]>
Commit | Line | Data |
---|---|---|
26e63931 | 1 | import collections |
31bd3925 | 2 | import contextlib |
31215122 | 3 | import copy |
9d2ecdbc | 4 | import datetime |
c1c9a79c | 5 | import errno |
31bd3925 | 6 | import fileinput |
31215122 | 7 | import http.cookiejar |
8222d8de | 8 | import io |
b82f815f | 9 | import itertools |
8694c600 | 10 | import json |
62fec3b2 | 11 | import locale |
083c9df9 | 12 | import operator |
8222d8de | 13 | import os |
f8271158 | 14 | import random |
8222d8de JMF |
15 | import re |
16 | import shutil | |
6f2287cb | 17 | import string |
dca08720 | 18 | import subprocess |
8222d8de | 19 | import sys |
21cd8fae | 20 | import tempfile |
8222d8de | 21 | import time |
67134eab | 22 | import tokenize |
8222d8de | 23 | import traceback |
524e2e4f | 24 | import unicodedata |
961ea474 | 25 | |
f8271158 | 26 | from .cache import Cache |
227bf1a3 | 27 | from .compat import functools, urllib # isort: split |
28 | from .compat import compat_os_name, compat_shlex_quote, urllib_req_to_req | |
31215122 | 29 | from .cookies import LenientSimpleCookie, load_cookies |
f8271158 | 30 | from .downloader import FFmpegFD, get_suitable_downloader, shorten_protocol_name |
31 | from .downloader.rtmp import rtmpdump_version | |
f8271158 | 32 | from .extractor import gen_extractor_classes, get_info_extractor |
fe7866d0 | 33 | from .extractor.common import UnsupportedURLIE |
f8271158 | 34 | from .extractor.openload import PhantomJSwrapper |
35 | from .minicurses import format_text | |
3d2623a8 | 36 | from .networking import HEADRequest, Request, RequestDirector |
db7b054a | 37 | from .networking.common import _REQUEST_HANDLERS, _RH_PREFERENCES |
227bf1a3 | 38 | from .networking.exceptions import ( |
39 | HTTPError, | |
40 | NoSupportingHandlers, | |
41 | RequestError, | |
42 | SSLError, | |
3d2623a8 | 43 | network_exceptions, |
227bf1a3 | 44 | ) |
8e40b9d1 | 45 | from .plugins import directories as plugin_directories |
e756f45b | 46 | from .postprocessor import _PLUGIN_CLASSES as plugin_pps |
f8271158 | 47 | from .postprocessor import ( |
48 | EmbedThumbnailPP, | |
49 | FFmpegFixupDuplicateMoovPP, | |
50 | FFmpegFixupDurationPP, | |
51 | FFmpegFixupM3u8PP, | |
52 | FFmpegFixupM4aPP, | |
53 | FFmpegFixupStretchedPP, | |
54 | FFmpegFixupTimestampPP, | |
55 | FFmpegMergerPP, | |
56 | FFmpegPostProcessor, | |
ca9def71 | 57 | FFmpegVideoConvertorPP, |
f8271158 | 58 | MoveFilesAfterDownloadPP, |
59 | get_postprocessor, | |
60 | ) | |
ca9def71 | 61 | from .postprocessor.ffmpeg import resolve_mapping as resolve_recode_mapping |
f9fb3ce8 SS |
62 | from .update import ( |
63 | REPOSITORY, | |
64 | _get_system_deprecation, | |
65 | _make_label, | |
66 | current_git_head, | |
67 | detect_variant, | |
68 | ) | |
8c25f81b | 69 | from .utils import ( |
f8271158 | 70 | DEFAULT_OUTTMPL, |
7b2c3f47 | 71 | IDENTITY, |
f8271158 | 72 | LINK_TEMPLATES, |
8dc59305 | 73 | MEDIA_EXTENSIONS, |
f8271158 | 74 | NO_DEFAULT, |
1d485a1a | 75 | NUMBER_RE, |
f8271158 | 76 | OUTTMPL_TYPES, |
77 | POSTPROCESS_WHEN, | |
78 | STR_FORMAT_RE_TMPL, | |
79 | STR_FORMAT_TYPES, | |
80 | ContentTooShortError, | |
81 | DateRange, | |
82 | DownloadCancelled, | |
83 | DownloadError, | |
84 | EntryNotInPlaylist, | |
85 | ExistingVideoReached, | |
86 | ExtractorError, | |
784320c9 | 87 | FormatSorter, |
f8271158 | 88 | GeoRestrictedError, |
f8271158 | 89 | ISO3166Utils, |
90 | LazyList, | |
91 | MaxDownloadsReached, | |
19a03940 | 92 | Namespace, |
f8271158 | 93 | PagedList, |
7e88d7d7 | 94 | PlaylistEntries, |
f8271158 | 95 | Popen, |
96 | PostProcessingError, | |
97 | ReExtractInfo, | |
98 | RejectedVideoReached, | |
99 | SameFileError, | |
100 | UnavailableVideoError, | |
693f0600 | 101 | UserNotLive, |
eedb7ba5 S |
102 | age_restricted, |
103 | args_to_str, | |
cb794ee0 | 104 | bug_reports_message, |
ce02ed60 | 105 | date_from_str, |
da4db748 | 106 | deprecation_warning, |
ce02ed60 | 107 | determine_ext, |
b5559424 | 108 | determine_protocol, |
c0384f22 | 109 | encode_compat_str, |
ce02ed60 | 110 | encodeFilename, |
a06916d9 | 111 | error_to_compat_str, |
47cdc68e | 112 | escapeHTML, |
590bc6f6 | 113 | expand_path, |
227bf1a3 | 114 | extract_basic_auth, |
90137ca4 | 115 | filter_dict, |
e29663c6 | 116 | float_or_none, |
02dbf93f | 117 | format_bytes, |
e0fd9573 | 118 | format_decimal_suffix, |
f8271158 | 119 | format_field, |
525ef922 | 120 | formatSeconds, |
fc61aff4 | 121 | get_compatible_ext, |
0bb322b9 | 122 | get_domain, |
c9969434 | 123 | int_or_none, |
732044af | 124 | iri_to_uri, |
941e881e | 125 | is_path_like, |
34921b43 | 126 | join_nonempty, |
ce02ed60 | 127 | locked_file, |
0647d925 | 128 | make_archive_id, |
0202b52a | 129 | make_dir, |
ec11a9f4 | 130 | number_of_digits, |
cd6fc19e | 131 | orderedSet, |
5314b521 | 132 | orderedSet_from_options, |
083c9df9 | 133 | parse_filesize, |
ce02ed60 | 134 | preferredencoding, |
eedb7ba5 | 135 | prepend_extension, |
3efb96a6 | 136 | remove_terminal_sequences, |
cfb56d1a | 137 | render_table, |
eedb7ba5 | 138 | replace_extension, |
ce02ed60 | 139 | sanitize_filename, |
1bb5c511 | 140 | sanitize_path, |
dcf77cf1 | 141 | sanitize_url, |
1211bb6d | 142 | str_or_none, |
e29663c6 | 143 | strftime_or_none, |
ce02ed60 | 144 | subtitles_filename, |
819e0531 | 145 | supports_terminal_sequences, |
b1f94422 | 146 | system_identifier, |
f2ebc5c7 | 147 | timetuple_from_msec, |
732044af | 148 | to_high_limit_path, |
324ad820 | 149 | traverse_obj, |
fc61aff4 | 150 | try_call, |
6033d980 | 151 | try_get, |
29eb5174 | 152 | url_basename, |
7d1eb38a | 153 | variadic, |
58b1f00d | 154 | version_tuple, |
53973b4d | 155 | windows_enable_vt_mode, |
ce02ed60 PH |
156 | write_json_file, |
157 | write_string, | |
4f026faf | 158 | ) |
227bf1a3 | 159 | from .utils._utils import _YDLLogger |
160 | from .utils.networking import ( | |
161 | HTTPHeaderDict, | |
162 | clean_headers, | |
163 | clean_proxies, | |
3d2623a8 | 164 | std_headers, |
227bf1a3 | 165 | ) |
20314dd4 | 166 | from .version import CHANNEL, ORIGIN, RELEASE_GIT_HEAD, VARIANT, __version__ |
8222d8de | 167 | |
e9c0cdd3 YCH |
168 | if compat_os_name == 'nt': |
169 | import ctypes | |
170 | ||
2459b6e1 | 171 | |
86e5f3ed | 172 | class YoutubeDL: |
8222d8de JMF |
173 | """YoutubeDL class. |
174 | ||
175 | YoutubeDL objects are the ones responsible of downloading the | |
176 | actual video file and writing it to disk if the user has requested | |
177 | it, among some other tasks. In most cases there should be one per | |
178 | program. As, given a video URL, the downloader doesn't know how to | |
179 | extract all the needed information, task that InfoExtractors do, it | |
180 | has to pass the URL to one of them. | |
181 | ||
182 | For this, YoutubeDL objects have a method that allows | |
183 | InfoExtractors to be registered in a given order. When it is passed | |
184 | a URL, the YoutubeDL object handles it to the first InfoExtractor it | |
185 | finds that reports being able to handle it. The InfoExtractor extracts | |
186 | all the information about the video or videos the URL refers to, and | |
187 | YoutubeDL process the extracted information, possibly using a File | |
188 | Downloader to download the video. | |
189 | ||
190 | YoutubeDL objects accept a lot of parameters. In order not to saturate | |
191 | the object constructor with arguments, it receives a dictionary of | |
192 | options instead. These options are available through the params | |
193 | attribute for the InfoExtractors to use. The YoutubeDL also | |
194 | registers itself as the downloader in charge for the InfoExtractors | |
195 | that are added to it, so this is a "mutual registration". | |
196 | ||
197 | Available options: | |
198 | ||
199 | username: Username for authentication purposes. | |
200 | password: Password for authentication purposes. | |
180940e0 | 201 | videopassword: Password for accessing a video. |
1da50aa3 S |
202 | ap_mso: Adobe Pass multiple-system operator identifier. |
203 | ap_username: Multiple-system operator account username. | |
204 | ap_password: Multiple-system operator account password. | |
8222d8de | 205 | usenetrc: Use netrc for authentication instead. |
c8bc203f | 206 | netrc_location: Location of the netrc file. Defaults to ~/.netrc. |
db3ad8a6 | 207 | netrc_cmd: Use a shell command to get credentials |
8222d8de JMF |
208 | verbose: Print additional info to stdout. |
209 | quiet: Do not print messages to stdout. | |
ad8915b7 | 210 | no_warnings: Do not print out anything for warnings. |
bb66c247 | 211 | forceprint: A dict with keys WHEN mapped to a list of templates to |
212 | print to stdout. The allowed keys are video or any of the | |
213 | items in utils.POSTPROCESS_WHEN. | |
ca30f449 | 214 | For compatibility, a single list is also accepted |
bb66c247 | 215 | print_to_file: A dict with keys WHEN (same as forceprint) mapped to |
216 | a list of tuples with (template, filename) | |
8694c600 | 217 | forcejson: Force printing info_dict as JSON. |
63e0be34 PH |
218 | dump_single_json: Force printing the info_dict of the whole playlist |
219 | (or video) as a single JSON line. | |
c25228e5 | 220 | force_write_download_archive: Force writing download archive regardless |
221 | of 'skip_download' or 'simulate'. | |
b7b04c78 | 222 | simulate: Do not download the video files. If unset (or None), |
223 | simulate only if listsubtitles, listformats or list_thumbnails is used | |
eb8a4433 | 224 | format: Video format code. see "FORMAT SELECTION" for more details. |
093a1710 | 225 | You can also pass a function. The function takes 'ctx' as |
226 | argument and returns the formats to download. | |
227 | See "build_format_selector" for an implementation | |
63ad4d43 | 228 | allow_unplayable_formats: Allow unplayable formats to be extracted and downloaded. |
b7da73eb | 229 | ignore_no_formats_error: Ignore "No video formats" error. Usefull for |
230 | extracting metadata even if the video is not actually | |
231 | available for download (experimental) | |
0930b11f | 232 | format_sort: A list of fields by which to sort the video formats. |
233 | See "Sorting Formats" for more details. | |
c25228e5 | 234 | format_sort_force: Force the given format_sort. see "Sorting Formats" |
235 | for more details. | |
08d30158 | 236 | prefer_free_formats: Whether to prefer video formats with free containers |
237 | over non-free ones of same quality. | |
c25228e5 | 238 | allow_multiple_video_streams: Allow multiple video streams to be merged |
239 | into a single file | |
240 | allow_multiple_audio_streams: Allow multiple audio streams to be merged | |
241 | into a single file | |
0ba692ac | 242 | check_formats Whether to test if the formats are downloadable. |
9f1a1c36 | 243 | Can be True (check all), False (check none), |
244 | 'selected' (check selected formats), | |
0ba692ac | 245 | or None (check only if requested by extractor) |
4524baf0 | 246 | paths: Dictionary of output paths. The allowed keys are 'home' |
5ca095cb | 247 | 'temp' and the keys of OUTTMPL_TYPES (in utils/_utils.py) |
de6000d9 | 248 | outtmpl: Dictionary of templates for output names. Allowed keys |
5ca095cb | 249 | are 'default' and the keys of OUTTMPL_TYPES (in utils/_utils.py). |
34488702 | 250 | For compatibility with youtube-dl, a single string can also be used |
a820dc72 RA |
251 | outtmpl_na_placeholder: Placeholder for unavailable meta fields. |
252 | restrictfilenames: Do not allow "&" and spaces in file names | |
253 | trim_file_name: Limit length of filename (extension excluded) | |
4524baf0 | 254 | windowsfilenames: Force the filenames to be windows compatible |
b1940459 | 255 | ignoreerrors: Do not stop on download/postprocessing errors. |
256 | Can be 'only_download' to ignore only download errors. | |
257 | Default is 'only_download' for CLI, but False for API | |
26e2805c | 258 | skip_playlist_after_errors: Number of allowed failures until the rest of |
259 | the playlist is skipped | |
fe7866d0 | 260 | allowed_extractors: List of regexes to match against extractor names that are allowed |
0c3d0f51 | 261 | overwrites: Overwrite all video and metadata files if True, |
262 | overwrite only non-video files if None | |
263 | and don't overwrite any file if False | |
c14e88f0 | 264 | playlist_items: Specific indices of playlist to download. |
75822ca7 | 265 | playlistrandom: Download playlist items in random order. |
7e9a6125 | 266 | lazy_playlist: Process playlist entries as they are received. |
8222d8de JMF |
267 | matchtitle: Download only matching titles. |
268 | rejecttitle: Reject downloads for matching titles. | |
8bf9319e | 269 | logger: Log messages to a logging.Logger instance. |
17ffed18 | 270 | logtostderr: Print everything to stderr instead of stdout. |
271 | consoletitle: Display progress in console window's titlebar. | |
8222d8de JMF |
272 | writedescription: Write the video description to a .description file |
273 | writeinfojson: Write the video description to a .info.json file | |
ad54c913 | 274 | clean_infojson: Remove internal metadata from the infojson |
34488702 | 275 | getcomments: Extract video comments. This will not be written to disk |
06167fbb | 276 | unless writeinfojson is also given |
1fb07d10 | 277 | writeannotations: Write the video annotations to a .annotations.xml file |
8222d8de | 278 | writethumbnail: Write the thumbnail image to a file |
c25228e5 | 279 | allow_playlist_files: Whether to write playlists' description, infojson etc |
280 | also to disk when using the 'write*' options | |
ec82d85a | 281 | write_all_thumbnails: Write all thumbnail formats to files |
732044af | 282 | writelink: Write an internet shortcut file, depending on the |
283 | current platform (.url/.webloc/.desktop) | |
284 | writeurllink: Write a Windows internet shortcut file (.url) | |
285 | writewebloclink: Write a macOS internet shortcut file (.webloc) | |
286 | writedesktoplink: Write a Linux internet shortcut file (.desktop) | |
8222d8de | 287 | writesubtitles: Write the video subtitles to a file |
741dd8ea | 288 | writeautomaticsub: Write the automatically generated subtitles to a file |
8222d8de | 289 | listsubtitles: Lists all available subtitles for the video |
a504ced0 | 290 | subtitlesformat: The format code for subtitles |
c32b0aab | 291 | subtitleslangs: List of languages of the subtitles to download (can be regex). |
292 | The list may contain "all" to refer to all the available | |
293 | subtitles. The language can be prefixed with a "-" to | |
62b58c09 | 294 | exclude it from the requested languages, e.g. ['all', '-live_chat'] |
8222d8de | 295 | keepvideo: Keep the video file after post-processing |
46f1370e | 296 | daterange: A utils.DateRange object, download only if the upload_date is in the range. |
8222d8de | 297 | skip_download: Skip the actual download of the video file |
c35f9e72 | 298 | cachedir: Location of the cache files in the filesystem. |
a0e07d31 | 299 | False to disable filesystem cache. |
47192f92 | 300 | noplaylist: Download single video instead of a playlist if in doubt. |
8dbe9899 PH |
301 | age_limit: An integer representing the user's age in years. |
302 | Unsuitable videos for the given age are skipped. | |
5fe18bdb PH |
303 | min_views: An integer representing the minimum view count the video |
304 | must have in order to not be skipped. | |
305 | Videos without view count information are always | |
306 | downloaded. None for no limit. | |
307 | max_views: An integer representing the maximum view count. | |
308 | Videos that are more popular than that are not | |
309 | downloaded. | |
310 | Videos without view count information are always | |
311 | downloaded. None for no limit. | |
ae103564 | 312 | download_archive: A set, or the name of a file where all downloads are recorded. |
313 | Videos already present in the file are not downloaded again. | |
8a51f564 | 314 | break_on_existing: Stop the download process after attempting to download a |
315 | file that is in the archive. | |
b222c271 | 316 | break_per_url: Whether break_on_reject and break_on_existing |
317 | should act on each input URL as opposed to for the entire queue | |
d76fa1f3 | 318 | cookiefile: File name or text stream from where cookies should be read and dumped to |
f59f5ef8 | 319 | cookiesfrombrowser: A tuple containing the name of the browser, the profile |
9bd13fe5 | 320 | name/path from where cookies are loaded, the name of the keyring, |
321 | and the container name, e.g. ('chrome', ) or | |
322 | ('vivaldi', 'default', 'BASICTEXT') or ('firefox', 'default', None, 'Meta') | |
f81c62a6 | 323 | legacyserverconnect: Explicitly allow HTTPS connection to servers that do not |
324 | support RFC 5746 secure renegotiation | |
f59f5ef8 | 325 | nocheckcertificate: Do not verify SSL certificates |
bb58c9ed | 326 | client_certificate: Path to client certificate file in PEM format. May include the private key |
327 | client_certificate_key: Path to private key file for client certificate | |
328 | client_certificate_password: Password for client certificate private key, if encrypted. | |
329 | If not provided and the key is encrypted, yt-dlp will ask interactively | |
7e8c0af0 | 330 | prefer_insecure: Use HTTP instead of HTTPS to retrieve information. |
c6e07cf1 | 331 | (Only supported by some extractors) |
8300774c | 332 | enable_file_urls: Enable file:// URLs. This is disabled by default for security reasons. |
8b7539d2 | 333 | http_headers: A dictionary of custom headers to be used for all requests |
a1ee09e8 | 334 | proxy: URL of the proxy server to use |
38cce791 | 335 | geo_verification_proxy: URL of the proxy to use for IP address verification |
504f20dd | 336 | on geo-restricted sites. |
e344693b | 337 | socket_timeout: Time to wait for unresponsive hosts, in seconds |
0783b09b PH |
338 | bidi_workaround: Work around buggy terminals without bidirectional text |
339 | support, using fridibi | |
a0ddb8a2 | 340 | debug_printtraffic:Print out sent and received HTTP traffic |
04b4d394 PH |
341 | default_search: Prepend this string if an input url is not valid. |
342 | 'auto' for elaborate guessing | |
62fec3b2 | 343 | encoding: Use this encoding instead of the system-specified. |
134c913c | 344 | extract_flat: Whether to resolve and process url_results further |
46f1370e | 345 | * False: Always process. Default for API |
134c913c | 346 | * True: Never process |
347 | * 'in_playlist': Do not process inside playlist/multi_video | |
348 | * 'discard': Always process, but don't return the result | |
349 | from inside playlist/multi_video | |
350 | * 'discard_in_playlist': Same as "discard", but only for | |
46f1370e | 351 | playlists (not multi_video). Default for CLI |
f2ebc5c7 | 352 | wait_for_video: If given, wait for scheduled streams to become available. |
353 | The value should be a tuple containing the range | |
354 | (min_secs, max_secs) to wait between retries | |
4f026faf | 355 | postprocessors: A list of dictionaries, each with an entry |
71b640cc | 356 | * key: The name of the postprocessor. See |
7a5c1cfe | 357 | yt_dlp/postprocessor/__init__.py for a list. |
bb66c247 | 358 | * when: When to run the postprocessor. Allowed values are |
359 | the entries of utils.POSTPROCESS_WHEN | |
56d868db | 360 | Assumed to be 'post_process' if not given |
71b640cc PH |
361 | progress_hooks: A list of functions that get called on download |
362 | progress, with a dictionary with the entries | |
5cda4eda | 363 | * status: One of "downloading", "error", or "finished". |
ee69b99a | 364 | Check this first and ignore unknown values. |
3ba7740d | 365 | * info_dict: The extracted info_dict |
71b640cc | 366 | |
5cda4eda | 367 | If status is one of "downloading", or "finished", the |
ee69b99a PH |
368 | following properties may also be present: |
369 | * filename: The final filename (always present) | |
5cda4eda | 370 | * tmpfilename: The filename we're currently writing to |
71b640cc PH |
371 | * downloaded_bytes: Bytes on disk |
372 | * total_bytes: Size of the whole file, None if unknown | |
5cda4eda PH |
373 | * total_bytes_estimate: Guess of the eventual file size, |
374 | None if unavailable. | |
375 | * elapsed: The number of seconds since download started. | |
71b640cc PH |
376 | * eta: The estimated time in seconds, None if unknown |
377 | * speed: The download speed in bytes/second, None if | |
378 | unknown | |
5cda4eda PH |
379 | * fragment_index: The counter of the currently |
380 | downloaded video fragment. | |
381 | * fragment_count: The number of fragments (= individual | |
382 | files that will be merged) | |
71b640cc PH |
383 | |
384 | Progress hooks are guaranteed to be called at least once | |
385 | (with status "finished") if the download is successful. | |
819e0531 | 386 | postprocessor_hooks: A list of functions that get called on postprocessing |
387 | progress, with a dictionary with the entries | |
388 | * status: One of "started", "processing", or "finished". | |
389 | Check this first and ignore unknown values. | |
390 | * postprocessor: Name of the postprocessor | |
391 | * info_dict: The extracted info_dict | |
392 | ||
393 | Progress hooks are guaranteed to be called at least twice | |
394 | (with status "started" and "finished") if the processing is successful. | |
fc61aff4 | 395 | merge_output_format: "/" separated list of extensions to use when merging formats. |
6b591b29 | 396 | final_ext: Expected final extension; used to detect when the file was |
59a7a13e | 397 | already downloaded and converted |
6271f1ca PH |
398 | fixup: Automatically correct known faults of the file. |
399 | One of: | |
400 | - "never": do nothing | |
401 | - "warn": only emit a warning | |
402 | - "detect_or_warn": check whether we can do anything | |
62cd676c | 403 | about it, warn otherwise (default) |
504f20dd | 404 | source_address: Client-side IP address to bind to. |
1cf376f5 | 405 | sleep_interval_requests: Number of seconds to sleep between requests |
406 | during extraction | |
7aa589a5 S |
407 | sleep_interval: Number of seconds to sleep before each download when |
408 | used alone or a lower bound of a range for randomized | |
409 | sleep before each download (minimum possible number | |
410 | of seconds to sleep) when used along with | |
411 | max_sleep_interval. | |
412 | max_sleep_interval:Upper bound of a range for randomized sleep before each | |
413 | download (maximum possible number of seconds to sleep). | |
414 | Must only be used along with sleep_interval. | |
415 | Actual sleep time will be a random float from range | |
416 | [sleep_interval; max_sleep_interval]. | |
1cf376f5 | 417 | sleep_interval_subtitles: Number of seconds to sleep before each subtitle download |
cfb56d1a PH |
418 | listformats: Print an overview of available video formats and exit. |
419 | list_thumbnails: Print a table of all thumbnails and exit. | |
0a41f331 | 420 | match_filter: A function that gets called for every video with the signature |
421 | (info_dict, *, incomplete: bool) -> Optional[str] | |
422 | For backward compatibility with youtube-dl, the signature | |
423 | (info_dict) -> Optional[str] is also allowed. | |
424 | - If it returns a message, the video is ignored. | |
425 | - If it returns None, the video is downloaded. | |
426 | - If it returns utils.NO_DEFAULT, the user is interactively | |
427 | asked whether to download the video. | |
fe2ce85a | 428 | - Raise utils.DownloadCancelled(msg) to abort remaining |
429 | downloads when a video is rejected. | |
5ca095cb | 430 | match_filter_func in utils/_utils.py is one example for this. |
8417f26b SS |
431 | color: A Dictionary with output stream names as keys |
432 | and their respective color policy as values. | |
433 | Can also just be a single color policy, | |
434 | in which case it applies to all outputs. | |
435 | Valid stream names are 'stdout' and 'stderr'. | |
436 | Valid color policies are one of 'always', 'auto', 'no_color' or 'never'. | |
0a840f58 | 437 | geo_bypass: Bypass geographic restriction via faking X-Forwarded-For |
504f20dd | 438 | HTTP header |
0a840f58 | 439 | geo_bypass_country: |
773f291d S |
440 | Two-letter ISO 3166-2 country code that will be used for |
441 | explicit geographic restriction bypassing via faking | |
504f20dd | 442 | X-Forwarded-For HTTP header |
5f95927a S |
443 | geo_bypass_ip_block: |
444 | IP range in CIDR notation that will be used similarly to | |
504f20dd | 445 | geo_bypass_country |
52a8a1e1 | 446 | external_downloader: A dictionary of protocol keys and the executable of the |
447 | external downloader to use for it. The allowed protocols | |
448 | are default|http|ftp|m3u8|dash|rtsp|rtmp|mms. | |
449 | Set the value to 'native' to use the native downloader | |
53ed7066 | 450 | compat_opts: Compatibility options. See "Differences in default behavior". |
3acf6d38 | 451 | The following options do not work when used through the API: |
b5ae35ee | 452 | filename, abort-on-error, multistreams, no-live-chat, format-sort |
dac5df5a | 453 | no-clean-infojson, no-playlist-metafiles, no-keep-subs, no-attach-info-json. |
e4f02757 | 454 | Refer __init__.py for their implementation |
819e0531 | 455 | progress_template: Dictionary of templates for progress outputs. |
456 | Allowed keys are 'download', 'postprocess', | |
457 | 'download-title' (console title) and 'postprocess-title'. | |
458 | The template is mapped on a dictionary with keys 'progress' and 'info' | |
23326151 | 459 | retry_sleep_functions: Dictionary of functions that takes the number of attempts |
460 | as argument and returns the time to sleep in seconds. | |
461 | Allowed keys are 'http', 'fragment', 'file_access' | |
0f446365 SW |
462 | download_ranges: A callback function that gets called for every video with |
463 | the signature (info_dict, ydl) -> Iterable[Section]. | |
464 | Only the returned sections will be downloaded. | |
465 | Each Section is a dict with the following keys: | |
5ec1b6b7 | 466 | * start_time: Start time of the section in seconds |
467 | * end_time: End time of the section in seconds | |
468 | * title: Section title (Optional) | |
469 | * index: Section number (Optional) | |
0f446365 | 470 | force_keyframes_at_cuts: Re-encode the video when downloading ranges to get precise cuts |
a7dc6a89 | 471 | noprogress: Do not print the progress bar |
a831c2ea | 472 | live_from_start: Whether to download livestreams videos from the start |
fe7e0c98 | 473 | |
8222d8de | 474 | The following parameters are not used by YoutubeDL itself, they are used by |
7a5c1cfe | 475 | the downloader (see yt_dlp/downloader/common.py): |
51d9739f | 476 | nopart, updatetime, buffersize, ratelimit, throttledratelimit, min_filesize, |
205a0654 | 477 | max_filesize, test, noresizebuffer, retries, file_access_retries, fragment_retries, |
a7dc6a89 | 478 | continuedl, xattr_set_filesize, hls_use_mpegts, http_chunk_size, |
59a7a13e | 479 | external_downloader_args, concurrent_fragment_downloads. |
76b1bd67 JMF |
480 | |
481 | The following options are used by the post processors: | |
c0b7d117 S |
482 | ffmpeg_location: Location of the ffmpeg/avconv binary; either the path |
483 | to the binary or its containing directory. | |
43820c03 | 484 | postprocessor_args: A dictionary of postprocessor/executable keys (in lower case) |
34488702 | 485 | and a list of additional command-line arguments for the |
486 | postprocessor/executable. The dict can also have "PP+EXE" keys | |
487 | which are used when the given exe is used by the given PP. | |
488 | Use 'default' as the name for arguments to passed to all PP | |
489 | For compatibility with youtube-dl, a single list of args | |
490 | can also be used | |
e409895f | 491 | |
492 | The following options are used by the extractors: | |
