]>
Commit | Line | Data |
---|---|---|
cc52de43 | 1 | #!/usr/bin/env python3 |
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 | 22 | import sys |
21cd8fae | 23 | import tempfile |
8222d8de | 24 | import time |
67134eab | 25 | import tokenize |
8222d8de | 26 | import traceback |
75822ca7 | 27 | import random |
8222d8de | 28 | |
961ea474 | 29 | from string import ascii_letters |
e5813e53 | 30 | from zipimport import zipimporter |
961ea474 | 31 | |
8c25f81b | 32 | from .compat import ( |
82d8a8b6 | 33 | compat_basestring, |
003c69a8 | 34 | compat_get_terminal_size, |
4f026faf | 35 | compat_kwargs, |
d0d9ade4 | 36 | compat_numeric_types, |
e9c0cdd3 | 37 | compat_os_name, |
edf65256 | 38 | compat_pycrypto_AES, |
7d1eb38a | 39 | compat_shlex_quote, |
ce02ed60 | 40 | compat_str, |
67134eab | 41 | compat_tokenize_tokenize, |
ce02ed60 PH |
42 | compat_urllib_error, |
43 | compat_urllib_request, | |
8b172c2e | 44 | compat_urllib_request_DataHandler, |
8c25f81b | 45 | ) |
982ee69a | 46 | from .cookies import load_cookies |
8c25f81b | 47 | from .utils import ( |
eedb7ba5 S |
48 | age_restricted, |
49 | args_to_str, | |
ce02ed60 PH |
50 | ContentTooShortError, |
51 | date_from_str, | |
52 | DateRange, | |
acd69589 | 53 | DEFAULT_OUTTMPL, |
ce02ed60 | 54 | determine_ext, |
b5559424 | 55 | determine_protocol, |
732044af | 56 | DOT_DESKTOP_LINK_TEMPLATE, |
57 | DOT_URL_LINK_TEMPLATE, | |
58 | DOT_WEBLOC_LINK_TEMPLATE, | |
ce02ed60 | 59 | DownloadError, |
c0384f22 | 60 | encode_compat_str, |
ce02ed60 | 61 | encodeFilename, |
498f5606 | 62 | EntryNotInPlaylist, |
a06916d9 | 63 | error_to_compat_str, |
8b0d7497 | 64 | ExistingVideoReached, |
590bc6f6 | 65 | expand_path, |
ce02ed60 | 66 | ExtractorError, |
e29663c6 | 67 | float_or_none, |
02dbf93f | 68 | format_bytes, |
76d321f6 | 69 | format_field, |
901130bb | 70 | STR_FORMAT_RE_TMPL, |
71 | STR_FORMAT_TYPES, | |
525ef922 | 72 | formatSeconds, |
773f291d | 73 | GeoRestrictedError, |
b0249bca | 74 | HEADRequest, |
c9969434 | 75 | int_or_none, |
732044af | 76 | iri_to_uri, |
773f291d | 77 | ISO3166Utils, |
56a8fb4f | 78 | LazyList, |
ce02ed60 | 79 | locked_file, |
0202b52a | 80 | make_dir, |
dca08720 | 81 | make_HTTPS_handler, |
ce02ed60 | 82 | MaxDownloadsReached, |
3158150c | 83 | network_exceptions, |
cd6fc19e | 84 | orderedSet, |
a06916d9 | 85 | OUTTMPL_TYPES, |
b7ab0590 | 86 | PagedList, |
083c9df9 | 87 | parse_filesize, |
91410c9b | 88 | PerRequestProxyHandler, |
dca08720 | 89 | platform_name, |
eedb7ba5 | 90 | PostProcessingError, |
ce02ed60 | 91 | preferredencoding, |
eedb7ba5 | 92 | prepend_extension, |
a06916d9 | 93 | process_communicate_or_kill, |
51fb4995 | 94 | register_socks_protocols, |
a06916d9 | 95 | RejectedVideoReached, |
cfb56d1a | 96 | render_table, |
eedb7ba5 | 97 | replace_extension, |
ce02ed60 PH |
98 | SameFileError, |
99 | sanitize_filename, | |
1bb5c511 | 100 | sanitize_path, |
dcf77cf1 | 101 | sanitize_url, |
67dda517 | 102 | sanitized_Request, |
e5660ee6 | 103 | std_headers, |
1211bb6d | 104 | str_or_none, |
e29663c6 | 105 | strftime_or_none, |
ce02ed60 | 106 | subtitles_filename, |
51d9739f | 107 | ThrottledDownload, |
732044af | 108 | to_high_limit_path, |
324ad820 | 109 | traverse_obj, |
6033d980 | 110 | try_get, |
ce02ed60 | 111 | UnavailableVideoError, |
29eb5174 | 112 | url_basename, |
7d1eb38a | 113 | variadic, |
58b1f00d | 114 | version_tuple, |
ce02ed60 PH |
115 | write_json_file, |
116 | write_string, | |
6a3f4c3f | 117 | YoutubeDLCookieProcessor, |
dca08720 | 118 | YoutubeDLHandler, |
fca6dba8 | 119 | YoutubeDLRedirectHandler, |
ce02ed60 | 120 | ) |
a0e07d31 | 121 | from .cache import Cache |
52a8a1e1 | 122 | from .extractor import ( |
123 | gen_extractor_classes, | |
124 | get_info_extractor, | |
125 | _LAZY_LOADER, | |
126 | _PLUGIN_CLASSES | |
127 | ) | |
4c54b89e | 128 | from .extractor.openload import PhantomJSwrapper |
52a8a1e1 | 129 | from .downloader import ( |
dbf5416a | 130 | FFmpegFD, |
52a8a1e1 | 131 | get_suitable_downloader, |
132 | shorten_protocol_name | |
133 | ) | |
4c83c967 | 134 | from .downloader.rtmp import rtmpdump_version |
4f026faf | 135 | from .postprocessor import ( |
e36d50c5 | 136 | get_postprocessor, |
137 | FFmpegFixupDurationPP, | |
f17f8651 | 138 | FFmpegFixupM3u8PP, |
62cd676c | 139 | FFmpegFixupM4aPP, |
6271f1ca | 140 | FFmpegFixupStretchedPP, |
e36d50c5 | 141 | FFmpegFixupTimestampPP, |
4f026faf PH |
142 | FFmpegMergerPP, |
143 | FFmpegPostProcessor, | |
0202b52a | 144 | MoveFilesAfterDownloadPP, |
4f026faf | 145 | ) |
dca08720 | 146 | from .version import __version__ |
8222d8de | 147 | |
e9c0cdd3 YCH |
148 | if compat_os_name == 'nt': |
149 | import ctypes | |
150 | ||
2459b6e1 | 151 | |
8222d8de JMF |
152 | class YoutubeDL(object): |
153 | """YoutubeDL class. | |
154 | ||
155 | YoutubeDL objects are the ones responsible of downloading the | |
156 | actual video file and writing it to disk if the user has requested | |
157 | it, among some other tasks. In most cases there should be one per | |
158 | program. As, given a video URL, the downloader doesn't know how to | |
159 | extract all the needed information, task that InfoExtractors do, it | |
160 | has to pass the URL to one of them. | |
161 | ||
162 | For this, YoutubeDL objects have a method that allows | |
163 | InfoExtractors to be registered in a given order. When it is passed | |
164 | a URL, the YoutubeDL object handles it to the first InfoExtractor it | |
165 | finds that reports being able to handle it. The InfoExtractor extracts | |
166 | all the information about the video or videos the URL refers to, and | |
167 | YoutubeDL process the extracted information, possibly using a File | |
168 | Downloader to download the video. | |
169 | ||
170 | YoutubeDL objects accept a lot of parameters. In order not to saturate | |
171 | the object constructor with arguments, it receives a dictionary of | |
172 | options instead. These options are available through the params | |
173 | attribute for the InfoExtractors to use. The YoutubeDL also | |
174 | registers itself as the downloader in charge for the InfoExtractors | |
175 | that are added to it, so this is a "mutual registration". | |
176 | ||
177 | Available options: | |
178 | ||
179 | username: Username for authentication purposes. | |
180 | password: Password for authentication purposes. | |
180940e0 | 181 | videopassword: Password for accessing a video. |
1da50aa3 S |
182 | ap_mso: Adobe Pass multiple-system operator identifier. |
183 | ap_username: Multiple-system operator account username. | |
184 | ap_password: Multiple-system operator account password. | |
8222d8de JMF |
185 | usenetrc: Use netrc for authentication instead. |
186 | verbose: Print additional info to stdout. | |
187 | quiet: Do not print messages to stdout. | |
ad8915b7 | 188 | no_warnings: Do not print out anything for warnings. |
53c18592 | 189 | forceprint: A list of templates to force print |
190 | forceurl: Force printing final URL. (Deprecated) | |
191 | forcetitle: Force printing title. (Deprecated) | |
192 | forceid: Force printing ID. (Deprecated) | |
193 | forcethumbnail: Force printing thumbnail URL. (Deprecated) | |
194 | forcedescription: Force printing description. (Deprecated) | |
195 | forcefilename: Force printing final filename. (Deprecated) | |
196 | forceduration: Force printing duration. (Deprecated) | |
8694c600 | 197 | forcejson: Force printing info_dict as JSON. |
63e0be34 PH |
198 | dump_single_json: Force printing the info_dict of the whole playlist |
199 | (or video) as a single JSON line. | |
c25228e5 | 200 | force_write_download_archive: Force writing download archive regardless |
201 | of 'skip_download' or 'simulate'. | |
b7b04c78 | 202 | simulate: Do not download the video files. If unset (or None), |
203 | simulate only if listsubtitles, listformats or list_thumbnails is used | |
eb8a4433 | 204 | format: Video format code. see "FORMAT SELECTION" for more details. |
63ad4d43 | 205 | allow_unplayable_formats: Allow unplayable formats to be extracted and downloaded. |
b7da73eb | 206 | ignore_no_formats_error: Ignore "No video formats" error. Usefull for |
207 | extracting metadata even if the video is not actually | |
208 | available for download (experimental) | |
c25228e5 | 209 | format_sort: How to sort the video formats. see "Sorting Formats" |
210 | for more details. | |
211 | format_sort_force: Force the given format_sort. see "Sorting Formats" | |
212 | for more details. | |
213 | allow_multiple_video_streams: Allow multiple video streams to be merged | |
214 | into a single file | |
215 | allow_multiple_audio_streams: Allow multiple audio streams to be merged | |
216 | into a single file | |
0ba692ac | 217 | check_formats Whether to test if the formats are downloadable. |
218 | Can be True (check all), False (check none) | |
219 | or None (check only if requested by extractor) | |
4524baf0 | 220 | paths: Dictionary of output paths. The allowed keys are 'home' |
221 | 'temp' and the keys of OUTTMPL_TYPES (in utils.py) | |
de6000d9 | 222 | outtmpl: Dictionary of templates for output names. Allowed keys |
4524baf0 | 223 | are 'default' and the keys of OUTTMPL_TYPES (in utils.py). |
34488702 | 224 | For compatibility with youtube-dl, a single string can also be used |
a820dc72 RA |
225 | outtmpl_na_placeholder: Placeholder for unavailable meta fields. |
226 | restrictfilenames: Do not allow "&" and spaces in file names | |
227 | trim_file_name: Limit length of filename (extension excluded) | |
4524baf0 | 228 | windowsfilenames: Force the filenames to be windows compatible |
b1940459 | 229 | ignoreerrors: Do not stop on download/postprocessing errors. |
230 | Can be 'only_download' to ignore only download errors. | |
231 | Default is 'only_download' for CLI, but False for API | |
26e2805c | 232 | skip_playlist_after_errors: Number of allowed failures until the rest of |
233 | the playlist is skipped | |
d22dec74 | 234 | force_generic_extractor: Force downloader to use the generic extractor |
0c3d0f51 | 235 | overwrites: Overwrite all video and metadata files if True, |
236 | overwrite only non-video files if None | |
237 | and don't overwrite any file if False | |
34488702 | 238 | For compatibility with youtube-dl, |
239 | "nooverwrites" may also be used instead | |
8222d8de JMF |
240 | playliststart: Playlist item to start at. |
241 | playlistend: Playlist item to end at. | |
c14e88f0 | 242 | playlist_items: Specific indices of playlist to download. |
ff815fe6 | 243 | playlistreverse: Download playlist items in reverse order. |
75822ca7 | 244 | playlistrandom: Download playlist items in random order. |
8222d8de JMF |
245 | matchtitle: Download only matching titles. |
246 | rejecttitle: Reject downloads for matching titles. | |
8bf9319e | 247 | logger: Log messages to a logging.Logger instance. |
8222d8de JMF |
248 | logtostderr: Log messages to stderr instead of stdout. |
249 | writedescription: Write the video description to a .description file | |
250 | writeinfojson: Write the video description to a .info.json file | |
75d43ca0 | 251 | clean_infojson: Remove private fields from the infojson |
34488702 | 252 | getcomments: Extract video comments. This will not be written to disk |
06167fbb | 253 | unless writeinfojson is also given |
1fb07d10 | 254 | writeannotations: Write the video annotations to a .annotations.xml file |
8222d8de | 255 | writethumbnail: Write the thumbnail image to a file |
c25228e5 | 256 | allow_playlist_files: Whether to write playlists' description, infojson etc |
257 | also to disk when using the 'write*' options | |
ec82d85a | 258 | write_all_thumbnails: Write all thumbnail formats to files |
732044af | 259 | writelink: Write an internet shortcut file, depending on the |
260 | current platform (.url/.webloc/.desktop) | |
261 | writeurllink: Write a Windows internet shortcut file (.url) | |
262 | writewebloclink: Write a macOS internet shortcut file (.webloc) | |
263 | writedesktoplink: Write a Linux internet shortcut file (.desktop) | |
8222d8de | 264 | writesubtitles: Write the video subtitles to a file |
741dd8ea | 265 | writeautomaticsub: Write the automatically generated subtitles to a file |
245524e6 | 266 | allsubtitles: Deprecated - Use subtitleslangs = ['all'] |
c32b0aab | 267 | Downloads all the subtitles of the video |
0b7f3118 | 268 | (requires writesubtitles or writeautomaticsub) |
8222d8de | 269 | listsubtitles: Lists all available subtitles for the video |
a504ced0 | 270 | subtitlesformat: The format code for subtitles |
c32b0aab | 271 | subtitleslangs: List of languages of the subtitles to download (can be regex). |
272 | The list may contain "all" to refer to all the available | |
273 | subtitles. The language can be prefixed with a "-" to | |
274 | exclude it from the requested languages. Eg: ['all', '-live_chat'] | |
8222d8de JMF |
275 | keepvideo: Keep the video file after post-processing |
276 | daterange: A DateRange object, download only if the upload_date is in the range. | |
277 | skip_download: Skip the actual download of the video file | |
c35f9e72 | 278 | cachedir: Location of the cache files in the filesystem. |
a0e07d31 | 279 | False to disable filesystem cache. |
47192f92 | 280 | noplaylist: Download single video instead of a playlist if in doubt. |
8dbe9899 PH |
281 | age_limit: An integer representing the user's age in years. |
282 | Unsuitable videos for the given age are skipped. | |
5fe18bdb PH |
283 | min_views: An integer representing the minimum view count the video |
284 | must have in order to not be skipped. | |
285 | Videos without view count information are always | |
286 | downloaded. None for no limit. | |
287 | max_views: An integer representing the maximum view count. | |
288 | Videos that are more popular than that are not | |
289 | downloaded. | |
290 | Videos without view count information are always | |
291 | downloaded. None for no limit. | |
292 | download_archive: File name of a file where all downloads are recorded. | |
c1c9a79c PH |
293 | Videos already present in the file are not downloaded |
294 | again. | |
8a51f564 | 295 | break_on_existing: Stop the download process after attempting to download a |
296 | file that is in the archive. | |
297 | break_on_reject: Stop the download process when encountering a video that | |
298 | has been filtered out. | |
299 | cookiefile: File name where cookies should be read from and dumped to | |
982ee69a MB |
300 | cookiesfrombrowser: A tuple containing the name of the browser and the profile |
301 | name/path from where cookies are loaded. | |
302 | Eg: ('chrome', ) or (vivaldi, 'default') | |
a1ee09e8 | 303 | nocheckcertificate:Do not verify SSL certificates |
7e8c0af0 PH |
304 | prefer_insecure: Use HTTP instead of HTTPS to retrieve information. |
305 | At the moment, this is only supported by YouTube. | |
a1ee09e8 | 306 | proxy: URL of the proxy server to use |
38cce791 | 307 | geo_verification_proxy: URL of the proxy to use for IP address verification |
504f20dd | 308 | on geo-restricted sites. |
e344693b | 309 | socket_timeout: Time to wait for unresponsive hosts, in seconds |
0783b09b PH |
310 | bidi_workaround: Work around buggy terminals without bidirectional text |
311 | support, using fridibi | |
a0ddb8a2 | 312 | debug_printtraffic:Print out sent and received HTTP traffic |
7b0817e8 | 313 | include_ads: Download ads as well |
04b4d394 PH |
314 | default_search: Prepend this string if an input url is not valid. |
315 | 'auto' for elaborate guessing | |
62fec3b2 | 316 | encoding: Use this encoding instead of the system-specified. |
e8ee972c | 317 | extract_flat: Do not resolve URLs, return the immediate result. |
057a5206 PH |
318 | Pass in 'in_playlist' to only show this behavior for |
319 | playlist items. | |
4f026faf | 320 | postprocessors: A list of dictionaries, each with an entry |
71b640cc | 321 | * key: The name of the postprocessor. See |
7a5c1cfe | 322 | yt_dlp/postprocessor/__init__.py for a list. |
56d868db | 323 | * when: When to run the postprocessor. Can be one of |
324 | pre_process|before_dl|post_process|after_move. | |
325 | Assumed to be 'post_process' if not given | |
ab8e5e51 AM |
326 | post_hooks: A list of functions that get called as the final step |
327 | for each video file, after all postprocessors have been | |
328 | called. The filename will be passed as the only argument. | |
71b640cc PH |
329 | progress_hooks: A list of functions that get called on download |
330 | progress, with a dictionary with the entries | |
5cda4eda | 331 | * status: One of "downloading", "error", or "finished". |
ee69b99a | 332 | Check this first and ignore unknown values. |
3ba7740d | 333 | * info_dict: The extracted info_dict |
71b640cc | 334 | |
5cda4eda | 335 | If status is one of "downloading", or "finished", the |
ee69b99a PH |
336 | following properties may also be present: |
337 | * filename: The final filename (always present) | |
5cda4eda | 338 | * tmpfilename: The filename we're currently writing to |
71b640cc PH |
339 | * downloaded_bytes: Bytes on disk |
340 | * total_bytes: Size of the whole file, None if unknown | |
5cda4eda PH |
341 | * total_bytes_estimate: Guess of the eventual file size, |
342 | None if unavailable. | |
343 | * elapsed: The number of seconds since download started. | |
71b640cc PH |
344 | * eta: The estimated time in seconds, None if unknown |
345 | * speed: The download speed in bytes/second, None if | |
346 | unknown | |
5cda4eda PH |
347 | * fragment_index: The counter of the currently |
348 | downloaded video fragment. | |
349 | * fragment_count: The number of fragments (= individual | |
350 | files that will be merged) | |
71b640cc PH |
351 | |
352 | Progress hooks are guaranteed to be called at least once | |
353 | (with status "finished") if the download is successful. | |
45598f15 | 354 | merge_output_format: Extension to use when merging formats. |
6b591b29 | 355 | final_ext: Expected final extension; used to detect when the file was |
356 | already downloaded and converted. "merge_output_format" is | |
357 | replaced by this extension when given | |
6271f1ca PH |
358 | fixup: Automatically correct known faults of the file. |
359 | One of: | |
360 | - "never": do nothing | |
361 | - "warn": only emit a warning | |
362 | - "detect_or_warn": check whether we can do anything | |
62cd676c | 363 | about it, warn otherwise (default) |
504f20dd | 364 | source_address: Client-side IP address to bind to. |
6ec6cb4e | 365 | call_home: Boolean, true iff we are allowed to contact the |
7a5c1cfe | 366 | yt-dlp servers for debugging. (BROKEN) |
1cf376f5 | 367 | sleep_interval_requests: Number of seconds to sleep between requests |
368 | during extraction | |
7aa589a5 S |
369 | sleep_interval: Number of seconds to sleep before each download when |
370 | used alone or a lower bound of a range for randomized | |
371 | sleep before each download (minimum possible number | |
372 | of seconds to sleep) when used along with | |
373 | max_sleep_interval. | |
374 | max_sleep_interval:Upper bound of a range for randomized sleep before each | |
375 | download (maximum possible number of seconds to sleep). | |
376 | Must only be used along with sleep_interval. | |
377 | Actual sleep time will be a random float from range | |
378 | [sleep_interval; max_sleep_interval]. | |
1cf376f5 | 379 | sleep_interval_subtitles: Number of seconds to sleep before each subtitle download |
cfb56d1a PH |
380 | listformats: Print an overview of available video formats and exit. |
381 | list_thumbnails: Print a table of all thumbnails and exit. | |
347de493 PH |
382 | match_filter: A function that gets called with the info_dict of |
383 | every video. | |
384 | If it returns a message, the video is ignored. | |
385 | If it returns None, the video is downloaded. | |
386 | match_filter_func in utils.py is one example for this. | |
7e5db8c9 | 387 | no_color: Do not emit color codes in output. |
0a840f58 | 388 | geo_bypass: Bypass geographic restriction via faking X-Forwarded-For |
504f20dd | 389 | HTTP header |
0a840f58 | 390 | geo_bypass_country: |
773f291d S |
391 | Two-letter ISO 3166-2 country code that will be used for |
392 | explicit geographic restriction bypassing via faking | |
504f20dd | 393 | X-Forwarded-For HTTP header |
5f95927a S |
394 | geo_bypass_ip_block: |
395 | IP range in CIDR notation that will be used similarly to | |
504f20dd | 396 | geo_bypass_country |
71b640cc | 397 | |
85729c51 | 398 | The following options determine which downloader is picked: |
52a8a1e1 | 399 | external_downloader: A dictionary of protocol keys and the executable of the |
400 | external downloader to use for it. The allowed protocols | |
401 | are default|http|ftp|m3u8|dash|rtsp|rtmp|mms. | |
402 | Set the value to 'native' to use the native downloader | |
403 | hls_prefer_native: Deprecated - Use external_downloader = {'m3u8': 'native'} | |
404 | or {'m3u8': 'ffmpeg'} instead. | |
405 | Use the native HLS downloader instead of ffmpeg/avconv | |
bf09af3a S |
406 | if True, otherwise use ffmpeg/avconv if False, otherwise |
407 | use downloader suggested by extractor if None. | |
