]>
Commit | Line | Data |
---|---|---|
8222d8de | 1 | #!/usr/bin/env python |
dcdb292f | 2 | # coding: utf-8 |
8222d8de | 3 | |
6febd1c1 | 4 | from __future__ import absolute_import, unicode_literals |
8222d8de | 5 | |
26e63931 | 6 | import collections |
31bd3925 | 7 | import contextlib |
317f7ab6 | 8 | import copy |
9d2ecdbc | 9 | import datetime |
c1c9a79c | 10 | import errno |
31bd3925 | 11 | import fileinput |
8222d8de | 12 | import io |
b82f815f | 13 | import itertools |
8694c600 | 14 | import json |
62fec3b2 | 15 | import locale |
083c9df9 | 16 | import operator |
8222d8de | 17 | import os |
dca08720 | 18 | import platform |
8222d8de JMF |
19 | import re |
20 | import shutil | |
dca08720 | 21 | import subprocess |
8222d8de JMF |
22 | import socket |
23 | import sys | |
24 | import time | |
67134eab | 25 | import tokenize |
8222d8de | 26 | import traceback |
75822ca7 | 27 | import random |
8222d8de | 28 | |
961ea474 | 29 | from string import ascii_letters |
e5813e53 | 30 | from zipimport import zipimporter |
961ea474 | 31 | |
8c25f81b | 32 | from .compat import ( |
82d8a8b6 | 33 | compat_basestring, |
dca08720 | 34 | compat_cookiejar, |
003c69a8 | 35 | compat_get_terminal_size, |
ce02ed60 | 36 | compat_http_client, |
4f026faf | 37 | compat_kwargs, |
d0d9ade4 | 38 | compat_numeric_types, |
e9c0cdd3 | 39 | compat_os_name, |
ce02ed60 | 40 | compat_str, |
67134eab | 41 | compat_tokenize_tokenize, |
ce02ed60 PH |
42 | compat_urllib_error, |
43 | compat_urllib_request, | |
8b172c2e | 44 | compat_urllib_request_DataHandler, |
8c25f81b PH |
45 | ) |
46 | from .utils import ( | |
eedb7ba5 S |
47 | age_restricted, |
48 | args_to_str, | |
ce02ed60 PH |
49 | ContentTooShortError, |
50 | date_from_str, | |
51 | DateRange, | |
acd69589 | 52 | DEFAULT_OUTTMPL, |
de6000d9 | 53 | OUTTMPL_TYPES, |
ce02ed60 | 54 | determine_ext, |
b5559424 | 55 | determine_protocol, |
732044af | 56 | DOT_DESKTOP_LINK_TEMPLATE, |
57 | DOT_URL_LINK_TEMPLATE, | |
58 | DOT_WEBLOC_LINK_TEMPLATE, | |
ce02ed60 | 59 | DownloadError, |
c0384f22 | 60 | encode_compat_str, |
ce02ed60 | 61 | encodeFilename, |
9b9c5355 | 62 | error_to_compat_str, |
498f5606 | 63 | EntryNotInPlaylist, |
8b0d7497 | 64 | ExistingVideoReached, |
590bc6f6 | 65 | expand_path, |
ce02ed60 | 66 | ExtractorError, |
e29663c6 | 67 | float_or_none, |
02dbf93f | 68 | format_bytes, |
76d321f6 | 69 | format_field, |
143db31d | 70 | FORMAT_RE, |
525ef922 | 71 | formatSeconds, |
773f291d | 72 | GeoRestrictedError, |
c9969434 | 73 | int_or_none, |
732044af | 74 | iri_to_uri, |
773f291d | 75 | ISO3166Utils, |
ce02ed60 | 76 | locked_file, |
0202b52a | 77 | make_dir, |
dca08720 | 78 | make_HTTPS_handler, |
ce02ed60 | 79 | MaxDownloadsReached, |
cd6fc19e | 80 | orderedSet, |
b7ab0590 | 81 | PagedList, |
083c9df9 | 82 | parse_filesize, |
91410c9b | 83 | PerRequestProxyHandler, |
dca08720 | 84 | platform_name, |
eedb7ba5 | 85 | PostProcessingError, |
ce02ed60 | 86 | preferredencoding, |
eedb7ba5 | 87 | prepend_extension, |
51fb4995 | 88 | register_socks_protocols, |
cfb56d1a | 89 | render_table, |
eedb7ba5 | 90 | replace_extension, |
8b0d7497 | 91 | RejectedVideoReached, |
ce02ed60 PH |
92 | SameFileError, |
93 | sanitize_filename, | |
1bb5c511 | 94 | sanitize_path, |
dcf77cf1 | 95 | sanitize_url, |
67dda517 | 96 | sanitized_Request, |
e5660ee6 | 97 | std_headers, |
1211bb6d | 98 | str_or_none, |
e29663c6 | 99 | strftime_or_none, |
ce02ed60 | 100 | subtitles_filename, |
732044af | 101 | to_high_limit_path, |
ce02ed60 | 102 | UnavailableVideoError, |
29eb5174 | 103 | url_basename, |
58b1f00d | 104 | version_tuple, |
ce02ed60 PH |
105 | write_json_file, |
106 | write_string, | |
1bab3437 | 107 | YoutubeDLCookieJar, |
6a3f4c3f | 108 | YoutubeDLCookieProcessor, |
dca08720 | 109 | YoutubeDLHandler, |
fca6dba8 | 110 | YoutubeDLRedirectHandler, |
f5b1bca9 | 111 | process_communicate_or_kill, |
ce02ed60 | 112 | ) |
a0e07d31 | 113 | from .cache import Cache |
52a8a1e1 | 114 | from .extractor import ( |
115 | gen_extractor_classes, | |
116 | get_info_extractor, | |
117 | _LAZY_LOADER, | |
118 | _PLUGIN_CLASSES | |
119 | ) | |
4c54b89e | 120 | from .extractor.openload import PhantomJSwrapper |
52a8a1e1 | 121 | from .downloader import ( |
122 | get_suitable_downloader, | |
123 | shorten_protocol_name | |
124 | ) | |
4c83c967 | 125 | from .downloader.rtmp import rtmpdump_version |
4f026faf | 126 | from .postprocessor import ( |
f17f8651 | 127 | FFmpegFixupM3u8PP, |
62cd676c | 128 | FFmpegFixupM4aPP, |
6271f1ca | 129 | FFmpegFixupStretchedPP, |
4f026faf PH |
130 | FFmpegMergerPP, |
131 | FFmpegPostProcessor, | |
0202b52a | 132 | # FFmpegSubtitlesConvertorPP, |
4f026faf | 133 | get_postprocessor, |
0202b52a | 134 | MoveFilesAfterDownloadPP, |
4f026faf | 135 | ) |
dca08720 | 136 | from .version import __version__ |
8222d8de | 137 | |
e9c0cdd3 YCH |
138 | if compat_os_name == 'nt': |
139 | import ctypes | |
140 | ||
2459b6e1 | 141 | |
8222d8de JMF |
142 | class YoutubeDL(object): |
143 | """YoutubeDL class. | |
144 | ||
145 | YoutubeDL objects are the ones responsible of downloading the | |
146 | actual video file and writing it to disk if the user has requested | |
147 | it, among some other tasks. In most cases there should be one per | |
148 | program. As, given a video URL, the downloader doesn't know how to | |
149 | extract all the needed information, task that InfoExtractors do, it | |
150 | has to pass the URL to one of them. | |
151 | ||
152 | For this, YoutubeDL objects have a method that allows | |
153 | InfoExtractors to be registered in a given order. When it is passed | |
154 | a URL, the YoutubeDL object handles it to the first InfoExtractor it | |
155 | finds that reports being able to handle it. The InfoExtractor extracts | |
156 | all the information about the video or videos the URL refers to, and | |
157 | YoutubeDL process the extracted information, possibly using a File | |
158 | Downloader to download the video. | |
159 | ||
160 | YoutubeDL objects accept a lot of parameters. In order not to saturate | |
161 | the object constructor with arguments, it receives a dictionary of | |
162 | options instead. These options are available through the params | |
163 | attribute for the InfoExtractors to use. The YoutubeDL also | |
164 | registers itself as the downloader in charge for the InfoExtractors | |
165 | that are added to it, so this is a "mutual registration". | |
166 | ||
167 | Available options: | |
168 | ||
169 | username: Username for authentication purposes. | |
170 | password: Password for authentication purposes. | |
180940e0 | 171 | videopassword: Password for accessing a video. |
1da50aa3 S |
172 | ap_mso: Adobe Pass multiple-system operator identifier. |
173 | ap_username: Multiple-system operator account username. | |
174 | ap_password: Multiple-system operator account password. | |
8222d8de JMF |
175 | usenetrc: Use netrc for authentication instead. |
176 | verbose: Print additional info to stdout. | |
177 | quiet: Do not print messages to stdout. | |
ad8915b7 | 178 | no_warnings: Do not print out anything for warnings. |
8222d8de JMF |
179 | forceurl: Force printing final URL. |
180 | forcetitle: Force printing title. | |
181 | forceid: Force printing ID. | |
182 | forcethumbnail: Force printing thumbnail URL. | |
183 | forcedescription: Force printing description. | |
184 | forcefilename: Force printing final filename. | |
525ef922 | 185 | forceduration: Force printing duration. |
8694c600 | 186 | forcejson: Force printing info_dict as JSON. |
63e0be34 PH |
187 | dump_single_json: Force printing the info_dict of the whole playlist |
188 | (or video) as a single JSON line. | |
c25228e5 | 189 | force_write_download_archive: Force writing download archive regardless |
190 | of 'skip_download' or 'simulate'. | |
8222d8de | 191 | simulate: Do not download the video files. |
eb8a4433 | 192 | format: Video format code. see "FORMAT SELECTION" for more details. |
63ad4d43 | 193 | allow_unplayable_formats: Allow unplayable formats to be extracted and downloaded. |
c25228e5 | 194 | format_sort: How to sort the video formats. see "Sorting Formats" |
195 | for more details. | |
196 | format_sort_force: Force the given format_sort. see "Sorting Formats" | |
197 | for more details. | |
198 | allow_multiple_video_streams: Allow multiple video streams to be merged | |
199 | into a single file | |
200 | allow_multiple_audio_streams: Allow multiple audio streams to be merged | |
201 | into a single file | |
4524baf0 | 202 | paths: Dictionary of output paths. The allowed keys are 'home' |
203 | 'temp' and the keys of OUTTMPL_TYPES (in utils.py) | |
de6000d9 | 204 | outtmpl: Dictionary of templates for output names. Allowed keys |
4524baf0 | 205 | are 'default' and the keys of OUTTMPL_TYPES (in utils.py). |
206 | A string a also accepted for backward compatibility | |
a820dc72 RA |
207 | outtmpl_na_placeholder: Placeholder for unavailable meta fields. |
208 | restrictfilenames: Do not allow "&" and spaces in file names | |
209 | trim_file_name: Limit length of filename (extension excluded) | |
4524baf0 | 210 | windowsfilenames: Force the filenames to be windows compatible |
a820dc72 | 211 | ignoreerrors: Do not stop on download errors |
7a5c1cfe | 212 | (Default True when running yt-dlp, |
a820dc72 | 213 | but False when directly accessing YoutubeDL class) |
d22dec74 | 214 | force_generic_extractor: Force downloader to use the generic extractor |
0c3d0f51 | 215 | overwrites: Overwrite all video and metadata files if True, |
216 | overwrite only non-video files if None | |
217 | and don't overwrite any file if False | |
8222d8de JMF |
218 | playliststart: Playlist item to start at. |
219 | playlistend: Playlist item to end at. | |
c14e88f0 | 220 | playlist_items: Specific indices of playlist to download. |
ff815fe6 | 221 | playlistreverse: Download playlist items in reverse order. |
75822ca7 | 222 | playlistrandom: Download playlist items in random order. |
8222d8de JMF |
223 | matchtitle: Download only matching titles. |
224 | rejecttitle: Reject downloads for matching titles. | |
8bf9319e | 225 | logger: Log messages to a logging.Logger instance. |
8222d8de JMF |
226 | logtostderr: Log messages to stderr instead of stdout. |
227 | writedescription: Write the video description to a .description file | |
228 | writeinfojson: Write the video description to a .info.json file | |
75d43ca0 | 229 | clean_infojson: Remove private fields from the infojson |
06167fbb | 230 | writecomments: Extract video comments. This will not be written to disk |
231 | unless writeinfojson is also given | |
1fb07d10 | 232 | writeannotations: Write the video annotations to a .annotations.xml file |
8222d8de | 233 | writethumbnail: Write the thumbnail image to a file |
c25228e5 | 234 | allow_playlist_files: Whether to write playlists' description, infojson etc |
235 | also to disk when using the 'write*' options | |
ec82d85a | 236 | write_all_thumbnails: Write all thumbnail formats to files |
732044af | 237 | writelink: Write an internet shortcut file, depending on the |
238 | current platform (.url/.webloc/.desktop) | |
239 | writeurllink: Write a Windows internet shortcut file (.url) | |
240 | writewebloclink: Write a macOS internet shortcut file (.webloc) | |
241 | writedesktoplink: Write a Linux internet shortcut file (.desktop) | |
8222d8de | 242 | writesubtitles: Write the video subtitles to a file |
741dd8ea | 243 | writeautomaticsub: Write the automatically generated subtitles to a file |
8222d8de | 244 | allsubtitles: Downloads all the subtitles of the video |
0b7f3118 | 245 | (requires writesubtitles or writeautomaticsub) |
8222d8de | 246 | listsubtitles: Lists all available subtitles for the video |
a504ced0 | 247 | subtitlesformat: The format code for subtitles |
aa6a10c4 | 248 | subtitleslangs: List of languages of the subtitles to download |
8222d8de JMF |
249 | keepvideo: Keep the video file after post-processing |
250 | daterange: A DateRange object, download only if the upload_date is in the range. | |
251 | skip_download: Skip the actual download of the video file | |
c35f9e72 | 252 | cachedir: Location of the cache files in the filesystem. |
a0e07d31 | 253 | False to disable filesystem cache. |
47192f92 | 254 | noplaylist: Download single video instead of a playlist if in doubt. |
8dbe9899 PH |
255 | age_limit: An integer representing the user's age in years. |
256 | Unsuitable videos for the given age are skipped. | |
5fe18bdb PH |
257 | min_views: An integer representing the minimum view count the video |
258 | must have in order to not be skipped. | |
259 | Videos without view count information are always | |
260 | downloaded. None for no limit. | |
261 | max_views: An integer representing the maximum view count. | |
262 | Videos that are more popular than that are not | |
263 | downloaded. | |