46f1370e | 493 | extractor_retries: Number of times to retry for known errors (default: 3) |
62bff2c1 | 494 | dynamic_mpd: Whether to process dynamic DASH manifests (default: True) |
e409895f | 495 | hls_split_discontinuity: Split HLS playlists to different formats at |
62bff2c1 | 496 | discontinuities such as ad breaks (default: False) |
5d3a0e79 | 497 | extractor_args: A dictionary of arguments to be passed to the extractors. |
498 | See "EXTRACTOR ARGUMENTS" for details. | |
62b58c09 | 499 | E.g. {'youtube': {'skip': ['dash', 'hls']}} |
88f23a18 | 500 | mark_watched: Mark videos watched (even with --simulate). Only for YouTube |
1890fc63 | 501 | |
502 | The following options are deprecated and may be removed in the future: | |
503 | ||
fe2ce85a | 504 | break_on_reject: Stop the download process when encountering a video that |
505 | has been filtered out. | |
506 | - `raise DownloadCancelled(msg)` in match_filter instead | |
fe7866d0 | 507 | force_generic_extractor: Force downloader to use the generic extractor |
508 | - Use allowed_extractors = ['generic', 'default'] | |
7e9a6125 | 509 | playliststart: - Use playlist_items |
510 | Playlist item to start at. | |
511 | playlistend: - Use playlist_items | |
512 | Playlist item to end at. | |
513 | playlistreverse: - Use playlist_items | |
514 | Download playlist items in reverse order. | |
1890fc63 | 515 | forceurl: - Use forceprint |
516 | Force printing final URL. | |
517 | forcetitle: - Use forceprint | |
518 | Force printing title. | |
519 | forceid: - Use forceprint | |
520 | Force printing ID. | |
521 | forcethumbnail: - Use forceprint | |
522 | Force printing thumbnail URL. | |
523 | forcedescription: - Use forceprint | |
524 | Force printing description. | |
525 | forcefilename: - Use forceprint | |
526 | Force printing final filename. | |
527 | forceduration: - Use forceprint | |
528 | Force printing duration. | |
529 | allsubtitles: - Use subtitleslangs = ['all'] | |
530 | Downloads all the subtitles of the video | |
531 | (requires writesubtitles or writeautomaticsub) | |
532 | include_ads: - Doesn't work | |
533 | Download ads as well | |
534 | call_home: - Not implemented | |
535 | Boolean, true iff we are allowed to contact the | |
536 | yt-dlp servers for debugging. | |
537 | post_hooks: - Register a custom postprocessor | |
538 | A list of functions that get called as the final step | |
539 | for each video file, after all postprocessors have been | |
540 | called. The filename will be passed as the only argument. | |
541 | hls_prefer_native: - Use external_downloader = {'m3u8': 'native'} or {'m3u8': 'ffmpeg'}. | |
542 | Use the native HLS downloader instead of ffmpeg/avconv | |
543 | if True, otherwise use ffmpeg/avconv if False, otherwise | |
544 | use downloader suggested by extractor if None. | |
545 | prefer_ffmpeg: - avconv support is deprecated | |
546 | If False, use avconv instead of ffmpeg if both are available, | |
547 | otherwise prefer ffmpeg. | |
548 | youtube_include_dash_manifest: - Use extractor_args | |
5d3a0e79 | 549 | If True (default), DASH manifests and related |
62bff2c1 | 550 | data will be downloaded and processed by extractor. |
551 | You can reduce network I/O by disabling it if you don't | |
552 | care about DASH. (only for youtube) | |
1890fc63 | 553 | youtube_include_hls_manifest: - Use extractor_args |
5d3a0e79 | 554 | If True (default), HLS manifests and related |
62bff2c1 | 555 | data will be downloaded and processed by extractor. |
556 | You can reduce network I/O by disabling it if you don't | |
557 | care about HLS. (only for youtube) | |
8417f26b | 558 | no_color: Same as `color='no_color'` |
6148833f | 559 | no_overwrites: Same as `overwrites=False` |
8222d8de JMF |
560 | """ |
561 | ||
86e5f3ed | 562 | _NUMERIC_FIELDS = { |
b8ed0f15 | 563 | 'width', 'height', 'asr', 'audio_channels', 'fps', |
564 | 'tbr', 'abr', 'vbr', 'filesize', 'filesize_approx', | |
e6f21b3d | 565 | 'timestamp', 'release_timestamp', |
c9969434 S |
566 | 'duration', 'view_count', 'like_count', 'dislike_count', 'repost_count', |
567 | 'average_rating', 'comment_count', 'age_limit', | |
568 | 'start_time', 'end_time', | |
569 | 'chapter_number', 'season_number', 'episode_number', | |
570 | 'track_number', 'disc_number', 'release_year', | |
86e5f3ed | 571 | } |
c9969434 | 572 | |
6db9c4d5 | 573 | _format_fields = { |
574 | # NB: Keep in sync with the docstring of extractor/common.py | |
a44ca5a4 | 575 | 'url', 'manifest_url', 'manifest_stream_number', 'ext', 'format', 'format_id', 'format_note', |
105bfd90 | 576 | 'width', 'height', 'aspect_ratio', 'resolution', 'dynamic_range', 'tbr', 'abr', 'acodec', 'asr', 'audio_channels', |
d5d1df8a | 577 | 'vbr', 'fps', 'vcodec', 'container', 'filesize', 'filesize_approx', 'rows', 'columns', |
6db9c4d5 | 578 | 'player_url', 'protocol', 'fragment_base_url', 'fragments', 'is_from_start', |
62b5c94c | 579 | 'preference', 'language', 'language_preference', 'quality', 'source_preference', 'cookies', |
7e68567e | 580 | 'http_headers', 'stretched_ratio', 'no_resume', 'has_drm', 'extra_param_to_segment_url', 'hls_aes', 'downloader_options', |
6db9c4d5 | 581 | 'page_url', 'app', 'play_path', 'tc_url', 'flash_version', 'rtmp_live', 'rtmp_conn', 'rtmp_protocol', 'rtmp_real_time' |
582 | } | |
48ee10ee | 583 | _format_selection_exts = { |
8dc59305 | 584 | 'audio': set(MEDIA_EXTENSIONS.common_audio), |
585 | 'video': set(MEDIA_EXTENSIONS.common_video + ('3gp', )), | |
586 | 'storyboards': set(MEDIA_EXTENSIONS.storyboards), | |
48ee10ee | 587 | } |
588 | ||
3511266b | 589 | def __init__(self, params=None, auto_init=True): |
883d4b1e | 590 | """Create a FileDownloader object with the given options. |
591 | @param auto_init Whether to load the default extractors and print header (if verbose). | |
49a57e70 | 592 | Set to 'no_verbose_header' to not print the header |
883d4b1e | 593 | """ |
e9f9a10f JMF |
594 | if params is None: |
595 | params = {} | |
592b7485 | 596 | self.params = params |
8b7491c8 | 597 | self._ies = {} |
56c73665 | 598 | self._ies_instances = {} |
1e43a6f7 | 599 | self._pps = {k: [] for k in POSTPROCESS_WHEN} |
b35496d8 | 600 | self._printed_messages = set() |
1cf376f5 | 601 | self._first_webpage_request = True |
ab8e5e51 | 602 | self._post_hooks = [] |
933605d7 | 603 | self._progress_hooks = [] |
819e0531 | 604 | self._postprocessor_hooks = [] |
8222d8de JMF |
605 | self._download_retcode = 0 |
606 | self._num_downloads = 0 | |
9c906919 | 607 | self._num_videos = 0 |
592b7485 | 608 | self._playlist_level = 0 |
609 | self._playlist_urls = set() | |
a0e07d31 | 610 | self.cache = Cache(self) |
6148833f | 611 | self.__header_cookies = [] |
34308b30 | 612 | |
591bb9d3 | 613 | stdout = sys.stderr if self.params.get('logtostderr') else sys.stdout |
614 | self._out_files = Namespace( | |
615 | out=stdout, | |
616 | error=sys.stderr, | |
617 | screen=sys.stderr if self.params.get('quiet') else stdout, | |
618 | console=None if compat_os_name == 'nt' else next( | |
cf4f42cb | 619 | filter(supports_terminal_sequences, (sys.stderr, sys.stdout)), None) |
591bb9d3 | 620 | ) |
f0795149 | 621 | |
622 | try: | |
623 | windows_enable_vt_mode() | |
624 | except Exception as e: | |
625 | self.write_debug(f'Failed to enable VT mode: {e}') | |
626 | ||
8417f26b SS |
627 | if self.params.get('no_color'): |
628 | if self.params.get('color') is not None: | |
62b5c94c | 629 | self.params.setdefault('_warnings', []).append( |
630 | 'Overwriting params from "color" with "no_color"') | |
8417f26b SS |
631 | self.params['color'] = 'no_color' |
632 | ||
a0b19d31 | 633 | term_allow_color = os.getenv('TERM', '').lower() != 'dumb' |
634 | no_color = bool(os.getenv('NO_COLOR')) | |
8417f26b SS |
635 | |
636 | def process_color_policy(stream): | |
637 | stream_name = {sys.stdout: 'stdout', sys.stderr: 'stderr'}[stream] | |
638 | policy = traverse_obj(self.params, ('color', (stream_name, None), {str}), get_all=False) | |
639 | if policy in ('auto', None): | |
a0b19d31 | 640 | if term_allow_color and supports_terminal_sequences(stream): |
641 | return 'no_color' if no_color else True | |
642 | return False | |
6148833f | 643 | assert policy in ('always', 'never', 'no_color'), policy |
8417f26b SS |
644 | return {'always': True, 'never': False}.get(policy, policy) |
645 | ||
591bb9d3 | 646 | self._allow_colors = Namespace(**{ |
8417f26b SS |
647 | name: process_color_policy(stream) |
648 | for name, stream in self._out_files.items_ if name != 'console' | |
591bb9d3 | 649 | }) |
819e0531 | 650 | |
61bdf15f SS |
651 | system_deprecation = _get_system_deprecation() |
652 | if system_deprecation: | |
653 | self.deprecated_feature(system_deprecation.replace('\n', '\n ')) | |
a61f4b28 | 654 | |
88acdbc2 | 655 | if self.params.get('allow_unplayable_formats'): |
656 | self.report_warning( | |
ec11a9f4 | 657 | f'You have asked for {self._format_err("UNPLAYABLE", self.Styles.EMPHASIS)} formats to be listed/downloaded. ' |
819e0531 | 658 | 'This is a developer option intended for debugging. \n' |
659 | ' If you experience any issues while using this option, ' | |
ec11a9f4 | 660 | f'{self._format_err("DO NOT", self.Styles.ERROR)} open a bug report') |
88acdbc2 | 661 | |
497074f0 | 662 | if self.params.get('bidi_workaround', False): |
663 | try: | |
664 | import pty | |
665 | master, slave = pty.openpty() | |
666 | width = shutil.get_terminal_size().columns | |
667 | width_args = [] if width is None else ['-w', str(width)] | |
668 | sp_kwargs = {'stdin': subprocess.PIPE, 'stdout': slave, 'stderr': self._out_files.error} | |
669 | try: | |
670 | self._output_process = Popen(['bidiv'] + width_args, **sp_kwargs) | |
671 | except OSError: | |
672 | self._output_process = Popen(['fribidi', '-c', 'UTF-8'] + width_args, **sp_kwargs) | |
673 | self._output_channel = os.fdopen(master, 'rb') | |
674 | except OSError as ose: | |
675 | if ose.errno == errno.ENOENT: | |
676 | self.report_warning( | |
677 | 'Could not find fribidi executable, ignoring --bidi-workaround. ' | |
678 | 'Make sure that fribidi is an executable file in one of the directories in your $PATH.') | |
679 | else: | |
680 | raise | |
681 | ||
682 | self.params['compat_opts'] = set(self.params.get('compat_opts', ())) | |
227bf1a3 | 683 | self.params['http_headers'] = HTTPHeaderDict(std_headers, self.params.get('http_headers')) |
6c5211ce | 684 | self._load_cookies(self.params['http_headers'].get('Cookie')) # compat |
685 | self.params['http_headers'].pop('Cookie', None) | |
db7b054a | 686 | self._request_director = self.build_request_director(_REQUEST_HANDLERS.values(), _RH_PREFERENCES) |
6c5211ce | 687 | |
497074f0 | 688 | if auto_init and auto_init != 'no_verbose_header': |
689 | self.print_debug_header() | |
690 | ||
be5df5ee S |
691 | def check_deprecated(param, option, suggestion): |
692 | if self.params.get(param) is not None: | |
86e5f3ed | 693 | self.report_warning(f'{option} is deprecated. Use {suggestion} instead') |
be5df5ee S |
694 | return True |
695 | return False | |
696 | ||
697 | if check_deprecated('cn_verification_proxy', '--cn-verification-proxy', '--geo-verification-proxy'): | |
38cce791 YCH |
698 | if self.params.get('geo_verification_proxy') is None: |
699 | self.params['geo_verification_proxy'] = self.params['cn_verification_proxy'] | |
700 | ||
0d1bb027 | 701 | check_deprecated('autonumber', '--auto-number', '-o "%(autonumber)s-%(title)s.%(ext)s"') |
702 | check_deprecated('usetitle', '--title', '-o "%(title)s-%(id)s.%(ext)s"') | |
53ed7066 | 703 | check_deprecated('useid', '--id', '-o "%(id)s.%(ext)s"') |
0d1bb027 | 704 | |
49a57e70 | 705 | for msg in self.params.get('_warnings', []): |
0d1bb027 | 706 | self.report_warning(msg) |
ee8dd27a | 707 | for msg in self.params.get('_deprecation_warnings', []): |
da4db748 | 708 | self.deprecated_feature(msg) |
0d1bb027 | 709 | |
8a82af35 | 710 | if 'list-formats' in self.params['compat_opts']: |
ec11a9f4 | 711 | self.params['listformats_table'] = False |
712 | ||
b5ae35ee | 713 | if 'overwrites' not in self.params and self.params.get('nooverwrites') is not None: |
b868936c | 714 | # nooverwrites was unnecessarily changed to overwrites |
715 | # in 0c3d0f51778b153f65c21906031c2e091fcfb641 | |
716 | # This ensures compatibility with both keys | |
717 | self.params['overwrites'] = not self.params['nooverwrites'] | |
b5ae35ee | 718 | elif self.params.get('overwrites') is None: |
719 | self.params.pop('overwrites', None) | |
b868936c | 720 | else: |
721 | self.params['nooverwrites'] = not self.params['overwrites'] | |
b9d973be | 722 | |
e4221b70 | 723 | if self.params.get('simulate') is None and any(( |
724 | self.params.get('list_thumbnails'), | |
725 | self.params.get('listformats'), | |
726 | self.params.get('listsubtitles'), | |
727 | )): | |
728 | self.params['simulate'] = 'list_only' | |
729 | ||
455a15e2 | 730 | self.params.setdefault('forceprint', {}) |
731 | self.params.setdefault('print_to_file', {}) | |
bb66c247 | 732 | |
733 | # Compatibility with older syntax | |
ca30f449 | 734 | if not isinstance(params['forceprint'], dict): |
455a15e2 | 735 | self.params['forceprint'] = {'video': params['forceprint']} |
ca30f449 | 736 | |
97ec5bc5 | 737 | if auto_init: |
97ec5bc5 | 738 | self.add_default_info_extractors() |
739 | ||
3089bc74 S |
740 | if (sys.platform != 'win32' |
741 | and sys.getfilesystemencoding() in ['ascii', 'ANSI_X3.4-1968'] | |
455a15e2 | 742 | and not self.params.get('restrictfilenames', False)): |
e9137224 | 743 | # Unicode filesystem API will throw errors (#1474, #13027) |
34308b30 | 744 | self.report_warning( |
6febd1c1 | 745 | 'Assuming --restrict-filenames since file system encoding ' |
1b725173 | 746 | 'cannot encode all characters. ' |
6febd1c1 | 747 | 'Set the LC_ALL environment variable to fix this.') |
4a98cdbf | 748 | self.params['restrictfilenames'] = True |
34308b30 | 749 | |
bf1824b3 | 750 | self._parse_outtmpl() |
486dd09e | 751 | |
187986a8 | 752 | # Creating format selector here allows us to catch syntax errors before the extraction |
753 | self.format_selector = ( | |
fa9f30b8 | 754 | self.params.get('format') if self.params.get('format') in (None, '-') |
093a1710 | 755 | else self.params['format'] if callable(self.params['format']) |
187986a8 | 756 | else self.build_format_selector(self.params['format'])) |
757 | ||
013b50b7 | 758 | hooks = { |
759 | 'post_hooks': self.add_post_hook, | |
760 | 'progress_hooks': self.add_progress_hook, | |
761 | 'postprocessor_hooks': self.add_postprocessor_hook, | |
762 | } | |
763 | for opt, fn in hooks.items(): | |
764 | for ph in self.params.get(opt, []): | |
765 | fn(ph) | |
71b640cc | 766 | |
5bfc8bee | 767 | for pp_def_raw in self.params.get('postprocessors', []): |
768 | pp_def = dict(pp_def_raw) | |
769 | when = pp_def.pop('when', 'post_process') | |
770 | self.add_post_processor( | |
f9934b96 | 771 | get_postprocessor(pp_def.pop('key'))(self, **pp_def), |
5bfc8bee | 772 | when=when) |
773 | ||
ed39cac5 | 774 | def preload_download_archive(fn): |
775 | """Preload the archive, if any is specified""" | |
ae103564 | 776 | archive = set() |
ed39cac5 | 777 | if fn is None: |
ae103564 | 778 | return archive |
941e881e | 779 | elif not is_path_like(fn): |
ae103564 | 780 | return fn |
781 | ||
49a57e70 | 782 | self.write_debug(f'Loading archive file {fn!r}') |
ed39cac5 | 783 | try: |
784 | with locked_file(fn, 'r', encoding='utf-8') as archive_file: | |
785 | for line in archive_file: | |
ae103564 | 786 | archive.add(line.strip()) |
86e5f3ed | 787 | except OSError as ioe: |
ed39cac5 | 788 | if ioe.errno != errno.ENOENT: |
789 | raise | |
ae103564 | 790 | return archive |
ed39cac5 | 791 | |
ae103564 | 792 | self.archive = preload_download_archive(self.params.get('download_archive')) |
ed39cac5 | 793 | |
7d4111ed PH |
794 | def warn_if_short_id(self, argv): |
795 | # short YouTube ID starting with dash? | |
796 | idxs = [ | |
797 | i for i, a in enumerate(argv) | |
798 | if re.match(r'^-[0-9A-Za-z_-]{10}$', a)] | |
799 | if idxs: | |
800 | correct_argv = ( | |
7a5c1cfe | 801 | ['yt-dlp'] |
3089bc74 S |
802 | + [a for i, a in enumerate(argv) if i not in idxs] |
803 | + ['--'] + [argv[i] for i in idxs] | |
7d4111ed PH |
804 | ) |
805 | self.report_warning( | |
806 | 'Long argument string detected. ' | |
49a57e70 | 807 | 'Use -- to separate parameters and URLs, like this:\n%s' % |
7d4111ed PH |
808 | args_to_str(correct_argv)) |
809 | ||
8222d8de JMF |
810 | def add_info_extractor(self, ie): |
811 | """Add an InfoExtractor object to the end of the list.""" | |
8b7491c8 | 812 | ie_key = ie.ie_key() |
813 | self._ies[ie_key] = ie | |
e52d7f85 | 814 | if not isinstance(ie, type): |
8b7491c8 | 815 | self._ies_instances[ie_key] = ie |
e52d7f85 | 816 | ie.set_downloader(self) |
8222d8de | 817 | |
56c73665 JMF |
818 | def get_info_extractor(self, ie_key): |
819 | """ | |
820 | Get an instance of an IE with name ie_key, it will try to get one from | |
821 | the _ies list, if there's no instance it will create a new one and add | |
822 | it to the extractor list. | |
823 | """ | |
824 | ie = self._ies_instances.get(ie_key) | |
825 | if ie is None: | |
826 | ie = get_info_extractor(ie_key)() | |
827 | self.add_info_extractor(ie) | |
828 | return ie | |
829 | ||
023fa8c4 JMF |
830 | def add_default_info_extractors(self): |
831 | """ | |
832 | Add the InfoExtractors returned by gen_extractors to the end of the list | |
833 | """ | |
fe7866d0 | 834 | all_ies = {ie.IE_NAME.lower(): ie for ie in gen_extractor_classes()} |
835 | all_ies['end'] = UnsupportedURLIE() | |
836 | try: | |
837 | ie_names = orderedSet_from_options( | |
838 | self.params.get('allowed_extractors', ['default']), { | |
839 | 'all': list(all_ies), | |
840 | 'default': [name for name, ie in all_ies.items() if ie._ENABLED], | |
841 | }, use_regex=True) | |
842 | except re.error as e: | |
843 | raise ValueError(f'Wrong regex for allowed_extractors: {e.pattern}') | |
844 | for name in ie_names: | |
845 | self.add_info_extractor(all_ies[name]) | |
846 | self.write_debug(f'Loaded {len(ie_names)} extractors') | |
023fa8c4 | 847 | |
56d868db | 848 | def add_post_processor(self, pp, when='post_process'): |
8222d8de | 849 | """Add a PostProcessor object to the end of the chain.""" |
8aa0e7cd | 850 | assert when in POSTPROCESS_WHEN, f'Invalid when={when}' |
5bfa4862 | 851 | self._pps[when].append(pp) |
8222d8de JMF |
852 | pp.set_downloader(self) |
853 | ||
ab8e5e51 AM |
854 | def add_post_hook(self, ph): |
855 | """Add the post hook""" | |
856 | self._post_hooks.append(ph) | |
857 | ||
933605d7 | 858 | def add_progress_hook(self, ph): |
819e0531 | 859 | """Add the download progress hook""" |
933605d7 | 860 | self._progress_hooks.append(ph) |
8ab470f1 | 861 | |
819e0531 | 862 | def add_postprocessor_hook(self, ph): |
863 | """Add the postprocessing progress hook""" | |
864 | self._postprocessor_hooks.append(ph) | |
5bfc8bee | 865 | for pps in self._pps.values(): |
866 | for pp in pps: | |
867 | pp.add_progress_hook(ph) | |
819e0531 | 868 | |
1c088fa8 | 869 | def _bidi_workaround(self, message): |
5d681e96 | 870 | if not hasattr(self, '_output_channel'): |
1c088fa8 PH |
871 | return message |
872 | ||
5d681e96 | 873 | assert hasattr(self, '_output_process') |
14f25df2 | 874 | assert isinstance(message, str) |
6febd1c1 | 875 | line_count = message.count('\n') + 1 |
0f06bcd7 | 876 | self._output_process.stdin.write((message + '\n').encode()) |
5d681e96 | 877 | self._output_process.stdin.flush() |
0f06bcd7 | 878 | res = ''.join(self._output_channel.readline().decode() |
9e1a5b84 | 879 | for _ in range(line_count)) |
6febd1c1 | 880 | return res[:-len('\n')] |
1c088fa8 | 881 | |
b35496d8 | 882 | def _write_string(self, message, out=None, only_once=False): |
883 | if only_once: | |
884 | if message in self._printed_messages: | |
885 | return | |
886 | self._printed_messages.add(message) | |
887 | write_string(message, out=out, encoding=self.params.get('encoding')) | |
734f90bb | 888 | |
cf4f42cb | 889 | def to_stdout(self, message, skip_eol=False, quiet=None): |
0760b0a7 | 890 | """Print message to stdout""" |
cf4f42cb | 891 | if quiet is not None: |
da4db748 | 892 | self.deprecation_warning('"YoutubeDL.to_stdout" no longer accepts the argument quiet. ' |
893 | 'Use "YoutubeDL.to_screen" instead') | |
8a82af35 | 894 | if skip_eol is not False: |
da4db748 | 895 | self.deprecation_warning('"YoutubeDL.to_stdout" no longer accepts the argument skip_eol. ' |
896 | 'Use "YoutubeDL.to_screen" instead') | |
0bf9dc1e | 897 | self._write_string(f'{self._bidi_workaround(message)}\n', self._out_files.out) |
cf4f42cb | 898 | |
dfea94f8 | 899 | def to_screen(self, message, skip_eol=False, quiet=None, only_once=False): |
cf4f42cb | 900 | """Print message to screen if not in quiet mode""" |
8bf9319e | 901 | if self.params.get('logger'): |
43afe285 | 902 | self.params['logger'].debug(message) |
cf4f42cb | 903 | return |
904 | if (self.params.get('quiet') if quiet is None else quiet) and not self.params.get('verbose'): | |
905 | return | |
906 | self._write_string( | |
907 | '%s%s' % (self._bidi_workaround(message), ('' if skip_eol else '\n')), | |
dfea94f8 | 908 | self._out_files.screen, only_once=only_once) |
8222d8de | 909 | |
b35496d8 | 910 | def to_stderr(self, message, only_once=False): |
0760b0a7 | 911 | """Print message to stderr""" |
14f25df2 | 912 | assert isinstance(message, str) |
8bf9319e | 913 | if self.params.get('logger'): |
43afe285 IB |
914 | self.params['logger'].error(message) |
915 | else: | |
5792c950 | 916 | self._write_string(f'{self._bidi_workaround(message)}\n', self._out_files.error, only_once=only_once) |
cf4f42cb | 917 | |
918 | def _send_console_code(self, code): | |
591bb9d3 | 919 | if compat_os_name == 'nt' or not self._out_files.console: |
cf4f42cb | 920 | return |
591bb9d3 | 921 | self._write_string(code, self._out_files.console) |
8222d8de | 922 | |
1e5b9a95 PH |
923 | def to_console_title(self, message): |
924 | if not self.params.get('consoletitle', False): | |
925 | return | |
3efb96a6 | 926 | message = remove_terminal_sequences(message) |
4bede0d8 C |
927 | if compat_os_name == 'nt': |
928 | if ctypes.windll.kernel32.GetConsoleWindow(): | |
929 | # c_wchar_p() might not be necessary if `message` is | |
930 | # already of type unicode() | |
931 | ctypes.windll.kernel32.SetConsoleTitleW(ctypes.c_wchar_p(message)) | |
cf4f42cb | 932 | else: |
933 | self._send_console_code(f'\033]0;{message}\007') | |
1e5b9a95 | 934 | |
bdde425c | 935 | def save_console_title(self): |
cf4f42cb | 936 | if not self.params.get('consoletitle') or self.params.get('simulate'): |
bdde425c | 937 | return |
592b7485 | 938 | self._send_console_code('\033[22;0t') # Save the title on stack |
bdde425c PH |
939 | |
940 | def restore_console_title(self): | |
cf4f42cb | 941 | if not self.params.get('consoletitle') or self.params.get('simulate'): |
bdde425c | 942 | return |
592b7485 | 943 | self._send_console_code('\033[23;0t') # Restore the title from stack |
bdde425c PH |
944 | |
945 | def __enter__(self): | |
946 | self.save_console_title() | |
947 | return self | |
948 | ||
c365dba8 | 949 | def save_cookies(self): |
dca08720 | 950 | if self.params.get('cookiefile') is not None: |
62b5c94c | 951 | self.cookiejar.save() |
bdde425c | 952 | |
c365dba8 | 953 | def __exit__(self, *args): |
954 | self.restore_console_title() | |
227bf1a3 | 955 | self.close() |
956 | ||
957 | def close(self): | |
c365dba8 | 958 | self.save_cookies() |
227bf1a3 | 959 | self._request_director.close() |
c365dba8 | 960 | |
fa9f30b8 | 961 | def trouble(self, message=None, tb=None, is_error=True): |
8222d8de JMF |
962 | """Determine action to take when a download problem appears. |
963 | ||
964 | Depending on if the downloader has been configured to ignore | |
965 | download errors or not, this method may throw an exception or | |
966 | not when errors are found, after printing the message. | |
967 | ||
fa9f30b8 | 968 | @param tb If given, is additional traceback information |
969 | @param is_error Whether to raise error according to ignorerrors | |
8222d8de JMF |
970 | """ |
971 | if message is not None: | |
972 | self.to_stderr(message) | |
973 | if self.params.get('verbose'): | |
974 | if tb is None: | |
975 | if sys.exc_info()[0]: # if .trouble has been called from an except block | |
6febd1c1 | 976 | tb = '' |
8222d8de | 977 | if hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]: |
6febd1c1 | 978 | tb += ''.join(traceback.format_exception(*sys.exc_info()[1].exc_info)) |
c0384f22 | 979 | tb += encode_compat_str(traceback.format_exc()) |
8222d8de JMF |
980 | else: |
981 | tb_data = traceback.format_list(traceback.extract_stack()) | |
6febd1c1 | 982 | tb = ''.join(tb_data) |
c19bc311 | 983 | if tb: |
984 | self.to_stderr(tb) | |
fa9f30b8 | 985 | if not is_error: |
986 | return | |
b1940459 | 987 | if not self.params.get('ignoreerrors'): |
8222d8de JMF |
988 | if sys.exc_info()[0] and hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]: |
989 | exc_info = sys.exc_info()[1].exc_info | |
990 | else: | |
991 | exc_info = sys.exc_info() | |
992 | raise DownloadError(message, exc_info) | |
993 | self._download_retcode = 1 | |
994 | ||
19a03940 | 995 | Styles = Namespace( |
996 | HEADERS='yellow', | |
997 | EMPHASIS='light blue', | |
492272fe | 998 | FILENAME='green', |
19a03940 | 999 | ID='green', |
1000 | DELIM='blue', | |
1001 | ERROR='red', | |
bc344cd4 | 1002 | BAD_FORMAT='light red', |
19a03940 | 1003 | WARNING='yellow', |
1004 | SUPPRESS='light black', | |
1005 | ) | |
ec11a9f4 | 1006 | |
7578d77d | 1007 | def _format_text(self, handle, allow_colors, text, f, fallback=None, *, test_encoding=False): |
e5a998f3 | 1008 | text = str(text) |
ec11a9f4 | 1009 | if test_encoding: |
1010 | original_text = text | |
5c104538 | 1011 | # handle.encoding can be None. See https://github.com/yt-dlp/yt-dlp/issues/2711 |
1012 | encoding = self.params.get('encoding') or getattr(handle, 'encoding', None) or 'ascii' | |
ec11a9f4 | 1013 | text = text.encode(encoding, 'ignore').decode(encoding) |
1014 | if fallback is not None and text != original_text: | |
1015 | text = fallback | |
8417f26b | 1016 | return format_text(text, f) if allow_colors is True else text if fallback is None else fallback |
ec11a9f4 | 1017 | |
591bb9d3 | 1018 | def _format_out(self, *args, **kwargs): |
1019 | return self._format_text(self._out_files.out, self._allow_colors.out, *args, **kwargs) | |
1020 | ||
ec11a9f4 | 1021 | def _format_screen(self, *args, **kwargs): |
591bb9d3 | 1022 | return self._format_text(self._out_files.screen, self._allow_colors.screen, *args, **kwargs) |
ec11a9f4 | 1023 | |
1024 | def _format_err(self, *args, **kwargs): | |
591bb9d3 | 1025 | return self._format_text(self._out_files.error, self._allow_colors.error, *args, **kwargs) |
819e0531 | 1026 | |
c84aeac6 | 1027 | def report_warning(self, message, only_once=False): |
8222d8de JMF |
1028 | ''' |
1029 | Print the message to stderr, it will be prefixed with 'WARNING:' | |
1030 | If stderr is a tty file the 'WARNING:' will be colored | |
1031 | ''' | |
6d07ce01 JMF |
1032 | if self.params.get('logger') is not None: |
1033 | self.params['logger'].warning(message) | |
8222d8de | 1034 | else: |
ad8915b7 PH |
1035 | if self.params.get('no_warnings'): |
1036 | return | |
ec11a9f4 | 1037 | self.to_stderr(f'{self._format_err("WARNING:", self.Styles.WARNING)} {message}', only_once) |
8222d8de | 1038 | |
da4db748 | 1039 | def deprecation_warning(self, message, *, stacklevel=0): |
1040 | deprecation_warning( | |
1041 | message, stacklevel=stacklevel + 1, printer=self.report_error, is_error=False) | |
1042 | ||
1043 | def deprecated_feature(self, message): | |