53ed7066 | 408 | compat_opts: Compatibility options. See "Differences in default behavior". |
3acf6d38 | 409 | The following options do not work when used through the API: |
410 | filename, abort-on-error, multistreams, no-live-chat, | |
b51d2ae3 | 411 | no-clean-infojson, no-playlist-metafiles, no-keep-subs. |
e4f02757 | 412 | Refer __init__.py for their implementation |
fe7e0c98 | 413 | |
8222d8de | 414 | The following parameters are not used by YoutubeDL itself, they are used by |
7a5c1cfe | 415 | the downloader (see yt_dlp/downloader/common.py): |
51d9739f | 416 | nopart, updatetime, buffersize, ratelimit, throttledratelimit, min_filesize, |
417 | max_filesize, test, noresizebuffer, retries, continuedl, noprogress, consoletitle, | |
418 | xattr_set_filesize, external_downloader_args, hls_use_mpegts, http_chunk_size. | |
76b1bd67 JMF |
419 | |
420 | The following options are used by the post processors: | |
d4a24f40 | 421 | prefer_ffmpeg: If False, use avconv instead of ffmpeg if both are available, |
e4172ac9 | 422 | otherwise prefer ffmpeg. (avconv support is deprecated) |
c0b7d117 S |
423 | ffmpeg_location: Location of the ffmpeg/avconv binary; either the path |
424 | to the binary or its containing directory. | |
43820c03 | 425 | postprocessor_args: A dictionary of postprocessor/executable keys (in lower case) |
34488702 | 426 | and a list of additional command-line arguments for the |
427 | postprocessor/executable. The dict can also have "PP+EXE" keys | |
428 | which are used when the given exe is used by the given PP. | |
429 | Use 'default' as the name for arguments to passed to all PP | |
430 | For compatibility with youtube-dl, a single list of args | |
431 | can also be used | |
e409895f | 432 | |
433 | The following options are used by the extractors: | |
62bff2c1 | 434 | extractor_retries: Number of times to retry for known errors |
435 | dynamic_mpd: Whether to process dynamic DASH manifests (default: True) | |
e409895f | 436 | hls_split_discontinuity: Split HLS playlists to different formats at |
62bff2c1 | 437 | discontinuities such as ad breaks (default: False) |
5d3a0e79 | 438 | extractor_args: A dictionary of arguments to be passed to the extractors. |
439 | See "EXTRACTOR ARGUMENTS" for details. | |
440 | Eg: {'youtube': {'skip': ['dash', 'hls']}} | |
441 | youtube_include_dash_manifest: Deprecated - Use extractor_args instead. | |
442 | If True (default), DASH manifests and related | |
62bff2c1 | 443 | data will be downloaded and processed by extractor. |
444 | You can reduce network I/O by disabling it if you don't | |
445 | care about DASH. (only for youtube) | |
5d3a0e79 | 446 | youtube_include_hls_manifest: Deprecated - Use extractor_args instead. |
447 | If True (default), HLS manifests and related | |
62bff2c1 | 448 | data will be downloaded and processed by extractor. |
449 | You can reduce network I/O by disabling it if you don't | |
450 | care about HLS. (only for youtube) | |
8222d8de JMF |
451 | """ |
452 | ||
c9969434 S |
453 | _NUMERIC_FIELDS = set(( |
454 | 'width', 'height', 'tbr', 'abr', 'asr', 'vbr', 'fps', 'filesize', 'filesize_approx', | |
455 | 'timestamp', 'upload_year', 'upload_month', 'upload_day', | |
456 | 'duration', 'view_count', 'like_count', 'dislike_count', 'repost_count', | |
457 | 'average_rating', 'comment_count', 'age_limit', | |
458 | 'start_time', 'end_time', | |
459 | 'chapter_number', 'season_number', 'episode_number', | |
460 | 'track_number', 'disc_number', 'release_year', | |
461 | 'playlist_index', | |
462 | )) | |
463 | ||
8222d8de | 464 | params = None |
8b7491c8 | 465 | _ies = {} |
56d868db | 466 | _pps = {'pre_process': [], 'before_dl': [], 'after_move': [], 'post_process': []} |
b35496d8 | 467 | _printed_messages = set() |
1cf376f5 | 468 | _first_webpage_request = True |
8222d8de JMF |
469 | _download_retcode = None |
470 | _num_downloads = None | |
30a074c2 | 471 | _playlist_level = 0 |
472 | _playlist_urls = set() | |
8222d8de JMF |
473 | _screen_file = None |
474 | ||
3511266b | 475 | def __init__(self, params=None, auto_init=True): |
8222d8de | 476 | """Create a FileDownloader object with the given options.""" |
e9f9a10f JMF |
477 | if params is None: |
478 | params = {} | |
8b7491c8 | 479 | self._ies = {} |
56c73665 | 480 | self._ies_instances = {} |
56d868db | 481 | self._pps = {'pre_process': [], 'before_dl': [], 'after_move': [], 'post_process': []} |
b35496d8 | 482 | self._printed_messages = set() |
1cf376f5 | 483 | self._first_webpage_request = True |
ab8e5e51 | 484 | self._post_hooks = [] |
933605d7 | 485 | self._progress_hooks = [] |
8222d8de JMF |
486 | self._download_retcode = 0 |
487 | self._num_downloads = 0 | |
488 | self._screen_file = [sys.stdout, sys.stderr][params.get('logtostderr', False)] | |
0783b09b | 489 | self._err_file = sys.stderr |
4abf617b S |
490 | self.params = { |
491 | # Default parameters | |
492 | 'nocheckcertificate': False, | |
493 | } | |
494 | self.params.update(params) | |
a0e07d31 | 495 | self.cache = Cache(self) |
34308b30 | 496 | |
a61f4b28 | 497 | if sys.version_info < (3, 6): |
498 | self.report_warning( | |
0181adef | 499 | 'Python version %d.%d is not supported! Please update to Python 3.6 or above' % sys.version_info[:2]) |
a61f4b28 | 500 | |
88acdbc2 | 501 | if self.params.get('allow_unplayable_formats'): |
502 | self.report_warning( | |
503 | 'You have asked for unplayable formats to be listed/downloaded. ' | |
504 | 'This is a developer option intended for debugging. ' | |
505 | 'If you experience any issues while using this option, DO NOT open a bug report') | |
506 | ||
be5df5ee S |
507 | def check_deprecated(param, option, suggestion): |
508 | if self.params.get(param) is not None: | |
53ed7066 | 509 | self.report_warning('%s is deprecated. Use %s instead' % (option, suggestion)) |
be5df5ee S |
510 | return True |
511 | return False | |
512 | ||
513 | if check_deprecated('cn_verification_proxy', '--cn-verification-proxy', '--geo-verification-proxy'): | |
38cce791 YCH |
514 | if self.params.get('geo_verification_proxy') is None: |
515 | self.params['geo_verification_proxy'] = self.params['cn_verification_proxy'] | |
516 | ||
0d1bb027 | 517 | check_deprecated('autonumber', '--auto-number', '-o "%(autonumber)s-%(title)s.%(ext)s"') |
518 | check_deprecated('usetitle', '--title', '-o "%(title)s-%(id)s.%(ext)s"') | |
53ed7066 | 519 | check_deprecated('useid', '--id', '-o "%(id)s.%(ext)s"') |
0d1bb027 | 520 | |
521 | for msg in self.params.get('warnings', []): | |
522 | self.report_warning(msg) | |
523 | ||
b868936c | 524 | if self.params.get('overwrites') is None: |
525 | self.params.pop('overwrites', None) | |
526 | elif self.params.get('nooverwrites') is not None: | |
527 | # nooverwrites was unnecessarily changed to overwrites | |
528 | # in 0c3d0f51778b153f65c21906031c2e091fcfb641 | |
529 | # This ensures compatibility with both keys | |
530 | self.params['overwrites'] = not self.params['nooverwrites'] | |
531 | else: | |
532 | self.params['nooverwrites'] = not self.params['overwrites'] | |
b9d973be | 533 | |
0783b09b | 534 | if params.get('bidi_workaround', False): |
1c088fa8 PH |
535 | try: |
536 | import pty | |
537 | master, slave = pty.openpty() | |
003c69a8 | 538 | width = compat_get_terminal_size().columns |
1c088fa8 PH |
539 | if width is None: |
540 | width_args = [] | |
541 | else: | |
542 | width_args = ['-w', str(width)] | |
5d681e96 | 543 | sp_kwargs = dict( |
1c088fa8 PH |
544 | stdin=subprocess.PIPE, |
545 | stdout=slave, | |
546 | stderr=self._err_file) | |
5d681e96 PH |
547 | try: |
548 | self._output_process = subprocess.Popen( | |
549 | ['bidiv'] + width_args, **sp_kwargs | |
550 | ) | |
551 | except OSError: | |
5d681e96 PH |
552 | self._output_process = subprocess.Popen( |
553 | ['fribidi', '-c', 'UTF-8'] + width_args, **sp_kwargs) | |
554 | self._output_channel = os.fdopen(master, 'rb') | |
1c088fa8 | 555 | except OSError as ose: |
66e7ace1 | 556 | if ose.errno == errno.ENOENT: |
6febd1c1 | 557 | 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 |
558 | else: |
559 | raise | |
0783b09b | 560 | |
3089bc74 S |
561 | if (sys.platform != 'win32' |
562 | and sys.getfilesystemencoding() in ['ascii', 'ANSI_X3.4-1968'] | |
563 | and not params.get('restrictfilenames', False)): | |
e9137224 | 564 | # Unicode filesystem API will throw errors (#1474, #13027) |
34308b30 | 565 | self.report_warning( |
6febd1c1 | 566 | 'Assuming --restrict-filenames since file system encoding ' |
1b725173 | 567 | 'cannot encode all characters. ' |
6febd1c1 | 568 | 'Set the LC_ALL environment variable to fix this.') |
4a98cdbf | 569 | self.params['restrictfilenames'] = True |
34308b30 | 570 | |
de6000d9 | 571 | self.outtmpl_dict = self.parse_outtmpl() |
486dd09e | 572 | |
187986a8 | 573 | # Creating format selector here allows us to catch syntax errors before the extraction |
574 | self.format_selector = ( | |
575 | None if self.params.get('format') is None | |
576 | else self.build_format_selector(self.params['format'])) | |
577 | ||
dca08720 PH |
578 | self._setup_opener() |
579 | ||
4cd0a709 | 580 | """Preload the archive, if any is specified""" |
581 | def preload_download_archive(fn): | |
582 | if fn is None: | |
583 | return False | |
0760b0a7 | 584 | self.write_debug('Loading archive file %r\n' % fn) |
4cd0a709 | 585 | try: |
586 | with locked_file(fn, 'r', encoding='utf-8') as archive_file: | |
587 | for line in archive_file: | |
588 | self.archive.add(line.strip()) | |
589 | except IOError as ioe: | |
590 | if ioe.errno != errno.ENOENT: | |
591 | raise | |
592 | return False | |
593 | return True | |
594 | ||
595 | self.archive = set() | |
596 | preload_download_archive(self.params.get('download_archive')) | |
597 | ||
3511266b PH |
598 | if auto_init: |
599 | self.print_debug_header() | |
600 | self.add_default_info_extractors() | |
601 | ||
4f026faf | 602 | for pp_def_raw in self.params.get('postprocessors', []): |
4f026faf | 603 | pp_def = dict(pp_def_raw) |
fd7cfb64 | 604 | when = pp_def.pop('when', 'post_process') |
605 | pp_class = get_postprocessor(pp_def.pop('key')) | |
4f026faf | 606 | pp = pp_class(self, **compat_kwargs(pp_def)) |
5bfa4862 | 607 | self.add_post_processor(pp, when=when) |
4f026faf | 608 | |
ab8e5e51 AM |
609 | for ph in self.params.get('post_hooks', []): |
610 | self.add_post_hook(ph) | |
611 | ||
71b640cc PH |
612 | for ph in self.params.get('progress_hooks', []): |
613 | self.add_progress_hook(ph) | |
614 | ||
51fb4995 YCH |
615 | register_socks_protocols() |
616 | ||
7d4111ed PH |
617 | def warn_if_short_id(self, argv): |
618 | # short YouTube ID starting with dash? | |
619 | idxs = [ | |
620 | i for i, a in enumerate(argv) | |
621 | if re.match(r'^-[0-9A-Za-z_-]{10}$', a)] | |
622 | if idxs: | |
623 | correct_argv = ( | |
7a5c1cfe | 624 | ['yt-dlp'] |
3089bc74 S |
625 | + [a for i, a in enumerate(argv) if i not in idxs] |
626 | + ['--'] + [argv[i] for i in idxs] | |
7d4111ed PH |
627 | ) |
628 | self.report_warning( | |
629 | 'Long argument string detected. ' | |
630 | 'Use -- to separate parameters and URLs, like this:\n%s\n' % | |
631 | args_to_str(correct_argv)) | |
632 | ||
8222d8de JMF |
633 | def add_info_extractor(self, ie): |
634 | """Add an InfoExtractor object to the end of the list.""" | |
8b7491c8 | 635 | ie_key = ie.ie_key() |
636 | self._ies[ie_key] = ie | |
e52d7f85 | 637 | if not isinstance(ie, type): |
8b7491c8 | 638 | self._ies_instances[ie_key] = ie |
e52d7f85 | 639 | ie.set_downloader(self) |
8222d8de | 640 | |
8b7491c8 | 641 | def _get_info_extractor_class(self, ie_key): |
642 | ie = self._ies.get(ie_key) | |
643 | if ie is None: | |
644 | ie = get_info_extractor(ie_key) | |
645 | self.add_info_extractor(ie) | |
646 | return ie | |
647 | ||
56c73665 JMF |
648 | def get_info_extractor(self, ie_key): |
649 | """ | |
650 | Get an instance of an IE with name ie_key, it will try to get one from | |
651 | the _ies list, if there's no instance it will create a new one and add | |
652 | it to the extractor list. | |
653 | """ | |
654 | ie = self._ies_instances.get(ie_key) | |
655 | if ie is None: | |
656 | ie = get_info_extractor(ie_key)() | |
657 | self.add_info_extractor(ie) | |
658 | return ie | |
659 | ||
023fa8c4 JMF |
660 | def add_default_info_extractors(self): |
661 | """ | |
662 | Add the InfoExtractors returned by gen_extractors to the end of the list | |
663 | """ | |
e52d7f85 | 664 | for ie in gen_extractor_classes(): |
023fa8c4 JMF |
665 | self.add_info_extractor(ie) |
666 | ||
56d868db | 667 | def add_post_processor(self, pp, when='post_process'): |
8222d8de | 668 | """Add a PostProcessor object to the end of the chain.""" |
5bfa4862 | 669 | self._pps[when].append(pp) |
8222d8de JMF |
670 | pp.set_downloader(self) |
671 | ||
ab8e5e51 AM |
672 | def add_post_hook(self, ph): |
673 | """Add the post hook""" | |
674 | self._post_hooks.append(ph) | |
675 | ||
933605d7 JMF |
676 | def add_progress_hook(self, ph): |
677 | """Add the progress hook (currently only for the file downloader)""" | |
678 | self._progress_hooks.append(ph) | |
8ab470f1 | 679 | |
1c088fa8 | 680 | def _bidi_workaround(self, message): |
5d681e96 | 681 | if not hasattr(self, '_output_channel'): |
1c088fa8 PH |
682 | return message |
683 | ||
5d681e96 | 684 | assert hasattr(self, '_output_process') |
11b85ce6 | 685 | assert isinstance(message, compat_str) |
6febd1c1 PH |
686 | line_count = message.count('\n') + 1 |
687 | self._output_process.stdin.write((message + '\n').encode('utf-8')) | |
5d681e96 | 688 | self._output_process.stdin.flush() |
6febd1c1 | 689 | res = ''.join(self._output_channel.readline().decode('utf-8') |
9e1a5b84 | 690 | for _ in range(line_count)) |
6febd1c1 | 691 | return res[:-len('\n')] |
1c088fa8 | 692 | |
b35496d8 | 693 | def _write_string(self, message, out=None, only_once=False): |
694 | if only_once: | |
695 | if message in self._printed_messages: | |
696 | return | |
697 | self._printed_messages.add(message) | |
698 | write_string(message, out=out, encoding=self.params.get('encoding')) | |
734f90bb | 699 | |
848887eb | 700 | def to_stdout(self, message, skip_eol=False, quiet=False): |
0760b0a7 | 701 | """Print message to stdout""" |
8bf9319e | 702 | if self.params.get('logger'): |
43afe285 | 703 | self.params['logger'].debug(message) |
835a1478 | 704 | elif not quiet or self.params.get('verbose'): |
705 | self._write_string( | |
706 | '%s%s' % (self._bidi_workaround(message), ('' if skip_eol else '\n')), | |
707 | self._err_file if quiet else self._screen_file) | |
8222d8de | 708 | |
b35496d8 | 709 | def to_stderr(self, message, only_once=False): |
0760b0a7 | 710 | """Print message to stderr""" |
11b85ce6 | 711 | assert isinstance(message, compat_str) |
8bf9319e | 712 | if self.params.get('logger'): |
43afe285 IB |
713 | self.params['logger'].error(message) |
714 | else: | |
b35496d8 | 715 | self._write_string('%s\n' % self._bidi_workaround(message), self._err_file, only_once=only_once) |
8222d8de | 716 | |
1e5b9a95 PH |
717 | def to_console_title(self, message): |
718 | if not self.params.get('consoletitle', False): | |
719 | return | |
4bede0d8 C |
720 | if compat_os_name == 'nt': |
721 | if ctypes.windll.kernel32.GetConsoleWindow(): | |
722 | # c_wchar_p() might not be necessary if `message` is | |
723 | # already of type unicode() | |
724 | ctypes.windll.kernel32.SetConsoleTitleW(ctypes.c_wchar_p(message)) | |
1e5b9a95 | 725 | elif 'TERM' in os.environ: |
b46696bd | 726 | self._write_string('\033]0;%s\007' % message, self._screen_file) |
1e5b9a95 | 727 | |
bdde425c PH |
728 | def save_console_title(self): |
729 | if not self.params.get('consoletitle', False): | |
730 | return | |
b7b04c78 | 731 | if self.params.get('simulate'): |
94c3442e | 732 | return |
4bede0d8 | 733 | if compat_os_name != 'nt' and 'TERM' in os.environ: |
efd6c574 | 734 | # Save the title on stack |
734f90bb | 735 | self._write_string('\033[22;0t', self._screen_file) |
bdde425c PH |
736 | |
737 | def restore_console_title(self): | |
738 | if not self.params.get('consoletitle', False): | |
739 | return | |
b7b04c78 | 740 | if self.params.get('simulate'): |
94c3442e | 741 | return |
4bede0d8 | 742 | if compat_os_name != 'nt' and 'TERM' in os.environ: |
efd6c574 | 743 | # Restore the title from stack |
734f90bb | 744 | self._write_string('\033[23;0t', self._screen_file) |
bdde425c PH |
745 | |
746 | def __enter__(self): | |
747 | self.save_console_title() | |
748 | return self | |
749 | ||
750 | def __exit__(self, *args): | |
751 | self.restore_console_title() | |
f89197d7 | 752 | |
dca08720 | 753 | if self.params.get('cookiefile') is not None: |
1bab3437 | 754 | self.cookiejar.save(ignore_discard=True, ignore_expires=True) |
bdde425c | 755 | |
8222d8de JMF |
756 | def trouble(self, message=None, tb=None): |
757 | """Determine action to take when a download problem appears. | |
758 | ||
759 | Depending on if the downloader has been configured to ignore | |
760 | download errors or not, this method may throw an exception or | |
761 | not when errors are found, after printing the message. | |
762 | ||
763 | tb, if given, is additional traceback information. | |
764 | """ | |
765 | if message is not None: | |
766 | self.to_stderr(message) | |
767 | if self.params.get('verbose'): | |
768 | if tb is None: | |
769 | if sys.exc_info()[0]: # if .trouble has been called from an except block | |
6febd1c1 | 770 | tb = '' |
8222d8de | 771 | if hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]: |
6febd1c1 | 772 | tb += ''.join(traceback.format_exception(*sys.exc_info()[1].exc_info)) |
c0384f22 | 773 | tb += encode_compat_str(traceback.format_exc()) |
8222d8de JMF |
774 | else: |
775 | tb_data = traceback.format_list(traceback.extract_stack()) | |
6febd1c1 | 776 | tb = ''.join(tb_data) |
c19bc311 | 777 | if tb: |
778 | self.to_stderr(tb) | |
b1940459 | 779 | if not self.params.get('ignoreerrors'): |
8222d8de JMF |
780 | if sys.exc_info()[0] and hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]: |
781 | exc_info = sys.exc_info()[1].exc_info | |
782 | else: | |
783 | exc_info = sys.exc_info() | |
784 | raise DownloadError(message, exc_info) | |
785 | self._download_retcode = 1 | |
786 | ||
0760b0a7 | 787 | def to_screen(self, message, skip_eol=False): |
788 | """Print message to stdout if not in quiet mode""" | |
789 | self.to_stdout( | |
790 | message, skip_eol, quiet=self.params.get('quiet', False)) | |
791 | ||
c84aeac6 | 792 | def report_warning(self, message, only_once=False): |
8222d8de JMF |
793 | ''' |
794 | Print the message to stderr, it will be prefixed with 'WARNING:' | |
795 | If stderr is a tty file the 'WARNING:' will be colored | |
796 | ''' | |
6d07ce01 JMF |
797 | if self.params.get('logger') is not None: |
798 | self.params['logger'].warning(message) | |
8222d8de | 799 | else: |
ad8915b7 PH |
800 | if self.params.get('no_warnings'): |
801 | return | |
e9c0cdd3 | 802 | if not self.params.get('no_color') and self._err_file.isatty() and compat_os_name != 'nt': |
6d07ce01 JMF |
803 | _msg_header = '\033[0;33mWARNING:\033[0m' |
804 | else: | |
805 | _msg_header = 'WARNING:' | |
806 | warning_message = '%s %s' % (_msg_header, message) | |
b35496d8 | 807 | self.to_stderr(warning_message, only_once) |
8222d8de JMF |
808 | |
809 | def report_error(self, message, tb=None): | |
810 | ''' | |
811 | Do the same as trouble, but prefixes the message with 'ERROR:', colored | |
812 | in red if stderr is a tty file. | |
813 | ''' | |
e9c0cdd3 | 814 | if not self.params.get('no_color') and self._err_file.isatty() and compat_os_name != 'nt': |
6febd1c1 | 815 | _msg_header = '\033[0;31mERROR:\033[0m' |
8222d8de | 816 | else: |
6febd1c1 PH |
817 | _msg_header = 'ERROR:' |
818 | error_message = '%s %s' % (_msg_header, message) | |
8222d8de JMF |
819 | self.trouble(error_message, tb) |
820 | ||
b35496d8 | 821 | def write_debug(self, message, only_once=False): |
0760b0a7 | 822 | '''Log debug message or Print message to stderr''' |
823 | if not self.params.get('verbose', False): | |
824 | return | |
825 | message = '[debug] %s' % message | |
826 | if self.params.get('logger'): | |
827 | self.params['logger'].debug(message) | |
828 | else: | |
b35496d8 | 829 | self.to_stderr(message, only_once) |
0760b0a7 | 830 | |
8222d8de JMF |
831 | def report_file_already_downloaded(self, file_name): |
832 | """Report file has already been fully downloaded.""" | |
833 | try: | |
6febd1c1 | 834 | self.to_screen('[download] %s has already been downloaded' % file_name) |
ce02ed60 | 835 | except UnicodeEncodeError: |
6febd1c1 | 836 | self.to_screen('[download] The file has already been downloaded') |
8222d8de | 837 | |
0c3d0f51 | 838 | def report_file_delete(self, file_name): |
839 | """Report that existing file will be deleted.""" | |
840 | try: | |
c25228e5 | 841 | self.to_screen('Deleting existing file %s' % file_name) |
0c3d0f51 | 842 | except UnicodeEncodeError: |
c25228e5 | 843 | self.to_screen('Deleting existing file') |
0c3d0f51 | 844 | |
1151c407 | 845 | def raise_no_formats(self, info, forced=False): |
846 | has_drm = info.get('__has_drm') | |
88acdbc2 | 847 | msg = 'This video is DRM protected' if has_drm else 'No video formats found!' |
