | raise | |
2748 | except ExistingVideoReached: | |
2749 | self.to_screen('[info] Encountered a file that is already in the archive, stopping due to --break-on-existing') | |
2750 | raise | |
2751 | except RejectedVideoReached: | |
2752 | self.to_screen('[info] Encountered a file that did not match filter, stopping due to --break-on-reject') | |
2753 | raise | |
2754 | else: | |
2755 | if self.params.get('dump_single_json', False): | |
2756 | self.post_extract(res) | |
2757 | self.to_stdout(json.dumps(res, default=repr)) | |
2758 | ||
2759 | return self._download_retcode | |
2760 | ||
2761 | def download_with_info_file(self, info_filename): | |
2762 | with contextlib.closing(fileinput.FileInput( | |
2763 | [info_filename], mode='r', | |
2764 | openhook=fileinput.hook_encoded('utf-8'))) as f: | |
2765 | # FileInput doesn't have a read method, we can't call json.load | |
2766 | info = self.filter_requested_info(json.loads('\n'.join(f)), self.params.get('clean_infojson', True)) | |
2767 | try: | |
2768 | self.process_ie_result(info, download=True) | |
2769 | except (DownloadError, EntryNotInPlaylist): | |
2770 | webpage_url = info.get('webpage_url') | |
2771 | if webpage_url is not None: | |
2772 | self.report_warning('The info failed to download, trying with "%s"' % webpage_url) | |
2773 | return self.download([webpage_url]) | |
2774 | else: | |
2775 | raise | |
2776 | return self._download_retcode | |
2777 | ||
2778 | @staticmethod | |
2779 | def filter_requested_info(info_dict, actually_filter=True): | |
2780 | remove_keys = ['__original_infodict'] # Always remove this since this may contain a copy of the entire dict | |
2781 | keep_keys = ['_type'], # Always keep this to facilitate load-info-json | |
2782 | if actually_filter: | |
2783 | remove_keys += ('requested_formats', 'requested_subtitles', 'requested_entries', 'filepath', 'entries') | |
2784 | empty_values = (None, {}, [], set(), tuple()) | |
2785 | reject = lambda k, v: k not in keep_keys and ( | |
2786 | k.startswith('_') or k in remove_keys or v in empty_values) | |
2787 | else: | |
2788 | info_dict['epoch'] = int(time.time()) | |
2789 | reject = lambda k, v: k in remove_keys | |
2790 | filter_fn = lambda obj: ( | |
2791 | list(map(filter_fn, obj)) if isinstance(obj, (list, tuple, set)) | |
2792 | else obj if not isinstance(obj, dict) | |
2793 | else dict((k, filter_fn(v)) for k, v in obj.items() if not reject(k, v))) | |
2794 | return filter_fn(info_dict) | |
2795 | ||
2796 | def run_pp(self, pp, infodict): | |
2797 | files_to_delete = [] | |
2798 | if '__files_to_move' not in infodict: | |
2799 | infodict['__files_to_move'] = {} | |
2800 | files_to_delete, infodict = pp.run(infodict) | |
2801 | if not files_to_delete: | |
2802 | return infodict | |
2803 | ||
2804 | if self.params.get('keepvideo', False): | |
2805 | for f in files_to_delete: | |
2806 | infodict['__files_to_move'].setdefault(f, '') | |
2807 | else: | |
2808 | for old_filename in set(files_to_delete): | |
2809 | self.to_screen('Deleting original file %s (pass -k to keep)' % old_filename) | |
2810 | try: | |
2811 | os.remove(encodeFilename(old_filename)) | |
2812 | except (IOError, OSError): | |
2813 | self.report_warning('Unable to remove downloaded original file') | |
2814 | if old_filename in infodict['__files_to_move']: | |
2815 | del infodict['__files_to_move'][old_filename] | |
2816 | return infodict | |
2817 | ||
2818 | @staticmethod | |
2819 | def post_extract(info_dict): | |
2820 | def actual_post_extract(info_dict): | |
2821 | if info_dict.get('_type') in ('playlist', 'multi_video'): | |
2822 | for video_dict in info_dict.get('entries', {}): | |
2823 | actual_post_extract(video_dict or {}) | |
2824 | return | |