ee8dd27a | 1044 | if self.params.get('logger') is not None: |
da4db748 | 1045 | self.params['logger'].warning(f'Deprecated Feature: {message}') |
1046 | self.to_stderr(f'{self._format_err("Deprecated Feature:", self.Styles.ERROR)} {message}', True) | |
ee8dd27a | 1047 | |
fa9f30b8 | 1048 | def report_error(self, message, *args, **kwargs): |
8222d8de JMF |
1049 | ''' |
1050 | Do the same as trouble, but prefixes the message with 'ERROR:', colored | |
1051 | in red if stderr is a tty file. | |
1052 | ''' | |
fa9f30b8 | 1053 | self.trouble(f'{self._format_err("ERROR:", self.Styles.ERROR)} {message}', *args, **kwargs) |
8222d8de | 1054 | |
b35496d8 | 1055 | def write_debug(self, message, only_once=False): |
0760b0a7 | 1056 | '''Log debug message or Print message to stderr''' |
1057 | if not self.params.get('verbose', False): | |
1058 | return | |
8a82af35 | 1059 | message = f'[debug] {message}' |
0760b0a7 | 1060 | if self.params.get('logger'): |
1061 | self.params['logger'].debug(message) | |
1062 | else: | |
b35496d8 | 1063 | self.to_stderr(message, only_once) |
0760b0a7 | 1064 | |
8222d8de JMF |
1065 | def report_file_already_downloaded(self, file_name): |
1066 | """Report file has already been fully downloaded.""" | |
1067 | try: | |
6febd1c1 | 1068 | self.to_screen('[download] %s has already been downloaded' % file_name) |
ce02ed60 | 1069 | except UnicodeEncodeError: |
6febd1c1 | 1070 | self.to_screen('[download] The file has already been downloaded') |
8222d8de | 1071 | |
0c3d0f51 | 1072 | def report_file_delete(self, file_name): |
1073 | """Report that existing file will be deleted.""" | |
1074 | try: | |
c25228e5 | 1075 | self.to_screen('Deleting existing file %s' % file_name) |
0c3d0f51 | 1076 | except UnicodeEncodeError: |
c25228e5 | 1077 | self.to_screen('Deleting existing file') |
0c3d0f51 | 1078 | |
319b6059 | 1079 | def raise_no_formats(self, info, forced=False, *, msg=None): |
0a5a191a | 1080 | has_drm = info.get('_has_drm') |
319b6059 | 1081 | ignored, expected = self.params.get('ignore_no_formats_error'), bool(msg) |
1082 | msg = msg or has_drm and 'This video is DRM protected' or 'No video formats found!' | |
1083 | if forced or not ignored: | |
1151c407 | 1084 | raise ExtractorError(msg, video_id=info['id'], ie=info['extractor'], |
319b6059 | 1085 | expected=has_drm or ignored or expected) |
88acdbc2 | 1086 | else: |
1087 | self.report_warning(msg) | |
1088 | ||
de6000d9 | 1089 | def parse_outtmpl(self): |
bf1824b3 | 1090 | self.deprecation_warning('"YoutubeDL.parse_outtmpl" is deprecated and may be removed in a future version') |
1091 | self._parse_outtmpl() | |
1092 | return self.params['outtmpl'] | |
1093 | ||
1094 | def _parse_outtmpl(self): | |
7b2c3f47 | 1095 | sanitize = IDENTITY |
bf1824b3 | 1096 | if self.params.get('restrictfilenames'): # Remove spaces in the default template |
71ce444a | 1097 | sanitize = lambda x: x.replace(' - ', ' ').replace(' ', '-') |
bf1824b3 | 1098 | |
1099 | outtmpl = self.params.setdefault('outtmpl', {}) | |
1100 | if not isinstance(outtmpl, dict): | |
1101 | self.params['outtmpl'] = outtmpl = {'default': outtmpl} | |
1102 | outtmpl.update({k: sanitize(v) for k, v in DEFAULT_OUTTMPL.items() if outtmpl.get(k) is None}) | |
de6000d9 | 1103 | |
21cd8fae | 1104 | def get_output_path(self, dir_type='', filename=None): |
1105 | paths = self.params.get('paths', {}) | |
d2c8aadf | 1106 | assert isinstance(paths, dict), '"paths" parameter must be a dictionary' |
21cd8fae | 1107 | path = os.path.join( |
1108 | expand_path(paths.get('home', '').strip()), | |
1109 | expand_path(paths.get(dir_type, '').strip()) if dir_type else '', | |
1110 | filename or '') | |
21cd8fae | 1111 | return sanitize_path(path, force=self.params.get('windowsfilenames')) |
1112 | ||
76a264ac | 1113 | @staticmethod |
901130bb | 1114 | def _outtmpl_expandpath(outtmpl): |
1115 | # expand_path translates '%%' into '%' and '$$' into '$' | |
1116 | # correspondingly that is not what we want since we need to keep | |
1117 | # '%%' intact for template dict substitution step. Working around | |
1118 | # with boundary-alike separator hack. | |
6f2287cb | 1119 | sep = ''.join(random.choices(string.ascii_letters, k=32)) |
86e5f3ed | 1120 | outtmpl = outtmpl.replace('%%', f'%{sep}%').replace('$$', f'${sep}$') |
901130bb | 1121 | |
1122 | # outtmpl should be expand_path'ed before template dict substitution | |
1123 | # because meta fields may contain env variables we don't want to | |
62b58c09 | 1124 | # be expanded. E.g. for outtmpl "%(title)s.%(ext)s" and |
901130bb | 1125 | # title "Hello $PATH", we don't want `$PATH` to be expanded. |
1126 | return expand_path(outtmpl).replace(sep, '') | |
1127 | ||
1128 | @staticmethod | |
1129 | def escape_outtmpl(outtmpl): | |
1130 | ''' Escape any remaining strings like %s, %abc% etc. ''' | |
1131 | return re.sub( | |
1132 | STR_FORMAT_RE_TMPL.format('', '(?![%(\0])'), | |
1133 | lambda mobj: ('' if mobj.group('has_key') else '%') + mobj.group(0), | |
1134 | outtmpl) | |
1135 | ||
1136 | @classmethod | |
1137 | def validate_outtmpl(cls, outtmpl): | |
76a264ac | 1138 | ''' @return None or Exception object ''' |
7d1eb38a | 1139 | outtmpl = re.sub( |
47cdc68e | 1140 | STR_FORMAT_RE_TMPL.format('[^)]*', '[ljhqBUDS]'), |
7d1eb38a | 1141 | lambda mobj: f'{mobj.group(0)[:-1]}s', |
1142 | cls._outtmpl_expandpath(outtmpl)) | |
76a264ac | 1143 | try: |
7d1eb38a | 1144 | cls.escape_outtmpl(outtmpl) % collections.defaultdict(int) |
76a264ac | 1145 | return None |
1146 | except ValueError as err: | |
1147 | return err | |
1148 | ||
03b4de72 | 1149 | @staticmethod |
1150 | def _copy_infodict(info_dict): | |
1151 | info_dict = dict(info_dict) | |
09b49e1f | 1152 | info_dict.pop('__postprocessors', None) |
415f8d51 | 1153 | info_dict.pop('__pending_error', None) |
03b4de72 | 1154 | return info_dict |
1155 | ||
e0fd9573 | 1156 | def prepare_outtmpl(self, outtmpl, info_dict, sanitize=False): |
1157 | """ Make the outtmpl and info_dict suitable for substitution: ydl.escape_outtmpl(outtmpl) % info_dict | |
1158 | @param sanitize Whether to sanitize the output as a filename. | |
1159 | For backward compatibility, a function can also be passed | |
1160 | """ | |
1161 | ||
6e84b215 | 1162 | info_dict.setdefault('epoch', int(time.time())) # keep epoch consistent once set |
143db31d | 1163 | |
03b4de72 | 1164 | info_dict = self._copy_infodict(info_dict) |
752cda38 | 1165 | info_dict['duration_string'] = ( # %(duration>%H-%M-%S)s is wrong if duration > 24hrs |
53c18592 | 1166 | formatSeconds(info_dict['duration'], '-' if sanitize else ':') |
143db31d | 1167 | if info_dict.get('duration', None) is not None |
1168 | else None) | |
1d485a1a | 1169 | info_dict['autonumber'] = int(self.params.get('autonumber_start', 1) - 1 + self._num_downloads) |
9c906919 | 1170 | info_dict['video_autonumber'] = self._num_videos |
752cda38 | 1171 | if info_dict.get('resolution') is None: |
1172 | info_dict['resolution'] = self.format_resolution(info_dict, default=None) | |
143db31d | 1173 | |
e6f21b3d | 1174 | # For fields playlist_index, playlist_autonumber and autonumber convert all occurrences |
143db31d | 1175 | # of %(field)s to %(field)0Nd for backward compatibility |
1176 | field_size_compat_map = { | |
0a5a191a | 1177 | 'playlist_index': number_of_digits(info_dict.get('__last_playlist_index') or 0), |
ec11a9f4 | 1178 | 'playlist_autonumber': number_of_digits(info_dict.get('n_entries') or 0), |
752cda38 | 1179 | 'autonumber': self.params.get('autonumber_size') or 5, |
143db31d | 1180 | } |
752cda38 | 1181 | |
385a27fa | 1182 | TMPL_DICT = {} |
47cdc68e | 1183 | EXTERNAL_FORMAT_RE = re.compile(STR_FORMAT_RE_TMPL.format('[^)]*', f'[{STR_FORMAT_TYPES}ljhqBUDS]')) |
385a27fa | 1184 | MATH_FUNCTIONS = { |
1185 | '+': float.__add__, | |
1186 | '-': float.__sub__, | |
993edd3f | 1187 | '*': float.__mul__, |
385a27fa | 1188 | } |
e625be0d | 1189 | # Field is of the form key1.key2... |
07a1250e | 1190 | # where keys (except first) can be string, int, slice or "{field, ...}" |
1191 | FIELD_INNER_RE = r'(?:\w+|%(num)s|%(num)s?(?::%(num)s?){1,2})' % {'num': r'(?:-?\d+)'} | |
1192 | FIELD_RE = r'\w*(?:\.(?:%(inner)s|{%(field)s(?:,%(field)s)*}))*' % { | |
1193 | 'inner': FIELD_INNER_RE, | |
1194 | 'field': rf'\w*(?:\.{FIELD_INNER_RE})*' | |
1195 | } | |
1d485a1a | 1196 | MATH_FIELD_RE = rf'(?:{FIELD_RE}|-?{NUMBER_RE})' |
385a27fa | 1197 | MATH_OPERATORS_RE = r'(?:%s)' % '|'.join(map(re.escape, MATH_FUNCTIONS.keys())) |
78fde6e3 | 1198 | INTERNAL_FORMAT_RE = re.compile(rf'''(?xs) |
e625be0d | 1199 | (?P<negate>-)? |
1d485a1a | 1200 | (?P<fields>{FIELD_RE}) |
1201 | (?P<maths>(?:{MATH_OPERATORS_RE}{MATH_FIELD_RE})*) | |
e625be0d | 1202 | (?:>(?P<strf_format>.+?))? |
34baa9fd | 1203 | (?P<remaining> |
1204 | (?P<alternate>(?<!\\),[^|&)]+)? | |
1205 | (?:&(?P<replacement>.*?))? | |
1206 | (?:\|(?P<default>.*?))? | |
1d485a1a | 1207 | )$''') |
752cda38 | 1208 | |
0b6f829b SS |
1209 | def _from_user_input(field): |
1210 | if field == ':': | |
1211 | return ... | |
1212 | elif ':' in field: | |
1213 | return slice(*map(int_or_none, field.split(':'))) | |
1214 | elif int_or_none(field) is not None: | |
1215 | return int(field) | |
1216 | return field | |
1217 | ||
07a1250e | 1218 | def _traverse_infodict(fields): |
1219 | fields = [f for x in re.split(r'\.({.+?})\.?', fields) | |
1220 | for f in ([x] if x.startswith('{') else x.split('.'))] | |
1221 | for i in (0, -1): | |
1222 | if fields and not fields[i]: | |
1223 | fields.pop(i) | |
1224 | ||
1225 | for i, f in enumerate(fields): | |
1226 | if not f.startswith('{'): | |
0b6f829b | 1227 | fields[i] = _from_user_input(f) |
07a1250e | 1228 | continue |
1229 | assert f.endswith('}'), f'No closing brace for {f} in {fields}' | |
0b6f829b | 1230 | fields[i] = {k: list(map(_from_user_input, k.split('.'))) for k in f[1:-1].split(',')} |
07a1250e | 1231 | |
0b6f829b | 1232 | return traverse_obj(info_dict, fields, traverse_string=True) |
76a264ac | 1233 | |
752cda38 | 1234 | def get_value(mdict): |
1235 | # Object traversal | |
2b8a2973 | 1236 | value = _traverse_infodict(mdict['fields']) |
752cda38 | 1237 | # Negative |
1238 | if mdict['negate']: | |
1239 | value = float_or_none(value) | |
1240 | if value is not None: | |
1241 | value *= -1 | |
1242 | # Do maths | |
385a27fa | 1243 | offset_key = mdict['maths'] |
1244 | if offset_key: | |
752cda38 | 1245 | value = float_or_none(value) |
1246 | operator = None | |
385a27fa | 1247 | while offset_key: |
1248 | item = re.match( | |
1249 | MATH_FIELD_RE if operator else MATH_OPERATORS_RE, | |
1250 | offset_key).group(0) | |
1251 | offset_key = offset_key[len(item):] | |
1252 | if operator is None: | |
752cda38 | 1253 | operator = MATH_FUNCTIONS[item] |
385a27fa | 1254 | continue |
1255 | item, multiplier = (item[1:], -1) if item[0] == '-' else (item, 1) | |
1256 | offset = float_or_none(item) | |
1257 | if offset is None: | |
2b8a2973 | 1258 | offset = float_or_none(_traverse_infodict(item)) |
385a27fa | 1259 | try: |
1260 | value = operator(value, multiplier * offset) | |
1261 | except (TypeError, ZeroDivisionError): | |
1262 | return None | |
1263 | operator = None | |
752cda38 | 1264 | # Datetime formatting |
1265 | if mdict['strf_format']: | |
7c37ff97 | 1266 | value = strftime_or_none(value, mdict['strf_format'].replace('\\,', ',')) |
752cda38 | 1267 | |
a6bcaf71 | 1268 | # XXX: Workaround for https://github.com/yt-dlp/yt-dlp/issues/4485 |
1269 | if sanitize and value == '': | |
1270 | value = None | |
752cda38 | 1271 | return value |
1272 | ||
b868936c | 1273 | na = self.params.get('outtmpl_na_placeholder', 'NA') |
1274 | ||
e0fd9573 | 1275 | def filename_sanitizer(key, value, restricted=self.params.get('restrictfilenames')): |
5c3895ff | 1276 | return sanitize_filename(str(value), restricted=restricted, is_id=( |
1277 | bool(re.search(r'(^|[_.])id(\.|$)', key)) | |
8a82af35 | 1278 | if 'filename-sanitization' in self.params['compat_opts'] |
5c3895ff | 1279 | else NO_DEFAULT)) |
e0fd9573 | 1280 | |
1281 | sanitizer = sanitize if callable(sanitize) else filename_sanitizer | |
1282 | sanitize = bool(sanitize) | |
1283 | ||
6e84b215 | 1284 | def _dumpjson_default(obj): |
1285 | if isinstance(obj, (set, LazyList)): | |
1286 | return list(obj) | |
adbc4ec4 | 1287 | return repr(obj) |
6e84b215 | 1288 | |
6f2287cb | 1289 | class _ReplacementFormatter(string.Formatter): |
ec9311c4 | 1290 | def get_field(self, field_name, args, kwargs): |
1291 | if field_name.isdigit(): | |
1292 | return args[0], -1 | |
1293 | raise ValueError('Unsupported field') | |
1294 | ||
1295 | replacement_formatter = _ReplacementFormatter() | |
1296 | ||
752cda38 | 1297 | def create_key(outer_mobj): |
1298 | if not outer_mobj.group('has_key'): | |
b836dc94 | 1299 | return outer_mobj.group(0) |
752cda38 | 1300 | key = outer_mobj.group('key') |
752cda38 | 1301 | mobj = re.match(INTERNAL_FORMAT_RE, key) |
47bcd437 | 1302 | value, replacement, default, last_field = None, None, na, '' |
7c37ff97 | 1303 | while mobj: |
e625be0d | 1304 | mobj = mobj.groupdict() |
7c37ff97 | 1305 | default = mobj['default'] if mobj['default'] is not None else default |
752cda38 | 1306 | value = get_value(mobj) |
47bcd437 | 1307 | last_field, replacement = mobj['fields'], mobj['replacement'] |
7c37ff97 | 1308 | if value is None and mobj['alternate']: |
34baa9fd | 1309 | mobj = re.match(INTERNAL_FORMAT_RE, mobj['remaining'][1:]) |
7c37ff97 | 1310 | else: |
1311 | break | |
752cda38 | 1312 | |
ebe1b4e3 | 1313 | if None not in (value, replacement): |
ec9311c4 | 1314 | try: |
1315 | value = replacement_formatter.format(replacement, value) | |
1316 | except ValueError: | |
ebe1b4e3 | 1317 | value, default = None, na |
752cda38 | 1318 | |
a264433c | 1319 | fmt = outer_mobj.group('format') |
1320 | if fmt == 's' and last_field in field_size_compat_map.keys() and isinstance(value, int): | |
1321 | fmt = f'0{field_size_compat_map[last_field]:d}d' | |
1322 | ||
4476d2c7 | 1323 | flags = outer_mobj.group('conversion') or '' |
7d1eb38a | 1324 | str_fmt = f'{fmt[:-1]}s' |
ebe1b4e3 | 1325 | if value is None: |
1326 | value, fmt = default, 's' | |
1327 | elif fmt[-1] == 'l': # list | |
4476d2c7 | 1328 | delim = '\n' if '#' in flags else ', ' |
9e907ebd | 1329 | value, fmt = delim.join(map(str, variadic(value, allowed_types=(str, bytes)))), str_fmt |
524e2e4f | 1330 | elif fmt[-1] == 'j': # json |
deae7c17 | 1331 | value, fmt = json.dumps( |
1332 | value, default=_dumpjson_default, | |
9b9dad11 | 1333 | indent=4 if '#' in flags else None, ensure_ascii='+' not in flags), str_fmt |
47cdc68e | 1334 | elif fmt[-1] == 'h': # html |
deae7c17 | 1335 | value, fmt = escapeHTML(str(value)), str_fmt |
524e2e4f | 1336 | elif fmt[-1] == 'q': # quoted |
4476d2c7 | 1337 | value = map(str, variadic(value) if '#' in flags else [value]) |
1338 | value, fmt = ' '.join(map(compat_shlex_quote, value)), str_fmt | |
524e2e4f | 1339 | elif fmt[-1] == 'B': # bytes |
0f06bcd7 | 1340 | value = f'%{str_fmt}'.encode() % str(value).encode() |
f5aa5cfb | 1341 | value, fmt = value.decode('utf-8', 'ignore'), 's' |
524e2e4f | 1342 | elif fmt[-1] == 'U': # unicode normalized |
524e2e4f | 1343 | value, fmt = unicodedata.normalize( |
1344 | # "+" = compatibility equivalence, "#" = NFD | |
4476d2c7 | 1345 | 'NF%s%s' % ('K' if '+' in flags else '', 'D' if '#' in flags else 'C'), |
524e2e4f | 1346 | value), str_fmt |
e0fd9573 | 1347 | elif fmt[-1] == 'D': # decimal suffix |
abbeeebc | 1348 | num_fmt, fmt = fmt[:-1].replace('#', ''), 's' |
1349 | value = format_decimal_suffix(value, f'%{num_fmt}f%s' if num_fmt else '%d%s', | |
1350 | factor=1024 if '#' in flags else 1000) | |
37893bb0 | 1351 | elif fmt[-1] == 'S': # filename sanitization |
47bcd437 | 1352 | value, fmt = filename_sanitizer(last_field, value, restricted='#' in flags), str_fmt |
7d1eb38a | 1353 | elif fmt[-1] == 'c': |
524e2e4f | 1354 | if value: |
1355 | value = str(value)[0] | |
76a264ac | 1356 | else: |
524e2e4f | 1357 | fmt = str_fmt |
1619ab3e | 1358 | elif fmt[-1] not in 'rsa': # numeric |
a439a3a4 | 1359 | value = float_or_none(value) |
752cda38 | 1360 | if value is None: |
1361 | value, fmt = default, 's' | |
901130bb | 1362 | |
752cda38 | 1363 | if sanitize: |
1619ab3e | 1364 | # If value is an object, sanitize might convert it to a string |
1365 | # So we convert it to repr first | |
752cda38 | 1366 | if fmt[-1] == 'r': |
7d1eb38a | 1367 | value, fmt = repr(value), str_fmt |
1619ab3e | 1368 | elif fmt[-1] == 'a': |
1369 | value, fmt = ascii(value), str_fmt | |
1370 | if fmt[-1] in 'csra': | |
47bcd437 | 1371 | value = sanitizer(last_field, value) |
901130bb | 1372 | |
b868936c | 1373 | key = '%s\0%s' % (key.replace('%', '%\0'), outer_mobj.group('format')) |
385a27fa | 1374 | TMPL_DICT[key] = value |
b868936c | 1375 | return '{prefix}%({key}){fmt}'.format(key=key, fmt=fmt, prefix=outer_mobj.group('prefix')) |
752cda38 | 1376 | |
385a27fa | 1377 | return EXTERNAL_FORMAT_RE.sub(create_key, outtmpl), TMPL_DICT |
143db31d | 1378 | |
819e0531 | 1379 | def evaluate_outtmpl(self, outtmpl, info_dict, *args, **kwargs): |
1380 | outtmpl, info_dict = self.prepare_outtmpl(outtmpl, info_dict, *args, **kwargs) | |
1381 | return self.escape_outtmpl(outtmpl) % info_dict | |
1382 | ||
5127e92a | 1383 | def _prepare_filename(self, info_dict, *, outtmpl=None, tmpl_type=None): |
1384 | assert None in (outtmpl, tmpl_type), 'outtmpl and tmpl_type are mutually exclusive' | |
1385 | if outtmpl is None: | |
bf1824b3 | 1386 | outtmpl = self.params['outtmpl'].get(tmpl_type or 'default', self.params['outtmpl']['default']) |
8222d8de | 1387 | try: |
5127e92a | 1388 | outtmpl = self._outtmpl_expandpath(outtmpl) |
e0fd9573 | 1389 | filename = self.evaluate_outtmpl(outtmpl, info_dict, True) |
6a0546e3 | 1390 | if not filename: |
1391 | return None | |
15da37c7 | 1392 | |
5127e92a | 1393 | if tmpl_type in ('', 'temp'): |
6a0546e3 | 1394 | final_ext, ext = self.params.get('final_ext'), info_dict.get('ext') |
1395 | if final_ext and ext and final_ext != ext and filename.endswith(f'.{final_ext}'): | |
1396 | filename = replace_extension(filename, ext, final_ext) | |
5127e92a | 1397 | elif tmpl_type: |
6a0546e3 | 1398 | force_ext = OUTTMPL_TYPES[tmpl_type] |
1399 | if force_ext: | |
1400 | filename = replace_extension(filename, force_ext, info_dict.get('ext')) | |
de6000d9 | 1401 | |
bdc3fd2f U |
1402 | # https://github.com/blackjack4494/youtube-dlc/issues/85 |
1403 | trim_file_name = self.params.get('trim_file_name', False) | |
1404 | if trim_file_name: | |
5c22c63d | 1405 | no_ext, *ext = filename.rsplit('.', 2) |
1406 | filename = join_nonempty(no_ext[:trim_file_name], *ext, delim='.') | |
bdc3fd2f | 1407 | |
0202b52a | 1408 | return filename |
8222d8de | 1409 | except ValueError as err: |
6febd1c1 | 1410 | self.report_error('Error in output template: ' + str(err) + ' (encoding: ' + repr(preferredencoding()) + ')') |
8222d8de JMF |
1411 | return None |
1412 | ||
5127e92a | 1413 | def prepare_filename(self, info_dict, dir_type='', *, outtmpl=None, warn=False): |
1414 | """Generate the output filename""" | |
1415 | if outtmpl: | |
1416 | assert not dir_type, 'outtmpl and dir_type are mutually exclusive' | |
1417 | dir_type = None | |
1418 | filename = self._prepare_filename(info_dict, tmpl_type=dir_type, outtmpl=outtmpl) | |
80c03fa9 | 1419 | if not filename and dir_type not in ('', 'temp'): |
1420 | return '' | |
de6000d9 | 1421 | |
c84aeac6 | 1422 | if warn: |
21cd8fae | 1423 | if not self.params.get('paths'): |
de6000d9 | 1424 | pass |
1425 | elif filename == '-': | |
c84aeac6 | 1426 | self.report_warning('--paths is ignored when an outputting to stdout', only_once=True) |
de6000d9 | 1427 | elif os.path.isabs(filename): |
c84aeac6 | 1428 | self.report_warning('--paths is ignored since an absolute path is given in output template', only_once=True) |
de6000d9 | 1429 | if filename == '-' or not filename: |
1430 | return filename | |
1431 | ||
21cd8fae | 1432 | return self.get_output_path(dir_type, filename) |
0202b52a | 1433 | |
120fe513 | 1434 | def _match_entry(self, info_dict, incomplete=False, silent=False): |
6368e2e6 | 1435 | """Returns None if the file should be downloaded""" |
93b39cdb | 1436 | _type = 'video' if 'playlist-match-filter' in self.params['compat_opts'] else info_dict.get('_type', 'video') |
d7b460d0 | 1437 | assert incomplete or _type == 'video', 'Only video result can be considered complete' |
8222d8de | 1438 | |
3bec830a | 1439 | video_title = info_dict.get('title', info_dict.get('id', 'entry')) |
c77495e3 | 1440 | |
8b0d7497 | 1441 | def check_filter(): |
d7b460d0 | 1442 | if _type in ('playlist', 'multi_video'): |
1443 | return | |
1444 | elif _type in ('url', 'url_transparent') and not try_call( | |
1445 | lambda: self.get_info_extractor(info_dict['ie_key']).is_single_video(info_dict['url'])): | |
1446 | return | |
1447 | ||
8b0d7497 | 1448 | if 'title' in info_dict: |
1449 | # This can happen when we're just evaluating the playlist | |
1450 | title = info_dict['title'] | |
1451 | matchtitle = self.params.get('matchtitle', False) | |
1452 | if matchtitle: | |
1453 | if not re.search(matchtitle, title, re.IGNORECASE): | |
1454 | return '"' + title + '" title did not match pattern "' + matchtitle + '"' | |
1455 | rejecttitle = self.params.get('rejecttitle', False) | |
1456 | if rejecttitle: | |
1457 | if re.search(rejecttitle, title, re.IGNORECASE): | |
1458 | return '"' + title + '" title matched reject pattern "' + rejecttitle + '"' | |
6368e2e6 | 1459 | |
8b0d7497 | 1460 | date = info_dict.get('upload_date') |
1461 | if date is not None: | |
1462 | dateRange = self.params.get('daterange', DateRange()) | |
1463 | if date not in dateRange: | |
86e5f3ed | 1464 | return f'{date_from_str(date).isoformat()} upload date is not in range {dateRange}' |
8b0d7497 | 1465 | view_count = info_dict.get('view_count') |
1466 | if view_count is not None: | |
1467 | min_views = self.params.get('min_views') | |
1468 | if min_views is not None and view_count < min_views: | |
1469 | return 'Skipping %s, because it has not reached minimum view count (%d/%d)' % (video_title, view_count, min_views) | |
1470 | max_views = self.params.get('max_views') | |
1471 | if max_views is not None and view_count > max_views: | |
1472 | return 'Skipping %s, because it has exceeded the maximum view count (%d/%d)' % (video_title, view_count, max_views) | |
1473 | if age_restricted(info_dict.get('age_limit'), self.params.get('age_limit')): | |
1474 | return 'Skipping "%s" because it is age restricted' % video_title | |
8b0d7497 | 1475 | |
8f18aca8 | 1476 | match_filter = self.params.get('match_filter') |
fe2ce85a | 1477 | if match_filter is None: |
1478 | return None | |
1479 | ||
1480 | cancelled = None | |
1481 | try: | |
8f18aca8 | 1482 | try: |
1483 | ret = match_filter(info_dict, incomplete=incomplete) | |
1484 | except TypeError: | |
1485 | # For backward compatibility | |
1486 | ret = None if incomplete else match_filter(info_dict) | |
fe2ce85a | 1487 | except DownloadCancelled as err: |
1488 | if err.msg is not NO_DEFAULT: | |
1489 | raise | |
1490 | ret, cancelled = err.msg, err | |
1491 | ||
1492 | if ret is NO_DEFAULT: | |
1493 | while True: | |
1494 | filename = self._format_screen(self.prepare_filename(info_dict), self.Styles.FILENAME) | |
1495 | reply = input(self._format_screen( | |
1496 | f'Download "{filename}"? (Y/n): ', self.Styles.EMPHASIS)).lower().strip() | |
1497 | if reply in {'y', ''}: | |
1498 | return None | |
1499 | elif reply == 'n': | |
1500 | if cancelled: | |
1501 | raise type(cancelled)(f'Skipping {video_title}') | |
1502 | return f'Skipping {video_title}' | |
1503 | return ret | |
8b0d7497 | 1504 | |
c77495e3 | 1505 | if self.in_download_archive(info_dict): |
2b029ca0 AK |
1506 | reason = ''.join(( |
1507 | format_field(info_dict, 'id', f'{self._format_screen("%s", self.Styles.ID)}: '), | |
1508 | format_field(info_dict, 'title', f'{self._format_screen("%s", self.Styles.EMPHASIS)} '), | |
1509 | 'has already been recorded in the archive')) | |
c77495e3 | 1510 | break_opt, break_err = 'break_on_existing', ExistingVideoReached |
1511 | else: | |
fe2ce85a | 1512 | try: |
1513 | reason = check_filter() | |
1514 | except DownloadCancelled as e: | |
1515 | reason, break_opt, break_err = e.msg, 'match_filter', type(e) | |
1516 | else: | |
1517 | break_opt, break_err = 'break_on_reject', RejectedVideoReached | |
8b0d7497 | 1518 | if reason is not None: |
120fe513 | 1519 | if not silent: |
1520 | self.to_screen('[download] ' + reason) | |
c77495e3 | 1521 | if self.params.get(break_opt, False): |
1522 | raise break_err() | |
8b0d7497 | 1523 | return reason |
fe7e0c98 | 1524 | |
b6c45014 JMF |
1525 | @staticmethod |
1526 | def add_extra_info(info_dict, extra_info): | |
1527 | '''Set the keys from extra_info in info dict if they are missing''' | |
1528 | for key, value in extra_info.items(): | |
1529 | info_dict.setdefault(key, value) | |
1530 | ||
409e1828 | 1531 | def extract_info(self, url, download=True, ie_key=None, extra_info=None, |
61aa5ba3 | 1532 | process=True, force_generic_extractor=False): |
41d1cca3 | 1533 | """ |
17ffed18 | 1534 | Extract and return the information dictionary of the URL |
41d1cca3 | 1535 | |
1536 | Arguments: | |
17ffed18 | 1537 | @param url URL to extract |
41d1cca3 | 1538 | |
1539 | Keyword arguments: | |
17ffed18 | 1540 | @param download Whether to download videos |
1541 | @param process Whether to resolve all unresolved references (URLs, playlist items). | |
1542 | Must be True for download to work | |
1543 | @param ie_key Use only the extractor with this key | |
1544 | ||
1545 | @param extra_info Dictionary containing the extra values to add to the info (For internal use only) | |
1546 | @force_generic_extractor Force using the generic extractor (Deprecated; use ie_key='Generic') | |
41d1cca3 | 1547 | """ |
fe7e0c98 | 1548 | |
409e1828 | 1549 | if extra_info is None: |
1550 | extra_info = {} | |
1551 | ||
61aa5ba3 | 1552 | if not ie_key and force_generic_extractor: |
d22dec74 S |
1553 | ie_key = 'Generic' |
1554 | ||
8222d8de | 1555 | if ie_key: |
fe7866d0 | 1556 | ies = {ie_key: self._ies[ie_key]} if ie_key in self._ies else {} |
8222d8de JMF |
1557 | else: |
1558 | ies = self._ies | |
1559 | ||
fe7866d0 | 1560 | for key, ie in ies.items(): |
8222d8de JMF |
1561 | if not ie.suitable(url): |
1562 | continue | |
1563 | ||
1564 | if not ie.working(): | |
6febd1c1 PH |
1565 | self.report_warning('The program functionality for this site has been marked as broken, ' |
1566 | 'and will probably not work.') | |
8222d8de | 1567 | |
1151c407 | 1568 | temp_id = ie.get_temp_id(url) |
fe7866d0 | 1569 | if temp_id is not None and self.in_download_archive({'id': temp_id, 'ie_key': key}): |
2b029ca0 AK |
1570 | self.to_screen(f'[download] {self._format_screen(temp_id, self.Styles.ID)}: ' |
1571 | 'has already been recorded in the archive') | |
5e5be0c0 | 1572 | if self.params.get('break_on_existing', False): |
1573 | raise ExistingVideoReached() | |
a0566bbf | 1574 | break |
fe7866d0 | 1575 | return self.__extract_info(url, self.get_info_extractor(key), download, extra_info, process) |
a0566bbf | 1576 | else: |
fe7866d0 | 1577 | extractors_restricted = self.params.get('allowed_extractors') not in (None, ['default']) |
1578 | self.report_error(f'No suitable extractor{format_field(ie_key, None, " (%s)")} found for URL {url}', | |
1579 | tb=False if extractors_restricted else None) | |
a0566bbf | 1580 | |
7e88d7d7 | 1581 | def _handle_extraction_exceptions(func): |
b5ae35ee | 1582 | @functools.wraps(func) |
a0566bbf | 1583 | def wrapper(self, *args, **kwargs): |
6da22e7d | 1584 | while True: |
1585 | try: | |