848 | expected = self.params.get('ignore_no_formats_error') | |
849 | if forced or not expected: | |
1151c407 | 850 | raise ExtractorError(msg, video_id=info['id'], ie=info['extractor'], |
851 | expected=has_drm or expected) | |
88acdbc2 | 852 | else: |
853 | self.report_warning(msg) | |
854 | ||
de6000d9 | 855 | def parse_outtmpl(self): |
856 | outtmpl_dict = self.params.get('outtmpl', {}) | |
857 | if not isinstance(outtmpl_dict, dict): | |
858 | outtmpl_dict = {'default': outtmpl_dict} | |
859 | outtmpl_dict.update({ | |
860 | k: v for k, v in DEFAULT_OUTTMPL.items() | |
861 | if not outtmpl_dict.get(k)}) | |
862 | for key, val in outtmpl_dict.items(): | |
863 | if isinstance(val, bytes): | |
864 | self.report_warning( | |
865 | 'Parameter outtmpl is bytes, but should be a unicode string. ' | |
866 | 'Put from __future__ import unicode_literals at the top of your code file or consider switching to Python 3.x.') | |
867 | return outtmpl_dict | |
868 | ||
21cd8fae | 869 | def get_output_path(self, dir_type='', filename=None): |
870 | paths = self.params.get('paths', {}) | |
871 | assert isinstance(paths, dict) | |
872 | path = os.path.join( | |
873 | expand_path(paths.get('home', '').strip()), | |
874 | expand_path(paths.get(dir_type, '').strip()) if dir_type else '', | |
875 | filename or '') | |
876 | ||
877 | # Temporary fix for #4787 | |
878 | # 'Treat' all problem characters by passing filename through preferredencoding | |
879 | # to workaround encoding issues with subprocess on python2 @ Windows | |
880 | if sys.version_info < (3, 0) and sys.platform == 'win32': | |
881 | path = encodeFilename(path, True).decode(preferredencoding()) | |
882 | return sanitize_path(path, force=self.params.get('windowsfilenames')) | |
883 | ||
76a264ac | 884 | @staticmethod |
901130bb | 885 | def _outtmpl_expandpath(outtmpl): |
886 | # expand_path translates '%%' into '%' and '$$' into '$' | |
887 | # correspondingly that is not what we want since we need to keep | |
888 | # '%%' intact for template dict substitution step. Working around | |
889 | # with boundary-alike separator hack. | |
890 | sep = ''.join([random.choice(ascii_letters) for _ in range(32)]) | |
891 | outtmpl = outtmpl.replace('%%', '%{0}%'.format(sep)).replace('$$', '${0}$'.format(sep)) | |
892 | ||
893 | # outtmpl should be expand_path'ed before template dict substitution | |
894 | # because meta fields may contain env variables we don't want to | |
895 | # be expanded. For example, for outtmpl "%(title)s.%(ext)s" and | |
896 | # title "Hello $PATH", we don't want `$PATH` to be expanded. | |
897 | return expand_path(outtmpl).replace(sep, '') | |
898 | ||
899 | @staticmethod | |
900 | def escape_outtmpl(outtmpl): | |
901 | ''' Escape any remaining strings like %s, %abc% etc. ''' | |
902 | return re.sub( | |
903 | STR_FORMAT_RE_TMPL.format('', '(?![%(\0])'), | |
904 | lambda mobj: ('' if mobj.group('has_key') else '%') + mobj.group(0), | |
905 | outtmpl) | |
906 | ||
907 | @classmethod | |
908 | def validate_outtmpl(cls, outtmpl): | |
76a264ac | 909 | ''' @return None or Exception object ''' |
7d1eb38a | 910 | outtmpl = re.sub( |
f5aa5cfb | 911 | STR_FORMAT_RE_TMPL.format('[^)]*', '[ljqB]'), |
7d1eb38a | 912 | lambda mobj: f'{mobj.group(0)[:-1]}s', |
913 | cls._outtmpl_expandpath(outtmpl)) | |
76a264ac | 914 | try: |
7d1eb38a | 915 | cls.escape_outtmpl(outtmpl) % collections.defaultdict(int) |
76a264ac | 916 | return None |
917 | except ValueError as err: | |
918 | return err | |
919 | ||
143db31d | 920 | def prepare_outtmpl(self, outtmpl, info_dict, sanitize=None): |
901130bb | 921 | """ Make the template and info_dict suitable for substitution : ydl.outtmpl_escape(outtmpl) % info_dict """ |
6e84b215 | 922 | info_dict.setdefault('epoch', int(time.time())) # keep epoch consistent once set |
143db31d | 923 | |
6e84b215 | 924 | info_dict = dict(info_dict) # Do not sanitize so as not to consume LazyList |
925 | for key in ('__original_infodict', '__postprocessors'): | |
926 | info_dict.pop(key, None) | |
752cda38 | 927 | info_dict['duration_string'] = ( # %(duration>%H-%M-%S)s is wrong if duration > 24hrs |
53c18592 | 928 | formatSeconds(info_dict['duration'], '-' if sanitize else ':') |
143db31d | 929 | if info_dict.get('duration', None) is not None |
930 | else None) | |
752cda38 | 931 | info_dict['autonumber'] = self.params.get('autonumber_start', 1) - 1 + self._num_downloads |
932 | if info_dict.get('resolution') is None: | |
933 | info_dict['resolution'] = self.format_resolution(info_dict, default=None) | |
143db31d | 934 | |
143db31d | 935 | # For fields playlist_index and autonumber convert all occurrences |
936 | # of %(field)s to %(field)0Nd for backward compatibility | |
937 | field_size_compat_map = { | |
752cda38 | 938 | 'playlist_index': len(str(info_dict.get('_last_playlist_index') or '')), |
939 | 'autonumber': self.params.get('autonumber_size') or 5, | |
143db31d | 940 | } |
752cda38 | 941 | |
385a27fa | 942 | TMPL_DICT = {} |
f5aa5cfb | 943 | EXTERNAL_FORMAT_RE = re.compile(STR_FORMAT_RE_TMPL.format('[^)]*', f'[{STR_FORMAT_TYPES}ljqB]')) |
385a27fa | 944 | MATH_FUNCTIONS = { |
945 | '+': float.__add__, | |
946 | '-': float.__sub__, | |
947 | } | |
e625be0d | 948 | # Field is of the form key1.key2... |
949 | # where keys (except first) can be string, int or slice | |
2b8a2973 | 950 | FIELD_RE = r'\w*(?:\.(?:\w+|{num}|{num}?(?::{num}?){{1,2}}))*'.format(num=r'(?:-?\d+)') |
385a27fa | 951 | MATH_FIELD_RE = r'''{field}|{num}'''.format(field=FIELD_RE, num=r'-?\d+(?:.\d+)?') |
952 | MATH_OPERATORS_RE = r'(?:%s)' % '|'.join(map(re.escape, MATH_FUNCTIONS.keys())) | |
e625be0d | 953 | INTERNAL_FORMAT_RE = re.compile(r'''(?x) |
954 | (?P<negate>-)? | |
385a27fa | 955 | (?P<fields>{field}) |
956 | (?P<maths>(?:{math_op}{math_field})*) | |
e625be0d | 957 | (?:>(?P<strf_format>.+?))? |
7c37ff97 | 958 | (?P<alternate>(?<!\\),[^|)]+)? |
e625be0d | 959 | (?:\|(?P<default>.*?))? |
385a27fa | 960 | $'''.format(field=FIELD_RE, math_op=MATH_OPERATORS_RE, math_field=MATH_FIELD_RE)) |
752cda38 | 961 | |
2b8a2973 | 962 | def _traverse_infodict(k): |
963 | k = k.split('.') | |
964 | if k[0] == '': | |
965 | k.pop(0) | |
966 | return traverse_obj(info_dict, k, is_user_input=True, traverse_string=True) | |
76a264ac | 967 | |
752cda38 | 968 | def get_value(mdict): |
969 | # Object traversal | |
2b8a2973 | 970 | value = _traverse_infodict(mdict['fields']) |
752cda38 | 971 | # Negative |
972 | if mdict['negate']: | |
973 | value = float_or_none(value) | |
974 | if value is not None: | |
975 | value *= -1 | |
976 | # Do maths | |
385a27fa | 977 | offset_key = mdict['maths'] |
978 | if offset_key: | |
752cda38 | 979 | value = float_or_none(value) |
980 | operator = None | |
385a27fa | 981 | while offset_key: |
982 | item = re.match( | |
983 | MATH_FIELD_RE if operator else MATH_OPERATORS_RE, | |
984 | offset_key).group(0) | |
985 | offset_key = offset_key[len(item):] | |
986 | if operator is None: | |
752cda38 | 987 | operator = MATH_FUNCTIONS[item] |
385a27fa | 988 | continue |
989 | item, multiplier = (item[1:], -1) if item[0] == '-' else (item, 1) | |
990 | offset = float_or_none(item) | |
991 | if offset is None: | |
2b8a2973 | 992 | offset = float_or_none(_traverse_infodict(item)) |
385a27fa | 993 | try: |
994 | value = operator(value, multiplier * offset) | |
995 | except (TypeError, ZeroDivisionError): | |
996 | return None | |
997 | operator = None | |
752cda38 | 998 | # Datetime formatting |
999 | if mdict['strf_format']: | |
7c37ff97 | 1000 | value = strftime_or_none(value, mdict['strf_format'].replace('\\,', ',')) |
752cda38 | 1001 | |
1002 | return value | |
1003 | ||
b868936c | 1004 | na = self.params.get('outtmpl_na_placeholder', 'NA') |
1005 | ||
6e84b215 | 1006 | def _dumpjson_default(obj): |
1007 | if isinstance(obj, (set, LazyList)): | |
1008 | return list(obj) | |
1009 | raise TypeError(f'Object of type {type(obj).__name__} is not JSON serializable') | |
1010 | ||
752cda38 | 1011 | def create_key(outer_mobj): |
1012 | if not outer_mobj.group('has_key'): | |
901130bb | 1013 | return f'%{outer_mobj.group(0)}' |
752cda38 | 1014 | key = outer_mobj.group('key') |
752cda38 | 1015 | mobj = re.match(INTERNAL_FORMAT_RE, key) |
7c37ff97 | 1016 | initial_field = mobj.group('fields').split('.')[-1] if mobj else '' |
1017 | value, default = None, na | |
1018 | while mobj: | |
e625be0d | 1019 | mobj = mobj.groupdict() |
7c37ff97 | 1020 | default = mobj['default'] if mobj['default'] is not None else default |
752cda38 | 1021 | value = get_value(mobj) |
7c37ff97 | 1022 | if value is None and mobj['alternate']: |
1023 | mobj = re.match(INTERNAL_FORMAT_RE, mobj['alternate'][1:]) | |
1024 | else: | |
1025 | break | |
752cda38 | 1026 | |
b868936c | 1027 | fmt = outer_mobj.group('format') |
752cda38 | 1028 | if fmt == 's' and value is not None and key in field_size_compat_map.keys(): |
1029 | fmt = '0{:d}d'.format(field_size_compat_map[key]) | |
1030 | ||
1031 | value = default if value is None else value | |
752cda38 | 1032 | |
7d1eb38a | 1033 | str_fmt = f'{fmt[:-1]}s' |
1034 | if fmt[-1] == 'l': | |
1035 | value, fmt = ', '.join(variadic(value)), str_fmt | |
1036 | elif fmt[-1] == 'j': | |
6e84b215 | 1037 | value, fmt = json.dumps(value, default=_dumpjson_default), str_fmt |
7d1eb38a | 1038 | elif fmt[-1] == 'q': |
1039 | value, fmt = compat_shlex_quote(str(value)), str_fmt | |
f5aa5cfb | 1040 | elif fmt[-1] == 'B': |
1041 | value = f'%{str_fmt}'.encode('utf-8') % str(value).encode('utf-8') | |
1042 | value, fmt = value.decode('utf-8', 'ignore'), 's' | |
7d1eb38a | 1043 | elif fmt[-1] == 'c': |
1044 | value = str(value) | |
76a264ac | 1045 | if value is None: |
1046 | value, fmt = default, 's' | |
1047 | else: | |
1048 | value = value[0] | |
1049 | elif fmt[-1] not in 'rs': # numeric | |
a439a3a4 | 1050 | value = float_or_none(value) |
752cda38 | 1051 | if value is None: |
1052 | value, fmt = default, 's' | |
901130bb | 1053 | |
752cda38 | 1054 | if sanitize: |
1055 | if fmt[-1] == 'r': | |
1056 | # If value is an object, sanitize might convert it to a string | |
1057 | # So we convert it to repr first | |
7d1eb38a | 1058 | value, fmt = repr(value), str_fmt |
639f1cea | 1059 | if fmt[-1] in 'csr': |
7c37ff97 | 1060 | value = sanitize(initial_field, value) |
901130bb | 1061 | |
b868936c | 1062 | key = '%s\0%s' % (key.replace('%', '%\0'), outer_mobj.group('format')) |
385a27fa | 1063 | TMPL_DICT[key] = value |
b868936c | 1064 | return '{prefix}%({key}){fmt}'.format(key=key, fmt=fmt, prefix=outer_mobj.group('prefix')) |
752cda38 | 1065 | |
385a27fa | 1066 | return EXTERNAL_FORMAT_RE.sub(create_key, outtmpl), TMPL_DICT |
143db31d | 1067 | |
de6000d9 | 1068 | def _prepare_filename(self, info_dict, tmpl_type='default'): |
8222d8de | 1069 | try: |
586a91b6 | 1070 | sanitize = lambda k, v: sanitize_filename( |
45598aab | 1071 | compat_str(v), |
1bb5c511 | 1072 | restricted=self.params.get('restrictfilenames'), |
40df485f | 1073 | is_id=(k == 'id' or k.endswith('_id'))) |
de6000d9 | 1074 | outtmpl = self.outtmpl_dict.get(tmpl_type, self.outtmpl_dict['default']) |
143db31d | 1075 | outtmpl, template_dict = self.prepare_outtmpl(outtmpl, info_dict, sanitize) |
901130bb | 1076 | outtmpl = self.escape_outtmpl(self._outtmpl_expandpath(outtmpl)) |
1077 | filename = outtmpl % template_dict | |
15da37c7 | 1078 | |
143db31d | 1079 | force_ext = OUTTMPL_TYPES.get(tmpl_type) |
de6000d9 | 1080 | if force_ext is not None: |
752cda38 | 1081 | filename = replace_extension(filename, force_ext, info_dict.get('ext')) |
de6000d9 | 1082 | |
bdc3fd2f U |
1083 | # https://github.com/blackjack4494/youtube-dlc/issues/85 |
1084 | trim_file_name = self.params.get('trim_file_name', False) | |
1085 | if trim_file_name: | |
1086 | fn_groups = filename.rsplit('.') | |
1087 | ext = fn_groups[-1] | |
1088 | sub_ext = '' | |
1089 | if len(fn_groups) > 2: | |
1090 | sub_ext = fn_groups[-2] | |
1091 | filename = '.'.join(filter(None, [fn_groups[0][:trim_file_name], sub_ext, ext])) | |
1092 | ||
0202b52a | 1093 | return filename |
8222d8de | 1094 | except ValueError as err: |
6febd1c1 | 1095 | self.report_error('Error in output template: ' + str(err) + ' (encoding: ' + repr(preferredencoding()) + ')') |
8222d8de JMF |
1096 | return None |
1097 | ||
de6000d9 | 1098 | def prepare_filename(self, info_dict, dir_type='', warn=False): |
1099 | """Generate the output filename.""" | |
21cd8fae | 1100 | |
de6000d9 | 1101 | filename = self._prepare_filename(info_dict, dir_type or 'default') |
1102 | ||
c84aeac6 | 1103 | if warn: |
21cd8fae | 1104 | if not self.params.get('paths'): |
de6000d9 | 1105 | pass |
1106 | elif filename == '-': | |
c84aeac6 | 1107 | self.report_warning('--paths is ignored when an outputting to stdout', only_once=True) |
de6000d9 | 1108 | elif os.path.isabs(filename): |
c84aeac6 | 1109 | self.report_warning('--paths is ignored since an absolute path is given in output template', only_once=True) |
de6000d9 | 1110 | if filename == '-' or not filename: |
1111 | return filename | |
1112 | ||
21cd8fae | 1113 | return self.get_output_path(dir_type, filename) |
0202b52a | 1114 | |
120fe513 | 1115 | def _match_entry(self, info_dict, incomplete=False, silent=False): |
ecdec191 | 1116 | """ Returns None if the file should be downloaded """ |
8222d8de | 1117 | |
c77495e3 | 1118 | video_title = info_dict.get('title', info_dict.get('id', 'video')) |
1119 | ||
8b0d7497 | 1120 | def check_filter(): |
8b0d7497 | 1121 | if 'title' in info_dict: |
1122 | # This can happen when we're just evaluating the playlist | |
1123 | title = info_dict['title'] | |
1124 | matchtitle = self.params.get('matchtitle', False) | |
1125 | if matchtitle: | |
1126 | if not re.search(matchtitle, title, re.IGNORECASE): | |
1127 | return '"' + title + '" title did not match pattern "' + matchtitle + '"' | |
1128 | rejecttitle = self.params.get('rejecttitle', False) | |
1129 | if rejecttitle: | |
1130 | if re.search(rejecttitle, title, re.IGNORECASE): | |
1131 | return '"' + title + '" title matched reject pattern "' + rejecttitle + '"' | |
1132 | date = info_dict.get('upload_date') | |
1133 | if date is not None: | |
1134 | dateRange = self.params.get('daterange', DateRange()) | |
1135 | if date not in dateRange: | |
1136 | return '%s upload date is not in range %s' % (date_from_str(date).isoformat(), dateRange) | |
1137 | view_count = info_dict.get('view_count') | |
1138 | if view_count is not None: | |
1139 | min_views = self.params.get('min_views') | |
1140 | if min_views is not None and view_count < min_views: | |
1141 | return 'Skipping %s, because it has not reached minimum view count (%d/%d)' % (video_title, view_count, min_views) | |
1142 | max_views = self.params.get('max_views') | |
1143 | if max_views is not None and view_count > max_views: | |
1144 | return 'Skipping %s, because it has exceeded the maximum view count (%d/%d)' % (video_title, view_count, max_views) | |
1145 | if age_restricted(info_dict.get('age_limit'), self.params.get('age_limit')): | |
1146 | return 'Skipping "%s" because it is age restricted' % video_title | |
8b0d7497 | 1147 | |
8f18aca8 | 1148 | match_filter = self.params.get('match_filter') |
1149 | if match_filter is not None: | |
1150 | try: | |
1151 | ret = match_filter(info_dict, incomplete=incomplete) | |
1152 | except TypeError: | |
1153 | # For backward compatibility | |
1154 | ret = None if incomplete else match_filter(info_dict) | |
1155 | if ret is not None: | |
1156 | return ret | |
8b0d7497 | 1157 | return None |
1158 | ||
c77495e3 | 1159 | if self.in_download_archive(info_dict): |
1160 | reason = '%s has already been recorded in the archive' % video_title | |
1161 | break_opt, break_err = 'break_on_existing', ExistingVideoReached | |
1162 | else: | |
1163 | reason = check_filter() | |
1164 | break_opt, break_err = 'break_on_reject', RejectedVideoReached | |
8b0d7497 | 1165 | if reason is not None: |
120fe513 | 1166 | if not silent: |
1167 | self.to_screen('[download] ' + reason) | |
c77495e3 | 1168 | if self.params.get(break_opt, False): |
1169 | raise break_err() | |
8b0d7497 | 1170 | return reason |
fe7e0c98 | 1171 | |
b6c45014 JMF |
1172 | @staticmethod |
1173 | def add_extra_info(info_dict, extra_info): | |
1174 | '''Set the keys from extra_info in info dict if they are missing''' | |
1175 | for key, value in extra_info.items(): | |
1176 | info_dict.setdefault(key, value) | |
1177 | ||
409e1828 | 1178 | def extract_info(self, url, download=True, ie_key=None, extra_info=None, |
61aa5ba3 | 1179 | process=True, force_generic_extractor=False): |
41d1cca3 | 1180 | """ |
1181 | Return a list with a dictionary for each video extracted. | |
1182 | ||
1183 | Arguments: | |
1184 | url -- URL to extract | |
1185 | ||
1186 | Keyword arguments: | |
1187 | download -- whether to download videos during extraction | |
1188 | ie_key -- extractor key hint | |
1189 | extra_info -- dictionary containing the extra values to add to each result | |
1190 | process -- whether to resolve all unresolved references (URLs, playlist items), | |
1191 | must be True for download to work. | |
1192 | force_generic_extractor -- force using the generic extractor | |
1193 | """ | |
fe7e0c98 | 1194 | |
409e1828 | 1195 | if extra_info is None: |
1196 | extra_info = {} | |
1197 | ||
61aa5ba3 | 1198 | if not ie_key and force_generic_extractor: |
d22dec74 S |
1199 | ie_key = 'Generic' |
1200 | ||
8222d8de | 1201 | if ie_key: |
8b7491c8 | 1202 | ies = {ie_key: self._get_info_extractor_class(ie_key)} |
8222d8de JMF |
1203 | else: |
1204 | ies = self._ies | |
1205 | ||
8b7491c8 | 1206 | for ie_key, ie in ies.items(): |
8222d8de JMF |
1207 | if not ie.suitable(url): |
1208 | continue | |
1209 | ||
1210 | if not ie.working(): | |
6febd1c1 PH |
1211 | self.report_warning('The program functionality for this site has been marked as broken, ' |
1212 | 'and will probably not work.') | |
8222d8de | 1213 | |
1151c407 | 1214 | temp_id = ie.get_temp_id(url) |
a0566bbf | 1215 | if temp_id is not None and self.in_download_archive({'id': temp_id, 'ie_key': ie_key}): |
1216 | self.to_screen("[%s] %s: has already been recorded in archive" % ( | |
1217 | ie_key, temp_id)) | |
1218 | break | |
8b7491c8 | 1219 | return self.__extract_info(url, self.get_info_extractor(ie_key), download, extra_info, process) |
a0566bbf | 1220 | else: |
1221 | self.report_error('no suitable InfoExtractor for URL %s' % url) | |
1222 | ||
8e5fecc8 | 1223 | def __handle_extraction_exceptions(func): |
1224 | ||
a0566bbf | 1225 | def wrapper(self, *args, **kwargs): |
1226 | try: | |
1227 | return func(self, *args, **kwargs) | |
773f291d S |
1228 | except GeoRestrictedError as e: |
1229 | msg = e.msg | |
1230 | if e.countries: | |
1231 | msg += '\nThis video is available in %s.' % ', '.join( | |
1232 | map(ISO3166Utils.short2full, e.countries)) | |
1233 | msg += '\nYou might want to use a VPN or a proxy server (with --proxy) to workaround.' | |
1234 | self.report_error(msg) | |
fb043a6e | 1235 | except ExtractorError as e: # An error we somewhat expected |
2c74e6fa | 1236 | self.report_error(compat_str(e), e.format_traceback()) |
51d9739f | 1237 | except ThrottledDownload: |
1238 | self.to_stderr('\r') | |
1239 | self.report_warning('The download speed is below throttle limit. Re-extracting data') | |
1240 | return wrapper(self, *args, **kwargs) | |
8e5fecc8 | 1241 | except (MaxDownloadsReached, ExistingVideoReached, RejectedVideoReached, LazyList.IndexError): |
d3e5bbf4 | 1242 | raise |
8222d8de | 1243 | except Exception as e: |
b1940459 | 1244 | if self.params.get('ignoreerrors'): |
9b9c5355 | 1245 | self.report_error(error_to_compat_str(e), tb=encode_compat_str(traceback.format_exc())) |
8222d8de JMF |
1246 | else: |
1247 | raise | |
a0566bbf | 1248 | return wrapper |
1249 | ||
1250 | @__handle_extraction_exceptions | |
58f197b7 | 1251 | def __extract_info(self, url, ie, download, extra_info, process): |
a0566bbf | 1252 | ie_result = ie.extract(url) |
1253 | if ie_result is None: # Finished already (backwards compatibility; listformats and friends should be moved here) | |
1254 | return | |
1255 | if isinstance(ie_result, list): | |
1256 | # Backwards compatibility: old IE result format | |
1257 | ie_result = { | |
1258 | '_type': 'compat_list', | |
1259 | 'entries': ie_result, | |
1260 | } | |
e37d0efb | 1261 | if extra_info.get('original_url'): |
1262 | ie_result.setdefault('original_url', extra_info['original_url']) | |
a0566bbf | 1263 | self.add_default_extra_info(ie_result, ie, url) |
1264 | if process: | |
1265 | return self.process_ie_result(ie_result, download, extra_info) | |
8222d8de | 1266 | else: |
a0566bbf | 1267 | return ie_result |
fe7e0c98 | 1268 | |
ea38e55f | 1269 | def add_default_extra_info(self, ie_result, ie, url): |
6033d980 | 1270 | if url is not None: |
1271 | self.add_extra_info(ie_result, { | |
1272 | 'webpage_url': url, | |
1273 | 'original_url': url, | |
1274 | 'webpage_url_basename': url_basename(url), | |
1275 | }) | |
1276 | if ie is not None: | |
1277 | self.add_extra_info(ie_result, { | |
1278 | 'extractor': ie.IE_NAME, | |
