]>
Commit | Line | Data |
---|---|---|
8222d8de | 1 | #!/usr/bin/env python |
dcdb292f | 2 | # coding: utf-8 |
8222d8de | 3 | |
6febd1c1 | 4 | from __future__ import absolute_import, unicode_literals |
8222d8de | 5 | |
26e63931 | 6 | import collections |
31bd3925 | 7 | import contextlib |
317f7ab6 | 8 | import copy |
9d2ecdbc | 9 | import datetime |
c1c9a79c | 10 | import errno |
31bd3925 | 11 | import fileinput |
8222d8de | 12 | import io |
b82f815f | 13 | import itertools |
8694c600 | 14 | import json |
62fec3b2 | 15 | import locale |
083c9df9 | 16 | import operator |
8222d8de | 17 | import os |
dca08720 | 18 | import platform |
8222d8de JMF |
19 | import re |
20 | import shutil | |
dca08720 | 21 | import subprocess |
8222d8de JMF |
22 | import socket |
23 | import sys | |
24 | import time | |
67134eab | 25 | import tokenize |
8222d8de | 26 | import traceback |
75822ca7 | 27 | import random |
8222d8de | 28 | |
961ea474 S |
29 | from string import ascii_letters |
30 | ||
8c25f81b | 31 | from .compat import ( |
82d8a8b6 | 32 | compat_basestring, |
dca08720 | 33 | compat_cookiejar, |
003c69a8 | 34 | compat_get_terminal_size, |
ce02ed60 | 35 | compat_http_client, |
4f026faf | 36 | compat_kwargs, |
d0d9ade4 | 37 | compat_numeric_types, |
e9c0cdd3 | 38 | compat_os_name, |
ce02ed60 | 39 | compat_str, |
67134eab | 40 | compat_tokenize_tokenize, |
ce02ed60 PH |
41 | compat_urllib_error, |
42 | compat_urllib_request, | |
8b172c2e | 43 | compat_urllib_request_DataHandler, |
8c25f81b PH |
44 | ) |
45 | from .utils import ( | |
eedb7ba5 S |
46 | age_restricted, |
47 | args_to_str, | |
ce02ed60 PH |
48 | ContentTooShortError, |
49 | date_from_str, | |
50 | DateRange, | |
acd69589 | 51 | DEFAULT_OUTTMPL, |
ce02ed60 | 52 | determine_ext, |
b5559424 | 53 | determine_protocol, |
732044af | 54 | DOT_DESKTOP_LINK_TEMPLATE, |
55 | DOT_URL_LINK_TEMPLATE, | |
56 | DOT_WEBLOC_LINK_TEMPLATE, | |
ce02ed60 | 57 | DownloadError, |
c0384f22 | 58 | encode_compat_str, |
ce02ed60 | 59 | encodeFilename, |
9b9c5355 | 60 | error_to_compat_str, |
8b0d7497 | 61 | ExistingVideoReached, |
590bc6f6 | 62 | expand_path, |
ce02ed60 | 63 | ExtractorError, |
02dbf93f | 64 | format_bytes, |
76d321f6 | 65 | format_field, |
525ef922 | 66 | formatSeconds, |
773f291d | 67 | GeoRestrictedError, |
c9969434 | 68 | int_or_none, |
732044af | 69 | iri_to_uri, |
773f291d | 70 | ISO3166Utils, |
ce02ed60 | 71 | locked_file, |
0202b52a | 72 | make_dir, |
dca08720 | 73 | make_HTTPS_handler, |
ce02ed60 | 74 | MaxDownloadsReached, |
cd6fc19e | 75 | orderedSet, |
b7ab0590 | 76 | PagedList, |
083c9df9 | 77 | parse_filesize, |
91410c9b | 78 | PerRequestProxyHandler, |
dca08720 | 79 | platform_name, |
eedb7ba5 | 80 | PostProcessingError, |
ce02ed60 | 81 | preferredencoding, |
eedb7ba5 | 82 | prepend_extension, |
51fb4995 | 83 | register_socks_protocols, |
cfb56d1a | 84 | render_table, |
eedb7ba5 | 85 | replace_extension, |
8b0d7497 | 86 | RejectedVideoReached, |
ce02ed60 PH |
87 | SameFileError, |
88 | sanitize_filename, | |
1bb5c511 | 89 | sanitize_path, |
dcf77cf1 | 90 | sanitize_url, |
67dda517 | 91 | sanitized_Request, |
e5660ee6 | 92 | std_headers, |
1211bb6d | 93 | str_or_none, |
ce02ed60 | 94 | subtitles_filename, |
732044af | 95 | to_high_limit_path, |
ce02ed60 | 96 | UnavailableVideoError, |
29eb5174 | 97 | url_basename, |
58b1f00d | 98 | version_tuple, |
ce02ed60 PH |
99 | write_json_file, |
100 | write_string, | |
1bab3437 | 101 | YoutubeDLCookieJar, |
6a3f4c3f | 102 | YoutubeDLCookieProcessor, |
dca08720 | 103 | YoutubeDLHandler, |
fca6dba8 | 104 | YoutubeDLRedirectHandler, |
f5b1bca9 | 105 | process_communicate_or_kill, |
ce02ed60 | 106 | ) |
a0e07d31 | 107 | from .cache import Cache |
f74980cb | 108 | from .extractor import get_info_extractor, gen_extractor_classes, _LAZY_LOADER, _PLUGIN_CLASSES |
4c54b89e | 109 | from .extractor.openload import PhantomJSwrapper |
3bc2ddcc | 110 | from .downloader import get_suitable_downloader |
4c83c967 | 111 | from .downloader.rtmp import rtmpdump_version |
4f026faf | 112 | from .postprocessor import ( |
f17f8651 | 113 | FFmpegFixupM3u8PP, |
62cd676c | 114 | FFmpegFixupM4aPP, |
6271f1ca | 115 | FFmpegFixupStretchedPP, |
4f026faf PH |
116 | FFmpegMergerPP, |
117 | FFmpegPostProcessor, | |
0202b52a | 118 | # FFmpegSubtitlesConvertorPP, |
4f026faf | 119 | get_postprocessor, |
0202b52a | 120 | MoveFilesAfterDownloadPP, |
4f026faf | 121 | ) |
dca08720 | 122 | from .version import __version__ |
8222d8de | 123 | |
e9c0cdd3 YCH |
124 | if compat_os_name == 'nt': |
125 | import ctypes | |
126 | ||
2459b6e1 | 127 | |
8222d8de JMF |
128 | class YoutubeDL(object): |
129 | """YoutubeDL class. | |
130 | ||
131 | YoutubeDL objects are the ones responsible of downloading the | |
132 | actual video file and writing it to disk if the user has requested | |
133 | it, among some other tasks. In most cases there should be one per | |
134 | program. As, given a video URL, the downloader doesn't know how to | |
135 | extract all the needed information, task that InfoExtractors do, it | |
136 | has to pass the URL to one of them. | |
137 | ||
138 | For this, YoutubeDL objects have a method that allows | |
139 | InfoExtractors to be registered in a given order. When it is passed | |
140 | a URL, the YoutubeDL object handles it to the first InfoExtractor it | |
141 | finds that reports being able to handle it. The InfoExtractor extracts | |
142 | all the information about the video or videos the URL refers to, and | |
143 | YoutubeDL process the extracted information, possibly using a File | |
144 | Downloader to download the video. | |
145 | ||
146 | YoutubeDL objects accept a lot of parameters. In order not to saturate | |
147 | the object constructor with arguments, it receives a dictionary of | |
148 | options instead. These options are available through the params | |
149 | attribute for the InfoExtractors to use. The YoutubeDL also | |
150 | registers itself as the downloader in charge for the InfoExtractors | |
151 | that are added to it, so this is a "mutual registration". | |
152 | ||
153 | Available options: | |
154 | ||
155 | username: Username for authentication purposes. | |
156 | password: Password for authentication purposes. | |
180940e0 | 157 | videopassword: Password for accessing a video. |
1da50aa3 S |
158 | ap_mso: Adobe Pass multiple-system operator identifier. |
159 | ap_username: Multiple-system operator account username. | |
160 | ap_password: Multiple-system operator account password. | |
8222d8de JMF |
161 | usenetrc: Use netrc for authentication instead. |
162 | verbose: Print additional info to stdout. | |
163 | quiet: Do not print messages to stdout. | |
ad8915b7 | 164 | no_warnings: Do not print out anything for warnings. |
8222d8de JMF |
165 | forceurl: Force printing final URL. |
166 | forcetitle: Force printing title. | |
167 | forceid: Force printing ID. | |
168 | forcethumbnail: Force printing thumbnail URL. | |
169 | forcedescription: Force printing description. | |
170 | forcefilename: Force printing final filename. | |
525ef922 | 171 | forceduration: Force printing duration. |
8694c600 | 172 | forcejson: Force printing info_dict as JSON. |
63e0be34 PH |
173 | dump_single_json: Force printing the info_dict of the whole playlist |
174 | (or video) as a single JSON line. | |
2d30509f | 175 | force_write_download_archive: Force writing download archive regardless of |
176 | 'skip_download' or 'simulate'. | |
8222d8de | 177 | simulate: Do not download the video files. |
eb8a4433 | 178 | format: Video format code. see "FORMAT SELECTION" for more details. |
179 | format_sort: How to sort the video formats. see "Sorting Formats" for more details. | |
180 | format_sort_force: Force the given format_sort. see "Sorting Formats" for more details. | |
909d24dd | 181 | allow_multiple_video_streams: Allow multiple video streams to be merged into a single file |
182 | allow_multiple_audio_streams: Allow multiple audio streams to be merged into a single file | |
8222d8de | 183 | outtmpl: Template for output names. |
a820dc72 RA |
184 | outtmpl_na_placeholder: Placeholder for unavailable meta fields. |
185 | restrictfilenames: Do not allow "&" and spaces in file names | |
186 | trim_file_name: Limit length of filename (extension excluded) | |
187 | ignoreerrors: Do not stop on download errors | |
188 | (Default True when running youtube-dlc, | |
189 | but False when directly accessing YoutubeDL class) | |
d22dec74 | 190 | force_generic_extractor: Force downloader to use the generic extractor |
0c3d0f51 | 191 | overwrites: Overwrite all video and metadata files if True, |
192 | overwrite only non-video files if None | |
193 | and don't overwrite any file if False | |
8222d8de JMF |
194 | playliststart: Playlist item to start at. |
195 | playlistend: Playlist item to end at. | |
c14e88f0 | 196 | playlist_items: Specific indices of playlist to download. |
ff815fe6 | 197 | playlistreverse: Download playlist items in reverse order. |
75822ca7 | 198 | playlistrandom: Download playlist items in random order. |
8222d8de JMF |
199 | matchtitle: Download only matching titles. |
200 | rejecttitle: Reject downloads for matching titles. | |
8bf9319e | 201 | logger: Log messages to a logging.Logger instance. |
8222d8de JMF |
202 | logtostderr: Log messages to stderr instead of stdout. |
203 | writedescription: Write the video description to a .description file | |
204 | writeinfojson: Write the video description to a .info.json file | |
1fb07d10 | 205 | writeannotations: Write the video annotations to a .annotations.xml file |
8222d8de | 206 | writethumbnail: Write the thumbnail image to a file |
ec82d85a | 207 | write_all_thumbnails: Write all thumbnail formats to files |
732044af | 208 | writelink: Write an internet shortcut file, depending on the |
209 | current platform (.url/.webloc/.desktop) | |
210 | writeurllink: Write a Windows internet shortcut file (.url) | |
211 | writewebloclink: Write a macOS internet shortcut file (.webloc) | |
212 | writedesktoplink: Write a Linux internet shortcut file (.desktop) | |
8222d8de | 213 | writesubtitles: Write the video subtitles to a file |
741dd8ea | 214 | writeautomaticsub: Write the automatically generated subtitles to a file |
8222d8de | 215 | allsubtitles: Downloads all the subtitles of the video |
0b7f3118 | 216 | (requires writesubtitles or writeautomaticsub) |
8222d8de | 217 | listsubtitles: Lists all available subtitles for the video |
a504ced0 | 218 | subtitlesformat: The format code for subtitles |
aa6a10c4 | 219 | subtitleslangs: List of languages of the subtitles to download |
8222d8de JMF |
220 | keepvideo: Keep the video file after post-processing |
221 | daterange: A DateRange object, download only if the upload_date is in the range. | |
222 | skip_download: Skip the actual download of the video file | |
c35f9e72 | 223 | cachedir: Location of the cache files in the filesystem. |
a0e07d31 | 224 | False to disable filesystem cache. |
47192f92 | 225 | noplaylist: Download single video instead of a playlist if in doubt. |
8dbe9899 PH |
226 | age_limit: An integer representing the user's age in years. |
227 | Unsuitable videos for the given age are skipped. | |
5fe18bdb PH |
228 | min_views: An integer representing the minimum view count the video |
229 | must have in order to not be skipped. | |
230 | Videos without view count information are always | |
231 | downloaded. None for no limit. | |
232 | max_views: An integer representing the maximum view count. | |
233 | Videos that are more popular than that are not | |
234 | downloaded. | |
235 | Videos without view count information are always | |
236 | downloaded. None for no limit. | |
237 | download_archive: File name of a file where all downloads are recorded. | |
c1c9a79c PH |
238 | Videos already present in the file are not downloaded |
239 | again. | |
8a51f564 | 240 | break_on_existing: Stop the download process after attempting to download a |
241 | file that is in the archive. | |
242 | break_on_reject: Stop the download process when encountering a video that | |
243 | has been filtered out. | |
244 | cookiefile: File name where cookies should be read from and dumped to | |
a1ee09e8 | 245 | nocheckcertificate:Do not verify SSL certificates |
7e8c0af0 PH |
246 | prefer_insecure: Use HTTP instead of HTTPS to retrieve information. |
247 | At the moment, this is only supported by YouTube. | |
a1ee09e8 | 248 | proxy: URL of the proxy server to use |
38cce791 | 249 | geo_verification_proxy: URL of the proxy to use for IP address verification |
504f20dd | 250 | on geo-restricted sites. |
e344693b | 251 | socket_timeout: Time to wait for unresponsive hosts, in seconds |
0783b09b PH |
252 | bidi_workaround: Work around buggy terminals without bidirectional text |
253 | support, using fridibi | |
a0ddb8a2 | 254 | debug_printtraffic:Print out sent and received HTTP traffic |
7b0817e8 | 255 | include_ads: Download ads as well |
04b4d394 PH |
256 | default_search: Prepend this string if an input url is not valid. |
257 | 'auto' for elaborate guessing | |
62fec3b2 | 258 | encoding: Use this encoding instead of the system-specified. |
e8ee972c | 259 | extract_flat: Do not resolve URLs, return the immediate result. |
057a5206 PH |
260 | Pass in 'in_playlist' to only show this behavior for |
261 | playlist items. | |
4f026faf | 262 | postprocessors: A list of dictionaries, each with an entry |
71b640cc | 263 | * key: The name of the postprocessor. See |
cefecac1 | 264 | youtube_dlc/postprocessor/__init__.py for a list. |
0202b52a | 265 | * _after_move: Optional. If True, run this post_processor |
266 | after 'MoveFilesAfterDownload' | |
4f026faf PH |
267 | as well as any further keyword arguments for the |
268 | postprocessor. | |
ab8e5e51 AM |
269 | post_hooks: A list of functions that get called as the final step |
270 | for each video file, after all postprocessors have been | |
271 | called. The filename will be passed as the only argument. | |
71b640cc PH |
272 | progress_hooks: A list of functions that get called on download |
273 | progress, with a dictionary with the entries | |
5cda4eda | 274 | * status: One of "downloading", "error", or "finished". |
ee69b99a | 275 | Check this first and ignore unknown values. |
71b640cc | 276 | |
5cda4eda | 277 | If status is one of "downloading", or "finished", the |
ee69b99a PH |
278 | following properties may also be present: |
279 | * filename: The final filename (always present) | |
5cda4eda | 280 | * tmpfilename: The filename we're currently writing to |
71b640cc PH |
281 | * downloaded_bytes: Bytes on disk |
282 | * total_bytes: Size of the whole file, None if unknown | |
5cda4eda PH |
283 | * total_bytes_estimate: Guess of the eventual file size, |
284 | None if unavailable. | |
285 | * elapsed: The number of seconds since download started. | |
71b640cc PH |
286 | * eta: The estimated time in seconds, None if unknown |
287 | * speed: The download speed in bytes/second, None if | |
288 | unknown | |
5cda4eda PH |
289 | * fragment_index: The counter of the currently |
290 | downloaded video fragment. | |
291 | * fragment_count: The number of fragments (= individual | |
292 | files that will be merged) | |
71b640cc PH |
293 | |
294 | Progress hooks are guaranteed to be called at least once | |
295 | (with status "finished") if the download is successful. | |
45598f15 | 296 | merge_output_format: Extension to use when merging formats. |
6271f1ca PH |
297 | fixup: Automatically correct known faults of the file. |
298 | One of: | |
299 | - "never": do nothing | |
300 | - "warn": only emit a warning | |
301 | - "detect_or_warn": check whether we can do anything | |
62cd676c | 302 | about it, warn otherwise (default) |
504f20dd | 303 | source_address: Client-side IP address to bind to. |
6ec6cb4e | 304 | call_home: Boolean, true iff we are allowed to contact the |
cefecac1 | 305 | youtube-dlc servers for debugging. |
7aa589a5 S |
306 | sleep_interval: Number of seconds to sleep before each download when |
307 | used alone or a lower bound of a range for randomized | |
308 | sleep before each download (minimum possible number | |
309 | of seconds to sleep) when used along with | |
310 | max_sleep_interval. | |
311 | max_sleep_interval:Upper bound of a range for randomized sleep before each | |
312 | download (maximum possible number of seconds to sleep). | |
313 | Must only be used along with sleep_interval. | |
314 | Actual sleep time will be a random float from range | |
315 | [sleep_interval; max_sleep_interval]. | |
cfb56d1a PH |
316 | listformats: Print an overview of available video formats and exit. |
317 | list_thumbnails: Print a table of all thumbnails and exit. | |
347de493 PH |
318 | match_filter: A function that gets called with the info_dict of |
319 | every video. | |
320 | If it returns a message, the video is ignored. | |
321 | If it returns None, the video is downloaded. | |
322 | match_filter_func in utils.py is one example for this. | |
7e5db8c9 | 323 | no_color: Do not emit color codes in output. |