2825 | ||
2826 | post_extractor = info_dict.get('__post_extractor') or (lambda: {}) | |
2827 | extra = post_extractor().items() | |
2828 | info_dict.update(extra) | |
2829 | info_dict.pop('__post_extractor', None) | |
2830 | ||
2831 | original_infodict = info_dict.get('__original_infodict') or {} | |
2832 | original_infodict.update(extra) | |
2833 | original_infodict.pop('__post_extractor', None) | |
2834 | ||
2835 | actual_post_extract(info_dict or {}) | |
2836 | ||
2837 | def pre_process(self, ie_info, key='pre_process', files_to_move=None): | |
2838 | info = dict(ie_info) | |
2839 | info['__files_to_move'] = files_to_move or {} | |
2840 | for pp in self._pps[key]: | |
2841 | info = self.run_pp(pp, info) | |
2842 | return info, info.pop('__files_to_move', None) | |
2843 | ||
2844 | def post_process(self, filename, ie_info, files_to_move=None): | |
2845 | """Run all the postprocessors on the given file.""" | |
2846 | info = dict(ie_info) | |
2847 | info['filepath'] = filename | |
2848 | info['__files_to_move'] = files_to_move or {} | |
2849 | ||
2850 | for pp in ie_info.get('__postprocessors', []) + self._pps['post_process']: | |
2851 | info = self.run_pp(pp, info) | |
2852 | info = self.run_pp(MoveFilesAfterDownloadPP(self), info) | |
2853 | del info['__files_to_move'] | |
2854 | for pp in self._pps['after_move']: | |
2855 | info = self.run_pp(pp, info) | |
2856 | return info | |
2857 | ||
2858 | def _make_archive_id(self, info_dict): | |
2859 | video_id = info_dict.get('id') | |
2860 | if not video_id: | |
2861 | return | |
2862 | # Future-proof against any change in case | |
2863 | # and backwards compatibility with prior versions | |
2864 | extractor = info_dict.get('extractor_key') or info_dict.get('ie_key') # key in a playlist | |
2865 | if extractor is None: | |
2866 | url = str_or_none(info_dict.get('url')) | |
2867 | if not url: | |
2868 | return | |
2869 | # Try to find matching extractor for the URL and take its ie_key | |
2870 | for ie in self._ies: | |
2871 | if ie.suitable(url): | |
2872 | extractor = ie.ie_key() | |
2873 | break | |
2874 | else: | |
2875 | return | |
2876 | return '%s %s' % (extractor.lower(), video_id) | |
2877 | ||
2878 | def in_download_archive(self, info_dict): | |
2879 | fn = self.params.get('download_archive') | |
2880 | if fn is None: | |
2881 | return False | |
2882 | ||
2883 | vid_id = self._make_archive_id(info_dict) | |
2884 | if not vid_id: | |
2885 | return False # Incomplete video information | |
2886 | ||
2887 | return vid_id in self.archive | |
2888 | ||
2889 | def record_download_archive(self, info_dict): | |
2890 | fn = self.params.get('download_archive') | |
2891 | if fn is None: | |
2892 | return | |
2893 | vid_id = self._make_archive_id(info_dict) | |
2894 | assert vid_id | |
2895 | with locked_file(fn, 'a', encoding='utf-8') as archive_file: | |
2896 | archive_file.write(vid_id + '\n') | |
2897 | self.archive.add(vid_id) | |
2898 | ||
2899 | @staticmethod | |
2900 | def format_resolution(format, default='unknown'): | |
2901 | if format.get('vcodec') == 'none': | |
2902 | return 'audio only' | |
2903 | if format.get('resolution') is not None: | |
2904 | return format['resolution'] | |
2905 | if format.get('width') and format.get('height'): | |
2906 | res = '%dx%d' % (format['width'], format['height']) | |
2907 | elif format.get('height'): | |
2908 | res = '%sp' % format['height'] | |
2909 | elif format.get('width'): | |
2910 | res = '%dx?' % format['width'] | |
2911 | else: | |
2912 | res = default | |
2913 | return res | |
2914 | ||
2915 | def _format_note(self, fdict): | |
2916 | res = '' | |
2917 | if fdict.get('ext') in ['f4f', 'f4m']: | |
2918 | res += '(unsupported) ' | |
2919 | if fdict.get('language'): | |