1586 | return func(self, *args, **kwargs) | |
1587 | except (DownloadCancelled, LazyList.IndexError, PagedList.IndexError): | |
8222d8de | 1588 | raise |
6da22e7d | 1589 | except ReExtractInfo as e: |
1590 | if e.expected: | |
1591 | self.to_screen(f'{e}; Re-extracting data') | |
1592 | else: | |
1593 | self.to_stderr('\r') | |
1594 | self.report_warning(f'{e}; Re-extracting data') | |
1595 | continue | |
1596 | except GeoRestrictedError as e: | |
1597 | msg = e.msg | |
1598 | if e.countries: | |
1599 | msg += '\nThis video is available in %s.' % ', '.join( | |
1600 | map(ISO3166Utils.short2full, e.countries)) | |
1601 | msg += '\nYou might want to use a VPN or a proxy server (with --proxy) to workaround.' | |
1602 | self.report_error(msg) | |
1603 | except ExtractorError as e: # An error we somewhat expected | |
1604 | self.report_error(str(e), e.format_traceback()) | |
1605 | except Exception as e: | |
1606 | if self.params.get('ignoreerrors'): | |
1607 | self.report_error(str(e), tb=encode_compat_str(traceback.format_exc())) | |
1608 | else: | |
1609 | raise | |
1610 | break | |
a0566bbf | 1611 | return wrapper |
1612 | ||
693f0600 | 1613 | def _wait_for_video(self, ie_result={}): |
f2ebc5c7 | 1614 | if (not self.params.get('wait_for_video') |
1615 | or ie_result.get('_type', 'video') != 'video' | |
1616 | or ie_result.get('formats') or ie_result.get('url')): | |
1617 | return | |
1618 | ||
1619 | format_dur = lambda dur: '%02d:%02d:%02d' % timetuple_from_msec(dur * 1000)[:-1] | |
1620 | last_msg = '' | |
1621 | ||
1622 | def progress(msg): | |
1623 | nonlocal last_msg | |
a7dc6a89 | 1624 | full_msg = f'{msg}\n' |
1625 | if not self.params.get('noprogress'): | |
1626 | full_msg = msg + ' ' * (len(last_msg) - len(msg)) + '\r' | |
1627 | elif last_msg: | |
1628 | return | |
1629 | self.to_screen(full_msg, skip_eol=True) | |
f2ebc5c7 | 1630 | last_msg = msg |
1631 | ||
1632 | min_wait, max_wait = self.params.get('wait_for_video') | |
1633 | diff = try_get(ie_result, lambda x: x['release_timestamp'] - time.time()) | |
1634 | if diff is None and ie_result.get('live_status') == 'is_upcoming': | |
16c620bc | 1635 | diff = round(random.uniform(min_wait, max_wait) if (max_wait and min_wait) else (max_wait or min_wait), 0) |
f2ebc5c7 | 1636 | self.report_warning('Release time of video is not known') |
693f0600 | 1637 | elif ie_result and (diff or 0) <= 0: |
f2ebc5c7 | 1638 | self.report_warning('Video should already be available according to extracted info') |
38d79fd1 | 1639 | diff = min(max(diff or 0, min_wait or 0), max_wait or float('inf')) |
f2ebc5c7 | 1640 | self.to_screen(f'[wait] Waiting for {format_dur(diff)} - Press Ctrl+C to try now') |
1641 | ||
1642 | wait_till = time.time() + diff | |
1643 | try: | |
1644 | while True: | |
1645 | diff = wait_till - time.time() | |
1646 | if diff <= 0: | |
1647 | progress('') | |
1648 | raise ReExtractInfo('[wait] Wait period ended', expected=True) | |
1649 | progress(f'[wait] Remaining time until next attempt: {self._format_screen(format_dur(diff), self.Styles.EMPHASIS)}') | |
1650 | time.sleep(1) | |
1651 | except KeyboardInterrupt: | |
1652 | progress('') | |
1653 | raise ReExtractInfo('[wait] Interrupted by user', expected=True) | |
1654 | except BaseException as e: | |
1655 | if not isinstance(e, ReExtractInfo): | |
1656 | self.to_screen('') | |
1657 | raise | |
1658 | ||
6c5211ce | 1659 | def _load_cookies(self, data, *, autoscope=True): |
31215122 SS |
1660 | """Loads cookies from a `Cookie` header |
1661 | ||
1662 | This tries to work around the security vulnerability of passing cookies to every domain. | |
1663 | See: https://github.com/yt-dlp/yt-dlp/security/advisories/GHSA-v8mc-9377-rwjj | |
31215122 SS |
1664 | |
1665 | @param data The Cookie header as string to load the cookies from | |
6c5211ce | 1666 | @param autoscope If `False`, scope cookies using Set-Cookie syntax and error for cookie without domains |
1667 | If `True`, save cookies for later to be stored in the jar with a limited scope | |
1668 | If a URL, save cookies in the jar with the domain of the URL | |
31215122 SS |
1669 | """ |
1670 | for cookie in LenientSimpleCookie(data).values(): | |
6c5211ce | 1671 | if autoscope and any(cookie.values()): |
31215122 SS |
1672 | raise ValueError('Invalid syntax in Cookie Header') |
1673 | ||
1674 | domain = cookie.get('domain') or '' | |
1675 | expiry = cookie.get('expires') | |
1676 | if expiry == '': # 0 is valid | |
1677 | expiry = None | |
1678 | prepared_cookie = http.cookiejar.Cookie( | |
1679 | cookie.get('version') or 0, cookie.key, cookie.value, None, False, | |
1680 | domain, True, True, cookie.get('path') or '', bool(cookie.get('path')), | |
1681 | cookie.get('secure') or False, expiry, False, None, None, {}) | |
1682 | ||
1683 | if domain: | |
1684 | self.cookiejar.set_cookie(prepared_cookie) | |
6c5211ce | 1685 | elif autoscope is True: |
31215122 SS |
1686 | self.deprecated_feature( |
1687 | 'Passing cookies as a header is a potential security risk; ' | |
1688 | 'they will be scoped to the domain of the downloaded urls. ' | |
1689 | 'Please consider loading cookies from a file or browser instead.') | |
1690 | self.__header_cookies.append(prepared_cookie) | |
6c5211ce | 1691 | elif autoscope: |
1692 | self.report_warning( | |
1693 | 'The extractor result contains an unscoped cookie as an HTTP header. ' | |
1694 | f'If you are using yt-dlp with an input URL{bug_reports_message(before=",")}', | |
1695 | only_once=True) | |
1696 | self._apply_header_cookies(autoscope, [prepared_cookie]) | |
31215122 SS |
1697 | else: |
1698 | self.report_error('Unscoped cookies are not allowed; please specify some sort of scoping', | |
1699 | tb=False, is_error=False) | |
1700 | ||
6c5211ce | 1701 | def _apply_header_cookies(self, url, cookies=None): |
31215122 SS |
1702 | """Applies stray header cookies to the provided url |
1703 | ||
1704 | This loads header cookies and scopes them to the domain provided in `url`. | |
1705 | While this is not ideal, it helps reduce the risk of them being sent | |
1706 | to an unintended destination while mostly maintaining compatibility. | |
1707 | """ | |
1708 | parsed = urllib.parse.urlparse(url) | |
1709 | if not parsed.hostname: | |
1710 | return | |
1711 | ||
6c5211ce | 1712 | for cookie in map(copy.copy, cookies or self.__header_cookies): |
31215122 SS |
1713 | cookie.domain = f'.{parsed.hostname}' |
1714 | self.cookiejar.set_cookie(cookie) | |
1715 | ||
7e88d7d7 | 1716 | @_handle_extraction_exceptions |
58f197b7 | 1717 | def __extract_info(self, url, ie, download, extra_info, process): |
31215122 SS |
1718 | self._apply_header_cookies(url) |
1719 | ||
693f0600 | 1720 | try: |
1721 | ie_result = ie.extract(url) | |
1722 | except UserNotLive as e: | |
1723 | if process: | |
1724 | if self.params.get('wait_for_video'): | |
1725 | self.report_warning(e) | |
1726 | self._wait_for_video() | |
1727 | raise | |
a0566bbf | 1728 | if ie_result is None: # Finished already (backwards compatibility; listformats and friends should be moved here) |
cb794ee0 | 1729 | self.report_warning(f'Extractor {ie.IE_NAME} returned nothing{bug_reports_message()}') |
a0566bbf | 1730 | return |
1731 | if isinstance(ie_result, list): | |
1732 | # Backwards compatibility: old IE result format | |
1733 | ie_result = { | |
1734 | '_type': 'compat_list', | |
1735 | 'entries': ie_result, | |
1736 | } | |
e37d0efb | 1737 | if extra_info.get('original_url'): |
1738 | ie_result.setdefault('original_url', extra_info['original_url']) | |
a0566bbf | 1739 | self.add_default_extra_info(ie_result, ie, url) |
1740 | if process: | |
f2ebc5c7 | 1741 | self._wait_for_video(ie_result) |
a0566bbf | 1742 | return self.process_ie_result(ie_result, download, extra_info) |
8222d8de | 1743 | else: |
a0566bbf | 1744 | return ie_result |
fe7e0c98 | 1745 | |
ea38e55f | 1746 | def add_default_extra_info(self, ie_result, ie, url): |
6033d980 | 1747 | if url is not None: |
1748 | self.add_extra_info(ie_result, { | |
1749 | 'webpage_url': url, | |
1750 | 'original_url': url, | |
57ebfca3 | 1751 | }) |
1752 | webpage_url = ie_result.get('webpage_url') | |
1753 | if webpage_url: | |
1754 | self.add_extra_info(ie_result, { | |
1755 | 'webpage_url_basename': url_basename(webpage_url), | |
1756 | 'webpage_url_domain': get_domain(webpage_url), | |
6033d980 | 1757 | }) |
1758 | if ie is not None: | |
1759 | self.add_extra_info(ie_result, { | |
1760 | 'extractor': ie.IE_NAME, | |
1761 | 'extractor_key': ie.ie_key(), | |
1762 | }) | |
ea38e55f | 1763 | |
58adec46 | 1764 | def process_ie_result(self, ie_result, download=True, extra_info=None): |
8222d8de JMF |
1765 | """ |
1766 | Take the result of the ie(may be modified) and resolve all unresolved | |
1767 | references (URLs, playlist items). | |
1768 | ||
1769 | It will also download the videos if 'download'. | |
1770 | Returns the resolved ie_result. | |
1771 | """ | |
58adec46 | 1772 | if extra_info is None: |
1773 | extra_info = {} | |
e8ee972c PH |
1774 | result_type = ie_result.get('_type', 'video') |
1775 | ||
057a5206 | 1776 | if result_type in ('url', 'url_transparent'): |
8f97a15d | 1777 | ie_result['url'] = sanitize_url( |
1778 | ie_result['url'], scheme='http' if self.params.get('prefer_insecure') else 'https') | |
8791e78c | 1779 | if ie_result.get('original_url') and not extra_info.get('original_url'): |
1780 | extra_info = {'original_url': ie_result['original_url'], **extra_info} | |
e37d0efb | 1781 | |
057a5206 | 1782 | extract_flat = self.params.get('extract_flat', False) |
3089bc74 S |
1783 | if ((extract_flat == 'in_playlist' and 'playlist' in extra_info) |
1784 | or extract_flat is True): | |
ecb54191 | 1785 | info_copy = ie_result.copy() |
6033d980 | 1786 | ie = try_get(ie_result.get('ie_key'), self.get_info_extractor) |
360167b9 | 1787 | if ie and not ie_result.get('id'): |
4614bc22 | 1788 | info_copy['id'] = ie.get_temp_id(ie_result['url']) |
6033d980 | 1789 | self.add_default_extra_info(info_copy, ie, ie_result['url']) |
4614bc22 | 1790 | self.add_extra_info(info_copy, extra_info) |
b5475f11 | 1791 | info_copy, _ = self.pre_process(info_copy) |
94dc8604 | 1792 | self._fill_common_fields(info_copy, False) |
17060584 | 1793 | self.__forced_printings(info_copy) |
415f8d51 | 1794 | self._raise_pending_errors(info_copy) |
4614bc22 | 1795 | if self.params.get('force_write_download_archive', False): |
1796 | self.record_download_archive(info_copy) | |
e8ee972c PH |
1797 | return ie_result |
1798 | ||
8222d8de | 1799 | if result_type == 'video': |
b6c45014 | 1800 | self.add_extra_info(ie_result, extra_info) |
9c2b75b5 | 1801 | ie_result = self.process_video_result(ie_result, download=download) |
415f8d51 | 1802 | self._raise_pending_errors(ie_result) |
28b0eb0f | 1803 | additional_urls = (ie_result or {}).get('additional_urls') |
9c2b75b5 | 1804 | if additional_urls: |
e9f4ccd1 | 1805 | # TODO: Improve MetadataParserPP to allow setting a list |
14f25df2 | 1806 | if isinstance(additional_urls, str): |
9c2b75b5 | 1807 | additional_urls = [additional_urls] |
1808 | self.to_screen( | |
1809 | '[info] %s: %d additional URL(s) requested' % (ie_result['id'], len(additional_urls))) | |
1810 | self.write_debug('Additional URLs: "%s"' % '", "'.join(additional_urls)) | |
1811 | ie_result['additional_entries'] = [ | |
1812 | self.extract_info( | |
b69fd25c | 1813 | url, download, extra_info=extra_info, |
9c2b75b5 | 1814 | force_generic_extractor=self.params.get('force_generic_extractor')) |
1815 | for url in additional_urls | |
1816 | ] | |
1817 | return ie_result | |
8222d8de JMF |
1818 | elif result_type == 'url': |
1819 | # We have to add extra_info to the results because it may be | |
1820 | # contained in a playlist | |
07cce701 | 1821 | return self.extract_info( |
1822 | ie_result['url'], download, | |
1823 | ie_key=ie_result.get('ie_key'), | |
1824 | extra_info=extra_info) | |
7fc3fa05 PH |
1825 | elif result_type == 'url_transparent': |
1826 | # Use the information from the embedding page | |
1827 | info = self.extract_info( | |
1828 | ie_result['url'], ie_key=ie_result.get('ie_key'), | |
1829 | extra_info=extra_info, download=False, process=False) | |
1830 | ||
1640eb09 S |
1831 | # extract_info may return None when ignoreerrors is enabled and |
1832 | # extraction failed with an error, don't crash and return early | |
1833 | # in this case | |
1834 | if not info: | |
1835 | return info | |
1836 | ||
3975b4d2 | 1837 | exempted_fields = {'_type', 'url', 'ie_key'} |
1838 | if not ie_result.get('section_end') and ie_result.get('section_start') is None: | |
1839 | # For video clips, the id etc of the clip extractor should be used | |
1840 | exempted_fields |= {'id', 'extractor', 'extractor_key'} | |
1841 | ||
412c617d | 1842 | new_result = info.copy() |
3975b4d2 | 1843 | new_result.update(filter_dict(ie_result, lambda k, v: v is not None and k not in exempted_fields)) |
7fc3fa05 | 1844 | |
0563f7ac S |
1845 | # Extracted info may not be a video result (i.e. |
1846 | # info.get('_type', 'video') != video) but rather an url or | |
1847 | # url_transparent. In such cases outer metadata (from ie_result) | |
1848 | # should be propagated to inner one (info). For this to happen | |
1849 | # _type of info should be overridden with url_transparent. This | |
067aa17e | 1850 | # fixes issue from https://github.com/ytdl-org/youtube-dl/pull/11163. |
0563f7ac S |
1851 | if new_result.get('_type') == 'url': |
1852 | new_result['_type'] = 'url_transparent' | |
7fc3fa05 PH |
1853 | |
1854 | return self.process_ie_result( | |
1855 | new_result, download=download, extra_info=extra_info) | |
40fcba5e | 1856 | elif result_type in ('playlist', 'multi_video'): |
30a074c2 | 1857 | # Protect from infinite recursion due to recursively nested playlists |
1858 | # (see https://github.com/ytdl-org/youtube-dl/issues/27833) | |
0bd5a039 | 1859 | webpage_url = ie_result.get('webpage_url') # Playlists maynot have webpage_url |
1860 | if webpage_url and webpage_url in self._playlist_urls: | |
7e85e872 | 1861 | self.to_screen( |
30a074c2 | 1862 | '[download] Skipping already downloaded playlist: %s' |
1863 | % ie_result.get('title') or ie_result.get('id')) | |
1864 | return | |
7e85e872 | 1865 | |
30a074c2 | 1866 | self._playlist_level += 1 |
1867 | self._playlist_urls.add(webpage_url) | |
03f83004 | 1868 | self._fill_common_fields(ie_result, False) |
bc516a3f | 1869 | self._sanitize_thumbnails(ie_result) |
30a074c2 | 1870 | try: |
1871 | return self.__process_playlist(ie_result, download) | |
1872 | finally: | |
1873 | self._playlist_level -= 1 | |
1874 | if not self._playlist_level: | |
1875 | self._playlist_urls.clear() | |
8222d8de | 1876 | elif result_type == 'compat_list': |
c9bf4114 PH |
1877 | self.report_warning( |
1878 | 'Extractor %s returned a compat_list result. ' | |
1879 | 'It needs to be updated.' % ie_result.get('extractor')) | |
5f6a1245 | 1880 | |
8222d8de | 1881 | def _fixup(r): |
b868936c | 1882 | self.add_extra_info(r, { |
1883 | 'extractor': ie_result['extractor'], | |
1884 | 'webpage_url': ie_result['webpage_url'], | |
1885 | 'webpage_url_basename': url_basename(ie_result['webpage_url']), | |
0bb322b9 | 1886 | 'webpage_url_domain': get_domain(ie_result['webpage_url']), |
b868936c | 1887 | 'extractor_key': ie_result['extractor_key'], |
1888 | }) | |
8222d8de JMF |
1889 | return r |
1890 | ie_result['entries'] = [ | |
b6c45014 | 1891 | self.process_ie_result(_fixup(r), download, extra_info) |
8222d8de JMF |
1892 | for r in ie_result['entries'] |
1893 | ] | |
1894 | return ie_result | |
1895 | else: | |
1896 | raise Exception('Invalid result type: %s' % result_type) | |
1897 | ||
e92caff5 | 1898 | def _ensure_dir_exists(self, path): |
1899 | return make_dir(path, self.report_error) | |
1900 | ||
3b603dbd | 1901 | @staticmethod |
3bec830a | 1902 | def _playlist_infodict(ie_result, strict=False, **kwargs): |
1903 | info = { | |
1904 | 'playlist_count': ie_result.get('playlist_count'), | |
3b603dbd | 1905 | 'playlist': ie_result.get('title') or ie_result.get('id'), |
1906 | 'playlist_id': ie_result.get('id'), | |
1907 | 'playlist_title': ie_result.get('title'), | |
1908 | 'playlist_uploader': ie_result.get('uploader'), | |
1909 | 'playlist_uploader_id': ie_result.get('uploader_id'), | |
3b603dbd | 1910 | **kwargs, |
1911 | } | |
3bec830a | 1912 | if strict: |
1913 | return info | |
0bd5a039 | 1914 | if ie_result.get('webpage_url'): |
1915 | info.update({ | |
1916 | 'webpage_url': ie_result['webpage_url'], | |
1917 | 'webpage_url_basename': url_basename(ie_result['webpage_url']), | |
1918 | 'webpage_url_domain': get_domain(ie_result['webpage_url']), | |
1919 | }) | |
3bec830a | 1920 | return { |
1921 | **info, | |
1922 | 'playlist_index': 0, | |
59d7de0d | 1923 | '__last_playlist_index': max(ie_result.get('requested_entries') or (0, 0)), |
3bec830a | 1924 | 'extractor': ie_result['extractor'], |
3bec830a | 1925 | 'extractor_key': ie_result['extractor_key'], |
1926 | } | |
3b603dbd | 1927 | |
30a074c2 | 1928 | def __process_playlist(self, ie_result, download): |
7e88d7d7 | 1929 | """Process each entry in the playlist""" |
f5ea4748 | 1930 | assert ie_result['_type'] in ('playlist', 'multi_video') |
1931 | ||
3bec830a | 1932 | common_info = self._playlist_infodict(ie_result, strict=True) |
3955b207 | 1933 | title = common_info.get('playlist') or '<Untitled>' |
3bec830a | 1934 | if self._match_entry(common_info, incomplete=True) is not None: |
1935 | return | |
c6e07cf1 | 1936 | self.to_screen(f'[download] Downloading {ie_result["_type"]}: {title}') |
f0d785d3 | 1937 | |
7e88d7d7 | 1938 | all_entries = PlaylistEntries(self, ie_result) |
7e9a6125 | 1939 | entries = orderedSet(all_entries.get_requested_items(), lazy=True) |
1940 | ||
1941 | lazy = self.params.get('lazy_playlist') | |
1942 | if lazy: | |
1943 | resolved_entries, n_entries = [], 'N/A' | |
1944 | ie_result['requested_entries'], ie_result['entries'] = None, None | |
1945 | else: | |
1946 | entries = resolved_entries = list(entries) | |
1947 | n_entries = len(resolved_entries) | |
1948 | ie_result['requested_entries'], ie_result['entries'] = tuple(zip(*resolved_entries)) or ([], []) | |
1949 | if not ie_result.get('playlist_count'): | |
1950 | # Better to do this after potentially exhausting entries | |
1951 | ie_result['playlist_count'] = all_entries.get_full_count() | |
498f5606 | 1952 | |
0647d925 | 1953 | extra = self._playlist_infodict(ie_result, n_entries=int_or_none(n_entries)) |
1954 | ie_copy = collections.ChainMap(ie_result, extra) | |
3bec830a | 1955 | |
e08a85d8 | 1956 | _infojson_written = False |
0bfc53d0 | 1957 | write_playlist_files = self.params.get('allow_playlist_files', True) |
1958 | if write_playlist_files and self.params.get('list_thumbnails'): | |
1959 | self.list_thumbnails(ie_result) | |
1960 | if write_playlist_files and not self.params.get('simulate'): | |
e08a85d8 | 1961 | _infojson_written = self._write_info_json( |
1962 | 'playlist', ie_result, self.prepare_filename(ie_copy, 'pl_infojson')) | |
1963 | if _infojson_written is None: | |
80c03fa9 | 1964 | return |
1965 | if self._write_description('playlist', ie_result, | |
1966 | self.prepare_filename(ie_copy, 'pl_description')) is None: | |
1967 | return | |
681de68e | 1968 | # TODO: This should be passed to ThumbnailsConvertor if necessary |
3bec830a | 1969 | self._write_thumbnails('playlist', ie_result, self.prepare_filename(ie_copy, 'pl_thumbnail')) |
30a074c2 | 1970 | |
7e9a6125 | 1971 | if lazy: |
1972 | if self.params.get('playlistreverse') or self.params.get('playlistrandom'): | |
1973 | self.report_warning('playlistreverse and playlistrandom are not supported with lazy_playlist', only_once=True) | |
1974 | elif self.params.get('playlistreverse'): | |
1975 | entries.reverse() | |
1976 | elif self.params.get('playlistrandom'): | |
30a074c2 | 1977 | random.shuffle(entries) |
1978 | ||
bc5c2f8a | 1979 | self.to_screen(f'[{ie_result["extractor"]}] Playlist {title}: Downloading {n_entries} items' |
7e88d7d7 | 1980 | f'{format_field(ie_result, "playlist_count", " of %s")}') |
30a074c2 | 1981 | |
134c913c | 1982 | keep_resolved_entries = self.params.get('extract_flat') != 'discard' |
1983 | if self.params.get('extract_flat') == 'discard_in_playlist': | |
1984 | keep_resolved_entries = ie_result['_type'] != 'playlist' | |
1985 | if keep_resolved_entries: | |
1986 | self.write_debug('The information of all playlist entries will be held in memory') | |
1987 | ||
26e2805c | 1988 | failures = 0 |
1989 | max_failures = self.params.get('skip_playlist_after_errors') or float('inf') | |
7e9a6125 | 1990 | for i, (playlist_index, entry) in enumerate(entries): |
1991 | if lazy: | |
1992 | resolved_entries.append((playlist_index, entry)) | |
3bec830a | 1993 | if not entry: |
7e88d7d7 | 1994 | continue |
1995 | ||
7e88d7d7 | 1996 | entry['__x_forwarded_for_ip'] = ie_result.get('__x_forwarded_for_ip') |
ad54c913 | 1997 | if not lazy and 'playlist-index' in self.params['compat_opts']: |
7e9a6125 | 1998 | playlist_index = ie_result['requested_entries'][i] |
1999 | ||
0647d925 | 2000 | entry_copy = collections.ChainMap(entry, { |
3bec830a | 2001 | **common_info, |
3955b207 | 2002 | 'n_entries': int_or_none(n_entries), |
71729754 | 2003 | 'playlist_index': playlist_index, |
7e9a6125 | 2004 | 'playlist_autonumber': i + 1, |
0647d925 | 2005 | }) |
3bec830a | 2006 | |
0647d925 | 2007 | if self._match_entry(entry_copy, incomplete=True) is not None: |
f0ad6f8c | 2008 | # For compatabilty with youtube-dl. See https://github.com/yt-dlp/yt-dlp/issues/4369 |
2009 | resolved_entries[i] = (playlist_index, NO_DEFAULT) | |
3bec830a | 2010 | continue |
2011 | ||
bc5c2f8a | 2012 | self.to_screen('[download] Downloading item %s of %s' % ( |
3bec830a | 2013 | self._format_screen(i + 1, self.Styles.ID), self._format_screen(n_entries, self.Styles.EMPHASIS))) |
2014 | ||
ec54bd43 | 2015 | entry_result = self.__process_iterable_entry(entry, download, collections.ChainMap({ |
a6ca61d4 | 2016 | 'playlist_index': playlist_index, |
2017 | 'playlist_autonumber': i + 1, | |
ec54bd43 | 2018 | }, extra)) |
26e2805c | 2019 | if not entry_result: |
2020 | failures += 1 | |
2021 | if failures >= max_failures: | |
2022 | self.report_error( | |
7e88d7d7 | 2023 | f'Skipping the remaining entries in playlist "{title}" since {failures} items failed extraction') |
26e2805c | 2024 | break |
134c913c | 2025 | if keep_resolved_entries: |
2026 | resolved_entries[i] = (playlist_index, entry_result) | |
7e88d7d7 | 2027 | |
2028 | # Update with processed data | |
f0ad6f8c | 2029 | ie_result['entries'] = [e for _, e in resolved_entries if e is not NO_DEFAULT] |
bc5c2f8a | 2030 | ie_result['requested_entries'] = [i for i, e in resolved_entries if e is not NO_DEFAULT] |
2031 | if ie_result['requested_entries'] == try_call(lambda: list(range(1, ie_result['playlist_count'] + 1))): | |
2032 | # Do not set for full playlist | |
2033 | ie_result.pop('requested_entries') | |
e08a85d8 | 2034 | |
2035 | # Write the updated info to json | |
cb96c5be | 2036 | if _infojson_written is True and self._write_info_json( |
e08a85d8 | 2037 | 'updated playlist', ie_result, |
2038 | self.prepare_filename(ie_copy, 'pl_infojson'), overwrite=True) is None: | |
2039 | return | |
ca30f449 | 2040 | |
ed5835b4 | 2041 | ie_result = self.run_all_pps('playlist', ie_result) |
7e88d7d7 | 2042 | self.to_screen(f'[download] Finished downloading playlist: {title}') |
30a074c2 | 2043 | return ie_result |
2044 | ||
7e88d7d7 | 2045 | @_handle_extraction_exceptions |
a0566bbf | 2046 | def __process_iterable_entry(self, entry, download, extra_info): |
2047 | return self.process_ie_result( | |
2048 | entry, download=download, extra_info=extra_info) | |
2049 | ||
67134eab JMF |
2050 | def _build_format_filter(self, filter_spec): |
2051 | " Returns a function to filter the formats according to the filter_spec " | |
083c9df9 PH |
2052 | |
2053 | OPERATORS = { | |
2054 | '<': operator.lt, | |
2055 | '<=': operator.le, | |
2056 | '>': operator.gt, | |
2057 | '>=': operator.ge, | |
2058 | '=': operator.eq, | |
2059 | '!=': operator.ne, | |
2060 | } | |
67134eab | 2061 | operator_rex = re.compile(r'''(?x)\s* |
c3f624ef | 2062 | (?P<key>[\w.-]+)\s* |
187986a8 | 2063 | (?P<op>%s)(?P<none_inclusive>\s*\?)?\s* |
2064 | (?P<value>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?)\s* | |
083c9df9 | 2065 | ''' % '|'.join(map(re.escape, OPERATORS.keys()))) |
187986a8 | 2066 | m = operator_rex.fullmatch(filter_spec) |
9ddb6925 S |
2067 | if m: |
2068 | try: | |
2069 | comparison_value = int(m.group('value')) | |
2070 | except ValueError: | |
2071 | comparison_value = parse_filesize(m.group('value')) | |
2072 | if comparison_value is None: | |
2073 | comparison_value = parse_filesize(m.group('value') + 'B') | |
2074 | if comparison_value is None: | |
2075 | raise ValueError( | |
2076 | 'Invalid value %r in format specification %r' % ( | |
67134eab | 2077 | m.group('value'), filter_spec)) |
9ddb6925 S |
2078 | op = OPERATORS[m.group('op')] |
2079 | ||
083c9df9 | 2080 | if not m: |
9ddb6925 S |
2081 | STR_OPERATORS = { |
2082 | '=': operator.eq, | |
10d33b34 YCH |
2083 | '^=': lambda attr, value: attr.startswith(value), |
2084 | '$=': lambda attr, value: attr.endswith(value), | |
2085 | '*=': lambda attr, value: value in attr, | |
1ce9a3cb | 2086 | '~=': lambda attr, value: value.search(attr) is not None |
9ddb6925 | 2087 | } |
187986a8 | 2088 | str_operator_rex = re.compile(r'''(?x)\s* |
2089 | (?P<key>[a-zA-Z0-9._-]+)\s* | |
1ce9a3cb LF |
2090 | (?P<negation>!\s*)?(?P<op>%s)\s*(?P<none_inclusive>\?\s*)? |
2091 | (?P<quote>["'])? | |
2092 | (?P<value>(?(quote)(?:(?!(?P=quote))[^\\]|\\.)+|[\w.-]+)) | |
2093 | (?(quote)(?P=quote))\s* | |
9ddb6925 | 2094 | ''' % '|'.join(map(re.escape, STR_OPERATORS.keys()))) |
187986a8 | 2095 | m = str_operator_rex.fullmatch(filter_spec) |
9ddb6925 | 2096 | if m: |
1ce9a3cb LF |
2097 | if m.group('op') == '~=': |
2098 | comparison_value = re.compile(m.group('value')) | |
2099 | else: | |
2100 | comparison_value = re.sub(r'''\\([\\"'])''', r'\1', m.group('value')) | |
2cc779f4 S |
2101 | str_op = STR_OPERATORS[m.group('op')] |
2102 | if m.group('negation'): | |
e118a879 | 2103 | op = lambda attr, value: not str_op(attr, value) |
2cc779f4 S |
2104 | else: |
2105 | op = str_op | |
083c9df9 | 2106 | |
9ddb6925 | 2107 | if not m: |
187986a8 | 2108 | raise SyntaxError('Invalid filter specification %r' % filter_spec) |
083c9df9 PH |
2109 | |
2110 | def _filter(f): | |
2111 | actual_value = f.get(m.group('key')) | |
2112 | if actual_value is None: | |
2113 | return m.group('none_inclusive') | |
2114 | return op(actual_value, comparison_value) | |
67134eab JMF |
2115 | return _filter |
2116 | ||
9f1a1c36 | 2117 | def _check_formats(self, formats): |
2118 | for f in formats: | |
2119 | self.to_screen('[info] Testing format %s' % f['format_id']) | |
75689fe5 | 2120 | path = self.get_output_path('temp') |
2121 | if not self._ensure_dir_exists(f'{path}/'): | |
2122 | continue | |
2123 | temp_file = tempfile.NamedTemporaryFile(suffix='.tmp', delete=False, dir=path or None) | |
9f1a1c36 | 2124 | temp_file.close() |
2125 | try: | |
2126 | success, _ = self.dl(temp_file.name, f, test=True) | |
8a82af35 | 2127 | except (DownloadError, OSError, ValueError) + network_exceptions: |
9f1a1c36 | 2128 | success = False |
2129 | finally: | |
2130 | if os.path.exists(temp_file.name): | |
2131 | try: | |
2132 | os.remove(temp_file.name) | |
2133 | except OSError: | |
2134 | self.report_warning('Unable to delete temporary file "%s"' % temp_file.name) | |
2135 | if success: | |
2136 | yield f | |
2137 | else: | |
2138 | self.to_screen('[info] Unable to download format %s. Skipping...' % f['format_id']) | |