0a840f58 | 324 | geo_bypass: Bypass geographic restriction via faking X-Forwarded-For |
504f20dd | 325 | HTTP header |
0a840f58 | 326 | geo_bypass_country: |
773f291d S |
327 | Two-letter ISO 3166-2 country code that will be used for |
328 | explicit geographic restriction bypassing via faking | |
504f20dd | 329 | X-Forwarded-For HTTP header |
5f95927a S |
330 | geo_bypass_ip_block: |
331 | IP range in CIDR notation that will be used similarly to | |
504f20dd | 332 | geo_bypass_country |
71b640cc | 333 | |
85729c51 PH |
334 | The following options determine which downloader is picked: |
335 | external_downloader: Executable of the external downloader to call. | |
336 | None or unset for standard (built-in) downloader. | |
bf09af3a S |
337 | hls_prefer_native: Use the native HLS downloader instead of ffmpeg/avconv |
338 | if True, otherwise use ffmpeg/avconv if False, otherwise | |
339 | use downloader suggested by extractor if None. | |
fe7e0c98 | 340 | |
8222d8de | 341 | The following parameters are not used by YoutubeDL itself, they are used by |
cefecac1 | 342 | the downloader (see youtube_dlc/downloader/common.py): |
8222d8de | 343 | nopart, updatetime, buffersize, ratelimit, min_filesize, max_filesize, test, |
881e6a1f | 344 | noresizebuffer, retries, continuedl, noprogress, consoletitle, |
b54d4a5c S |
345 | xattr_set_filesize, external_downloader_args, hls_use_mpegts, |
346 | http_chunk_size. | |
76b1bd67 JMF |
347 | |
348 | The following options are used by the post processors: | |
d4a24f40 | 349 | prefer_ffmpeg: If False, use avconv instead of ffmpeg if both are available, |
e4172ac9 | 350 | otherwise prefer ffmpeg. (avconv support is deprecated) |
c0b7d117 S |
351 | ffmpeg_location: Location of the ffmpeg/avconv binary; either the path |
352 | to the binary or its containing directory. | |
43820c03 | 353 | postprocessor_args: A dictionary of postprocessor/executable keys (in lower case) |
354 | and a list of additional command-line arguments for the | |
355 | postprocessor/executable. The dict can also have "PP+EXE" keys | |
356 | which are used when the given exe is used by the given PP. | |
357 | Use 'default' as the name for arguments to passed to all PP | |
3600fd59 S |
358 | The following options are used by the Youtube extractor: |
359 | youtube_include_dash_manifest: If True (default), DASH manifests and related | |
360 | data will be downloaded and processed by extractor. | |
361 | You can reduce network I/O by disabling it if you don't | |
362 | care about DASH. | |
8222d8de JMF |
363 | """ |
364 | ||
c9969434 S |
365 | _NUMERIC_FIELDS = set(( |
366 | 'width', 'height', 'tbr', 'abr', 'asr', 'vbr', 'fps', 'filesize', 'filesize_approx', | |
367 | 'timestamp', 'upload_year', 'upload_month', 'upload_day', | |
368 | 'duration', 'view_count', 'like_count', 'dislike_count', 'repost_count', | |
369 | 'average_rating', 'comment_count', 'age_limit', | |
370 | 'start_time', 'end_time', | |
371 | 'chapter_number', 'season_number', 'episode_number', | |
372 | 'track_number', 'disc_number', 'release_year', | |
373 | 'playlist_index', | |
374 | )) | |
375 | ||
8222d8de JMF |
376 | params = None |
377 | _ies = [] | |
5bfa4862 | 378 | _pps = {'beforedl': [], 'aftermove': [], 'normal': []} |
0202b52a | 379 | __prepare_filename_warned = False |
8222d8de JMF |
380 | _download_retcode = None |
381 | _num_downloads = None | |
30a074c2 | 382 | _playlist_level = 0 |
383 | _playlist_urls = set() | |
8222d8de JMF |
384 | _screen_file = None |
385 | ||
3511266b | 386 | def __init__(self, params=None, auto_init=True): |
8222d8de | 387 | """Create a FileDownloader object with the given options.""" |
e9f9a10f JMF |
388 | if params is None: |
389 | params = {} | |
8222d8de | 390 | self._ies = [] |
56c73665 | 391 | self._ies_instances = {} |
5bfa4862 | 392 | self._pps = {'beforedl': [], 'aftermove': [], 'normal': []} |
0202b52a | 393 | self.__prepare_filename_warned = False |
ab8e5e51 | 394 | self._post_hooks = [] |
933605d7 | 395 | self._progress_hooks = [] |
8222d8de JMF |
396 | self._download_retcode = 0 |
397 | self._num_downloads = 0 | |
398 | self._screen_file = [sys.stdout, sys.stderr][params.get('logtostderr', False)] | |
0783b09b | 399 | self._err_file = sys.stderr |
4abf617b S |
400 | self.params = { |
401 | # Default parameters | |
402 | 'nocheckcertificate': False, | |
403 | } | |
404 | self.params.update(params) | |
a0e07d31 | 405 | self.cache = Cache(self) |
a45e8619 | 406 | self.archive = set() |
ecdec191 JB |
407 | |
408 | """Preload the archive, if any is specified""" | |
409 | def preload_download_archive(self): | |
410 | fn = self.params.get('download_archive') | |
411 | if fn is None: | |
412 | return False | |
413 | try: | |
414 | with locked_file(fn, 'r', encoding='utf-8') as archive_file: | |
415 | for line in archive_file: | |
a45e8619 | 416 | self.archive.add(line.strip()) |
ecdec191 JB |
417 | except IOError as ioe: |
418 | if ioe.errno != errno.ENOENT: | |
419 | raise | |
1d74d8d9 | 420 | return False |
ecdec191 | 421 | return True |
34308b30 | 422 | |
be5df5ee S |
423 | def check_deprecated(param, option, suggestion): |
424 | if self.params.get(param) is not None: | |
425 | self.report_warning( | |
426 | '%s is deprecated. Use %s instead.' % (option, suggestion)) | |
427 | return True | |
428 | return False | |
429 | ||
1de7ea76 JB |
430 | if self.params.get('verbose'): |
431 | self.to_stdout('[debug] Loading archive file %r' % self.params.get('download_archive')) | |
432 | ||
ecdec191 JB |
433 | preload_download_archive(self) |
434 | ||
be5df5ee | 435 | if check_deprecated('cn_verification_proxy', '--cn-verification-proxy', '--geo-verification-proxy'): |
38cce791 YCH |
436 | if self.params.get('geo_verification_proxy') is None: |
437 | self.params['geo_verification_proxy'] = self.params['cn_verification_proxy'] | |
438 | ||
be5df5ee S |
439 | check_deprecated('autonumber_size', '--autonumber-size', 'output template with %(autonumber)0Nd, where N in the number of digits') |
440 | check_deprecated('autonumber', '--auto-number', '-o "%(autonumber)s-%(title)s.%(ext)s"') | |
441 | check_deprecated('usetitle', '--title', '-o "%(title)s-%(id)s.%(ext)s"') | |
442 | ||
0783b09b | 443 | if params.get('bidi_workaround', False): |
1c088fa8 PH |
444 | try: |
445 | import pty | |
446 | master, slave = pty.openpty() | |
003c69a8 | 447 | width = compat_get_terminal_size().columns |
1c088fa8 PH |
448 | if width is None: |
449 | width_args = [] | |
450 | else: | |
451 | width_args = ['-w', str(width)] | |
5d681e96 | 452 | sp_kwargs = dict( |
1c088fa8 PH |
453 | stdin=subprocess.PIPE, |
454 | stdout=slave, | |
455 | stderr=self._err_file) | |
5d681e96 PH |
456 | try: |
457 | self._output_process = subprocess.Popen( | |
458 | ['bidiv'] + width_args, **sp_kwargs | |
459 | ) | |
460 | except OSError: | |
5d681e96 PH |
461 | self._output_process = subprocess.Popen( |
462 | ['fribidi', '-c', 'UTF-8'] + width_args, **sp_kwargs) | |
463 | self._output_channel = os.fdopen(master, 'rb') | |
1c088fa8 | 464 | except OSError as ose: |
66e7ace1 | 465 | if ose.errno == errno.ENOENT: |
6febd1c1 | 466 | 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.') |
1c088fa8 PH |
467 | else: |
468 | raise | |
0783b09b | 469 | |
3089bc74 S |
470 | if (sys.platform != 'win32' |
471 | and sys.getfilesystemencoding() in ['ascii', 'ANSI_X3.4-1968'] | |
472 | and not params.get('restrictfilenames', False)): | |
e9137224 | 473 | # Unicode filesystem API will throw errors (#1474, #13027) |
34308b30 | 474 | self.report_warning( |
6febd1c1 | 475 | 'Assuming --restrict-filenames since file system encoding ' |
1b725173 | 476 | 'cannot encode all characters. ' |
6febd1c1 | 477 | 'Set the LC_ALL environment variable to fix this.') |
4a98cdbf | 478 | self.params['restrictfilenames'] = True |
34308b30 | 479 | |
486dd09e PH |
480 | if isinstance(params.get('outtmpl'), bytes): |
481 | self.report_warning( | |
482 | 'Parameter outtmpl is bytes, but should be a unicode string. ' | |
483 | 'Put from __future__ import unicode_literals at the top of your code file or consider switching to Python 3.x.') | |
484 | ||
dca08720 PH |
485 | self._setup_opener() |
486 | ||
3511266b PH |
487 | if auto_init: |
488 | self.print_debug_header() | |
489 | self.add_default_info_extractors() | |
490 | ||
4f026faf PH |
491 | for pp_def_raw in self.params.get('postprocessors', []): |
492 | pp_class = get_postprocessor(pp_def_raw['key']) | |
493 | pp_def = dict(pp_def_raw) | |
494 | del pp_def['key'] | |
5bfa4862 | 495 | if 'when' in pp_def: |
496 | when = pp_def['when'] | |
497 | del pp_def['when'] | |
498 | else: | |
499 | when = 'normal' | |
4f026faf | 500 | pp = pp_class(self, **compat_kwargs(pp_def)) |
5bfa4862 | 501 | self.add_post_processor(pp, when=when) |
4f026faf | 502 | |
ab8e5e51 AM |
503 | for ph in self.params.get('post_hooks', []): |
504 | self.add_post_hook(ph) | |
505 | ||
71b640cc PH |
506 | for ph in self.params.get('progress_hooks', []): |
507 | self.add_progress_hook(ph) | |
508 | ||
51fb4995 YCH |
509 | register_socks_protocols() |
510 | ||
7d4111ed PH |
511 | def warn_if_short_id(self, argv): |
512 | # short YouTube ID starting with dash? | |
513 | idxs = [ | |
514 | i for i, a in enumerate(argv) | |
515 | if re.match(r'^-[0-9A-Za-z_-]{10}$', a)] | |
516 | if idxs: | |
517 | correct_argv = ( | |
cefecac1 | 518 | ['youtube-dlc'] |
3089bc74 S |
519 | + [a for i, a in enumerate(argv) if i not in idxs] |
520 | + ['--'] + [argv[i] for i in idxs] | |
7d4111ed PH |
521 | ) |
522 | self.report_warning( | |
523 | 'Long argument string detected. ' | |
524 | 'Use -- to separate parameters and URLs, like this:\n%s\n' % | |
525 | args_to_str(correct_argv)) | |
526 | ||
8222d8de JMF |
527 | def add_info_extractor(self, ie): |
528 | """Add an InfoExtractor object to the end of the list.""" | |
529 | self._ies.append(ie) | |
e52d7f85 JMF |
530 | if not isinstance(ie, type): |
531 | self._ies_instances[ie.ie_key()] = ie | |
532 | ie.set_downloader(self) | |
8222d8de | 533 | |
56c73665 JMF |
534 | def get_info_extractor(self, ie_key): |
535 | """ | |
536 | Get an instance of an IE with name ie_key, it will try to get one from | |
537 | the _ies list, if there's no instance it will create a new one and add | |
538 | it to the extractor list. | |
539 | """ | |
540 | ie = self._ies_instances.get(ie_key) | |
541 | if ie is None: | |
542 | ie = get_info_extractor(ie_key)() | |
543 | self.add_info_extractor(ie) | |
544 | return ie | |
545 | ||
023fa8c4 JMF |
546 | def add_default_info_extractors(self): |
547 | """ | |
548 | Add the InfoExtractors returned by gen_extractors to the end of the list | |
549 | """ | |
e52d7f85 | 550 | for ie in gen_extractor_classes(): |
023fa8c4 JMF |
551 | self.add_info_extractor(ie) |
552 | ||
5bfa4862 | 553 | def add_post_processor(self, pp, when='normal'): |
8222d8de | 554 | """Add a PostProcessor object to the end of the chain.""" |
5bfa4862 | 555 | self._pps[when].append(pp) |
8222d8de JMF |
556 | pp.set_downloader(self) |
557 | ||
ab8e5e51 AM |
558 | def add_post_hook(self, ph): |
559 | """Add the post hook""" | |
560 | self._post_hooks.append(ph) | |
561 | ||
933605d7 JMF |
562 | def add_progress_hook(self, ph): |
563 | """Add the progress hook (currently only for the file downloader)""" | |
564 | self._progress_hooks.append(ph) | |
8ab470f1 | 565 | |
1c088fa8 | 566 | def _bidi_workaround(self, message): |
5d681e96 | 567 | if not hasattr(self, '_output_channel'): |
1c088fa8 PH |
568 | return message |
569 | ||
5d681e96 | 570 | assert hasattr(self, '_output_process') |
11b85ce6 | 571 | assert isinstance(message, compat_str) |
6febd1c1 PH |
572 | line_count = message.count('\n') + 1 |
573 | self._output_process.stdin.write((message + '\n').encode('utf-8')) | |
5d681e96 | 574 | self._output_process.stdin.flush() |
6febd1c1 | 575 | res = ''.join(self._output_channel.readline().decode('utf-8') |
9e1a5b84 | 576 | for _ in range(line_count)) |
6febd1c1 | 577 | return res[:-len('\n')] |
1c088fa8 | 578 | |
8222d8de | 579 | def to_screen(self, message, skip_eol=False): |
0783b09b PH |
580 | """Print message to stdout if not in quiet mode.""" |
581 | return self.to_stdout(message, skip_eol, check_quiet=True) | |
582 | ||
734f90bb | 583 | def _write_string(self, s, out=None): |
b58ddb32 | 584 | write_string(s, out=out, encoding=self.params.get('encoding')) |
734f90bb | 585 | |
0783b09b | 586 | def to_stdout(self, message, skip_eol=False, check_quiet=False): |
8222d8de | 587 | """Print message to stdout if not in quiet mode.""" |
8bf9319e | 588 | if self.params.get('logger'): |
43afe285 | 589 | self.params['logger'].debug(message) |
0783b09b | 590 | elif not check_quiet or not self.params.get('quiet', False): |
1c088fa8 | 591 | message = self._bidi_workaround(message) |
6febd1c1 | 592 | terminator = ['\n', ''][skip_eol] |
8222d8de | 593 | output = message + terminator |
1c088fa8 | 594 | |
734f90bb | 595 | self._write_string(output, self._screen_file) |
8222d8de JMF |
596 | |
597 | def to_stderr(self, message): | |
598 | """Print message to stderr.""" | |
11b85ce6 | 599 | assert isinstance(message, compat_str) |
8bf9319e | 600 | if self.params.get('logger'): |
43afe285 IB |
601 | self.params['logger'].error(message) |
602 | else: | |
1c088fa8 | 603 | message = self._bidi_workaround(message) |
6febd1c1 | 604 | output = message + '\n' |
734f90bb | 605 | self._write_string(output, self._err_file) |
8222d8de | 606 | |
1e5b9a95 PH |
607 | def to_console_title(self, message): |
608 | if not self.params.get('consoletitle', False): | |
609 | return | |
4bede0d8 C |
610 | if compat_os_name == 'nt': |
611 | if ctypes.windll.kernel32.GetConsoleWindow(): | |
612 | # c_wchar_p() might not be necessary if `message` is | |
613 | # already of type unicode() | |
614 | ctypes.windll.kernel32.SetConsoleTitleW(ctypes.c_wchar_p(message)) | |
1e5b9a95 | 615 | elif 'TERM' in os.environ: |
b46696bd | 616 | self._write_string('\033]0;%s\007' % message, self._screen_file) |
1e5b9a95 | 617 | |
bdde425c PH |
618 | def save_console_title(self): |
619 | if not self.params.get('consoletitle', False): | |
620 | return | |
94c3442e S |
621 | if self.params.get('simulate', False): |
622 | return | |
4bede0d8 | 623 | if compat_os_name != 'nt' and 'TERM' in os.environ: |
efd6c574 | 624 | # Save the title on stack |
734f90bb | 625 | self._write_string('\033[22;0t', self._screen_file) |
bdde425c PH |
626 | |
627 | def restore_console_title(self): | |
628 | if not self.params.get('consoletitle', False): | |
629 | return | |
94c3442e S |
630 | if self.params.get('simulate', False): |
631 | return | |
4bede0d8 | 632 | if compat_os_name != 'nt' and 'TERM' in os.environ: |
efd6c574 | 633 | # Restore the title from stack |
734f90bb | 634 | self._write_string('\033[23;0t', self._screen_file) |
bdde425c PH |
635 | |
636 | def __enter__(self): | |
637 | self.save_console_title() | |
638 | return self | |
639 | ||
640 | def __exit__(self, *args): | |
641 | self.restore_console_title() | |
f89197d7 | 642 | |
dca08720 | 643 | if self.params.get('cookiefile') is not None: |
1bab3437 | 644 | self.cookiejar.save(ignore_discard=True, ignore_expires=True) |
bdde425c | 645 | |
8222d8de JMF |
646 | def trouble(self, message=None, tb=None): |
647 | """Determine action to take when a download problem appears. | |
648 | ||
649 | Depending on if the downloader has been configured to ignore | |
650 | download errors or not, this method may throw an exception or | |
651 | not when errors are found, after printing the message. | |
652 | ||
653 | tb, if given, is additional traceback information. | |
654 | """ | |
655 | if message is not None: | |
656 | self.to_stderr(message) | |
657 | if self.params.get('verbose'): | |
658 | if tb is None: | |
659 | if sys.exc_info()[0]: # if .trouble has been called from an except block | |
6febd1c1 | 660 | tb = '' |
8222d8de | 661 | if hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]: |
6febd1c1 | 662 | tb += ''.join(traceback.format_exception(*sys.exc_info()[1].exc_info)) |
c0384f22 | 663 | tb += encode_compat_str(traceback.format_exc()) |
8222d8de JMF |
664 | else: |
665 | tb_data = traceback.format_list(traceback.extract_stack()) | |
6febd1c1 | 666 | tb = ''.join(tb_data) |
8222d8de JMF |
667 | self.to_stderr(tb) |
668 | if not self.params.get('ignoreerrors', False): | |
669 | if sys.exc_info()[0] and hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]: | |
670 | exc_info = sys.exc_info()[1].exc_info | |
671 | else: | |
672 | exc_info = sys.exc_info() | |
673 | raise DownloadError(message, exc_info) | |
674 | self._download_retcode = 1 | |
675 | ||
676 | def report_warning(self, message): | |
677 | ''' | |
678 | Print the message to stderr, it will be prefixed with 'WARNING:' | |
679 | If stderr is a tty file the 'WARNING:' will be colored | |
680 | ''' | |
6d07ce01 JMF |
681 | if self.params.get('logger') is not None: |
682 | self.params['logger'].warning(message) | |
8222d8de | 683 | else: |
ad8915b7 PH |
684 | if self.params.get('no_warnings'): |
685 | return | |