2920 | if res: | |
2921 | res += ' ' | |
2922 | res += '[%s] ' % fdict['language'] | |
2923 | if fdict.get('format_note') is not None: | |
2924 | res += fdict['format_note'] + ' ' | |
2925 | if fdict.get('tbr') is not None: | |
2926 | res += '%4dk ' % fdict['tbr'] | |
2927 | if fdict.get('container') is not None: | |
2928 | if res: | |
2929 | res += ', ' | |
2930 | res += '%s container' % fdict['container'] | |
2931 | if (fdict.get('vcodec') is not None | |
2932 | and fdict.get('vcodec') != 'none'): | |
2933 | if res: | |
2934 | res += ', ' | |
2935 | res += fdict['vcodec'] | |
2936 | if fdict.get('vbr') is not None: | |
2937 | res += '@' | |
2938 | elif fdict.get('vbr') is not None and fdict.get('abr') is not None: | |
2939 | res += 'video@' | |
2940 | if fdict.get('vbr') is not None: | |
2941 | res += '%4dk' % fdict['vbr'] | |
2942 | if fdict.get('fps') is not None: | |
2943 | if res: | |
2944 | res += ', ' | |
2945 | res += '%sfps' % fdict['fps'] | |
2946 | if fdict.get('acodec') is not None: | |
2947 | if res: | |
2948 | res += ', ' | |
2949 | if fdict['acodec'] == 'none': | |
2950 | res += 'video only' | |
2951 | else: | |
2952 | res += '%-5s' % fdict['acodec'] | |
2953 | elif fdict.get('abr') is not None: | |
2954 | if res: | |
2955 | res += ', ' | |
2956 | res += 'audio' | |
2957 | if fdict.get('abr') is not None: | |
2958 | res += '@%3dk' % fdict['abr'] | |
2959 | if fdict.get('asr') is not None: | |
2960 | res += ' (%5dHz)' % fdict['asr'] | |
2961 | if fdict.get('filesize') is not None: | |
2962 | if res: | |
2963 | res += ', ' | |
2964 | res += format_bytes(fdict['filesize']) | |
2965 | elif fdict.get('filesize_approx') is not None: | |
2966 | if res: | |
2967 | res += ', ' | |
2968 | res += '~' + format_bytes(fdict['filesize_approx']) | |
2969 | return res | |
2970 | ||
2971 | def _format_note_table(self, f): | |
2972 | def join_fields(*vargs): | |
2973 | return ', '.join((val for val in vargs if val != '')) | |
2974 | ||
2975 | return join_fields( | |
2976 | 'UNSUPPORTED' if f.get('ext') in ('f4f', 'f4m') else '', | |
2977 | format_field(f, 'language', '[%s]'), | |
2978 | format_field(f, 'format_note'), | |
2979 | format_field(f, 'container', ignore=(None, f.get('ext'))), | |
2980 | format_field(f, 'asr', '%5dHz')) | |
2981 | ||
2982 | def list_formats(self, info_dict): | |
2983 | formats = info_dict.get('formats', [info_dict]) | |
2984 | new_format = ( | |
2985 | 'list-formats' not in self.params.get('compat_opts', []) | |
2986 | and self.params.get('list_formats_as_table', True) is not False) | |
2987 | if new_format: | |
2988 | table = [ | |
2989 | [ | |
2990 | format_field(f, 'format_id'), | |
2991 | format_field(f, 'ext'), | |
2992 | self.format_resolution(f), | |
2993 | format_field(f, 'fps', '%d'), | |
2994 | '|', | |
2995 | format_field(f, 'filesize', ' %s', func=format_bytes) + format_field(f, 'filesize_approx', '~%s', func=format_bytes), | |
2996 | format_field(f, 'tbr', '%4dk'), | |
2997 | shorten_protocol_name(f.get('protocol', '').replace("native", "n")), | |
2998 | '|', | |
2999 | format_field(f, 'vcodec', default='unknown').replace('none', ''), | |
3000 | format_field(f, 'vbr', '%4dk'), | |
3001 | format_field(f, 'acodec', default='unknown').replace('none', ''), | |
3002 | format_field(f, 'abr', '%3dk'), | |
3003 | format_field(f, 'asr', '%5dHz'), | |
3004 | self._format_note_table(f)] | |
3005 | for f in formats | |
3006 | if f.get('preference') is None or f['preference'] >= -1000] | |
3007 | header_line = ['ID', 'EXT', 'RESOLUTION', 'FPS', '|', ' FILESIZE', ' TBR', 'PROTO', | |
3008 | '|', 'VCODEC', ' VBR', 'ACODEC', ' ABR', ' ASR', 'NOTE'] | |
3009 | else: | |
3010 | table = [ | |
3011 | [ | |