1279 | 'extractor_key': ie.ie_key(), | |
1280 | }) | |
ea38e55f | 1281 | |
58adec46 | 1282 | def process_ie_result(self, ie_result, download=True, extra_info=None): |
8222d8de JMF |
1283 | """ |
1284 | Take the result of the ie(may be modified) and resolve all unresolved | |
1285 | references (URLs, playlist items). | |
1286 | ||
1287 | It will also download the videos if 'download'. | |
1288 | Returns the resolved ie_result. | |
1289 | """ | |
58adec46 | 1290 | if extra_info is None: |
1291 | extra_info = {} | |
e8ee972c PH |
1292 | result_type = ie_result.get('_type', 'video') |
1293 | ||
057a5206 | 1294 | if result_type in ('url', 'url_transparent'): |
134c6ea8 | 1295 | ie_result['url'] = sanitize_url(ie_result['url']) |
e37d0efb | 1296 | if ie_result.get('original_url'): |
1297 | extra_info.setdefault('original_url', ie_result['original_url']) | |
1298 | ||
057a5206 | 1299 | extract_flat = self.params.get('extract_flat', False) |
3089bc74 S |
1300 | if ((extract_flat == 'in_playlist' and 'playlist' in extra_info) |
1301 | or extract_flat is True): | |
ecb54191 | 1302 | info_copy = ie_result.copy() |
6033d980 | 1303 | ie = try_get(ie_result.get('ie_key'), self.get_info_extractor) |
4614bc22 | 1304 | if not ie_result.get('id'): |
1305 | info_copy['id'] = ie.get_temp_id(ie_result['url']) | |
6033d980 | 1306 | self.add_default_extra_info(info_copy, ie, ie_result['url']) |
4614bc22 | 1307 | self.add_extra_info(info_copy, extra_info) |
ecb54191 | 1308 | self.__forced_printings(info_copy, self.prepare_filename(info_copy), incomplete=True) |
4614bc22 | 1309 | if self.params.get('force_write_download_archive', False): |
1310 | self.record_download_archive(info_copy) | |
e8ee972c PH |
1311 | return ie_result |
1312 | ||
8222d8de | 1313 | if result_type == 'video': |
b6c45014 | 1314 | self.add_extra_info(ie_result, extra_info) |
9c2b75b5 | 1315 | ie_result = self.process_video_result(ie_result, download=download) |
28b0eb0f | 1316 | additional_urls = (ie_result or {}).get('additional_urls') |
9c2b75b5 | 1317 | if additional_urls: |
e9f4ccd1 | 1318 | # TODO: Improve MetadataParserPP to allow setting a list |
9c2b75b5 | 1319 | if isinstance(additional_urls, compat_str): |
1320 | additional_urls = [additional_urls] | |
1321 | self.to_screen( | |
1322 | '[info] %s: %d additional URL(s) requested' % (ie_result['id'], len(additional_urls))) | |
1323 | self.write_debug('Additional URLs: "%s"' % '", "'.join(additional_urls)) | |
1324 | ie_result['additional_entries'] = [ | |
1325 | self.extract_info( | |
1326 | url, download, extra_info, | |
1327 | force_generic_extractor=self.params.get('force_generic_extractor')) | |
1328 | for url in additional_urls | |
1329 | ] | |
1330 | return ie_result | |
8222d8de JMF |
1331 | elif result_type == 'url': |
1332 | # We have to add extra_info to the results because it may be | |
1333 | # contained in a playlist | |
07cce701 | 1334 | return self.extract_info( |
1335 | ie_result['url'], download, | |
1336 | ie_key=ie_result.get('ie_key'), | |
1337 | extra_info=extra_info) | |
7fc3fa05 PH |
1338 | elif result_type == 'url_transparent': |
1339 | # Use the information from the embedding page | |
1340 | info = self.extract_info( | |
1341 | ie_result['url'], ie_key=ie_result.get('ie_key'), | |
1342 | extra_info=extra_info, download=False, process=False) | |
1343 | ||
1640eb09 S |
1344 | # extract_info may return None when ignoreerrors is enabled and |
1345 | # extraction failed with an error, don't crash and return early | |
1346 | # in this case | |
1347 | if not info: | |
1348 | return info | |
1349 | ||
412c617d PH |
1350 | force_properties = dict( |
1351 | (k, v) for k, v in ie_result.items() if v is not None) | |
0396806f | 1352 | for f in ('_type', 'url', 'id', 'extractor', 'extractor_key', 'ie_key'): |
412c617d PH |
1353 | if f in force_properties: |
1354 | del force_properties[f] | |
1355 | new_result = info.copy() | |
1356 | new_result.update(force_properties) | |
7fc3fa05 | 1357 | |
0563f7ac S |
1358 | # Extracted info may not be a video result (i.e. |
1359 | # info.get('_type', 'video') != video) but rather an url or | |
1360 | # url_transparent. In such cases outer metadata (from ie_result) | |
1361 | # should be propagated to inner one (info). For this to happen | |
1362 | # _type of info should be overridden with url_transparent. This | |
067aa17e | 1363 | # fixes issue from https://github.com/ytdl-org/youtube-dl/pull/11163. |
0563f7ac S |
1364 | if new_result.get('_type') == 'url': |
1365 | new_result['_type'] = 'url_transparent' | |
7fc3fa05 PH |
1366 | |
1367 | return self.process_ie_result( | |
1368 | new_result, download=download, extra_info=extra_info) | |
40fcba5e | 1369 | elif result_type in ('playlist', 'multi_video'): |
30a074c2 | 1370 | # Protect from infinite recursion due to recursively nested playlists |
1371 | # (see https://github.com/ytdl-org/youtube-dl/issues/27833) | |
1372 | webpage_url = ie_result['webpage_url'] | |
1373 | if webpage_url in self._playlist_urls: | |
7e85e872 | 1374 | self.to_screen( |
30a074c2 | 1375 | '[download] Skipping already downloaded playlist: %s' |
1376 | % ie_result.get('title') or ie_result.get('id')) | |
1377 | return | |
7e85e872 | 1378 | |
30a074c2 | 1379 | self._playlist_level += 1 |
1380 | self._playlist_urls.add(webpage_url) | |
bc516a3f | 1381 | self._sanitize_thumbnails(ie_result) |
30a074c2 | 1382 | try: |
1383 | return self.__process_playlist(ie_result, download) | |
1384 | finally: | |
1385 | self._playlist_level -= 1 | |
1386 | if not self._playlist_level: | |
1387 | self._playlist_urls.clear() | |
8222d8de | 1388 | elif result_type == 'compat_list': |
c9bf4114 PH |
1389 | self.report_warning( |
1390 | 'Extractor %s returned a compat_list result. ' | |
1391 | 'It needs to be updated.' % ie_result.get('extractor')) | |
5f6a1245 | 1392 | |
8222d8de | 1393 | def _fixup(r): |
b868936c | 1394 | self.add_extra_info(r, { |
1395 | 'extractor': ie_result['extractor'], | |
1396 | 'webpage_url': ie_result['webpage_url'], | |
1397 | 'webpage_url_basename': url_basename(ie_result['webpage_url']), | |
1398 | 'extractor_key': ie_result['extractor_key'], | |
1399 | }) | |
8222d8de JMF |
1400 | return r |
1401 | ie_result['entries'] = [ | |
b6c45014 | 1402 | self.process_ie_result(_fixup(r), download, extra_info) |
8222d8de JMF |
1403 | for r in ie_result['entries'] |
1404 | ] | |
1405 | return ie_result | |
1406 | else: | |
1407 | raise Exception('Invalid result type: %s' % result_type) | |
1408 | ||
e92caff5 | 1409 | def _ensure_dir_exists(self, path): |
1410 | return make_dir(path, self.report_error) | |
1411 | ||
30a074c2 | 1412 | def __process_playlist(self, ie_result, download): |
1413 | # We process each entry in the playlist | |
1414 | playlist = ie_result.get('title') or ie_result.get('id') | |
1415 | self.to_screen('[download] Downloading playlist: %s' % playlist) | |
1416 | ||
498f5606 | 1417 | if 'entries' not in ie_result: |
1418 | raise EntryNotInPlaylist() | |
1419 | incomplete_entries = bool(ie_result.get('requested_entries')) | |
1420 | if incomplete_entries: | |
1421 | def fill_missing_entries(entries, indexes): | |
1422 | ret = [None] * max(*indexes) | |
1423 | for i, entry in zip(indexes, entries): | |
1424 | ret[i - 1] = entry | |
1425 | return ret | |
1426 | ie_result['entries'] = fill_missing_entries(ie_result['entries'], ie_result['requested_entries']) | |
02fd60d3 | 1427 | |
30a074c2 | 1428 | playlist_results = [] |
1429 | ||
56a8fb4f | 1430 | playliststart = self.params.get('playliststart', 1) |
30a074c2 | 1431 | playlistend = self.params.get('playlistend') |
1432 | # For backwards compatibility, interpret -1 as whole list | |
1433 | if playlistend == -1: | |
1434 | playlistend = None | |
1435 | ||
1436 | playlistitems_str = self.params.get('playlist_items') | |
1437 | playlistitems = None | |
1438 | if playlistitems_str is not None: | |
1439 | def iter_playlistitems(format): | |
1440 | for string_segment in format.split(','): | |
1441 | if '-' in string_segment: | |
1442 | start, end = string_segment.split('-') | |
1443 | for item in range(int(start), int(end) + 1): | |
1444 | yield int(item) | |
1445 | else: | |
1446 | yield int(string_segment) | |
1447 | playlistitems = orderedSet(iter_playlistitems(playlistitems_str)) | |
1448 | ||
1449 | ie_entries = ie_result['entries'] | |
56a8fb4f | 1450 | msg = ( |
1451 | 'Downloading %d videos' if not isinstance(ie_entries, list) | |
1452 | else 'Collected %d videos; downloading %%d of them' % len(ie_entries)) | |
8e5fecc8 | 1453 | |
1454 | if isinstance(ie_entries, list): | |
1455 | def get_entry(i): | |
1456 | return ie_entries[i - 1] | |
1457 | else: | |
1458 | if not isinstance(ie_entries, PagedList): | |
1459 | ie_entries = LazyList(ie_entries) | |
1460 | ||
1461 | def get_entry(i): | |
1462 | return YoutubeDL.__handle_extraction_exceptions( | |
1463 | lambda self, i: ie_entries[i - 1] | |
1464 | )(self, i) | |
50fed816 | 1465 | |
56a8fb4f | 1466 | entries = [] |
1467 | for i in playlistitems or itertools.count(playliststart): | |
1468 | if playlistitems is None and playlistend is not None and playlistend < i: | |
1469 | break | |
1470 | entry = None | |
1471 | try: | |
50fed816 | 1472 | entry = get_entry(i) |
56a8fb4f | 1473 | if entry is None: |
498f5606 | 1474 | raise EntryNotInPlaylist() |
56a8fb4f | 1475 | except (IndexError, EntryNotInPlaylist): |
1476 | if incomplete_entries: | |
1477 | raise EntryNotInPlaylist() | |
1478 | elif not playlistitems: | |
1479 | break | |
1480 | entries.append(entry) | |
120fe513 | 1481 | try: |
1482 | if entry is not None: | |
1483 | self._match_entry(entry, incomplete=True, silent=True) | |
1484 | except (ExistingVideoReached, RejectedVideoReached): | |
1485 | break | |
56a8fb4f | 1486 | ie_result['entries'] = entries |
30a074c2 | 1487 | |
56a8fb4f | 1488 | # Save playlist_index before re-ordering |
1489 | entries = [ | |
9e598870 | 1490 | ((playlistitems[i - 1] if playlistitems else i + playliststart - 1), entry) |
56a8fb4f | 1491 | for i, entry in enumerate(entries, 1) |
1492 | if entry is not None] | |
1493 | n_entries = len(entries) | |
498f5606 | 1494 | |
498f5606 | 1495 | if not playlistitems and (playliststart or playlistend): |
56a8fb4f | 1496 | playlistitems = list(range(playliststart, playliststart + n_entries)) |
498f5606 | 1497 | ie_result['requested_entries'] = playlistitems |
1498 | ||
1499 | if self.params.get('allow_playlist_files', True): | |
1500 | ie_copy = { | |
1501 | 'playlist': playlist, | |
1502 | 'playlist_id': ie_result.get('id'), | |
1503 | 'playlist_title': ie_result.get('title'), | |
1504 | 'playlist_uploader': ie_result.get('uploader'), | |
1505 | 'playlist_uploader_id': ie_result.get('uploader_id'), | |
71729754 | 1506 | 'playlist_index': 0, |
498f5606 | 1507 | } |
1508 | ie_copy.update(dict(ie_result)) | |
1509 | ||
1510 | if self.params.get('writeinfojson', False): | |
1511 | infofn = self.prepare_filename(ie_copy, 'pl_infojson') | |
1512 | if not self._ensure_dir_exists(encodeFilename(infofn)): | |
1513 | return | |
1514 | if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(infofn)): | |
1515 | self.to_screen('[info] Playlist metadata is already present') | |
1516 | else: | |
1517 | self.to_screen('[info] Writing playlist metadata as JSON to: ' + infofn) | |
1518 | try: | |
8012d892 | 1519 | write_json_file(self.sanitize_info(ie_result, self.params.get('clean_infojson', True)), infofn) |
498f5606 | 1520 | except (OSError, IOError): |
1521 | self.report_error('Cannot write playlist metadata to JSON file ' + infofn) | |
1522 | ||
681de68e | 1523 | # TODO: This should be passed to ThumbnailsConvertor if necessary |
1524 | self._write_thumbnails(ie_copy, self.prepare_filename(ie_copy, 'pl_thumbnail')) | |
1525 | ||
498f5606 | 1526 | if self.params.get('writedescription', False): |
1527 | descfn = self.prepare_filename(ie_copy, 'pl_description') | |
1528 | if not self._ensure_dir_exists(encodeFilename(descfn)): | |
1529 | return | |
1530 | if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(descfn)): | |
1531 | self.to_screen('[info] Playlist description is already present') | |
1532 | elif ie_result.get('description') is None: | |
1533 | self.report_warning('There\'s no playlist description to write.') | |
1534 | else: | |
1535 | try: | |
1536 | self.to_screen('[info] Writing playlist description to: ' + descfn) | |
1537 | with io.open(encodeFilename(descfn), 'w', encoding='utf-8') as descfile: | |
1538 | descfile.write(ie_result['description']) | |
1539 | except (OSError, IOError): | |
1540 | self.report_error('Cannot write playlist description file ' + descfn) | |
1541 | return | |
30a074c2 | 1542 | |
1543 | if self.params.get('playlistreverse', False): | |
1544 | entries = entries[::-1] | |
30a074c2 | 1545 | if self.params.get('playlistrandom', False): |
1546 | random.shuffle(entries) | |
1547 | ||
1548 | x_forwarded_for = ie_result.get('__x_forwarded_for_ip') | |
1549 | ||
56a8fb4f | 1550 | self.to_screen('[%s] playlist %s: %s' % (ie_result['extractor'], playlist, msg % n_entries)) |
26e2805c | 1551 | failures = 0 |
1552 | max_failures = self.params.get('skip_playlist_after_errors') or float('inf') | |
71729754 | 1553 | for i, entry_tuple in enumerate(entries, 1): |
1554 | playlist_index, entry = entry_tuple | |
81139999 | 1555 | if 'playlist-index' in self.params.get('compat_opts', []): |
1556 | playlist_index = playlistitems[i - 1] if playlistitems else i + playliststart - 1 | |
30a074c2 | 1557 | self.to_screen('[download] Downloading video %s of %s' % (i, n_entries)) |
1558 | # This __x_forwarded_for_ip thing is a bit ugly but requires | |
1559 | # minimal changes | |
1560 | if x_forwarded_for: | |
1561 | entry['__x_forwarded_for_ip'] = x_forwarded_for | |
1562 | extra = { | |
1563 | 'n_entries': n_entries, | |
f59ae581 | 1564 | '_last_playlist_index': max(playlistitems) if playlistitems else (playlistend or n_entries), |
71729754 | 1565 | 'playlist_index': playlist_index, |
1566 | 'playlist_autonumber': i, | |
30a074c2 | 1567 | 'playlist': playlist, |
1568 | 'playlist_id': ie_result.get('id'), | |
1569 | 'playlist_title': ie_result.get('title'), | |
1570 | 'playlist_uploader': ie_result.get('uploader'), | |
1571 | 'playlist_uploader_id': ie_result.get('uploader_id'), | |
30a074c2 | 1572 | 'extractor': ie_result['extractor'], |
1573 | 'webpage_url': ie_result['webpage_url'], | |
1574 | 'webpage_url_basename': url_basename(ie_result['webpage_url']), | |
1575 | 'extractor_key': ie_result['extractor_key'], | |
1576 | } | |
1577 | ||
1578 | if self._match_entry(entry, incomplete=True) is not None: | |
1579 | continue | |
1580 | ||
1581 | entry_result = self.__process_iterable_entry(entry, download, extra) | |
26e2805c | 1582 | if not entry_result: |
1583 | failures += 1 | |
1584 | if failures >= max_failures: | |
1585 | self.report_error( | |
1586 | 'Skipping the remaining entries in playlist "%s" since %d items failed extraction' % (playlist, failures)) | |
1587 | break | |
30a074c2 | 1588 | # TODO: skip failed (empty) entries? |
1589 | playlist_results.append(entry_result) | |
1590 | ie_result['entries'] = playlist_results | |
1591 | self.to_screen('[download] Finished downloading playlist: %s' % playlist) | |
1592 | return ie_result | |
1593 | ||
a0566bbf | 1594 | @__handle_extraction_exceptions |
1595 | def __process_iterable_entry(self, entry, download, extra_info): | |
1596 | return self.process_ie_result( | |
1597 | entry, download=download, extra_info=extra_info) | |
1598 | ||
67134eab JMF |
1599 | def _build_format_filter(self, filter_spec): |
1600 | " Returns a function to filter the formats according to the filter_spec " | |
083c9df9 PH |
1601 | |
1602 | OPERATORS = { | |
1603 | '<': operator.lt, | |
1604 | '<=': operator.le, | |
1605 | '>': operator.gt, | |
1606 | '>=': operator.ge, | |
1607 | '=': operator.eq, | |
1608 | '!=': operator.ne, | |
1609 | } | |
67134eab | 1610 | operator_rex = re.compile(r'''(?x)\s* |
187986a8 | 1611 | (?P<key>width|height|tbr|abr|vbr|asr|filesize|filesize_approx|fps)\s* |
1612 | (?P<op>%s)(?P<none_inclusive>\s*\?)?\s* | |
1613 | (?P<value>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?)\s* | |
083c9df9 | 1614 | ''' % '|'.join(map(re.escape, OPERATORS.keys()))) |
187986a8 | 1615 | m = operator_rex.fullmatch(filter_spec) |
9ddb6925 S |
1616 | if m: |
1617 | try: | |
1618 | comparison_value = int(m.group('value')) | |
1619 | except ValueError: | |
1620 | comparison_value = parse_filesize(m.group('value')) | |
1621 | if comparison_value is None: | |
1622 | comparison_value = parse_filesize(m.group('value') + 'B') | |
1623 | if comparison_value is None: | |
1624 | raise ValueError( | |
1625 | 'Invalid value %r in format specification %r' % ( | |
67134eab | 1626 | m.group('value'), filter_spec)) |
9ddb6925 S |
1627 | op = OPERATORS[m.group('op')] |
1628 | ||
083c9df9 | 1629 | if not m: |
9ddb6925 S |
1630 | STR_OPERATORS = { |
1631 | '=': operator.eq, | |
10d33b34 YCH |
1632 | '^=': lambda attr, value: attr.startswith(value), |
1633 | '$=': lambda attr, value: attr.endswith(value), | |
1634 | '*=': lambda attr, value: value in attr, | |
9ddb6925 | 1635 | } |
187986a8 | 1636 | str_operator_rex = re.compile(r'''(?x)\s* |
1637 | (?P<key>[a-zA-Z0-9._-]+)\s* | |
1638 | (?P<negation>!\s*)?(?P<op>%s)(?P<none_inclusive>\s*\?)?\s* | |
1639 | (?P<value>[a-zA-Z0-9._-]+)\s* | |
9ddb6925 | 1640 | ''' % '|'.join(map(re.escape, STR_OPERATORS.keys()))) |
187986a8 | 1641 | m = str_operator_rex.fullmatch(filter_spec) |
9ddb6925 S |
1642 | if m: |
1643 | comparison_value = m.group('value') | |
2cc779f4 S |
1644 | str_op = STR_OPERATORS[m.group('op')] |
1645 | if m.group('negation'): | |
e118a879 | 1646 | op = lambda attr, value: not str_op(attr, value) |
2cc779f4 S |
1647 | else: |
1648 | op = str_op | |
083c9df9 | 1649 | |
9ddb6925 | 1650 | if not m: |
187986a8 | 1651 | raise SyntaxError('Invalid filter specification %r' % filter_spec) |
083c9df9 PH |
1652 | |
1653 | def _filter(f): | |
1654 | actual_value = f.get(m.group('key')) | |
1655 | if actual_value is None: | |
1656 | return m.group('none_inclusive') | |
1657 | return op(actual_value, comparison_value) | |
67134eab JMF |
1658 | return _filter |
1659 | ||
0017d9ad | 1660 | def _default_format_spec(self, info_dict, download=True): |
0017d9ad | 1661 | |
af0f7428 S |
1662 | def can_merge(): |
1663 | merger = FFmpegMergerPP(self) | |
1664 | return merger.available and merger.can_merge() | |
1665 | ||
91ebc640 | 1666 | prefer_best = ( |
b7b04c78 | 1667 | not self.params.get('simulate') |
91ebc640 | 1668 | and download |
1669 | and ( | |
1670 | not can_merge() | |
19807826 | 1671 | or info_dict.get('is_live', False) |
de6000d9 | 1672 | or self.outtmpl_dict['default'] == '-')) |
53ed7066 | 1673 | compat = ( |
1674 | prefer_best | |
1675 | or self.params.get('allow_multiple_audio_streams', False) | |
1676 | or 'format-spec' in self.params.get('compat_opts', [])) | |
91ebc640 | 1677 | |
1678 | return ( | |
53ed7066 | 1679 | 'best/bestvideo+bestaudio' if prefer_best |
1680 | else 'bestvideo*+bestaudio/best' if not compat | |
91ebc640 | 1681 | else 'bestvideo+bestaudio/best') |
0017d9ad | 1682 | |
67134eab JMF |
1683 | def build_format_selector(self, format_spec): |
1684 | def syntax_error(note, start): | |
1685 | message = ( | |
1686 | 'Invalid format specification: ' | |
1687 | '{0}\n\t{1}\n\t{2}^'.format(note, format_spec, ' ' * start[1])) | |
1688 | return SyntaxError(message) | |
1689 | ||
1690 | PICKFIRST = 'PICKFIRST' | |
1691 | MERGE = 'MERGE' | |
1692 | SINGLE = 'SINGLE' | |
0130afb7 | 1693 | GROUP = 'GROUP' |
67134eab JMF |
1694 | FormatSelector = collections.namedtuple('FormatSelector', ['type', 'selector', 'filters']) |
1695 | ||
91ebc640 | 1696 | allow_multiple_streams = {'audio': self.params.get('allow_multiple_audio_streams', False), |
1697 | 'video': self.params.get('allow_multiple_video_streams', False)} | |
909d24dd | 1698 | |
e8e73840 | 1699 | check_formats = self.params.get('check_formats') |
1700 | ||
67134eab JMF |
1701 | def _parse_filter(tokens): |
1702 | filter_parts = [] | |
1703 | for type, string, start, _, _ in tokens: | |
1704 | if type == tokenize.OP and string == ']': | |
1705 | return ''.join(filter_parts) | |
1706 | else: | |
1707 | filter_parts.append(string) | |
1708 | ||
232541df | 1709 | def _remove_unused_ops(tokens): |
17cc1534 | 1710 | # Remove operators that we don't use and join them with the surrounding strings |
232541df JMF |
1711 | # for example: 'mp4' '-' 'baseline' '-' '16x9' is converted to 'mp4-baseline-16x9' |
1712 | ALLOWED_OPS = ('/', '+', ',', '(', ')') | |
1713 | last_string, last_start, last_end, last_line = None, None, None, None | |
1714 | for type, string, start, end, line in tokens: | |
1715 | if type == tokenize.OP and string == '[': | |
1716 | if last_string: | |
1717 | yield tokenize.NAME, last_string, last_start, last_end, last_line | |
1718 | last_string = None | |
1719 | yield type, string, start, end, line | |
1720 | # everything inside brackets will be handled by _parse_filter | |
1721 | for type, string, start, end, line in tokens: | |
1722 | yield type, string, start, end, line | |