e9c0cdd3 | 686 | if not self.params.get('no_color') and self._err_file.isatty() and compat_os_name != 'nt': |
6d07ce01 JMF |
687 | _msg_header = '\033[0;33mWARNING:\033[0m' |
688 | else: | |
689 | _msg_header = 'WARNING:' | |
690 | warning_message = '%s %s' % (_msg_header, message) | |
691 | self.to_stderr(warning_message) | |
8222d8de JMF |
692 | |
693 | def report_error(self, message, tb=None): | |
694 | ''' | |
695 | Do the same as trouble, but prefixes the message with 'ERROR:', colored | |
696 | in red if stderr is a tty file. | |
697 | ''' | |
e9c0cdd3 | 698 | if not self.params.get('no_color') and self._err_file.isatty() and compat_os_name != 'nt': |
6febd1c1 | 699 | _msg_header = '\033[0;31mERROR:\033[0m' |
8222d8de | 700 | else: |
6febd1c1 PH |
701 | _msg_header = 'ERROR:' |
702 | error_message = '%s %s' % (_msg_header, message) | |
8222d8de JMF |
703 | self.trouble(error_message, tb) |
704 | ||
8222d8de JMF |
705 | def report_file_already_downloaded(self, file_name): |
706 | """Report file has already been fully downloaded.""" | |
707 | try: | |
6febd1c1 | 708 | self.to_screen('[download] %s has already been downloaded' % file_name) |
ce02ed60 | 709 | except UnicodeEncodeError: |
6febd1c1 | 710 | self.to_screen('[download] The file has already been downloaded') |
8222d8de | 711 | |
0c3d0f51 | 712 | def report_file_delete(self, file_name): |
713 | """Report that existing file will be deleted.""" | |
714 | try: | |
715 | self.to_screen('Deleting already existent file %s' % file_name) | |
716 | except UnicodeEncodeError: | |
717 | self.to_screen('Deleting already existent file') | |
718 | ||
0202b52a | 719 | def prepare_filename(self, info_dict, warn=False): |
8222d8de JMF |
720 | """Generate the output filename.""" |
721 | try: | |
722 | template_dict = dict(info_dict) | |
723 | ||
724 | template_dict['epoch'] = int(time.time()) | |
725 | autonumber_size = self.params.get('autonumber_size') | |
726 | if autonumber_size is None: | |
727 | autonumber_size = 5 | |
89db639d | 728 | template_dict['autonumber'] = self.params.get('autonumber_start', 1) - 1 + self._num_downloads |
17b75c0d PH |
729 | if template_dict.get('resolution') is None: |
730 | if template_dict.get('width') and template_dict.get('height'): | |
731 | template_dict['resolution'] = '%dx%d' % (template_dict['width'], template_dict['height']) | |
732 | elif template_dict.get('height'): | |
805ef3c6 | 733 | template_dict['resolution'] = '%sp' % template_dict['height'] |
17b75c0d | 734 | elif template_dict.get('width'): |
51ce9117 | 735 | template_dict['resolution'] = '%dx?' % template_dict['width'] |
8222d8de | 736 | |
586a91b6 | 737 | sanitize = lambda k, v: sanitize_filename( |
45598aab | 738 | compat_str(v), |
1bb5c511 | 739 | restricted=self.params.get('restrictfilenames'), |
40df485f | 740 | is_id=(k == 'id' or k.endswith('_id'))) |
d0d9ade4 | 741 | template_dict = dict((k, v if isinstance(v, compat_numeric_types) else sanitize(k, v)) |
45598aab | 742 | for k, v in template_dict.items() |
f0e14fdd | 743 | if v is not None and not isinstance(v, (list, tuple, dict))) |
a820dc72 | 744 | template_dict = collections.defaultdict(lambda: self.params.get('outtmpl_na_placeholder', 'NA'), template_dict) |
8222d8de | 745 | |
b3613d36 | 746 | outtmpl = self.params.get('outtmpl', DEFAULT_OUTTMPL) |
d0d9ade4 | 747 | |
89db639d S |
748 | # For fields playlist_index and autonumber convert all occurrences |
749 | # of %(field)s to %(field)0Nd for backward compatibility | |
750 | field_size_compat_map = { | |
751 | 'playlist_index': len(str(template_dict['n_entries'])), | |
752 | 'autonumber': autonumber_size, | |
753 | } | |
754 | FIELD_SIZE_COMPAT_RE = r'(?<!%)%\((?P<field>autonumber|playlist_index)\)s' | |
755 | mobj = re.search(FIELD_SIZE_COMPAT_RE, outtmpl) | |
756 | if mobj: | |
757 | outtmpl = re.sub( | |
758 | FIELD_SIZE_COMPAT_RE, | |
759 | r'%%(\1)0%dd' % field_size_compat_map[mobj.group('field')], | |
760 | outtmpl) | |
761 | ||
d0d9ade4 S |
762 | # Missing numeric fields used together with integer presentation types |
763 | # in format specification will break the argument substitution since | |
a820dc72 RA |
764 | # string NA placeholder is returned for missing fields. We will patch |
765 | # output template for missing fields to meet string presentation type. | |
c9969434 | 766 | for numeric_field in self._NUMERIC_FIELDS: |
d0d9ade4 S |
767 | if numeric_field not in template_dict: |
768 | # As of [1] format syntax is: | |
769 | # %[mapping_key][conversion_flags][minimum_width][.precision][length_modifier]type | |
770 | # 1. https://docs.python.org/2/library/stdtypes.html#string-formatting | |
771 | FORMAT_RE = r'''(?x) | |
772 | (?<!%) | |
773 | % | |
774 | \({0}\) # mapping key | |
775 | (?:[#0\-+ ]+)? # conversion flags (optional) | |
776 | (?:\d+)? # minimum field width (optional) | |
777 | (?:\.\d+)? # precision (optional) | |
778 | [hlL]? # length modifier (optional) | |
779 | [diouxXeEfFgGcrs%] # conversion type | |
780 | ''' | |
781 | outtmpl = re.sub( | |
782 | FORMAT_RE.format(numeric_field), | |
783 | r'%({0})s'.format(numeric_field), outtmpl) | |
784 | ||
15da37c7 S |
785 | # expand_path translates '%%' into '%' and '$$' into '$' |
786 | # correspondingly that is not what we want since we need to keep | |
787 | # '%%' intact for template dict substitution step. Working around | |
788 | # with boundary-alike separator hack. | |
961ea474 | 789 | sep = ''.join([random.choice(ascii_letters) for _ in range(32)]) |
15da37c7 S |
790 | outtmpl = outtmpl.replace('%%', '%{0}%'.format(sep)).replace('$$', '${0}$'.format(sep)) |
791 | ||
792 | # outtmpl should be expand_path'ed before template dict substitution | |
793 | # because meta fields may contain env variables we don't want to | |
794 | # be expanded. For example, for outtmpl "%(title)s.%(ext)s" and | |
795 | # title "Hello $PATH", we don't want `$PATH` to be expanded. | |
796 | filename = expand_path(outtmpl).replace(sep, '') % template_dict | |
797 | ||
bdc3fd2f U |
798 | # https://github.com/blackjack4494/youtube-dlc/issues/85 |
799 | trim_file_name = self.params.get('trim_file_name', False) | |
800 | if trim_file_name: | |
801 | fn_groups = filename.rsplit('.') | |
802 | ext = fn_groups[-1] | |
803 | sub_ext = '' | |
804 | if len(fn_groups) > 2: | |
805 | sub_ext = fn_groups[-2] | |
806 | filename = '.'.join(filter(None, [fn_groups[0][:trim_file_name], sub_ext, ext])) | |
807 | ||
3a0d2f52 S |
808 | # Temporary fix for #4787 |
809 | # 'Treat' all problem characters by passing filename through preferredencoding | |
810 | # to workaround encoding issues with subprocess on python2 @ Windows | |
811 | if sys.version_info < (3, 0) and sys.platform == 'win32': | |
812 | filename = encodeFilename(filename, True).decode(preferredencoding()) | |
0202b52a | 813 | filename = sanitize_path(filename) |
814 | ||
815 | if warn and not self.__prepare_filename_warned: | |
816 | if not self.params.get('paths'): | |
817 | pass | |
818 | elif filename == '-': | |
819 | self.report_warning('--paths is ignored when an outputting to stdout') | |
820 | elif os.path.isabs(filename): | |
821 | self.report_warning('--paths is ignored since an absolute path is given in output template') | |
822 | self.__prepare_filename_warned = True | |
823 | ||
824 | return filename | |
8222d8de | 825 | except ValueError as err: |
6febd1c1 | 826 | self.report_error('Error in output template: ' + str(err) + ' (encoding: ' + repr(preferredencoding()) + ')') |
8222d8de JMF |
827 | return None |
828 | ||
0202b52a | 829 | def prepare_filepath(self, filename, dir_type=''): |
830 | if filename == '-': | |
831 | return filename | |
832 | paths = self.params.get('paths', {}) | |
833 | assert isinstance(paths, dict) | |
834 | homepath = expand_path(paths.get('home', '').strip()) | |
835 | assert isinstance(homepath, compat_str) | |
836 | subdir = expand_path(paths.get(dir_type, '').strip()) if dir_type else '' | |
837 | assert isinstance(subdir, compat_str) | |
838 | return sanitize_path(os.path.join(homepath, subdir, filename)) | |
839 | ||
442c37b7 | 840 | def _match_entry(self, info_dict, incomplete): |
ecdec191 | 841 | """ Returns None if the file should be downloaded """ |
8222d8de | 842 | |
8b0d7497 | 843 | def check_filter(): |
844 | video_title = info_dict.get('title', info_dict.get('id', 'video')) | |
845 | if 'title' in info_dict: | |
846 | # This can happen when we're just evaluating the playlist | |
847 | title = info_dict['title'] | |
848 | matchtitle = self.params.get('matchtitle', False) | |
849 | if matchtitle: | |
850 | if not re.search(matchtitle, title, re.IGNORECASE): | |
851 | return '"' + title + '" title did not match pattern "' + matchtitle + '"' | |
852 | rejecttitle = self.params.get('rejecttitle', False) | |
853 | if rejecttitle: | |
854 | if re.search(rejecttitle, title, re.IGNORECASE): | |
855 | return '"' + title + '" title matched reject pattern "' + rejecttitle + '"' | |
856 | date = info_dict.get('upload_date') | |
857 | if date is not None: | |
858 | dateRange = self.params.get('daterange', DateRange()) | |
859 | if date not in dateRange: | |
860 | return '%s upload date is not in range %s' % (date_from_str(date).isoformat(), dateRange) | |
861 | view_count = info_dict.get('view_count') | |
862 | if view_count is not None: | |
863 | min_views = self.params.get('min_views') | |
864 | if min_views is not None and view_count < min_views: | |
865 | return 'Skipping %s, because it has not reached minimum view count (%d/%d)' % (video_title, view_count, min_views) | |
866 | max_views = self.params.get('max_views') | |
867 | if max_views is not None and view_count > max_views: | |
868 | return 'Skipping %s, because it has exceeded the maximum view count (%d/%d)' % (video_title, view_count, max_views) | |
869 | if age_restricted(info_dict.get('age_limit'), self.params.get('age_limit')): | |
870 | return 'Skipping "%s" because it is age restricted' % video_title | |
871 | if self.in_download_archive(info_dict): | |
872 | return '%s has already been recorded in archive' % video_title | |
873 | ||
874 | if not incomplete: | |
875 | match_filter = self.params.get('match_filter') | |
876 | if match_filter is not None: | |
877 | ret = match_filter(info_dict) | |
878 | if ret is not None: | |
879 | return ret | |
880 | return None | |
881 | ||
882 | reason = check_filter() | |
883 | if reason is not None: | |
884 | self.to_screen('[download] ' + reason) | |
d83cb531 | 885 | if reason.endswith('has already been recorded in the archive') and self.params.get('break_on_existing', False): |
8b0d7497 | 886 | raise ExistingVideoReached() |
d83cb531 | 887 | elif self.params.get('break_on_reject', False): |
8b0d7497 | 888 | raise RejectedVideoReached() |
889 | return reason | |
fe7e0c98 | 890 | |
b6c45014 JMF |
891 | @staticmethod |
892 | def add_extra_info(info_dict, extra_info): | |
893 | '''Set the keys from extra_info in info dict if they are missing''' | |
894 | for key, value in extra_info.items(): | |
895 | info_dict.setdefault(key, value) | |
896 | ||
0704d222 | 897 | def extract_info(self, url, download=True, ie_key=None, info_dict=None, extra_info={}, |
61aa5ba3 | 898 | process=True, force_generic_extractor=False): |
8222d8de JMF |
899 | ''' |
900 | Returns a list with a dictionary for each video we find. | |
901 | If 'download', also downloads the videos. | |
902 | extra_info is a dict containing the extra values to add to each result | |
613b2d9d | 903 | ''' |
fe7e0c98 | 904 | |
61aa5ba3 | 905 | if not ie_key and force_generic_extractor: |
d22dec74 S |
906 | ie_key = 'Generic' |
907 | ||
8222d8de | 908 | if ie_key: |
56c73665 | 909 | ies = [self.get_info_extractor(ie_key)] |
8222d8de JMF |
910 | else: |
911 | ies = self._ies | |
912 | ||
913 | for ie in ies: | |
914 | if not ie.suitable(url): | |
915 | continue | |
916 | ||
9a68de12 | 917 | ie_key = ie.ie_key() |
918 | ie = self.get_info_extractor(ie_key) | |
8222d8de | 919 | if not ie.working(): |
6febd1c1 PH |
920 | self.report_warning('The program functionality for this site has been marked as broken, ' |
921 | 'and will probably not work.') | |
8222d8de JMF |
922 | |
923 | try: | |
d0757229 | 924 | temp_id = str_or_none( |
63be1aab | 925 | ie.extract_id(url) if callable(getattr(ie, 'extract_id', None)) |
926 | else ie._match_id(url)) | |
a0566bbf | 927 | except (AssertionError, IndexError, AttributeError): |
928 | temp_id = None | |
929 | if temp_id is not None and self.in_download_archive({'id': temp_id, 'ie_key': ie_key}): | |
930 | self.to_screen("[%s] %s: has already been recorded in archive" % ( | |
931 | ie_key, temp_id)) | |
932 | break | |
933 | ||
934 | return self.__extract_info(url, ie, download, extra_info, process, info_dict) | |
935 | ||
936 | else: | |
937 | self.report_error('no suitable InfoExtractor for URL %s' % url) | |
938 | ||
939 | def __handle_extraction_exceptions(func): | |
940 | def wrapper(self, *args, **kwargs): | |
941 | try: | |
942 | return func(self, *args, **kwargs) | |
773f291d S |
943 | except GeoRestrictedError as e: |
944 | msg = e.msg | |
945 | if e.countries: | |
946 | msg += '\nThis video is available in %s.' % ', '.join( | |
947 | map(ISO3166Utils.short2full, e.countries)) | |
948 | msg += '\nYou might want to use a VPN or a proxy server (with --proxy) to workaround.' | |
949 | self.report_error(msg) | |
fb043a6e | 950 | except ExtractorError as e: # An error we somewhat expected |
2c74e6fa | 951 | self.report_error(compat_str(e), e.format_traceback()) |
8b0d7497 | 952 | except (MaxDownloadsReached, ExistingVideoReached, RejectedVideoReached): |
d3e5bbf4 | 953 | raise |
8222d8de JMF |
954 | except Exception as e: |
955 | if self.params.get('ignoreerrors', False): | |
9b9c5355 | 956 | self.report_error(error_to_compat_str(e), tb=encode_compat_str(traceback.format_exc())) |
8222d8de JMF |
957 | else: |
958 | raise | |
a0566bbf | 959 | return wrapper |
960 | ||
961 | @__handle_extraction_exceptions | |
962 | def __extract_info(self, url, ie, download, extra_info, process, info_dict): | |
963 | ie_result = ie.extract(url) | |
964 | if ie_result is None: # Finished already (backwards compatibility; listformats and friends should be moved here) | |
965 | return | |
966 | if isinstance(ie_result, list): | |
967 | # Backwards compatibility: old IE result format | |
968 | ie_result = { | |
969 | '_type': 'compat_list', | |
970 | 'entries': ie_result, | |
971 | } | |
972 | if info_dict: | |
973 | if info_dict.get('id'): | |
974 | ie_result['id'] = info_dict['id'] | |
975 | if info_dict.get('title'): | |
976 | ie_result['title'] = info_dict['title'] | |
977 | self.add_default_extra_info(ie_result, ie, url) | |
978 | if process: | |
979 | return self.process_ie_result(ie_result, download, extra_info) | |
8222d8de | 980 | else: |
a0566bbf | 981 | return ie_result |
fe7e0c98 | 982 | |
ea38e55f PH |
983 | def add_default_extra_info(self, ie_result, ie, url): |
984 | self.add_extra_info(ie_result, { | |
985 | 'extractor': ie.IE_NAME, | |
986 | 'webpage_url': url, | |
dbbbe555 | 987 | 'duration_string': ( |
988 | formatSeconds(ie_result['duration'], '-') | |
989 | if ie_result.get('duration', None) is not None | |
990 | else None), | |
ea38e55f PH |
991 | 'webpage_url_basename': url_basename(url), |
992 | 'extractor_key': ie.ie_key(), | |
993 | }) | |
994 | ||
8222d8de JMF |
995 | def process_ie_result(self, ie_result, download=True, extra_info={}): |
996 | """ | |
997 | Take the result of the ie(may be modified) and resolve all unresolved | |
998 | references (URLs, playlist items). | |
999 | ||
1000 | It will also download the videos if 'download'. | |
1001 | Returns the resolved ie_result. | |
1002 | """ | |
e8ee972c PH |
1003 | result_type = ie_result.get('_type', 'video') |
1004 | ||
057a5206 | 1005 | if result_type in ('url', 'url_transparent'): |
134c6ea8 | 1006 | ie_result['url'] = sanitize_url(ie_result['url']) |
057a5206 | 1007 | extract_flat = self.params.get('extract_flat', False) |
3089bc74 S |
1008 | if ((extract_flat == 'in_playlist' and 'playlist' in extra_info) |
1009 | or extract_flat is True): | |
d06daf23 | 1010 | self.__forced_printings( |
0202b52a | 1011 | ie_result, |
1012 | self.prepare_filepath(self.prepare_filename(ie_result)), | |
d06daf23 | 1013 | incomplete=True) |
e8ee972c PH |
1014 | return ie_result |
1015 | ||
8222d8de | 1016 | if result_type == 'video': |
b6c45014 | 1017 | self.add_extra_info(ie_result, extra_info) |
feee2ecf | 1018 | return self.process_video_result(ie_result, download=download) |
8222d8de JMF |
1019 | elif result_type == 'url': |
1020 | # We have to add extra_info to the results because it may be | |
1021 | # contained in a playlist | |
1022 | return self.extract_info(ie_result['url'], | |
0704d222 | 1023 | download, info_dict=ie_result, |
8222d8de JMF |
1024 | ie_key=ie_result.get('ie_key'), |
1025 | extra_info=extra_info) | |
7fc3fa05 PH |
1026 | elif result_type == 'url_transparent': |
1027 | # Use the information from the embedding page | |
1028 | info = self.extract_info( | |
1029 | ie_result['url'], ie_key=ie_result.get('ie_key'), | |