3012 | format_field(f, 'format_id'), | |
3013 | format_field(f, 'ext'), | |
3014 | self.format_resolution(f), | |
3015 | self._format_note(f)] | |
3016 | for f in formats | |
3017 | if f.get('preference') is None or f['preference'] >= -1000] | |
3018 | header_line = ['format code', 'extension', 'resolution', 'note'] | |
3019 | ||
3020 | self.to_screen( | |
3021 | '[info] Available formats for %s:\n%s' % (info_dict['id'], render_table( | |
3022 | header_line, | |
3023 | table, | |
3024 | delim=new_format, | |
3025 | extraGap=(0 if new_format else 1), | |
3026 | hideEmpty=new_format))) | |
3027 | ||
3028 | def list_thumbnails(self, info_dict): | |
3029 | thumbnails = info_dict.get('thumbnails') | |
3030 | if not thumbnails: | |
3031 | self.to_screen('[info] No thumbnails present for %s' % info_dict['id']) | |
3032 | return | |
3033 | ||
3034 | self.to_screen( | |
3035 | '[info] Thumbnails for %s:' % info_dict['id']) | |
3036 | self.to_screen(render_table( | |
3037 | ['ID', 'width', 'height', 'URL'], | |
3038 | [[t['id'], t.get('width', 'unknown'), t.get('height', 'unknown'), t['url']] for t in thumbnails])) | |
3039 | ||
3040 | def list_subtitles(self, video_id, subtitles, name='subtitles'): | |
3041 | if not subtitles: | |
3042 | self.to_screen('%s has no %s' % (video_id, name)) | |
3043 | return | |
3044 | self.to_screen( | |
3045 | 'Available %s for %s:' % (name, video_id)) | |
3046 | ||
3047 | def _row(lang, formats): | |
3048 | exts, names = zip(*((f['ext'], f.get('name', 'unknown')) for f in reversed(formats))) | |
3049 | if len(set(names)) == 1: | |
3050 | names = [] if names[0] == 'unknown' else names[:1] | |
3051 | return [lang, ', '.join(names), ', '.join(exts)] | |
3052 | ||
3053 | self.to_screen(render_table( | |
3054 | ['Language', 'Name', 'Formats'], | |
3055 | [_row(lang, formats) for lang, formats in subtitles.items()], | |
3056 | hideEmpty=True)) | |
3057 | ||
3058 | def urlopen(self, req): | |
3059 | """ Start an HTTP download """ | |
3060 | if isinstance(req, compat_basestring): | |
3061 | req = sanitized_Request(req) | |
3062 | return self._opener.open(req, timeout=self._socket_timeout) | |
3063 | ||
3064 | def print_debug_header(self): | |
3065 | if not self.params.get('verbose'): | |
3066 | return | |
3067 | ||
3068 | if type('') is not compat_str: | |
3069 | # Python 2.6 on SLES11 SP1 (https://github.com/ytdl-org/youtube-dl/issues/3326) | |
3070 | self.report_warning( | |
3071 | 'Your Python is broken! Update to a newer and supported version') | |
3072 | ||
3073 | stdout_encoding = getattr( | |
3074 | sys.stdout, 'encoding', 'missing (%s)' % type(sys.stdout).__name__) | |
3075 | encoding_str = ( | |
3076 | '[debug] Encodings: locale %s, fs %s, out %s, pref %s\n' % ( | |
3077 | locale.getpreferredencoding(), | |
3078 | sys.getfilesystemencoding(), | |
3079 | stdout_encoding, | |
3080 | self.get_encoding())) | |
3081 | write_string(encoding_str, encoding=None) | |
3082 | ||
3083 | source = ( | |
3084 | '(exe)' if hasattr(sys, 'frozen') | |
3085 | else '(zip)' if isinstance(globals().get('__loader__'), zipimporter) | |
3086 | else '(source)' if os.path.basename(sys.argv[0]) == '__main__.py' | |
3087 | else '') | |
3088 | self._write_string('[debug] yt-dlp version %s %s\n' % (__version__, source)) | |
3089 | if _LAZY_LOADER: | |
3090 | self._write_string('[debug] Lazy loading extractors enabled\n') | |
3091 | if _PLUGIN_CLASSES: | |
3092 | self._write_string( | |
3093 | '[debug] Plugin Extractors: %s\n' % [ie.ie_key() for ie in _PLUGIN_CLASSES]) | |
3094 | if self.params.get('compat_opts'): | |
3095 | self._write_string( | |
3096 | '[debug] Compatibility options: %s\n' % ', '.join(self.params.get('compat_opts'))) | |
3097 | try: | |