264 | Videos without view count information are always | |
265 | downloaded. None for no limit. | |
266 | download_archive: File name of a file where all downloads are recorded. | |
c1c9a79c PH |
267 | Videos already present in the file are not downloaded |
268 | again. | |
8a51f564 | 269 | break_on_existing: Stop the download process after attempting to download a |
270 | file that is in the archive. | |
271 | break_on_reject: Stop the download process when encountering a video that | |
272 | has been filtered out. | |
273 | cookiefile: File name where cookies should be read from and dumped to | |
a1ee09e8 | 274 | nocheckcertificate:Do not verify SSL certificates |
7e8c0af0 PH |
275 | prefer_insecure: Use HTTP instead of HTTPS to retrieve information. |
276 | At the moment, this is only supported by YouTube. | |
a1ee09e8 | 277 | proxy: URL of the proxy server to use |
38cce791 | 278 | geo_verification_proxy: URL of the proxy to use for IP address verification |
504f20dd | 279 | on geo-restricted sites. |
e344693b | 280 | socket_timeout: Time to wait for unresponsive hosts, in seconds |
0783b09b PH |
281 | bidi_workaround: Work around buggy terminals without bidirectional text |
282 | support, using fridibi | |
a0ddb8a2 | 283 | debug_printtraffic:Print out sent and received HTTP traffic |
7b0817e8 | 284 | include_ads: Download ads as well |
04b4d394 PH |
285 | default_search: Prepend this string if an input url is not valid. |
286 | 'auto' for elaborate guessing | |
62fec3b2 | 287 | encoding: Use this encoding instead of the system-specified. |
e8ee972c | 288 | extract_flat: Do not resolve URLs, return the immediate result. |
057a5206 PH |
289 | Pass in 'in_playlist' to only show this behavior for |
290 | playlist items. | |
4f026faf | 291 | postprocessors: A list of dictionaries, each with an entry |
71b640cc | 292 | * key: The name of the postprocessor. See |
7a5c1cfe | 293 | yt_dlp/postprocessor/__init__.py for a list. |
56d868db | 294 | * when: When to run the postprocessor. Can be one of |
295 | pre_process|before_dl|post_process|after_move. | |
296 | Assumed to be 'post_process' if not given | |
ab8e5e51 AM |
297 | post_hooks: A list of functions that get called as the final step |
298 | for each video file, after all postprocessors have been | |
299 | called. The filename will be passed as the only argument. | |
71b640cc PH |
300 | progress_hooks: A list of functions that get called on download |
301 | progress, with a dictionary with the entries | |
5cda4eda | 302 | * status: One of "downloading", "error", or "finished". |
ee69b99a | 303 | Check this first and ignore unknown values. |
71b640cc | 304 | |
5cda4eda | 305 | If status is one of "downloading", or "finished", the |
ee69b99a PH |
306 | following properties may also be present: |
307 | * filename: The final filename (always present) | |
5cda4eda | 308 | * tmpfilename: The filename we're currently writing to |
71b640cc PH |
309 | * downloaded_bytes: Bytes on disk |
310 | * total_bytes: Size of the whole file, None if unknown | |
5cda4eda PH |
311 | * total_bytes_estimate: Guess of the eventual file size, |
312 | None if unavailable. | |
313 | * elapsed: The number of seconds since download started. | |
71b640cc PH |
314 | * eta: The estimated time in seconds, None if unknown |
315 | * speed: The download speed in bytes/second, None if | |
316 | unknown | |
5cda4eda PH |
317 | * fragment_index: The counter of the currently |
318 | downloaded video fragment. | |
319 | * fragment_count: The number of fragments (= individual | |
320 | files that will be merged) | |
71b640cc PH |
321 | |
322 | Progress hooks are guaranteed to be called at least once | |
323 | (with status "finished") if the download is successful. | |
45598f15 | 324 | merge_output_format: Extension to use when merging formats. |
6b591b29 | 325 | final_ext: Expected final extension; used to detect when the file was |
326 | already downloaded and converted. "merge_output_format" is | |
327 | replaced by this extension when given | |
6271f1ca PH |
328 | fixup: Automatically correct known faults of the file. |
329 | One of: | |
330 | - "never": do nothing | |
331 | - "warn": only emit a warning | |
332 | - "detect_or_warn": check whether we can do anything | |
62cd676c | 333 | about it, warn otherwise (default) |
504f20dd | 334 | source_address: Client-side IP address to bind to. |
6ec6cb4e | 335 | call_home: Boolean, true iff we are allowed to contact the |
7a5c1cfe | 336 | yt-dlp servers for debugging. (BROKEN) |
1cf376f5 | 337 | sleep_interval_requests: Number of seconds to sleep between requests |
338 | during extraction | |
7aa589a5 S |
339 | sleep_interval: Number of seconds to sleep before each download when |
340 | used alone or a lower bound of a range for randomized | |
341 | sleep before each download (minimum possible number | |
342 | of seconds to sleep) when used along with | |
343 | max_sleep_interval. | |
344 | max_sleep_interval:Upper bound of a range for randomized sleep before each | |
345 | download (maximum possible number of seconds to sleep). | |
346 | Must only be used along with sleep_interval. | |
347 | Actual sleep time will be a random float from range | |
348 | [sleep_interval; max_sleep_interval]. | |
1cf376f5 | 349 | sleep_interval_subtitles: Number of seconds to sleep before each subtitle download |
cfb56d1a PH |
350 | listformats: Print an overview of available video formats and exit. |
351 | list_thumbnails: Print a table of all thumbnails and exit. | |
347de493 PH |
352 | match_filter: A function that gets called with the info_dict of |
353 | every video. | |
354 | If it returns a message, the video is ignored. | |
355 | If it returns None, the video is downloaded. | |
356 | match_filter_func in utils.py is one example for this. | |
7e5db8c9 | 357 | no_color: Do not emit color codes in output. |
0a840f58 | 358 | geo_bypass: Bypass geographic restriction via faking X-Forwarded-For |
504f20dd | 359 | HTTP header |
0a840f58 | 360 | geo_bypass_country: |
773f291d S |
361 | Two-letter ISO 3166-2 country code that will be used for |
362 | explicit geographic restriction bypassing via faking | |
504f20dd | 363 | X-Forwarded-For HTTP header |
5f95927a S |
364 | geo_bypass_ip_block: |
365 | IP range in CIDR notation that will be used similarly to | |
504f20dd | 366 | geo_bypass_country |
71b640cc | 367 | |
85729c51 | 368 | The following options determine which downloader is picked: |
52a8a1e1 | 369 | external_downloader: A dictionary of protocol keys and the executable of the |
370 | external downloader to use for it. The allowed protocols | |
371 | are default|http|ftp|m3u8|dash|rtsp|rtmp|mms. | |
372 | Set the value to 'native' to use the native downloader | |
373 | hls_prefer_native: Deprecated - Use external_downloader = {'m3u8': 'native'} | |
374 | or {'m3u8': 'ffmpeg'} instead. | |
375 | Use the native HLS downloader instead of ffmpeg/avconv | |
bf09af3a S |
376 | if True, otherwise use ffmpeg/avconv if False, otherwise |
377 | use downloader suggested by extractor if None. | |
fe7e0c98 | 378 | |
8222d8de | 379 | The following parameters are not used by YoutubeDL itself, they are used by |
7a5c1cfe | 380 | the downloader (see yt_dlp/downloader/common.py): |
8222d8de | 381 | nopart, updatetime, buffersize, ratelimit, min_filesize, max_filesize, test, |
881e6a1f | 382 | noresizebuffer, retries, continuedl, noprogress, consoletitle, |
b54d4a5c | 383 | xattr_set_filesize, external_downloader_args, hls_use_mpegts, |
e409895f | 384 | http_chunk_size. |
76b1bd67 JMF |
385 | |
386 | The following options are used by the post processors: | |
d4a24f40 | 387 | prefer_ffmpeg: If False, use avconv instead of ffmpeg if both are available, |
e4172ac9 | 388 | otherwise prefer ffmpeg. (avconv support is deprecated) |
c0b7d117 S |
389 | ffmpeg_location: Location of the ffmpeg/avconv binary; either the path |
390 | to the binary or its containing directory. | |
43820c03 | 391 | postprocessor_args: A dictionary of postprocessor/executable keys (in lower case) |
392 | and a list of additional command-line arguments for the | |
393 | postprocessor/executable. The dict can also have "PP+EXE" keys | |
394 | which are used when the given exe is used by the given PP. | |
395 | Use 'default' as the name for arguments to passed to all PP | |
e409895f | 396 | |
397 | The following options are used by the extractors: | |
62bff2c1 | 398 | extractor_retries: Number of times to retry for known errors |
399 | dynamic_mpd: Whether to process dynamic DASH manifests (default: True) | |
e409895f | 400 | hls_split_discontinuity: Split HLS playlists to different formats at |
62bff2c1 | 401 | discontinuities such as ad breaks (default: False) |
3600fd59 | 402 | youtube_include_dash_manifest: If True (default), DASH manifests and related |
62bff2c1 | 403 | data will be downloaded and processed by extractor. |
404 | You can reduce network I/O by disabling it if you don't | |
405 | care about DASH. (only for youtube) | |
e409895f | 406 | youtube_include_hls_manifest: If True (default), HLS manifests and related |
62bff2c1 | 407 | data will be downloaded and processed by extractor. |
408 | You can reduce network I/O by disabling it if you don't | |
409 | care about HLS. (only for youtube) | |
8222d8de JMF |
410 | """ |
411 | ||
c9969434 S |
412 | _NUMERIC_FIELDS = set(( |
413 | 'width', 'height', 'tbr', 'abr', 'asr', 'vbr', 'fps', 'filesize', 'filesize_approx', | |
414 | 'timestamp', 'upload_year', 'upload_month', 'upload_day', | |
415 | 'duration', 'view_count', 'like_count', 'dislike_count', 'repost_count', | |
416 | 'average_rating', 'comment_count', 'age_limit', | |
417 | 'start_time', 'end_time', | |
418 | 'chapter_number', 'season_number', 'episode_number', | |
419 | 'track_number', 'disc_number', 'release_year', | |
420 | 'playlist_index', | |
421 | )) | |
422 | ||
8222d8de JMF |
423 | params = None |
424 | _ies = [] | |
56d868db | 425 | _pps = {'pre_process': [], 'before_dl': [], 'after_move': [], 'post_process': []} |
0202b52a | 426 | __prepare_filename_warned = False |
1cf376f5 | 427 | _first_webpage_request = True |
8222d8de JMF |
428 | _download_retcode = None |
429 | _num_downloads = None | |
30a074c2 | 430 | _playlist_level = 0 |
431 | _playlist_urls = set() | |
8222d8de JMF |
432 | _screen_file = None |
433 | ||
3511266b | 434 | def __init__(self, params=None, auto_init=True): |
8222d8de | 435 | """Create a FileDownloader object with the given options.""" |
e9f9a10f JMF |
436 | if params is None: |
437 | params = {} | |
8222d8de | 438 | self._ies = [] |
56c73665 | 439 | self._ies_instances = {} |
56d868db | 440 | self._pps = {'pre_process': [], 'before_dl': [], 'after_move': [], 'post_process': []} |
0202b52a | 441 | self.__prepare_filename_warned = False |
1cf376f5 | 442 | self._first_webpage_request = True |
ab8e5e51 | 443 | self._post_hooks = [] |
933605d7 | 444 | self._progress_hooks = [] |
8222d8de JMF |
445 | self._download_retcode = 0 |
446 | self._num_downloads = 0 | |
447 | self._screen_file = [sys.stdout, sys.stderr][params.get('logtostderr', False)] | |
0783b09b | 448 | self._err_file = sys.stderr |
4abf617b S |
449 | self.params = { |
450 | # Default parameters | |
451 | 'nocheckcertificate': False, | |
452 | } | |
453 | self.params.update(params) | |
a0e07d31 | 454 | self.cache = Cache(self) |
a45e8619 | 455 | self.archive = set() |
ecdec191 JB |
456 | |
457 | """Preload the archive, if any is specified""" | |
458 | def preload_download_archive(self): | |
459 | fn = self.params.get('download_archive') | |
460 | if fn is None: | |
461 | return False | |
462 | try: | |
463 | with locked_file(fn, 'r', encoding='utf-8') as archive_file: | |
464 | for line in archive_file: | |
a45e8619 | 465 | self.archive.add(line.strip()) |
ecdec191 JB |
466 | except IOError as ioe: |
467 | if ioe.errno != errno.ENOENT: | |
468 | raise | |
1d74d8d9 | 469 | return False |
ecdec191 | 470 | return True |
34308b30 | 471 | |
be5df5ee S |
472 | def check_deprecated(param, option, suggestion): |
473 | if self.params.get(param) is not None: | |
474 | self.report_warning( | |
475 | '%s is deprecated. Use %s instead.' % (option, suggestion)) | |
476 | return True | |
477 | return False | |
478 | ||
1de7ea76 JB |
479 | if self.params.get('verbose'): |
480 | self.to_stdout('[debug] Loading archive file %r' % self.params.get('download_archive')) | |
481 | ||
ecdec191 JB |
482 | preload_download_archive(self) |
483 | ||
be5df5ee | 484 | if check_deprecated('cn_verification_proxy', '--cn-verification-proxy', '--geo-verification-proxy'): |
38cce791 YCH |
485 | if self.params.get('geo_verification_proxy') is None: |
486 | self.params['geo_verification_proxy'] = self.params['cn_verification_proxy'] | |
487 | ||
6b591b29 | 488 | if self.params.get('final_ext'): |
489 | if self.params.get('merge_output_format'): | |