1030 | extra_info=extra_info, download=False, process=False) | |
1031 | ||
1640eb09 S |
1032 | # extract_info may return None when ignoreerrors is enabled and |
1033 | # extraction failed with an error, don't crash and return early | |
1034 | # in this case | |
1035 | if not info: | |
1036 | return info | |
1037 | ||
412c617d PH |
1038 | force_properties = dict( |
1039 | (k, v) for k, v in ie_result.items() if v is not None) | |
0396806f | 1040 | for f in ('_type', 'url', 'id', 'extractor', 'extractor_key', 'ie_key'): |
412c617d PH |
1041 | if f in force_properties: |
1042 | del force_properties[f] | |
1043 | new_result = info.copy() | |
1044 | new_result.update(force_properties) | |
7fc3fa05 | 1045 | |
0563f7ac S |
1046 | # Extracted info may not be a video result (i.e. |
1047 | # info.get('_type', 'video') != video) but rather an url or | |
1048 | # url_transparent. In such cases outer metadata (from ie_result) | |
1049 | # should be propagated to inner one (info). For this to happen | |
1050 | # _type of info should be overridden with url_transparent. This | |
067aa17e | 1051 | # fixes issue from https://github.com/ytdl-org/youtube-dl/pull/11163. |
0563f7ac S |
1052 | if new_result.get('_type') == 'url': |
1053 | new_result['_type'] = 'url_transparent' | |
7fc3fa05 PH |
1054 | |
1055 | return self.process_ie_result( | |
1056 | new_result, download=download, extra_info=extra_info) | |
40fcba5e | 1057 | elif result_type in ('playlist', 'multi_video'): |
30a074c2 | 1058 | # Protect from infinite recursion due to recursively nested playlists |
1059 | # (see https://github.com/ytdl-org/youtube-dl/issues/27833) | |
1060 | webpage_url = ie_result['webpage_url'] | |
1061 | if webpage_url in self._playlist_urls: | |
7e85e872 | 1062 | self.to_screen( |
30a074c2 | 1063 | '[download] Skipping already downloaded playlist: %s' |
1064 | % ie_result.get('title') or ie_result.get('id')) | |
1065 | return | |
7e85e872 | 1066 | |
30a074c2 | 1067 | self._playlist_level += 1 |
1068 | self._playlist_urls.add(webpage_url) | |
1069 | try: | |
1070 | return self.__process_playlist(ie_result, download) | |
1071 | finally: | |
1072 | self._playlist_level -= 1 | |
1073 | if not self._playlist_level: | |
1074 | self._playlist_urls.clear() | |
8222d8de | 1075 | elif result_type == 'compat_list': |
c9bf4114 PH |
1076 | self.report_warning( |
1077 | 'Extractor %s returned a compat_list result. ' | |
1078 | 'It needs to be updated.' % ie_result.get('extractor')) | |
5f6a1245 | 1079 | |
8222d8de | 1080 | def _fixup(r): |
9e1a5b84 JW |
1081 | self.add_extra_info( |
1082 | r, | |
9103bbc5 JMF |
1083 | { |
1084 | 'extractor': ie_result['extractor'], | |
1085 | 'webpage_url': ie_result['webpage_url'], | |
29eb5174 | 1086 | 'webpage_url_basename': url_basename(ie_result['webpage_url']), |
be97abc2 | 1087 | 'extractor_key': ie_result['extractor_key'], |
9e1a5b84 JW |
1088 | } |
1089 | ) | |
8222d8de JMF |
1090 | return r |
1091 | ie_result['entries'] = [ | |
b6c45014 | 1092 | self.process_ie_result(_fixup(r), download, extra_info) |
8222d8de JMF |
1093 | for r in ie_result['entries'] |
1094 | ] | |
1095 | return ie_result | |
1096 | else: | |
1097 | raise Exception('Invalid result type: %s' % result_type) | |
1098 | ||
30a074c2 | 1099 | def __process_playlist(self, ie_result, download): |
1100 | # We process each entry in the playlist | |
1101 | playlist = ie_result.get('title') or ie_result.get('id') | |
1102 | self.to_screen('[download] Downloading playlist: %s' % playlist) | |
1103 | ||
1104 | playlist_results = [] | |
1105 | ||
1106 | playliststart = self.params.get('playliststart', 1) - 1 | |
1107 | playlistend = self.params.get('playlistend') | |
1108 | # For backwards compatibility, interpret -1 as whole list | |
1109 | if playlistend == -1: | |
1110 | playlistend = None | |
1111 | ||
1112 | playlistitems_str = self.params.get('playlist_items') | |
1113 | playlistitems = None | |
1114 | if playlistitems_str is not None: | |
1115 | def iter_playlistitems(format): | |
1116 | for string_segment in format.split(','): | |
1117 | if '-' in string_segment: | |
1118 | start, end = string_segment.split('-') | |
1119 | for item in range(int(start), int(end) + 1): | |
1120 | yield int(item) | |
1121 | else: | |
1122 | yield int(string_segment) | |
1123 | playlistitems = orderedSet(iter_playlistitems(playlistitems_str)) | |
1124 | ||
1125 | ie_entries = ie_result['entries'] | |
1126 | ||
1127 | def make_playlistitems_entries(list_ie_entries): | |
1128 | num_entries = len(list_ie_entries) | |
1129 | return [ | |
1130 | list_ie_entries[i - 1] for i in playlistitems | |
1131 | if -num_entries <= i - 1 < num_entries] | |
1132 | ||
1133 | def report_download(num_entries): | |
1134 | self.to_screen( | |
1135 | '[%s] playlist %s: Downloading %d videos' % | |
1136 | (ie_result['extractor'], playlist, num_entries)) | |
1137 | ||
1138 | if isinstance(ie_entries, list): | |
1139 | n_all_entries = len(ie_entries) | |
1140 | if playlistitems: | |
1141 | entries = make_playlistitems_entries(ie_entries) | |
1142 | else: | |
1143 | entries = ie_entries[playliststart:playlistend] | |
1144 | n_entries = len(entries) | |
1145 | self.to_screen( | |
1146 | '[%s] playlist %s: Collected %d video ids (downloading %d of them)' % | |
1147 | (ie_result['extractor'], playlist, n_all_entries, n_entries)) | |
1148 | elif isinstance(ie_entries, PagedList): | |
1149 | if playlistitems: | |
1150 | entries = [] | |
1151 | for item in playlistitems: | |
1152 | entries.extend(ie_entries.getslice( | |
1153 | item - 1, item | |
1154 | )) | |
1155 | else: | |
1156 | entries = ie_entries.getslice( | |
1157 | playliststart, playlistend) | |
1158 | n_entries = len(entries) | |
1159 | report_download(n_entries) | |
1160 | else: # iterable | |
1161 | if playlistitems: | |
1162 | entries = make_playlistitems_entries(list(itertools.islice( | |
1163 | ie_entries, 0, max(playlistitems)))) | |
1164 | else: | |
1165 | entries = list(itertools.islice( | |
1166 | ie_entries, playliststart, playlistend)) | |
1167 | n_entries = len(entries) | |
1168 | report_download(n_entries) | |
1169 | ||
1170 | if self.params.get('playlistreverse', False): | |
1171 | entries = entries[::-1] | |
1172 | ||
1173 | if self.params.get('playlistrandom', False): | |
1174 | random.shuffle(entries) | |
1175 | ||
1176 | x_forwarded_for = ie_result.get('__x_forwarded_for_ip') | |
1177 | ||
1178 | for i, entry in enumerate(entries, 1): | |
1179 | self.to_screen('[download] Downloading video %s of %s' % (i, n_entries)) | |
1180 | # This __x_forwarded_for_ip thing is a bit ugly but requires | |
1181 | # minimal changes | |
1182 | if x_forwarded_for: | |
1183 | entry['__x_forwarded_for_ip'] = x_forwarded_for | |
1184 | extra = { | |
1185 | 'n_entries': n_entries, | |
1186 | 'playlist': playlist, | |
1187 | 'playlist_id': ie_result.get('id'), | |
1188 | 'playlist_title': ie_result.get('title'), | |
1189 | 'playlist_uploader': ie_result.get('uploader'), | |
1190 | 'playlist_uploader_id': ie_result.get('uploader_id'), | |
1191 | 'playlist_index': playlistitems[i - 1] if playlistitems else i + playliststart, | |
1192 | 'extractor': ie_result['extractor'], | |
1193 | 'webpage_url': ie_result['webpage_url'], | |
1194 | 'webpage_url_basename': url_basename(ie_result['webpage_url']), | |
1195 | 'extractor_key': ie_result['extractor_key'], | |
1196 | } | |
1197 | ||
1198 | if self._match_entry(entry, incomplete=True) is not None: | |
1199 | continue | |
1200 | ||
1201 | entry_result = self.__process_iterable_entry(entry, download, extra) | |
1202 | # TODO: skip failed (empty) entries? | |
1203 | playlist_results.append(entry_result) | |
1204 | ie_result['entries'] = playlist_results | |
1205 | self.to_screen('[download] Finished downloading playlist: %s' % playlist) | |
1206 | return ie_result | |
1207 | ||
a0566bbf | 1208 | @__handle_extraction_exceptions |
1209 | def __process_iterable_entry(self, entry, download, extra_info): | |
1210 | return self.process_ie_result( | |
1211 | entry, download=download, extra_info=extra_info) | |
1212 | ||
67134eab JMF |
1213 | def _build_format_filter(self, filter_spec): |
1214 | " Returns a function to filter the formats according to the filter_spec " | |
083c9df9 PH |
1215 | |
1216 | OPERATORS = { | |
1217 | '<': operator.lt, | |
1218 | '<=': operator.le, | |
1219 | '>': operator.gt, | |
1220 | '>=': operator.ge, | |
1221 | '=': operator.eq, | |
1222 | '!=': operator.ne, | |
1223 | } | |
67134eab | 1224 | operator_rex = re.compile(r'''(?x)\s* |
a03a3c80 | 1225 | (?P<key>width|height|tbr|abr|vbr|asr|filesize|filesize_approx|fps) |
083c9df9 PH |
1226 | \s*(?P<op>%s)(?P<none_inclusive>\s*\?)?\s* |
1227 | (?P<value>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?) | |
67134eab | 1228 | $ |
083c9df9 | 1229 | ''' % '|'.join(map(re.escape, OPERATORS.keys()))) |
67134eab | 1230 | m = operator_rex.search(filter_spec) |
9ddb6925 S |
1231 | if m: |
1232 | try: | |
1233 | comparison_value = int(m.group('value')) | |
1234 | except ValueError: | |
1235 | comparison_value = parse_filesize(m.group('value')) | |
1236 | if comparison_value is None: | |
1237 | comparison_value = parse_filesize(m.group('value') + 'B') | |
1238 | if comparison_value is None: | |
1239 | raise ValueError( | |
1240 | 'Invalid value %r in format specification %r' % ( | |
67134eab | 1241 | m.group('value'), filter_spec)) |
9ddb6925 S |
1242 | op = OPERATORS[m.group('op')] |
1243 | ||
083c9df9 | 1244 | if not m: |
9ddb6925 S |
1245 | STR_OPERATORS = { |
1246 | '=': operator.eq, | |
10d33b34 YCH |
1247 | '^=': lambda attr, value: attr.startswith(value), |
1248 | '$=': lambda attr, value: attr.endswith(value), | |
1249 | '*=': lambda attr, value: value in attr, | |
9ddb6925 | 1250 | } |
67134eab | 1251 | str_operator_rex = re.compile(r'''(?x) |
f96bff99 | 1252 | \s*(?P<key>[a-zA-Z0-9._-]+) |
2cc779f4 | 1253 | \s*(?P<negation>!\s*)?(?P<op>%s)(?P<none_inclusive>\s*\?)? |
b0df5223 | 1254 | \s*(?P<value>[a-zA-Z0-9._-]+) |
67134eab | 1255 | \s*$ |
9ddb6925 | 1256 | ''' % '|'.join(map(re.escape, STR_OPERATORS.keys()))) |
67134eab | 1257 | m = str_operator_rex.search(filter_spec) |
9ddb6925 S |
1258 | if m: |
1259 | comparison_value = m.group('value') | |
2cc779f4 S |
1260 | str_op = STR_OPERATORS[m.group('op')] |
1261 | if m.group('negation'): | |
e118a879 | 1262 | op = lambda attr, value: not str_op(attr, value) |
2cc779f4 S |
1263 | else: |
1264 | op = str_op | |
083c9df9 | 1265 | |
9ddb6925 | 1266 | if not m: |
67134eab | 1267 | raise ValueError('Invalid filter specification %r' % filter_spec) |
083c9df9 PH |
1268 | |
1269 | def _filter(f): | |
1270 | actual_value = f.get(m.group('key')) | |
1271 | if actual_value is None: | |
1272 | return m.group('none_inclusive') | |
1273 | return op(actual_value, comparison_value) | |
67134eab JMF |
1274 | return _filter |
1275 | ||
0017d9ad | 1276 | def _default_format_spec(self, info_dict, download=True): |
0017d9ad | 1277 | |
af0f7428 S |
1278 | def can_merge(): |
1279 | merger = FFmpegMergerPP(self) | |
1280 | return merger.available and merger.can_merge() | |
1281 | ||
91ebc640 | 1282 | prefer_best = ( |
1283 | not self.params.get('simulate', False) | |
1284 | and download | |
1285 | and ( | |
1286 | not can_merge() | |
19807826 | 1287 | or info_dict.get('is_live', False) |
91ebc640 | 1288 | or self.params.get('outtmpl', DEFAULT_OUTTMPL) == '-')) |
1289 | ||
1290 | return ( | |
1291 | 'best/bestvideo+bestaudio' | |
1292 | if prefer_best | |
1293 | else 'bestvideo*+bestaudio/best' | |
19807826 | 1294 | if not self.params.get('allow_multiple_audio_streams', False) |
91ebc640 | 1295 | else 'bestvideo+bestaudio/best') |
0017d9ad | 1296 | |
67134eab JMF |
1297 | def build_format_selector(self, format_spec): |
1298 | def syntax_error(note, start): | |
1299 | message = ( | |
1300 | 'Invalid format specification: ' | |
1301 | '{0}\n\t{1}\n\t{2}^'.format(note, format_spec, ' ' * start[1])) | |
1302 | return SyntaxError(message) | |
1303 | ||
1304 | PICKFIRST = 'PICKFIRST' | |
1305 | MERGE = 'MERGE' | |
1306 | SINGLE = 'SINGLE' | |
0130afb7 | 1307 | GROUP = 'GROUP' |
67134eab JMF |
1308 | FormatSelector = collections.namedtuple('FormatSelector', ['type', 'selector', 'filters']) |
1309 | ||
91ebc640 | 1310 | allow_multiple_streams = {'audio': self.params.get('allow_multiple_audio_streams', False), |
1311 | 'video': self.params.get('allow_multiple_video_streams', False)} | |
909d24dd | 1312 | |
67134eab JMF |
1313 | def _parse_filter(tokens): |
1314 | filter_parts = [] | |
1315 | for type, string, start, _, _ in tokens: | |
1316 | if type == tokenize.OP and string == ']': | |
1317 | return ''.join(filter_parts) | |
1318 | else: | |
1319 | filter_parts.append(string) | |
1320 | ||
232541df | 1321 | def _remove_unused_ops(tokens): |
17cc1534 | 1322 | # Remove operators that we don't use and join them with the surrounding strings |
232541df JMF |
1323 | # for example: 'mp4' '-' 'baseline' '-' '16x9' is converted to 'mp4-baseline-16x9' |
1324 | ALLOWED_OPS = ('/', '+', ',', '(', ')') | |
1325 | last_string, last_start, last_end, last_line = None, None, None, None | |
1326 | for type, string, start, end, line in tokens: | |
1327 | if type == tokenize.OP and string == '[': | |
1328 | if last_string: | |
1329 | yield tokenize.NAME, last_string, last_start, last_end, last_line | |
1330 | last_string = None | |
1331 | yield type, string, start, end, line | |
1332 | # everything inside brackets will be handled by _parse_filter | |
1333 | for type, string, start, end, line in tokens: | |
1334 | yield type, string, start, end, line | |
1335 | if type == tokenize.OP and string == ']': | |
1336 | break | |
1337 | elif type == tokenize.OP and string in ALLOWED_OPS: | |
1338 | if last_string: | |
1339 | yield tokenize.NAME, last_string, last_start, last_end, last_line | |
1340 | last_string = None | |
1341 | yield type, string, start, end, line | |
1342 | elif type in [tokenize.NAME, tokenize.NUMBER, tokenize.OP]: | |
1343 | if not last_string: | |
1344 | last_string = string | |
1345 | last_start = start | |
1346 | last_end = end | |
1347 | else: | |
1348 | last_string += string | |
1349 | if last_string: | |
1350 | yield tokenize.NAME, last_string, last_start, last_end, last_line | |
1351 | ||
cf2ac6df | 1352 | def _parse_format_selection(tokens, inside_merge=False, inside_choice=False, inside_group=False): |
67134eab JMF |
1353 | selectors = [] |
1354 | current_selector = None | |
1355 | for type, string, start, _, _ in tokens: | |
1356 | # ENCODING is only defined in python 3.x | |
1357 | if type == getattr(tokenize, 'ENCODING', None): | |
1358 | continue | |
1359 | elif type in [tokenize.NAME, tokenize.NUMBER]: | |
1360 | current_selector = FormatSelector(SINGLE, string, []) | |
1361 | elif type == tokenize.OP: | |
cf2ac6df JMF |
1362 | if string == ')': |
1363 | if not inside_group: | |
1364 | # ')' will be handled by the parentheses group | |
1365 | tokens.restore_last_token() | |
67134eab | 1366 | break |
cf2ac6df | 1367 | elif inside_merge and string in ['/', ',']: |
0130afb7 JMF |
1368 | tokens.restore_last_token() |
1369 | break | |
cf2ac6df JMF |
1370 | elif inside_choice and string == ',': |
1371 | tokens.restore_last_token() | |
1372 | break | |
1373 | elif string == ',': | |
0a31a350 JMF |
1374 | if not current_selector: |
1375 | raise syntax_error('"," must follow a format selector', start) | |
67134eab JMF |
1376 | selectors.append(current_selector) |
1377 | current_selector = None | |
1378 | elif string == '/': | |
d96d604e JMF |
1379 | if not current_selector: |
1380 | raise syntax_error('"/" must follow a format selector', start) | |
67134eab | 1381 | first_choice = current_selector |
cf2ac6df | 1382 | second_choice = _parse_format_selection(tokens, inside_choice=True) |
f5f4a27a | 1383 | current_selector = FormatSelector(PICKFIRST, (first_choice, second_choice), []) |
67134eab JMF |
1384 | elif string == '[': |
1385 | if not current_selector: | |
1386 | current_selector = FormatSelector(SINGLE, 'best', []) | |
1387 | format_filter = _parse_filter(tokens) | |
1388 | current_selector.filters.append(format_filter) | |
0130afb7 JMF |
1389 | elif string == '(': |
1390 | if current_selector: | |
1391 | raise syntax_error('Unexpected "("', start) | |
cf2ac6df JMF |
1392 | group = _parse_format_selection(tokens, inside_group=True) |
1393 | current_selector = FormatSelector(GROUP, group, []) | |
67134eab | 1394 | elif string == '+': |
d03cfdce | 1395 | if not current_selector: |
1396 | raise syntax_error('Unexpected "+"', start) | |
1397 | selector_1 = current_selector | |
1398 | selector_2 = _parse_format_selection(tokens, inside_merge=True) | |
1399 | if not selector_2: | |
1400 | raise syntax_error('Expected a selector', start) | |
1401 | current_selector = FormatSelector(MERGE, (selector_1, selector_2), []) | |
67134eab JMF |
1402 | else: |
1403 | raise syntax_error('Operator not recognized: "{0}"'.format(string), start) | |
1404 | elif type == tokenize.ENDMARKER: | |
1405 | break | |
1406 | if current_selector: | |