3098 | sp = subprocess.Popen( | |
3099 | ['git', 'rev-parse', '--short', 'HEAD'], | |
3100 | stdout=subprocess.PIPE, stderr=subprocess.PIPE, | |
3101 | cwd=os.path.dirname(os.path.abspath(__file__))) | |
3102 | out, err = process_communicate_or_kill(sp) | |
3103 | out = out.decode().strip() | |
3104 | if re.match('[0-9a-f]+', out): | |
3105 | self._write_string('[debug] Git HEAD: %s\n' % out) | |
3106 | except Exception: | |
3107 | try: | |
3108 | sys.exc_clear() | |
3109 | except Exception: | |
3110 | pass | |
3111 | ||
3112 | def python_implementation(): | |
3113 | impl_name = platform.python_implementation() | |
3114 | if impl_name == 'PyPy' and hasattr(sys, 'pypy_version_info'): | |
3115 | return impl_name + ' version %d.%d.%d' % sys.pypy_version_info[:3] | |
3116 | return impl_name | |
3117 | ||
3118 | self._write_string('[debug] Python version %s (%s %s) - %s\n' % ( | |
3119 | platform.python_version(), | |
3120 | python_implementation(), | |
3121 | platform.architecture()[0], | |
3122 | platform_name())) | |
3123 | ||
3124 | exe_versions = FFmpegPostProcessor.get_versions(self) | |
3125 | exe_versions['rtmpdump'] = rtmpdump_version() | |
3126 | exe_versions['phantomjs'] = PhantomJSwrapper._version() | |
3127 | exe_str = ', '.join( | |
3128 | '%s %s' % (exe, v) | |
3129 | for exe, v in sorted(exe_versions.items()) | |
3130 | if v | |
3131 | ) | |
3132 | if not exe_str: | |
3133 | exe_str = 'none' | |
3134 | self._write_string('[debug] exe versions: %s\n' % exe_str) | |
3135 | ||
3136 | proxy_map = {} | |
3137 | for handler in self._opener.handlers: | |
3138 | if hasattr(handler, 'proxies'): | |
3139 | proxy_map.update(handler.proxies) | |
3140 | self._write_string('[debug] Proxy map: ' + compat_str(proxy_map) + '\n') | |
3141 | ||
3142 | if self.params.get('call_home', False): | |
3143 | ipaddr = self.urlopen('https://yt-dl.org/ip').read().decode('utf-8') | |
3144 | self._write_string('[debug] Public IP address: %s\n' % ipaddr) | |
3145 | return | |
3146 | latest_version = self.urlopen( | |
3147 | 'https://yt-dl.org/latest/version').read().decode('utf-8') | |
3148 | if version_tuple(latest_version) > version_tuple(__version__): | |
3149 | self.report_warning( | |
3150 | 'You are using an outdated version (newest version: %s)! ' | |
3151 | 'See https://yt-dl.org/update if you need help updating.' % | |
3152 | latest_version) | |
3153 | ||
3154 | def _setup_opener(self): | |
3155 | timeout_val = self.params.get('socket_timeout') | |
3156 | self._socket_timeout = 600 if timeout_val is None else float(timeout_val) | |
3157 | ||
3158 | opts_cookiefile = self.params.get('cookiefile') | |
3159 | opts_proxy = self.params.get('proxy') | |
3160 | ||
3161 | if opts_cookiefile is None: | |
3162 | self.cookiejar = compat_cookiejar.CookieJar() | |
3163 | else: | |
3164 | opts_cookiefile = expand_path(opts_cookiefile) | |
3165 | self.cookiejar = YoutubeDLCookieJar(opts_cookiefile) | |
3166 | if os.access(opts_cookiefile, os.R_OK): | |
3167 | self.cookiejar.load(ignore_discard=True, ignore_expires=True) | |
3168 | ||
3169 | cookie_processor = YoutubeDLCookieProcessor(self.cookiejar) | |
3170 | if opts_proxy is not None: | |
3171 | if opts_proxy == '': | |
3172 | proxies = {} | |
3173 | else: | |
3174 | proxies = {'http': opts_proxy, 'https': opts_proxy} | |
3175 | else: | |
3176 | proxies = compat_urllib_request.getproxies() | |
3177 | # Set HTTPS proxy to HTTP one if given (https://github.com/ytdl-org/youtube-dl/issues/805) | |
3178 | if 'http' in proxies and 'https' not in proxies: | |
3179 | proxies['https'] = proxies['http'] | |
3180 | proxy_handler = PerRequestProxyHandler(proxies) | |
3181 | ||
3182 | debuglevel = 1 if self.params.get('debug_printtraffic') else 0 | |