1723 | if type == tokenize.OP and string == ']': | |
1724 | break | |
1725 | elif type == tokenize.OP and string in ALLOWED_OPS: | |
1726 | if last_string: | |
1727 | yield tokenize.NAME, last_string, last_start, last_end, last_line | |
1728 | last_string = None | |
1729 | yield type, string, start, end, line | |
1730 | elif type in [tokenize.NAME, tokenize.NUMBER, tokenize.OP]: | |
1731 | if not last_string: | |
1732 | last_string = string | |
1733 | last_start = start | |
1734 | last_end = end | |
1735 | else: | |
1736 | last_string += string | |
1737 | if last_string: | |
1738 | yield tokenize.NAME, last_string, last_start, last_end, last_line | |
1739 | ||
cf2ac6df | 1740 | def _parse_format_selection(tokens, inside_merge=False, inside_choice=False, inside_group=False): |
67134eab JMF |
1741 | selectors = [] |
1742 | current_selector = None | |
1743 | for type, string, start, _, _ in tokens: | |
1744 | # ENCODING is only defined in python 3.x | |
1745 | if type == getattr(tokenize, 'ENCODING', None): | |
1746 | continue | |
1747 | elif type in [tokenize.NAME, tokenize.NUMBER]: | |
1748 | current_selector = FormatSelector(SINGLE, string, []) | |
1749 | elif type == tokenize.OP: | |
cf2ac6df JMF |
1750 | if string == ')': |
1751 | if not inside_group: | |
1752 | # ')' will be handled by the parentheses group | |
1753 | tokens.restore_last_token() | |
67134eab | 1754 | break |
cf2ac6df | 1755 | elif inside_merge and string in ['/', ',']: |
0130afb7 JMF |
1756 | tokens.restore_last_token() |
1757 | break | |
cf2ac6df JMF |
1758 | elif inside_choice and string == ',': |
1759 | tokens.restore_last_token() | |
1760 | break | |
1761 | elif string == ',': | |
0a31a350 JMF |
1762 | if not current_selector: |
1763 | raise syntax_error('"," must follow a format selector', start) | |
67134eab JMF |
1764 | selectors.append(current_selector) |
1765 | current_selector = None | |
1766 | elif string == '/': | |
d96d604e JMF |
1767 | if not current_selector: |
1768 | raise syntax_error('"/" must follow a format selector', start) | |
67134eab | 1769 | first_choice = current_selector |
cf2ac6df | 1770 | second_choice = _parse_format_selection(tokens, inside_choice=True) |
f5f4a27a | 1771 | current_selector = FormatSelector(PICKFIRST, (first_choice, second_choice), []) |
67134eab JMF |
1772 | elif string == '[': |
1773 | if not current_selector: | |
1774 | current_selector = FormatSelector(SINGLE, 'best', []) | |
1775 | format_filter = _parse_filter(tokens) | |
1776 | current_selector.filters.append(format_filter) | |
0130afb7 JMF |
1777 | elif string == '(': |
1778 | if current_selector: | |
1779 | raise syntax_error('Unexpected "("', start) | |
cf2ac6df JMF |
1780 | group = _parse_format_selection(tokens, inside_group=True) |
1781 | current_selector = FormatSelector(GROUP, group, []) | |
67134eab | 1782 | elif string == '+': |
d03cfdce | 1783 | if not current_selector: |
1784 | raise syntax_error('Unexpected "+"', start) | |
1785 | selector_1 = current_selector | |
1786 | selector_2 = _parse_format_selection(tokens, inside_merge=True) | |
1787 | if not selector_2: | |
1788 | raise syntax_error('Expected a selector', start) | |
1789 | current_selector = FormatSelector(MERGE, (selector_1, selector_2), []) | |
67134eab JMF |
1790 | else: |
1791 | raise syntax_error('Operator not recognized: "{0}"'.format(string), start) | |
1792 | elif type == tokenize.ENDMARKER: | |
1793 | break | |
1794 | if current_selector: | |
1795 | selectors.append(current_selector) | |
1796 | return selectors | |
1797 | ||
f8d4ad9a | 1798 | def _merge(formats_pair): |
1799 | format_1, format_2 = formats_pair | |
1800 | ||
1801 | formats_info = [] | |
1802 | formats_info.extend(format_1.get('requested_formats', (format_1,))) | |
1803 | formats_info.extend(format_2.get('requested_formats', (format_2,))) | |
1804 | ||
1805 | if not allow_multiple_streams['video'] or not allow_multiple_streams['audio']: | |
551f9388 | 1806 | get_no_more = {'video': False, 'audio': False} |
f8d4ad9a | 1807 | for (i, fmt_info) in enumerate(formats_info): |
551f9388 | 1808 | if fmt_info.get('acodec') == fmt_info.get('vcodec') == 'none': |
1809 | formats_info.pop(i) | |
1810 | continue | |
1811 | for aud_vid in ['audio', 'video']: | |
f8d4ad9a | 1812 | if not allow_multiple_streams[aud_vid] and fmt_info.get(aud_vid[0] + 'codec') != 'none': |
1813 | if get_no_more[aud_vid]: | |
1814 | formats_info.pop(i) | |
f5510afe | 1815 | break |
f8d4ad9a | 1816 | get_no_more[aud_vid] = True |
1817 | ||
1818 | if len(formats_info) == 1: | |
1819 | return formats_info[0] | |
1820 | ||
1821 | video_fmts = [fmt_info for fmt_info in formats_info if fmt_info.get('vcodec') != 'none'] | |
1822 | audio_fmts = [fmt_info for fmt_info in formats_info if fmt_info.get('acodec') != 'none'] | |
1823 | ||
1824 | the_only_video = video_fmts[0] if len(video_fmts) == 1 else None | |
1825 | the_only_audio = audio_fmts[0] if len(audio_fmts) == 1 else None | |
1826 | ||
1827 | output_ext = self.params.get('merge_output_format') | |
1828 | if not output_ext: | |
1829 | if the_only_video: | |
1830 | output_ext = the_only_video['ext'] | |
1831 | elif the_only_audio and not video_fmts: | |
1832 | output_ext = the_only_audio['ext'] | |
1833 | else: | |
1834 | output_ext = 'mkv' | |
1835 | ||
1836 | new_dict = { | |
1837 | 'requested_formats': formats_info, | |
1838 | 'format': '+'.join(fmt_info.get('format') for fmt_info in formats_info), | |
1839 | 'format_id': '+'.join(fmt_info.get('format_id') for fmt_info in formats_info), | |
1840 | 'ext': output_ext, | |
1841 | } | |
1842 | ||
1843 | if the_only_video: | |
1844 | new_dict.update({ | |
1845 | 'width': the_only_video.get('width'), | |
1846 | 'height': the_only_video.get('height'), | |
1847 | 'resolution': the_only_video.get('resolution') or self.format_resolution(the_only_video), | |
1848 | 'fps': the_only_video.get('fps'), | |
1849 | 'vcodec': the_only_video.get('vcodec'), | |
1850 | 'vbr': the_only_video.get('vbr'), | |
1851 | 'stretched_ratio': the_only_video.get('stretched_ratio'), | |
1852 | }) | |
1853 | ||
1854 | if the_only_audio: | |
1855 | new_dict.update({ | |
1856 | 'acodec': the_only_audio.get('acodec'), | |
1857 | 'abr': the_only_audio.get('abr'), | |
1858 | }) | |
1859 | ||
1860 | return new_dict | |
1861 | ||
e8e73840 | 1862 | def _check_formats(formats): |
981052c9 | 1863 | if not check_formats: |
1864 | yield from formats | |
b5ac45b1 | 1865 | return |
e8e73840 | 1866 | for f in formats: |
1867 | self.to_screen('[info] Testing format %s' % f['format_id']) | |
21cd8fae | 1868 | temp_file = tempfile.NamedTemporaryFile( |
1869 | suffix='.tmp', delete=False, | |
1870 | dir=self.get_output_path('temp') or None) | |
1871 | temp_file.close() | |
fe346461 | 1872 | try: |
981052c9 | 1873 | success, _ = self.dl(temp_file.name, f, test=True) |
1874 | except (DownloadError, IOError, OSError, ValueError) + network_exceptions: | |
1875 | success = False | |
fe346461 | 1876 | finally: |
21cd8fae | 1877 | if os.path.exists(temp_file.name): |
1878 | try: | |
1879 | os.remove(temp_file.name) | |
1880 | except OSError: | |
1881 | self.report_warning('Unable to delete temporary file "%s"' % temp_file.name) | |
981052c9 | 1882 | if success: |
e8e73840 | 1883 | yield f |
1884 | else: | |
1885 | self.to_screen('[info] Unable to download format %s. Skipping...' % f['format_id']) | |
1886 | ||
67134eab | 1887 | def _build_selector_function(selector): |
909d24dd | 1888 | if isinstance(selector, list): # , |
67134eab JMF |
1889 | fs = [_build_selector_function(s) for s in selector] |
1890 | ||
317f7ab6 | 1891 | def selector_function(ctx): |
67134eab | 1892 | for f in fs: |
981052c9 | 1893 | yield from f(ctx) |
67134eab | 1894 | return selector_function |
909d24dd | 1895 | |
1896 | elif selector.type == GROUP: # () | |
0130afb7 | 1897 | selector_function = _build_selector_function(selector.selector) |
909d24dd | 1898 | |
1899 | elif selector.type == PICKFIRST: # / | |
67134eab JMF |
1900 | fs = [_build_selector_function(s) for s in selector.selector] |
1901 | ||
317f7ab6 | 1902 | def selector_function(ctx): |
67134eab | 1903 | for f in fs: |
317f7ab6 | 1904 | picked_formats = list(f(ctx)) |
67134eab JMF |
1905 | if picked_formats: |
1906 | return picked_formats | |
1907 | return [] | |
67134eab | 1908 | |
981052c9 | 1909 | elif selector.type == MERGE: # + |
1910 | selector_1, selector_2 = map(_build_selector_function, selector.selector) | |
1911 | ||
1912 | def selector_function(ctx): | |
1913 | for pair in itertools.product( | |
1914 | selector_1(copy.deepcopy(ctx)), selector_2(copy.deepcopy(ctx))): | |
1915 | yield _merge(pair) | |
1916 | ||
909d24dd | 1917 | elif selector.type == SINGLE: # atom |
598d185d | 1918 | format_spec = selector.selector or 'best' |
909d24dd | 1919 | |
f8d4ad9a | 1920 | # TODO: Add allvideo, allaudio etc by generalizing the code with best/worst selector |
909d24dd | 1921 | if format_spec == 'all': |
1922 | def selector_function(ctx): | |
981052c9 | 1923 | yield from _check_formats(ctx['formats']) |
f8d4ad9a | 1924 | elif format_spec == 'mergeall': |
1925 | def selector_function(ctx): | |
981052c9 | 1926 | formats = list(_check_formats(ctx['formats'])) |
e01d6aa4 | 1927 | if not formats: |
1928 | return | |
921b76ca | 1929 | merged_format = formats[-1] |
1930 | for f in formats[-2::-1]: | |
f8d4ad9a | 1931 | merged_format = _merge((merged_format, f)) |
1932 | yield merged_format | |
909d24dd | 1933 | |
1934 | else: | |
e8e73840 | 1935 | format_fallback, format_reverse, format_idx = False, True, 1 |
eff63539 | 1936 | mobj = re.match( |
1937 | r'(?P<bw>best|worst|b|w)(?P<type>video|audio|v|a)?(?P<mod>\*)?(?:\.(?P<n>[1-9]\d*))?$', | |
1938 | format_spec) | |
1939 | if mobj is not None: | |
1940 | format_idx = int_or_none(mobj.group('n'), default=1) | |
e8e73840 | 1941 | format_reverse = mobj.group('bw')[0] == 'b' |
eff63539 | 1942 | format_type = (mobj.group('type') or [None])[0] |
1943 | not_format_type = {'v': 'a', 'a': 'v'}.get(format_type) | |
1944 | format_modified = mobj.group('mod') is not None | |
909d24dd | 1945 | |
1946 | format_fallback = not format_type and not format_modified # for b, w | |
8326b00a | 1947 | _filter_f = ( |
eff63539 | 1948 | (lambda f: f.get('%scodec' % format_type) != 'none') |
1949 | if format_type and format_modified # bv*, ba*, wv*, wa* | |
1950 | else (lambda f: f.get('%scodec' % not_format_type) == 'none') | |
1951 | if format_type # bv, ba, wv, wa | |
1952 | else (lambda f: f.get('vcodec') != 'none' and f.get('acodec') != 'none') | |
1953 | if not format_modified # b, w | |
8326b00a | 1954 | else lambda f: True) # b*, w* |
1955 | filter_f = lambda f: _filter_f(f) and ( | |
1956 | f.get('vcodec') != 'none' or f.get('acodec') != 'none') | |
67134eab | 1957 | else: |
909d24dd | 1958 | filter_f = ((lambda f: f.get('ext') == format_spec) |
1959 | if format_spec in ['mp4', 'flv', 'webm', '3gp', 'm4a', 'mp3', 'ogg', 'aac', 'wav'] # extension | |
1960 | else (lambda f: f.get('format_id') == format_spec)) # id | |
1961 | ||
1962 | def selector_function(ctx): | |
1963 | formats = list(ctx['formats']) | |
909d24dd | 1964 | matches = list(filter(filter_f, formats)) if filter_f is not None else formats |
e8e73840 | 1965 | if format_fallback and ctx['incomplete_formats'] and not matches: |
909d24dd | 1966 | # for extractors with incomplete formats (audio only (soundcloud) |
1967 | # or video only (imgur)) best/worst will fallback to | |
1968 | # best/worst {video,audio}-only format | |
e8e73840 | 1969 | matches = formats |
981052c9 | 1970 | matches = LazyList(_check_formats(matches[::-1 if format_reverse else 1])) |
1971 | try: | |
e8e73840 | 1972 | yield matches[format_idx - 1] |
981052c9 | 1973 | except IndexError: |
1974 | return | |
083c9df9 | 1975 | |
67134eab | 1976 | filters = [self._build_format_filter(f) for f in selector.filters] |
083c9df9 | 1977 | |
317f7ab6 S |
1978 | def final_selector(ctx): |
1979 | ctx_copy = copy.deepcopy(ctx) | |
67134eab | 1980 | for _filter in filters: |
317f7ab6 S |
1981 | ctx_copy['formats'] = list(filter(_filter, ctx_copy['formats'])) |
1982 | return selector_function(ctx_copy) | |
67134eab | 1983 | return final_selector |
083c9df9 | 1984 | |
67134eab | 1985 | stream = io.BytesIO(format_spec.encode('utf-8')) |
0130afb7 | 1986 | try: |
232541df | 1987 | tokens = list(_remove_unused_ops(compat_tokenize_tokenize(stream.readline))) |
0130afb7 JMF |
1988 | except tokenize.TokenError: |
1989 | raise syntax_error('Missing closing/opening brackets or parenthesis', (0, len(format_spec))) | |
1990 | ||
1991 | class TokenIterator(object): | |
1992 | def __init__(self, tokens): | |
1993 | self.tokens = tokens | |
1994 | self.counter = 0 | |
1995 | ||
1996 | def __iter__(self): | |
1997 | return self | |
1998 | ||
1999 | def __next__(self): | |
2000 | if self.counter >= len(self.tokens): | |
2001 | raise StopIteration() | |
2002 | value = self.tokens[self.counter] | |
2003 | self.counter += 1 | |
2004 | return value | |
2005 | ||
2006 | next = __next__ | |
2007 | ||
2008 | def restore_last_token(self): | |
2009 | self.counter -= 1 | |
2010 | ||
2011 | parsed_selector = _parse_format_selection(iter(TokenIterator(tokens))) | |
67134eab | 2012 | return _build_selector_function(parsed_selector) |
a9c58ad9 | 2013 | |
e5660ee6 JMF |
2014 | def _calc_headers(self, info_dict): |
2015 | res = std_headers.copy() | |
2016 | ||
2017 | add_headers = info_dict.get('http_headers') | |
2018 | if add_headers: | |
2019 | res.update(add_headers) | |
2020 | ||
2021 | cookies = self._calc_cookies(info_dict) | |
2022 | if cookies: | |
2023 | res['Cookie'] = cookies | |
2024 | ||
0016b84e S |
2025 | if 'X-Forwarded-For' not in res: |
2026 | x_forwarded_for_ip = info_dict.get('__x_forwarded_for_ip') | |
2027 | if x_forwarded_for_ip: | |
2028 | res['X-Forwarded-For'] = x_forwarded_for_ip | |
2029 | ||
e5660ee6 JMF |
2030 | return res |
2031 | ||
2032 | def _calc_cookies(self, info_dict): | |
5c2266df | 2033 | pr = sanitized_Request(info_dict['url']) |
e5660ee6 | 2034 | self.cookiejar.add_cookie_header(pr) |
662435f7 | 2035 | return pr.get_header('Cookie') |
e5660ee6 | 2036 | |
b0249bca | 2037 | def _sanitize_thumbnails(self, info_dict): |
bc516a3f | 2038 | thumbnails = info_dict.get('thumbnails') |
2039 | if thumbnails is None: | |
2040 | thumbnail = info_dict.get('thumbnail') | |
2041 | if thumbnail: | |
2042 | info_dict['thumbnails'] = thumbnails = [{'url': thumbnail}] | |
2043 | if thumbnails: | |
2044 | thumbnails.sort(key=lambda t: ( | |
2045 | t.get('preference') if t.get('preference') is not None else -1, | |
2046 | t.get('width') if t.get('width') is not None else -1, | |
2047 | t.get('height') if t.get('height') is not None else -1, | |
2048 | t.get('id') if t.get('id') is not None else '', | |
2049 | t.get('url'))) | |
b0249bca | 2050 | |
0ba692ac | 2051 | def thumbnail_tester(): |
2052 | if self.params.get('check_formats'): | |
cca80fe6 | 2053 | test_all = True |
2054 | to_screen = lambda msg: self.to_screen(f'[info] {msg}') | |
0ba692ac | 2055 | else: |
cca80fe6 | 2056 | test_all = False |
0ba692ac | 2057 | to_screen = self.write_debug |
2058 | ||
2059 | def test_thumbnail(t): | |
cca80fe6 | 2060 | if not test_all and not t.get('_test_url'): |
2061 | return True | |
0ba692ac | 2062 | to_screen('Testing thumbnail %s' % t['id']) |
2063 | try: | |
2064 | self.urlopen(HEADRequest(t['url'])) | |
2065 | except network_exceptions as err: | |
2066 | to_screen('Unable to connect to thumbnail %s URL "%s" - %s. Skipping...' % ( | |
2067 | t['id'], t['url'], error_to_compat_str(err))) | |
2068 | return False | |
2069 | return True | |
2070 | ||
2071 | return test_thumbnail | |
b0249bca | 2072 | |
bc516a3f | 2073 | for i, t in enumerate(thumbnails): |
bc516a3f | 2074 | if t.get('id') is None: |
2075 | t['id'] = '%d' % i | |
b0249bca | 2076 | if t.get('width') and t.get('height'): |
2077 | t['resolution'] = '%dx%d' % (t['width'], t['height']) | |
2078 | t['url'] = sanitize_url(t['url']) | |
0ba692ac | 2079 | |
2080 | if self.params.get('check_formats') is not False: | |
2081 | info_dict['thumbnails'] = LazyList(filter(thumbnail_tester(), thumbnails[::-1])).reverse() | |
2082 | else: | |
2083 | info_dict['thumbnails'] = thumbnails | |
bc516a3f | 2084 | |
dd82ffea JMF |
2085 | def process_video_result(self, info_dict, download=True): |
2086 | assert info_dict.get('_type', 'video') == 'video' | |
2087 | ||
bec1fad2 PH |
2088 | if 'id' not in info_dict: |
2089 | raise ExtractorError('Missing "id" field in extractor result') | |
2090 | if 'title' not in info_dict: | |
1151c407 | 2091 | raise ExtractorError('Missing "title" field in extractor result', |
2092 | video_id=info_dict['id'], ie=info_dict['extractor']) | |
bec1fad2 | 2093 | |
c9969434 S |
2094 | def report_force_conversion(field, field_not, conversion): |
2095 | self.report_warning( | |
2096 | '"%s" field is not %s - forcing %s conversion, there is an error in extractor' | |
2097 | % (field, field_not, conversion)) | |
2098 | ||
2099 | def sanitize_string_field(info, string_field): | |
2100 | field = info.get(string_field) | |
2101 | if field is None or isinstance(field, compat_str): | |
2102 | return | |
2103 | report_force_conversion(string_field, 'a string', 'string') | |
2104 | info[string_field] = compat_str(field) | |
2105 | ||
2106 | def sanitize_numeric_fields(info): | |
2107 | for numeric_field in self._NUMERIC_FIELDS: | |
2108 | field = info.get(numeric_field) | |
2109 | if field is None or isinstance(field, compat_numeric_types): | |
2110 | continue | |
2111 | report_force_conversion(numeric_field, 'numeric', 'int') | |
2112 | info[numeric_field] = int_or_none(field) | |
2113 | ||
2114 | sanitize_string_field(info_dict, 'id') | |
2115 | sanitize_numeric_fields(info_dict) | |
be6217b2 | 2116 | |
dd82ffea JMF |
2117 | if 'playlist' not in info_dict: |
2118 | # It isn't part of a playlist | |
2119 | info_dict['playlist'] = None | |
2120 | info_dict['playlist_index'] = None | |
2121 | ||
bc516a3f | 2122 | self._sanitize_thumbnails(info_dict) |
d5519808 | 2123 | |
536a55da | 2124 | thumbnail = info_dict.get('thumbnail') |
bc516a3f | 2125 | thumbnails = info_dict.get('thumbnails') |
536a55da S |
2126 | if thumbnail: |
2127 | info_dict['thumbnail'] = sanitize_url(thumbnail) | |
2128 | elif thumbnails: | |
d5519808 PH |
2129 | info_dict['thumbnail'] = thumbnails[-1]['url'] |
2130 | ||
ae30b840 | 2131 | if info_dict.get('display_id') is None and 'id' in info_dict: |
0afef30b PH |
2132 | info_dict['display_id'] = info_dict['id'] |
2133 | ||
10db0d2f | 2134 | for ts_key, date_key in ( |
2135 | ('timestamp', 'upload_date'), | |
2136 | ('release_timestamp', 'release_date'), | |
2137 | ): | |
2138 | if info_dict.get(date_key) is None and info_dict.get(ts_key) is not None: | |
2139 | # Working around out-of-range timestamp values (e.g. negative ones on Windows, | |
2140 | # see http://bugs.python.org/issue1646728) | |
2141 | try: | |
2142 | upload_date = datetime.datetime.utcfromtimestamp(info_dict[ts_key]) | |
2143 | info_dict[date_key] = upload_date.strftime('%Y%m%d') | |
2144 | except (ValueError, OverflowError, OSError): | |
2145 | pass | |
9d2ecdbc | 2146 | |
ae30b840 | 2147 | live_keys = ('is_live', 'was_live') |
2148 | live_status = info_dict.get('live_status') | |
2149 | if live_status is None: | |
2150 | for key in live_keys: | |
2151 | if info_dict.get(key) is False: | |
2152 | continue | |
2153 | if info_dict.get(key): | |
2154 | live_status = key | |
2155 | break | |
2156 | if all(info_dict.get(key) is False for key in live_keys): | |
2157 | live_status = 'not_live' | |
2158 | if live_status: | |
2159 | info_dict['live_status'] = live_status | |
2160 | for key in live_keys: | |
2161 | if info_dict.get(key) is None: | |
2162 | info_dict[key] = (live_status == key) | |
2163 | ||
33d2fc2f S |
2164 | # Auto generate title fields corresponding to the *_number fields when missing |
2165 | # in order to always have clean titles. This is very common for TV series. | |
2166 | for field in ('chapter', 'season', 'episode'): | |
2167 | if info_dict.get('%s_number' % field) is not None and not info_dict.get(field): | |
2168 | info_dict[field] = '%s %d' % (field.capitalize(), info_dict['%s_number' % field]) | |
2169 | ||
05108a49 S |
2170 | for cc_kind in ('subtitles', 'automatic_captions'): |
2171 | cc = info_dict.get(cc_kind) | |
2172 | if cc: | |
2173 | for _, subtitle in cc.items(): | |
2174 | for subtitle_format in subtitle: | |
2175 | if subtitle_format.get('url'): | |
2176 | subtitle_format['url'] = sanitize_url(subtitle_format['url']) | |
2177 | if subtitle_format.get('ext') is None: | |
2178 | subtitle_format['ext'] = determine_ext(subtitle_format['url']).lower() | |
2179 | ||
2180 | automatic_captions = info_dict.get('automatic_captions') | |
4bba3716 | 2181 | subtitles = info_dict.get('subtitles') |
4bba3716 | 2182 | |
360e1ca5 | 2183 | info_dict['requested_subtitles'] = self.process_subtitles( |
05108a49 | 2184 | info_dict['id'], subtitles, automatic_captions) |
a504ced0 | 2185 | |
dd82ffea JMF |
2186 | # We now pick which formats have to be downloaded |