1407 | selectors.append(current_selector) | |
1408 | return selectors | |
1409 | ||
1410 | def _build_selector_function(selector): | |
909d24dd | 1411 | if isinstance(selector, list): # , |
67134eab JMF |
1412 | fs = [_build_selector_function(s) for s in selector] |
1413 | ||
317f7ab6 | 1414 | def selector_function(ctx): |
67134eab | 1415 | for f in fs: |
317f7ab6 | 1416 | for format in f(ctx): |
67134eab JMF |
1417 | yield format |
1418 | return selector_function | |
909d24dd | 1419 | |
1420 | elif selector.type == GROUP: # () | |
0130afb7 | 1421 | selector_function = _build_selector_function(selector.selector) |
909d24dd | 1422 | |
1423 | elif selector.type == PICKFIRST: # / | |
67134eab JMF |
1424 | fs = [_build_selector_function(s) for s in selector.selector] |
1425 | ||
317f7ab6 | 1426 | def selector_function(ctx): |
67134eab | 1427 | for f in fs: |
317f7ab6 | 1428 | picked_formats = list(f(ctx)) |
67134eab JMF |
1429 | if picked_formats: |
1430 | return picked_formats | |
1431 | return [] | |
67134eab | 1432 | |
909d24dd | 1433 | elif selector.type == SINGLE: # atom |
1434 | format_spec = selector.selector if selector.selector is not None else 'best' | |
1435 | ||
1436 | if format_spec == 'all': | |
1437 | def selector_function(ctx): | |
1438 | formats = list(ctx['formats']) | |
1439 | if formats: | |
1440 | for f in formats: | |
1441 | yield f | |
1442 | ||
1443 | else: | |
1444 | format_fallback = False | |
1445 | format_spec_obj = re.match(r'(best|worst|b|w)(video|audio|v|a)?(\*)?$', format_spec) | |
1446 | if format_spec_obj is not None: | |
1447 | format_idx = 0 if format_spec_obj.group(1)[0] == 'w' else -1 | |
1448 | format_type = format_spec_obj.group(2)[0] if format_spec_obj.group(2) else False | |
1449 | not_format_type = 'v' if format_type == 'a' else 'a' | |
1450 | format_modified = format_spec_obj.group(3) is not None | |
1451 | ||
1452 | format_fallback = not format_type and not format_modified # for b, w | |
1453 | filter_f = ((lambda f: f.get(format_type + 'codec') != 'none') | |
1454 | if format_type and format_modified # bv*, ba*, wv*, wa* | |
1455 | else (lambda f: f.get(not_format_type + 'codec') == 'none') | |
1456 | if format_type # bv, ba, wv, wa | |
1457 | else (lambda f: f.get('vcodec') != 'none' and f.get('acodec') != 'none') | |
1458 | if not format_modified # b, w | |
1459 | else None) # b*, w* | |
67134eab | 1460 | else: |
909d24dd | 1461 | format_idx = -1 |
1462 | filter_f = ((lambda f: f.get('ext') == format_spec) | |
1463 | if format_spec in ['mp4', 'flv', 'webm', '3gp', 'm4a', 'mp3', 'ogg', 'aac', 'wav'] # extension | |
1464 | else (lambda f: f.get('format_id') == format_spec)) # id | |
1465 | ||
1466 | def selector_function(ctx): | |
1467 | formats = list(ctx['formats']) | |
1468 | if not formats: | |
1469 | return | |
1470 | matches = list(filter(filter_f, formats)) if filter_f is not None else formats | |
67134eab | 1471 | if matches: |
909d24dd | 1472 | yield matches[format_idx] |
1473 | elif format_fallback == 'force' or (format_fallback and ctx['incomplete_formats']): | |
1474 | # for extractors with incomplete formats (audio only (soundcloud) | |
1475 | # or video only (imgur)) best/worst will fallback to | |
1476 | # best/worst {video,audio}-only format | |
1477 | yield formats[format_idx] | |
1478 | ||
1479 | elif selector.type == MERGE: # + | |
d03cfdce | 1480 | def _merge(formats_pair): |
1481 | format_1, format_2 = formats_pair | |
1482 | ||
1483 | formats_info = [] | |
1484 | formats_info.extend(format_1.get('requested_formats', (format_1,))) | |
1485 | formats_info.extend(format_2.get('requested_formats', (format_2,))) | |
1486 | ||
909d24dd | 1487 | if not allow_multiple_streams['video'] or not allow_multiple_streams['audio']: |
1488 | get_no_more = {"video": False, "audio": False} | |
1489 | for (i, fmt_info) in enumerate(formats_info): | |
1490 | for aud_vid in ["audio", "video"]: | |
1491 | if not allow_multiple_streams[aud_vid] and fmt_info.get(aud_vid[0] + 'codec') != 'none': | |
1492 | if get_no_more[aud_vid]: | |
1493 | formats_info.pop(i) | |
1494 | get_no_more[aud_vid] = True | |
1495 | ||
1496 | if len(formats_info) == 1: | |
1497 | return formats_info[0] | |
1498 | ||
d03cfdce | 1499 | video_fmts = [fmt_info for fmt_info in formats_info if fmt_info.get('vcodec') != 'none'] |
1500 | audio_fmts = [fmt_info for fmt_info in formats_info if fmt_info.get('acodec') != 'none'] | |
1501 | ||
1502 | the_only_video = video_fmts[0] if len(video_fmts) == 1 else None | |
1503 | the_only_audio = audio_fmts[0] if len(audio_fmts) == 1 else None | |
1504 | ||
1505 | output_ext = self.params.get('merge_output_format') | |
1506 | if not output_ext: | |
1507 | if the_only_video: | |
1508 | output_ext = the_only_video['ext'] | |
1509 | elif the_only_audio and not video_fmts: | |
1510 | output_ext = the_only_audio['ext'] | |
1511 | else: | |
1512 | output_ext = 'mkv' | |
1513 | ||
1514 | new_dict = { | |
67134eab | 1515 | 'requested_formats': formats_info, |
d03cfdce | 1516 | 'format': '+'.join(fmt_info.get('format') for fmt_info in formats_info), |
1517 | 'format_id': '+'.join(fmt_info.get('format_id') for fmt_info in formats_info), | |
67134eab JMF |
1518 | 'ext': output_ext, |
1519 | } | |
d03cfdce | 1520 | |
1521 | if the_only_video: | |
1522 | new_dict.update({ | |
1523 | 'width': the_only_video.get('width'), | |
1524 | 'height': the_only_video.get('height'), | |
1525 | 'resolution': the_only_video.get('resolution'), | |
1526 | 'fps': the_only_video.get('fps'), | |
1527 | 'vcodec': the_only_video.get('vcodec'), | |
1528 | 'vbr': the_only_video.get('vbr'), | |
1529 | 'stretched_ratio': the_only_video.get('stretched_ratio'), | |
1530 | }) | |
1531 | ||
1532 | if the_only_audio: | |
1533 | new_dict.update({ | |
1534 | 'acodec': the_only_audio.get('acodec'), | |
1535 | 'abr': the_only_audio.get('abr'), | |
1536 | }) | |
1537 | ||
1538 | return new_dict | |
1539 | ||
1540 | selector_1, selector_2 = map(_build_selector_function, selector.selector) | |
083c9df9 | 1541 | |
317f7ab6 S |
1542 | def selector_function(ctx): |
1543 | for pair in itertools.product( | |
d03cfdce | 1544 | selector_1(copy.deepcopy(ctx)), selector_2(copy.deepcopy(ctx))): |
67134eab | 1545 | yield _merge(pair) |
083c9df9 | 1546 | |
67134eab | 1547 | filters = [self._build_format_filter(f) for f in selector.filters] |
083c9df9 | 1548 | |
317f7ab6 S |
1549 | def final_selector(ctx): |
1550 | ctx_copy = copy.deepcopy(ctx) | |
67134eab | 1551 | for _filter in filters: |
317f7ab6 S |
1552 | ctx_copy['formats'] = list(filter(_filter, ctx_copy['formats'])) |
1553 | return selector_function(ctx_copy) | |
67134eab | 1554 | return final_selector |
083c9df9 | 1555 | |
67134eab | 1556 | stream = io.BytesIO(format_spec.encode('utf-8')) |
0130afb7 | 1557 | try: |
232541df | 1558 | tokens = list(_remove_unused_ops(compat_tokenize_tokenize(stream.readline))) |
0130afb7 JMF |
1559 | except tokenize.TokenError: |
1560 | raise syntax_error('Missing closing/opening brackets or parenthesis', (0, len(format_spec))) | |
1561 | ||
1562 | class TokenIterator(object): | |
1563 | def __init__(self, tokens): | |
1564 | self.tokens = tokens | |
1565 | self.counter = 0 | |
1566 | ||
1567 | def __iter__(self): | |
1568 | return self | |
1569 | ||
1570 | def __next__(self): | |
1571 | if self.counter >= len(self.tokens): | |
1572 | raise StopIteration() | |
1573 | value = self.tokens[self.counter] | |
1574 | self.counter += 1 | |
1575 | return value | |
1576 | ||
1577 | next = __next__ | |
1578 | ||
1579 | def restore_last_token(self): | |
1580 | self.counter -= 1 | |
1581 | ||
1582 | parsed_selector = _parse_format_selection(iter(TokenIterator(tokens))) | |
67134eab | 1583 | return _build_selector_function(parsed_selector) |
a9c58ad9 | 1584 | |
e5660ee6 JMF |
1585 | def _calc_headers(self, info_dict): |
1586 | res = std_headers.copy() | |
1587 | ||
1588 | add_headers = info_dict.get('http_headers') | |
1589 | if add_headers: | |
1590 | res.update(add_headers) | |
1591 | ||
1592 | cookies = self._calc_cookies(info_dict) | |
1593 | if cookies: | |
1594 | res['Cookie'] = cookies | |
1595 | ||
0016b84e S |
1596 | if 'X-Forwarded-For' not in res: |
1597 | x_forwarded_for_ip = info_dict.get('__x_forwarded_for_ip') | |
1598 | if x_forwarded_for_ip: | |
1599 | res['X-Forwarded-For'] = x_forwarded_for_ip | |
1600 | ||
e5660ee6 JMF |
1601 | return res |
1602 | ||
1603 | def _calc_cookies(self, info_dict): | |
5c2266df | 1604 | pr = sanitized_Request(info_dict['url']) |
e5660ee6 | 1605 | self.cookiejar.add_cookie_header(pr) |
662435f7 | 1606 | return pr.get_header('Cookie') |
e5660ee6 | 1607 | |
dd82ffea JMF |
1608 | def process_video_result(self, info_dict, download=True): |
1609 | assert info_dict.get('_type', 'video') == 'video' | |
1610 | ||
bec1fad2 PH |
1611 | if 'id' not in info_dict: |
1612 | raise ExtractorError('Missing "id" field in extractor result') | |
1613 | if 'title' not in info_dict: | |
1614 | raise ExtractorError('Missing "title" field in extractor result') | |
1615 | ||
c9969434 S |
1616 | def report_force_conversion(field, field_not, conversion): |
1617 | self.report_warning( | |
1618 | '"%s" field is not %s - forcing %s conversion, there is an error in extractor' | |
1619 | % (field, field_not, conversion)) | |
1620 | ||
1621 | def sanitize_string_field(info, string_field): | |
1622 | field = info.get(string_field) | |
1623 | if field is None or isinstance(field, compat_str): | |
1624 | return | |
1625 | report_force_conversion(string_field, 'a string', 'string') | |
1626 | info[string_field] = compat_str(field) | |
1627 | ||
1628 | def sanitize_numeric_fields(info): | |
1629 | for numeric_field in self._NUMERIC_FIELDS: | |
1630 | field = info.get(numeric_field) | |
1631 | if field is None or isinstance(field, compat_numeric_types): | |
1632 | continue | |
1633 | report_force_conversion(numeric_field, 'numeric', 'int') | |
1634 | info[numeric_field] = int_or_none(field) | |
1635 | ||
1636 | sanitize_string_field(info_dict, 'id') | |
1637 | sanitize_numeric_fields(info_dict) | |
be6217b2 | 1638 | |
dd82ffea JMF |
1639 | if 'playlist' not in info_dict: |
1640 | # It isn't part of a playlist | |
1641 | info_dict['playlist'] = None | |
1642 | info_dict['playlist_index'] = None | |
1643 | ||
d5519808 | 1644 | thumbnails = info_dict.get('thumbnails') |
cfb56d1a PH |
1645 | if thumbnails is None: |
1646 | thumbnail = info_dict.get('thumbnail') | |
1647 | if thumbnail: | |
a7a14d95 | 1648 | info_dict['thumbnails'] = thumbnails = [{'url': thumbnail}] |
d5519808 | 1649 | if thumbnails: |
be6d7229 | 1650 | thumbnails.sort(key=lambda t: ( |
d37708fc RA |
1651 | t.get('preference') if t.get('preference') is not None else -1, |
1652 | t.get('width') if t.get('width') is not None else -1, | |
1653 | t.get('height') if t.get('height') is not None else -1, | |
1654 | t.get('id') if t.get('id') is not None else '', t.get('url'))) | |
f6c24009 | 1655 | for i, t in enumerate(thumbnails): |
dcf77cf1 | 1656 | t['url'] = sanitize_url(t['url']) |
9603e8a7 | 1657 | if t.get('width') and t.get('height'): |
d5519808 | 1658 | t['resolution'] = '%dx%d' % (t['width'], t['height']) |
f6c24009 PH |
1659 | if t.get('id') is None: |
1660 | t['id'] = '%d' % i | |
d5519808 | 1661 | |
b7b72db9 | 1662 | if self.params.get('list_thumbnails'): |
1663 | self.list_thumbnails(info_dict) | |
1664 | return | |
1665 | ||
536a55da S |
1666 | thumbnail = info_dict.get('thumbnail') |
1667 | if thumbnail: | |
1668 | info_dict['thumbnail'] = sanitize_url(thumbnail) | |
1669 | elif thumbnails: | |
d5519808 PH |
1670 | info_dict['thumbnail'] = thumbnails[-1]['url'] |
1671 | ||
c9ae7b95 | 1672 | if 'display_id' not in info_dict and 'id' in info_dict: |
0afef30b PH |
1673 | info_dict['display_id'] = info_dict['id'] |
1674 | ||
955c4514 | 1675 | if info_dict.get('upload_date') is None and info_dict.get('timestamp') is not None: |
a55e36f4 S |
1676 | # Working around out-of-range timestamp values (e.g. negative ones on Windows, |
1677 | # see http://bugs.python.org/issue1646728) | |
1678 | try: | |
1679 | upload_date = datetime.datetime.utcfromtimestamp(info_dict['timestamp']) | |
1680 | info_dict['upload_date'] = upload_date.strftime('%Y%m%d') | |
1681 | except (ValueError, OverflowError, OSError): | |
1682 | pass | |
9d2ecdbc | 1683 | |
33d2fc2f S |
1684 | # Auto generate title fields corresponding to the *_number fields when missing |
1685 | # in order to always have clean titles. This is very common for TV series. | |
1686 | for field in ('chapter', 'season', 'episode'): | |
1687 | if info_dict.get('%s_number' % field) is not None and not info_dict.get(field): | |
1688 | info_dict[field] = '%s %d' % (field.capitalize(), info_dict['%s_number' % field]) | |
1689 | ||
05108a49 S |
1690 | for cc_kind in ('subtitles', 'automatic_captions'): |
1691 | cc = info_dict.get(cc_kind) | |
1692 | if cc: | |
1693 | for _, subtitle in cc.items(): | |
1694 | for subtitle_format in subtitle: | |
1695 | if subtitle_format.get('url'): | |
1696 | subtitle_format['url'] = sanitize_url(subtitle_format['url']) | |
1697 | if subtitle_format.get('ext') is None: | |
1698 | subtitle_format['ext'] = determine_ext(subtitle_format['url']).lower() | |
1699 | ||
1700 | automatic_captions = info_dict.get('automatic_captions') | |
4bba3716 | 1701 | subtitles = info_dict.get('subtitles') |
4bba3716 | 1702 | |
a504ced0 | 1703 | if self.params.get('listsubtitles', False): |
360e1ca5 | 1704 | if 'automatic_captions' in info_dict: |
05108a49 S |
1705 | self.list_subtitles( |
1706 | info_dict['id'], automatic_captions, 'automatic captions') | |
4bba3716 | 1707 | self.list_subtitles(info_dict['id'], subtitles, 'subtitles') |
a504ced0 | 1708 | return |
05108a49 | 1709 | |
360e1ca5 | 1710 | info_dict['requested_subtitles'] = self.process_subtitles( |
05108a49 | 1711 | info_dict['id'], subtitles, automatic_captions) |
a504ced0 | 1712 | |
dd82ffea JMF |
1713 | # We now pick which formats have to be downloaded |
1714 | if info_dict.get('formats') is None: | |
1715 | # There's only one format available | |
1716 | formats = [info_dict] | |
1717 | else: | |
1718 | formats = info_dict['formats'] | |
1719 | ||
db95dc13 PH |
1720 | if not formats: |
1721 | raise ExtractorError('No video formats found!') | |
1722 | ||
73af5cc8 S |
1723 | def is_wellformed(f): |
1724 | url = f.get('url') | |
a5ac0c47 | 1725 | if not url: |
73af5cc8 S |
1726 | self.report_warning( |
1727 | '"url" field is missing or empty - skipping format, ' | |
1728 | 'there is an error in extractor') | |
a5ac0c47 S |
1729 | return False |
1730 | if isinstance(url, bytes): | |
1731 | sanitize_string_field(f, 'url') | |
1732 | return True | |
73af5cc8 S |
1733 | |
1734 | # Filter out malformed formats for better extraction robustness | |
1735 | formats = list(filter(is_wellformed, formats)) | |
1736 | ||
181c7053 S |
1737 | formats_dict = {} |
1738 | ||
dd82ffea | 1739 | # We check that all the formats have the format and format_id fields |
db95dc13 | 1740 | for i, format in enumerate(formats): |
c9969434 S |
1741 | sanitize_string_field(format, 'format_id') |
1742 | sanitize_numeric_fields(format) | |
dcf77cf1 | 1743 | format['url'] = sanitize_url(format['url']) |
e74e3b63 | 1744 | if not format.get('format_id'): |
8016c922 | 1745 | format['format_id'] = compat_str(i) |
e2effb08 S |
1746 | else: |
1747 | # Sanitize format_id from characters used in format selector expression | |
ec85ded8 | 1748 | format['format_id'] = re.sub(r'[\s,/+\[\]()]', '_', format['format_id']) |
181c7053 S |
1749 | format_id = format['format_id'] |
1750 | if format_id not in formats_dict: | |
1751 | formats_dict[format_id] = [] | |
1752 | formats_dict[format_id].append(format) | |
1753 | ||
1754 | # Make sure all formats have unique format_id | |
1755 | for format_id, ambiguous_formats in formats_dict.items(): | |
1756 | if len(ambiguous_formats) > 1: | |
1757 | for i, format in enumerate(ambiguous_formats): | |
1758 | format['format_id'] = '%s-%d' % (format_id, i) | |
1759 | ||
1760 | for i, format in enumerate(formats): | |
8c51aa65 | 1761 | if format.get('format') is None: |
6febd1c1 | 1762 | format['format'] = '{id} - {res}{note}'.format( |
8c51aa65 JMF |
1763 | id=format['format_id'], |
1764 | res=self.format_resolution(format), | |
6febd1c1 | 1765 | note=' ({0})'.format(format['format_note']) if format.get('format_note') is not None else '', |
8c51aa65 | 1766 | ) |
c1002e96 | 1767 | # Automatically determine file extension if missing |
5b1d8575 | 1768 | if format.get('ext') is None: |
cce929ea | 1769 | format['ext'] = determine_ext(format['url']).lower() |
b5559424 S |
1770 | # Automatically determine protocol if missing (useful for format |
1771 | # selection purposes) | |
6f0be937 | 1772 | if format.get('protocol') is None: |
b5559424 | 1773 | format['protocol'] = determine_protocol(format) |
e5660ee6 JMF |
1774 | # Add HTTP headers, so that external programs can use them from the |
1775 | # json output | |
1776 | full_format_info = info_dict.copy() | |
1777 | full_format_info.update(format) | |