2139 | ||
0017d9ad | 2140 | def _default_format_spec(self, info_dict, download=True): |
0017d9ad | 2141 | |
af0f7428 S |
2142 | def can_merge(): |
2143 | merger = FFmpegMergerPP(self) | |
2144 | return merger.available and merger.can_merge() | |
2145 | ||
91ebc640 | 2146 | prefer_best = ( |
b7b04c78 | 2147 | not self.params.get('simulate') |
91ebc640 | 2148 | and download |
2149 | and ( | |
2150 | not can_merge() | |
21633673 | 2151 | or info_dict.get('is_live') and not self.params.get('live_from_start') |
bf1824b3 | 2152 | or self.params['outtmpl']['default'] == '-')) |
53ed7066 | 2153 | compat = ( |
2154 | prefer_best | |
2155 | or self.params.get('allow_multiple_audio_streams', False) | |
8a82af35 | 2156 | or 'format-spec' in self.params['compat_opts']) |
91ebc640 | 2157 | |
2158 | return ( | |
53ed7066 | 2159 | 'best/bestvideo+bestaudio' if prefer_best |
2160 | else 'bestvideo*+bestaudio/best' if not compat | |
91ebc640 | 2161 | else 'bestvideo+bestaudio/best') |
0017d9ad | 2162 | |
67134eab JMF |
2163 | def build_format_selector(self, format_spec): |
2164 | def syntax_error(note, start): | |
2165 | message = ( | |
2166 | 'Invalid format specification: ' | |
86e5f3ed | 2167 | '{}\n\t{}\n\t{}^'.format(note, format_spec, ' ' * start[1])) |
67134eab JMF |
2168 | return SyntaxError(message) |
2169 | ||
2170 | PICKFIRST = 'PICKFIRST' | |
2171 | MERGE = 'MERGE' | |
2172 | SINGLE = 'SINGLE' | |
0130afb7 | 2173 | GROUP = 'GROUP' |
67134eab JMF |
2174 | FormatSelector = collections.namedtuple('FormatSelector', ['type', 'selector', 'filters']) |
2175 | ||
91ebc640 | 2176 | allow_multiple_streams = {'audio': self.params.get('allow_multiple_audio_streams', False), |
2177 | 'video': self.params.get('allow_multiple_video_streams', False)} | |
909d24dd | 2178 | |
67134eab JMF |
2179 | def _parse_filter(tokens): |
2180 | filter_parts = [] | |
6f2287cb | 2181 | for type, string_, start, _, _ in tokens: |
2182 | if type == tokenize.OP and string_ == ']': | |
67134eab JMF |
2183 | return ''.join(filter_parts) |
2184 | else: | |
6f2287cb | 2185 | filter_parts.append(string_) |
67134eab | 2186 | |
232541df | 2187 | def _remove_unused_ops(tokens): |
62b58c09 L |
2188 | # Remove operators that we don't use and join them with the surrounding strings. |
2189 | # E.g. 'mp4' '-' 'baseline' '-' '16x9' is converted to 'mp4-baseline-16x9' | |
232541df JMF |
2190 | ALLOWED_OPS = ('/', '+', ',', '(', ')') |
2191 | last_string, last_start, last_end, last_line = None, None, None, None | |
6f2287cb | 2192 | for type, string_, start, end, line in tokens: |
2193 | if type == tokenize.OP and string_ == '[': | |
232541df JMF |
2194 | if last_string: |
2195 | yield tokenize.NAME, last_string, last_start, last_end, last_line | |
2196 | last_string = None | |
6f2287cb | 2197 | yield type, string_, start, end, line |
232541df | 2198 | # everything inside brackets will be handled by _parse_filter |
6f2287cb | 2199 | for type, string_, start, end, line in tokens: |
2200 | yield type, string_, start, end, line | |
2201 | if type == tokenize.OP and string_ == ']': | |
232541df | 2202 | break |
6f2287cb | 2203 | elif type == tokenize.OP and string_ in ALLOWED_OPS: |
232541df JMF |
2204 | if last_string: |
2205 | yield tokenize.NAME, last_string, last_start, last_end, last_line | |
2206 | last_string = None | |
6f2287cb | 2207 | yield type, string_, start, end, line |
232541df JMF |
2208 | elif type in [tokenize.NAME, tokenize.NUMBER, tokenize.OP]: |
2209 | if not last_string: | |
6f2287cb | 2210 | last_string = string_ |
232541df JMF |
2211 | last_start = start |
2212 | last_end = end | |
2213 | else: | |
6f2287cb | 2214 | last_string += string_ |
232541df JMF |
2215 | if last_string: |
2216 | yield tokenize.NAME, last_string, last_start, last_end, last_line | |
2217 | ||
cf2ac6df | 2218 | def _parse_format_selection(tokens, inside_merge=False, inside_choice=False, inside_group=False): |
67134eab JMF |
2219 | selectors = [] |
2220 | current_selector = None | |
6f2287cb | 2221 | for type, string_, start, _, _ in tokens: |
67134eab JMF |
2222 | # ENCODING is only defined in python 3.x |
2223 | if type == getattr(tokenize, 'ENCODING', None): | |
2224 | continue | |
2225 | elif type in [tokenize.NAME, tokenize.NUMBER]: | |
6f2287cb | 2226 | current_selector = FormatSelector(SINGLE, string_, []) |
67134eab | 2227 | elif type == tokenize.OP: |
6f2287cb | 2228 | if string_ == ')': |
cf2ac6df JMF |
2229 | if not inside_group: |
2230 | # ')' will be handled by the parentheses group | |
2231 | tokens.restore_last_token() | |
67134eab | 2232 | break |
6f2287cb | 2233 | elif inside_merge and string_ in ['/', ',']: |
0130afb7 JMF |
2234 | tokens.restore_last_token() |
2235 | break | |
6f2287cb | 2236 | elif inside_choice and string_ == ',': |
cf2ac6df JMF |
2237 | tokens.restore_last_token() |
2238 | break | |
6f2287cb | 2239 | elif string_ == ',': |
0a31a350 JMF |
2240 | if not current_selector: |
2241 | raise syntax_error('"," must follow a format selector', start) | |
67134eab JMF |
2242 | selectors.append(current_selector) |
2243 | current_selector = None | |
6f2287cb | 2244 | elif string_ == '/': |
d96d604e JMF |
2245 | if not current_selector: |
2246 | raise syntax_error('"/" must follow a format selector', start) | |
67134eab | 2247 | first_choice = current_selector |
cf2ac6df | 2248 | second_choice = _parse_format_selection(tokens, inside_choice=True) |
f5f4a27a | 2249 | current_selector = FormatSelector(PICKFIRST, (first_choice, second_choice), []) |
6f2287cb | 2250 | elif string_ == '[': |
67134eab JMF |
2251 | if not current_selector: |
2252 | current_selector = FormatSelector(SINGLE, 'best', []) | |
2253 | format_filter = _parse_filter(tokens) | |
2254 | current_selector.filters.append(format_filter) | |
6f2287cb | 2255 | elif string_ == '(': |
0130afb7 JMF |
2256 | if current_selector: |
2257 | raise syntax_error('Unexpected "("', start) | |
cf2ac6df JMF |
2258 | group = _parse_format_selection(tokens, inside_group=True) |
2259 | current_selector = FormatSelector(GROUP, group, []) | |
6f2287cb | 2260 | elif string_ == '+': |
d03cfdce | 2261 | if not current_selector: |
2262 | raise syntax_error('Unexpected "+"', start) | |
2263 | selector_1 = current_selector | |
2264 | selector_2 = _parse_format_selection(tokens, inside_merge=True) | |
2265 | if not selector_2: | |
2266 | raise syntax_error('Expected a selector', start) | |
2267 | current_selector = FormatSelector(MERGE, (selector_1, selector_2), []) | |
67134eab | 2268 | else: |
6f2287cb | 2269 | raise syntax_error(f'Operator not recognized: "{string_}"', start) |
67134eab JMF |
2270 | elif type == tokenize.ENDMARKER: |
2271 | break | |
2272 | if current_selector: | |
2273 | selectors.append(current_selector) | |
2274 | return selectors | |
2275 | ||
f8d4ad9a | 2276 | def _merge(formats_pair): |
2277 | format_1, format_2 = formats_pair | |
2278 | ||
2279 | formats_info = [] | |
2280 | formats_info.extend(format_1.get('requested_formats', (format_1,))) | |
2281 | formats_info.extend(format_2.get('requested_formats', (format_2,))) | |
2282 | ||
2283 | if not allow_multiple_streams['video'] or not allow_multiple_streams['audio']: | |
551f9388 | 2284 | get_no_more = {'video': False, 'audio': False} |
f8d4ad9a | 2285 | for (i, fmt_info) in enumerate(formats_info): |
551f9388 | 2286 | if fmt_info.get('acodec') == fmt_info.get('vcodec') == 'none': |
2287 | formats_info.pop(i) | |
2288 | continue | |
2289 | for aud_vid in ['audio', 'video']: | |
f8d4ad9a | 2290 | if not allow_multiple_streams[aud_vid] and fmt_info.get(aud_vid[0] + 'codec') != 'none': |
2291 | if get_no_more[aud_vid]: | |
2292 | formats_info.pop(i) | |
f5510afe | 2293 | break |
f8d4ad9a | 2294 | get_no_more[aud_vid] = True |
2295 | ||
2296 | if len(formats_info) == 1: | |
2297 | return formats_info[0] | |
2298 | ||
2299 | video_fmts = [fmt_info for fmt_info in formats_info if fmt_info.get('vcodec') != 'none'] | |
2300 | audio_fmts = [fmt_info for fmt_info in formats_info if fmt_info.get('acodec') != 'none'] | |
2301 | ||
2302 | the_only_video = video_fmts[0] if len(video_fmts) == 1 else None | |
2303 | the_only_audio = audio_fmts[0] if len(audio_fmts) == 1 else None | |
2304 | ||
fc61aff4 LL |
2305 | output_ext = get_compatible_ext( |
2306 | vcodecs=[f.get('vcodec') for f in video_fmts], | |
2307 | acodecs=[f.get('acodec') for f in audio_fmts], | |
2308 | vexts=[f['ext'] for f in video_fmts], | |
2309 | aexts=[f['ext'] for f in audio_fmts], | |
2310 | preferences=(try_call(lambda: self.params['merge_output_format'].split('/')) | |
2311 | or self.params.get('prefer_free_formats') and ('webm', 'mkv'))) | |
f8d4ad9a | 2312 | |
975a0d0d | 2313 | filtered = lambda *keys: filter(None, (traverse_obj(fmt, *keys) for fmt in formats_info)) |
2314 | ||
f8d4ad9a | 2315 | new_dict = { |
2316 | 'requested_formats': formats_info, | |
975a0d0d | 2317 | 'format': '+'.join(filtered('format')), |
2318 | 'format_id': '+'.join(filtered('format_id')), | |
f8d4ad9a | 2319 | 'ext': output_ext, |
975a0d0d | 2320 | 'protocol': '+'.join(map(determine_protocol, formats_info)), |
093a1710 | 2321 | 'language': '+'.join(orderedSet(filtered('language'))) or None, |
2322 | 'format_note': '+'.join(orderedSet(filtered('format_note'))) or None, | |
2323 | 'filesize_approx': sum(filtered('filesize', 'filesize_approx')) or None, | |
975a0d0d | 2324 | 'tbr': sum(filtered('tbr', 'vbr', 'abr')), |
f8d4ad9a | 2325 | } |
2326 | ||
2327 | if the_only_video: | |
2328 | new_dict.update({ | |
2329 | 'width': the_only_video.get('width'), | |
2330 | 'height': the_only_video.get('height'), | |
2331 | 'resolution': the_only_video.get('resolution') or self.format_resolution(the_only_video), | |
2332 | 'fps': the_only_video.get('fps'), | |
49a57e70 | 2333 | 'dynamic_range': the_only_video.get('dynamic_range'), |
f8d4ad9a | 2334 | 'vcodec': the_only_video.get('vcodec'), |
2335 | 'vbr': the_only_video.get('vbr'), | |
2336 | 'stretched_ratio': the_only_video.get('stretched_ratio'), | |
105bfd90 | 2337 | 'aspect_ratio': the_only_video.get('aspect_ratio'), |
f8d4ad9a | 2338 | }) |
2339 | ||
2340 | if the_only_audio: | |
2341 | new_dict.update({ | |
2342 | 'acodec': the_only_audio.get('acodec'), | |
2343 | 'abr': the_only_audio.get('abr'), | |
975a0d0d | 2344 | 'asr': the_only_audio.get('asr'), |
b8ed0f15 | 2345 | 'audio_channels': the_only_audio.get('audio_channels') |
f8d4ad9a | 2346 | }) |
2347 | ||
2348 | return new_dict | |
2349 | ||
e8e73840 | 2350 | def _check_formats(formats): |
8cb7fc44 | 2351 | if self.params.get('check_formats') == 'selected': |
2352 | yield from self._check_formats(formats) | |
2353 | return | |
2354 | elif (self.params.get('check_formats') is not None | |
bc344cd4 | 2355 | or self.params.get('allow_unplayable_formats')): |
981052c9 | 2356 | yield from formats |
b5ac45b1 | 2357 | return |
bc344cd4 | 2358 | |
2359 | for f in formats: | |
ef79d20d | 2360 | if f.get('has_drm') or f.get('__needs_testing'): |
bc344cd4 | 2361 | yield from self._check_formats([f]) |
2362 | else: | |
2363 | yield f | |
e8e73840 | 2364 | |
67134eab | 2365 | def _build_selector_function(selector): |
909d24dd | 2366 | if isinstance(selector, list): # , |
67134eab JMF |
2367 | fs = [_build_selector_function(s) for s in selector] |
2368 | ||
317f7ab6 | 2369 | def selector_function(ctx): |
67134eab | 2370 | for f in fs: |
981052c9 | 2371 | yield from f(ctx) |
67134eab | 2372 | return selector_function |
909d24dd | 2373 | |
2374 | elif selector.type == GROUP: # () | |
0130afb7 | 2375 | selector_function = _build_selector_function(selector.selector) |
909d24dd | 2376 | |
2377 | elif selector.type == PICKFIRST: # / | |
67134eab JMF |
2378 | fs = [_build_selector_function(s) for s in selector.selector] |
2379 | ||
317f7ab6 | 2380 | def selector_function(ctx): |
67134eab | 2381 | for f in fs: |
317f7ab6 | 2382 | picked_formats = list(f(ctx)) |
67134eab JMF |
2383 | if picked_formats: |
2384 | return picked_formats | |
2385 | return [] | |
67134eab | 2386 | |
981052c9 | 2387 | elif selector.type == MERGE: # + |
2388 | selector_1, selector_2 = map(_build_selector_function, selector.selector) | |
2389 | ||
2390 | def selector_function(ctx): | |
adbc4ec4 | 2391 | for pair in itertools.product(selector_1(ctx), selector_2(ctx)): |
981052c9 | 2392 | yield _merge(pair) |
2393 | ||
909d24dd | 2394 | elif selector.type == SINGLE: # atom |
598d185d | 2395 | format_spec = selector.selector or 'best' |
909d24dd | 2396 | |
f8d4ad9a | 2397 | # TODO: Add allvideo, allaudio etc by generalizing the code with best/worst selector |
909d24dd | 2398 | if format_spec == 'all': |
2399 | def selector_function(ctx): | |
9222c381 | 2400 | yield from _check_formats(ctx['formats'][::-1]) |
f8d4ad9a | 2401 | elif format_spec == 'mergeall': |
2402 | def selector_function(ctx): | |
316f2650 | 2403 | formats = list(_check_formats( |
2404 | f for f in ctx['formats'] if f.get('vcodec') != 'none' or f.get('acodec') != 'none')) | |
e01d6aa4 | 2405 | if not formats: |
2406 | return | |
921b76ca | 2407 | merged_format = formats[-1] |
2408 | for f in formats[-2::-1]: | |
f8d4ad9a | 2409 | merged_format = _merge((merged_format, f)) |
2410 | yield merged_format | |
909d24dd | 2411 | |
2412 | else: | |
85e801a9 | 2413 | format_fallback, seperate_fallback, format_reverse, format_idx = False, None, True, 1 |
eff63539 | 2414 | mobj = re.match( |
2415 | r'(?P<bw>best|worst|b|w)(?P<type>video|audio|v|a)?(?P<mod>\*)?(?:\.(?P<n>[1-9]\d*))?$', | |
2416 | format_spec) | |
2417 | if mobj is not None: | |
2418 | format_idx = int_or_none(mobj.group('n'), default=1) | |
e8e73840 | 2419 | format_reverse = mobj.group('bw')[0] == 'b' |
eff63539 | 2420 | format_type = (mobj.group('type') or [None])[0] |
2421 | not_format_type = {'v': 'a', 'a': 'v'}.get(format_type) | |
2422 | format_modified = mobj.group('mod') is not None | |
909d24dd | 2423 | |
2424 | format_fallback = not format_type and not format_modified # for b, w | |
8326b00a | 2425 | _filter_f = ( |
eff63539 | 2426 | (lambda f: f.get('%scodec' % format_type) != 'none') |
2427 | if format_type and format_modified # bv*, ba*, wv*, wa* | |
2428 | else (lambda f: f.get('%scodec' % not_format_type) == 'none') | |
2429 | if format_type # bv, ba, wv, wa | |
2430 | else (lambda f: f.get('vcodec') != 'none' and f.get('acodec') != 'none') | |
2431 | if not format_modified # b, w | |
8326b00a | 2432 | else lambda f: True) # b*, w* |
2433 | filter_f = lambda f: _filter_f(f) and ( | |
2434 | f.get('vcodec') != 'none' or f.get('acodec') != 'none') | |
67134eab | 2435 | else: |
48ee10ee | 2436 | if format_spec in self._format_selection_exts['audio']: |
b11c04a8 | 2437 | filter_f = lambda f: f.get('ext') == format_spec and f.get('acodec') != 'none' |
48ee10ee | 2438 | elif format_spec in self._format_selection_exts['video']: |
b11c04a8 | 2439 | filter_f = lambda f: f.get('ext') == format_spec and f.get('acodec') != 'none' and f.get('vcodec') != 'none' |
85e801a9 | 2440 | seperate_fallback = lambda f: f.get('ext') == format_spec and f.get('vcodec') != 'none' |
48ee10ee | 2441 | elif format_spec in self._format_selection_exts['storyboards']: |
b11c04a8 | 2442 | filter_f = lambda f: f.get('ext') == format_spec and f.get('acodec') == 'none' and f.get('vcodec') == 'none' |
2443 | else: | |
b5ae35ee | 2444 | filter_f = lambda f: f.get('format_id') == format_spec # id |
909d24dd | 2445 | |
2446 | def selector_function(ctx): | |
2447 | formats = list(ctx['formats']) | |
909d24dd | 2448 | matches = list(filter(filter_f, formats)) if filter_f is not None else formats |
85e801a9 | 2449 | if not matches: |
2450 | if format_fallback and ctx['incomplete_formats']: | |
2451 | # for extractors with incomplete formats (audio only (soundcloud) | |
2452 | # or video only (imgur)) best/worst will fallback to | |
2453 | # best/worst {video,audio}-only format | |
d63eae7e | 2454 | matches = list(filter(lambda f: f.get('vcodec') != 'none' or f.get('acodec') != 'none', formats)) |
85e801a9 | 2455 | elif seperate_fallback and not ctx['has_merged_format']: |
2456 | # for compatibility with youtube-dl when there is no pre-merged format | |
2457 | matches = list(filter(seperate_fallback, formats)) | |
981052c9 | 2458 | matches = LazyList(_check_formats(matches[::-1 if format_reverse else 1])) |
2459 | try: | |
e8e73840 | 2460 | yield matches[format_idx - 1] |
4abea8ca | 2461 | except LazyList.IndexError: |
981052c9 | 2462 | return |
083c9df9 | 2463 | |
67134eab | 2464 | filters = [self._build_format_filter(f) for f in selector.filters] |
083c9df9 | 2465 | |
317f7ab6 | 2466 | def final_selector(ctx): |
adbc4ec4 | 2467 | ctx_copy = dict(ctx) |
67134eab | 2468 | for _filter in filters: |
317f7ab6 S |
2469 | ctx_copy['formats'] = list(filter(_filter, ctx_copy['formats'])) |
2470 | return selector_function(ctx_copy) | |
67134eab | 2471 | return final_selector |
083c9df9 | 2472 | |
00cdda4f SS |
2473 | # HACK: Python 3.12 changed the underlying parser, rendering '7_a' invalid |
2474 | # Prefix numbers with random letters to avoid it being classified as a number | |
2475 | # See: https://github.com/yt-dlp/yt-dlp/pulls/8797 | |
2476 | # TODO: Implement parser not reliant on tokenize.tokenize | |
2477 | prefix = ''.join(random.choices(string.ascii_letters, k=32)) | |
2478 | stream = io.BytesIO(re.sub(r'\d[_\d]*', rf'{prefix}\g<0>', format_spec).encode()) | |
0130afb7 | 2479 | try: |
00cdda4f SS |
2480 | tokens = list(_remove_unused_ops( |
2481 | token._replace(string=token.string.replace(prefix, '')) | |
2482 | for token in tokenize.tokenize(stream.readline))) | |
0130afb7 JMF |
2483 | except tokenize.TokenError: |
2484 | raise syntax_error('Missing closing/opening brackets or parenthesis', (0, len(format_spec))) | |
2485 | ||
86e5f3ed | 2486 | class TokenIterator: |
0130afb7 JMF |
2487 | def __init__(self, tokens): |
2488 | self.tokens = tokens | |
2489 | self.counter = 0 | |
2490 | ||
2491 | def __iter__(self): | |
2492 | return self | |
2493 | ||
2494 | def __next__(self): | |
2495 | if self.counter >= len(self.tokens): | |
2496 | raise StopIteration() | |
2497 | value = self.tokens[self.counter] | |
2498 | self.counter += 1 | |
2499 | return value | |
2500 | ||
2501 | next = __next__ | |
2502 | ||
2503 | def restore_last_token(self): | |
2504 | self.counter -= 1 | |
2505 | ||
2506 | parsed_selector = _parse_format_selection(iter(TokenIterator(tokens))) | |
67134eab | 2507 | return _build_selector_function(parsed_selector) |
a9c58ad9 | 2508 | |
6c5211ce | 2509 | def _calc_headers(self, info_dict, load_cookies=False): |
227bf1a3 | 2510 | res = HTTPHeaderDict(self.params['http_headers'], info_dict.get('http_headers')) |
c365dba8 | 2511 | clean_headers(res) |
6c5211ce | 2512 | |
2513 | if load_cookies: # For --load-info-json | |
2514 | self._load_cookies(res.get('Cookie'), autoscope=info_dict['url']) # compat | |
2515 | self._load_cookies(info_dict.get('cookies'), autoscope=False) | |
2516 | # The `Cookie` header is removed to prevent leaks and unscoped cookies. | |
2517 | # See: https://github.com/yt-dlp/yt-dlp/security/advisories/GHSA-v8mc-9377-rwjj | |
2518 | res.pop('Cookie', None) | |
31215122 | 2519 | cookies = self.cookiejar.get_cookies_for_url(info_dict['url']) |
e5660ee6 | 2520 | if cookies: |
31215122 SS |
2521 | encoder = LenientSimpleCookie() |
2522 | values = [] | |
2523 | for cookie in cookies: | |
2524 | _, value = encoder.value_encode(cookie.value) | |
2525 | values.append(f'{cookie.name}={value}') | |
2526 | if cookie.domain: | |
2527 | values.append(f'Domain={cookie.domain}') | |
2528 | if cookie.path: | |
2529 | values.append(f'Path={cookie.path}') | |
2530 | if cookie.secure: | |
2531 | values.append('Secure') | |
2532 | if cookie.expires: | |
2533 | values.append(f'Expires={cookie.expires}') | |
2534 | if cookie.version: | |
2535 | values.append(f'Version={cookie.version}') | |
2536 | info_dict['cookies'] = '; '.join(values) | |
e5660ee6 | 2537 | |
0016b84e S |
2538 | if 'X-Forwarded-For' not in res: |
2539 | x_forwarded_for_ip = info_dict.get('__x_forwarded_for_ip') | |
2540 | if x_forwarded_for_ip: | |
2541 | res['X-Forwarded-For'] = x_forwarded_for_ip | |
2542 | ||
e5660ee6 JMF |
2543 | return res |
2544 | ||
c487cf00 | 2545 | def _calc_cookies(self, url): |
b87e01c1 | 2546 | self.deprecation_warning('"YoutubeDL._calc_cookies" is deprecated and may be removed in a future version') |
2547 | return self.cookiejar.get_cookie_header(url) | |
e5660ee6 | 2548 | |
9f1a1c36 | 2549 | def _sort_thumbnails(self, thumbnails): |
2550 | thumbnails.sort(key=lambda t: ( | |
2551 | t.get('preference') if t.get('preference') is not None else -1, | |
2552 | t.get('width') if t.get('width') is not None else -1, | |
2553 | t.get('height') if t.get('height') is not None else -1, | |
2554 | t.get('id') if t.get('id') is not None else '', | |
2555 | t.get('url'))) | |
2556 | ||
b0249bca | 2557 | def _sanitize_thumbnails(self, info_dict): |
bc516a3f | 2558 | thumbnails = info_dict.get('thumbnails') |
2559 | if thumbnails is None: | |
2560 | thumbnail = info_dict.get('thumbnail') | |
2561 | if thumbnail: | |
2562 | info_dict['thumbnails'] = thumbnails = [{'url': thumbnail}] | |
9f1a1c36 | 2563 | if not thumbnails: |
2564 | return | |
2565 | ||
2566 | def check_thumbnails(thumbnails): | |
2567 | for t in thumbnails: | |
2568 | self.to_screen(f'[info] Testing thumbnail {t["id"]}') | |
2569 | try: | |
2570 | self.urlopen(HEADRequest(t['url'])) | |
2571 | except network_exceptions as err: | |
2572 | self.to_screen(f'[info] Unable to connect to thumbnail {t["id"]} URL {t["url"]!r} - {err}. Skipping...') | |
2573 | continue | |
2574 | yield t | |
2575 | ||
2576 | self._sort_thumbnails(thumbnails) | |
2577 | for i, t in enumerate(thumbnails): | |
2578 | if t.get('id') is None: | |
2579 | t['id'] = '%d' % i | |
2580 | if t.get('width') and t.get('height'): | |
2581 | t['resolution'] = '%dx%d' % (t['width'], t['height']) | |
2582 | t['url'] = sanitize_url(t['url']) | |
2583 | ||
2584 | if self.params.get('check_formats') is True: | |
282f5709 | 2585 | info_dict['thumbnails'] = LazyList(check_thumbnails(thumbnails[::-1]), reverse=True) |
9f1a1c36 | 2586 | else: |
2587 | info_dict['thumbnails'] = thumbnails | |
bc516a3f | 2588 | |
94dc8604 | 2589 | def _fill_common_fields(self, info_dict, final=True): |
03f83004 | 2590 | # TODO: move sanitization here |
94dc8604 | 2591 | if final: |
7aefd19a | 2592 | title = info_dict['fulltitle'] = info_dict.get('title') |
d4736fdb | 2593 | if not title: |
2594 | if title == '': | |
2595 | self.write_debug('Extractor gave empty title. Creating a generic title') | |
2596 | else: | |
2597 | self.report_warning('Extractor failed to obtain "title". Creating a generic title instead') | |
1d485a1a | 2598 | info_dict['title'] = f'{info_dict["extractor"].replace(":", "-")} video #{info_dict["id"]}' |
03f83004 LNO |
2599 | |
2600 | if info_dict.get('duration') is not None: | |
2601 | info_dict['duration_string'] = formatSeconds(info_dict['duration']) | |
2602 | ||
2603 | for ts_key, date_key in ( | |
2604 | ('timestamp', 'upload_date'), | |
2605 | ('release_timestamp', 'release_date'), | |
2606 | ('modified_timestamp', 'modified_date'), | |
2607 | ): | |
2608 | if info_dict.get(date_key) is None and info_dict.get(ts_key) is not None: | |
2609 | # Working around out-of-range timestamp values (e.g. negative ones on Windows, | |
2610 | # see http://bugs.python.org/issue1646728) | |
19a03940 | 2611 | with contextlib.suppress(ValueError, OverflowError, OSError): |
836e06d2 | 2612 | upload_date = datetime.datetime.fromtimestamp(info_dict[ts_key], datetime.timezone.utc) |
03f83004 | 2613 | info_dict[date_key] = upload_date.strftime('%Y%m%d') |
03f83004 | 2614 | |
1732eccc | 2615 | if not info_dict.get('release_year'): |
2616 | info_dict['release_year'] = traverse_obj(info_dict, ('release_date', {lambda x: int(x[:4])})) | |
2617 | ||
03f83004 LNO |
2618 | live_keys = ('is_live', 'was_live') |
2619 | live_status = info_dict.get('live_status') | |
2620 | if live_status is None: | |
2621 | for key in live_keys: | |
2622 | if info_dict.get(key) is False: | |
2623 | continue | |
2624 | if info_dict.get(key): | |
2625 | live_status = key | |
2626 | break | |
2627 | if all(info_dict.get(key) is False for key in live_keys): | |
2628 | live_status = 'not_live' | |
2629 | if live_status: | |
2630 | info_dict['live_status'] = live_status | |
2631 | for key in live_keys: | |
2632 | if info_dict.get(key) is None: | |
2633 | info_dict[key] = (live_status == key) | |
a057779d | 2634 | if live_status == 'post_live': |
2635 | info_dict['was_live'] = True | |
03f83004 LNO |
2636 | |
2637 | # Auto generate title fields corresponding to the *_number fields when missing | |
2638 | # in order to always have clean titles. This is very common for TV series. | |
2639 | for field in ('chapter', 'season', 'episode'): | |
94dc8604 | 2640 | if final and info_dict.get('%s_number' % field) is not None and not info_dict.get(field): |
03f83004 LNO |
2641 | info_dict[field] = '%s %d' % (field.capitalize(), info_dict['%s_number' % field]) |
2642 | ||
415f8d51 | 2643 | def _raise_pending_errors(self, info): |
2644 | err = info.pop('__pending_error', None) | |
2645 | if err: | |
2646 | self.report_error(err, tb=False) | |
2647 | ||
784320c9 | 2648 | def sort_formats(self, info_dict): |
2649 | formats = self._get_formats(info_dict) | |
784320c9 | 2650 | formats.sort(key=FormatSorter( |
c154302c | 2651 | self, info_dict.get('_format_sort_fields') or []).calculate_preference) |
784320c9 | 2652 | |
dd82ffea JMF |
2653 | def process_video_result(self, info_dict, download=True): |
2654 | assert info_dict.get('_type', 'video') == 'video' | |
9c906919 | 2655 | self._num_videos += 1 |
dd82ffea | 2656 | |
bec1fad2 | 2657 | if 'id' not in info_dict: |
fc08bdd6 | 2658 | raise ExtractorError('Missing "id" field in extractor result', ie=info_dict['extractor']) |
2659 | elif not info_dict.get('id'): | |
2660 | raise ExtractorError('Extractor failed to obtain "id"', ie=info_dict['extractor']) | |
455a15e2 | 2661 | |
c9969434 S |
2662 | def report_force_conversion(field, field_not, conversion): |
2663 | self.report_warning( | |
2664 | '"%s" field is not %s - forcing %s conversion, there is an error in extractor' | |
2665 | % (field, field_not, conversion)) | |
2666 | ||
2667 | def sanitize_string_field(info, string_field): | |
2668 | field = info.get(string_field) | |
14f25df2 | 2669 | if field is None or isinstance(field, str): |
c9969434 S |
2670 | return |
2671 | report_force_conversion(string_field, 'a string', 'string') | |
14f25df2 | 2672 | info[string_field] = str(field) |
c9969434 S |
2673 | |
2674 | def sanitize_numeric_fields(info): | |
2675 | for numeric_field in self._NUMERIC_FIELDS: | |
2676 | field = info.get(numeric_field) | |
f9934b96 | 2677 | if field is None or isinstance(field, (int, float)): |
c9969434 S |
2678 | continue |
2679 | report_force_conversion(numeric_field, 'numeric', 'int') | |
2680 | info[numeric_field] = int_or_none(field) | |
2681 | ||
2682 | sanitize_string_field(info_dict, 'id') | |
2683 | sanitize_numeric_fields(info_dict) | |
3975b4d2 | 2684 | if info_dict.get('section_end') and info_dict.get('section_start') is not None: |
2685 | info_dict['duration'] = round(info_dict['section_end'] - info_dict['section_start'], 3) | |
4c3f8c3f | 2686 | if (info_dict.get('duration') or 0) <= 0 and info_dict.pop('duration', None): |
50e93e03 | 2687 | self.report_warning('"duration" field is negative, there is an error in extractor') |
be6217b2 | 2688 | |
9eef7c4e | 2689 | chapters = info_dict.get('chapters') or [] |
a3976e07 | 2690 | if chapters and chapters[0].get('start_time'): |