3183 | https_handler = make_HTTPS_handler(self.params, debuglevel=debuglevel) | |
3184 | ydlh = YoutubeDLHandler(self.params, debuglevel=debuglevel) | |
3185 | redirect_handler = YoutubeDLRedirectHandler() | |
3186 | data_handler = compat_urllib_request_DataHandler() | |
3187 | ||
3188 | # When passing our own FileHandler instance, build_opener won't add the | |
3189 | # default FileHandler and allows us to disable the file protocol, which | |
3190 | # can be used for malicious purposes (see | |
3191 | # https://github.com/ytdl-org/youtube-dl/issues/8227) | |
3192 | file_handler = compat_urllib_request.FileHandler() | |
3193 | ||
3194 | def file_open(*args, **kwargs): | |
3195 | raise compat_urllib_error.URLError('file:// scheme is explicitly disabled in yt-dlp for security reasons') | |
3196 | file_handler.file_open = file_open | |
3197 | ||
3198 | opener = compat_urllib_request.build_opener( | |
3199 | proxy_handler, https_handler, cookie_processor, ydlh, redirect_handler, data_handler, file_handler) | |
3200 | ||
3201 | # Delete the default user-agent header, which would otherwise apply in | |
3202 | # cases where our custom HTTP handler doesn't come into play | |
3203 | # (See https://github.com/ytdl-org/youtube-dl/issues/1309 for details) | |
3204 | opener.addheaders = [] | |
3205 | self._opener = opener | |
3206 | ||
3207 | def encode(self, s): | |
3208 | if isinstance(s, bytes): | |
3209 | return s # Already encoded | |
3210 | ||
3211 | try: | |
3212 | return s.encode(self.get_encoding()) | |
3213 | except UnicodeEncodeError as err: | |
3214 | err.reason = err.reason + '. Check your system encoding configuration or use the --encoding option.' | |
3215 | raise | |
3216 | ||
3217 | def get_encoding(self): | |
3218 | encoding = self.params.get('encoding') | |
3219 | if encoding is None: | |
3220 | encoding = preferredencoding() | |
3221 | return encoding | |
3222 | ||
3223 | def _write_thumbnails(self, info_dict, filename): # return the extensions | |
3224 | write_all = self.params.get('write_all_thumbnails', False) | |
3225 | thumbnails = [] | |
3226 | if write_all or self.params.get('writethumbnail', False): | |
3227 | thumbnails = info_dict.get('thumbnails') or [] | |
3228 | multiple = write_all and len(thumbnails) > 1 | |
3229 | ||
3230 | ret = [] | |
3231 | for t in thumbnails[::1 if write_all else -1]: | |
3232 | thumb_ext = determine_ext(t['url'], 'jpg') | |
3233 | suffix = '%s.' % t['id'] if multiple else '' | |
3234 | thumb_display_id = '%s ' % t['id'] if multiple else '' | |
3235 | thumb_filename = replace_extension(filename, suffix + thumb_ext, info_dict.get('ext')) | |
3236 | ||
3237 | if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(thumb_filename)): | |
3238 | ret.append(suffix + thumb_ext) | |
3239 | self.to_screen('[%s] %s: Thumbnail %sis already present' % | |
3240 | (info_dict['extractor'], info_dict['id'], thumb_display_id)) | |
3241 | else: | |
3242 | self.to_screen('[%s] %s: Downloading thumbnail %s ...' % | |
3243 | (info_dict['extractor'], info_dict['id'], thumb_display_id)) | |
3244 | try: | |
3245 | uf = self.urlopen(t['url']) | |
3246 | with open(encodeFilename(thumb_filename), 'wb') as thumbf: | |
3247 | shutil.copyfileobj(uf, thumbf) | |
3248 | ret.append(suffix + thumb_ext) | |
3249 | self.to_screen('[%s] %s: Writing thumbnail %sto: %s' % | |
3250 | (info_dict['extractor'], info_dict['id'], thumb_display_id, thumb_filename)) | |
3251 | t['filepath'] = thumb_filename | |
3252 | except network_exceptions as err: | |
3253 | self.report_warning('Unable to download thumbnail "%s": %s' % | |
3254 | (t['url'], error_to_compat_str(err))) | |
3255 | if ret and not write_all: | |
3256 | break | |
3257 | return ret |