2187 | if info_dict.get('formats') is None: | |
2188 | # There's only one format available | |
2189 | formats = [info_dict] | |
2190 | else: | |
2191 | formats = info_dict['formats'] | |
2192 | ||
e0493e90 | 2193 | info_dict['__has_drm'] = any(f.get('has_drm') for f in formats) |
88acdbc2 | 2194 | if not self.params.get('allow_unplayable_formats'): |
2195 | formats = [f for f in formats if not f.get('has_drm')] | |
88acdbc2 | 2196 | |
db95dc13 | 2197 | if not formats: |
1151c407 | 2198 | self.raise_no_formats(info_dict) |
db95dc13 | 2199 | |
73af5cc8 S |
2200 | def is_wellformed(f): |
2201 | url = f.get('url') | |
a5ac0c47 | 2202 | if not url: |
73af5cc8 S |
2203 | self.report_warning( |
2204 | '"url" field is missing or empty - skipping format, ' | |
2205 | 'there is an error in extractor') | |
a5ac0c47 S |
2206 | return False |
2207 | if isinstance(url, bytes): | |
2208 | sanitize_string_field(f, 'url') | |
2209 | return True | |
73af5cc8 S |
2210 | |
2211 | # Filter out malformed formats for better extraction robustness | |
2212 | formats = list(filter(is_wellformed, formats)) | |
2213 | ||
181c7053 S |
2214 | formats_dict = {} |
2215 | ||
dd82ffea | 2216 | # We check that all the formats have the format and format_id fields |
db95dc13 | 2217 | for i, format in enumerate(formats): |
c9969434 S |
2218 | sanitize_string_field(format, 'format_id') |
2219 | sanitize_numeric_fields(format) | |
dcf77cf1 | 2220 | format['url'] = sanitize_url(format['url']) |
e74e3b63 | 2221 | if not format.get('format_id'): |
8016c922 | 2222 | format['format_id'] = compat_str(i) |
e2effb08 S |
2223 | else: |
2224 | # Sanitize format_id from characters used in format selector expression | |
ec85ded8 | 2225 | format['format_id'] = re.sub(r'[\s,/+\[\]()]', '_', format['format_id']) |
181c7053 S |
2226 | format_id = format['format_id'] |
2227 | if format_id not in formats_dict: | |
2228 | formats_dict[format_id] = [] | |
2229 | formats_dict[format_id].append(format) | |
2230 | ||
2231 | # Make sure all formats have unique format_id | |
2232 | for format_id, ambiguous_formats in formats_dict.items(): | |
2233 | if len(ambiguous_formats) > 1: | |
2234 | for i, format in enumerate(ambiguous_formats): | |
2235 | format['format_id'] = '%s-%d' % (format_id, i) | |
2236 | ||
2237 | for i, format in enumerate(formats): | |
8c51aa65 | 2238 | if format.get('format') is None: |
6febd1c1 | 2239 | format['format'] = '{id} - {res}{note}'.format( |
8c51aa65 JMF |
2240 | id=format['format_id'], |
2241 | res=self.format_resolution(format), | |
b868936c | 2242 | note=format_field(format, 'format_note', ' (%s)'), |
8c51aa65 | 2243 | ) |
c1002e96 | 2244 | # Automatically determine file extension if missing |
5b1d8575 | 2245 | if format.get('ext') is None: |
cce929ea | 2246 | format['ext'] = determine_ext(format['url']).lower() |
b5559424 S |
2247 | # Automatically determine protocol if missing (useful for format |
2248 | # selection purposes) | |
6f0be937 | 2249 | if format.get('protocol') is None: |
b5559424 | 2250 | format['protocol'] = determine_protocol(format) |
e5660ee6 JMF |
2251 | # Add HTTP headers, so that external programs can use them from the |
2252 | # json output | |
2253 | full_format_info = info_dict.copy() | |
2254 | full_format_info.update(format) | |
2255 | format['http_headers'] = self._calc_headers(full_format_info) | |
0016b84e S |
2256 | # Remove private housekeeping stuff |
2257 | if '__x_forwarded_for_ip' in info_dict: | |
2258 | del info_dict['__x_forwarded_for_ip'] | |
dd82ffea | 2259 | |
4bcc7bd1 | 2260 | # TODO Central sorting goes here |
99e206d5 | 2261 | |
88acdbc2 | 2262 | if not formats or formats[0] is not info_dict: |
b3d9ef88 JMF |
2263 | # only set the 'formats' fields if the original info_dict list them |
2264 | # otherwise we end up with a circular reference, the first (and unique) | |
f89197d7 | 2265 | # element in the 'formats' field in info_dict is info_dict itself, |
dfb1b146 | 2266 | # which can't be exported to json |
b3d9ef88 | 2267 | info_dict['formats'] = formats |
4ec82a72 | 2268 | |
2269 | info_dict, _ = self.pre_process(info_dict) | |
2270 | ||
b7b04c78 | 2271 | if self.params.get('list_thumbnails'): |
2272 | self.list_thumbnails(info_dict) | |
2273 | if self.params.get('listformats'): | |
86c66b2d | 2274 | if not info_dict.get('formats') and not info_dict.get('url'): |
88acdbc2 | 2275 | self.to_screen('%s has no formats' % info_dict['id']) |
2276 | else: | |
2277 | self.list_formats(info_dict) | |
b7b04c78 | 2278 | if self.params.get('listsubtitles'): |
2279 | if 'automatic_captions' in info_dict: | |
2280 | self.list_subtitles( | |
2281 | info_dict['id'], automatic_captions, 'automatic captions') | |
2282 | self.list_subtitles(info_dict['id'], subtitles, 'subtitles') | |
2283 | list_only = self.params.get('simulate') is None and ( | |
2284 | self.params.get('list_thumbnails') or self.params.get('listformats') or self.params.get('listsubtitles')) | |
169dbde9 | 2285 | if list_only: |
b7b04c78 | 2286 | # Without this printing, -F --print-json will not work |
169dbde9 | 2287 | self.__forced_printings(info_dict, self.prepare_filename(info_dict), incomplete=True) |
bfaae0a7 | 2288 | return |
2289 | ||
187986a8 | 2290 | format_selector = self.format_selector |
2291 | if format_selector is None: | |
0017d9ad | 2292 | req_format = self._default_format_spec(info_dict, download=download) |
0760b0a7 | 2293 | self.write_debug('Default format spec: %s' % req_format) |
187986a8 | 2294 | format_selector = self.build_format_selector(req_format) |
317f7ab6 S |
2295 | |
2296 | # While in format selection we may need to have an access to the original | |
2297 | # format set in order to calculate some metrics or do some processing. | |
2298 | # For now we need to be able to guess whether original formats provided | |
2299 | # by extractor are incomplete or not (i.e. whether extractor provides only | |
2300 | # video-only or audio-only formats) for proper formats selection for | |
2301 | # extractors with such incomplete formats (see | |
067aa17e | 2302 | # https://github.com/ytdl-org/youtube-dl/pull/5556). |
317f7ab6 S |
2303 | # Since formats may be filtered during format selection and may not match |
2304 | # the original formats the results may be incorrect. Thus original formats | |
2305 | # or pre-calculated metrics should be passed to format selection routines | |
2306 | # as well. | |
2307 | # We will pass a context object containing all necessary additional data | |
2308 | # instead of just formats. | |
2309 | # This fixes incorrect format selection issue (see | |
067aa17e | 2310 | # https://github.com/ytdl-org/youtube-dl/issues/10083). |
2e221ca3 | 2311 | incomplete_formats = ( |
317f7ab6 | 2312 | # All formats are video-only or |
3089bc74 | 2313 | all(f.get('vcodec') != 'none' and f.get('acodec') == 'none' for f in formats) |
317f7ab6 | 2314 | # all formats are audio-only |
3089bc74 | 2315 | or all(f.get('vcodec') == 'none' and f.get('acodec') != 'none' for f in formats)) |
317f7ab6 S |
2316 | |
2317 | ctx = { | |
2318 | 'formats': formats, | |
2319 | 'incomplete_formats': incomplete_formats, | |
2320 | } | |
2321 | ||
2322 | formats_to_download = list(format_selector(ctx)) | |
dd82ffea | 2323 | if not formats_to_download: |
b7da73eb | 2324 | if not self.params.get('ignore_no_formats_error'): |
1151c407 | 2325 | raise ExtractorError('Requested format is not available', expected=True, |
2326 | video_id=info_dict['id'], ie=info_dict['extractor']) | |
b7da73eb | 2327 | else: |
2328 | self.report_warning('Requested format is not available') | |
4513a41a A |
2329 | # Process what we can, even without any available formats. |
2330 | self.process_info(dict(info_dict)) | |
b7da73eb | 2331 | elif download: |
2332 | self.to_screen( | |
07cce701 | 2333 | '[info] %s: Downloading %d format(s): %s' % ( |
2334 | info_dict['id'], len(formats_to_download), | |
2335 | ", ".join([f['format_id'] for f in formats_to_download]))) | |
b7da73eb | 2336 | for fmt in formats_to_download: |
dd82ffea | 2337 | new_info = dict(info_dict) |
4ec82a72 | 2338 | # Save a reference to the original info_dict so that it can be modified in process_info if needed |
2339 | new_info['__original_infodict'] = info_dict | |
b7da73eb | 2340 | new_info.update(fmt) |
dd82ffea JMF |
2341 | self.process_info(new_info) |
2342 | # We update the info dict with the best quality format (backwards compatibility) | |
b7da73eb | 2343 | if formats_to_download: |
2344 | info_dict.update(formats_to_download[-1]) | |
dd82ffea JMF |
2345 | return info_dict |
2346 | ||
98c70d6f | 2347 | def process_subtitles(self, video_id, normal_subtitles, automatic_captions): |
a504ced0 | 2348 | """Select the requested subtitles and their format""" |
98c70d6f JMF |
2349 | available_subs = {} |
2350 | if normal_subtitles and self.params.get('writesubtitles'): | |
2351 | available_subs.update(normal_subtitles) | |
2352 | if automatic_captions and self.params.get('writeautomaticsub'): | |
2353 | for lang, cap_info in automatic_captions.items(): | |
360e1ca5 JMF |
2354 | if lang not in available_subs: |
2355 | available_subs[lang] = cap_info | |
2356 | ||
4d171848 JMF |
2357 | if (not self.params.get('writesubtitles') and not |
2358 | self.params.get('writeautomaticsub') or not | |
2359 | available_subs): | |
2360 | return None | |
a504ced0 | 2361 | |
c32b0aab | 2362 | all_sub_langs = available_subs.keys() |
a504ced0 | 2363 | if self.params.get('allsubtitles', False): |
c32b0aab | 2364 | requested_langs = all_sub_langs |
2365 | elif self.params.get('subtitleslangs', False): | |
77c4a9ef | 2366 | # A list is used so that the order of languages will be the same as |
2367 | # given in subtitleslangs. See https://github.com/yt-dlp/yt-dlp/issues/1041 | |
2368 | requested_langs = [] | |
2369 | for lang_re in self.params.get('subtitleslangs'): | |
2370 | if lang_re == 'all': | |
2371 | requested_langs.extend(all_sub_langs) | |
c32b0aab | 2372 | continue |
77c4a9ef | 2373 | discard = lang_re[0] == '-' |
c32b0aab | 2374 | if discard: |
77c4a9ef | 2375 | lang_re = lang_re[1:] |
2376 | current_langs = filter(re.compile(lang_re + '$').match, all_sub_langs) | |
c32b0aab | 2377 | if discard: |
2378 | for lang in current_langs: | |
77c4a9ef | 2379 | while lang in requested_langs: |
2380 | requested_langs.remove(lang) | |
c32b0aab | 2381 | else: |
77c4a9ef | 2382 | requested_langs.extend(current_langs) |
2383 | requested_langs = orderedSet(requested_langs) | |
c32b0aab | 2384 | elif 'en' in available_subs: |
2385 | requested_langs = ['en'] | |
a504ced0 | 2386 | else: |
c32b0aab | 2387 | requested_langs = [list(all_sub_langs)[0]] |
ad3dc496 | 2388 | if requested_langs: |
2389 | self.write_debug('Downloading subtitles: %s' % ', '.join(requested_langs)) | |
a504ced0 JMF |
2390 | |
2391 | formats_query = self.params.get('subtitlesformat', 'best') | |
2392 | formats_preference = formats_query.split('/') if formats_query else [] | |
2393 | subs = {} | |
2394 | for lang in requested_langs: | |
2395 | formats = available_subs.get(lang) | |
2396 | if formats is None: | |
2397 | self.report_warning('%s subtitles not available for %s' % (lang, video_id)) | |
2398 | continue | |
a504ced0 JMF |
2399 | for ext in formats_preference: |
2400 | if ext == 'best': | |
2401 | f = formats[-1] | |
2402 | break | |
2403 | matches = list(filter(lambda f: f['ext'] == ext, formats)) | |
2404 | if matches: | |
2405 | f = matches[-1] | |
2406 | break | |
2407 | else: | |
2408 | f = formats[-1] | |
2409 | self.report_warning( | |
2410 | 'No subtitle format found matching "%s" for language %s, ' | |
2411 | 'using %s' % (formats_query, lang, f['ext'])) | |
2412 | subs[lang] = f | |
2413 | return subs | |
2414 | ||
d06daf23 | 2415 | def __forced_printings(self, info_dict, filename, incomplete): |
53c18592 | 2416 | def print_mandatory(field, actual_field=None): |
2417 | if actual_field is None: | |
2418 | actual_field = field | |
d06daf23 | 2419 | if (self.params.get('force%s' % field, False) |
53c18592 | 2420 | and (not incomplete or info_dict.get(actual_field) is not None)): |
2421 | self.to_stdout(info_dict[actual_field]) | |
d06daf23 S |
2422 | |
2423 | def print_optional(field): | |
2424 | if (self.params.get('force%s' % field, False) | |
2425 | and info_dict.get(field) is not None): | |
2426 | self.to_stdout(info_dict[field]) | |
2427 | ||
53c18592 | 2428 | info_dict = info_dict.copy() |
2429 | if filename is not None: | |
2430 | info_dict['filename'] = filename | |
2431 | if info_dict.get('requested_formats') is not None: | |
2432 | # For RTMP URLs, also include the playpath | |
2433 | info_dict['urls'] = '\n'.join(f['url'] + f.get('play_path', '') for f in info_dict['requested_formats']) | |
2434 | elif 'url' in info_dict: | |
2435 | info_dict['urls'] = info_dict['url'] + info_dict.get('play_path', '') | |
2436 | ||
2b8a2973 | 2437 | if self.params.get('forceprint') or self.params.get('forcejson'): |
2438 | self.post_extract(info_dict) | |
53c18592 | 2439 | for tmpl in self.params.get('forceprint', []): |
2440 | if re.match(r'\w+$', tmpl): | |
2441 | tmpl = '%({})s'.format(tmpl) | |
2442 | tmpl, info_copy = self.prepare_outtmpl(tmpl, info_dict) | |
901130bb | 2443 | self.to_stdout(self.escape_outtmpl(tmpl) % info_copy) |
53c18592 | 2444 | |
d06daf23 S |
2445 | print_mandatory('title') |
2446 | print_mandatory('id') | |
53c18592 | 2447 | print_mandatory('url', 'urls') |
d06daf23 S |
2448 | print_optional('thumbnail') |
2449 | print_optional('description') | |
53c18592 | 2450 | print_optional('filename') |
b868936c | 2451 | if self.params.get('forceduration') and info_dict.get('duration') is not None: |
d06daf23 S |
2452 | self.to_stdout(formatSeconds(info_dict['duration'])) |
2453 | print_mandatory('format') | |
53c18592 | 2454 | |
2b8a2973 | 2455 | if self.params.get('forcejson'): |
6e84b215 | 2456 | self.to_stdout(json.dumps(self.sanitize_info(info_dict))) |
d06daf23 | 2457 | |
e8e73840 | 2458 | def dl(self, name, info, subtitle=False, test=False): |
88acdbc2 | 2459 | if not info.get('url'): |
1151c407 | 2460 | self.raise_no_formats(info, True) |
e8e73840 | 2461 | |
2462 | if test: | |
2463 | verbose = self.params.get('verbose') | |
2464 | params = { | |
2465 | 'test': True, | |
2466 | 'quiet': not verbose, | |
2467 | 'verbose': verbose, | |
2468 | 'noprogress': not verbose, | |
2469 | 'nopart': True, | |
2470 | 'skip_unavailable_fragments': False, | |
2471 | 'keep_fragments': False, | |
2472 | 'overwrites': True, | |
2473 | '_no_ytdl_file': True, | |
2474 | } | |
2475 | else: | |
2476 | params = self.params | |
96fccc10 | 2477 | fd = get_suitable_downloader(info, params, to_stdout=(name == '-'))(self, params) |
e8e73840 | 2478 | if not test: |
2479 | for ph in self._progress_hooks: | |
2480 | fd.add_progress_hook(ph) | |
18e674b4 | 2481 | urls = '", "'.join([f['url'] for f in info.get('requested_formats', [])] or [info['url']]) |
2482 | self.write_debug('Invoking downloader on "%s"' % urls) | |
e8e73840 | 2483 | new_info = dict(info) |
2484 | if new_info.get('http_headers') is None: | |
2485 | new_info['http_headers'] = self._calc_headers(new_info) | |
2486 | return fd.download(name, new_info, subtitle) | |
2487 | ||
8222d8de JMF |
2488 | def process_info(self, info_dict): |
2489 | """Process a single resolved IE result.""" | |
2490 | ||
2491 | assert info_dict.get('_type', 'video') == 'video' | |
fd288278 PH |
2492 | |
2493 | max_downloads = self.params.get('max_downloads') | |
2494 | if max_downloads is not None: | |
2495 | if self._num_downloads >= int(max_downloads): | |
2496 | raise MaxDownloadsReached() | |
8222d8de | 2497 | |
d06daf23 | 2498 | # TODO: backward compatibility, to be removed |
8222d8de | 2499 | info_dict['fulltitle'] = info_dict['title'] |
8222d8de | 2500 | |
4513a41a | 2501 | if 'format' not in info_dict and 'ext' in info_dict: |
8222d8de JMF |
2502 | info_dict['format'] = info_dict['ext'] |
2503 | ||
c77495e3 | 2504 | if self._match_entry(info_dict) is not None: |
8222d8de JMF |
2505 | return |
2506 | ||
277d6ff5 | 2507 | self.post_extract(info_dict) |
fd288278 | 2508 | self._num_downloads += 1 |
8222d8de | 2509 | |
dcf64d43 | 2510 | # info_dict['_filename'] needs to be set for backward compatibility |
de6000d9 | 2511 | info_dict['_filename'] = full_filename = self.prepare_filename(info_dict, warn=True) |
2512 | temp_filename = self.prepare_filename(info_dict, 'temp') | |
0202b52a | 2513 | files_to_move = {} |
8222d8de JMF |
2514 | |
2515 | # Forced printings | |
4513a41a | 2516 | self.__forced_printings(info_dict, full_filename, incomplete=('format' not in info_dict)) |
8222d8de | 2517 | |
b7b04c78 | 2518 | if self.params.get('simulate'): |
2d30509f | 2519 | if self.params.get('force_write_download_archive', False): |
2520 | self.record_download_archive(info_dict) | |
2521 | ||
2522 | # Do nothing else if in simulate mode | |
8222d8de JMF |
2523 | return |
2524 | ||
de6000d9 | 2525 | if full_filename is None: |
8222d8de JMF |
2526 | return |
2527 | ||
e92caff5 | 2528 | if not self._ensure_dir_exists(encodeFilename(full_filename)): |
0202b52a | 2529 | return |
e92caff5 | 2530 | if not self._ensure_dir_exists(encodeFilename(temp_filename)): |
8222d8de JMF |
2531 | return |
2532 | ||
2533 | if self.params.get('writedescription', False): | |
de6000d9 | 2534 | descfn = self.prepare_filename(info_dict, 'description') |
e92caff5 | 2535 | if not self._ensure_dir_exists(encodeFilename(descfn)): |
0202b52a | 2536 | return |
0c3d0f51 | 2537 | if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(descfn)): |
6febd1c1 | 2538 | self.to_screen('[info] Video description is already present') |
f00fd51d JMF |
2539 | elif info_dict.get('description') is None: |
2540 | self.report_warning('There\'s no description to write.') | |
7b6fefc9 PH |
2541 | else: |
2542 | try: | |
6febd1c1 | 2543 | self.to_screen('[info] Writing video description to: ' + descfn) |
7b6fefc9 PH |
2544 | with io.open(encodeFilename(descfn), 'w', encoding='utf-8') as descfile: |
2545 | descfile.write(info_dict['description']) | |
7b6fefc9 | 2546 | except (OSError, IOError): |
6febd1c1 | 2547 | self.report_error('Cannot write description file ' + descfn) |
7b6fefc9 | 2548 | return |
8222d8de | 2549 | |
1fb07d10 | 2550 | if self.params.get('writeannotations', False): |
de6000d9 | 2551 | annofn = self.prepare_filename(info_dict, 'annotation') |
e92caff5 | 2552 | if not self._ensure_dir_exists(encodeFilename(annofn)): |
0202b52a | 2553 | return |
0c3d0f51 | 2554 | if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(annofn)): |
6febd1c1 | 2555 | self.to_screen('[info] Video annotations are already present') |
ffddb112 RA |
2556 | elif not info_dict.get('annotations'): |
2557 | self.report_warning('There are no annotations to write.') | |
7b6fefc9 PH |
2558 | else: |
2559 | try: | |
6febd1c1 | 2560 | self.to_screen('[info] Writing video annotations to: ' + annofn) |
7b6fefc9 PH |
2561 | with io.open(encodeFilename(annofn), 'w', encoding='utf-8') as annofile: |
2562 | annofile.write(info_dict['annotations']) | |
2563 | except (KeyError, TypeError): | |
6febd1c1 | 2564 | self.report_warning('There are no annotations to write.') |
7b6fefc9 | 2565 | except (OSError, IOError): |
6febd1c1 | 2566 | self.report_error('Cannot write annotations file: ' + annofn) |
7b6fefc9 | 2567 | return |
1fb07d10 | 2568 | |
c4a91be7 | 2569 | subtitles_are_requested = any([self.params.get('writesubtitles', False), |
0b7f3118 | 2570 | self.params.get('writeautomaticsub')]) |
c4a91be7 | 2571 | |
c84dd8a9 | 2572 | if subtitles_are_requested and info_dict.get('requested_subtitles'): |
8222d8de JMF |
2573 | # subtitles download errors are already managed as troubles in relevant IE |
2574 | # that way it will silently go on when used with unsupporting IE | |
c84dd8a9 | 2575 | subtitles = info_dict['requested_subtitles'] |
fa57af1e | 2576 | # ie = self.get_info_extractor(info_dict['extractor_key']) |
a504ced0 JMF |
2577 | for sub_lang, sub_info in subtitles.items(): |
2578 | sub_format = sub_info['ext'] | |
56d868db | 2579 | sub_filename = subtitles_filename(temp_filename, sub_lang, sub_format, info_dict.get('ext')) |
2580 | sub_filename_final = subtitles_filename( | |
2581 | self.prepare_filename(info_dict, 'subtitle'), sub_lang, sub_format, info_dict.get('ext')) | |
0c3d0f51 | 2582 | if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(sub_filename)): |
5ff1bc0c | 2583 | self.to_screen('[info] Video subtitle %s.%s is already present' % (sub_lang, sub_format)) |
dcf64d43 | 2584 | sub_info['filepath'] = sub_filename |
0202b52a | 2585 | files_to_move[sub_filename] = sub_filename_final |
a504ced0 | 2586 | else: |
0c9df79e | 2587 | self.to_screen('[info] Writing video subtitles to: ' + sub_filename) |
5ff1bc0c RA |
2588 | if sub_info.get('data') is not None: |
2589 | try: | |
2590 | # Use newline='' to prevent conversion of newline characters | |
067aa17e | 2591 | # See https://github.com/ytdl-org/youtube-dl/issues/10268 |
5ff1bc0c RA |
2592 | with io.open(encodeFilename(sub_filename), 'w', encoding='utf-8', newline='') as subfile: |
2593 | subfile.write(sub_info['data']) | |
dcf64d43 | 2594 | sub_info['filepath'] = sub_filename |