2691 | chapters.insert(0, {'start_time': 0}) | |
2692 | ||
9eef7c4e | 2693 | dummy_chapter = {'end_time': 0, 'start_time': info_dict.get('duration')} |
a3976e07 | 2694 | for idx, (prev, current, next_) in enumerate(zip( |
2695 | (dummy_chapter, *chapters), chapters, (*chapters[1:], dummy_chapter)), 1): | |
9eef7c4e | 2696 | if current.get('start_time') is None: |
2697 | current['start_time'] = prev.get('end_time') | |
2698 | if not current.get('end_time'): | |
2699 | current['end_time'] = next_.get('start_time') | |
a3976e07 | 2700 | if not current.get('title'): |
2701 | current['title'] = f'<Untitled Chapter {idx}>' | |
9eef7c4e | 2702 | |
dd82ffea JMF |
2703 | if 'playlist' not in info_dict: |
2704 | # It isn't part of a playlist | |
2705 | info_dict['playlist'] = None | |
2706 | info_dict['playlist_index'] = None | |
2707 | ||
bc516a3f | 2708 | self._sanitize_thumbnails(info_dict) |
d5519808 | 2709 | |
536a55da | 2710 | thumbnail = info_dict.get('thumbnail') |
bc516a3f | 2711 | thumbnails = info_dict.get('thumbnails') |
536a55da S |
2712 | if thumbnail: |
2713 | info_dict['thumbnail'] = sanitize_url(thumbnail) | |
2714 | elif thumbnails: | |
d5519808 PH |
2715 | info_dict['thumbnail'] = thumbnails[-1]['url'] |
2716 | ||
ae30b840 | 2717 | if info_dict.get('display_id') is None and 'id' in info_dict: |
0afef30b PH |
2718 | info_dict['display_id'] = info_dict['id'] |
2719 | ||
03f83004 | 2720 | self._fill_common_fields(info_dict) |
33d2fc2f | 2721 | |
05108a49 S |
2722 | for cc_kind in ('subtitles', 'automatic_captions'): |
2723 | cc = info_dict.get(cc_kind) | |
2724 | if cc: | |
2725 | for _, subtitle in cc.items(): | |
2726 | for subtitle_format in subtitle: | |
2727 | if subtitle_format.get('url'): | |
2728 | subtitle_format['url'] = sanitize_url(subtitle_format['url']) | |
2729 | if subtitle_format.get('ext') is None: | |
2730 | subtitle_format['ext'] = determine_ext(subtitle_format['url']).lower() | |
2731 | ||
2732 | automatic_captions = info_dict.get('automatic_captions') | |
4bba3716 | 2733 | subtitles = info_dict.get('subtitles') |
4bba3716 | 2734 | |
360e1ca5 | 2735 | info_dict['requested_subtitles'] = self.process_subtitles( |
05108a49 | 2736 | info_dict['id'], subtitles, automatic_captions) |
a504ced0 | 2737 | |
aebb4f4b | 2738 | formats = self._get_formats(info_dict) |
dd82ffea | 2739 | |
c154302c | 2740 | # Backward compatibility with InfoExtractor._sort_formats |
9ebac355 | 2741 | field_preference = (formats or [{}])[0].pop('__sort_fields', None) |
c154302c | 2742 | if field_preference: |
2743 | info_dict['_format_sort_fields'] = field_preference | |
2744 | ||
bc344cd4 | 2745 | info_dict['_has_drm'] = any( # or None ensures --clean-infojson removes it |
2746 | f.get('has_drm') and f['has_drm'] != 'maybe' for f in formats) or None | |
88acdbc2 | 2747 | if not self.params.get('allow_unplayable_formats'): |
bc344cd4 | 2748 | formats = [f for f in formats if not f.get('has_drm') or f['has_drm'] == 'maybe'] |
17ffed18 | 2749 | |
2750 | if formats and all(f.get('acodec') == f.get('vcodec') == 'none' for f in formats): | |
2751 | self.report_warning( | |
2752 | f'{"This video is DRM protected and " if info_dict["_has_drm"] else ""}' | |
2753 | 'only images are available for download. Use --list-formats to see them'.capitalize()) | |
88acdbc2 | 2754 | |
319b6059 | 2755 | get_from_start = not info_dict.get('is_live') or bool(self.params.get('live_from_start')) |
2756 | if not get_from_start: | |
2757 | info_dict['title'] += ' ' + datetime.datetime.now().strftime('%Y-%m-%d %H:%M') | |
2758 | if info_dict.get('is_live') and formats: | |
adbc4ec4 | 2759 | formats = [f for f in formats if bool(f.get('is_from_start')) == get_from_start] |
319b6059 | 2760 | if get_from_start and not formats: |
a44ca5a4 | 2761 | self.raise_no_formats(info_dict, msg=( |
2762 | '--live-from-start is passed, but there are no formats that can be downloaded from the start. ' | |
2763 | 'If you want to download from the current time, use --no-live-from-start')) | |
adbc4ec4 | 2764 | |
73af5cc8 S |
2765 | def is_wellformed(f): |
2766 | url = f.get('url') | |
a5ac0c47 | 2767 | if not url: |
73af5cc8 S |
2768 | self.report_warning( |
2769 | '"url" field is missing or empty - skipping format, ' | |
2770 | 'there is an error in extractor') | |
a5ac0c47 S |
2771 | return False |
2772 | if isinstance(url, bytes): | |
2773 | sanitize_string_field(f, 'url') | |
2774 | return True | |
73af5cc8 S |
2775 | |
2776 | # Filter out malformed formats for better extraction robustness | |
1ac7f461 | 2777 | formats = list(filter(is_wellformed, formats or [])) |
2778 | ||
2779 | if not formats: | |
2780 | self.raise_no_formats(info_dict) | |
73af5cc8 | 2781 | |
39f32f17 | 2782 | for format in formats: |
c9969434 S |
2783 | sanitize_string_field(format, 'format_id') |
2784 | sanitize_numeric_fields(format) | |
dcf77cf1 | 2785 | format['url'] = sanitize_url(format['url']) |
39f32f17 | 2786 | if format.get('ext') is None: |
2787 | format['ext'] = determine_ext(format['url']).lower() | |
2788 | if format.get('protocol') is None: | |
2789 | format['protocol'] = determine_protocol(format) | |
2790 | if format.get('resolution') is None: | |
2791 | format['resolution'] = self.format_resolution(format, default=None) | |
2792 | if format.get('dynamic_range') is None and format.get('vcodec') != 'none': | |
2793 | format['dynamic_range'] = 'SDR' | |
2794 | if format.get('aspect_ratio') is None: | |
2795 | format['aspect_ratio'] = try_call(lambda: round(format['width'] / format['height'], 2)) | |
10025b71 | 2796 | # For fragmented formats, "tbr" is often max bitrate and not average |
2797 | if (('manifest-filesize-approx' in self.params['compat_opts'] or not format.get('manifest_url')) | |
0dff8e4d | 2798 | and info_dict.get('duration') and format.get('tbr') |
39f32f17 | 2799 | and not format.get('filesize') and not format.get('filesize_approx')): |
2800 | format['filesize_approx'] = int(info_dict['duration'] * format['tbr'] * (1024 / 8)) | |
6c5211ce | 2801 | format['http_headers'] = self._calc_headers(collections.ChainMap(format, info_dict), load_cookies=True) |
2802 | ||
2803 | # Safeguard against old/insecure infojson when using --load-info-json | |
2804 | if info_dict.get('http_headers'): | |
2805 | info_dict['http_headers'] = HTTPHeaderDict(info_dict['http_headers']) | |
2806 | info_dict['http_headers'].pop('Cookie', None) | |
39f32f17 | 2807 | |
2808 | # This is copied to http_headers by the above _calc_headers and can now be removed | |
2809 | if '__x_forwarded_for_ip' in info_dict: | |
2810 | del info_dict['__x_forwarded_for_ip'] | |
2811 | ||
c154302c | 2812 | self.sort_formats({ |
2813 | 'formats': formats, | |
2814 | '_format_sort_fields': info_dict.get('_format_sort_fields') | |
2815 | }) | |
39f32f17 | 2816 | |
2817 | # Sanitize and group by format_id | |
2818 | formats_dict = {} | |
2819 | for i, format in enumerate(formats): | |
e74e3b63 | 2820 | if not format.get('format_id'): |
14f25df2 | 2821 | format['format_id'] = str(i) |
e2effb08 S |
2822 | else: |
2823 | # Sanitize format_id from characters used in format selector expression | |
ec85ded8 | 2824 | format['format_id'] = re.sub(r'[\s,/+\[\]()]', '_', format['format_id']) |
39f32f17 | 2825 | formats_dict.setdefault(format['format_id'], []).append(format) |
181c7053 S |
2826 | |
2827 | # Make sure all formats have unique format_id | |
03b4de72 | 2828 | common_exts = set(itertools.chain(*self._format_selection_exts.values())) |
181c7053 | 2829 | for format_id, ambiguous_formats in formats_dict.items(): |
48ee10ee | 2830 | ambigious_id = len(ambiguous_formats) > 1 |
2831 | for i, format in enumerate(ambiguous_formats): | |
2832 | if ambigious_id: | |
181c7053 | 2833 | format['format_id'] = '%s-%d' % (format_id, i) |
48ee10ee | 2834 | # Ensure there is no conflict between id and ext in format selection |
2835 | # See https://github.com/yt-dlp/yt-dlp/issues/1282 | |
2836 | if format['format_id'] != format['ext'] and format['format_id'] in common_exts: | |
2837 | format['format_id'] = 'f%s' % format['format_id'] | |
181c7053 | 2838 | |
39f32f17 | 2839 | if format.get('format') is None: |
2840 | format['format'] = '{id} - {res}{note}'.format( | |
2841 | id=format['format_id'], | |
2842 | res=self.format_resolution(format), | |
2843 | note=format_field(format, 'format_note', ' (%s)'), | |
2844 | ) | |
dd82ffea | 2845 | |
9f1a1c36 | 2846 | if self.params.get('check_formats') is True: |
282f5709 | 2847 | formats = LazyList(self._check_formats(formats[::-1]), reverse=True) |
9f1a1c36 | 2848 | |
88acdbc2 | 2849 | if not formats or formats[0] is not info_dict: |
b3d9ef88 JMF |
2850 | # only set the 'formats' fields if the original info_dict list them |
2851 | # otherwise we end up with a circular reference, the first (and unique) | |
f89197d7 | 2852 | # element in the 'formats' field in info_dict is info_dict itself, |
dfb1b146 | 2853 | # which can't be exported to json |
b3d9ef88 | 2854 | info_dict['formats'] = formats |
4ec82a72 | 2855 | |
2856 | info_dict, _ = self.pre_process(info_dict) | |
2857 | ||
6db9c4d5 | 2858 | if self._match_entry(info_dict, incomplete=self._format_fields) is not None: |
09b49e1f | 2859 | return info_dict |
2860 | ||
2861 | self.post_extract(info_dict) | |
2862 | info_dict, _ = self.pre_process(info_dict, 'after_filter') | |
2863 | ||
093a1710 | 2864 | # The pre-processors may have modified the formats |
aebb4f4b | 2865 | formats = self._get_formats(info_dict) |
093a1710 | 2866 | |
e4221b70 | 2867 | list_only = self.params.get('simulate') == 'list_only' |
fa9f30b8 | 2868 | interactive_format_selection = not list_only and self.format_selector == '-' |
b7b04c78 | 2869 | if self.params.get('list_thumbnails'): |
2870 | self.list_thumbnails(info_dict) | |
b7b04c78 | 2871 | if self.params.get('listsubtitles'): |
2872 | if 'automatic_captions' in info_dict: | |
2873 | self.list_subtitles( | |
2874 | info_dict['id'], automatic_captions, 'automatic captions') | |
2875 | self.list_subtitles(info_dict['id'], subtitles, 'subtitles') | |
fa9f30b8 | 2876 | if self.params.get('listformats') or interactive_format_selection: |
b69fd25c | 2877 | self.list_formats(info_dict) |
169dbde9 | 2878 | if list_only: |
b7b04c78 | 2879 | # Without this printing, -F --print-json will not work |
17060584 | 2880 | self.__forced_printings(info_dict) |
c487cf00 | 2881 | return info_dict |
bfaae0a7 | 2882 | |
187986a8 | 2883 | format_selector = self.format_selector |
fa9f30b8 | 2884 | while True: |
2885 | if interactive_format_selection: | |
372a0f3b IS |
2886 | req_format = input(self._format_screen('\nEnter format selector ', self.Styles.EMPHASIS) |
2887 | + '(Press ENTER for default, or Ctrl+C to quit)' | |
2888 | + self._format_screen(': ', self.Styles.EMPHASIS)) | |
fa9f30b8 | 2889 | try: |
372a0f3b | 2890 | format_selector = self.build_format_selector(req_format) if req_format else None |
fa9f30b8 | 2891 | except SyntaxError as err: |
2892 | self.report_error(err, tb=False, is_error=False) | |
2893 | continue | |
2894 | ||
372a0f3b IS |
2895 | if format_selector is None: |
2896 | req_format = self._default_format_spec(info_dict, download=download) | |
2897 | self.write_debug(f'Default format spec: {req_format}') | |
2898 | format_selector = self.build_format_selector(req_format) | |
2899 | ||
85e801a9 | 2900 | formats_to_download = list(format_selector({ |
fa9f30b8 | 2901 | 'formats': formats, |
85e801a9 | 2902 | 'has_merged_format': any('none' not in (f.get('acodec'), f.get('vcodec')) for f in formats), |
906c0bdc | 2903 | 'incomplete_formats': (all(f.get('vcodec') == 'none' for f in formats) # No formats with video |
2904 | or all(f.get('acodec') == 'none' for f in formats)), # OR, No formats with audio | |
85e801a9 | 2905 | })) |
fa9f30b8 | 2906 | if interactive_format_selection and not formats_to_download: |
2907 | self.report_error('Requested format is not available', tb=False, is_error=False) | |
2908 | continue | |
2909 | break | |
317f7ab6 | 2910 | |
dd82ffea | 2911 | if not formats_to_download: |
b7da73eb | 2912 | if not self.params.get('ignore_no_formats_error'): |
c0b6e5c7 | 2913 | raise ExtractorError( |
2914 | 'Requested format is not available. Use --list-formats for a list of available formats', | |
2915 | expected=True, video_id=info_dict['id'], ie=info_dict['extractor']) | |
b62fa6d7 | 2916 | self.report_warning('Requested format is not available') |
2917 | # Process what we can, even without any available formats. | |
2918 | formats_to_download = [{}] | |
a13e6848 | 2919 | |
0500ee3d | 2920 | requested_ranges = tuple(self.params.get('download_ranges', lambda *_: [{}])(info_dict, self)) |
5ec1b6b7 | 2921 | best_format, downloaded_formats = formats_to_download[-1], [] |
b62fa6d7 | 2922 | if download: |
0500ee3d | 2923 | if best_format and requested_ranges: |
5ec1b6b7 | 2924 | def to_screen(*msg): |
2925 | self.to_screen(f'[info] {info_dict["id"]}: {" ".join(", ".join(variadic(m)) for m in msg)}') | |
2926 | ||
2927 | to_screen(f'Downloading {len(formats_to_download)} format(s):', | |
2928 | (f['format_id'] for f in formats_to_download)) | |
0500ee3d | 2929 | if requested_ranges != ({}, ): |
5ec1b6b7 | 2930 | to_screen(f'Downloading {len(requested_ranges)} time ranges:', |
fc2ba496 | 2931 | (f'{c["start_time"]:.1f}-{c["end_time"]:.1f}' for c in requested_ranges)) |
a13e6848 | 2932 | max_downloads_reached = False |
5ec1b6b7 | 2933 | |
0500ee3d | 2934 | for fmt, chapter in itertools.product(formats_to_download, requested_ranges): |
5ec1b6b7 | 2935 | new_info = self._copy_infodict(info_dict) |
b7da73eb | 2936 | new_info.update(fmt) |
3975b4d2 | 2937 | offset, duration = info_dict.get('section_start') or 0, info_dict.get('duration') or float('inf') |
fc2ba496 | 2938 | end_time = offset + min(chapter.get('end_time', duration), duration) |
b4e0d758 | 2939 | # duration may not be accurate. So allow deviations <1sec |
2940 | if end_time == float('inf') or end_time > offset + duration + 1: | |
2941 | end_time = None | |
3975b4d2 | 2942 | if chapter or offset: |
5ec1b6b7 | 2943 | new_info.update({ |
3975b4d2 | 2944 | 'section_start': offset + chapter.get('start_time', 0), |
b4e0d758 | 2945 | 'section_end': end_time, |
5ec1b6b7 | 2946 | 'section_title': chapter.get('title'), |
2947 | 'section_number': chapter.get('index'), | |
2948 | }) | |
2949 | downloaded_formats.append(new_info) | |
a13e6848 | 2950 | try: |
2951 | self.process_info(new_info) | |
2952 | except MaxDownloadsReached: | |
2953 | max_downloads_reached = True | |
415f8d51 | 2954 | self._raise_pending_errors(new_info) |
f46e2f9d | 2955 | # Remove copied info |
2956 | for key, val in tuple(new_info.items()): | |
2957 | if info_dict.get(key) == val: | |
2958 | new_info.pop(key) | |
a13e6848 | 2959 | if max_downloads_reached: |
2960 | break | |
ebed8b37 | 2961 | |
5ec1b6b7 | 2962 | write_archive = {f.get('__write_download_archive', False) for f in downloaded_formats} |
a13e6848 | 2963 | assert write_archive.issubset({True, False, 'ignore'}) |
2964 | if True in write_archive and False not in write_archive: | |
2965 | self.record_download_archive(info_dict) | |
be72c624 | 2966 | |
5ec1b6b7 | 2967 | info_dict['requested_downloads'] = downloaded_formats |
ed5835b4 | 2968 | info_dict = self.run_all_pps('after_video', info_dict) |
a13e6848 | 2969 | if max_downloads_reached: |
2970 | raise MaxDownloadsReached() | |
ebed8b37 | 2971 | |
49a57e70 | 2972 | # We update the info dict with the selected best quality format (backwards compatibility) |
be72c624 | 2973 | info_dict.update(best_format) |
dd82ffea JMF |
2974 | return info_dict |
2975 | ||
98c70d6f | 2976 | def process_subtitles(self, video_id, normal_subtitles, automatic_captions): |
a504ced0 | 2977 | """Select the requested subtitles and their format""" |
d8a58ddc | 2978 | available_subs, normal_sub_langs = {}, [] |
98c70d6f JMF |
2979 | if normal_subtitles and self.params.get('writesubtitles'): |
2980 | available_subs.update(normal_subtitles) | |
d8a58ddc | 2981 | normal_sub_langs = tuple(normal_subtitles.keys()) |
98c70d6f JMF |
2982 | if automatic_captions and self.params.get('writeautomaticsub'): |
2983 | for lang, cap_info in automatic_captions.items(): | |
360e1ca5 JMF |
2984 | if lang not in available_subs: |
2985 | available_subs[lang] = cap_info | |
2986 | ||
d2c8aadf | 2987 | if not available_subs or ( |
2988 | not self.params.get('writesubtitles') | |
2989 | and not self.params.get('writeautomaticsub')): | |
4d171848 | 2990 | return None |
a504ced0 | 2991 | |
d8a58ddc | 2992 | all_sub_langs = tuple(available_subs.keys()) |
a504ced0 | 2993 | if self.params.get('allsubtitles', False): |
c32b0aab | 2994 | requested_langs = all_sub_langs |
2995 | elif self.params.get('subtitleslangs', False): | |
5314b521 | 2996 | try: |
2997 | requested_langs = orderedSet_from_options( | |
2998 | self.params.get('subtitleslangs'), {'all': all_sub_langs}, use_regex=True) | |
2999 | except re.error as e: | |
3000 | raise ValueError(f'Wrong regex for subtitlelangs: {e.pattern}') | |
a504ced0 | 3001 | else: |
376aa24b SS |
3002 | requested_langs = LazyList(itertools.chain( |
3003 | ['en'] if 'en' in normal_sub_langs else [], | |
3004 | filter(lambda f: f.startswith('en'), normal_sub_langs), | |
3005 | ['en'] if 'en' in all_sub_langs else [], | |
3006 | filter(lambda f: f.startswith('en'), all_sub_langs), | |
3007 | normal_sub_langs, all_sub_langs, | |
3008 | ))[:1] | |
ad3dc496 | 3009 | if requested_langs: |
d2c8aadf | 3010 | self.to_screen(f'[info] {video_id}: Downloading subtitles: {", ".join(requested_langs)}') |
a504ced0 JMF |
3011 | |
3012 | formats_query = self.params.get('subtitlesformat', 'best') | |
3013 | formats_preference = formats_query.split('/') if formats_query else [] | |
3014 | subs = {} | |
3015 | for lang in requested_langs: | |
3016 | formats = available_subs.get(lang) | |
3017 | if formats is None: | |
86e5f3ed | 3018 | self.report_warning(f'{lang} subtitles not available for {video_id}') |
a504ced0 | 3019 | continue |
a504ced0 JMF |
3020 | for ext in formats_preference: |
3021 | if ext == 'best': | |
3022 | f = formats[-1] | |
3023 | break | |
3024 | matches = list(filter(lambda f: f['ext'] == ext, formats)) | |
3025 | if matches: | |
3026 | f = matches[-1] | |
3027 | break | |
3028 | else: | |
3029 | f = formats[-1] | |
3030 | self.report_warning( | |
3031 | 'No subtitle format found matching "%s" for language %s, ' | |
3032 | 'using %s' % (formats_query, lang, f['ext'])) | |
3033 | subs[lang] = f | |
3034 | return subs | |
3035 | ||
bb66c247 | 3036 | def _forceprint(self, key, info_dict): |
3037 | if info_dict is None: | |
3038 | return | |
3039 | info_copy = info_dict.copy() | |
17060584 | 3040 | info_copy.setdefault('filename', self.prepare_filename(info_dict)) |
3041 | if info_dict.get('requested_formats') is not None: | |
3042 | # For RTMP URLs, also include the playpath | |
3043 | info_copy['urls'] = '\n'.join(f['url'] + f.get('play_path', '') for f in info_dict['requested_formats']) | |
3044 | elif info_dict.get('url'): | |
3045 | info_copy['urls'] = info_dict['url'] + info_dict.get('play_path', '') | |
bb66c247 | 3046 | info_copy['formats_table'] = self.render_formats_table(info_dict) |
3047 | info_copy['thumbnails_table'] = self.render_thumbnails_table(info_dict) | |
3048 | info_copy['subtitles_table'] = self.render_subtitles_table(info_dict.get('id'), info_dict.get('subtitles')) | |
3049 | info_copy['automatic_captions_table'] = self.render_subtitles_table(info_dict.get('id'), info_dict.get('automatic_captions')) | |
3050 | ||
3051 | def format_tmpl(tmpl): | |
48c8424b | 3052 | mobj = re.fullmatch(r'([\w.:,]|-\d|(?P<dict>{([\w.:,]|-\d)+}))+=?', tmpl) |
07a1250e | 3053 | if not mobj: |
3054 | return tmpl | |
48c8424b | 3055 | |
3056 | fmt = '%({})s' | |
3057 | if tmpl.startswith('{'): | |
6f2287cb | 3058 | tmpl, fmt = f'.{tmpl}', '%({})j' |
48c8424b | 3059 | if tmpl.endswith('='): |
3060 | tmpl, fmt = tmpl[:-1], '{0} = %({0})#j' | |
3061 | return '\n'.join(map(fmt.format, [tmpl] if mobj.group('dict') else tmpl.split(','))) | |
8130779d | 3062 | |
bb66c247 | 3063 | for tmpl in self.params['forceprint'].get(key, []): |
3064 | self.to_stdout(self.evaluate_outtmpl(format_tmpl(tmpl), info_copy)) | |
3065 | ||
3066 | for tmpl, file_tmpl in self.params['print_to_file'].get(key, []): | |
5127e92a | 3067 | filename = self.prepare_filename(info_dict, outtmpl=file_tmpl) |
bb66c247 | 3068 | tmpl = format_tmpl(tmpl) |
3069 | self.to_screen(f'[info] Writing {tmpl!r} to: {filename}') | |
8d93e69d | 3070 | if self._ensure_dir_exists(filename): |
9874e82b | 3071 | with open(filename, 'a', encoding='utf-8', newline='') as f: |
3072 | f.write(self.evaluate_outtmpl(tmpl, info_copy) + os.linesep) | |
ca30f449 | 3073 | |
17060584 | 3074 | return info_copy |
3075 | ||
3076 | def __forced_printings(self, info_dict, filename=None, incomplete=True): | |
bb66c247 | 3077 | if (self.params.get('forcejson') |
3078 | or self.params['forceprint'].get('video') | |
3079 | or self.params['print_to_file'].get('video')): | |
2b8a2973 | 3080 | self.post_extract(info_dict) |
17060584 | 3081 | if filename: |
3082 | info_dict['filename'] = filename | |
b5f61b69 | 3083 | info_copy = self._forceprint('video', info_dict) |
3084 | ||
3085 | def print_field(field, actual_field=None, optional=False): | |
3086 | if actual_field is None: | |
3087 | actual_field = field | |
3088 | if self.params.get(f'force{field}') and ( | |
3089 | info_copy.get(field) is not None or (not optional and not incomplete)): | |
3090 | self.to_stdout(info_copy[actual_field]) | |
3091 | ||
3092 | print_field('title') | |
3093 | print_field('id') | |
3094 | print_field('url', 'urls') | |
3095 | print_field('thumbnail', optional=True) | |
3096 | print_field('description', optional=True) | |
ad54c913 | 3097 | print_field('filename') |
b5f61b69 | 3098 | if self.params.get('forceduration') and info_copy.get('duration') is not None: |
3099 | self.to_stdout(formatSeconds(info_copy['duration'])) | |
3100 | print_field('format') | |
53c18592 | 3101 | |
2b8a2973 | 3102 | if self.params.get('forcejson'): |
6e84b215 | 3103 | self.to_stdout(json.dumps(self.sanitize_info(info_dict))) |
d06daf23 | 3104 | |
e8e73840 | 3105 | def dl(self, name, info, subtitle=False, test=False): |
88acdbc2 | 3106 | if not info.get('url'): |
1151c407 | 3107 | self.raise_no_formats(info, True) |
e8e73840 | 3108 | |
3109 | if test: | |
3110 | verbose = self.params.get('verbose') | |
3111 | params = { | |
3112 | 'test': True, | |
a169858f | 3113 | 'quiet': self.params.get('quiet') or not verbose, |
e8e73840 | 3114 | 'verbose': verbose, |
3115 | 'noprogress': not verbose, | |
3116 | 'nopart': True, | |
3117 | 'skip_unavailable_fragments': False, | |
3118 | 'keep_fragments': False, | |
3119 | 'overwrites': True, | |
3120 | '_no_ytdl_file': True, | |
3121 | } | |
3122 | else: | |
3123 | params = self.params | |
96fccc10 | 3124 | fd = get_suitable_downloader(info, params, to_stdout=(name == '-'))(self, params) |
e8e73840 | 3125 | if not test: |
3126 | for ph in self._progress_hooks: | |
3127 | fd.add_progress_hook(ph) | |
42676437 M |
3128 | urls = '", "'.join( |
3129 | (f['url'].split(',')[0] + ',<data>' if f['url'].startswith('data:') else f['url']) | |
3130 | for f in info.get('requested_formats', []) or [info]) | |
3a408f9d | 3131 | self.write_debug(f'Invoking {fd.FD_NAME} downloader on "{urls}"') |
03b4de72 | 3132 | |
adbc4ec4 THD |
3133 | # Note: Ideally info should be a deep-copied so that hooks cannot modify it. |
3134 | # But it may contain objects that are not deep-copyable | |
3135 | new_info = self._copy_infodict(info) | |
e8e73840 | 3136 | if new_info.get('http_headers') is None: |
3137 | new_info['http_headers'] = self._calc_headers(new_info) | |
3138 | return fd.download(name, new_info, subtitle) | |
3139 | ||
e04938ab | 3140 | def existing_file(self, filepaths, *, default_overwrite=True): |
3141 | existing_files = list(filter(os.path.exists, orderedSet(filepaths))) | |
3142 | if existing_files and not self.params.get('overwrites', default_overwrite): | |
3143 | return existing_files[0] | |
3144 | ||
3145 | for file in existing_files: | |
3146 | self.report_file_delete(file) | |
3147 | os.remove(file) | |
3148 | return None | |
3149 | ||
8222d8de | 3150 | def process_info(self, info_dict): |
09b49e1f | 3151 | """Process a single resolved IE result. (Modifies it in-place)""" |
8222d8de JMF |
3152 | |
3153 | assert info_dict.get('_type', 'video') == 'video' | |
f46e2f9d | 3154 | original_infodict = info_dict |
fd288278 | 3155 | |
4513a41a | 3156 | if 'format' not in info_dict and 'ext' in info_dict: |
8222d8de JMF |
3157 | info_dict['format'] = info_dict['ext'] |
3158 | ||
c77495e3 | 3159 | if self._match_entry(info_dict) is not None: |
9e907ebd | 3160 | info_dict['__write_download_archive'] = 'ignore' |
8222d8de JMF |
3161 | return |
3162 | ||
09b49e1f | 3163 | # Does nothing under normal operation - for backward compatibility of process_info |
277d6ff5 | 3164 | self.post_extract(info_dict) |
119e40ef | 3165 | |
3166 | def replace_info_dict(new_info): | |
3167 | nonlocal info_dict | |
3168 | if new_info == info_dict: | |
3169 | return | |
3170 | info_dict.clear() | |
3171 | info_dict.update(new_info) | |
3172 | ||
3173 | new_info, _ = self.pre_process(info_dict, 'video') | |
3174 | replace_info_dict(new_info) | |
0c14d66a | 3175 | self._num_downloads += 1 |
8222d8de | 3176 | |
dcf64d43 | 3177 | # info_dict['_filename'] needs to be set for backward compatibility |
de6000d9 | 3178 | info_dict['_filename'] = full_filename = self.prepare_filename(info_dict, warn=True) |
3179 | temp_filename = self.prepare_filename(info_dict, 'temp') | |
0202b52a | 3180 | files_to_move = {} |
8222d8de JMF |
3181 | |
3182 | # Forced printings | |
4513a41a | 3183 | self.__forced_printings(info_dict, full_filename, incomplete=('format' not in info_dict)) |
8222d8de | 3184 | |
ca6d59d2 | 3185 | def check_max_downloads(): |
3186 | if self._num_downloads >= float(self.params.get('max_downloads') or 'inf'): | |
3187 | raise MaxDownloadsReached() | |
3188 | ||
b7b04c78 | 3189 | if self.params.get('simulate'): |
9e907ebd | 3190 | info_dict['__write_download_archive'] = self.params.get('force_write_download_archive') |
ca6d59d2 | 3191 | check_max_downloads() |
8222d8de JMF |
3192 | return |
3193 | ||
de6000d9 | 3194 | if full_filename is None: |
8222d8de | 3195 | return |
e92caff5 | 3196 | if not self._ensure_dir_exists(encodeFilename(full_filename)): |
0202b52a | 3197 | return |
e92caff5 | 3198 | if not self._ensure_dir_exists(encodeFilename(temp_filename)): |
8222d8de JMF |
3199 | return |
3200 | ||
80c03fa9 | 3201 | if self._write_description('video', info_dict, |
3202 | self.prepare_filename(info_dict, 'description')) is None: | |
3203 | return | |
3204 | ||
3205 | sub_files = self._write_subtitles(info_dict, temp_filename) | |
3206 | if sub_files is None: | |
3207 | return | |
3208 | files_to_move.update(dict(sub_files)) | |
3209 | ||
3210 | thumb_files = self._write_thumbnails( | |
3211 | 'video', info_dict, temp_filename, self.prepare_filename(info_dict, 'thumbnail')) | |
3212 | if thumb_files is None: | |
3213 | return | |
3214 | files_to_move.update(dict(thumb_files)) | |
8222d8de | 3215 | |
80c03fa9 | 3216 | infofn = self.prepare_filename(info_dict, 'infojson') |
3217 | _infojson_written = self._write_info_json('video', info_dict, infofn) | |
3218 | if _infojson_written: | |
dac5df5a | 3219 | info_dict['infojson_filename'] = infofn |
e75bb0d6 | 3220 | # For backward compatibility, even though it was a private field |
80c03fa9 | 3221 | info_dict['__infojson_filename'] = infofn |
3222 | elif _infojson_written is None: | |
3223 | return | |