490 | self.report_warning('--merge-output-format will be ignored since --remux-video or --recode-video is given') | |
491 | self.params['merge_output_format'] = self.params['final_ext'] | |
492 | ||
b9d973be | 493 | if 'overwrites' in self.params and self.params['overwrites'] is None: |
494 | del self.params['overwrites'] | |
495 | ||
be5df5ee S |
496 | check_deprecated('autonumber_size', '--autonumber-size', 'output template with %(autonumber)0Nd, where N in the number of digits') |
497 | check_deprecated('autonumber', '--auto-number', '-o "%(autonumber)s-%(title)s.%(ext)s"') | |
498 | check_deprecated('usetitle', '--title', '-o "%(title)s-%(id)s.%(ext)s"') | |
499 | ||
0783b09b | 500 | if params.get('bidi_workaround', False): |
1c088fa8 PH |
501 | try: |
502 | import pty | |
503 | master, slave = pty.openpty() | |
003c69a8 | 504 | width = compat_get_terminal_size().columns |
1c088fa8 PH |
505 | if width is None: |
506 | width_args = [] | |
507 | else: | |
508 | width_args = ['-w', str(width)] | |
5d681e96 | 509 | sp_kwargs = dict( |
1c088fa8 PH |
510 | stdin=subprocess.PIPE, |
511 | stdout=slave, | |
512 | stderr=self._err_file) | |
5d681e96 PH |
513 | try: |
514 | self._output_process = subprocess.Popen( | |
515 | ['bidiv'] + width_args, **sp_kwargs | |
516 | ) | |
517 | except OSError: | |
5d681e96 PH |
518 | self._output_process = subprocess.Popen( |
519 | ['fribidi', '-c', 'UTF-8'] + width_args, **sp_kwargs) | |
520 | self._output_channel = os.fdopen(master, 'rb') | |
1c088fa8 | 521 | except OSError as ose: |
66e7ace1 | 522 | if ose.errno == errno.ENOENT: |
6febd1c1 | 523 | self.report_warning('Could not find fribidi executable, ignoring --bidi-workaround . Make sure that fribidi is an executable file in one of the directories in your $PATH.') |
1c088fa8 PH |
524 | else: |
525 | raise | |
0783b09b | 526 | |
3089bc74 S |
527 | if (sys.platform != 'win32' |
528 | and sys.getfilesystemencoding() in ['ascii', 'ANSI_X3.4-1968'] | |
529 | and not params.get('restrictfilenames', False)): | |
e9137224 | 530 | # Unicode filesystem API will throw errors (#1474, #13027) |
34308b30 | 531 | self.report_warning( |
6febd1c1 | 532 | 'Assuming --restrict-filenames since file system encoding ' |
1b725173 | 533 | 'cannot encode all characters. ' |
6febd1c1 | 534 | 'Set the LC_ALL environment variable to fix this.') |
4a98cdbf | 535 | self.params['restrictfilenames'] = True |
34308b30 | 536 | |
de6000d9 | 537 | self.outtmpl_dict = self.parse_outtmpl() |
486dd09e | 538 | |
dca08720 PH |
539 | self._setup_opener() |
540 | ||
3511266b PH |
541 | if auto_init: |
542 | self.print_debug_header() | |
543 | self.add_default_info_extractors() | |
544 | ||
4f026faf PH |
545 | for pp_def_raw in self.params.get('postprocessors', []): |
546 | pp_class = get_postprocessor(pp_def_raw['key']) | |
547 | pp_def = dict(pp_def_raw) | |
548 | del pp_def['key'] | |
5bfa4862 | 549 | if 'when' in pp_def: |
550 | when = pp_def['when'] | |
551 | del pp_def['when'] | |
552 | else: | |
56d868db | 553 | when = 'post_process' |
4f026faf | 554 | pp = pp_class(self, **compat_kwargs(pp_def)) |
5bfa4862 | 555 | self.add_post_processor(pp, when=when) |
4f026faf | 556 | |
ab8e5e51 AM |
557 | for ph in self.params.get('post_hooks', []): |
558 | self.add_post_hook(ph) | |
559 | ||
71b640cc PH |
560 | for ph in self.params.get('progress_hooks', []): |
561 | self.add_progress_hook(ph) | |
562 | ||
51fb4995 YCH |
563 | register_socks_protocols() |
564 | ||
7d4111ed PH |
565 | def warn_if_short_id(self, argv): |
566 | # short YouTube ID starting with dash? | |
567 | idxs = [ | |
568 | i for i, a in enumerate(argv) | |
569 | if re.match(r'^-[0-9A-Za-z_-]{10}$', a)] | |
570 | if idxs: | |
571 | correct_argv = ( | |
7a5c1cfe | 572 | ['yt-dlp'] |
3089bc74 S |
573 | + [a for i, a in enumerate(argv) if i not in idxs] |
574 | + ['--'] + [argv[i] for i in idxs] | |
7d4111ed PH |
575 | ) |
576 | self.report_warning( | |
577 | 'Long argument string detected. ' | |
578 | 'Use -- to separate parameters and URLs, like this:\n%s\n' % | |
579 | args_to_str(correct_argv)) | |
580 | ||
8222d8de JMF |
581 | def add_info_extractor(self, ie): |
582 | """Add an InfoExtractor object to the end of the list.""" | |
583 | self._ies.append(ie) | |
e52d7f85 JMF |
584 | if not isinstance(ie, type): |
585 | self._ies_instances[ie.ie_key()] = ie | |
586 | ie.set_downloader(self) | |
8222d8de | 587 | |
56c73665 JMF |
588 | def get_info_extractor(self, ie_key): |
589 | """ | |
590 | Get an instance of an IE with name ie_key, it will try to get one from | |
591 | the _ies list, if there's no instance it will create a new one and add | |
592 | it to the extractor list. | |
593 | """ | |
594 | ie = self._ies_instances.get(ie_key) | |
595 | if ie is None: | |
596 | ie = get_info_extractor(ie_key)() | |
597 | self.add_info_extractor(ie) | |
598 | return ie | |
599 | ||
023fa8c4 JMF |
600 | def add_default_info_extractors(self): |
601 | """ | |
602 | Add the InfoExtractors returned by gen_extractors to the end of the list | |
603 | """ | |
e52d7f85 | 604 | for ie in gen_extractor_classes(): |
023fa8c4 JMF |
605 | self.add_info_extractor(ie) |
606 | ||
56d868db | 607 | def add_post_processor(self, pp, when='post_process'): |
8222d8de | 608 | """Add a PostProcessor object to the end of the chain.""" |
5bfa4862 | 609 | self._pps[when].append(pp) |
8222d8de JMF |
610 | pp.set_downloader(self) |
611 | ||
ab8e5e51 AM |
612 | def add_post_hook(self, ph): |
613 | """Add the post hook""" | |
614 | self._post_hooks.append(ph) | |
615 | ||
933605d7 JMF |
616 | def add_progress_hook(self, ph): |
617 | """Add the progress hook (currently only for the file downloader)""" | |
618 | self._progress_hooks.append(ph) | |
8ab470f1 | 619 | |
1c088fa8 | 620 | def _bidi_workaround(self, message): |
5d681e96 | 621 | if not hasattr(self, '_output_channel'): |
1c088fa8 PH |
622 | return message |
623 | ||
5d681e96 | 624 | assert hasattr(self, '_output_process') |
11b85ce6 | 625 | assert isinstance(message, compat_str) |
6febd1c1 PH |
626 | line_count = message.count('\n') + 1 |
627 | self._output_process.stdin.write((message + '\n').encode('utf-8')) | |
5d681e96 | 628 | self._output_process.stdin.flush() |
6febd1c1 | 629 | res = ''.join(self._output_channel.readline().decode('utf-8') |
9e1a5b84 | 630 | for _ in range(line_count)) |
6febd1c1 | 631 | return res[:-len('\n')] |
1c088fa8 | 632 | |
8222d8de | 633 | def to_screen(self, message, skip_eol=False): |
0783b09b PH |
634 | """Print message to stdout if not in quiet mode.""" |
635 | return self.to_stdout(message, skip_eol, check_quiet=True) | |
636 | ||
734f90bb | 637 | def _write_string(self, s, out=None): |
b58ddb32 | 638 | write_string(s, out=out, encoding=self.params.get('encoding')) |
734f90bb | 639 | |
0783b09b | 640 | def to_stdout(self, message, skip_eol=False, check_quiet=False): |
8222d8de | 641 | """Print message to stdout if not in quiet mode.""" |
8bf9319e | 642 | if self.params.get('logger'): |
43afe285 | 643 | self.params['logger'].debug(message) |
0783b09b | 644 | elif not check_quiet or not self.params.get('quiet', False): |
1c088fa8 | 645 | message = self._bidi_workaround(message) |
6febd1c1 | 646 | terminator = ['\n', ''][skip_eol] |
8222d8de | 647 | output = message + terminator |
1c088fa8 | 648 | |
734f90bb | 649 | self._write_string(output, self._screen_file) |
8222d8de JMF |
650 | |
651 | def to_stderr(self, message): | |
652 | """Print message to stderr.""" | |
11b85ce6 | 653 | assert isinstance(message, compat_str) |
8bf9319e | 654 | if self.params.get('logger'): |
43afe285 IB |
655 | self.params['logger'].error(message) |
656 | else: | |
1c088fa8 | 657 | message = self._bidi_workaround(message) |
6febd1c1 | 658 | output = message + '\n' |
734f90bb | 659 | self._write_string(output, self._err_file) |
8222d8de | 660 | |
1e5b9a95 PH |
661 | def to_console_title(self, message): |
662 | if not self.params.get('consoletitle', False): | |
663 | return | |
4bede0d8 C |
664 | if compat_os_name == 'nt': |
665 | if ctypes.windll.kernel32.GetConsoleWindow(): | |
666 | # c_wchar_p() might not be necessary if `message` is | |
667 | # already of type unicode() | |
668 | ctypes.windll.kernel32.SetConsoleTitleW(ctypes.c_wchar_p(message)) | |
1e5b9a95 | 669 | elif 'TERM' in os.environ: |
b46696bd | 670 | self._write_string('\033]0;%s\007' % message, self._screen_file) |
1e5b9a95 | 671 | |
bdde425c PH |
672 | def save_console_title(self): |
673 | if not self.params.get('consoletitle', False): | |
674 | return | |
94c3442e S |
675 | if self.params.get('simulate', False): |
676 | return | |
4bede0d8 | 677 | if compat_os_name != 'nt' and 'TERM' in os.environ: |
efd6c574 | 678 | # Save the title on stack |
734f90bb | 679 | self._write_string('\033[22;0t', self._screen_file) |
bdde425c PH |
680 | |
681 | def restore_console_title(self): | |
682 | if not self.params.get('consoletitle', False): | |
683 | return | |
94c3442e S |
684 | if self.params.get('simulate', False): |
685 | return | |
4bede0d8 | 686 | if compat_os_name != 'nt' and 'TERM' in os.environ: |
efd6c574 | 687 | # Restore the title from stack |
734f90bb | 688 | self._write_string('\033[23;0t', self._screen_file) |
bdde425c PH |
689 | |
690 | def __enter__(self): | |
691 | self.save_console_title() | |
692 | return self | |
693 | ||
694 | def __exit__(self, *args): | |
695 | self.restore_console_title() | |
f89197d7 | 696 | |
dca08720 | 697 | if self.params.get('cookiefile') is not None: |
1bab3437 | 698 | self.cookiejar.save(ignore_discard=True, ignore_expires=True) |
bdde425c | 699 | |
8222d8de JMF |
700 | def trouble(self, message=None, tb=None): |
701 | """Determine action to take when a download problem appears. | |
702 | ||
703 | Depending on if the downloader has been configured to ignore | |
704 | download errors or not, this method may throw an exception or | |
705 | not when errors are found, after printing the message. | |
706 | ||
707 | tb, if given, is additional traceback information. | |
708 | """ | |
709 | if message is not None: | |
710 | self.to_stderr(message) | |
711 | if self.params.get('verbose'): | |
712 | if tb is None: | |
713 | if sys.exc_info()[0]: # if .trouble has been called from an except block | |
6febd1c1 | 714 | tb = '' |
8222d8de | 715 | if hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]: |
6febd1c1 | 716 | tb += ''.join(traceback.format_exception(*sys.exc_info()[1].exc_info)) |
c0384f22 | 717 | tb += encode_compat_str(traceback.format_exc()) |
8222d8de JMF |
718 | else: |
719 | tb_data = traceback.format_list(traceback.extract_stack()) | |
6febd1c1 | 720 | tb = ''.join(tb_data) |
8222d8de JMF |
721 | self.to_stderr(tb) |
722 | if not self.params.get('ignoreerrors', False): | |
723 | if sys.exc_info()[0] and hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]: | |
724 | exc_info = sys.exc_info()[1].exc_info | |
725 | else: | |
726 | exc_info = sys.exc_info() | |
727 | raise DownloadError(message, exc_info) | |
728 | self._download_retcode = 1 | |
729 | ||
730 | def report_warning(self, message): | |
731 | ''' | |
732 | Print the message to stderr, it will be prefixed with 'WARNING:' | |
733 | If stderr is a tty file the 'WARNING:' will be colored | |
734 | ''' | |
6d07ce01 JMF |
735 | if self.params.get('logger') is not None: |
736 | self.params['logger'].warning(message) | |
8222d8de | 737 | else: |
ad8915b7 PH |
738 | if self.params.get('no_warnings'): |
739 | return | |
e9c0cdd3 | 740 | if not self.params.get('no_color') and self._err_file.isatty() and compat_os_name != 'nt': |
6d07ce01 JMF |
741 | _msg_header = '\033[0;33mWARNING:\033[0m' |
742 | else: | |
743 | _msg_header = 'WARNING:' | |
744 | warning_message = '%s %s' % (_msg_header, message) | |
745 | self.to_stderr(warning_message) | |
8222d8de JMF |
746 | |
747 | def report_error(self, message, tb=None): | |
748 | ''' | |
749 | Do the same as trouble, but prefixes the message with 'ERROR:', colored | |
750 | in red if stderr is a tty file. | |
751 | ''' | |
e9c0cdd3 | 752 | if not self.params.get('no_color') and self._err_file.isatty() and compat_os_name != 'nt': |
6febd1c1 | 753 | _msg_header = '\033[0;31mERROR:\033[0m' |
8222d8de | 754 | else: |
6febd1c1 PH |
755 | _msg_header = 'ERROR:' |
756 | error_message = '%s %s' % (_msg_header, message) | |
8222d8de JMF |
757 | self.trouble(error_message, tb) |
758 | ||
8222d8de JMF |
759 | def report_file_already_downloaded(self, file_name): |
760 | """Report file has already been fully downloaded.""" | |
761 | try: | |
6febd1c1 | 762 | self.to_screen('[download] %s has already been downloaded' % file_name) |
ce02ed60 | 763 | except UnicodeEncodeError: |
6febd1c1 | 764 | self.to_screen('[download] The file has already been downloaded') |
8222d8de | 765 | |
0c3d0f51 | 766 | def report_file_delete(self, file_name): |
767 | """Report that existing file will be deleted.""" | |