0202b52a | 2595 | files_to_move[sub_filename] = sub_filename_final |
5ff1bc0c RA |
2596 | except (OSError, IOError): |
2597 | self.report_error('Cannot write subtitles file ' + sub_filename) | |
2598 | return | |
7b6fefc9 | 2599 | else: |
5ff1bc0c | 2600 | try: |
388bc4a6 AG |
2601 | sub_copy = sub_info.copy() |
2602 | sub_copy.setdefault('http_headers', info_dict.get('http_headers')) | |
2603 | self.dl(sub_filename, sub_copy, subtitle=True) | |
dcf64d43 | 2604 | sub_info['filepath'] = sub_filename |
0202b52a | 2605 | files_to_move[sub_filename] = sub_filename_final |
fe346461 | 2606 | except (ExtractorError, IOError, OSError, ValueError) + network_exceptions as err: |
5ff1bc0c RA |
2607 | self.report_warning('Unable to download subtitle for "%s": %s' % |
2608 | (sub_lang, error_to_compat_str(err))) | |
2609 | continue | |
8222d8de | 2610 | |
8222d8de | 2611 | if self.params.get('writeinfojson', False): |
de6000d9 | 2612 | infofn = self.prepare_filename(info_dict, 'infojson') |
e92caff5 | 2613 | if not self._ensure_dir_exists(encodeFilename(infofn)): |
0202b52a | 2614 | return |
0c3d0f51 | 2615 | if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(infofn)): |
66c935fb | 2616 | self.to_screen('[info] Video metadata is already present') |
7b6fefc9 | 2617 | else: |
66c935fb | 2618 | self.to_screen('[info] Writing video metadata as JSON to: ' + infofn) |
7b6fefc9 | 2619 | try: |
8012d892 | 2620 | write_json_file(self.sanitize_info(info_dict, self.params.get('clean_infojson', True)), infofn) |
7b6fefc9 | 2621 | except (OSError, IOError): |
66c935fb | 2622 | self.report_error('Cannot write video metadata to JSON file ' + infofn) |
7b6fefc9 | 2623 | return |
de6000d9 | 2624 | info_dict['__infojson_filename'] = infofn |
8222d8de | 2625 | |
56d868db | 2626 | for thumb_ext in self._write_thumbnails(info_dict, temp_filename): |
2627 | thumb_filename_temp = replace_extension(temp_filename, thumb_ext, info_dict.get('ext')) | |
2628 | thumb_filename = replace_extension( | |
2629 | self.prepare_filename(info_dict, 'thumbnail'), thumb_ext, info_dict.get('ext')) | |
dcf64d43 | 2630 | files_to_move[thumb_filename_temp] = thumb_filename |
8222d8de | 2631 | |
732044af | 2632 | # Write internet shortcut files |
2633 | url_link = webloc_link = desktop_link = False | |
2634 | if self.params.get('writelink', False): | |
2635 | if sys.platform == "darwin": # macOS. | |
2636 | webloc_link = True | |
2637 | elif sys.platform.startswith("linux"): | |
2638 | desktop_link = True | |
2639 | else: # if sys.platform in ['win32', 'cygwin']: | |
2640 | url_link = True | |
2641 | if self.params.get('writeurllink', False): | |
2642 | url_link = True | |
2643 | if self.params.get('writewebloclink', False): | |
2644 | webloc_link = True | |
2645 | if self.params.get('writedesktoplink', False): | |
2646 | desktop_link = True | |
2647 | ||
2648 | if url_link or webloc_link or desktop_link: | |
2649 | if 'webpage_url' not in info_dict: | |
2650 | self.report_error('Cannot write internet shortcut file because the "webpage_url" field is missing in the media information') | |
2651 | return | |
2652 | ascii_url = iri_to_uri(info_dict['webpage_url']) | |
2653 | ||
2654 | def _write_link_file(extension, template, newline, embed_filename): | |
0202b52a | 2655 | linkfn = replace_extension(full_filename, extension, info_dict.get('ext')) |
10e3742e | 2656 | if self.params.get('overwrites', True) and os.path.exists(encodeFilename(linkfn)): |
732044af | 2657 | self.to_screen('[info] Internet shortcut is already present') |
2658 | else: | |
2659 | try: | |
2660 | self.to_screen('[info] Writing internet shortcut to: ' + linkfn) | |
2661 | with io.open(encodeFilename(to_high_limit_path(linkfn)), 'w', encoding='utf-8', newline=newline) as linkfile: | |
2662 | template_vars = {'url': ascii_url} | |
2663 | if embed_filename: | |
2664 | template_vars['filename'] = linkfn[:-(len(extension) + 1)] | |
2665 | linkfile.write(template % template_vars) | |
2666 | except (OSError, IOError): | |
2667 | self.report_error('Cannot write internet shortcut ' + linkfn) | |
2668 | return False | |
2669 | return True | |
2670 | ||
2671 | if url_link: | |
2672 | if not _write_link_file('url', DOT_URL_LINK_TEMPLATE, '\r\n', embed_filename=False): | |
2673 | return | |
2674 | if webloc_link: | |
2675 | if not _write_link_file('webloc', DOT_WEBLOC_LINK_TEMPLATE, '\n', embed_filename=False): | |
2676 | return | |
2677 | if desktop_link: | |
2678 | if not _write_link_file('desktop', DOT_DESKTOP_LINK_TEMPLATE, '\n', embed_filename=True): | |
2679 | return | |
2680 | ||
56d868db | 2681 | try: |
2682 | info_dict, files_to_move = self.pre_process(info_dict, 'before_dl', files_to_move) | |
2683 | except PostProcessingError as err: | |
2684 | self.report_error('Preprocessing: %s' % str(err)) | |
2685 | return | |
2686 | ||
732044af | 2687 | must_record_download_archive = False |
56d868db | 2688 | if self.params.get('skip_download', False): |
2689 | info_dict['filepath'] = temp_filename | |
2690 | info_dict['__finaldir'] = os.path.dirname(os.path.abspath(encodeFilename(full_filename))) | |
2691 | info_dict['__files_to_move'] = files_to_move | |
2692 | info_dict = self.run_pp(MoveFilesAfterDownloadPP(self, False), info_dict) | |
2693 | else: | |
2694 | # Download | |
b868936c | 2695 | info_dict.setdefault('__postprocessors', []) |
4340deca | 2696 | try: |
0202b52a | 2697 | |
6b591b29 | 2698 | def existing_file(*filepaths): |
2699 | ext = info_dict.get('ext') | |
2700 | final_ext = self.params.get('final_ext', ext) | |
2701 | existing_files = [] | |
2702 | for file in orderedSet(filepaths): | |
2703 | if final_ext != ext: | |
2704 | converted = replace_extension(file, final_ext, ext) | |
2705 | if os.path.exists(encodeFilename(converted)): | |
2706 | existing_files.append(converted) | |
2707 | if os.path.exists(encodeFilename(file)): | |
2708 | existing_files.append(file) | |
2709 | ||
2710 | if not existing_files or self.params.get('overwrites', False): | |
2711 | for file in orderedSet(existing_files): | |
2712 | self.report_file_delete(file) | |
2713 | os.remove(encodeFilename(file)) | |
2714 | return None | |
2715 | ||
6b591b29 | 2716 | info_dict['ext'] = os.path.splitext(existing_files[0])[1][1:] |
2717 | return existing_files[0] | |
0202b52a | 2718 | |
2719 | success = True | |
4340deca | 2720 | if info_dict.get('requested_formats') is not None: |
81cd954a S |
2721 | |
2722 | def compatible_formats(formats): | |
d03cfdce | 2723 | # TODO: some formats actually allow this (mkv, webm, ogg, mp4), but not all of them. |
2724 | video_formats = [format for format in formats if format.get('vcodec') != 'none'] | |
2725 | audio_formats = [format for format in formats if format.get('acodec') != 'none'] | |
2726 | if len(video_formats) > 2 or len(audio_formats) > 2: | |
2727 | return False | |
2728 | ||
81cd954a | 2729 | # Check extension |
d03cfdce | 2730 | exts = set(format.get('ext') for format in formats) |
2731 | COMPATIBLE_EXTS = ( | |
2732 | set(('mp3', 'mp4', 'm4a', 'm4p', 'm4b', 'm4r', 'm4v', 'ismv', 'isma')), | |
2733 | set(('webm',)), | |
2734 | ) | |
2735 | for ext_sets in COMPATIBLE_EXTS: | |
2736 | if ext_sets.issuperset(exts): | |
2737 | return True | |
81cd954a S |
2738 | # TODO: Check acodec/vcodec |
2739 | return False | |
2740 | ||
2741 | requested_formats = info_dict['requested_formats'] | |
0202b52a | 2742 | old_ext = info_dict['ext'] |
3b297919 | 2743 | if self.params.get('merge_output_format') is None and not compatible_formats(requested_formats): |
2744 | info_dict['ext'] = 'mkv' | |
2745 | self.report_warning( | |
2746 | 'Requested formats are incompatible for merge and will be merged into mkv.') | |
124bc071 | 2747 | new_ext = info_dict['ext'] |
0202b52a | 2748 | |
124bc071 | 2749 | def correct_ext(filename, ext=new_ext): |
96fccc10 | 2750 | if filename == '-': |
2751 | return filename | |
0202b52a | 2752 | filename_real_ext = os.path.splitext(filename)[1][1:] |
2753 | filename_wo_ext = ( | |
2754 | os.path.splitext(filename)[0] | |
124bc071 | 2755 | if filename_real_ext in (old_ext, new_ext) |
0202b52a | 2756 | else filename) |
124bc071 | 2757 | return '%s.%s' % (filename_wo_ext, ext) |
0202b52a | 2758 | |
38c6902b | 2759 | # Ensure filename always has a correct extension for successful merge |
0202b52a | 2760 | full_filename = correct_ext(full_filename) |
2761 | temp_filename = correct_ext(temp_filename) | |
2762 | dl_filename = existing_file(full_filename, temp_filename) | |
1ea24129 | 2763 | info_dict['__real_download'] = False |
18e674b4 | 2764 | |
2765 | _protocols = set(determine_protocol(f) for f in requested_formats) | |
dbf5416a | 2766 | if len(_protocols) == 1: # All requested formats have same protocol |
18e674b4 | 2767 | info_dict['protocol'] = _protocols.pop() |
d5fe04f5 | 2768 | directly_mergable = FFmpegFD.can_merge_formats(info_dict, self.params) |
dbf5416a | 2769 | if dl_filename is not None: |
6c7274ec | 2770 | self.report_file_already_downloaded(dl_filename) |
96fccc10 | 2771 | elif (directly_mergable and get_suitable_downloader( |
a46a815b | 2772 | info_dict, self.params, to_stdout=(temp_filename == '-')) == FFmpegFD): |
dbf5416a | 2773 | info_dict['url'] = '\n'.join(f['url'] for f in requested_formats) |
2774 | success, real_download = self.dl(temp_filename, info_dict) | |
2775 | info_dict['__real_download'] = real_download | |
18e674b4 | 2776 | else: |
2777 | downloaded = [] | |
2778 | merger = FFmpegMergerPP(self) | |
2779 | if self.params.get('allow_unplayable_formats'): | |
2780 | self.report_warning( | |
2781 | 'You have requested merging of multiple formats ' | |
2782 | 'while also allowing unplayable formats to be downloaded. ' | |
2783 | 'The formats won\'t be merged to prevent data corruption.') | |
2784 | elif not merger.available: | |
2785 | self.report_warning( | |
2786 | 'You have requested merging of multiple formats but ffmpeg is not installed. ' | |
2787 | 'The formats won\'t be merged.') | |
2788 | ||
96fccc10 | 2789 | if temp_filename == '-': |
2790 | reason = ('using a downloader other than ffmpeg' if directly_mergable | |
2791 | else 'but the formats are incompatible for simultaneous download' if merger.available | |
2792 | else 'but ffmpeg is not installed') | |
2793 | self.report_warning( | |
2794 | f'You have requested downloading multiple formats to stdout {reason}. ' | |
2795 | 'The formats will be streamed one after the other') | |
2796 | fname = temp_filename | |
dbf5416a | 2797 | for f in requested_formats: |
2798 | new_info = dict(info_dict) | |
2799 | del new_info['requested_formats'] | |
2800 | new_info.update(f) | |
96fccc10 | 2801 | if temp_filename != '-': |
124bc071 | 2802 | fname = prepend_extension( |
2803 | correct_ext(temp_filename, new_info['ext']), | |
2804 | 'f%s' % f['format_id'], new_info['ext']) | |
96fccc10 | 2805 | if not self._ensure_dir_exists(fname): |
2806 | return | |
a21e0ab1 | 2807 | f['filepath'] = fname |
96fccc10 | 2808 | downloaded.append(fname) |
dbf5416a | 2809 | partial_success, real_download = self.dl(fname, new_info) |
2810 | info_dict['__real_download'] = info_dict['__real_download'] or real_download | |
2811 | success = success and partial_success | |
2812 | if merger.available and not self.params.get('allow_unplayable_formats'): | |
2813 | info_dict['__postprocessors'].append(merger) | |
2814 | info_dict['__files_to_merge'] = downloaded | |
2815 | # Even if there were no downloads, it is being merged only now | |
2816 | info_dict['__real_download'] = True | |
2817 | else: | |
2818 | for file in downloaded: | |
2819 | files_to_move[file] = None | |
4340deca P |
2820 | else: |
2821 | # Just a single file | |
0202b52a | 2822 | dl_filename = existing_file(full_filename, temp_filename) |
6c7274ec | 2823 | if dl_filename is None or dl_filename == temp_filename: |
2824 | # dl_filename == temp_filename could mean that the file was partially downloaded with --no-part. | |
2825 | # So we should try to resume the download | |
e8e73840 | 2826 | success, real_download = self.dl(temp_filename, info_dict) |
0202b52a | 2827 | info_dict['__real_download'] = real_download |
6c7274ec | 2828 | else: |
2829 | self.report_file_already_downloaded(dl_filename) | |
0202b52a | 2830 | |
0202b52a | 2831 | dl_filename = dl_filename or temp_filename |
c571435f | 2832 | info_dict['__finaldir'] = os.path.dirname(os.path.abspath(encodeFilename(full_filename))) |
0202b52a | 2833 | |
3158150c | 2834 | except network_exceptions as err: |
7960b056 | 2835 | self.report_error('unable to download video data: %s' % error_to_compat_str(err)) |
4340deca P |
2836 | return |
2837 | except (OSError, IOError) as err: | |
2838 | raise UnavailableVideoError(err) | |
2839 | except (ContentTooShortError, ) as err: | |
2840 | self.report_error('content too short (expected %s bytes and served %s)' % (err.expected, err.downloaded)) | |
2841 | return | |
8222d8de | 2842 | |
de6000d9 | 2843 | if success and full_filename != '-': |
f17f8651 | 2844 | |
fd7cfb64 | 2845 | def fixup(): |
2846 | do_fixup = True | |
2847 | fixup_policy = self.params.get('fixup') | |
2848 | vid = info_dict['id'] | |
2849 | ||
2850 | if fixup_policy in ('ignore', 'never'): | |
2851 | return | |
2852 | elif fixup_policy == 'warn': | |
2853 | do_fixup = False | |
f89b3e2d | 2854 | elif fixup_policy != 'force': |
2855 | assert fixup_policy in ('detect_or_warn', None) | |
2856 | if not info_dict.get('__real_download'): | |
2857 | do_fixup = False | |
fd7cfb64 | 2858 | |
2859 | def ffmpeg_fixup(cndn, msg, cls): | |
2860 | if not cndn: | |
2861 | return | |
2862 | if not do_fixup: | |
2863 | self.report_warning(f'{vid}: {msg}') | |
2864 | return | |
2865 | pp = cls(self) | |
2866 | if pp.available: | |
2867 | info_dict['__postprocessors'].append(pp) | |
2868 | else: | |
2869 | self.report_warning(f'{vid}: {msg}. Install ffmpeg to fix this automatically') | |
2870 | ||
2871 | stretched_ratio = info_dict.get('stretched_ratio') | |
2872 | ffmpeg_fixup( | |
2873 | stretched_ratio not in (1, None), | |
2874 | f'Non-uniform pixel ratio {stretched_ratio}', | |
2875 | FFmpegFixupStretchedPP) | |
2876 | ||
2877 | ffmpeg_fixup( | |
2878 | (info_dict.get('requested_formats') is None | |
2879 | and info_dict.get('container') == 'm4a_dash' | |
2880 | and info_dict.get('ext') == 'm4a'), | |
2881 | 'writing DASH m4a. Only some players support this container', | |
2882 | FFmpegFixupM4aPP) | |
2883 | ||
2884 | downloader = (get_suitable_downloader(info_dict, self.params).__name__ | |
2885 | if 'protocol' in info_dict else None) | |
2886 | ffmpeg_fixup(downloader == 'HlsFD', 'malformed AAC bitstream detected', FFmpegFixupM3u8PP) | |
e36d50c5 | 2887 | ffmpeg_fixup(downloader == 'WebSocketFragmentFD', 'malformed timestamps detected', FFmpegFixupTimestampPP) |
2888 | ffmpeg_fixup(downloader == 'WebSocketFragmentFD', 'malformed duration detected', FFmpegFixupDurationPP) | |
fd7cfb64 | 2889 | |
2890 | fixup() | |
8222d8de | 2891 | try: |
23c1a667 | 2892 | info_dict = self.post_process(dl_filename, info_dict, files_to_move) |
af819c21 | 2893 | except PostProcessingError as err: |
2894 | self.report_error('Postprocessing: %s' % str(err)) | |
8222d8de | 2895 | return |
ab8e5e51 AM |
2896 | try: |
2897 | for ph in self._post_hooks: | |
23c1a667 | 2898 | ph(info_dict['filepath']) |
ab8e5e51 AM |
2899 | except Exception as err: |
2900 | self.report_error('post hooks: %s' % str(err)) | |
2901 | return | |
2d30509f | 2902 | must_record_download_archive = True |
2903 | ||
2904 | if must_record_download_archive or self.params.get('force_write_download_archive', False): | |
2905 | self.record_download_archive(info_dict) | |
c3e6ffba | 2906 | max_downloads = self.params.get('max_downloads') |
2907 | if max_downloads is not None and self._num_downloads >= int(max_downloads): | |
2908 | raise MaxDownloadsReached() | |
8222d8de JMF |
2909 | |
2910 | def download(self, url_list): | |
2911 | """Download a given list of URLs.""" | |
de6000d9 | 2912 | outtmpl = self.outtmpl_dict['default'] |
3089bc74 S |
2913 | if (len(url_list) > 1 |
2914 | and outtmpl != '-' | |
2915 | and '%' not in outtmpl | |
2916 | and self.params.get('max_downloads') != 1): | |
acd69589 | 2917 | raise SameFileError(outtmpl) |
8222d8de JMF |
2918 | |
2919 | for url in url_list: | |
2920 | try: | |
5f6a1245 | 2921 | # It also downloads the videos |
61aa5ba3 S |
2922 | res = self.extract_info( |
2923 | url, force_generic_extractor=self.params.get('force_generic_extractor', False)) | |
8222d8de | 2924 | except UnavailableVideoError: |
6febd1c1 | 2925 | self.report_error('unable to download video') |
8222d8de | 2926 | except MaxDownloadsReached: |
8f18aca8 | 2927 | self.to_screen('[info] Maximum number of downloads reached') |
8b0d7497 | 2928 | raise |
2929 | except ExistingVideoReached: | |
8f18aca8 | 2930 | self.to_screen('[info] Encountered a video that is already in the archive, stopping due to --break-on-existing') |
8b0d7497 | 2931 | raise |
2932 | except RejectedVideoReached: | |
8f18aca8 | 2933 | self.to_screen('[info] Encountered a video that did not match filter, stopping due to --break-on-reject') |
8222d8de | 2934 | raise |
63e0be34 PH |
2935 | else: |
2936 | if self.params.get('dump_single_json', False): | |
277d6ff5 | 2937 | self.post_extract(res) |
6e84b215 | 2938 | self.to_stdout(json.dumps(self.sanitize_info(res))) |
8222d8de JMF |
2939 | |
2940 | return self._download_retcode | |
2941 | ||
1dcc4c0c | 2942 | def download_with_info_file(self, info_filename): |
31bd3925 JMF |
2943 | with contextlib.closing(fileinput.FileInput( |
2944 | [info_filename], mode='r', | |
2945 | openhook=fileinput.hook_encoded('utf-8'))) as f: | |
2946 | # FileInput doesn't have a read method, we can't call json.load | |
8012d892 | 2947 | info = self.sanitize_info(json.loads('\n'.join(f)), self.params.get('clean_infojson', True)) |
d4943898 JMF |
2948 | try: |
2949 | self.process_ie_result(info, download=True) | |
d3f62c19 | 2950 | except (DownloadError, EntryNotInPlaylist, ThrottledDownload): |
d4943898 JMF |
2951 | webpage_url = info.get('webpage_url') |
2952 | if webpage_url is not None: | |
6febd1c1 | 2953 | self.report_warning('The info failed to download, trying with "%s"' % webpage_url) |
d4943898 JMF |
2954 | return self.download([webpage_url]) |
2955 | else: | |
2956 | raise | |
2957 | return self._download_retcode | |
1dcc4c0c | 2958 | |
cb202fd2 | 2959 | @staticmethod |
8012d892 | 2960 | def sanitize_info(info_dict, remove_private_keys=False): |
2961 | ''' Sanitize the infodict for converting to json ''' | |
3ad56b42 | 2962 | if info_dict is None: |
2963 | return info_dict | |
6e84b215 | 2964 | info_dict.setdefault('epoch', int(time.time())) |
2965 | remove_keys = {'__original_infodict'} # Always remove this since this may contain a copy of the entire dict | |
ae8f99e6 | 2966 | keep_keys = ['_type'], # Always keep this to facilitate load-info-json |
8012d892 | 2967 | if remove_private_keys: |
6e84b215 | 2968 | remove_keys |= { |
2969 | 'requested_formats', 'requested_subtitles', 'requested_entries', | |
2970 | 'filepath', 'entries', 'original_url', 'playlist_autonumber', | |
2971 | } | |
ae8f99e6 | 2972 | empty_values = (None, {}, [], set(), tuple()) |
2973 | reject = lambda k, v: k not in keep_keys and ( | |
2974 | k.startswith('_') or k in remove_keys or v in empty_values) | |
2975 | else: | |
ae8f99e6 | 2976 | reject = lambda k, v: k in remove_keys |
5226731e | 2977 | filter_fn = lambda obj: ( |
b0249bca | 2978 | list(map(filter_fn, obj)) if isinstance(obj, (LazyList, list, tuple, set)) |
a515a78d | 2979 | else obj if not isinstance(obj, dict) |
ae8f99e6 | 2980 | else dict((k, filter_fn(v)) for k, v in obj.items() if not reject(k, v))) |
5226731e | 2981 | return filter_fn(info_dict) |
cb202fd2 | 2982 | |
8012d892 | 2983 | @staticmethod |
2984 | def filter_requested_info(info_dict, actually_filter=True): | |
2985 | ''' Alias of sanitize_info for backward compatibility ''' | |
2986 | return YoutubeDL.sanitize_info(info_dict, actually_filter) | |
2987 | ||
dcf64d43 | 2988 | def run_pp(self, pp, infodict): |
5bfa4862 | 2989 | files_to_delete = [] |
dcf64d43 | 2990 | if '__files_to_move' not in infodict: |
2991 | infodict['__files_to_move'] = {} | |
b1940459 | 2992 | try: |
2993 | files_to_delete, infodict = pp.run(infodict) | |
2994 | except PostProcessingError as e: | |
2995 | # Must be True and not 'only_download' | |
2996 | if self.params.get('ignoreerrors') is True: | |
2997 | self.report_error(e) | |
2998 | return infodict | |
2999 | raise | |
3000 | ||
5bfa4862 | 3001 | if not files_to_delete: |
dcf64d43 | 3002 | return infodict |
5bfa4862 | 3003 | if self.params.get('keepvideo', False): |
3004 | for f in files_to_delete: | |
dcf64d43 | 3005 | infodict['__files_to_move'].setdefault(f, '') |
5bfa4862 | 3006 | else: |
3007 | for old_filename in set(files_to_delete): | |
3008 | self.to_screen('Deleting original file %s (pass -k to keep)' % old_filename) | |
3009 | try: | |
3010 | os.remove(encodeFilename(old_filename)) | |
3011 | except (IOError, OSError): | |
3012 | self.report_warning('Unable to remove downloaded original file') | |