3224 | ||
3225 | # Note: Annotations are deprecated | |
3226 | annofn = None | |
1fb07d10 | 3227 | if self.params.get('writeannotations', False): |
de6000d9 | 3228 | annofn = self.prepare_filename(info_dict, 'annotation') |
80c03fa9 | 3229 | if annofn: |
e92caff5 | 3230 | if not self._ensure_dir_exists(encodeFilename(annofn)): |
0202b52a | 3231 | return |
0c3d0f51 | 3232 | if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(annofn)): |
6febd1c1 | 3233 | self.to_screen('[info] Video annotations are already present') |
ffddb112 RA |
3234 | elif not info_dict.get('annotations'): |
3235 | self.report_warning('There are no annotations to write.') | |
7b6fefc9 PH |
3236 | else: |
3237 | try: | |
6febd1c1 | 3238 | self.to_screen('[info] Writing video annotations to: ' + annofn) |
86e5f3ed | 3239 | with open(encodeFilename(annofn), 'w', encoding='utf-8') as annofile: |
7b6fefc9 PH |
3240 | annofile.write(info_dict['annotations']) |
3241 | except (KeyError, TypeError): | |
6febd1c1 | 3242 | self.report_warning('There are no annotations to write.') |
86e5f3ed | 3243 | except OSError: |
6febd1c1 | 3244 | self.report_error('Cannot write annotations file: ' + annofn) |
7b6fefc9 | 3245 | return |
1fb07d10 | 3246 | |
732044af | 3247 | # Write internet shortcut files |
08438d2c | 3248 | def _write_link_file(link_type): |
60f3e995 | 3249 | url = try_get(info_dict['webpage_url'], iri_to_uri) |
3250 | if not url: | |
3251 | self.report_warning( | |
3252 | f'Cannot write internet shortcut file because the actual URL of "{info_dict["webpage_url"]}" is unknown') | |
3253 | return True | |
08438d2c | 3254 | linkfn = replace_extension(self.prepare_filename(info_dict, 'link'), link_type, info_dict.get('ext')) |
0e6b018a Z |
3255 | if not self._ensure_dir_exists(encodeFilename(linkfn)): |
3256 | return False | |
10e3742e | 3257 | if self.params.get('overwrites', True) and os.path.exists(encodeFilename(linkfn)): |
08438d2c | 3258 | self.to_screen(f'[info] Internet shortcut (.{link_type}) is already present') |
3259 | return True | |
3260 | try: | |
3261 | self.to_screen(f'[info] Writing internet shortcut (.{link_type}) to: {linkfn}') | |
86e5f3ed | 3262 | with open(encodeFilename(to_high_limit_path(linkfn)), 'w', encoding='utf-8', |
3263 | newline='\r\n' if link_type == 'url' else '\n') as linkfile: | |
60f3e995 | 3264 | template_vars = {'url': url} |
08438d2c | 3265 | if link_type == 'desktop': |
3266 | template_vars['filename'] = linkfn[:-(len(link_type) + 1)] | |
3267 | linkfile.write(LINK_TEMPLATES[link_type] % template_vars) | |
86e5f3ed | 3268 | except OSError: |
08438d2c | 3269 | self.report_error(f'Cannot write internet shortcut {linkfn}') |
3270 | return False | |
732044af | 3271 | return True |
3272 | ||
08438d2c | 3273 | write_links = { |
3274 | 'url': self.params.get('writeurllink'), | |
3275 | 'webloc': self.params.get('writewebloclink'), | |
3276 | 'desktop': self.params.get('writedesktoplink'), | |
3277 | } | |
3278 | if self.params.get('writelink'): | |
3279 | link_type = ('webloc' if sys.platform == 'darwin' | |
3280 | else 'desktop' if sys.platform.startswith('linux') | |
3281 | else 'url') | |
3282 | write_links[link_type] = True | |
3283 | ||
3284 | if any(should_write and not _write_link_file(link_type) | |
3285 | for link_type, should_write in write_links.items()): | |
3286 | return | |
732044af | 3287 | |
415f8d51 | 3288 | new_info, files_to_move = self.pre_process(info_dict, 'before_dl', files_to_move) |
3289 | replace_info_dict(new_info) | |
56d868db | 3290 | |
a13e6848 | 3291 | if self.params.get('skip_download'): |
56d868db | 3292 | info_dict['filepath'] = temp_filename |
3293 | info_dict['__finaldir'] = os.path.dirname(os.path.abspath(encodeFilename(full_filename))) | |
3294 | info_dict['__files_to_move'] = files_to_move | |
f46e2f9d | 3295 | replace_info_dict(self.run_pp(MoveFilesAfterDownloadPP(self, False), info_dict)) |
9e907ebd | 3296 | info_dict['__write_download_archive'] = self.params.get('force_write_download_archive') |
56d868db | 3297 | else: |
3298 | # Download | |
b868936c | 3299 | info_dict.setdefault('__postprocessors', []) |
4340deca | 3300 | try: |
0202b52a | 3301 | |
e04938ab | 3302 | def existing_video_file(*filepaths): |
6b591b29 | 3303 | ext = info_dict.get('ext') |
e04938ab | 3304 | converted = lambda file: replace_extension(file, self.params.get('final_ext') or ext, ext) |
3305 | file = self.existing_file(itertools.chain(*zip(map(converted, filepaths), filepaths)), | |
3306 | default_overwrite=False) | |
3307 | if file: | |
3308 | info_dict['ext'] = os.path.splitext(file)[1][1:] | |
3309 | return file | |
0202b52a | 3310 | |
7b2c3f47 | 3311 | fd, success = None, True |
fccf90e7 | 3312 | if info_dict.get('protocol') or info_dict.get('url'): |
56ba69e4 | 3313 | fd = get_suitable_downloader(info_dict, self.params, to_stdout=temp_filename == '-') |
62b5c94c | 3314 | if fd != FFmpegFD and 'no-direct-merge' not in self.params['compat_opts'] and ( |
56ba69e4 | 3315 | info_dict.get('section_start') or info_dict.get('section_end')): |
7b2c3f47 | 3316 | msg = ('This format cannot be partially downloaded' if FFmpegFD.available() |
56ba69e4 | 3317 | else 'You have requested downloading the video partially, but ffmpeg is not installed') |
3318 | self.report_error(f'{msg}. Aborting') | |
5ec1b6b7 | 3319 | return |
5ec1b6b7 | 3320 | |
4340deca | 3321 | if info_dict.get('requested_formats') is not None: |
0202b52a | 3322 | old_ext = info_dict['ext'] |
4e3b637d | 3323 | if self.params.get('merge_output_format') is None: |
4e3b637d | 3324 | if (info_dict['ext'] == 'webm' |
3325 | and info_dict.get('thumbnails') | |
3326 | # check with type instead of pp_key, __name__, or isinstance | |
3327 | # since we dont want any custom PPs to trigger this | |
c487cf00 | 3328 | and any(type(pp) == EmbedThumbnailPP for pp in self._pps['post_process'])): # noqa: E721 |
4e3b637d | 3329 | info_dict['ext'] = 'mkv' |
3330 | self.report_warning( | |
3331 | 'webm doesn\'t support embedding a thumbnail, mkv will be used') | |
124bc071 | 3332 | new_ext = info_dict['ext'] |
0202b52a | 3333 | |
124bc071 | 3334 | def correct_ext(filename, ext=new_ext): |
96fccc10 | 3335 | if filename == '-': |
3336 | return filename | |
0202b52a | 3337 | filename_real_ext = os.path.splitext(filename)[1][1:] |
3338 | filename_wo_ext = ( | |
3339 | os.path.splitext(filename)[0] | |
124bc071 | 3340 | if filename_real_ext in (old_ext, new_ext) |
0202b52a | 3341 | else filename) |
86e5f3ed | 3342 | return f'{filename_wo_ext}.{ext}' |
0202b52a | 3343 | |
38c6902b | 3344 | # Ensure filename always has a correct extension for successful merge |
0202b52a | 3345 | full_filename = correct_ext(full_filename) |
3346 | temp_filename = correct_ext(temp_filename) | |
e04938ab | 3347 | dl_filename = existing_video_file(full_filename, temp_filename) |
ad54c913 | 3348 | |
1ea24129 | 3349 | info_dict['__real_download'] = False |
84078a8b | 3350 | # NOTE: Copy so that original format dicts are not modified |
3351 | info_dict['requested_formats'] = list(map(dict, info_dict['requested_formats'])) | |
18e674b4 | 3352 | |
7b2c3f47 | 3353 | merger = FFmpegMergerPP(self) |
adbc4ec4 | 3354 | downloaded = [] |
dbf5416a | 3355 | if dl_filename is not None: |
6c7274ec | 3356 | self.report_file_already_downloaded(dl_filename) |
adbc4ec4 | 3357 | elif fd: |
ad54c913 | 3358 | for f in info_dict['requested_formats'] if fd != FFmpegFD else []: |
adbc4ec4 THD |
3359 | f['filepath'] = fname = prepend_extension( |
3360 | correct_ext(temp_filename, info_dict['ext']), | |
3361 | 'f%s' % f['format_id'], info_dict['ext']) | |
3362 | downloaded.append(fname) | |
ad54c913 | 3363 | info_dict['url'] = '\n'.join(f['url'] for f in info_dict['requested_formats']) |
dbf5416a | 3364 | success, real_download = self.dl(temp_filename, info_dict) |
3365 | info_dict['__real_download'] = real_download | |
18e674b4 | 3366 | else: |
18e674b4 | 3367 | if self.params.get('allow_unplayable_formats'): |
3368 | self.report_warning( | |
3369 | 'You have requested merging of multiple formats ' | |
3370 | 'while also allowing unplayable formats to be downloaded. ' | |
3371 | 'The formats won\'t be merged to prevent data corruption.') | |
3372 | elif not merger.available: | |
e8969bda | 3373 | msg = 'You have requested merging of multiple formats but ffmpeg is not installed' |
3374 | if not self.params.get('ignoreerrors'): | |
3375 | self.report_error(f'{msg}. Aborting due to --abort-on-error') | |
3376 | return | |
3377 | self.report_warning(f'{msg}. The formats won\'t be merged') | |
18e674b4 | 3378 | |
96fccc10 | 3379 | if temp_filename == '-': |
adbc4ec4 | 3380 | reason = ('using a downloader other than ffmpeg' if FFmpegFD.can_merge_formats(info_dict, self.params) |
96fccc10 | 3381 | else 'but the formats are incompatible for simultaneous download' if merger.available |
3382 | else 'but ffmpeg is not installed') | |
3383 | self.report_warning( | |
3384 | f'You have requested downloading multiple formats to stdout {reason}. ' | |
3385 | 'The formats will be streamed one after the other') | |
3386 | fname = temp_filename | |
ad54c913 | 3387 | for f in info_dict['requested_formats']: |
dbf5416a | 3388 | new_info = dict(info_dict) |
3389 | del new_info['requested_formats'] | |
3390 | new_info.update(f) | |
96fccc10 | 3391 | if temp_filename != '-': |
124bc071 | 3392 | fname = prepend_extension( |
3393 | correct_ext(temp_filename, new_info['ext']), | |
3394 | 'f%s' % f['format_id'], new_info['ext']) | |
96fccc10 | 3395 | if not self._ensure_dir_exists(fname): |
3396 | return | |
a21e0ab1 | 3397 | f['filepath'] = fname |
96fccc10 | 3398 | downloaded.append(fname) |
dbf5416a | 3399 | partial_success, real_download = self.dl(fname, new_info) |
3400 | info_dict['__real_download'] = info_dict['__real_download'] or real_download | |
3401 | success = success and partial_success | |
adbc4ec4 THD |
3402 | |
3403 | if downloaded and merger.available and not self.params.get('allow_unplayable_formats'): | |
3404 | info_dict['__postprocessors'].append(merger) | |
3405 | info_dict['__files_to_merge'] = downloaded | |
3406 | # Even if there were no downloads, it is being merged only now | |
3407 | info_dict['__real_download'] = True | |
3408 | else: | |
3409 | for file in downloaded: | |
3410 | files_to_move[file] = None | |
4340deca P |
3411 | else: |
3412 | # Just a single file | |
e04938ab | 3413 | dl_filename = existing_video_file(full_filename, temp_filename) |
6c7274ec | 3414 | if dl_filename is None or dl_filename == temp_filename: |
3415 | # dl_filename == temp_filename could mean that the file was partially downloaded with --no-part. | |
3416 | # So we should try to resume the download | |
e8e73840 | 3417 | success, real_download = self.dl(temp_filename, info_dict) |
0202b52a | 3418 | info_dict['__real_download'] = real_download |
6c7274ec | 3419 | else: |
3420 | self.report_file_already_downloaded(dl_filename) | |
0202b52a | 3421 | |
0202b52a | 3422 | dl_filename = dl_filename or temp_filename |
c571435f | 3423 | info_dict['__finaldir'] = os.path.dirname(os.path.abspath(encodeFilename(full_filename))) |
0202b52a | 3424 | |
3158150c | 3425 | except network_exceptions as err: |
7960b056 | 3426 | self.report_error('unable to download video data: %s' % error_to_compat_str(err)) |
4340deca | 3427 | return |
86e5f3ed | 3428 | except OSError as err: |
4340deca P |
3429 | raise UnavailableVideoError(err) |
3430 | except (ContentTooShortError, ) as err: | |
86e5f3ed | 3431 | self.report_error(f'content too short (expected {err.expected} bytes and served {err.downloaded})') |
4340deca | 3432 | return |
8222d8de | 3433 | |
415f8d51 | 3434 | self._raise_pending_errors(info_dict) |
de6000d9 | 3435 | if success and full_filename != '-': |
f17f8651 | 3436 | |
fd7cfb64 | 3437 | def fixup(): |
3438 | do_fixup = True | |
3439 | fixup_policy = self.params.get('fixup') | |
3440 | vid = info_dict['id'] | |
3441 | ||
3442 | if fixup_policy in ('ignore', 'never'): | |
3443 | return | |
3444 | elif fixup_policy == 'warn': | |
3fe75fdc | 3445 | do_fixup = 'warn' |
f89b3e2d | 3446 | elif fixup_policy != 'force': |
3447 | assert fixup_policy in ('detect_or_warn', None) | |
3448 | if not info_dict.get('__real_download'): | |
3449 | do_fixup = False | |
fd7cfb64 | 3450 | |
3451 | def ffmpeg_fixup(cndn, msg, cls): | |
3fe75fdc | 3452 | if not (do_fixup and cndn): |
fd7cfb64 | 3453 | return |
3fe75fdc | 3454 | elif do_fixup == 'warn': |
fd7cfb64 | 3455 | self.report_warning(f'{vid}: {msg}') |
3456 | return | |
3457 | pp = cls(self) | |
3458 | if pp.available: | |
3459 | info_dict['__postprocessors'].append(pp) | |
3460 | else: | |
3461 | self.report_warning(f'{vid}: {msg}. Install ffmpeg to fix this automatically') | |
3462 | ||
3463 | stretched_ratio = info_dict.get('stretched_ratio') | |
ca9def71 LNO |
3464 | ffmpeg_fixup(stretched_ratio not in (1, None), |
3465 | f'Non-uniform pixel ratio {stretched_ratio}', | |
3466 | FFmpegFixupStretchedPP) | |
fd7cfb64 | 3467 | |
993191c0 | 3468 | downloader = get_suitable_downloader(info_dict, self.params) if 'protocol' in info_dict else None |
24146491 | 3469 | downloader = downloader.FD_NAME if downloader else None |
adbc4ec4 | 3470 | |
ca9def71 LNO |
3471 | ext = info_dict.get('ext') |
3472 | postprocessed_by_ffmpeg = info_dict.get('requested_formats') or any(( | |
3473 | isinstance(pp, FFmpegVideoConvertorPP) | |
3474 | and resolve_recode_mapping(ext, pp.mapping)[0] not in (ext, None) | |
f73c1180 | 3475 | ) for pp in self._pps['post_process']) |
ca9def71 LNO |
3476 | |
3477 | if not postprocessed_by_ffmpeg: | |
f73c1180 | 3478 | ffmpeg_fixup(fd != FFmpegFD and ext == 'm4a' |
3479 | and info_dict.get('container') == 'm4a_dash', | |
f2df4071 | 3480 | 'writing DASH m4a. Only some players support this container', |
3481 | FFmpegFixupM4aPP) | |
24146491 | 3482 | ffmpeg_fixup(downloader == 'hlsnative' and not self.params.get('hls_use_mpegts') |
494f5230 | 3483 | or info_dict.get('is_live') and self.params.get('hls_use_mpegts') is None, |
adbc4ec4 THD |
3484 | 'Possible MPEG-TS in MP4 container or malformed AAC timestamps', |
3485 | FFmpegFixupM3u8PP) | |
26010b5c | 3486 | ffmpeg_fixup(info_dict.get('is_live') and downloader == 'dashsegments', |
adbc4ec4 THD |
3487 | 'Possible duplicate MOOV atoms', FFmpegFixupDuplicateMoovPP) |
3488 | ||
24146491 | 3489 | ffmpeg_fixup(downloader == 'web_socket_fragment', 'Malformed timestamps detected', FFmpegFixupTimestampPP) |
3490 | ffmpeg_fixup(downloader == 'web_socket_fragment', 'Malformed duration detected', FFmpegFixupDurationPP) | |
fd7cfb64 | 3491 | |
3492 | fixup() | |
8222d8de | 3493 | try: |
f46e2f9d | 3494 | replace_info_dict(self.post_process(dl_filename, info_dict, files_to_move)) |
af819c21 | 3495 | except PostProcessingError as err: |
3496 | self.report_error('Postprocessing: %s' % str(err)) | |
8222d8de | 3497 | return |
ab8e5e51 AM |
3498 | try: |
3499 | for ph in self._post_hooks: | |
23c1a667 | 3500 | ph(info_dict['filepath']) |
ab8e5e51 AM |
3501 | except Exception as err: |
3502 | self.report_error('post hooks: %s' % str(err)) | |
3503 | return | |
9e907ebd | 3504 | info_dict['__write_download_archive'] = True |
2d30509f | 3505 | |
c487cf00 | 3506 | assert info_dict is original_infodict # Make sure the info_dict was modified in-place |
a13e6848 | 3507 | if self.params.get('force_write_download_archive'): |
9e907ebd | 3508 | info_dict['__write_download_archive'] = True |
ca6d59d2 | 3509 | check_max_downloads() |
8222d8de | 3510 | |
aa9369a2 | 3511 | def __download_wrapper(self, func): |
3512 | @functools.wraps(func) | |
3513 | def wrapper(*args, **kwargs): | |
3514 | try: | |
3515 | res = func(*args, **kwargs) | |
3516 | except UnavailableVideoError as e: | |
3517 | self.report_error(e) | |
b222c271 | 3518 | except DownloadCancelled as e: |
3519 | self.to_screen(f'[info] {e}') | |
3520 | if not self.params.get('break_per_url'): | |
3521 | raise | |
fd404bec | 3522 | self._num_downloads = 0 |
aa9369a2 | 3523 | else: |
3524 | if self.params.get('dump_single_json', False): | |
3525 | self.post_extract(res) | |
3526 | self.to_stdout(json.dumps(self.sanitize_info(res))) | |
3527 | return wrapper | |
3528 | ||
8222d8de JMF |
3529 | def download(self, url_list): |
3530 | """Download a given list of URLs.""" | |
aa9369a2 | 3531 | url_list = variadic(url_list) # Passing a single URL is a common mistake |
bf1824b3 | 3532 | outtmpl = self.params['outtmpl']['default'] |
3089bc74 S |
3533 | if (len(url_list) > 1 |
3534 | and outtmpl != '-' | |
3535 | and '%' not in outtmpl | |
3536 | and self.params.get('max_downloads') != 1): | |
acd69589 | 3537 | raise SameFileError(outtmpl) |
8222d8de JMF |
3538 | |
3539 | for url in url_list: | |
aa9369a2 | 3540 | self.__download_wrapper(self.extract_info)( |
3541 | url, force_generic_extractor=self.params.get('force_generic_extractor', False)) | |
8222d8de JMF |
3542 | |
3543 | return self._download_retcode | |
3544 | ||
1dcc4c0c | 3545 | def download_with_info_file(self, info_filename): |
31bd3925 JMF |
3546 | with contextlib.closing(fileinput.FileInput( |
3547 | [info_filename], mode='r', | |
3548 | openhook=fileinput.hook_encoded('utf-8'))) as f: | |
3549 | # FileInput doesn't have a read method, we can't call json.load | |
ab1de9cb | 3550 | infos = [self.sanitize_info(info, self.params.get('clean_infojson', True)) |
3551 | for info in variadic(json.loads('\n'.join(f)))] | |
3552 | for info in infos: | |
3553 | try: | |
3554 | self.__download_wrapper(self.process_ie_result)(info, download=True) | |
3555 | except (DownloadError, EntryNotInPlaylist, ReExtractInfo) as e: | |
3556 | if not isinstance(e, EntryNotInPlaylist): | |
3557 | self.to_stderr('\r') | |
3558 | webpage_url = info.get('webpage_url') | |
3559 | if webpage_url is None: | |
3560 | raise | |
aa9369a2 | 3561 | self.report_warning(f'The info failed to download: {e}; trying with URL {webpage_url}') |
ab1de9cb | 3562 | self.download([webpage_url]) |
d4943898 | 3563 | return self._download_retcode |
1dcc4c0c | 3564 | |
cb202fd2 | 3565 | @staticmethod |
8012d892 | 3566 | def sanitize_info(info_dict, remove_private_keys=False): |
3567 | ''' Sanitize the infodict for converting to json ''' | |
3ad56b42 | 3568 | if info_dict is None: |
3569 | return info_dict | |
6e84b215 | 3570 | info_dict.setdefault('epoch', int(time.time())) |
6a5a30f9 | 3571 | info_dict.setdefault('_type', 'video') |
b5e7a2e6 | 3572 | info_dict.setdefault('_version', { |
3573 | 'version': __version__, | |
3574 | 'current_git_head': current_git_head(), | |
3575 | 'release_git_head': RELEASE_GIT_HEAD, | |
20314dd4 | 3576 | 'repository': ORIGIN, |
b5e7a2e6 | 3577 | }) |
09b49e1f | 3578 | |
8012d892 | 3579 | if remove_private_keys: |
0a5a191a | 3580 | reject = lambda k, v: v is None or k.startswith('__') or k in { |
f46e2f9d | 3581 | 'requested_downloads', 'requested_formats', 'requested_subtitles', 'requested_entries', |
6f2287cb | 3582 | 'entries', 'filepath', '_filename', 'filename', 'infojson_filename', 'original_url', |
595ea4a9 | 3583 | 'playlist_autonumber', |
6e84b215 | 3584 | } |
ae8f99e6 | 3585 | else: |
09b49e1f | 3586 | reject = lambda k, v: False |
adbc4ec4 THD |
3587 | |
3588 | def filter_fn(obj): | |
3589 | if isinstance(obj, dict): | |
3590 | return {k: filter_fn(v) for k, v in obj.items() if not reject(k, v)} | |
3591 | elif isinstance(obj, (list, tuple, set, LazyList)): | |
3592 | return list(map(filter_fn, obj)) | |
3593 | elif obj is None or isinstance(obj, (str, int, float, bool)): | |
3594 | return obj | |
3595 | else: | |
3596 | return repr(obj) | |
3597 | ||
5226731e | 3598 | return filter_fn(info_dict) |
cb202fd2 | 3599 | |
8012d892 | 3600 | @staticmethod |
3601 | def filter_requested_info(info_dict, actually_filter=True): | |
3602 | ''' Alias of sanitize_info for backward compatibility ''' | |
3603 | return YoutubeDL.sanitize_info(info_dict, actually_filter) | |
3604 | ||
43d7f5a5 | 3605 | def _delete_downloaded_files(self, *files_to_delete, info={}, msg=None): |
3606 | for filename in set(filter(None, files_to_delete)): | |
3607 | if msg: | |
3608 | self.to_screen(msg % filename) | |
3609 | try: | |
3610 | os.remove(filename) | |
3611 | except OSError: | |
3612 | self.report_warning(f'Unable to delete file {filename}') | |
3613 | if filename in info.get('__files_to_move', []): # NB: Delete even if None | |
3614 | del info['__files_to_move'][filename] | |
3615 | ||
ed5835b4 | 3616 | @staticmethod |
3617 | def post_extract(info_dict): | |
3618 | def actual_post_extract(info_dict): | |
3619 | if info_dict.get('_type') in ('playlist', 'multi_video'): | |
3620 | for video_dict in info_dict.get('entries', {}): | |
3621 | actual_post_extract(video_dict or {}) | |
3622 | return | |
3623 | ||
09b49e1f | 3624 | post_extractor = info_dict.pop('__post_extractor', None) or (lambda: {}) |
3625 | info_dict.update(post_extractor()) | |
ed5835b4 | 3626 | |
3627 | actual_post_extract(info_dict or {}) | |
3628 | ||
dcf64d43 | 3629 | def run_pp(self, pp, infodict): |
5bfa4862 | 3630 | files_to_delete = [] |
dcf64d43 | 3631 | if '__files_to_move' not in infodict: |
3632 | infodict['__files_to_move'] = {} | |
b1940459 | 3633 | try: |
3634 | files_to_delete, infodict = pp.run(infodict) | |
3635 | except PostProcessingError as e: | |
3636 | # Must be True and not 'only_download' | |
3637 | if self.params.get('ignoreerrors') is True: | |
3638 | self.report_error(e) | |
3639 | return infodict | |
3640 | raise | |
3641 | ||
5bfa4862 | 3642 | if not files_to_delete: |
dcf64d43 | 3643 | return infodict |
5bfa4862 | 3644 | if self.params.get('keepvideo', False): |
3645 | for f in files_to_delete: | |
dcf64d43 | 3646 | infodict['__files_to_move'].setdefault(f, '') |
5bfa4862 | 3647 | else: |
43d7f5a5 | 3648 | self._delete_downloaded_files( |
3649 | *files_to_delete, info=infodict, msg='Deleting original file %s (pass -k to keep)') | |
dcf64d43 | 3650 | return infodict |
5bfa4862 | 3651 | |
6f2287cb | 3652 | def run_all_pps(self, key, info, *, additional_pps=None): |
17ba4343 | 3653 | if key != 'video': |
3654 | self._forceprint(key, info) | |
3655 | for pp in (additional_pps or []) + self._pps[key]: | |
3656 | info = self.run_pp(pp, info) | |
ed5835b4 | 3657 | return info |
277d6ff5 | 3658 | |
56d868db | 3659 | def pre_process(self, ie_info, key='pre_process', files_to_move=None): |
5bfa4862 | 3660 | info = dict(ie_info) |
56d868db | 3661 | info['__files_to_move'] = files_to_move or {} |
415f8d51 | 3662 | try: |
3663 | info = self.run_all_pps(key, info) | |
3664 | except PostProcessingError as err: | |
3665 | msg = f'Preprocessing: {err}' | |
3666 | info.setdefault('__pending_error', msg) | |
3667 | self.report_error(msg, is_error=False) | |
56d868db | 3668 | return info, info.pop('__files_to_move', None) |
5bfa4862 | 3669 | |
f46e2f9d | 3670 | def post_process(self, filename, info, files_to_move=None): |
8222d8de | 3671 | """Run all the postprocessors on the given file.""" |
8222d8de | 3672 | info['filepath'] = filename |
dcf64d43 | 3673 | info['__files_to_move'] = files_to_move or {} |
ed5835b4 | 3674 | info = self.run_all_pps('post_process', info, additional_pps=info.get('__postprocessors')) |
dcf64d43 | 3675 | info = self.run_pp(MoveFilesAfterDownloadPP(self), info) |
3676 | del info['__files_to_move'] | |
ed5835b4 | 3677 | return self.run_all_pps('after_move', info) |
c1c9a79c | 3678 | |
5db07df6 | 3679 | def _make_archive_id(self, info_dict): |
e9fef7ee S |
3680 | video_id = info_dict.get('id') |
3681 | if not video_id: | |
3682 | return | |
5db07df6 PH |
3683 | # Future-proof against any change in case |
3684 | # and backwards compatibility with prior versions | |
e9fef7ee | 3685 | extractor = info_dict.get('extractor_key') or info_dict.get('ie_key') # key in a playlist |
7012b23c | 3686 | if extractor is None: |
1211bb6d S |
3687 | url = str_or_none(info_dict.get('url')) |
3688 | if not url: | |
3689 | return | |
e9fef7ee | 3690 | # Try to find matching extractor for the URL and take its ie_key |
8b7491c8 | 3691 | for ie_key, ie in self._ies.items(): |
1211bb6d | 3692 | if ie.suitable(url): |
8b7491c8 | 3693 | extractor = ie_key |
e9fef7ee S |
3694 | break |
3695 | else: | |
3696 | return | |
0647d925 | 3697 | return make_archive_id(extractor, video_id) |
5db07df6 PH |
3698 | |
3699 | def in_download_archive(self, info_dict): | |
ae103564 | 3700 | if not self.archive: |
5db07df6 PH |
3701 | return False |
3702 | ||
1e8fe57e | 3703 | vid_ids = [self._make_archive_id(info_dict)] |
c200096c | 3704 | vid_ids.extend(info_dict.get('_old_archive_ids') or []) |
1e8fe57e | 3705 | return any(id_ in self.archive for id_ in vid_ids) |
c1c9a79c PH |
3706 | |
3707 | def record_download_archive(self, info_dict): | |
3708 | fn = self.params.get('download_archive') | |
3709 | if fn is None: | |
3710 | return | |
5db07df6 PH |
3711 | vid_id = self._make_archive_id(info_dict) |
3712 | assert vid_id | |
ae103564 | 3713 | |
a13e6848 | 3714 | self.write_debug(f'Adding to archive: {vid_id}') |
9c935fbc | 3715 | if is_path_like(fn): |
ae103564 | 3716 | with locked_file(fn, 'a', encoding='utf-8') as archive_file: |
3717 | archive_file.write(vid_id + '\n') | |
a45e8619 | 3718 | self.archive.add(vid_id) |
dd82ffea | 3719 | |
8c51aa65 | 3720 | @staticmethod |
8abeeb94 | 3721 | def format_resolution(format, default='unknown'): |
9359f3d4 | 3722 | if format.get('vcodec') == 'none' and format.get('acodec') != 'none': |
fb04e403 | 3723 | return 'audio only' |
f49d89ee PH |
3724 | if format.get('resolution') is not None: |
3725 | return format['resolution'] | |
35615307 | 3726 | if format.get('width') and format.get('height'): |
ff51ed58 | 3727 | return '%dx%d' % (format['width'], format['height']) |
35615307 | 3728 | elif format.get('height'): |
ff51ed58 | 3729 | return '%sp' % format['height'] |
35615307 | 3730 | elif format.get('width'): |
ff51ed58 | 3731 | return '%dx?' % format['width'] |
3732 | return default | |
8c51aa65 | 3733 | |
8130779d | 3734 | def _list_format_headers(self, *headers): |
3735 | if self.params.get('listformats_table', True) is not False: | |
591bb9d3 | 3736 | return [self._format_out(header, self.Styles.HEADERS) for header in headers] |
8130779d | 3737 | return headers |
3738 | ||
c57f7757 PH |
3739 | def _format_note(self, fdict): |
3740 | res = '' | |
3741 | if fdict.get('ext') in ['f4f', 'f4m']: | |
f304da8a | 3742 | res += '(unsupported)' |
32f90364 PH |
3743 | if fdict.get('language'): |
3744 | if res: | |
3745 | res += ' ' | |
f304da8a | 3746 | res += '[%s]' % fdict['language'] |
c57f7757 | 3747 | if fdict.get('format_note') is not None: |
f304da8a | 3748 | if res: |
3749 | res += ' ' | |
3750 | res += fdict['format_note'] | |
c57f7757 | 3751 | if fdict.get('tbr') is not None: |
f304da8a | 3752 | if res: |
3753 | res += ', ' | |
3754 | res += '%4dk' % fdict['tbr'] | |
c57f7757 PH |
3755 | if fdict.get('container') is not None: |
3756 | if res: | |
3757 | res += ', ' | |
3758 | res += '%s container' % fdict['container'] | |
3089bc74 S |
3759 | if (fdict.get('vcodec') is not None |
3760 | and fdict.get('vcodec') != 'none'): | |
c57f7757 PH |
3761 | if res: |
3762 | res += ', ' | |
3763 | res += fdict['vcodec'] | |
91c7271a | 3764 | if fdict.get('vbr') is not None: |
c57f7757 PH |
3765 | res += '@' |
3766 | elif fdict.get('vbr') is not None and fdict.get('abr') is not None: | |
3767 | res += 'video@' | |
3768 | if fdict.get('vbr') is not None: | |
3769 | res += '%4dk' % fdict['vbr'] | |
fbb21cf5 | 3770 | if fdict.get('fps') is not None: |
5d583bdf S |
3771 | if res: |
3772 | res += ', ' | |
3773 | res += '%sfps' % fdict['fps'] | |
c57f7757 PH |
3774 | if fdict.get('acodec') is not None: |
3775 | if res: | |
3776 | res += ', ' | |