1778 | format['http_headers'] = self._calc_headers(full_format_info) | |
0016b84e S |
1779 | # Remove private housekeeping stuff |
1780 | if '__x_forwarded_for_ip' in info_dict: | |
1781 | del info_dict['__x_forwarded_for_ip'] | |
dd82ffea | 1782 | |
4bcc7bd1 | 1783 | # TODO Central sorting goes here |
99e206d5 | 1784 | |
f89197d7 | 1785 | if formats[0] is not info_dict: |
b3d9ef88 JMF |
1786 | # only set the 'formats' fields if the original info_dict list them |
1787 | # otherwise we end up with a circular reference, the first (and unique) | |
f89197d7 | 1788 | # element in the 'formats' field in info_dict is info_dict itself, |
dfb1b146 | 1789 | # which can't be exported to json |
b3d9ef88 | 1790 | info_dict['formats'] = formats |
cfb56d1a | 1791 | if self.params.get('listformats'): |
bfaae0a7 | 1792 | self.list_formats(info_dict) |
1793 | return | |
1794 | ||
de3ef3ed | 1795 | req_format = self.params.get('format') |
a9c58ad9 | 1796 | if req_format is None: |
0017d9ad S |
1797 | req_format = self._default_format_spec(info_dict, download=download) |
1798 | if self.params.get('verbose'): | |
29f7c58a | 1799 | self._write_string('[debug] Default format spec: %s\n' % req_format) |
0017d9ad | 1800 | |
5acfa126 | 1801 | format_selector = self.build_format_selector(req_format) |
317f7ab6 S |
1802 | |
1803 | # While in format selection we may need to have an access to the original | |
1804 | # format set in order to calculate some metrics or do some processing. | |
1805 | # For now we need to be able to guess whether original formats provided | |
1806 | # by extractor are incomplete or not (i.e. whether extractor provides only | |
1807 | # video-only or audio-only formats) for proper formats selection for | |
1808 | # extractors with such incomplete formats (see | |
067aa17e | 1809 | # https://github.com/ytdl-org/youtube-dl/pull/5556). |
317f7ab6 S |
1810 | # Since formats may be filtered during format selection and may not match |
1811 | # the original formats the results may be incorrect. Thus original formats | |
1812 | # or pre-calculated metrics should be passed to format selection routines | |
1813 | # as well. | |
1814 | # We will pass a context object containing all necessary additional data | |
1815 | # instead of just formats. | |
1816 | # This fixes incorrect format selection issue (see | |
067aa17e | 1817 | # https://github.com/ytdl-org/youtube-dl/issues/10083). |
2e221ca3 | 1818 | incomplete_formats = ( |
317f7ab6 | 1819 | # All formats are video-only or |
3089bc74 | 1820 | all(f.get('vcodec') != 'none' and f.get('acodec') == 'none' for f in formats) |
317f7ab6 | 1821 | # all formats are audio-only |
3089bc74 | 1822 | or all(f.get('vcodec') == 'none' and f.get('acodec') != 'none' for f in formats)) |
317f7ab6 S |
1823 | |
1824 | ctx = { | |
1825 | 'formats': formats, | |
1826 | 'incomplete_formats': incomplete_formats, | |
1827 | } | |
1828 | ||
1829 | formats_to_download = list(format_selector(ctx)) | |
dd82ffea | 1830 | if not formats_to_download: |
6febd1c1 | 1831 | raise ExtractorError('requested format not available', |
78a3a9f8 | 1832 | expected=True) |
dd82ffea JMF |
1833 | |
1834 | if download: | |
909d24dd | 1835 | self.to_screen('[info] Downloading format(s) %s' % ", ".join([f['format_id'] for f in formats_to_download])) |
dd82ffea | 1836 | if len(formats_to_download) > 1: |
6febd1c1 | 1837 | self.to_screen('[info] %s: downloading video in %s formats' % (info_dict['id'], len(formats_to_download))) |
dd82ffea JMF |
1838 | for format in formats_to_download: |
1839 | new_info = dict(info_dict) | |
1840 | new_info.update(format) | |
1841 | self.process_info(new_info) | |
1842 | # We update the info dict with the best quality format (backwards compatibility) | |
1843 | info_dict.update(formats_to_download[-1]) | |
1844 | return info_dict | |
1845 | ||
98c70d6f | 1846 | def process_subtitles(self, video_id, normal_subtitles, automatic_captions): |
a504ced0 | 1847 | """Select the requested subtitles and their format""" |
98c70d6f JMF |
1848 | available_subs = {} |
1849 | if normal_subtitles and self.params.get('writesubtitles'): | |
1850 | available_subs.update(normal_subtitles) | |
1851 | if automatic_captions and self.params.get('writeautomaticsub'): | |
1852 | for lang, cap_info in automatic_captions.items(): | |
360e1ca5 JMF |
1853 | if lang not in available_subs: |
1854 | available_subs[lang] = cap_info | |
1855 | ||
4d171848 JMF |
1856 | if (not self.params.get('writesubtitles') and not |
1857 | self.params.get('writeautomaticsub') or not | |
1858 | available_subs): | |
1859 | return None | |
a504ced0 JMF |
1860 | |
1861 | if self.params.get('allsubtitles', False): | |
1862 | requested_langs = available_subs.keys() | |
1863 | else: | |
1864 | if self.params.get('subtitleslangs', False): | |
1865 | requested_langs = self.params.get('subtitleslangs') | |
1866 | elif 'en' in available_subs: | |
1867 | requested_langs = ['en'] | |
1868 | else: | |
1869 | requested_langs = [list(available_subs.keys())[0]] | |
1870 | ||
1871 | formats_query = self.params.get('subtitlesformat', 'best') | |
1872 | formats_preference = formats_query.split('/') if formats_query else [] | |
1873 | subs = {} | |
1874 | for lang in requested_langs: | |
1875 | formats = available_subs.get(lang) | |
1876 | if formats is None: | |
1877 | self.report_warning('%s subtitles not available for %s' % (lang, video_id)) | |
1878 | continue | |
a504ced0 JMF |
1879 | for ext in formats_preference: |
1880 | if ext == 'best': | |
1881 | f = formats[-1] | |
1882 | break | |
1883 | matches = list(filter(lambda f: f['ext'] == ext, formats)) | |
1884 | if matches: | |
1885 | f = matches[-1] | |
1886 | break | |
1887 | else: | |
1888 | f = formats[-1] | |
1889 | self.report_warning( | |
1890 | 'No subtitle format found matching "%s" for language %s, ' | |
1891 | 'using %s' % (formats_query, lang, f['ext'])) | |
1892 | subs[lang] = f | |
1893 | return subs | |
1894 | ||
d06daf23 S |
1895 | def __forced_printings(self, info_dict, filename, incomplete): |
1896 | def print_mandatory(field): | |
1897 | if (self.params.get('force%s' % field, False) | |
1898 | and (not incomplete or info_dict.get(field) is not None)): | |
1899 | self.to_stdout(info_dict[field]) | |
1900 | ||
1901 | def print_optional(field): | |
1902 | if (self.params.get('force%s' % field, False) | |
1903 | and info_dict.get(field) is not None): | |
1904 | self.to_stdout(info_dict[field]) | |
1905 | ||
1906 | print_mandatory('title') | |
1907 | print_mandatory('id') | |
1908 | if self.params.get('forceurl', False) and not incomplete: | |
1909 | if info_dict.get('requested_formats') is not None: | |
1910 | for f in info_dict['requested_formats']: | |
1911 | self.to_stdout(f['url'] + f.get('play_path', '')) | |
1912 | else: | |
1913 | # For RTMP URLs, also include the playpath | |
1914 | self.to_stdout(info_dict['url'] + info_dict.get('play_path', '')) | |
1915 | print_optional('thumbnail') | |
1916 | print_optional('description') | |
1917 | if self.params.get('forcefilename', False) and filename is not None: | |
1918 | self.to_stdout(filename) | |
1919 | if self.params.get('forceduration', False) and info_dict.get('duration') is not None: | |
1920 | self.to_stdout(formatSeconds(info_dict['duration'])) | |
1921 | print_mandatory('format') | |
1922 | if self.params.get('forcejson', False): | |
1923 | self.to_stdout(json.dumps(info_dict)) | |
1924 | ||
8222d8de JMF |
1925 | def process_info(self, info_dict): |
1926 | """Process a single resolved IE result.""" | |
1927 | ||
1928 | assert info_dict.get('_type', 'video') == 'video' | |
fd288278 | 1929 | |
0202b52a | 1930 | info_dict.setdefault('__postprocessors', []) |
1931 | ||
fd288278 PH |
1932 | max_downloads = self.params.get('max_downloads') |
1933 | if max_downloads is not None: | |
1934 | if self._num_downloads >= int(max_downloads): | |
1935 | raise MaxDownloadsReached() | |
8222d8de | 1936 | |
d06daf23 | 1937 | # TODO: backward compatibility, to be removed |
8222d8de | 1938 | info_dict['fulltitle'] = info_dict['title'] |
8222d8de | 1939 | |
11b85ce6 | 1940 | if 'format' not in info_dict: |
8222d8de JMF |
1941 | info_dict['format'] = info_dict['ext'] |
1942 | ||
8b0d7497 | 1943 | if self._match_entry(info_dict, incomplete=False) is not None: |
8222d8de JMF |
1944 | return |
1945 | ||
fd288278 | 1946 | self._num_downloads += 1 |
8222d8de | 1947 | |
5bfa4862 | 1948 | info_dict = self.pre_process(info_dict) |
1949 | ||
0202b52a | 1950 | filename = self.prepare_filename(info_dict, warn=True) |
1951 | info_dict['_filename'] = full_filename = self.prepare_filepath(filename) | |
1952 | temp_filename = self.prepare_filepath(filename, 'temp') | |
1953 | files_to_move = {} | |
8222d8de JMF |
1954 | |
1955 | # Forced printings | |
0202b52a | 1956 | self.__forced_printings(info_dict, full_filename, incomplete=False) |
8222d8de | 1957 | |
8222d8de | 1958 | if self.params.get('simulate', False): |
2d30509f | 1959 | if self.params.get('force_write_download_archive', False): |
1960 | self.record_download_archive(info_dict) | |
1961 | ||
1962 | # Do nothing else if in simulate mode | |
8222d8de JMF |
1963 | return |
1964 | ||
1965 | if filename is None: | |
1966 | return | |
1967 | ||
c5c9bf0c | 1968 | def ensure_dir_exists(path): |
0202b52a | 1969 | return make_dir(path, self.report_error) |
c5c9bf0c | 1970 | |
0202b52a | 1971 | if not ensure_dir_exists(encodeFilename(full_filename)): |
1972 | return | |
1973 | if not ensure_dir_exists(encodeFilename(temp_filename)): | |
8222d8de JMF |
1974 | return |
1975 | ||
1976 | if self.params.get('writedescription', False): | |
0202b52a | 1977 | descfn = replace_extension( |
1978 | self.prepare_filepath(filename, 'description'), | |
1979 | 'description', info_dict.get('ext')) | |
1980 | if not ensure_dir_exists(encodeFilename(descfn)): | |
1981 | return | |
0c3d0f51 | 1982 | if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(descfn)): |
6febd1c1 | 1983 | self.to_screen('[info] Video description is already present') |
f00fd51d JMF |
1984 | elif info_dict.get('description') is None: |
1985 | self.report_warning('There\'s no description to write.') | |
7b6fefc9 PH |
1986 | else: |
1987 | try: | |
6febd1c1 | 1988 | self.to_screen('[info] Writing video description to: ' + descfn) |
7b6fefc9 PH |
1989 | with io.open(encodeFilename(descfn), 'w', encoding='utf-8') as descfile: |
1990 | descfile.write(info_dict['description']) | |
7b6fefc9 | 1991 | except (OSError, IOError): |
6febd1c1 | 1992 | self.report_error('Cannot write description file ' + descfn) |
7b6fefc9 | 1993 | return |
8222d8de | 1994 | |
1fb07d10 | 1995 | if self.params.get('writeannotations', False): |
0202b52a | 1996 | annofn = replace_extension( |
1997 | self.prepare_filepath(filename, 'annotation'), | |
1998 | 'annotations.xml', info_dict.get('ext')) | |
1999 | if not ensure_dir_exists(encodeFilename(annofn)): | |
2000 | return | |
0c3d0f51 | 2001 | if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(annofn)): |
6febd1c1 | 2002 | self.to_screen('[info] Video annotations are already present') |
ffddb112 RA |
2003 | elif not info_dict.get('annotations'): |
2004 | self.report_warning('There are no annotations to write.') | |
7b6fefc9 PH |
2005 | else: |
2006 | try: | |
6febd1c1 | 2007 | self.to_screen('[info] Writing video annotations to: ' + annofn) |
7b6fefc9 PH |
2008 | with io.open(encodeFilename(annofn), 'w', encoding='utf-8') as annofile: |
2009 | annofile.write(info_dict['annotations']) | |
2010 | except (KeyError, TypeError): | |
6febd1c1 | 2011 | self.report_warning('There are no annotations to write.') |
7b6fefc9 | 2012 | except (OSError, IOError): |
6febd1c1 | 2013 | self.report_error('Cannot write annotations file: ' + annofn) |
7b6fefc9 | 2014 | return |
1fb07d10 | 2015 | |
9f448fcb | 2016 | def dl(name, info, subtitle=False): |
98b69821 | 2017 | fd = get_suitable_downloader(info, self.params)(self, self.params) |
2018 | for ph in self._progress_hooks: | |
2019 | fd.add_progress_hook(ph) | |
2020 | if self.params.get('verbose'): | |
29f7c58a | 2021 | self.to_screen('[debug] Invoking downloader on %r' % info.get('url')) |
9f448fcb | 2022 | return fd.download(name, info, subtitle) |
98b69821 | 2023 | |
c4a91be7 | 2024 | subtitles_are_requested = any([self.params.get('writesubtitles', False), |
0b7f3118 | 2025 | self.params.get('writeautomaticsub')]) |
c4a91be7 | 2026 | |
c84dd8a9 | 2027 | if subtitles_are_requested and info_dict.get('requested_subtitles'): |
8222d8de JMF |
2028 | # subtitles download errors are already managed as troubles in relevant IE |
2029 | # that way it will silently go on when used with unsupporting IE | |
c84dd8a9 | 2030 | subtitles = info_dict['requested_subtitles'] |
fa57af1e | 2031 | # ie = self.get_info_extractor(info_dict['extractor_key']) |
a504ced0 JMF |
2032 | for sub_lang, sub_info in subtitles.items(): |
2033 | sub_format = sub_info['ext'] | |
0202b52a | 2034 | sub_filename = subtitles_filename(temp_filename, sub_lang, sub_format, info_dict.get('ext')) |
2035 | sub_filename_final = subtitles_filename( | |
2036 | self.prepare_filepath(filename, 'subtitle'), | |
2037 | sub_lang, sub_format, info_dict.get('ext')) | |
0c3d0f51 | 2038 | if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(sub_filename)): |
5ff1bc0c | 2039 | self.to_screen('[info] Video subtitle %s.%s is already present' % (sub_lang, sub_format)) |
0202b52a | 2040 | files_to_move[sub_filename] = sub_filename_final |
a504ced0 | 2041 | else: |
0c9df79e | 2042 | self.to_screen('[info] Writing video subtitles to: ' + sub_filename) |
5ff1bc0c RA |
2043 | if sub_info.get('data') is not None: |
2044 | try: | |
2045 | # Use newline='' to prevent conversion of newline characters | |
067aa17e | 2046 | # See https://github.com/ytdl-org/youtube-dl/issues/10268 |
5ff1bc0c RA |
2047 | with io.open(encodeFilename(sub_filename), 'w', encoding='utf-8', newline='') as subfile: |
2048 | subfile.write(sub_info['data']) | |
0202b52a | 2049 | files_to_move[sub_filename] = sub_filename_final |
5ff1bc0c RA |
2050 | except (OSError, IOError): |
2051 | self.report_error('Cannot write subtitles file ' + sub_filename) | |
2052 | return | |
7b6fefc9 | 2053 | else: |
5ff1bc0c | 2054 | try: |
9f448fcb U |
2055 | dl(sub_filename, sub_info, subtitle=True) |
2056 | ''' | |
0c9df79e U |
2057 | if self.params.get('sleep_interval_subtitles', False): |
2058 | dl(sub_filename, sub_info) | |
2059 | else: | |
2060 | sub_data = ie._request_webpage( | |
2061 | sub_info['url'], info_dict['id'], note=False).read() | |
2062 | with io.open(encodeFilename(sub_filename), 'wb') as subfile: | |
2063 | subfile.write(sub_data) | |
9f448fcb | 2064 | ''' |
0202b52a | 2065 | files_to_move[sub_filename] = sub_filename_final |
0c9df79e | 2066 | except (ExtractorError, IOError, OSError, ValueError, compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: |
5ff1bc0c RA |
2067 | self.report_warning('Unable to download subtitle for "%s": %s' % |
2068 | (sub_lang, error_to_compat_str(err))) | |
2069 | continue | |
8222d8de | 2070 | |
57df9f53 U |
2071 | if self.params.get('skip_download', False): |
2072 | if self.params.get('convertsubtitles', False): | |
0202b52a | 2073 | # subconv = FFmpegSubtitlesConvertorPP(self, format=self.params.get('convertsubtitles')) |
57df9f53 U |
2074 | filename_real_ext = os.path.splitext(filename)[1][1:] |
2075 | filename_wo_ext = ( | |
0202b52a | 2076 | os.path.splitext(full_filename)[0] |
57df9f53 | 2077 | if filename_real_ext == info_dict['ext'] |
0202b52a | 2078 | else full_filename) |
57df9f53 | 2079 | afilename = '%s.%s' % (filename_wo_ext, self.params.get('convertsubtitles')) |
0202b52a | 2080 | # if subconv.available: |
2081 | # info_dict['__postprocessors'].append(subconv) | |
57df9f53 | 2082 | if os.path.exists(encodeFilename(afilename)): |
f791b419 U |
2083 | self.to_screen( |
2084 | '[download] %s has already been downloaded and ' | |
2085 | 'converted' % afilename) | |
57df9f53 U |
2086 | else: |
2087 | try: | |
0202b52a | 2088 | self.post_process(full_filename, info_dict, files_to_move) |
57df9f53 U |
2089 | except (PostProcessingError) as err: |
2090 | self.report_error('postprocessing: %s' % str(err)) | |
2091 | return | |
2092 | ||
8222d8de | 2093 | if self.params.get('writeinfojson', False): |
0202b52a | 2094 | infofn = replace_extension( |
2095 | self.prepare_filepath(filename, 'infojson'), | |
2096 | 'info.json', info_dict.get('ext')) | |
2097 | if not ensure_dir_exists(encodeFilename(infofn)): | |
2098 | return | |
0c3d0f51 | 2099 | if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(infofn)): |
6febd1c1 | 2100 | self.to_screen('[info] Video description metadata is already present') |
7b6fefc9 | 2101 | else: |
6febd1c1 | 2102 | self.to_screen('[info] Writing video description metadata as JSON to: ' + infofn) |
7b6fefc9 | 2103 | try: |
cb202fd2 | 2104 | write_json_file(self.filter_requested_info(info_dict), infofn) |
7b6fefc9 | 2105 | except (OSError, IOError): |
6febd1c1 | 2106 | self.report_error('Cannot write metadata to JSON file ' + infofn) |