dcf64d43 | 3013 | if old_filename in infodict['__files_to_move']: |
3014 | del infodict['__files_to_move'][old_filename] | |
3015 | return infodict | |
5bfa4862 | 3016 | |
277d6ff5 | 3017 | @staticmethod |
3018 | def post_extract(info_dict): | |
3019 | def actual_post_extract(info_dict): | |
3020 | if info_dict.get('_type') in ('playlist', 'multi_video'): | |
3021 | for video_dict in info_dict.get('entries', {}): | |
b050d210 | 3022 | actual_post_extract(video_dict or {}) |
277d6ff5 | 3023 | return |
3024 | ||
07cce701 | 3025 | post_extractor = info_dict.get('__post_extractor') or (lambda: {}) |
4ec82a72 | 3026 | extra = post_extractor().items() |
3027 | info_dict.update(extra) | |
07cce701 | 3028 | info_dict.pop('__post_extractor', None) |
277d6ff5 | 3029 | |
4ec82a72 | 3030 | original_infodict = info_dict.get('__original_infodict') or {} |
3031 | original_infodict.update(extra) | |
3032 | original_infodict.pop('__post_extractor', None) | |
3033 | ||
b050d210 | 3034 | actual_post_extract(info_dict or {}) |
277d6ff5 | 3035 | |
56d868db | 3036 | def pre_process(self, ie_info, key='pre_process', files_to_move=None): |
5bfa4862 | 3037 | info = dict(ie_info) |
56d868db | 3038 | info['__files_to_move'] = files_to_move or {} |
3039 | for pp in self._pps[key]: | |
dcf64d43 | 3040 | info = self.run_pp(pp, info) |
56d868db | 3041 | return info, info.pop('__files_to_move', None) |
5bfa4862 | 3042 | |
dcf64d43 | 3043 | def post_process(self, filename, ie_info, files_to_move=None): |
8222d8de JMF |
3044 | """Run all the postprocessors on the given file.""" |
3045 | info = dict(ie_info) | |
3046 | info['filepath'] = filename | |
dcf64d43 | 3047 | info['__files_to_move'] = files_to_move or {} |
0202b52a | 3048 | |
56d868db | 3049 | for pp in ie_info.get('__postprocessors', []) + self._pps['post_process']: |
dcf64d43 | 3050 | info = self.run_pp(pp, info) |
3051 | info = self.run_pp(MoveFilesAfterDownloadPP(self), info) | |
3052 | del info['__files_to_move'] | |
56d868db | 3053 | for pp in self._pps['after_move']: |
dcf64d43 | 3054 | info = self.run_pp(pp, info) |
23c1a667 | 3055 | return info |
c1c9a79c | 3056 | |
5db07df6 | 3057 | def _make_archive_id(self, info_dict): |
e9fef7ee S |
3058 | video_id = info_dict.get('id') |
3059 | if not video_id: | |
3060 | return | |
5db07df6 PH |
3061 | # Future-proof against any change in case |
3062 | # and backwards compatibility with prior versions | |
e9fef7ee | 3063 | extractor = info_dict.get('extractor_key') or info_dict.get('ie_key') # key in a playlist |
7012b23c | 3064 | if extractor is None: |
1211bb6d S |
3065 | url = str_or_none(info_dict.get('url')) |
3066 | if not url: | |
3067 | return | |
e9fef7ee | 3068 | # Try to find matching extractor for the URL and take its ie_key |
8b7491c8 | 3069 | for ie_key, ie in self._ies.items(): |
1211bb6d | 3070 | if ie.suitable(url): |
8b7491c8 | 3071 | extractor = ie_key |
e9fef7ee S |
3072 | break |
3073 | else: | |
3074 | return | |
d0757229 | 3075 | return '%s %s' % (extractor.lower(), video_id) |
5db07df6 PH |
3076 | |
3077 | def in_download_archive(self, info_dict): | |
3078 | fn = self.params.get('download_archive') | |
3079 | if fn is None: | |
3080 | return False | |
3081 | ||
3082 | vid_id = self._make_archive_id(info_dict) | |
e9fef7ee | 3083 | if not vid_id: |
7012b23c | 3084 | return False # Incomplete video information |
5db07df6 | 3085 | |
a45e8619 | 3086 | return vid_id in self.archive |
c1c9a79c PH |
3087 | |
3088 | def record_download_archive(self, info_dict): | |
3089 | fn = self.params.get('download_archive') | |
3090 | if fn is None: | |
3091 | return | |
5db07df6 PH |
3092 | vid_id = self._make_archive_id(info_dict) |
3093 | assert vid_id | |
c1c9a79c | 3094 | with locked_file(fn, 'a', encoding='utf-8') as archive_file: |
6febd1c1 | 3095 | archive_file.write(vid_id + '\n') |
a45e8619 | 3096 | self.archive.add(vid_id) |
dd82ffea | 3097 | |
8c51aa65 | 3098 | @staticmethod |
8abeeb94 | 3099 | def format_resolution(format, default='unknown'): |
fb04e403 | 3100 | if format.get('vcodec') == 'none': |
8326b00a | 3101 | if format.get('acodec') == 'none': |
3102 | return 'images' | |
fb04e403 | 3103 | return 'audio only' |
f49d89ee PH |
3104 | if format.get('resolution') is not None: |
3105 | return format['resolution'] | |
35615307 DA |
3106 | if format.get('width') and format.get('height'): |
3107 | res = '%dx%d' % (format['width'], format['height']) | |
3108 | elif format.get('height'): | |
3109 | res = '%sp' % format['height'] | |
3110 | elif format.get('width'): | |
388ae76b | 3111 | res = '%dx?' % format['width'] |
8c51aa65 | 3112 | else: |
8abeeb94 | 3113 | res = default |
8c51aa65 JMF |
3114 | return res |
3115 | ||
c57f7757 PH |
3116 | def _format_note(self, fdict): |
3117 | res = '' | |
3118 | if fdict.get('ext') in ['f4f', 'f4m']: | |
3119 | res += '(unsupported) ' | |
32f90364 PH |
3120 | if fdict.get('language'): |
3121 | if res: | |
3122 | res += ' ' | |
9016d76f | 3123 | res += '[%s] ' % fdict['language'] |
c57f7757 PH |
3124 | if fdict.get('format_note') is not None: |
3125 | res += fdict['format_note'] + ' ' | |
3126 | if fdict.get('tbr') is not None: | |
3127 | res += '%4dk ' % fdict['tbr'] | |
3128 | if fdict.get('container') is not None: | |
3129 | if res: | |
3130 | res += ', ' | |
3131 | res += '%s container' % fdict['container'] | |
3089bc74 S |
3132 | if (fdict.get('vcodec') is not None |
3133 | and fdict.get('vcodec') != 'none'): | |
c57f7757 PH |
3134 | if res: |
3135 | res += ', ' | |
3136 | res += fdict['vcodec'] | |
91c7271a | 3137 | if fdict.get('vbr') is not None: |
c57f7757 PH |
3138 | res += '@' |
3139 | elif fdict.get('vbr') is not None and fdict.get('abr') is not None: | |
3140 | res += 'video@' | |
3141 | if fdict.get('vbr') is not None: | |
3142 | res += '%4dk' % fdict['vbr'] | |
fbb21cf5 | 3143 | if fdict.get('fps') is not None: |
5d583bdf S |
3144 | if res: |
3145 | res += ', ' | |
3146 | res += '%sfps' % fdict['fps'] | |
c57f7757 PH |
3147 | if fdict.get('acodec') is not None: |
3148 | if res: | |
3149 | res += ', ' | |
3150 | if fdict['acodec'] == 'none': | |
3151 | res += 'video only' | |
3152 | else: | |
3153 | res += '%-5s' % fdict['acodec'] | |
3154 | elif fdict.get('abr') is not None: | |
3155 | if res: | |
3156 | res += ', ' | |
3157 | res += 'audio' | |
3158 | if fdict.get('abr') is not None: | |
3159 | res += '@%3dk' % fdict['abr'] | |
3160 | if fdict.get('asr') is not None: | |
3161 | res += ' (%5dHz)' % fdict['asr'] | |
3162 | if fdict.get('filesize') is not None: | |
3163 | if res: | |
3164 | res += ', ' | |
3165 | res += format_bytes(fdict['filesize']) | |
9732d77e PH |
3166 | elif fdict.get('filesize_approx') is not None: |
3167 | if res: | |
3168 | res += ', ' | |
3169 | res += '~' + format_bytes(fdict['filesize_approx']) | |
c57f7757 | 3170 | return res |
91c7271a | 3171 | |
c57f7757 | 3172 | def list_formats(self, info_dict): |
94badb25 | 3173 | formats = info_dict.get('formats', [info_dict]) |
53ed7066 | 3174 | new_format = ( |
3175 | 'list-formats' not in self.params.get('compat_opts', []) | |
169dbde9 | 3176 | and self.params.get('listformats_table', True) is not False) |
76d321f6 | 3177 | if new_format: |
3178 | table = [ | |
3179 | [ | |
3180 | format_field(f, 'format_id'), | |
3181 | format_field(f, 'ext'), | |
3182 | self.format_resolution(f), | |
3183 | format_field(f, 'fps', '%d'), | |
3184 | '|', | |
3185 | format_field(f, 'filesize', ' %s', func=format_bytes) + format_field(f, 'filesize_approx', '~%s', func=format_bytes), | |
3186 | format_field(f, 'tbr', '%4dk'), | |
52a8a1e1 | 3187 | shorten_protocol_name(f.get('protocol', '').replace("native", "n")), |
76d321f6 | 3188 | '|', |
3189 | format_field(f, 'vcodec', default='unknown').replace('none', ''), | |
3190 | format_field(f, 'vbr', '%4dk'), | |
3191 | format_field(f, 'acodec', default='unknown').replace('none', ''), | |
3192 | format_field(f, 'abr', '%3dk'), | |
3193 | format_field(f, 'asr', '%5dHz'), | |
3f698246 | 3194 | ', '.join(filter(None, ( |
3195 | 'UNSUPPORTED' if f.get('ext') in ('f4f', 'f4m') else '', | |
3196 | format_field(f, 'language', '[%s]'), | |
3197 | format_field(f, 'format_note'), | |
3198 | format_field(f, 'container', ignore=(None, f.get('ext'))), | |
ea05b302 | 3199 | ))), |
3f698246 | 3200 | ] for f in formats if f.get('preference') is None or f['preference'] >= -1000] |
76d321f6 | 3201 | header_line = ['ID', 'EXT', 'RESOLUTION', 'FPS', '|', ' FILESIZE', ' TBR', 'PROTO', |
3f698246 | 3202 | '|', 'VCODEC', ' VBR', 'ACODEC', ' ABR', ' ASR', 'MORE INFO'] |
76d321f6 | 3203 | else: |
3204 | table = [ | |
3205 | [ | |
3206 | format_field(f, 'format_id'), | |
3207 | format_field(f, 'ext'), | |
3208 | self.format_resolution(f), | |
3209 | self._format_note(f)] | |
3210 | for f in formats | |
3211 | if f.get('preference') is None or f['preference'] >= -1000] | |
3212 | header_line = ['format code', 'extension', 'resolution', 'note'] | |
57dd9a8f | 3213 | |
cfb56d1a | 3214 | self.to_screen( |
169dbde9 | 3215 | '[info] Available formats for %s:' % info_dict['id']) |
3216 | self.to_stdout(render_table( | |
bc97cdae | 3217 | header_line, table, delim=new_format, extraGap=(0 if new_format else 1), hideEmpty=new_format)) |
cfb56d1a PH |
3218 | |
3219 | def list_thumbnails(self, info_dict): | |
b0249bca | 3220 | thumbnails = list(info_dict.get('thumbnails')) |
cfb56d1a | 3221 | if not thumbnails: |
b7b72db9 | 3222 | self.to_screen('[info] No thumbnails present for %s' % info_dict['id']) |
3223 | return | |
cfb56d1a PH |
3224 | |
3225 | self.to_screen( | |
3226 | '[info] Thumbnails for %s:' % info_dict['id']) | |
169dbde9 | 3227 | self.to_stdout(render_table( |
cfb56d1a PH |
3228 | ['ID', 'width', 'height', 'URL'], |
3229 | [[t['id'], t.get('width', 'unknown'), t.get('height', 'unknown'), t['url']] for t in thumbnails])) | |
dca08720 | 3230 | |
360e1ca5 | 3231 | def list_subtitles(self, video_id, subtitles, name='subtitles'): |
a504ced0 | 3232 | if not subtitles: |
360e1ca5 | 3233 | self.to_screen('%s has no %s' % (video_id, name)) |
a504ced0 | 3234 | return |
a504ced0 | 3235 | self.to_screen( |
edab9dbf | 3236 | 'Available %s for %s:' % (name, video_id)) |
2412044c | 3237 | |
3238 | def _row(lang, formats): | |
49c258e1 | 3239 | exts, names = zip(*((f['ext'], f.get('name') or 'unknown') for f in reversed(formats))) |
2412044c | 3240 | if len(set(names)) == 1: |
7aee40c1 | 3241 | names = [] if names[0] == 'unknown' else names[:1] |
2412044c | 3242 | return [lang, ', '.join(names), ', '.join(exts)] |
3243 | ||
169dbde9 | 3244 | self.to_stdout(render_table( |
2412044c | 3245 | ['Language', 'Name', 'Formats'], |
3246 | [_row(lang, formats) for lang, formats in subtitles.items()], | |
3247 | hideEmpty=True)) | |
a504ced0 | 3248 | |
dca08720 PH |
3249 | def urlopen(self, req): |
3250 | """ Start an HTTP download """ | |
82d8a8b6 | 3251 | if isinstance(req, compat_basestring): |
67dda517 | 3252 | req = sanitized_Request(req) |
19a41fc6 | 3253 | return self._opener.open(req, timeout=self._socket_timeout) |
dca08720 PH |
3254 | |
3255 | def print_debug_header(self): | |
3256 | if not self.params.get('verbose'): | |
3257 | return | |
62fec3b2 | 3258 | |
c6afed48 PH |
3259 | stdout_encoding = getattr( |
3260 | sys.stdout, 'encoding', 'missing (%s)' % type(sys.stdout).__name__) | |
b0472057 | 3261 | encoding_str = ( |
734f90bb PH |
3262 | '[debug] Encodings: locale %s, fs %s, out %s, pref %s\n' % ( |
3263 | locale.getpreferredencoding(), | |
3264 | sys.getfilesystemencoding(), | |
c6afed48 | 3265 | stdout_encoding, |
b0472057 | 3266 | self.get_encoding())) |
4192b51c | 3267 | write_string(encoding_str, encoding=None) |
734f90bb | 3268 | |
e5813e53 | 3269 | source = ( |
3270 | '(exe)' if hasattr(sys, 'frozen') | |
3271 | else '(zip)' if isinstance(globals().get('__loader__'), zipimporter) | |
3272 | else '(source)' if os.path.basename(sys.argv[0]) == '__main__.py' | |
3273 | else '') | |
3274 | self._write_string('[debug] yt-dlp version %s %s\n' % (__version__, source)) | |
e0986e31 | 3275 | if _LAZY_LOADER: |
f74980cb | 3276 | self._write_string('[debug] Lazy loading extractors enabled\n') |
3277 | if _PLUGIN_CLASSES: | |
3278 | self._write_string( | |
3279 | '[debug] Plugin Extractors: %s\n' % [ie.ie_key() for ie in _PLUGIN_CLASSES]) | |
53ed7066 | 3280 | if self.params.get('compat_opts'): |
3281 | self._write_string( | |
3282 | '[debug] Compatibility options: %s\n' % ', '.join(self.params.get('compat_opts'))) | |
dca08720 PH |
3283 | try: |
3284 | sp = subprocess.Popen( | |
3285 | ['git', 'rev-parse', '--short', 'HEAD'], | |
3286 | stdout=subprocess.PIPE, stderr=subprocess.PIPE, | |
3287 | cwd=os.path.dirname(os.path.abspath(__file__))) | |
f5b1bca9 | 3288 | out, err = process_communicate_or_kill(sp) |
dca08720 PH |
3289 | out = out.decode().strip() |
3290 | if re.match('[0-9a-f]+', out): | |
f74980cb | 3291 | self._write_string('[debug] Git HEAD: %s\n' % out) |
70a1165b | 3292 | except Exception: |
dca08720 PH |
3293 | try: |
3294 | sys.exc_clear() | |
70a1165b | 3295 | except Exception: |
dca08720 | 3296 | pass |
b300cda4 S |
3297 | |
3298 | def python_implementation(): | |
3299 | impl_name = platform.python_implementation() | |
3300 | if impl_name == 'PyPy' and hasattr(sys, 'pypy_version_info'): | |
3301 | return impl_name + ' version %d.%d.%d' % sys.pypy_version_info[:3] | |
3302 | return impl_name | |
3303 | ||
e5813e53 | 3304 | self._write_string('[debug] Python version %s (%s %s) - %s\n' % ( |
3305 | platform.python_version(), | |
3306 | python_implementation(), | |
3307 | platform.architecture()[0], | |
b300cda4 | 3308 | platform_name())) |
d28b5171 | 3309 | |
73fac4e9 | 3310 | exe_versions = FFmpegPostProcessor.get_versions(self) |
4c83c967 | 3311 | exe_versions['rtmpdump'] = rtmpdump_version() |
feee8d32 | 3312 | exe_versions['phantomjs'] = PhantomJSwrapper._version() |
d28b5171 | 3313 | exe_str = ', '.join( |
2831b468 | 3314 | f'{exe} {v}' for exe, v in sorted(exe_versions.items()) if v |
3315 | ) or 'none' | |
d28b5171 | 3316 | self._write_string('[debug] exe versions: %s\n' % exe_str) |
dca08720 | 3317 | |
2831b468 | 3318 | from .downloader.websocket import has_websockets |
3319 | from .postprocessor.embedthumbnail import has_mutagen | |
3320 | from .cookies import SQLITE_AVAILABLE, KEYRING_AVAILABLE | |
3321 | ||
ad3dc496 | 3322 | lib_str = ', '.join(sorted(filter(None, ( |
edf65256 | 3323 | compat_pycrypto_AES and compat_pycrypto_AES.__name__.split('.')[0], |
2831b468 | 3324 | has_websockets and 'websockets', |
3325 | has_mutagen and 'mutagen', | |
3326 | SQLITE_AVAILABLE and 'sqlite', | |
3327 | KEYRING_AVAILABLE and 'keyring', | |
ad3dc496 | 3328 | )))) or 'none' |
2831b468 | 3329 | self._write_string('[debug] Optional libraries: %s\n' % lib_str) |
3330 | ||
dca08720 PH |
3331 | proxy_map = {} |
3332 | for handler in self._opener.handlers: | |
3333 | if hasattr(handler, 'proxies'): | |
3334 | proxy_map.update(handler.proxies) | |
734f90bb | 3335 | self._write_string('[debug] Proxy map: ' + compat_str(proxy_map) + '\n') |
dca08720 | 3336 | |
58b1f00d PH |
3337 | if self.params.get('call_home', False): |
3338 | ipaddr = self.urlopen('https://yt-dl.org/ip').read().decode('utf-8') | |
3339 | self._write_string('[debug] Public IP address: %s\n' % ipaddr) | |
f5546c0b | 3340 | return |
58b1f00d PH |
3341 | latest_version = self.urlopen( |
3342 | 'https://yt-dl.org/latest/version').read().decode('utf-8') | |
3343 | if version_tuple(latest_version) > version_tuple(__version__): | |
3344 | self.report_warning( | |
3345 | 'You are using an outdated version (newest version: %s)! ' | |
3346 | 'See https://yt-dl.org/update if you need help updating.' % | |
3347 | latest_version) | |
3348 | ||
e344693b | 3349 | def _setup_opener(self): |
6ad14cab | 3350 | timeout_val = self.params.get('socket_timeout') |
19a41fc6 | 3351 | self._socket_timeout = 600 if timeout_val is None else float(timeout_val) |
6ad14cab | 3352 | |
982ee69a | 3353 | opts_cookiesfrombrowser = self.params.get('cookiesfrombrowser') |
dca08720 PH |
3354 | opts_cookiefile = self.params.get('cookiefile') |
3355 | opts_proxy = self.params.get('proxy') | |
3356 | ||
982ee69a | 3357 | self.cookiejar = load_cookies(opts_cookiefile, opts_cookiesfrombrowser, self) |
dca08720 | 3358 | |
6a3f4c3f | 3359 | cookie_processor = YoutubeDLCookieProcessor(self.cookiejar) |
dca08720 PH |
3360 | if opts_proxy is not None: |
3361 | if opts_proxy == '': | |
3362 | proxies = {} | |
3363 | else: | |
3364 | proxies = {'http': opts_proxy, 'https': opts_proxy} | |
3365 | else: | |
3366 | proxies = compat_urllib_request.getproxies() | |
067aa17e | 3367 | # Set HTTPS proxy to HTTP one if given (https://github.com/ytdl-org/youtube-dl/issues/805) |
dca08720 PH |
3368 | if 'http' in proxies and 'https' not in proxies: |
3369 | proxies['https'] = proxies['http'] | |
91410c9b | 3370 | proxy_handler = PerRequestProxyHandler(proxies) |
a0ddb8a2 PH |
3371 | |
3372 | debuglevel = 1 if self.params.get('debug_printtraffic') else 0 | |
be4a824d PH |
3373 | https_handler = make_HTTPS_handler(self.params, debuglevel=debuglevel) |
3374 | ydlh = YoutubeDLHandler(self.params, debuglevel=debuglevel) | |
fca6dba8 | 3375 | redirect_handler = YoutubeDLRedirectHandler() |
8b172c2e | 3376 | data_handler = compat_urllib_request_DataHandler() |
6240b0a2 JMF |
3377 | |
3378 | # When passing our own FileHandler instance, build_opener won't add the | |
3379 | # default FileHandler and allows us to disable the file protocol, which | |
3380 | # can be used for malicious purposes (see | |
067aa17e | 3381 | # https://github.com/ytdl-org/youtube-dl/issues/8227) |
6240b0a2 JMF |
3382 | file_handler = compat_urllib_request.FileHandler() |
3383 | ||
3384 | def file_open(*args, **kwargs): | |
7a5c1cfe | 3385 | raise compat_urllib_error.URLError('file:// scheme is explicitly disabled in yt-dlp for security reasons') |
6240b0a2 JMF |
3386 | file_handler.file_open = file_open |
3387 | ||
3388 | opener = compat_urllib_request.build_opener( | |
fca6dba8 | 3389 | proxy_handler, https_handler, cookie_processor, ydlh, redirect_handler, data_handler, file_handler) |
2461f79d | 3390 | |
dca08720 PH |
3391 | # Delete the default user-agent header, which would otherwise apply in |
3392 | # cases where our custom HTTP handler doesn't come into play | |
067aa17e | 3393 | # (See https://github.com/ytdl-org/youtube-dl/issues/1309 for details) |
dca08720 PH |
3394 | opener.addheaders = [] |
3395 | self._opener = opener | |
62fec3b2 PH |
3396 | |
3397 | def encode(self, s): | |
3398 | if isinstance(s, bytes): | |
3399 | return s # Already encoded | |
3400 | ||
3401 | try: | |
3402 | return s.encode(self.get_encoding()) | |
3403 | except UnicodeEncodeError as err: | |
3404 | err.reason = err.reason + '. Check your system encoding configuration or use the --encoding option.' | |
3405 | raise | |
3406 | ||
3407 | def get_encoding(self): | |
3408 | encoding = self.params.get('encoding') | |
3409 | if encoding is None: | |
3410 | encoding = preferredencoding() | |
3411 | return encoding | |
ec82d85a | 3412 | |
de6000d9 | 3413 | def _write_thumbnails(self, info_dict, filename): # return the extensions |
6c4fd172 | 3414 | write_all = self.params.get('write_all_thumbnails', False) |
3415 | thumbnails = [] | |
3416 | if write_all or self.params.get('writethumbnail', False): | |
0202b52a | 3417 | thumbnails = info_dict.get('thumbnails') or [] |
6c4fd172 | 3418 | multiple = write_all and len(thumbnails) > 1 |
ec82d85a | 3419 | |
0202b52a | 3420 | ret = [] |
981052c9 | 3421 | for t in thumbnails[::-1]: |
ec82d85a | 3422 | thumb_ext = determine_ext(t['url'], 'jpg') |
6c4fd172 | 3423 | suffix = '%s.' % t['id'] if multiple else '' |
3424 | thumb_display_id = '%s ' % t['id'] if multiple else '' | |
885cc0b7 | 3425 | thumb_filename = replace_extension(filename, suffix + thumb_ext, info_dict.get('ext')) |
ec82d85a | 3426 | |
0c3d0f51 | 3427 | if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(thumb_filename)): |
de6000d9 | 3428 | ret.append(suffix + thumb_ext) |
8ba87148 | 3429 | t['filepath'] = thumb_filename |
ec82d85a PH |
3430 | self.to_screen('[%s] %s: Thumbnail %sis already present' % |
3431 | (info_dict['extractor'], info_dict['id'], thumb_display_id)) | |
3432 | else: | |
5ef7d9bd | 3433 | self.to_screen('[%s] %s: Downloading thumbnail %s ...' % |
ec82d85a PH |
3434 | (info_dict['extractor'], info_dict['id'], thumb_display_id)) |
3435 | try: | |
3436 | uf = self.urlopen(t['url']) | |
d3d89c32 | 3437 | with open(encodeFilename(thumb_filename), 'wb') as thumbf: |
ec82d85a | 3438 | shutil.copyfileobj(uf, thumbf) |
de6000d9 | 3439 | ret.append(suffix + thumb_ext) |
ec82d85a PH |
3440 | self.to_screen('[%s] %s: Writing thumbnail %sto: %s' % |
3441 | (info_dict['extractor'], info_dict['id'], thumb_display_id, thumb_filename)) | |
885cc0b7 | 3442 | t['filepath'] = thumb_filename |
3158150c | 3443 | except network_exceptions as err: |
ec82d85a | 3444 | self.report_warning('Unable to download thumbnail "%s": %s' % |
9b9c5355 | 3445 | (t['url'], error_to_compat_str(err))) |
6c4fd172 | 3446 | if ret and not write_all: |
3447 | break | |
0202b52a | 3448 | return ret |