3777 | if fdict['acodec'] == 'none': | |
3778 | res += 'video only' | |
3779 | else: | |
3780 | res += '%-5s' % fdict['acodec'] | |
3781 | elif fdict.get('abr') is not None: | |
3782 | if res: | |
3783 | res += ', ' | |
3784 | res += 'audio' | |
3785 | if fdict.get('abr') is not None: | |
3786 | res += '@%3dk' % fdict['abr'] | |
3787 | if fdict.get('asr') is not None: | |
3788 | res += ' (%5dHz)' % fdict['asr'] | |
3789 | if fdict.get('filesize') is not None: | |
3790 | if res: | |
3791 | res += ', ' | |
3792 | res += format_bytes(fdict['filesize']) | |
9732d77e PH |
3793 | elif fdict.get('filesize_approx') is not None: |
3794 | if res: | |
3795 | res += ', ' | |
3796 | res += '~' + format_bytes(fdict['filesize_approx']) | |
c57f7757 | 3797 | return res |
91c7271a | 3798 | |
aebb4f4b | 3799 | def _get_formats(self, info_dict): |
3800 | if info_dict.get('formats') is None: | |
3801 | if info_dict.get('url') and info_dict.get('_type', 'video') == 'video': | |
3802 | return [info_dict] | |
3803 | return [] | |
3804 | return info_dict['formats'] | |
b69fd25c | 3805 | |
aebb4f4b | 3806 | def render_formats_table(self, info_dict): |
3807 | formats = self._get_formats(info_dict) | |
3808 | if not formats: | |
3809 | return | |
8130779d | 3810 | if not self.params.get('listformats_table', True) is not False: |
76d321f6 | 3811 | table = [ |
3812 | [ | |
3813 | format_field(f, 'format_id'), | |
3814 | format_field(f, 'ext'), | |
3815 | self.format_resolution(f), | |
8130779d | 3816 | self._format_note(f) |
d5d1df8a | 3817 | ] for f in formats if (f.get('preference') or 0) >= -1000] |
8130779d | 3818 | return render_table(['format code', 'extension', 'resolution', 'note'], table, extra_gap=1) |
3819 | ||
d816f61f | 3820 | def simplified_codec(f, field): |
3821 | assert field in ('acodec', 'vcodec') | |
337734d4 | 3822 | codec = f.get(field) |
f5ea4748 | 3823 | if not codec: |
3824 | return 'unknown' | |
3825 | elif codec != 'none': | |
d816f61f | 3826 | return '.'.join(codec.split('.')[:4]) |
3827 | ||
3828 | if field == 'vcodec' and f.get('acodec') == 'none': | |
3829 | return 'images' | |
3830 | elif field == 'acodec' and f.get('vcodec') == 'none': | |
3831 | return '' | |
3832 | return self._format_out('audio only' if field == 'vcodec' else 'video only', | |
3833 | self.Styles.SUPPRESS) | |
3834 | ||
591bb9d3 | 3835 | delim = self._format_out('\u2502', self.Styles.DELIM, '|', test_encoding=True) |
8130779d | 3836 | table = [ |
3837 | [ | |
591bb9d3 | 3838 | self._format_out(format_field(f, 'format_id'), self.Styles.ID), |
8130779d | 3839 | format_field(f, 'ext'), |
3840 | format_field(f, func=self.format_resolution, ignore=('audio only', 'images')), | |
563e0bf8 | 3841 | format_field(f, 'fps', '\t%d', func=round), |
8130779d | 3842 | format_field(f, 'dynamic_range', '%s', ignore=(None, 'SDR')).replace('HDR', ''), |
b8ed0f15 | 3843 | format_field(f, 'audio_channels', '\t%s'), |
0dff8e4d | 3844 | delim, ( |
3845 | format_field(f, 'filesize', ' \t%s', func=format_bytes) | |
3846 | or format_field(f, 'filesize_approx', '≈\t%s', func=format_bytes) | |
3847 | or format_field(try_call(lambda: format_bytes(int(info_dict['duration'] * f['tbr'] * (1024 / 8)))), | |
3848 | None, self._format_out('~\t%s', self.Styles.SUPPRESS))), | |
563e0bf8 | 3849 | format_field(f, 'tbr', '\t%dk', func=round), |
8130779d | 3850 | shorten_protocol_name(f.get('protocol', '')), |
3851 | delim, | |
d816f61f | 3852 | simplified_codec(f, 'vcodec'), |
563e0bf8 | 3853 | format_field(f, 'vbr', '\t%dk', func=round), |
d816f61f | 3854 | simplified_codec(f, 'acodec'), |
563e0bf8 | 3855 | format_field(f, 'abr', '\t%dk', func=round), |
ae61d108 | 3856 | format_field(f, 'asr', '\t%s', func=format_decimal_suffix), |
bc344cd4 | 3857 | join_nonempty(format_field(f, 'language', '[%s]'), join_nonempty( |
3858 | self._format_out('UNSUPPORTED', self.Styles.BAD_FORMAT) if f.get('ext') in ('f4f', 'f4m') else None, | |
3859 | (self._format_out('Maybe DRM', self.Styles.WARNING) if f.get('has_drm') == 'maybe' | |
3860 | else self._format_out('DRM', self.Styles.BAD_FORMAT) if f.get('has_drm') else None), | |
3861 | format_field(f, 'format_note'), | |
3862 | format_field(f, 'container', ignore=(None, f.get('ext'))), | |
3863 | delim=', '), delim=' '), | |
8130779d | 3864 | ] for f in formats if f.get('preference') is None or f['preference'] >= -1000] |
3865 | header_line = self._list_format_headers( | |
b8ed0f15 | 3866 | 'ID', 'EXT', 'RESOLUTION', '\tFPS', 'HDR', 'CH', delim, '\tFILESIZE', '\tTBR', 'PROTO', |
8130779d | 3867 | delim, 'VCODEC', '\tVBR', 'ACODEC', '\tABR', '\tASR', 'MORE INFO') |
3868 | ||
3869 | return render_table( | |
3870 | header_line, table, hide_empty=True, | |
591bb9d3 | 3871 | delim=self._format_out('\u2500', self.Styles.DELIM, '-', test_encoding=True)) |
8130779d | 3872 | |
3873 | def render_thumbnails_table(self, info_dict): | |
88f23a18 | 3874 | thumbnails = list(info_dict.get('thumbnails') or []) |
cfb56d1a | 3875 | if not thumbnails: |
8130779d | 3876 | return None |
3877 | return render_table( | |
ec11a9f4 | 3878 | self._list_format_headers('ID', 'Width', 'Height', 'URL'), |
177662e0 | 3879 | [[t.get('id'), t.get('width') or 'unknown', t.get('height') or 'unknown', t['url']] for t in thumbnails]) |
2412044c | 3880 | |
8130779d | 3881 | def render_subtitles_table(self, video_id, subtitles): |
2412044c | 3882 | def _row(lang, formats): |
49c258e1 | 3883 | exts, names = zip(*((f['ext'], f.get('name') or 'unknown') for f in reversed(formats))) |
2412044c | 3884 | if len(set(names)) == 1: |
7aee40c1 | 3885 | names = [] if names[0] == 'unknown' else names[:1] |
2412044c | 3886 | return [lang, ', '.join(names), ', '.join(exts)] |
3887 | ||
8130779d | 3888 | if not subtitles: |
3889 | return None | |
3890 | return render_table( | |
ec11a9f4 | 3891 | self._list_format_headers('Language', 'Name', 'Formats'), |
2412044c | 3892 | [_row(lang, formats) for lang, formats in subtitles.items()], |
8130779d | 3893 | hide_empty=True) |
3894 | ||
3895 | def __list_table(self, video_id, name, func, *args): | |
3896 | table = func(*args) | |
3897 | if not table: | |
3898 | self.to_screen(f'{video_id} has no {name}') | |
3899 | return | |
3900 | self.to_screen(f'[info] Available {name} for {video_id}:') | |
3901 | self.to_stdout(table) | |
3902 | ||
3903 | def list_formats(self, info_dict): | |
3904 | self.__list_table(info_dict['id'], 'formats', self.render_formats_table, info_dict) | |
3905 | ||
3906 | def list_thumbnails(self, info_dict): | |
3907 | self.__list_table(info_dict['id'], 'thumbnails', self.render_thumbnails_table, info_dict) | |
3908 | ||
3909 | def list_subtitles(self, video_id, subtitles, name='subtitles'): | |
3910 | self.__list_table(video_id, name, self.render_subtitles_table, video_id, subtitles) | |
a504ced0 | 3911 | |
dca08720 PH |
3912 | def print_debug_header(self): |
3913 | if not self.params.get('verbose'): | |
3914 | return | |
49a57e70 | 3915 | |
a057779d | 3916 | from . import _IN_CLI # Must be delayed import |
3917 | ||
560738f3 | 3918 | # These imports can be slow. So import them only as needed |
3919 | from .extractor.extractors import _LAZY_LOADER | |
e756f45b M |
3920 | from .extractor.extractors import ( |
3921 | _PLUGIN_CLASSES as plugin_ies, | |
3922 | _PLUGIN_OVERRIDES as plugin_ie_overrides | |
3923 | ) | |
560738f3 | 3924 | |
49a57e70 | 3925 | def get_encoding(stream): |
2a938746 | 3926 | ret = str(getattr(stream, 'encoding', 'missing (%s)' % type(stream).__name__)) |
8417f26b SS |
3927 | additional_info = [] |
3928 | if os.environ.get('TERM', '').lower() == 'dumb': | |
3929 | additional_info.append('dumb') | |
49a57e70 | 3930 | if not supports_terminal_sequences(stream): |
53973b4d | 3931 | from .utils import WINDOWS_VT_MODE # Must be imported locally |
8417f26b SS |
3932 | additional_info.append('No VT' if WINDOWS_VT_MODE is False else 'No ANSI') |
3933 | if additional_info: | |
3934 | ret = f'{ret} ({",".join(additional_info)})' | |
49a57e70 | 3935 | return ret |
3936 | ||
591bb9d3 | 3937 | encoding_str = 'Encodings: locale %s, fs %s, pref %s, %s' % ( |
49a57e70 | 3938 | locale.getpreferredencoding(), |
3939 | sys.getfilesystemencoding(), | |
591bb9d3 | 3940 | self.get_encoding(), |
3941 | ', '.join( | |
64fa820c | 3942 | f'{key} {get_encoding(stream)}' for key, stream in self._out_files.items_ |
591bb9d3 | 3943 | if stream is not None and key != 'console') |
3944 | ) | |
883d4b1e | 3945 | |
3946 | logger = self.params.get('logger') | |
3947 | if logger: | |
3948 | write_debug = lambda msg: logger.debug(f'[debug] {msg}') | |
3949 | write_debug(encoding_str) | |
3950 | else: | |
96565c7e | 3951 | write_string(f'[debug] {encoding_str}\n', encoding=None) |
49a57e70 | 3952 | write_debug = lambda msg: self._write_string(f'[debug] {msg}\n') |
734f90bb | 3953 | |
4c88ff87 | 3954 | source = detect_variant() |
70b23409 | 3955 | if VARIANT not in (None, 'pip'): |
3956 | source += '*' | |
a5387729 | 3957 | klass = type(self) |
36eaf303 | 3958 | write_debug(join_nonempty( |
20314dd4 | 3959 | f'{REPOSITORY.rpartition("/")[2]} version', |
0b6ad22e | 3960 | _make_label(ORIGIN, CHANNEL.partition('@')[2] or __version__, __version__), |
29cb20bd | 3961 | f'[{RELEASE_GIT_HEAD[:9]}]' if RELEASE_GIT_HEAD else '', |
36eaf303 | 3962 | '' if source == 'unknown' else f'({source})', |
a5387729 | 3963 | '' if _IN_CLI else 'API' if klass == YoutubeDL else f'API:{self.__module__}.{klass.__qualname__}', |
36eaf303 | 3964 | delim=' ')) |
497074f0 | 3965 | |
3966 | if not _IN_CLI: | |
3967 | write_debug(f'params: {self.params}') | |
3968 | ||
6e21fdd2 | 3969 | if not _LAZY_LOADER: |
3970 | if os.environ.get('YTDLP_NO_LAZY_EXTRACTORS'): | |
49a57e70 | 3971 | write_debug('Lazy loading extractors is forcibly disabled') |
6e21fdd2 | 3972 | else: |
49a57e70 | 3973 | write_debug('Lazy loading extractors is disabled') |
8a82af35 | 3974 | if self.params['compat_opts']: |
3975 | write_debug('Compatibility options: %s' % ', '.join(self.params['compat_opts'])) | |
36eaf303 | 3976 | |
b5e7a2e6 | 3977 | if current_git_head(): |
3978 | write_debug(f'Git HEAD: {current_git_head()}') | |
b1f94422 | 3979 | write_debug(system_identifier()) |
d28b5171 | 3980 | |
8913ef74 | 3981 | exe_versions, ffmpeg_features = FFmpegPostProcessor.get_versions_and_features(self) |
3982 | ffmpeg_features = {key for key, val in ffmpeg_features.items() if val} | |
3983 | if ffmpeg_features: | |
19a03940 | 3984 | exe_versions['ffmpeg'] += ' (%s)' % ','.join(sorted(ffmpeg_features)) |
8913ef74 | 3985 | |
4c83c967 | 3986 | exe_versions['rtmpdump'] = rtmpdump_version() |
feee8d32 | 3987 | exe_versions['phantomjs'] = PhantomJSwrapper._version() |
d28b5171 | 3988 | exe_str = ', '.join( |
2831b468 | 3989 | f'{exe} {v}' for exe, v in sorted(exe_versions.items()) if v |
3990 | ) or 'none' | |
49a57e70 | 3991 | write_debug('exe versions: %s' % exe_str) |
dca08720 | 3992 | |
1d485a1a | 3993 | from .compat.compat_utils import get_package_info |
9b8ee23b | 3994 | from .dependencies import available_dependencies |
3995 | ||
3996 | write_debug('Optional libraries: %s' % (', '.join(sorted({ | |
1d485a1a | 3997 | join_nonempty(*get_package_info(m)) for m in available_dependencies.values() |
9b8ee23b | 3998 | })) or 'none')) |
2831b468 | 3999 | |
227bf1a3 | 4000 | write_debug(f'Proxy map: {self.proxies}') |
8a8b5452 | 4001 | write_debug(f'Request Handlers: {", ".join(rh.RH_NAME for rh in self._request_director.handlers.values())}') |
e756f45b M |
4002 | for plugin_type, plugins in {'Extractor': plugin_ies, 'Post-Processor': plugin_pps}.items(): |
4003 | display_list = ['%s%s' % ( | |
8e40b9d1 | 4004 | klass.__name__, '' if klass.__name__ == name else f' as {name}') |
e756f45b M |
4005 | for name, klass in plugins.items()] |
4006 | if plugin_type == 'Extractor': | |
4007 | display_list.extend(f'{plugins[-1].IE_NAME.partition("+")[2]} ({parent.__name__})' | |
4008 | for parent, plugins in plugin_ie_overrides.items()) | |
4009 | if not display_list: | |
4010 | continue | |
4011 | write_debug(f'{plugin_type} Plugins: {", ".join(sorted(display_list))}') | |
4012 | ||
8e40b9d1 M |
4013 | plugin_dirs = plugin_directories() |
4014 | if plugin_dirs: | |
4015 | write_debug(f'Plugin directories: {plugin_dirs}') | |
4016 | ||
49a57e70 | 4017 | # Not implemented |
4018 | if False and self.params.get('call_home'): | |
0f06bcd7 | 4019 | ipaddr = self.urlopen('https://yt-dl.org/ip').read().decode() |
49a57e70 | 4020 | write_debug('Public IP address: %s' % ipaddr) |
58b1f00d | 4021 | latest_version = self.urlopen( |
0f06bcd7 | 4022 | 'https://yt-dl.org/latest/version').read().decode() |
58b1f00d PH |
4023 | if version_tuple(latest_version) > version_tuple(__version__): |
4024 | self.report_warning( | |
4025 | 'You are using an outdated version (newest version: %s)! ' | |
4026 | 'See https://yt-dl.org/update if you need help updating.' % | |
4027 | latest_version) | |
4028 | ||
227bf1a3 | 4029 | @functools.cached_property |
4030 | def proxies(self): | |
4031 | """Global proxy configuration""" | |
dca08720 | 4032 | opts_proxy = self.params.get('proxy') |
dca08720 PH |
4033 | if opts_proxy is not None: |
4034 | if opts_proxy == '': | |
227bf1a3 | 4035 | opts_proxy = '__noproxy__' |
4036 | proxies = {'all': opts_proxy} | |
dca08720 | 4037 | else: |
ac668111 | 4038 | proxies = urllib.request.getproxies() |
227bf1a3 | 4039 | # compat. Set HTTPS_PROXY to __noproxy__ to revert |
dca08720 PH |
4040 | if 'http' in proxies and 'https' not in proxies: |
4041 | proxies['https'] = proxies['http'] | |
227bf1a3 | 4042 | |
4043 | return proxies | |
62fec3b2 | 4044 | |
c365dba8 | 4045 | @functools.cached_property |
4046 | def cookiejar(self): | |
4047 | """Global cookiejar instance""" | |
4048 | return load_cookies( | |
4049 | self.params.get('cookiefile'), self.params.get('cookiesfrombrowser'), self) | |
4050 | ||
227bf1a3 | 4051 | @property |
4052 | def _opener(self): | |
4053 | """ | |
4054 | Get a urllib OpenerDirector from the Urllib handler (deprecated). | |
4055 | """ | |
62b5c94c | 4056 | self.deprecation_warning('YoutubeDL._opener is deprecated, use YoutubeDL.urlopen()') |
227bf1a3 | 4057 | handler = self._request_director.handlers['Urllib'] |
4058 | return handler._get_instance(cookiejar=self.cookiejar, proxies=self.proxies) | |
4059 | ||
c365dba8 | 4060 | def urlopen(self, req): |
4061 | """ Start an HTTP download """ | |
4062 | if isinstance(req, str): | |
227bf1a3 | 4063 | req = Request(req) |
4064 | elif isinstance(req, urllib.request.Request): | |
3d2623a8 | 4065 | self.deprecation_warning( |
4066 | 'Passing a urllib.request.Request object to YoutubeDL.urlopen() is deprecated. ' | |
4067 | 'Use yt_dlp.networking.common.Request instead.') | |
227bf1a3 | 4068 | req = urllib_req_to_req(req) |
4069 | assert isinstance(req, Request) | |
4070 | ||
4071 | # compat: Assume user:pass url params are basic auth | |
4072 | url, basic_auth_header = extract_basic_auth(req.url) | |
4073 | if basic_auth_header: | |
4074 | req.headers['Authorization'] = basic_auth_header | |
4075 | req.url = sanitize_url(url) | |
4076 | ||
4077 | clean_proxies(proxies=req.proxies, headers=req.headers) | |
4078 | clean_headers(req.headers) | |
4079 | ||
4080 | try: | |
4081 | return self._request_director.send(req) | |
4082 | except NoSupportingHandlers as e: | |
4083 | for ue in e.unsupported_errors: | |
ccfd70f4 | 4084 | # FIXME: This depends on the order of errors. |
227bf1a3 | 4085 | if not (ue.handler and ue.msg): |
4086 | continue | |
4087 | if ue.handler.RH_KEY == 'Urllib' and 'unsupported url scheme: "file"' in ue.msg.lower(): | |
4088 | raise RequestError( | |
4089 | 'file:// URLs are disabled by default in yt-dlp for security reasons. ' | |
4090 | 'Use --enable-file-urls to enable at your own risk.', cause=ue) from ue | |
8a8b5452 | 4091 | if 'unsupported proxy type: "https"' in ue.msg.lower(): |
4092 | raise RequestError( | |
4093 | 'To use an HTTPS proxy for this request, one of the following dependencies needs to be installed: requests') | |
ccfd70f4 | 4094 | |
4095 | elif ( | |
4096 | re.match(r'unsupported url scheme: "wss?"', ue.msg.lower()) | |
4097 | and 'websockets' not in self._request_director.handlers | |
4098 | ): | |
4099 | raise RequestError( | |
4100 | 'This request requires WebSocket support. ' | |
4101 | 'Ensure one of the following dependencies are installed: websockets', | |
4102 | cause=ue) from ue | |
227bf1a3 | 4103 | raise |
4104 | except SSLError as e: | |
4105 | if 'UNSAFE_LEGACY_RENEGOTIATION_DISABLED' in str(e): | |
4106 | raise RequestError('UNSAFE_LEGACY_RENEGOTIATION_DISABLED: Try using --legacy-server-connect', cause=e) from e | |
4107 | elif 'SSLV3_ALERT_HANDSHAKE_FAILURE' in str(e): | |
4108 | raise RequestError( | |
4109 | 'SSLV3_ALERT_HANDSHAKE_FAILURE: The server may not support the current cipher list. ' | |
4110 | 'Try using --legacy-server-connect', cause=e) from e | |
4111 | raise | |
227bf1a3 | 4112 | |
db7b054a | 4113 | def build_request_director(self, handlers, preferences=None): |
227bf1a3 | 4114 | logger = _YDLLogger(self) |
6148833f | 4115 | headers = self.params['http_headers'].copy() |
227bf1a3 | 4116 | proxies = self.proxies.copy() |
4117 | clean_headers(headers) | |
4118 | clean_proxies(proxies, headers) | |
4119 | ||
4120 | director = RequestDirector(logger=logger, verbose=self.params.get('debug_printtraffic')) | |
4121 | for handler in handlers: | |
4122 | director.add_handler(handler( | |
4123 | logger=logger, | |
4124 | headers=headers, | |
4125 | cookiejar=self.cookiejar, | |
4126 | proxies=proxies, | |
4127 | prefer_system_certs='no-certifi' in self.params['compat_opts'], | |
4128 | verify=not self.params.get('nocheckcertificate'), | |
4129 | **traverse_obj(self.params, { | |
4130 | 'verbose': 'debug_printtraffic', | |
4131 | 'source_address': 'source_address', | |
4132 | 'timeout': 'socket_timeout', | |
75dc8e67 | 4133 | 'legacy_ssl_support': 'legacyserverconnect', |
227bf1a3 | 4134 | 'enable_file_urls': 'enable_file_urls', |
4135 | 'client_cert': { | |
4136 | 'client_certificate': 'client_certificate', | |
4137 | 'client_certificate_key': 'client_certificate_key', | |
4138 | 'client_certificate_password': 'client_certificate_password', | |
4139 | }, | |
4140 | }), | |
4141 | )) | |
db7b054a | 4142 | director.preferences.update(preferences or []) |
8a8b5452 | 4143 | if 'prefer-legacy-http-handler' in self.params['compat_opts']: |
4144 | director.preferences.add(lambda rh, _: 500 if rh.RH_KEY == 'Urllib' else 0) | |
227bf1a3 | 4145 | return director |
c365dba8 | 4146 | |
62fec3b2 PH |
4147 | def encode(self, s): |
4148 | if isinstance(s, bytes): | |
4149 | return s # Already encoded | |
4150 | ||
4151 | try: | |
4152 | return s.encode(self.get_encoding()) | |
4153 | except UnicodeEncodeError as err: | |
4154 | err.reason = err.reason + '. Check your system encoding configuration or use the --encoding option.' | |
4155 | raise | |
4156 | ||
4157 | def get_encoding(self): | |
4158 | encoding = self.params.get('encoding') | |
4159 | if encoding is None: | |
4160 | encoding = preferredencoding() | |
4161 | return encoding | |
ec82d85a | 4162 | |
e08a85d8 | 4163 | def _write_info_json(self, label, ie_result, infofn, overwrite=None): |
cb96c5be | 4164 | ''' Write infojson and returns True = written, 'exists' = Already exists, False = skip, None = error ''' |
e08a85d8 | 4165 | if overwrite is None: |
4166 | overwrite = self.params.get('overwrites', True) | |
80c03fa9 | 4167 | if not self.params.get('writeinfojson'): |
4168 | return False | |
4169 | elif not infofn: | |
4170 | self.write_debug(f'Skipping writing {label} infojson') | |
4171 | return False | |
4172 | elif not self._ensure_dir_exists(infofn): | |
4173 | return None | |
e08a85d8 | 4174 | elif not overwrite and os.path.exists(infofn): |
80c03fa9 | 4175 | self.to_screen(f'[info] {label.title()} metadata is already present') |
cb96c5be | 4176 | return 'exists' |
4177 | ||
4178 | self.to_screen(f'[info] Writing {label} metadata as JSON to: {infofn}') | |
4179 | try: | |
4180 | write_json_file(self.sanitize_info(ie_result, self.params.get('clean_infojson', True)), infofn) | |
4181 | return True | |
86e5f3ed | 4182 | except OSError: |
cb96c5be | 4183 | self.report_error(f'Cannot write {label} metadata to JSON file {infofn}') |
4184 | return None | |
80c03fa9 | 4185 | |
4186 | def _write_description(self, label, ie_result, descfn): | |
4187 | ''' Write description and returns True = written, False = skip, None = error ''' | |
4188 | if not self.params.get('writedescription'): | |
4189 | return False | |
4190 | elif not descfn: | |
4191 | self.write_debug(f'Skipping writing {label} description') | |
4192 | return False | |
4193 | elif not self._ensure_dir_exists(descfn): | |
4194 | return None | |
4195 | elif not self.params.get('overwrites', True) and os.path.exists(descfn): | |
4196 | self.to_screen(f'[info] {label.title()} description is already present') | |
4197 | elif ie_result.get('description') is None: | |
88fb9425 | 4198 | self.to_screen(f'[info] There\'s no {label} description to write') |
80c03fa9 | 4199 | return False |
4200 | else: | |
4201 | try: | |
4202 | self.to_screen(f'[info] Writing {label} description to: {descfn}') | |
86e5f3ed | 4203 | with open(encodeFilename(descfn), 'w', encoding='utf-8') as descfile: |
80c03fa9 | 4204 | descfile.write(ie_result['description']) |
86e5f3ed | 4205 | except OSError: |
80c03fa9 | 4206 | self.report_error(f'Cannot write {label} description file {descfn}') |
4207 | return None | |
4208 | return True | |
4209 | ||
4210 | def _write_subtitles(self, info_dict, filename): | |
4211 | ''' Write subtitles to file and return list of (sub_filename, final_sub_filename); or None if error''' | |
4212 | ret = [] | |
4213 | subtitles = info_dict.get('requested_subtitles') | |
88fb9425 | 4214 | if not (self.params.get('writesubtitles') or self.params.get('writeautomaticsub')): |
80c03fa9 | 4215 | # subtitles download errors are already managed as troubles in relevant IE |
4216 | # that way it will silently go on when used with unsupporting IE | |
4217 | return ret | |
88fb9425 | 4218 | elif not subtitles: |
c8bc203f | 4219 | self.to_screen('[info] There are no subtitles for the requested languages') |
88fb9425 | 4220 | return ret |
80c03fa9 | 4221 | sub_filename_base = self.prepare_filename(info_dict, 'subtitle') |
4222 | if not sub_filename_base: | |
4223 | self.to_screen('[info] Skipping writing video subtitles') | |
4224 | return ret | |
88fb9425 | 4225 | |
80c03fa9 | 4226 | for sub_lang, sub_info in subtitles.items(): |
4227 | sub_format = sub_info['ext'] | |
4228 | sub_filename = subtitles_filename(filename, sub_lang, sub_format, info_dict.get('ext')) | |
4229 | sub_filename_final = subtitles_filename(sub_filename_base, sub_lang, sub_format, info_dict.get('ext')) | |
e04938ab | 4230 | existing_sub = self.existing_file((sub_filename_final, sub_filename)) |
4231 | if existing_sub: | |
80c03fa9 | 4232 | self.to_screen(f'[info] Video subtitle {sub_lang}.{sub_format} is already present') |
e04938ab | 4233 | sub_info['filepath'] = existing_sub |
4234 | ret.append((existing_sub, sub_filename_final)) | |
80c03fa9 | 4235 | continue |
4236 | ||
4237 | self.to_screen(f'[info] Writing video subtitles to: {sub_filename}') | |
4238 | if sub_info.get('data') is not None: | |
4239 | try: | |
4240 | # Use newline='' to prevent conversion of newline characters | |
4241 | # See https://github.com/ytdl-org/youtube-dl/issues/10268 | |
86e5f3ed | 4242 | with open(sub_filename, 'w', encoding='utf-8', newline='') as subfile: |
80c03fa9 | 4243 | subfile.write(sub_info['data']) |
4244 | sub_info['filepath'] = sub_filename | |
4245 | ret.append((sub_filename, sub_filename_final)) | |
4246 | continue | |
86e5f3ed | 4247 | except OSError: |
80c03fa9 | 4248 | self.report_error(f'Cannot write video subtitles file {sub_filename}') |
4249 | return None | |
4250 | ||
4251 | try: | |
4252 | sub_copy = sub_info.copy() | |
4253 | sub_copy.setdefault('http_headers', info_dict.get('http_headers')) | |
4254 | self.dl(sub_filename, sub_copy, subtitle=True) | |
4255 | sub_info['filepath'] = sub_filename | |
4256 | ret.append((sub_filename, sub_filename_final)) | |
6020e05d | 4257 | except (DownloadError, ExtractorError, IOError, OSError, ValueError) + network_exceptions as err: |
c70c418d | 4258 | msg = f'Unable to download video subtitles for {sub_lang!r}: {err}' |
6020e05d | 4259 | if self.params.get('ignoreerrors') is not True: # False or 'only_download' |
c70c418d | 4260 | if not self.params.get('ignoreerrors'): |
4261 | self.report_error(msg) | |
4262 | raise DownloadError(msg) | |
4263 | self.report_warning(msg) | |
519804a9 | 4264 | return ret |
80c03fa9 | 4265 | |
4266 | def _write_thumbnails(self, label, info_dict, filename, thumb_filename_base=None): | |
2acd1d55 | 4267 | ''' Write thumbnails to file and return list of (thumb_filename, final_thumb_filename); or None if error ''' |
6c4fd172 | 4268 | write_all = self.params.get('write_all_thumbnails', False) |
80c03fa9 | 4269 | thumbnails, ret = [], [] |
6c4fd172 | 4270 | if write_all or self.params.get('writethumbnail', False): |
0202b52a | 4271 | thumbnails = info_dict.get('thumbnails') or [] |
88fb9425 | 4272 | if not thumbnails: |
c8bc203f | 4273 | self.to_screen(f'[info] There are no {label} thumbnails to download') |
88fb9425 | 4274 | return ret |
6c4fd172 | 4275 | multiple = write_all and len(thumbnails) > 1 |
ec82d85a | 4276 | |
80c03fa9 | 4277 | if thumb_filename_base is None: |
4278 | thumb_filename_base = filename | |
4279 | if thumbnails and not thumb_filename_base: | |
4280 | self.write_debug(f'Skipping writing {label} thumbnail') | |
4281 | return ret | |
4282 | ||
a40e0b37 | 4283 | if thumbnails and not self._ensure_dir_exists(filename): |
2acd1d55 R |
4284 | return None |
4285 | ||
dd0228ce | 4286 | for idx, t in list(enumerate(thumbnails))[::-1]: |
80c03fa9 | 4287 | thumb_ext = (f'{t["id"]}.' if multiple else '') + determine_ext(t['url'], 'jpg') |
aa9369a2 | 4288 | thumb_display_id = f'{label} thumbnail {t["id"]}' |
80c03fa9 | 4289 | thumb_filename = replace_extension(filename, thumb_ext, info_dict.get('ext')) |
4290 | thumb_filename_final = replace_extension(thumb_filename_base, thumb_ext, info_dict.get('ext')) | |
ec82d85a | 4291 | |
e04938ab | 4292 | existing_thumb = self.existing_file((thumb_filename_final, thumb_filename)) |
4293 | if existing_thumb: | |
aa9369a2 | 4294 | self.to_screen('[info] %s is already present' % ( |
4295 | thumb_display_id if multiple else f'{label} thumbnail').capitalize()) | |
e04938ab | 4296 | t['filepath'] = existing_thumb |
4297 | ret.append((existing_thumb, thumb_filename_final)) | |
ec82d85a | 4298 | else: |
80c03fa9 | 4299 | self.to_screen(f'[info] Downloading {thumb_display_id} ...') |
ec82d85a | 4300 | try: |
227bf1a3 | 4301 | uf = self.urlopen(Request(t['url'], headers=t.get('http_headers', {}))) |
80c03fa9 | 4302 | self.to_screen(f'[info] Writing {thumb_display_id} to: {thumb_filename}') |
d3d89c32 | 4303 | with open(encodeFilename(thumb_filename), 'wb') as thumbf: |
ec82d85a | 4304 | shutil.copyfileobj(uf, thumbf) |
80c03fa9 | 4305 | ret.append((thumb_filename, thumb_filename_final)) |
885cc0b7 | 4306 | t['filepath'] = thumb_filename |
3158150c | 4307 | except network_exceptions as err: |
3d2623a8 | 4308 | if isinstance(err, HTTPError) and err.status == 404: |
ad54c913 | 4309 | self.to_screen(f'[info] {thumb_display_id.title()} does not exist') |
4310 | else: | |
4311 | self.report_warning(f'Unable to download {thumb_display_id}: {err}') | |
dd0228ce | 4312 | thumbnails.pop(idx) |
6c4fd172 | 4313 | if ret and not write_all: |
4314 | break | |
0202b52a | 4315 | return ret |