7b6fefc9 | 2107 | return |
8222d8de | 2108 | |
0202b52a | 2109 | thumbdir = os.path.dirname(self.prepare_filepath(filename, 'thumbnail')) |
2110 | for thumbfn in self._write_thumbnails(info_dict, temp_filename): | |
2111 | files_to_move[thumbfn] = os.path.join(thumbdir, os.path.basename(thumbfn)) | |
8222d8de | 2112 | |
732044af | 2113 | # Write internet shortcut files |
2114 | url_link = webloc_link = desktop_link = False | |
2115 | if self.params.get('writelink', False): | |
2116 | if sys.platform == "darwin": # macOS. | |
2117 | webloc_link = True | |
2118 | elif sys.platform.startswith("linux"): | |
2119 | desktop_link = True | |
2120 | else: # if sys.platform in ['win32', 'cygwin']: | |
2121 | url_link = True | |
2122 | if self.params.get('writeurllink', False): | |
2123 | url_link = True | |
2124 | if self.params.get('writewebloclink', False): | |
2125 | webloc_link = True | |
2126 | if self.params.get('writedesktoplink', False): | |
2127 | desktop_link = True | |
2128 | ||
2129 | if url_link or webloc_link or desktop_link: | |
2130 | if 'webpage_url' not in info_dict: | |
2131 | self.report_error('Cannot write internet shortcut file because the "webpage_url" field is missing in the media information') | |
2132 | return | |
2133 | ascii_url = iri_to_uri(info_dict['webpage_url']) | |
2134 | ||
2135 | def _write_link_file(extension, template, newline, embed_filename): | |
0202b52a | 2136 | linkfn = replace_extension(full_filename, extension, info_dict.get('ext')) |
10e3742e | 2137 | if self.params.get('overwrites', True) and os.path.exists(encodeFilename(linkfn)): |
732044af | 2138 | self.to_screen('[info] Internet shortcut is already present') |
2139 | else: | |
2140 | try: | |
2141 | self.to_screen('[info] Writing internet shortcut to: ' + linkfn) | |
2142 | with io.open(encodeFilename(to_high_limit_path(linkfn)), 'w', encoding='utf-8', newline=newline) as linkfile: | |
2143 | template_vars = {'url': ascii_url} | |
2144 | if embed_filename: | |
2145 | template_vars['filename'] = linkfn[:-(len(extension) + 1)] | |
2146 | linkfile.write(template % template_vars) | |
2147 | except (OSError, IOError): | |
2148 | self.report_error('Cannot write internet shortcut ' + linkfn) | |
2149 | return False | |
2150 | return True | |
2151 | ||
2152 | if url_link: | |
2153 | if not _write_link_file('url', DOT_URL_LINK_TEMPLATE, '\r\n', embed_filename=False): | |
2154 | return | |
2155 | if webloc_link: | |
2156 | if not _write_link_file('webloc', DOT_WEBLOC_LINK_TEMPLATE, '\n', embed_filename=False): | |
2157 | return | |
2158 | if desktop_link: | |
2159 | if not _write_link_file('desktop', DOT_DESKTOP_LINK_TEMPLATE, '\n', embed_filename=True): | |
2160 | return | |
2161 | ||
2162 | # Download | |
2163 | must_record_download_archive = False | |
8222d8de | 2164 | if not self.params.get('skip_download', False): |
4340deca | 2165 | try: |
0202b52a | 2166 | |
2167 | def existing_file(filename, temp_filename): | |
2168 | file_exists = os.path.exists(encodeFilename(filename)) | |
2169 | tempfile_exists = ( | |
2170 | False if temp_filename == filename | |
2171 | else os.path.exists(encodeFilename(temp_filename))) | |
2172 | if not self.params.get('overwrites', False) and (file_exists or tempfile_exists): | |
2173 | existing_filename = temp_filename if tempfile_exists else filename | |
2174 | self.to_screen('[download] %s has already been downloaded and merged' % existing_filename) | |
2175 | return existing_filename | |
2176 | if tempfile_exists: | |
2177 | self.report_file_delete(temp_filename) | |
2178 | os.remove(encodeFilename(temp_filename)) | |
2179 | if file_exists: | |
2180 | self.report_file_delete(filename) | |
2181 | os.remove(encodeFilename(filename)) | |
2182 | return None | |
2183 | ||
2184 | success = True | |
4340deca P |
2185 | if info_dict.get('requested_formats') is not None: |
2186 | downloaded = [] | |
d47aeb22 | 2187 | merger = FFmpegMergerPP(self) |
f740fae2 | 2188 | if not merger.available: |
4340deca P |
2189 | postprocessors = [] |
2190 | self.report_warning('You have requested multiple ' | |
e4172ac9 | 2191 | 'formats but ffmpeg is not installed.' |
4a5a898a | 2192 | ' The formats won\'t be merged.') |
6350728b | 2193 | else: |
4340deca | 2194 | postprocessors = [merger] |
81cd954a S |
2195 | |
2196 | def compatible_formats(formats): | |
d03cfdce | 2197 | # TODO: some formats actually allow this (mkv, webm, ogg, mp4), but not all of them. |
2198 | video_formats = [format for format in formats if format.get('vcodec') != 'none'] | |
2199 | audio_formats = [format for format in formats if format.get('acodec') != 'none'] | |
2200 | if len(video_formats) > 2 or len(audio_formats) > 2: | |
2201 | return False | |
2202 | ||
81cd954a | 2203 | # Check extension |
d03cfdce | 2204 | exts = set(format.get('ext') for format in formats) |
2205 | COMPATIBLE_EXTS = ( | |
2206 | set(('mp3', 'mp4', 'm4a', 'm4p', 'm4b', 'm4r', 'm4v', 'ismv', 'isma')), | |
2207 | set(('webm',)), | |
2208 | ) | |
2209 | for ext_sets in COMPATIBLE_EXTS: | |
2210 | if ext_sets.issuperset(exts): | |
2211 | return True | |
81cd954a S |
2212 | # TODO: Check acodec/vcodec |
2213 | return False | |
2214 | ||
2215 | requested_formats = info_dict['requested_formats'] | |
0202b52a | 2216 | old_ext = info_dict['ext'] |
c0dea0a7 | 2217 | if self.params.get('merge_output_format') is None and not compatible_formats(requested_formats): |
38c6902b | 2218 | info_dict['ext'] = 'mkv' |
4a5a898a S |
2219 | self.report_warning( |
2220 | 'Requested formats are incompatible for merge and will be merged into mkv.') | |
0202b52a | 2221 | |
2222 | def correct_ext(filename): | |
2223 | filename_real_ext = os.path.splitext(filename)[1][1:] | |
2224 | filename_wo_ext = ( | |
2225 | os.path.splitext(filename)[0] | |
2226 | if filename_real_ext == old_ext | |
2227 | else filename) | |
2228 | return '%s.%s' % (filename_wo_ext, info_dict['ext']) | |
2229 | ||
38c6902b | 2230 | # Ensure filename always has a correct extension for successful merge |
0202b52a | 2231 | full_filename = correct_ext(full_filename) |
2232 | temp_filename = correct_ext(temp_filename) | |
2233 | dl_filename = existing_file(full_filename, temp_filename) | |
2234 | if dl_filename is None: | |
81cd954a | 2235 | for f in requested_formats: |
5b5fbc08 JMF |
2236 | new_info = dict(info_dict) |
2237 | new_info.update(f) | |
c5c9bf0c | 2238 | fname = prepend_extension( |
0202b52a | 2239 | self.prepare_filepath(self.prepare_filename(new_info), 'temp'), |
c5c9bf0c S |
2240 | 'f%s' % f['format_id'], new_info['ext']) |
2241 | if not ensure_dir_exists(fname): | |
2242 | return | |
5b5fbc08 | 2243 | downloaded.append(fname) |
a9e7f546 | 2244 | partial_success, real_download = dl(fname, new_info) |
5b5fbc08 JMF |
2245 | success = success and partial_success |
2246 | info_dict['__postprocessors'] = postprocessors | |
2247 | info_dict['__files_to_merge'] = downloaded | |
a9e7f546 | 2248 | # Even if there were no downloads, it is being merged only now |
2249 | info_dict['__real_download'] = True | |
4340deca P |
2250 | else: |
2251 | # Just a single file | |
0202b52a | 2252 | dl_filename = existing_file(full_filename, temp_filename) |
2253 | if dl_filename is None: | |
2254 | success, real_download = dl(temp_filename, info_dict) | |
2255 | info_dict['__real_download'] = real_download | |
2256 | ||
0202b52a | 2257 | dl_filename = dl_filename or temp_filename |
c571435f | 2258 | info_dict['__finaldir'] = os.path.dirname(os.path.abspath(encodeFilename(full_filename))) |
0202b52a | 2259 | |
4340deca | 2260 | except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: |
7960b056 | 2261 | self.report_error('unable to download video data: %s' % error_to_compat_str(err)) |
4340deca P |
2262 | return |
2263 | except (OSError, IOError) as err: | |
2264 | raise UnavailableVideoError(err) | |
2265 | except (ContentTooShortError, ) as err: | |
2266 | self.report_error('content too short (expected %s bytes and served %s)' % (err.expected, err.downloaded)) | |
2267 | return | |
8222d8de | 2268 | |
e38cafe9 | 2269 | if success and filename != '-': |
6271f1ca | 2270 | # Fixup content |
62cd676c PH |
2271 | fixup_policy = self.params.get('fixup') |
2272 | if fixup_policy is None: | |
2273 | fixup_policy = 'detect_or_warn' | |
2274 | ||
e4172ac9 | 2275 | INSTALL_FFMPEG_MESSAGE = 'Install ffmpeg to fix this automatically.' |
d1e4a464 | 2276 | |
6271f1ca PH |
2277 | stretched_ratio = info_dict.get('stretched_ratio') |
2278 | if stretched_ratio is not None and stretched_ratio != 1: | |
6271f1ca PH |
2279 | if fixup_policy == 'warn': |
2280 | self.report_warning('%s: Non-uniform pixel ratio (%s)' % ( | |
2281 | info_dict['id'], stretched_ratio)) | |
2282 | elif fixup_policy == 'detect_or_warn': | |
2283 | stretched_pp = FFmpegFixupStretchedPP(self) | |
2284 | if stretched_pp.available: | |
6271f1ca PH |
2285 | info_dict['__postprocessors'].append(stretched_pp) |
2286 | else: | |
2287 | self.report_warning( | |
d1e4a464 S |
2288 | '%s: Non-uniform pixel ratio (%s). %s' |
2289 | % (info_dict['id'], stretched_ratio, INSTALL_FFMPEG_MESSAGE)) | |
6271f1ca | 2290 | else: |
62cd676c PH |
2291 | assert fixup_policy in ('ignore', 'never') |
2292 | ||
3089bc74 S |
2293 | if (info_dict.get('requested_formats') is None |
2294 | and info_dict.get('container') == 'm4a_dash'): | |
62cd676c | 2295 | if fixup_policy == 'warn': |
d1e4a464 S |
2296 | self.report_warning( |
2297 | '%s: writing DASH m4a. ' | |
2298 | 'Only some players support this container.' | |
2299 | % info_dict['id']) | |
62cd676c PH |
2300 | elif fixup_policy == 'detect_or_warn': |
2301 | fixup_pp = FFmpegFixupM4aPP(self) | |
2302 | if fixup_pp.available: | |
62cd676c PH |
2303 | info_dict['__postprocessors'].append(fixup_pp) |
2304 | else: | |
2305 | self.report_warning( | |
d1e4a464 S |
2306 | '%s: writing DASH m4a. ' |
2307 | 'Only some players support this container. %s' | |
2308 | % (info_dict['id'], INSTALL_FFMPEG_MESSAGE)) | |
62cd676c PH |
2309 | else: |
2310 | assert fixup_policy in ('ignore', 'never') | |
6271f1ca | 2311 | |
3089bc74 S |
2312 | if (info_dict.get('protocol') == 'm3u8_native' |
2313 | or info_dict.get('protocol') == 'm3u8' | |
2314 | and self.params.get('hls_prefer_native')): | |
f17f8651 | 2315 | if fixup_policy == 'warn': |
a02682fd | 2316 | self.report_warning('%s: malformed AAC bitstream detected.' % ( |
f17f8651 | 2317 | info_dict['id'])) |
2318 | elif fixup_policy == 'detect_or_warn': | |
2319 | fixup_pp = FFmpegFixupM3u8PP(self) | |
2320 | if fixup_pp.available: | |
f17f8651 | 2321 | info_dict['__postprocessors'].append(fixup_pp) |
2322 | else: | |
2323 | self.report_warning( | |
a02682fd | 2324 | '%s: malformed AAC bitstream detected. %s' |
d1e4a464 | 2325 | % (info_dict['id'], INSTALL_FFMPEG_MESSAGE)) |
f17f8651 | 2326 | else: |
2327 | assert fixup_policy in ('ignore', 'never') | |
2328 | ||
8222d8de | 2329 | try: |
0202b52a | 2330 | self.post_process(dl_filename, info_dict, files_to_move) |
8222d8de | 2331 | except (PostProcessingError) as err: |
6febd1c1 | 2332 | self.report_error('postprocessing: %s' % str(err)) |
8222d8de | 2333 | return |
ab8e5e51 AM |
2334 | try: |
2335 | for ph in self._post_hooks: | |
0202b52a | 2336 | ph(full_filename) |
ab8e5e51 AM |
2337 | except Exception as err: |
2338 | self.report_error('post hooks: %s' % str(err)) | |
2339 | return | |
2d30509f | 2340 | must_record_download_archive = True |
2341 | ||
2342 | if must_record_download_archive or self.params.get('force_write_download_archive', False): | |
2343 | self.record_download_archive(info_dict) | |
c3e6ffba | 2344 | max_downloads = self.params.get('max_downloads') |
2345 | if max_downloads is not None and self._num_downloads >= int(max_downloads): | |
2346 | raise MaxDownloadsReached() | |
8222d8de JMF |
2347 | |
2348 | def download(self, url_list): | |
2349 | """Download a given list of URLs.""" | |
acd69589 | 2350 | outtmpl = self.params.get('outtmpl', DEFAULT_OUTTMPL) |
3089bc74 S |
2351 | if (len(url_list) > 1 |
2352 | and outtmpl != '-' | |
2353 | and '%' not in outtmpl | |
2354 | and self.params.get('max_downloads') != 1): | |
acd69589 | 2355 | raise SameFileError(outtmpl) |
8222d8de JMF |
2356 | |
2357 | for url in url_list: | |
2358 | try: | |
5f6a1245 | 2359 | # It also downloads the videos |
61aa5ba3 S |
2360 | res = self.extract_info( |
2361 | url, force_generic_extractor=self.params.get('force_generic_extractor', False)) | |
8222d8de | 2362 | except UnavailableVideoError: |
6febd1c1 | 2363 | self.report_error('unable to download video') |
8222d8de | 2364 | except MaxDownloadsReached: |
8b0d7497 | 2365 | self.to_screen('[info] Maximum number of downloaded files reached') |
2366 | raise | |
2367 | except ExistingVideoReached: | |
d83cb531 | 2368 | self.to_screen('[info] Encountered a file that is already in the archive, stopping due to --break-on-existing') |
8b0d7497 | 2369 | raise |
2370 | except RejectedVideoReached: | |
d83cb531 | 2371 | self.to_screen('[info] Encountered a file that did not match filter, stopping due to --break-on-reject') |
8222d8de | 2372 | raise |
63e0be34 PH |
2373 | else: |
2374 | if self.params.get('dump_single_json', False): | |
2375 | self.to_stdout(json.dumps(res)) | |
8222d8de JMF |
2376 | |
2377 | return self._download_retcode | |
2378 | ||
1dcc4c0c | 2379 | def download_with_info_file(self, info_filename): |
31bd3925 JMF |
2380 | with contextlib.closing(fileinput.FileInput( |
2381 | [info_filename], mode='r', | |
2382 | openhook=fileinput.hook_encoded('utf-8'))) as f: | |
2383 | # FileInput doesn't have a read method, we can't call json.load | |
cb202fd2 | 2384 | info = self.filter_requested_info(json.loads('\n'.join(f))) |
d4943898 JMF |
2385 | try: |
2386 | self.process_ie_result(info, download=True) | |
2387 | except DownloadError: | |
2388 | webpage_url = info.get('webpage_url') | |
2389 | if webpage_url is not None: | |
6febd1c1 | 2390 | self.report_warning('The info failed to download, trying with "%s"' % webpage_url) |
d4943898 JMF |
2391 | return self.download([webpage_url]) |
2392 | else: | |
2393 | raise | |
2394 | return self._download_retcode | |
1dcc4c0c | 2395 | |
cb202fd2 S |
2396 | @staticmethod |
2397 | def filter_requested_info(info_dict): | |
2398 | return dict( | |
2399 | (k, v) for k, v in info_dict.items() | |
2400 | if k not in ['requested_formats', 'requested_subtitles']) | |
2401 | ||
5bfa4862 | 2402 | def run_pp(self, pp, infodict, files_to_move={}): |
2403 | files_to_delete = [] | |
2404 | try: | |
2405 | files_to_delete, infodict = pp.run(infodict) | |
2406 | except PostProcessingError as e: | |
2407 | self.report_error(e.msg) | |
2408 | if not files_to_delete: | |
2409 | return files_to_move, infodict | |
2410 | ||
2411 | if self.params.get('keepvideo', False): | |
2412 | for f in files_to_delete: | |
2413 | files_to_move.setdefault(f, '') | |
2414 | else: | |
2415 | for old_filename in set(files_to_delete): | |
2416 | self.to_screen('Deleting original file %s (pass -k to keep)' % old_filename) | |
2417 | try: | |
2418 | os.remove(encodeFilename(old_filename)) | |
2419 | except (IOError, OSError): | |
2420 | self.report_warning('Unable to remove downloaded original file') | |
2421 | if old_filename in files_to_move: | |
2422 | del files_to_move[old_filename] | |
2423 | return files_to_move, infodict | |
2424 | ||
2425 | def pre_process(self, ie_info): | |
2426 | info = dict(ie_info) | |
2427 | for pp in self._pps['beforedl']: | |
2428 | info = self.run_pp(pp, info)[1] | |
2429 | return info | |
2430 | ||
0202b52a | 2431 | def post_process(self, filename, ie_info, files_to_move={}): |
8222d8de JMF |
2432 | """Run all the postprocessors on the given file.""" |
2433 | info = dict(ie_info) | |
2434 | info['filepath'] = filename | |
0202b52a | 2435 | |
5bfa4862 | 2436 | for pp in ie_info.get('__postprocessors', []) + self._pps['normal']: |
2437 | files_to_move, info = self.run_pp(pp, info, files_to_move) | |
2438 | info = self.run_pp(MoveFilesAfterDownloadPP(self, files_to_move), info, files_to_move)[1] | |
2439 | for pp in self._pps['aftermove']: | |
2440 | files_to_move, info = self.run_pp(pp, info, {}) | |
c1c9a79c | 2441 | |
5db07df6 | 2442 | def _make_archive_id(self, info_dict): |
e9fef7ee S |
2443 | video_id = info_dict.get('id') |
2444 | if not video_id: | |
2445 | return | |
5db07df6 PH |
2446 | # Future-proof against any change in case |
2447 | # and backwards compatibility with prior versions | |
e9fef7ee | 2448 | extractor = info_dict.get('extractor_key') or info_dict.get('ie_key') # key in a playlist |
7012b23c | 2449 | if extractor is None: |
1211bb6d S |
2450 | url = str_or_none(info_dict.get('url')) |
2451 | if not url: | |
2452 | return | |
e9fef7ee S |
2453 | # Try to find matching extractor for the URL and take its ie_key |
2454 | for ie in self._ies: | |
1211bb6d | 2455 | if ie.suitable(url): |
e9fef7ee S |
2456 | extractor = ie.ie_key() |
2457 | break | |
2458 | else: | |
2459 | return | |
d0757229 | 2460 | return '%s %s' % (extractor.lower(), video_id) |
5db07df6 PH |
2461 | |
2462 | def in_download_archive(self, info_dict): | |