768 | try: | |
c25228e5 | 769 | self.to_screen('Deleting existing file %s' % file_name) |
0c3d0f51 | 770 | except UnicodeEncodeError: |
c25228e5 | 771 | self.to_screen('Deleting existing file') |
0c3d0f51 | 772 | |
de6000d9 | 773 | def parse_outtmpl(self): |
774 | outtmpl_dict = self.params.get('outtmpl', {}) | |
775 | if not isinstance(outtmpl_dict, dict): | |
776 | outtmpl_dict = {'default': outtmpl_dict} | |
777 | outtmpl_dict.update({ | |
778 | k: v for k, v in DEFAULT_OUTTMPL.items() | |
779 | if not outtmpl_dict.get(k)}) | |
780 | for key, val in outtmpl_dict.items(): | |
781 | if isinstance(val, bytes): | |
782 | self.report_warning( | |
783 | 'Parameter outtmpl is bytes, but should be a unicode string. ' | |
784 | 'Put from __future__ import unicode_literals at the top of your code file or consider switching to Python 3.x.') | |
785 | return outtmpl_dict | |
786 | ||
143db31d | 787 | def prepare_outtmpl(self, outtmpl, info_dict, sanitize=None): |
788 | """ Make the template and info_dict suitable for substitution (outtmpl % info_dict)""" | |
789 | template_dict = dict(info_dict) | |
790 | ||
791 | # duration_string | |
792 | template_dict['duration_string'] = ( # %(duration>%H-%M-%S)s is wrong if duration > 24hrs | |
793 | formatSeconds(info_dict['duration'], '-') | |
794 | if info_dict.get('duration', None) is not None | |
795 | else None) | |
796 | ||
797 | # epoch | |
798 | template_dict['epoch'] = int(time.time()) | |
799 | ||
800 | # autonumber | |
801 | autonumber_size = self.params.get('autonumber_size') | |
802 | if autonumber_size is None: | |
803 | autonumber_size = 5 | |
804 | template_dict['autonumber'] = self.params.get('autonumber_start', 1) - 1 + self._num_downloads | |
805 | ||
806 | # resolution if not defined | |
807 | if template_dict.get('resolution') is None: | |
808 | if template_dict.get('width') and template_dict.get('height'): | |
809 | template_dict['resolution'] = '%dx%d' % (template_dict['width'], template_dict['height']) | |
810 | elif template_dict.get('height'): | |
811 | template_dict['resolution'] = '%sp' % template_dict['height'] | |
812 | elif template_dict.get('width'): | |
813 | template_dict['resolution'] = '%dx?' % template_dict['width'] | |
814 | ||
815 | if sanitize is None: | |
816 | sanitize = lambda k, v: v | |
817 | template_dict = dict((k, v if isinstance(v, compat_numeric_types) else sanitize(k, v)) | |
818 | for k, v in template_dict.items() | |
819 | if v is not None and not isinstance(v, (list, tuple, dict))) | |
820 | na = self.params.get('outtmpl_na_placeholder', 'NA') | |
821 | template_dict = collections.defaultdict(lambda: na, template_dict) | |
822 | ||
823 | # For fields playlist_index and autonumber convert all occurrences | |
824 | # of %(field)s to %(field)0Nd for backward compatibility | |
825 | field_size_compat_map = { | |
826 | 'playlist_index': len(str(template_dict['n_entries'])), | |
827 | 'autonumber': autonumber_size, | |
828 | } | |
829 | FIELD_SIZE_COMPAT_RE = r'(?<!%)%\((?P<field>autonumber|playlist_index)\)s' | |
830 | mobj = re.search(FIELD_SIZE_COMPAT_RE, outtmpl) | |
831 | if mobj: | |
832 | outtmpl = re.sub( | |
833 | FIELD_SIZE_COMPAT_RE, | |
834 | r'%%(\1)0%dd' % field_size_compat_map[mobj.group('field')], | |
835 | outtmpl) | |
836 | ||
837 | numeric_fields = list(self._NUMERIC_FIELDS) | |
838 | ||
839 | # Format date | |
840 | FORMAT_DATE_RE = FORMAT_RE.format(r'(?P<key>(?P<field>\w+)>(?P<format>.+?))') | |
841 | for mobj in re.finditer(FORMAT_DATE_RE, outtmpl): | |
842 | conv_type, field, frmt, key = mobj.group('type', 'field', 'format', 'key') | |
843 | if key in template_dict: | |
844 | continue | |
845 | value = strftime_or_none(template_dict.get(field), frmt, na) | |
846 | if conv_type in 'crs': # string | |
847 | value = sanitize(field, value) | |
848 | else: # number | |
849 | numeric_fields.append(key) | |
850 | value = float_or_none(value, default=None) | |
851 | if value is not None: | |
852 | template_dict[key] = value | |
853 | ||
854 | # Missing numeric fields used together with integer presentation types | |
855 | # in format specification will break the argument substitution since | |
856 | # string NA placeholder is returned for missing fields. We will patch | |
857 | # output template for missing fields to meet string presentation type. | |
858 | for numeric_field in numeric_fields: | |
859 | if numeric_field not in template_dict: | |
860 | outtmpl = re.sub( | |
861 | FORMAT_RE.format(re.escape(numeric_field)), | |
862 | r'%({0})s'.format(numeric_field), outtmpl) | |
863 | ||
864 | return outtmpl, template_dict | |
865 | ||
de6000d9 | 866 | def _prepare_filename(self, info_dict, tmpl_type='default'): |
8222d8de | 867 | try: |
586a91b6 | 868 | sanitize = lambda k, v: sanitize_filename( |
45598aab | 869 | compat_str(v), |
1bb5c511 | 870 | restricted=self.params.get('restrictfilenames'), |
40df485f | 871 | is_id=(k == 'id' or k.endswith('_id'))) |
de6000d9 | 872 | outtmpl = self.outtmpl_dict.get(tmpl_type, self.outtmpl_dict['default']) |
143db31d | 873 | outtmpl, template_dict = self.prepare_outtmpl(outtmpl, info_dict, sanitize) |
d0d9ade4 | 874 | |
15da37c7 S |
875 | # expand_path translates '%%' into '%' and '$$' into '$' |
876 | # correspondingly that is not what we want since we need to keep | |
877 | # '%%' intact for template dict substitution step. Working around | |
878 | # with boundary-alike separator hack. | |
961ea474 | 879 | sep = ''.join([random.choice(ascii_letters) for _ in range(32)]) |
15da37c7 S |
880 | outtmpl = outtmpl.replace('%%', '%{0}%'.format(sep)).replace('$$', '${0}$'.format(sep)) |
881 | ||
882 | # outtmpl should be expand_path'ed before template dict substitution | |
883 | # because meta fields may contain env variables we don't want to | |
884 | # be expanded. For example, for outtmpl "%(title)s.%(ext)s" and | |
885 | # title "Hello $PATH", we don't want `$PATH` to be expanded. | |
886 | filename = expand_path(outtmpl).replace(sep, '') % template_dict | |
887 | ||
143db31d | 888 | force_ext = OUTTMPL_TYPES.get(tmpl_type) |
de6000d9 | 889 | if force_ext is not None: |
890 | filename = replace_extension(filename, force_ext, template_dict.get('ext')) | |
891 | ||
bdc3fd2f U |
892 | # https://github.com/blackjack4494/youtube-dlc/issues/85 |
893 | trim_file_name = self.params.get('trim_file_name', False) | |
894 | if trim_file_name: | |
895 | fn_groups = filename.rsplit('.') | |
896 | ext = fn_groups[-1] | |
897 | sub_ext = '' | |
898 | if len(fn_groups) > 2: | |
899 | sub_ext = fn_groups[-2] | |
900 | filename = '.'.join(filter(None, [fn_groups[0][:trim_file_name], sub_ext, ext])) | |
901 | ||
0202b52a | 902 | return filename |
8222d8de | 903 | except ValueError as err: |
6febd1c1 | 904 | self.report_error('Error in output template: ' + str(err) + ' (encoding: ' + repr(preferredencoding()) + ')') |
8222d8de JMF |
905 | return None |
906 | ||
de6000d9 | 907 | def prepare_filename(self, info_dict, dir_type='', warn=False): |
908 | """Generate the output filename.""" | |
0202b52a | 909 | paths = self.params.get('paths', {}) |
910 | assert isinstance(paths, dict) | |
de6000d9 | 911 | filename = self._prepare_filename(info_dict, dir_type or 'default') |
912 | ||
913 | if warn and not self.__prepare_filename_warned: | |
914 | if not paths: | |
915 | pass | |
916 | elif filename == '-': | |
917 | self.report_warning('--paths is ignored when an outputting to stdout') | |
918 | elif os.path.isabs(filename): | |
919 | self.report_warning('--paths is ignored since an absolute path is given in output template') | |
920 | self.__prepare_filename_warned = True | |
921 | if filename == '-' or not filename: | |
922 | return filename | |
923 | ||
0202b52a | 924 | homepath = expand_path(paths.get('home', '').strip()) |
925 | assert isinstance(homepath, compat_str) | |
926 | subdir = expand_path(paths.get(dir_type, '').strip()) if dir_type else '' | |
927 | assert isinstance(subdir, compat_str) | |
c2934512 | 928 | path = os.path.join(homepath, subdir, filename) |
929 | ||
930 | # Temporary fix for #4787 | |
931 | # 'Treat' all problem characters by passing filename through preferredencoding | |
932 | # to workaround encoding issues with subprocess on python2 @ Windows | |
933 | if sys.version_info < (3, 0) and sys.platform == 'win32': | |
934 | path = encodeFilename(path, True).decode(preferredencoding()) | |
935 | return sanitize_path(path, force=self.params.get('windowsfilenames')) | |
0202b52a | 936 | |
442c37b7 | 937 | def _match_entry(self, info_dict, incomplete): |
ecdec191 | 938 | """ Returns None if the file should be downloaded """ |
8222d8de | 939 | |
8b0d7497 | 940 | def check_filter(): |
941 | video_title = info_dict.get('title', info_dict.get('id', 'video')) | |
942 | if 'title' in info_dict: | |
943 | # This can happen when we're just evaluating the playlist | |
944 | title = info_dict['title'] | |
945 | matchtitle = self.params.get('matchtitle', False) | |
946 | if matchtitle: | |
947 | if not re.search(matchtitle, title, re.IGNORECASE): | |
948 | return '"' + title + '" title did not match pattern "' + matchtitle + '"' | |
949 | rejecttitle = self.params.get('rejecttitle', False) | |
950 | if rejecttitle: | |
951 | if re.search(rejecttitle, title, re.IGNORECASE): | |
952 | return '"' + title + '" title matched reject pattern "' + rejecttitle + '"' | |
953 | date = info_dict.get('upload_date') | |
954 | if date is not None: | |
955 | dateRange = self.params.get('daterange', DateRange()) | |
956 | if date not in dateRange: | |
957 | return '%s upload date is not in range %s' % (date_from_str(date).isoformat(), dateRange) | |
958 | view_count = info_dict.get('view_count') | |
959 | if view_count is not None: | |
960 | min_views = self.params.get('min_views') | |
961 | if min_views is not None and view_count < min_views: | |
962 | return 'Skipping %s, because it has not reached minimum view count (%d/%d)' % (video_title, view_count, min_views) | |
963 | max_views = self.params.get('max_views') | |
964 | if max_views is not None and view_count > max_views: | |
965 | return 'Skipping %s, because it has exceeded the maximum view count (%d/%d)' % (video_title, view_count, max_views) | |
966 | if age_restricted(info_dict.get('age_limit'), self.params.get('age_limit')): | |
967 | return 'Skipping "%s" because it is age restricted' % video_title | |
968 | if self.in_download_archive(info_dict): | |
969 | return '%s has already been recorded in archive' % video_title | |
970 | ||
971 | if not incomplete: | |
972 | match_filter = self.params.get('match_filter') | |
973 | if match_filter is not None: | |
974 | ret = match_filter(info_dict) | |
975 | if ret is not None: | |
976 | return ret | |
977 | return None | |
978 | ||
979 | reason = check_filter() | |
980 | if reason is not None: | |
981 | self.to_screen('[download] ' + reason) | |
d83cb531 | 982 | if reason.endswith('has already been recorded in the archive') and self.params.get('break_on_existing', False): |
8b0d7497 | 983 | raise ExistingVideoReached() |
d83cb531 | 984 | elif self.params.get('break_on_reject', False): |
8b0d7497 | 985 | raise RejectedVideoReached() |
986 | return reason | |
fe7e0c98 | 987 | |
b6c45014 JMF |
988 | @staticmethod |
989 | def add_extra_info(info_dict, extra_info): | |
990 | '''Set the keys from extra_info in info dict if they are missing''' | |
991 | for key, value in extra_info.items(): | |
992 | info_dict.setdefault(key, value) | |
993 | ||
0704d222 | 994 | def extract_info(self, url, download=True, ie_key=None, info_dict=None, extra_info={}, |
61aa5ba3 | 995 | process=True, force_generic_extractor=False): |
8222d8de JMF |
996 | ''' |
997 | Returns a list with a dictionary for each video we find. | |
998 | If 'download', also downloads the videos. | |
999 | extra_info is a dict containing the extra values to add to each result | |
613b2d9d | 1000 | ''' |
fe7e0c98 | 1001 | |
61aa5ba3 | 1002 | if not ie_key and force_generic_extractor: |
d22dec74 S |
1003 | ie_key = 'Generic' |
1004 | ||
8222d8de | 1005 | if ie_key: |
56c73665 | 1006 | ies = [self.get_info_extractor(ie_key)] |
8222d8de JMF |
1007 | else: |
1008 | ies = self._ies | |
1009 | ||
1010 | for ie in ies: | |
1011 | if not ie.suitable(url): | |
1012 | continue | |
1013 | ||
9a68de12 | 1014 | ie_key = ie.ie_key() |
1015 | ie = self.get_info_extractor(ie_key) | |
8222d8de | 1016 | if not ie.working(): |
6febd1c1 PH |
1017 | self.report_warning('The program functionality for this site has been marked as broken, ' |
1018 | 'and will probably not work.') | |
8222d8de JMF |
1019 | |
1020 | try: | |
d0757229 | 1021 | temp_id = str_or_none( |
63be1aab | 1022 | ie.extract_id(url) if callable(getattr(ie, 'extract_id', None)) |
1023 | else ie._match_id(url)) | |
a0566bbf | 1024 | except (AssertionError, IndexError, AttributeError): |
1025 | temp_id = None | |
1026 | if temp_id is not None and self.in_download_archive({'id': temp_id, 'ie_key': ie_key}): | |
1027 | self.to_screen("[%s] %s: has already been recorded in archive" % ( | |
1028 | ie_key, temp_id)) | |
1029 | break | |
a0566bbf | 1030 | return self.__extract_info(url, ie, download, extra_info, process, info_dict) |
a0566bbf | 1031 | else: |
1032 | self.report_error('no suitable InfoExtractor for URL %s' % url) | |
1033 | ||
1034 | def __handle_extraction_exceptions(func): | |
1035 | def wrapper(self, *args, **kwargs): | |
1036 | try: | |
1037 | return func(self, *args, **kwargs) | |
773f291d S |
1038 | except GeoRestrictedError as e: |
1039 | msg = e.msg | |
1040 | if e.countries: | |
1041 | msg += '\nThis video is available in %s.' % ', '.join( | |
1042 | map(ISO3166Utils.short2full, e.countries)) | |
1043 | msg += '\nYou might want to use a VPN or a proxy server (with --proxy) to workaround.' | |
1044 | self.report_error(msg) | |
fb043a6e | 1045 | except ExtractorError as e: # An error we somewhat expected |
2c74e6fa | 1046 | self.report_error(compat_str(e), e.format_traceback()) |
8b0d7497 | 1047 | except (MaxDownloadsReached, ExistingVideoReached, RejectedVideoReached): |
d3e5bbf4 | 1048 | raise |
8222d8de JMF |
1049 | except Exception as e: |
1050 | if self.params.get('ignoreerrors', False): | |
9b9c5355 | 1051 | self.report_error(error_to_compat_str(e), tb=encode_compat_str(traceback.format_exc())) |
8222d8de JMF |
1052 | else: |
1053 | raise | |
a0566bbf | 1054 | return wrapper |
1055 | ||
1056 | @__handle_extraction_exceptions | |
1057 | def __extract_info(self, url, ie, download, extra_info, process, info_dict): | |
1058 | ie_result = ie.extract(url) | |
1059 | if ie_result is None: # Finished already (backwards compatibility; listformats and friends should be moved here) | |
1060 | return | |
1061 | if isinstance(ie_result, list): | |
1062 | # Backwards compatibility: old IE result format | |
1063 | ie_result = { | |
1064 | '_type': 'compat_list', | |
1065 | 'entries': ie_result, | |
1066 | } | |
1067 | if info_dict: | |
1068 | if info_dict.get('id'): | |
1069 | ie_result['id'] = info_dict['id'] | |
1070 | if info_dict.get('title'): | |
1071 | ie_result['title'] = info_dict['title'] | |
1072 | self.add_default_extra_info(ie_result, ie, url) | |
1073 | if process: | |
1074 | return self.process_ie_result(ie_result, download, extra_info) | |
8222d8de | 1075 | else: |
a0566bbf | 1076 | return ie_result |
fe7e0c98 | 1077 | |
ea38e55f PH |
1078 | def add_default_extra_info(self, ie_result, ie, url): |
1079 | self.add_extra_info(ie_result, { | |
1080 | 'extractor': ie.IE_NAME, | |
1081 | 'webpage_url': url, | |
1082 | 'webpage_url_basename': url_basename(url), | |
1083 | 'extractor_key': ie.ie_key(), | |
1084 | }) | |
1085 | ||
8222d8de JMF |
1086 | def process_ie_result(self, ie_result, download=True, extra_info={}): |
1087 | """ | |
1088 | Take the result of the ie(may be modified) and resolve all unresolved | |
1089 | references (URLs, playlist items). | |
1090 | ||
1091 | It will also download the videos if 'download'. | |
1092 | Returns the resolved ie_result. | |
1093 | """ | |
e8ee972c PH |
1094 | result_type = ie_result.get('_type', 'video') |
1095 | ||
057a5206 | 1096 | if result_type in ('url', 'url_transparent'): |
134c6ea8 | 1097 | ie_result['url'] = sanitize_url(ie_result['url']) |
057a5206 | 1098 | extract_flat = self.params.get('extract_flat', False) |
3089bc74 S |
1099 | if ((extract_flat == 'in_playlist' and 'playlist' in extra_info) |
1100 | or extract_flat is True): | |
de6000d9 | 1101 | self.__forced_printings(ie_result, self.prepare_filename(ie_result), incomplete=True) |
e8ee972c PH |
1102 | return ie_result |
1103 | ||
8222d8de | 1104 | if result_type == 'video': |
b6c45014 | 1105 | self.add_extra_info(ie_result, extra_info) |
feee2ecf | 1106 | return self.process_video_result(ie_result, download=download) |
8222d8de JMF |
1107 | elif result_type == 'url': |
1108 | # We have to add extra_info to the results because it may be | |
1109 | # contained in a playlist | |
1110 | return self.extract_info(ie_result['url'], | |
0704d222 | 1111 | download, info_dict=ie_result, |
8222d8de JMF |
1112 | ie_key=ie_result.get('ie_key'), |
1113 | extra_info=extra_info) | |
7fc3fa05 PH |
1114 | elif result_type == 'url_transparent': |
1115 | # Use the information from the embedding page | |
1116 | info = self.extract_info( | |
1117 | ie_result['url'], ie_key=ie_result.get('ie_key'), | |
1118 | extra_info=extra_info, download=False, process=False) | |
1119 | ||
1640eb09 S |
1120 | # extract_info may return None when ignoreerrors is enabled and |
1121 | # extraction failed with an error, don't crash and return early | |
1122 | # in this case | |
1123 | if not info: | |
1124 | return info | |
1125 | ||
412c617d PH |
1126 | force_properties = dict( |
1127 | (k, v) for k, v in ie_result.items() if v is not None) | |
0396806f | 1128 | for f in ('_type', 'url', 'id', 'extractor', 'extractor_key', 'ie_key'): |
412c617d PH |
1129 | if f in force_properties: |
1130 | del force_properties[f] | |
1131 | new_result = info.copy() | |
1132 | new_result.update(force_properties) | |
7fc3fa05 | 1133 | |
0563f7ac S |
1134 | # Extracted info may not be a video result (i.e. |
1135 | # info.get('_type', 'video') != video) but rather an url or | |
1136 | # url_transparent. In such cases outer metadata (from ie_result) | |
1137 | # should be propagated to inner one (info). For this to happen | |
1138 | # _type of info should be overridden with url_transparent. This | |
067aa17e | 1139 | # fixes issue from https://github.com/ytdl-org/youtube-dl/pull/11163. |
0563f7ac S |
1140 | if new_result.get('_type') == 'url': |
1141 | new_result['_type'] = 'url_transparent' | |
7fc3fa05 PH |
1142 | |
1143 | return self.process_ie_result( | |
1144 | new_result, download=download, extra_info=extra_info) | |
40fcba5e | 1145 | elif result_type in ('playlist', 'multi_video'): |
30a074c2 | 1146 | # Protect from infinite recursion due to recursively nested playlists |
1147 | # (see https://github.com/ytdl-org/youtube-dl/issues/27833) | |
1148 | webpage_url = ie_result['webpage_url'] | |
1149 | if webpage_url in self._playlist_urls: | |
7e85e872 | 1150 | self.to_screen( |
30a074c2 | 1151 | '[download] Skipping already downloaded playlist: %s' |
1152 | % ie_result.get('title') or ie_result.get('id')) | |
1153 | return | |
7e85e872 | 1154 | |
30a074c2 | 1155 | self._playlist_level += 1 |
1156 | self._playlist_urls.add(webpage_url) | |
1157 | try: | |
1158 | return self.__process_playlist(ie_result, download) | |
1159 | finally: | |
1160 | self._playlist_level -= 1 | |
1161 | if not self._playlist_level: | |
1162 | self._playlist_urls.clear() | |
8222d8de | 1163 | elif result_type == 'compat_list': |
c9bf4114 PH |
1164 | self.report_warning( |
1165 | 'Extractor %s returned a compat_list result. ' | |
1166 | 'It needs to be updated.' % ie_result.get('extractor')) | |
5f6a1245 | 1167 | |
8222d8de | 1168 | def _fixup(r): |
9e1a5b84 JW |
1169 | self.add_extra_info( |
1170 | r, | |
9103bbc5 JMF |
1171 | { |
1172 | 'extractor': ie_result['extractor'], | |
1173 | 'webpage_url': ie_result['webpage_url'], | |
29eb5174 | 1174 | 'webpage_url_basename': url_basename(ie_result['webpage_url']), |
be97abc2 | 1175 | 'extractor_key': ie_result['extractor_key'], |
9e1a5b84 JW |
1176 | } |
1177 | ) | |
8222d8de JMF |
1178 | return r |
1179 | ie_result['entries'] = [ | |
b6c45014 | 1180 | self.process_ie_result(_fixup(r), download, extra_info) |
8222d8de JMF |
1181 | for r in ie_result['entries'] |
1182 | ] | |
1183 | return ie_result | |
1184 | else: | |
1185 | raise Exception('Invalid result type: %s' % result_type) | |
1186 | ||
e92caff5 | 1187 | def _ensure_dir_exists(self, path): |
1188 | return make_dir(path, self.report_error) | |
1189 | ||
30a074c2 | 1190 | def __process_playlist(self, ie_result, download): |
1191 | # We process each entry in the playlist | |
1192 | playlist = ie_result.get('title') or ie_result.get('id') | |
1193 | self.to_screen('[download] Downloading playlist: %s' % playlist) | |
1194 | ||
498f5606 | 1195 | if 'entries' not in ie_result: |
1196 | raise EntryNotInPlaylist() | |
1197 | incomplete_entries = bool(ie_result.get('requested_entries')) | |
1198 | if incomplete_entries: | |
1199 | def fill_missing_entries(entries, indexes): | |
1200 | ret = [None] * max(*indexes) | |
1201 | for i, entry in zip(indexes, entries): | |
1202 | ret[i - 1] = entry | |
1203 | return ret | |
1204 | ie_result['entries'] = fill_missing_entries(ie_result['entries'], ie_result['requested_entries']) | |
02fd60d3 | 1205 | |
30a074c2 | 1206 | playlist_results = [] |
1207 | ||
1208 | playliststart = self.params.get('playliststart', 1) - 1 | |
1209 | playlistend = self.params.get('playlistend') | |
1210 | # For backwards compatibility, interpret -1 as whole list | |
1211 | if playlistend == -1: | |
1212 | playlistend = None | |
1213 | ||
1214 | playlistitems_str = self.params.get('playlist_items') | |
1215 | playlistitems = None | |
1216 | if playlistitems_str is not None: | |
1217 | def iter_playlistitems(format): | |
1218 | for string_segment in format.split(','): | |
1219 | if '-' in string_segment: | |
1220 | start, end = string_segment.split('-') | |
1221 | for item in range(int(start), int(end) + 1): | |
1222 | yield int(item) | |
1223 | else: | |
1224 | yield int(string_segment) | |
1225 | playlistitems = orderedSet(iter_playlistitems(playlistitems_str)) | |
1226 | ||
1227 | ie_entries = ie_result['entries'] | |
1228 | ||
1229 | def make_playlistitems_entries(list_ie_entries): | |
1230 | num_entries = len(list_ie_entries) | |
498f5606 | 1231 | for i in playlistitems: |
1232 | if -num_entries < i <= num_entries: | |
1233 | yield list_ie_entries[i - 1] | |
1234 | elif incomplete_entries: | |
1235 | raise EntryNotInPlaylist() | |
30a074c2 | 1236 | |
1237 | if isinstance(ie_entries, list): | |
1238 | n_all_entries = len(ie_entries) | |
1239 | if playlistitems: | |
498f5606 | 1240 | entries = list(make_playlistitems_entries(ie_entries)) |
30a074c2 | 1241 | else: |
1242 | entries = ie_entries[playliststart:playlistend] | |
1243 | n_entries = len(entries) | |
498f5606 | 1244 | msg = 'Collected %d videos; downloading %d of them' % (n_all_entries, n_entries) |
30a074c2 | 1245 | elif isinstance(ie_entries, PagedList): |
1246 | if playlistitems: | |
1247 | entries = [] | |
1248 | for item in playlistitems: | |
1249 | entries.extend(ie_entries.getslice( | |
1250 | item - 1, item | |
1251 | )) | |
1252 | else: | |
1253 | entries = ie_entries.getslice( | |
1254 | playliststart, playlistend) | |
1255 | n_entries = len(entries) | |
498f5606 | 1256 | msg = 'Downloading %d videos' % n_entries |
30a074c2 | 1257 | else: # iterable |
1258 | if playlistitems: | |
498f5606 | 1259 | entries = list(make_playlistitems_entries(list(itertools.islice( |
1260 | ie_entries, 0, max(playlistitems))))) | |
30a074c2 | 1261 | else: |
1262 | entries = list(itertools.islice( | |
1263 | ie_entries, playliststart, playlistend)) | |
1264 | n_entries = len(entries) | |
498f5606 | 1265 | msg = 'Downloading %d videos' % n_entries |
1266 | ||
1267 | if any((entry is None for entry in entries)): | |
1268 | raise EntryNotInPlaylist() | |
1269 | if not playlistitems and (playliststart or playlistend): | |
1270 | playlistitems = list(range(1 + playliststart, 1 + playliststart + len(entries))) | |
1271 | ie_result['entries'] = entries | |
1272 | ie_result['requested_entries'] = playlistitems | |
1273 | ||
1274 | if self.params.get('allow_playlist_files', True): | |
1275 | ie_copy = { | |
1276 | 'playlist': playlist, | |
1277 | 'playlist_id': ie_result.get('id'), | |
1278 | 'playlist_title': ie_result.get('title'), | |
1279 | 'playlist_uploader': ie_result.get('uploader'), | |
1280 | 'playlist_uploader_id': ie_result.get('uploader_id'), | |
1281 | 'playlist_index': 0 | |
1282 | } | |
1283 | ie_copy.update(dict(ie_result)) | |
1284 | ||
1285 | if self.params.get('writeinfojson', False): | |
1286 | infofn = self.prepare_filename(ie_copy, 'pl_infojson') | |
1287 | if not self._ensure_dir_exists(encodeFilename(infofn)): | |
1288 | return | |
1289 | if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(infofn)): | |
1290 | self.to_screen('[info] Playlist metadata is already present') | |
1291 | else: | |
1292 | self.to_screen('[info] Writing playlist metadata as JSON to: ' + infofn) | |
1293 | try: | |
1294 | write_json_file(self.filter_requested_info(ie_result, self.params.get('clean_infojson', True)), infofn) | |
1295 | except (OSError, IOError): | |
1296 | self.report_error('Cannot write playlist metadata to JSON file ' + infofn) | |
1297 | ||
1298 | if self.params.get('writedescription', False): | |
1299 | descfn = self.prepare_filename(ie_copy, 'pl_description') | |