2463 | fn = self.params.get('download_archive') | |
2464 | if fn is None: | |
2465 | return False | |
2466 | ||
2467 | vid_id = self._make_archive_id(info_dict) | |
e9fef7ee | 2468 | if not vid_id: |
7012b23c | 2469 | return False # Incomplete video information |
5db07df6 | 2470 | |
a45e8619 | 2471 | return vid_id in self.archive |
c1c9a79c PH |
2472 | |
2473 | def record_download_archive(self, info_dict): | |
2474 | fn = self.params.get('download_archive') | |
2475 | if fn is None: | |
2476 | return | |
5db07df6 PH |
2477 | vid_id = self._make_archive_id(info_dict) |
2478 | assert vid_id | |
c1c9a79c | 2479 | with locked_file(fn, 'a', encoding='utf-8') as archive_file: |
6febd1c1 | 2480 | archive_file.write(vid_id + '\n') |
a45e8619 | 2481 | self.archive.add(vid_id) |
dd82ffea | 2482 | |
8c51aa65 | 2483 | @staticmethod |
8abeeb94 | 2484 | def format_resolution(format, default='unknown'): |
fb04e403 PH |
2485 | if format.get('vcodec') == 'none': |
2486 | return 'audio only' | |
f49d89ee PH |
2487 | if format.get('resolution') is not None: |
2488 | return format['resolution'] | |
8c51aa65 JMF |
2489 | if format.get('height') is not None: |
2490 | if format.get('width') is not None: | |
6febd1c1 | 2491 | res = '%sx%s' % (format['width'], format['height']) |
8c51aa65 | 2492 | else: |
6febd1c1 | 2493 | res = '%sp' % format['height'] |
f49d89ee | 2494 | elif format.get('width') is not None: |
388ae76b | 2495 | res = '%dx?' % format['width'] |
8c51aa65 | 2496 | else: |
8abeeb94 | 2497 | res = default |
8c51aa65 JMF |
2498 | return res |
2499 | ||
c57f7757 PH |
2500 | def _format_note(self, fdict): |
2501 | res = '' | |
2502 | if fdict.get('ext') in ['f4f', 'f4m']: | |
2503 | res += '(unsupported) ' | |
32f90364 PH |
2504 | if fdict.get('language'): |
2505 | if res: | |
2506 | res += ' ' | |
9016d76f | 2507 | res += '[%s] ' % fdict['language'] |
c57f7757 PH |
2508 | if fdict.get('format_note') is not None: |
2509 | res += fdict['format_note'] + ' ' | |
2510 | if fdict.get('tbr') is not None: | |
2511 | res += '%4dk ' % fdict['tbr'] | |
2512 | if fdict.get('container') is not None: | |
2513 | if res: | |
2514 | res += ', ' | |
2515 | res += '%s container' % fdict['container'] | |
3089bc74 S |
2516 | if (fdict.get('vcodec') is not None |
2517 | and fdict.get('vcodec') != 'none'): | |
c57f7757 PH |
2518 | if res: |
2519 | res += ', ' | |
2520 | res += fdict['vcodec'] | |
91c7271a | 2521 | if fdict.get('vbr') is not None: |
c57f7757 PH |
2522 | res += '@' |
2523 | elif fdict.get('vbr') is not None and fdict.get('abr') is not None: | |
2524 | res += 'video@' | |
2525 | if fdict.get('vbr') is not None: | |
2526 | res += '%4dk' % fdict['vbr'] | |
fbb21cf5 | 2527 | if fdict.get('fps') is not None: |
5d583bdf S |
2528 | if res: |
2529 | res += ', ' | |
2530 | res += '%sfps' % fdict['fps'] | |
c57f7757 PH |
2531 | if fdict.get('acodec') is not None: |
2532 | if res: | |
2533 | res += ', ' | |
2534 | if fdict['acodec'] == 'none': | |
2535 | res += 'video only' | |
2536 | else: | |
2537 | res += '%-5s' % fdict['acodec'] | |
2538 | elif fdict.get('abr') is not None: | |
2539 | if res: | |
2540 | res += ', ' | |
2541 | res += 'audio' | |
2542 | if fdict.get('abr') is not None: | |
2543 | res += '@%3dk' % fdict['abr'] | |
2544 | if fdict.get('asr') is not None: | |
2545 | res += ' (%5dHz)' % fdict['asr'] | |
2546 | if fdict.get('filesize') is not None: | |
2547 | if res: | |
2548 | res += ', ' | |
2549 | res += format_bytes(fdict['filesize']) | |
9732d77e PH |
2550 | elif fdict.get('filesize_approx') is not None: |
2551 | if res: | |
2552 | res += ', ' | |
2553 | res += '~' + format_bytes(fdict['filesize_approx']) | |
c57f7757 | 2554 | return res |
91c7271a | 2555 | |
76d321f6 | 2556 | def _format_note_table(self, f): |
2557 | def join_fields(*vargs): | |
2558 | return ', '.join((val for val in vargs if val != '')) | |
2559 | ||
2560 | return join_fields( | |
2561 | 'UNSUPPORTED' if f.get('ext') in ('f4f', 'f4m') else '', | |
2562 | format_field(f, 'language', '[%s]'), | |
2563 | format_field(f, 'format_note'), | |
2564 | format_field(f, 'container', ignore=(None, f.get('ext'))), | |
2565 | format_field(f, 'asr', '%5dHz')) | |
2566 | ||
c57f7757 | 2567 | def list_formats(self, info_dict): |
94badb25 | 2568 | formats = info_dict.get('formats', [info_dict]) |
76d321f6 | 2569 | new_format = self.params.get('listformats_table', False) |
2570 | if new_format: | |
2571 | table = [ | |
2572 | [ | |
2573 | format_field(f, 'format_id'), | |
2574 | format_field(f, 'ext'), | |
2575 | self.format_resolution(f), | |
2576 | format_field(f, 'fps', '%d'), | |
2577 | '|', | |
2578 | format_field(f, 'filesize', ' %s', func=format_bytes) + format_field(f, 'filesize_approx', '~%s', func=format_bytes), | |
2579 | format_field(f, 'tbr', '%4dk'), | |
2580 | f.get('protocol').replace('http_dash_segments', 'dash').replace("native", "n"), | |
2581 | '|', | |
2582 | format_field(f, 'vcodec', default='unknown').replace('none', ''), | |
2583 | format_field(f, 'vbr', '%4dk'), | |
2584 | format_field(f, 'acodec', default='unknown').replace('none', ''), | |
2585 | format_field(f, 'abr', '%3dk'), | |
2586 | format_field(f, 'asr', '%5dHz'), | |
2587 | self._format_note_table(f)] | |
2588 | for f in formats | |
2589 | if f.get('preference') is None or f['preference'] >= -1000] | |
2590 | header_line = ['ID', 'EXT', 'RESOLUTION', 'FPS', '|', ' FILESIZE', ' TBR', 'PROTO', | |
2591 | '|', 'VCODEC', ' VBR', 'ACODEC', ' ABR', ' ASR', 'NOTE'] | |
2592 | else: | |
2593 | table = [ | |
2594 | [ | |
2595 | format_field(f, 'format_id'), | |
2596 | format_field(f, 'ext'), | |
2597 | self.format_resolution(f), | |
2598 | self._format_note(f)] | |
2599 | for f in formats | |
2600 | if f.get('preference') is None or f['preference'] >= -1000] | |
2601 | header_line = ['format code', 'extension', 'resolution', 'note'] | |
57dd9a8f | 2602 | |
76d321f6 | 2603 | # if len(formats) > 1: |
2604 | # table[-1][-1] += (' ' if table[-1][-1] else '') + '(best)' | |
cfb56d1a | 2605 | self.to_screen( |
76d321f6 | 2606 | '[info] Available formats for %s:\n%s' % (info_dict['id'], render_table( |
2607 | header_line, | |
2608 | table, | |
2609 | delim=new_format, | |
2610 | extraGap=(0 if new_format else 1), | |
2611 | hideEmpty=new_format))) | |
cfb56d1a PH |
2612 | |
2613 | def list_thumbnails(self, info_dict): | |
2614 | thumbnails = info_dict.get('thumbnails') | |
2615 | if not thumbnails: | |
b7b72db9 | 2616 | self.to_screen('[info] No thumbnails present for %s' % info_dict['id']) |
2617 | return | |
cfb56d1a PH |
2618 | |
2619 | self.to_screen( | |
2620 | '[info] Thumbnails for %s:' % info_dict['id']) | |
2621 | self.to_screen(render_table( | |
2622 | ['ID', 'width', 'height', 'URL'], | |
2623 | [[t['id'], t.get('width', 'unknown'), t.get('height', 'unknown'), t['url']] for t in thumbnails])) | |
dca08720 | 2624 | |
360e1ca5 | 2625 | def list_subtitles(self, video_id, subtitles, name='subtitles'): |
a504ced0 | 2626 | if not subtitles: |
360e1ca5 | 2627 | self.to_screen('%s has no %s' % (video_id, name)) |
a504ced0 | 2628 | return |
a504ced0 | 2629 | self.to_screen( |
edab9dbf JMF |
2630 | 'Available %s for %s:' % (name, video_id)) |
2631 | self.to_screen(render_table( | |
2632 | ['Language', 'formats'], | |
2633 | [[lang, ', '.join(f['ext'] for f in reversed(formats))] | |
2634 | for lang, formats in subtitles.items()])) | |
a504ced0 | 2635 | |
dca08720 PH |
2636 | def urlopen(self, req): |
2637 | """ Start an HTTP download """ | |
82d8a8b6 | 2638 | if isinstance(req, compat_basestring): |
67dda517 | 2639 | req = sanitized_Request(req) |
19a41fc6 | 2640 | return self._opener.open(req, timeout=self._socket_timeout) |
dca08720 PH |
2641 | |
2642 | def print_debug_header(self): | |
2643 | if not self.params.get('verbose'): | |
2644 | return | |
62fec3b2 | 2645 | |
4192b51c | 2646 | if type('') is not compat_str: |
067aa17e | 2647 | # Python 2.6 on SLES11 SP1 (https://github.com/ytdl-org/youtube-dl/issues/3326) |
4192b51c PH |
2648 | self.report_warning( |
2649 | 'Your Python is broken! Update to a newer and supported version') | |
2650 | ||
c6afed48 PH |
2651 | stdout_encoding = getattr( |
2652 | sys.stdout, 'encoding', 'missing (%s)' % type(sys.stdout).__name__) | |
b0472057 | 2653 | encoding_str = ( |
734f90bb PH |
2654 | '[debug] Encodings: locale %s, fs %s, out %s, pref %s\n' % ( |
2655 | locale.getpreferredencoding(), | |
2656 | sys.getfilesystemencoding(), | |
c6afed48 | 2657 | stdout_encoding, |
b0472057 | 2658 | self.get_encoding())) |
4192b51c | 2659 | write_string(encoding_str, encoding=None) |
734f90bb | 2660 | |
f74980cb | 2661 | self._write_string('[debug] yt-dlp version %s\n' % __version__) |
e0986e31 | 2662 | if _LAZY_LOADER: |
f74980cb | 2663 | self._write_string('[debug] Lazy loading extractors enabled\n') |
2664 | if _PLUGIN_CLASSES: | |
2665 | self._write_string( | |
2666 | '[debug] Plugin Extractors: %s\n' % [ie.ie_key() for ie in _PLUGIN_CLASSES]) | |
dca08720 PH |
2667 | try: |
2668 | sp = subprocess.Popen( | |
2669 | ['git', 'rev-parse', '--short', 'HEAD'], | |
2670 | stdout=subprocess.PIPE, stderr=subprocess.PIPE, | |
2671 | cwd=os.path.dirname(os.path.abspath(__file__))) | |
f5b1bca9 | 2672 | out, err = process_communicate_or_kill(sp) |
dca08720 PH |
2673 | out = out.decode().strip() |
2674 | if re.match('[0-9a-f]+', out): | |
f74980cb | 2675 | self._write_string('[debug] Git HEAD: %s\n' % out) |
70a1165b | 2676 | except Exception: |
dca08720 PH |
2677 | try: |
2678 | sys.exc_clear() | |
70a1165b | 2679 | except Exception: |
dca08720 | 2680 | pass |
b300cda4 S |
2681 | |
2682 | def python_implementation(): | |
2683 | impl_name = platform.python_implementation() | |
2684 | if impl_name == 'PyPy' and hasattr(sys, 'pypy_version_info'): | |
2685 | return impl_name + ' version %d.%d.%d' % sys.pypy_version_info[:3] | |
2686 | return impl_name | |
2687 | ||
2688 | self._write_string('[debug] Python version %s (%s) - %s\n' % ( | |
2689 | platform.python_version(), python_implementation(), | |
2690 | platform_name())) | |
d28b5171 | 2691 | |
73fac4e9 | 2692 | exe_versions = FFmpegPostProcessor.get_versions(self) |
4c83c967 | 2693 | exe_versions['rtmpdump'] = rtmpdump_version() |
feee8d32 | 2694 | exe_versions['phantomjs'] = PhantomJSwrapper._version() |
d28b5171 PH |
2695 | exe_str = ', '.join( |
2696 | '%s %s' % (exe, v) | |
2697 | for exe, v in sorted(exe_versions.items()) | |
2698 | if v | |
2699 | ) | |
2700 | if not exe_str: | |
2701 | exe_str = 'none' | |
2702 | self._write_string('[debug] exe versions: %s\n' % exe_str) | |
dca08720 PH |
2703 | |
2704 | proxy_map = {} | |
2705 | for handler in self._opener.handlers: | |
2706 | if hasattr(handler, 'proxies'): | |
2707 | proxy_map.update(handler.proxies) | |
734f90bb | 2708 | self._write_string('[debug] Proxy map: ' + compat_str(proxy_map) + '\n') |
dca08720 | 2709 | |
58b1f00d PH |
2710 | if self.params.get('call_home', False): |
2711 | ipaddr = self.urlopen('https://yt-dl.org/ip').read().decode('utf-8') | |
2712 | self._write_string('[debug] Public IP address: %s\n' % ipaddr) | |
f5546c0b | 2713 | return |
58b1f00d PH |
2714 | latest_version = self.urlopen( |
2715 | 'https://yt-dl.org/latest/version').read().decode('utf-8') | |
2716 | if version_tuple(latest_version) > version_tuple(__version__): | |
2717 | self.report_warning( | |
2718 | 'You are using an outdated version (newest version: %s)! ' | |
2719 | 'See https://yt-dl.org/update if you need help updating.' % | |
2720 | latest_version) | |
2721 | ||
e344693b | 2722 | def _setup_opener(self): |
6ad14cab | 2723 | timeout_val = self.params.get('socket_timeout') |
19a41fc6 | 2724 | self._socket_timeout = 600 if timeout_val is None else float(timeout_val) |
6ad14cab | 2725 | |
dca08720 PH |
2726 | opts_cookiefile = self.params.get('cookiefile') |
2727 | opts_proxy = self.params.get('proxy') | |
2728 | ||
2729 | if opts_cookiefile is None: | |
2730 | self.cookiejar = compat_cookiejar.CookieJar() | |
2731 | else: | |
590bc6f6 | 2732 | opts_cookiefile = expand_path(opts_cookiefile) |
1bab3437 | 2733 | self.cookiejar = YoutubeDLCookieJar(opts_cookiefile) |
dca08720 | 2734 | if os.access(opts_cookiefile, os.R_OK): |
1d88b3e6 | 2735 | self.cookiejar.load(ignore_discard=True, ignore_expires=True) |
dca08720 | 2736 | |
6a3f4c3f | 2737 | cookie_processor = YoutubeDLCookieProcessor(self.cookiejar) |
dca08720 PH |
2738 | if opts_proxy is not None: |
2739 | if opts_proxy == '': | |
2740 | proxies = {} | |
2741 | else: | |
2742 | proxies = {'http': opts_proxy, 'https': opts_proxy} | |
2743 | else: | |
2744 | proxies = compat_urllib_request.getproxies() | |
067aa17e | 2745 | # Set HTTPS proxy to HTTP one if given (https://github.com/ytdl-org/youtube-dl/issues/805) |
dca08720 PH |
2746 | if 'http' in proxies and 'https' not in proxies: |
2747 | proxies['https'] = proxies['http'] | |
91410c9b | 2748 | proxy_handler = PerRequestProxyHandler(proxies) |
a0ddb8a2 PH |
2749 | |
2750 | debuglevel = 1 if self.params.get('debug_printtraffic') else 0 | |
be4a824d PH |
2751 | https_handler = make_HTTPS_handler(self.params, debuglevel=debuglevel) |
2752 | ydlh = YoutubeDLHandler(self.params, debuglevel=debuglevel) | |
fca6dba8 | 2753 | redirect_handler = YoutubeDLRedirectHandler() |
8b172c2e | 2754 | data_handler = compat_urllib_request_DataHandler() |
6240b0a2 JMF |
2755 | |
2756 | # When passing our own FileHandler instance, build_opener won't add the | |
2757 | # default FileHandler and allows us to disable the file protocol, which | |
2758 | # can be used for malicious purposes (see | |
067aa17e | 2759 | # https://github.com/ytdl-org/youtube-dl/issues/8227) |
6240b0a2 JMF |
2760 | file_handler = compat_urllib_request.FileHandler() |
2761 | ||
2762 | def file_open(*args, **kwargs): | |
cefecac1 | 2763 | raise compat_urllib_error.URLError('file:// scheme is explicitly disabled in youtube-dlc for security reasons') |
6240b0a2 JMF |
2764 | file_handler.file_open = file_open |
2765 | ||
2766 | opener = compat_urllib_request.build_opener( | |
fca6dba8 | 2767 | proxy_handler, https_handler, cookie_processor, ydlh, redirect_handler, data_handler, file_handler) |
2461f79d | 2768 | |
dca08720 PH |
2769 | # Delete the default user-agent header, which would otherwise apply in |
2770 | # cases where our custom HTTP handler doesn't come into play | |
067aa17e | 2771 | # (See https://github.com/ytdl-org/youtube-dl/issues/1309 for details) |
dca08720 PH |
2772 | opener.addheaders = [] |
2773 | self._opener = opener | |
62fec3b2 PH |
2774 | |
2775 | def encode(self, s): | |
2776 | if isinstance(s, bytes): | |
2777 | return s # Already encoded | |
2778 | ||
2779 | try: | |
2780 | return s.encode(self.get_encoding()) | |
2781 | except UnicodeEncodeError as err: | |
2782 | err.reason = err.reason + '. Check your system encoding configuration or use the --encoding option.' | |
2783 | raise | |
2784 | ||
2785 | def get_encoding(self): | |
2786 | encoding = self.params.get('encoding') | |
2787 | if encoding is None: | |
2788 | encoding = preferredencoding() | |
2789 | return encoding | |
ec82d85a PH |
2790 | |
2791 | def _write_thumbnails(self, info_dict, filename): | |
2792 | if self.params.get('writethumbnail', False): | |
2793 | thumbnails = info_dict.get('thumbnails') | |
2794 | if thumbnails: | |
2795 | thumbnails = [thumbnails[-1]] | |
2796 | elif self.params.get('write_all_thumbnails', False): | |
0202b52a | 2797 | thumbnails = info_dict.get('thumbnails') or [] |
ec82d85a | 2798 | else: |
0202b52a | 2799 | thumbnails = [] |
ec82d85a | 2800 | |
0202b52a | 2801 | ret = [] |
ec82d85a PH |
2802 | for t in thumbnails: |
2803 | thumb_ext = determine_ext(t['url'], 'jpg') | |
2804 | suffix = '_%s' % t['id'] if len(thumbnails) > 1 else '' | |
2805 | thumb_display_id = '%s ' % t['id'] if len(thumbnails) > 1 else '' | |
29f7c58a | 2806 | t['filename'] = thumb_filename = replace_extension(filename + suffix, thumb_ext, info_dict.get('ext')) |
ec82d85a | 2807 | |
0c3d0f51 | 2808 | if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(thumb_filename)): |
0202b52a | 2809 | ret.append(thumb_filename) |
ec82d85a PH |
2810 | self.to_screen('[%s] %s: Thumbnail %sis already present' % |
2811 | (info_dict['extractor'], info_dict['id'], thumb_display_id)) | |
2812 | else: | |
2813 | self.to_screen('[%s] %s: Downloading thumbnail %s...' % | |
2814 | (info_dict['extractor'], info_dict['id'], thumb_display_id)) | |
2815 | try: | |
2816 | uf = self.urlopen(t['url']) | |
d3d89c32 | 2817 | with open(encodeFilename(thumb_filename), 'wb') as thumbf: |
ec82d85a | 2818 | shutil.copyfileobj(uf, thumbf) |
0202b52a | 2819 | ret.append(thumb_filename) |
ec82d85a PH |
2820 | self.to_screen('[%s] %s: Writing thumbnail %sto: %s' % |
2821 | (info_dict['extractor'], info_dict['id'], thumb_display_id, thumb_filename)) | |
2822 | except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: | |
2823 | self.report_warning('Unable to download thumbnail "%s": %s' % | |
9b9c5355 | 2824 | (t['url'], error_to_compat_str(err))) |
0202b52a | 2825 | return ret |