1300 | if not self._ensure_dir_exists(encodeFilename(descfn)): | |
1301 | return | |
1302 | if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(descfn)): | |
1303 | self.to_screen('[info] Playlist description is already present') | |
1304 | elif ie_result.get('description') is None: | |
1305 | self.report_warning('There\'s no playlist description to write.') | |
1306 | else: | |
1307 | try: | |
1308 | self.to_screen('[info] Writing playlist description to: ' + descfn) | |
1309 | with io.open(encodeFilename(descfn), 'w', encoding='utf-8') as descfile: | |
1310 | descfile.write(ie_result['description']) | |
1311 | except (OSError, IOError): | |
1312 | self.report_error('Cannot write playlist description file ' + descfn) | |
1313 | return | |
30a074c2 | 1314 | |
1315 | if self.params.get('playlistreverse', False): | |
1316 | entries = entries[::-1] | |
30a074c2 | 1317 | if self.params.get('playlistrandom', False): |
1318 | random.shuffle(entries) | |
1319 | ||
1320 | x_forwarded_for = ie_result.get('__x_forwarded_for_ip') | |
1321 | ||
498f5606 | 1322 | self.to_screen('[%s] playlist %s: %s' % (ie_result['extractor'], playlist, msg)) |
30a074c2 | 1323 | for i, entry in enumerate(entries, 1): |
1324 | self.to_screen('[download] Downloading video %s of %s' % (i, n_entries)) | |
1325 | # This __x_forwarded_for_ip thing is a bit ugly but requires | |
1326 | # minimal changes | |
1327 | if x_forwarded_for: | |
1328 | entry['__x_forwarded_for_ip'] = x_forwarded_for | |
1329 | extra = { | |
1330 | 'n_entries': n_entries, | |
1331 | 'playlist': playlist, | |
1332 | 'playlist_id': ie_result.get('id'), | |
1333 | 'playlist_title': ie_result.get('title'), | |
1334 | 'playlist_uploader': ie_result.get('uploader'), | |
1335 | 'playlist_uploader_id': ie_result.get('uploader_id'), | |
498f5606 | 1336 | 'playlist_index': playlistitems[i - 1] if playlistitems else i, |
30a074c2 | 1337 | 'extractor': ie_result['extractor'], |
1338 | 'webpage_url': ie_result['webpage_url'], | |
1339 | 'webpage_url_basename': url_basename(ie_result['webpage_url']), | |
1340 | 'extractor_key': ie_result['extractor_key'], | |
1341 | } | |
1342 | ||
1343 | if self._match_entry(entry, incomplete=True) is not None: | |
1344 | continue | |
1345 | ||
1346 | entry_result = self.__process_iterable_entry(entry, download, extra) | |
1347 | # TODO: skip failed (empty) entries? | |
1348 | playlist_results.append(entry_result) | |
1349 | ie_result['entries'] = playlist_results | |
1350 | self.to_screen('[download] Finished downloading playlist: %s' % playlist) | |
1351 | return ie_result | |
1352 | ||
a0566bbf | 1353 | @__handle_extraction_exceptions |
1354 | def __process_iterable_entry(self, entry, download, extra_info): | |
1355 | return self.process_ie_result( | |
1356 | entry, download=download, extra_info=extra_info) | |
1357 | ||
67134eab JMF |
1358 | def _build_format_filter(self, filter_spec): |
1359 | " Returns a function to filter the formats according to the filter_spec " | |
083c9df9 PH |
1360 | |
1361 | OPERATORS = { | |
1362 | '<': operator.lt, | |
1363 | '<=': operator.le, | |
1364 | '>': operator.gt, | |
1365 | '>=': operator.ge, | |
1366 | '=': operator.eq, | |
1367 | '!=': operator.ne, | |
1368 | } | |
67134eab | 1369 | operator_rex = re.compile(r'''(?x)\s* |
a03a3c80 | 1370 | (?P<key>width|height|tbr|abr|vbr|asr|filesize|filesize_approx|fps) |
083c9df9 PH |
1371 | \s*(?P<op>%s)(?P<none_inclusive>\s*\?)?\s* |
1372 | (?P<value>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?) | |
67134eab | 1373 | $ |
083c9df9 | 1374 | ''' % '|'.join(map(re.escape, OPERATORS.keys()))) |
67134eab | 1375 | m = operator_rex.search(filter_spec) |
9ddb6925 S |
1376 | if m: |
1377 | try: | |
1378 | comparison_value = int(m.group('value')) | |
1379 | except ValueError: | |
1380 | comparison_value = parse_filesize(m.group('value')) | |
1381 | if comparison_value is None: | |
1382 | comparison_value = parse_filesize(m.group('value') + 'B') | |
1383 | if comparison_value is None: | |
1384 | raise ValueError( | |
1385 | 'Invalid value %r in format specification %r' % ( | |
67134eab | 1386 | m.group('value'), filter_spec)) |
9ddb6925 S |
1387 | op = OPERATORS[m.group('op')] |
1388 | ||
083c9df9 | 1389 | if not m: |
9ddb6925 S |
1390 | STR_OPERATORS = { |
1391 | '=': operator.eq, | |
10d33b34 YCH |
1392 | '^=': lambda attr, value: attr.startswith(value), |
1393 | '$=': lambda attr, value: attr.endswith(value), | |
1394 | '*=': lambda attr, value: value in attr, | |
9ddb6925 | 1395 | } |
67134eab | 1396 | str_operator_rex = re.compile(r'''(?x) |
f96bff99 | 1397 | \s*(?P<key>[a-zA-Z0-9._-]+) |
2cc779f4 | 1398 | \s*(?P<negation>!\s*)?(?P<op>%s)(?P<none_inclusive>\s*\?)? |
b0df5223 | 1399 | \s*(?P<value>[a-zA-Z0-9._-]+) |
67134eab | 1400 | \s*$ |
9ddb6925 | 1401 | ''' % '|'.join(map(re.escape, STR_OPERATORS.keys()))) |
67134eab | 1402 | m = str_operator_rex.search(filter_spec) |
9ddb6925 S |
1403 | if m: |
1404 | comparison_value = m.group('value') | |
2cc779f4 S |
1405 | str_op = STR_OPERATORS[m.group('op')] |
1406 | if m.group('negation'): | |
e118a879 | 1407 | op = lambda attr, value: not str_op(attr, value) |
2cc779f4 S |
1408 | else: |
1409 | op = str_op | |
083c9df9 | 1410 | |
9ddb6925 | 1411 | if not m: |
67134eab | 1412 | raise ValueError('Invalid filter specification %r' % filter_spec) |
083c9df9 PH |
1413 | |
1414 | def _filter(f): | |
1415 | actual_value = f.get(m.group('key')) | |
1416 | if actual_value is None: | |
1417 | return m.group('none_inclusive') | |
1418 | return op(actual_value, comparison_value) | |
67134eab JMF |
1419 | return _filter |
1420 | ||
0017d9ad | 1421 | def _default_format_spec(self, info_dict, download=True): |
0017d9ad | 1422 | |
af0f7428 S |
1423 | def can_merge(): |
1424 | merger = FFmpegMergerPP(self) | |
1425 | return merger.available and merger.can_merge() | |
1426 | ||
91ebc640 | 1427 | prefer_best = ( |
1428 | not self.params.get('simulate', False) | |
1429 | and download | |
1430 | and ( | |
1431 | not can_merge() | |
19807826 | 1432 | or info_dict.get('is_live', False) |
de6000d9 | 1433 | or self.outtmpl_dict['default'] == '-')) |
91ebc640 | 1434 | |
1435 | return ( | |
1436 | 'best/bestvideo+bestaudio' | |
1437 | if prefer_best | |
1438 | else 'bestvideo*+bestaudio/best' | |
19807826 | 1439 | if not self.params.get('allow_multiple_audio_streams', False) |
91ebc640 | 1440 | else 'bestvideo+bestaudio/best') |
0017d9ad | 1441 | |
67134eab JMF |
1442 | def build_format_selector(self, format_spec): |
1443 | def syntax_error(note, start): | |
1444 | message = ( | |
1445 | 'Invalid format specification: ' | |
1446 | '{0}\n\t{1}\n\t{2}^'.format(note, format_spec, ' ' * start[1])) | |
1447 | return SyntaxError(message) | |
1448 | ||
1449 | PICKFIRST = 'PICKFIRST' | |
1450 | MERGE = 'MERGE' | |
1451 | SINGLE = 'SINGLE' | |
0130afb7 | 1452 | GROUP = 'GROUP' |
67134eab JMF |
1453 | FormatSelector = collections.namedtuple('FormatSelector', ['type', 'selector', 'filters']) |
1454 | ||
91ebc640 | 1455 | allow_multiple_streams = {'audio': self.params.get('allow_multiple_audio_streams', False), |
1456 | 'video': self.params.get('allow_multiple_video_streams', False)} | |
909d24dd | 1457 | |
67134eab JMF |
1458 | def _parse_filter(tokens): |
1459 | filter_parts = [] | |
1460 | for type, string, start, _, _ in tokens: | |
1461 | if type == tokenize.OP and string == ']': | |
1462 | return ''.join(filter_parts) | |
1463 | else: | |
1464 | filter_parts.append(string) | |
1465 | ||
232541df | 1466 | def _remove_unused_ops(tokens): |
17cc1534 | 1467 | # Remove operators that we don't use and join them with the surrounding strings |
232541df JMF |
1468 | # for example: 'mp4' '-' 'baseline' '-' '16x9' is converted to 'mp4-baseline-16x9' |
1469 | ALLOWED_OPS = ('/', '+', ',', '(', ')') | |
1470 | last_string, last_start, last_end, last_line = None, None, None, None | |
1471 | for type, string, start, end, line in tokens: | |
1472 | if type == tokenize.OP and string == '[': | |
1473 | if last_string: | |
1474 | yield tokenize.NAME, last_string, last_start, last_end, last_line | |
1475 | last_string = None | |
1476 | yield type, string, start, end, line | |
1477 | # everything inside brackets will be handled by _parse_filter | |
1478 | for type, string, start, end, line in tokens: | |
1479 | yield type, string, start, end, line | |
1480 | if type == tokenize.OP and string == ']': | |
1481 | break | |
1482 | elif type == tokenize.OP and string in ALLOWED_OPS: | |
1483 | if last_string: | |
1484 | yield tokenize.NAME, last_string, last_start, last_end, last_line | |
1485 | last_string = None | |
1486 | yield type, string, start, end, line | |
1487 | elif type in [tokenize.NAME, tokenize.NUMBER, tokenize.OP]: | |
1488 | if not last_string: | |
1489 | last_string = string | |
1490 | last_start = start | |
1491 | last_end = end | |
1492 | else: | |
1493 | last_string += string | |
1494 | if last_string: | |
1495 | yield tokenize.NAME, last_string, last_start, last_end, last_line | |
1496 | ||
cf2ac6df | 1497 | def _parse_format_selection(tokens, inside_merge=False, inside_choice=False, inside_group=False): |
67134eab JMF |
1498 | selectors = [] |
1499 | current_selector = None | |
1500 | for type, string, start, _, _ in tokens: | |
1501 | # ENCODING is only defined in python 3.x | |
1502 | if type == getattr(tokenize, 'ENCODING', None): | |
1503 | continue | |
1504 | elif type in [tokenize.NAME, tokenize.NUMBER]: | |
1505 | current_selector = FormatSelector(SINGLE, string, []) | |
1506 | elif type == tokenize.OP: | |
cf2ac6df JMF |
1507 | if string == ')': |
1508 | if not inside_group: | |
1509 | # ')' will be handled by the parentheses group | |
1510 | tokens.restore_last_token() | |
67134eab | 1511 | break |
cf2ac6df | 1512 | elif inside_merge and string in ['/', ',']: |
0130afb7 JMF |
1513 | tokens.restore_last_token() |
1514 | break | |
cf2ac6df JMF |
1515 | elif inside_choice and string == ',': |
1516 | tokens.restore_last_token() | |
1517 | break | |
1518 | elif string == ',': | |
0a31a350 JMF |
1519 | if not current_selector: |
1520 | raise syntax_error('"," must follow a format selector', start) | |
67134eab JMF |
1521 | selectors.append(current_selector) |
1522 | current_selector = None | |
1523 | elif string == '/': | |
d96d604e JMF |
1524 | if not current_selector: |
1525 | raise syntax_error('"/" must follow a format selector', start) | |
67134eab | 1526 | first_choice = current_selector |
cf2ac6df | 1527 | second_choice = _parse_format_selection(tokens, inside_choice=True) |
f5f4a27a | 1528 | current_selector = FormatSelector(PICKFIRST, (first_choice, second_choice), []) |
67134eab JMF |
1529 | elif string == '[': |
1530 | if not current_selector: | |
1531 | current_selector = FormatSelector(SINGLE, 'best', []) | |
1532 | format_filter = _parse_filter(tokens) | |
1533 | current_selector.filters.append(format_filter) | |
0130afb7 JMF |
1534 | elif string == '(': |
1535 | if current_selector: | |
1536 | raise syntax_error('Unexpected "("', start) | |
cf2ac6df JMF |
1537 | group = _parse_format_selection(tokens, inside_group=True) |
1538 | current_selector = FormatSelector(GROUP, group, []) | |
67134eab | 1539 | elif string == '+': |
d03cfdce | 1540 | if not current_selector: |
1541 | raise syntax_error('Unexpected "+"', start) | |
1542 | selector_1 = current_selector | |
1543 | selector_2 = _parse_format_selection(tokens, inside_merge=True) | |
1544 | if not selector_2: | |
1545 | raise syntax_error('Expected a selector', start) | |
1546 | current_selector = FormatSelector(MERGE, (selector_1, selector_2), []) | |
67134eab JMF |
1547 | else: |
1548 | raise syntax_error('Operator not recognized: "{0}"'.format(string), start) | |
1549 | elif type == tokenize.ENDMARKER: | |
1550 | break | |
1551 | if current_selector: | |
1552 | selectors.append(current_selector) | |
1553 | return selectors | |
1554 | ||
f8d4ad9a | 1555 | def _merge(formats_pair): |
1556 | format_1, format_2 = formats_pair | |
1557 | ||
1558 | formats_info = [] | |
1559 | formats_info.extend(format_1.get('requested_formats', (format_1,))) | |
1560 | formats_info.extend(format_2.get('requested_formats', (format_2,))) | |
1561 | ||
1562 | if not allow_multiple_streams['video'] or not allow_multiple_streams['audio']: | |
1563 | get_no_more = {"video": False, "audio": False} | |
1564 | for (i, fmt_info) in enumerate(formats_info): | |
1565 | for aud_vid in ["audio", "video"]: | |
1566 | if not allow_multiple_streams[aud_vid] and fmt_info.get(aud_vid[0] + 'codec') != 'none': | |
1567 | if get_no_more[aud_vid]: | |
1568 | formats_info.pop(i) | |
1569 | get_no_more[aud_vid] = True | |
1570 | ||
1571 | if len(formats_info) == 1: | |
1572 | return formats_info[0] | |
1573 | ||
1574 | video_fmts = [fmt_info for fmt_info in formats_info if fmt_info.get('vcodec') != 'none'] | |
1575 | audio_fmts = [fmt_info for fmt_info in formats_info if fmt_info.get('acodec') != 'none'] | |
1576 | ||
1577 | the_only_video = video_fmts[0] if len(video_fmts) == 1 else None | |
1578 | the_only_audio = audio_fmts[0] if len(audio_fmts) == 1 else None | |
1579 | ||
1580 | output_ext = self.params.get('merge_output_format') | |
1581 | if not output_ext: | |
1582 | if the_only_video: | |
1583 | output_ext = the_only_video['ext'] | |
1584 | elif the_only_audio and not video_fmts: | |
1585 | output_ext = the_only_audio['ext'] | |
1586 | else: | |
1587 | output_ext = 'mkv' | |
1588 | ||
1589 | new_dict = { | |
1590 | 'requested_formats': formats_info, | |
1591 | 'format': '+'.join(fmt_info.get('format') for fmt_info in formats_info), | |
1592 | 'format_id': '+'.join(fmt_info.get('format_id') for fmt_info in formats_info), | |
1593 | 'ext': output_ext, | |
1594 | } | |
1595 | ||
1596 | if the_only_video: | |
1597 | new_dict.update({ | |
1598 | 'width': the_only_video.get('width'), | |
1599 | 'height': the_only_video.get('height'), | |
1600 | 'resolution': the_only_video.get('resolution') or self.format_resolution(the_only_video), | |
1601 | 'fps': the_only_video.get('fps'), | |
1602 | 'vcodec': the_only_video.get('vcodec'), | |
1603 | 'vbr': the_only_video.get('vbr'), | |
1604 | 'stretched_ratio': the_only_video.get('stretched_ratio'), | |
1605 | }) | |
1606 | ||
1607 | if the_only_audio: | |
1608 | new_dict.update({ | |
1609 | 'acodec': the_only_audio.get('acodec'), | |
1610 | 'abr': the_only_audio.get('abr'), | |
1611 | }) | |
1612 | ||
1613 | return new_dict | |
1614 | ||
67134eab | 1615 | def _build_selector_function(selector): |
909d24dd | 1616 | if isinstance(selector, list): # , |
67134eab JMF |
1617 | fs = [_build_selector_function(s) for s in selector] |
1618 | ||
317f7ab6 | 1619 | def selector_function(ctx): |
67134eab | 1620 | for f in fs: |
317f7ab6 | 1621 | for format in f(ctx): |
67134eab JMF |
1622 | yield format |
1623 | return selector_function | |
909d24dd | 1624 | |
1625 | elif selector.type == GROUP: # () | |
0130afb7 | 1626 | selector_function = _build_selector_function(selector.selector) |
909d24dd | 1627 | |
1628 | elif selector.type == PICKFIRST: # / | |
67134eab JMF |
1629 | fs = [_build_selector_function(s) for s in selector.selector] |
1630 | ||
317f7ab6 | 1631 | def selector_function(ctx): |
67134eab | 1632 | for f in fs: |
317f7ab6 | 1633 | picked_formats = list(f(ctx)) |
67134eab JMF |
1634 | if picked_formats: |
1635 | return picked_formats | |
1636 | return [] | |
67134eab | 1637 | |
909d24dd | 1638 | elif selector.type == SINGLE: # atom |
e01d6aa4 | 1639 | format_spec = (selector.selector or 'best').lower() |
909d24dd | 1640 | |
f8d4ad9a | 1641 | # TODO: Add allvideo, allaudio etc by generalizing the code with best/worst selector |
909d24dd | 1642 | if format_spec == 'all': |
1643 | def selector_function(ctx): | |
1644 | formats = list(ctx['formats']) | |
1645 | if formats: | |
1646 | for f in formats: | |
1647 | yield f | |
f8d4ad9a | 1648 | elif format_spec == 'mergeall': |
1649 | def selector_function(ctx): | |
1650 | formats = list(ctx['formats']) | |
e01d6aa4 | 1651 | if not formats: |
1652 | return | |
921b76ca | 1653 | merged_format = formats[-1] |
1654 | for f in formats[-2::-1]: | |
f8d4ad9a | 1655 | merged_format = _merge((merged_format, f)) |
1656 | yield merged_format | |
909d24dd | 1657 | |
1658 | else: | |
1659 | format_fallback = False | |
eff63539 | 1660 | mobj = re.match( |
1661 | r'(?P<bw>best|worst|b|w)(?P<type>video|audio|v|a)?(?P<mod>\*)?(?:\.(?P<n>[1-9]\d*))?$', | |
1662 | format_spec) | |
1663 | if mobj is not None: | |
1664 | format_idx = int_or_none(mobj.group('n'), default=1) | |
1665 | format_idx = format_idx - 1 if mobj.group('bw')[0] == 'w' else -format_idx | |
1666 | format_type = (mobj.group('type') or [None])[0] | |
1667 | not_format_type = {'v': 'a', 'a': 'v'}.get(format_type) | |
1668 | format_modified = mobj.group('mod') is not None | |
909d24dd | 1669 | |
1670 | format_fallback = not format_type and not format_modified # for b, w | |
eff63539 | 1671 | filter_f = ( |
1672 | (lambda f: f.get('%scodec' % format_type) != 'none') | |
1673 | if format_type and format_modified # bv*, ba*, wv*, wa* | |
1674 | else (lambda f: f.get('%scodec' % not_format_type) == 'none') | |
1675 | if format_type # bv, ba, wv, wa | |
1676 | else (lambda f: f.get('vcodec') != 'none' and f.get('acodec') != 'none') | |
1677 | if not format_modified # b, w | |
1678 | else None) # b*, w* | |
67134eab | 1679 | else: |
909d24dd | 1680 | format_idx = -1 |
1681 | filter_f = ((lambda f: f.get('ext') == format_spec) | |
1682 | if format_spec in ['mp4', 'flv', 'webm', '3gp', 'm4a', 'mp3', 'ogg', 'aac', 'wav'] # extension | |
1683 | else (lambda f: f.get('format_id') == format_spec)) # id | |
1684 | ||
1685 | def selector_function(ctx): | |
1686 | formats = list(ctx['formats']) | |
1687 | if not formats: | |
1688 | return | |
1689 | matches = list(filter(filter_f, formats)) if filter_f is not None else formats | |
eff63539 | 1690 | n = len(matches) |
1691 | if -n <= format_idx < n: | |
909d24dd | 1692 | yield matches[format_idx] |
eff63539 | 1693 | elif format_fallback and ctx['incomplete_formats']: |
909d24dd | 1694 | # for extractors with incomplete formats (audio only (soundcloud) |
1695 | # or video only (imgur)) best/worst will fallback to | |
1696 | # best/worst {video,audio}-only format | |
eff63539 | 1697 | n = len(formats) |
1698 | if -n <= format_idx < n: | |
1699 | yield formats[format_idx] | |
909d24dd | 1700 | |
1701 | elif selector.type == MERGE: # + | |
d03cfdce | 1702 | selector_1, selector_2 = map(_build_selector_function, selector.selector) |
083c9df9 | 1703 | |
317f7ab6 S |
1704 | def selector_function(ctx): |
1705 | for pair in itertools.product( | |
d03cfdce | 1706 | selector_1(copy.deepcopy(ctx)), selector_2(copy.deepcopy(ctx))): |
67134eab | 1707 | yield _merge(pair) |
083c9df9 | 1708 | |
67134eab | 1709 | filters = [self._build_format_filter(f) for f in selector.filters] |
083c9df9 | 1710 | |
317f7ab6 S |
1711 | def final_selector(ctx): |
1712 | ctx_copy = copy.deepcopy(ctx) | |
67134eab | 1713 | for _filter in filters: |
317f7ab6 S |
1714 | ctx_copy['formats'] = list(filter(_filter, ctx_copy['formats'])) |
1715 | return selector_function(ctx_copy) | |
67134eab | 1716 | return final_selector |
083c9df9 | 1717 | |
67134eab | 1718 | stream = io.BytesIO(format_spec.encode('utf-8')) |
0130afb7 | 1719 | try: |
232541df | 1720 | tokens = list(_remove_unused_ops(compat_tokenize_tokenize(stream.readline))) |
0130afb7 JMF |
1721 | except tokenize.TokenError: |
1722 | raise syntax_error('Missing closing/opening brackets or parenthesis', (0, len(format_spec))) | |
1723 | ||
1724 | class TokenIterator(object): | |
1725 | def __init__(self, tokens): | |
1726 | self.tokens = tokens | |
1727 | self.counter = 0 | |
1728 | ||
1729 | def __iter__(self): | |
1730 | return self | |
1731 | ||
1732 | def __next__(self): | |
1733 | if self.counter >= len(self.tokens): | |
1734 | raise StopIteration() | |
1735 | value = self.tokens[self.counter] | |
1736 | self.counter += 1 | |
1737 | return value | |
1738 | ||
1739 | next = __next__ | |
1740 | ||
1741 | def restore_last_token(self): | |
1742 | self.counter -= 1 | |
1743 | ||
1744 | parsed_selector = _parse_format_selection(iter(TokenIterator(tokens))) | |
67134eab | 1745 | return _build_selector_function(parsed_selector) |
a9c58ad9 | 1746 | |
e5660ee6 JMF |
1747 | def _calc_headers(self, info_dict): |
1748 | res = std_headers.copy() | |
1749 | ||
1750 | add_headers = info_dict.get('http_headers') | |
1751 | if add_headers: | |
1752 | res.update(add_headers) | |
1753 | ||
1754 | cookies = self._calc_cookies(info_dict) | |
1755 | if cookies: | |
1756 | res['Cookie'] = cookies | |
1757 | ||
0016b84e S |
1758 | if 'X-Forwarded-For' not in res: |
1759 | x_forwarded_for_ip = info_dict.get('__x_forwarded_for_ip') | |
1760 | if x_forwarded_for_ip: | |
1761 | res['X-Forwarded-For'] = x_forwarded_for_ip | |
1762 | ||
e5660ee6 JMF |
1763 | return res |
1764 | ||
1765 | def _calc_cookies(self, info_dict): | |
5c2266df | 1766 | pr = sanitized_Request(info_dict['url']) |
e5660ee6 | 1767 | self.cookiejar.add_cookie_header(pr) |
662435f7 | 1768 | return pr.get_header('Cookie') |
e5660ee6 | 1769 | |
dd82ffea JMF |
1770 | def process_video_result(self, info_dict, download=True): |
1771 | assert info_dict.get('_type', 'video') == 'video' | |
1772 | ||
bec1fad2 PH |
1773 | if 'id' not in info_dict: |
1774 | raise ExtractorError('Missing "id" field in extractor result') | |
1775 | if 'title' not in info_dict: | |
1776 | raise ExtractorError('Missing "title" field in extractor result') | |
1777 | ||
c9969434 S |
1778 | def report_force_conversion(field, field_not, conversion): |
1779 | self.report_warning( | |
1780 | '"%s" field is not %s - forcing %s conversion, there is an error in extractor' | |
1781 | % (field, field_not, conversion)) | |
1782 | ||
1783 | def sanitize_string_field(info, string_field): | |
1784 | field = info.get(string_field) | |
1785 | if field is None or isinstance(field, compat_str): | |
1786 | return | |
1787 | report_force_conversion(string_field, 'a string', 'string') | |
1788 | info[string_field] = compat_str(field) | |
1789 | ||
1790 | def sanitize_numeric_fields(info): | |
1791 | for numeric_field in self._NUMERIC_FIELDS: | |
1792 | field = info.get(numeric_field) | |
1793 | if field is None or isinstance(field, compat_numeric_types): | |
1794 | continue | |
1795 | report_force_conversion(numeric_field, 'numeric', 'int') | |
1796 | info[numeric_field] = int_or_none(field) | |
1797 | ||
1798 | sanitize_string_field(info_dict, 'id') | |
1799 | sanitize_numeric_fields(info_dict) | |
be6217b2 | 1800 | |
dd82ffea JMF |
1801 | if 'playlist' not in info_dict: |
1802 | # It isn't part of a playlist | |
1803 | info_dict['playlist'] = None | |
1804 | info_dict['playlist_index'] = None | |
1805 | ||
d5519808 | 1806 | thumbnails = info_dict.get('thumbnails') |
cfb56d1a PH |
1807 | if thumbnails is None: |
1808 | thumbnail = info_dict.get('thumbnail') | |
1809 | if thumbnail: | |
a7a14d95 | 1810 | info_dict['thumbnails'] = thumbnails = [{'url': thumbnail}] |
d5519808 | 1811 | if thumbnails: |
be6d7229 | 1812 | thumbnails.sort(key=lambda t: ( |
d37708fc RA |
1813 | t.get('preference') if t.get('preference') is not None else -1, |
1814 | t.get('width') if t.get('width') is not None else -1, | |
1815 | t.get('height') if t.get('height') is not None else -1, | |
1816 | t.get('id') if t.get('id') is not None else '', t.get('url'))) | |
f6c24009 | 1817 | for i, t in enumerate(thumbnails): |
dcf77cf1 | 1818 | t['url'] = sanitize_url(t['url']) |
9603e8a7 | 1819 | if t.get('width') and t.get('height'): |
d5519808 | 1820 | t['resolution'] = '%dx%d' % (t['width'], t['height']) |
f6c24009 PH |
1821 | if t.get('id') is None: |
1822 | t['id'] = '%d' % i | |
d5519808 | 1823 | |
b7b72db9 | 1824 | if self.params.get('list_thumbnails'): |
1825 | self.list_thumbnails(info_dict) | |
1826 | return | |
1827 | ||
536a55da S |
1828 | thumbnail = info_dict.get('thumbnail') |
1829 | if thumbnail: | |
1830 | info_dict['thumbnail'] = sanitize_url(thumbnail) | |
1831 | elif thumbnails: | |
d5519808 PH |
1832 | info_dict['thumbnail'] = thumbnails[-1]['url'] |
1833 | ||
c9ae7b95 | 1834 | if 'display_id' not in info_dict and 'id' in info_dict: |
0afef30b PH |
1835 | info_dict['display_id'] = info_dict['id'] |
1836 | ||
10db0d2f | 1837 | for ts_key, date_key in ( |
1838 | ('timestamp', 'upload_date'), | |
1839 | ('release_timestamp', 'release_date'), | |
1840 | ): | |
1841 | if info_dict.get(date_key) is None and info_dict.get(ts_key) is not None: | |
1842 | # Working around out-of-range timestamp values (e.g. negative ones on Windows, | |
1843 | # see http://bugs.python.org/issue1646728) | |
1844 | try: | |
1845 | upload_date = datetime.datetime.utcfromtimestamp(info_dict[ts_key]) | |
1846 | info_dict[date_key] = upload_date.strftime('%Y%m%d') | |
1847 | except (ValueError, OverflowError, OSError): | |
1848 | pass | |
9d2ecdbc | 1849 | |
33d2fc2f S |
1850 | # Auto generate title fields corresponding to the *_number fields when missing |
1851 | # in order to always have clean titles. This is very common for TV series. | |
1852 | for field in ('chapter', 'season', 'episode'): | |
1853 | if info_dict.get('%s_number' % field) is not None and not info_dict.get(field): | |
1854 | info_dict[field] = '%s %d' % (field.capitalize(), info_dict['%s_number' % field]) | |
1855 | ||
05108a49 S |
1856 | for cc_kind in ('subtitles', 'automatic_captions'): |
1857 | cc = info_dict.get(cc_kind) | |
1858 | if cc: | |
1859 | for _, subtitle in cc.items(): | |
1860 | for subtitle_format in subtitle: | |
1861 | if subtitle_format.get('url'): | |
1862 | subtitle_format['url'] = sanitize_url(subtitle_format['url']) | |
1863 | if subtitle_format.get('ext') is None: | |
1864 | subtitle_format['ext'] = determine_ext(subtitle_format['url']).lower() | |
1865 | ||
1866 | automatic_captions = info_dict.get('automatic_captions') | |
4bba3716 | 1867 | subtitles = info_dict.get('subtitles') |
4bba3716 | 1868 | |
a504ced0 | 1869 | if self.params.get('listsubtitles', False): |
360e1ca5 | 1870 | if 'automatic_captions' in info_dict: |
05108a49 S |
1871 | self.list_subtitles( |
1872 | info_dict['id'], automatic_captions, 'automatic captions') | |
4bba3716 | 1873 | self.list_subtitles(info_dict['id'], subtitles, 'subtitles') |
a504ced0 | 1874 | return |
05108a49 | 1875 | |
360e1ca5 | 1876 | info_dict['requested_subtitles'] = self.process_subtitles( |
05108a49 | 1877 | info_dict['id'], subtitles, automatic_captions) |
a504ced0 | 1878 | |
dd82ffea JMF |
1879 | # We now pick which formats have to be downloaded |
1880 | if info_dict.get('formats') is None: | |
1881 | # There's only one format available | |
1882 | formats = [info_dict] | |
1883 | else: | |
1884 | formats = info_dict['formats'] | |
1885 | ||
db95dc13 PH |
1886 | if not formats: |
1887 | raise ExtractorError('No video formats found!') | |
1888 | ||
73af5cc8 S |
1889 | def is_wellformed(f): |
1890 | url = f.get('url') | |
a5ac0c47 | 1891 | if not url: |
73af5cc8 S |
1892 | self.report_warning( |
1893 | '"url" field is missing or empty - skipping format, ' | |
1894 | 'there is an error in extractor') | |
a5ac0c47 S |
1895 | return False |
1896 | if isinstance(url, bytes): | |
1897 | sanitize_string_field(f, 'url') | |
1898 | return True | |
73af5cc8 S |
1899 | |
1900 | # Filter out malformed formats for better extraction robustness | |
1901 | formats = list(filter(is_wellformed, formats)) | |
1902 | ||
181c7053 S |
1903 | formats_dict = {} |
1904 | ||
dd82ffea | 1905 | # We check that all the formats have the format and format_id fields |
db95dc13 | 1906 | for i, format in enumerate(formats): |
c9969434 S |
1907 | sanitize_string_field(format, 'format_id') |
1908 | sanitize_numeric_fields(format) | |
dcf77cf1 | 1909 | format['url'] = sanitize_url(format['url']) |
e74e3b63 | 1910 | if not format.get('format_id'): |
8016c922 | 1911 | format['format_id'] = compat_str(i) |
e2effb08 S |
1912 | else: |
1913 | # Sanitize format_id from characters used in format selector expression | |
ec85ded8 | 1914 | format['format_id'] = re.sub(r'[\s,/+\[\]()]', '_', format['format_id']) |
181c7053 S |
1915 | format_id = format['format_id'] |
1916 | if format_id not in formats_dict: | |
1917 | formats_dict[format_id] = [] | |
1918 | formats_dict[format_id].append(format) | |
1919 | ||
1920 | # Make sure all formats have unique format_id | |
1921 | for format_id, ambiguous_formats in formats_dict.items(): | |
1922 | if len(ambiguous_formats) > 1: | |
1923 | for i, format in enumerate(ambiguous_formats): | |
1924 | format['format_id'] = '%s-%d' % (format_id, i) | |
1925 | ||
1926 | for i, format in enumerate(formats): | |
8c51aa65 | 1927 | if format.get('format') is None: |
6febd1c1 | 1928 | format['format'] = '{id} - {res}{note}'.format( |
8c51aa65 JMF |
1929 | id=format['format_id'], |
1930 | res=self.format_resolution(format), | |
6febd1c1 | 1931 | note=' ({0})'.format(format['format_note']) if format.get('format_note') is not None else '', |
8c51aa65 | 1932 | ) |
c1002e96 | 1933 | # Automatically determine file extension if missing |
5b1d8575 | 1934 | if format.get('ext') is None: |
cce929ea | 1935 | format['ext'] = determine_ext(format['url']).lower() |
b5559424 S |
1936 | # Automatically determine protocol if missing (useful for format |
1937 | # selection purposes) | |
6f0be937 | 1938 | if format.get('protocol') is None: |
b5559424 | 1939 | format['protocol'] = determine_protocol(format) |
e5660ee6 JMF |
1940 | # Add HTTP headers, so that external programs can use them from the |
1941 | # json output | |
1942 | full_format_info = info_dict.copy() | |
1943 | full_format_info.update(format) | |
1944 | format['http_headers'] = self._calc_headers(full_format_info) | |
0016b84e S |
1945 | # Remove private housekeeping stuff |
1946 | if '__x_forwarded_for_ip' in info_dict: | |
1947 | del info_dict['__x_forwarded_for_ip'] | |
dd82ffea | 1948 | |
4bcc7bd1 | 1949 | # TODO Central sorting goes here |
99e206d5 | 1950 | |
f89197d7 | 1951 | if formats[0] is not info_dict: |
b3d9ef88 JMF |
1952 | # only set the 'formats' fields if the original info_dict list them |
1953 | # otherwise we end up with a circular reference, the first (and unique) | |
f89197d7 | 1954 | # element in the 'formats' field in info_dict is info_dict itself, |
dfb1b146 | 1955 | # which can't be exported to json |
b3d9ef88 | 1956 | info_dict['formats'] = formats |
cfb56d1a | 1957 | if self.params.get('listformats'): |
bfaae0a7 | 1958 | self.list_formats(info_dict) |
1959 | return | |
1960 | ||
de3ef3ed | 1961 | req_format = self.params.get('format') |
a9c58ad9 | 1962 | if req_format is None: |
0017d9ad S |
1963 | req_format = self._default_format_spec(info_dict, download=download) |
1964 | if self.params.get('verbose'): | |
e8be92f9 | 1965 | self.to_screen('[debug] Default format spec: %s' % req_format) |
0017d9ad | 1966 | |
5acfa126 | 1967 | format_selector = self.build_format_selector(req_format) |
317f7ab6 S |
1968 | |
1969 | # While in format selection we may need to have an access to the original | |
1970 | # format set in order to calculate some metrics or do some processing. | |
1971 | # For now we need to be able to guess whether original formats provided | |
1972 | # by extractor are incomplete or not (i.e. whether extractor provides only | |
1973 | # video-only or audio-only formats) for proper formats selection for | |
1974 | # extractors with such incomplete formats (see | |
067aa17e | 1975 | # https://github.com/ytdl-org/youtube-dl/pull/5556). |
317f7ab6 S |
1976 | # Since formats may be filtered during format selection and may not match |
1977 | # the original formats the results may be incorrect. Thus original formats | |
1978 | # or pre-calculated metrics should be passed to format selection routines | |
1979 | # as well. | |
1980 | # We will pass a context object containing all necessary additional data | |
1981 | # instead of just formats. | |
1982 | # This fixes incorrect format selection issue (see | |
067aa17e | 1983 | # https://github.com/ytdl-org/youtube-dl/issues/10083). |
2e221ca3 | 1984 | incomplete_formats = ( |
317f7ab6 | 1985 | # All formats are video-only or |
3089bc74 | 1986 | all(f.get('vcodec') != 'none' and f.get('acodec') == 'none' for f in formats) |
317f7ab6 | 1987 | # all formats are audio-only |
3089bc74 | 1988 | or all(f.get('vcodec') == 'none' and f.get('acodec') != 'none' for f in formats)) |
317f7ab6 S |
1989 | |
1990 | ctx = { | |
1991 | 'formats': formats, | |
1992 | 'incomplete_formats': incomplete_formats, | |
1993 | } | |
1994 | ||
1995 | formats_to_download = list(format_selector(ctx)) | |
dd82ffea | 1996 | if not formats_to_download: |
6febd1c1 | 1997 | raise ExtractorError('requested format not available', |
78a3a9f8 | 1998 | expected=True) |
dd82ffea JMF |
1999 | |
2000 | if download: | |
909d24dd | 2001 | self.to_screen('[info] Downloading format(s) %s' % ", ".join([f['format_id'] for f in formats_to_download])) |
dd82ffea | 2002 | if len(formats_to_download) > 1: |
6febd1c1 | 2003 | self.to_screen('[info] %s: downloading video in %s formats' % (info_dict['id'], len(formats_to_download))) |
dd82ffea JMF |
2004 | for format in formats_to_download: |
2005 | new_info = dict(info_dict) | |
2006 | new_info.update(format) | |
2007 | self.process_info(new_info) | |
2008 | # We update the info dict with the best quality format (backwards compatibility) | |
2009 | info_dict.update(formats_to_download[-1]) | |
2010 | return info_dict | |
2011 | ||
98c70d6f | 2012 | def process_subtitles(self, video_id, normal_subtitles, automatic_captions): |
a504ced0 | 2013 | """Select the requested subtitles and their format""" |
98c70d6f JMF |
2014 | available_subs = {} |
2015 | if normal_subtitles and self.params.get('writesubtitles'): | |
2016 | available_subs.update(normal_subtitles) | |
2017 | if automatic_captions and self.params.get('writeautomaticsub'): | |
2018 | for lang, cap_info in automatic_captions.items(): | |
360e1ca5 JMF |
2019 | if lang not in available_subs: |
2020 | available_subs[lang] = cap_info | |
2021 | ||
4d171848 JMF |
2022 | if (not self.params.get('writesubtitles') and not |
2023 | self.params.get('writeautomaticsub') or not | |
2024 | available_subs): | |
2025 | return None | |
a504ced0 JMF |
2026 | |
2027 | if self.params.get('allsubtitles', False): | |
2028 | requested_langs = available_subs.keys() | |
2029 | else: | |
2030 | if self.params.get('subtitleslangs', False): | |
2031 | requested_langs = self.params.get('subtitleslangs') | |
2032 | elif 'en' in available_subs: | |
2033 | requested_langs = ['en'] | |
2034 | else: | |
2035 | requested_langs = [list(available_subs.keys())[0]] | |
2036 | ||
2037 | formats_query = self.params.get('subtitlesformat', 'best') | |
2038 | formats_preference = formats_query.split('/') if formats_query else [] | |
2039 | subs = {} | |
2040 | for lang in requested_langs: | |
2041 | formats = available_subs.get(lang) | |
2042 | if formats is None: | |
2043 | self.report_warning('%s subtitles not available for %s' % (lang, video_id)) | |
2044 | continue | |
a504ced0 JMF |
2045 | for ext in formats_preference: |
2046 | if ext == 'best': | |
2047 | f = formats[-1] | |
2048 | break | |
2049 | matches = list(filter(lambda f: f['ext'] == ext, formats)) | |
2050 | if matches: | |
2051 | f = matches[-1] | |
2052 | break | |
2053 | else: | |
2054 | f = formats[-1] | |
2055 | self.report_warning( | |
2056 | 'No subtitle format found matching "%s" for language %s, ' | |
2057 | 'using %s' % (formats_query, lang, f['ext'])) | |
2058 | subs[lang] = f | |
2059 | return subs | |
2060 | ||
d06daf23 S |
2061 | def __forced_printings(self, info_dict, filename, incomplete): |
2062 | def print_mandatory(field): | |
2063 | if (self.params.get('force%s' % field, False) | |
2064 | and (not incomplete or info_dict.get(field) is not None)): | |
2065 | self.to_stdout(info_dict[field]) | |
2066 | ||
2067 | def print_optional(field): | |
2068 | if (self.params.get('force%s' % field, False) | |
2069 | and info_dict.get(field) is not None): | |
2070 | self.to_stdout(info_dict[field]) | |
2071 | ||
2072 | print_mandatory('title') | |
2073 | print_mandatory('id') | |
2074 | if self.params.get('forceurl', False) and not incomplete: | |
2075 | if info_dict.get('requested_formats') is not None: | |
2076 | for f in info_dict['requested_formats']: | |
2077 | self.to_stdout(f['url'] + f.get('play_path', '')) | |
2078 | else: | |
2079 | # For RTMP URLs, also include the playpath | |
2080 | self.to_stdout(info_dict['url'] + info_dict.get('play_path', '')) | |
2081 | print_optional('thumbnail') | |
2082 | print_optional('description') | |
2083 | if self.params.get('forcefilename', False) and filename is not None: | |
2084 | self.to_stdout(filename) | |
2085 | if self.params.get('forceduration', False) and info_dict.get('duration') is not None: | |
2086 | self.to_stdout(formatSeconds(info_dict['duration'])) | |
2087 | print_mandatory('format') | |
2088 | if self.params.get('forcejson', False): | |
277d6ff5 | 2089 | self.post_extract(info_dict) |
75d43ca0 | 2090 | self.to_stdout(json.dumps(info_dict, default=repr)) |
d06daf23 | 2091 | |
8222d8de JMF |
2092 | def process_info(self, info_dict): |
2093 | """Process a single resolved IE result.""" | |
2094 | ||
2095 | assert info_dict.get('_type', 'video') == 'video' | |
fd288278 | 2096 | |
0202b52a | 2097 | info_dict.setdefault('__postprocessors', []) |
2098 | ||
fd288278 PH |
2099 | max_downloads = self.params.get('max_downloads') |
2100 | if max_downloads is not None: | |
2101 | if self._num_downloads >= int(max_downloads): | |
2102 | raise MaxDownloadsReached() | |
8222d8de | 2103 | |
d06daf23 | 2104 | # TODO: backward compatibility, to be removed |
8222d8de | 2105 | info_dict['fulltitle'] = info_dict['title'] |
8222d8de | 2106 | |
11b85ce6 | 2107 | if 'format' not in info_dict: |
8222d8de JMF |
2108 | info_dict['format'] = info_dict['ext'] |
2109 | ||
8b0d7497 | 2110 | if self._match_entry(info_dict, incomplete=False) is not None: |
8222d8de JMF |
2111 | return |
2112 | ||
277d6ff5 | 2113 | self.post_extract(info_dict) |
fd288278 | 2114 | self._num_downloads += 1 |
8222d8de | 2115 | |
56d868db | 2116 | info_dict, _ = self.pre_process(info_dict) |
5bfa4862 | 2117 | |
dcf64d43 | 2118 | # info_dict['_filename'] needs to be set for backward compatibility |
de6000d9 | 2119 | info_dict['_filename'] = full_filename = self.prepare_filename(info_dict, warn=True) |
2120 | temp_filename = self.prepare_filename(info_dict, 'temp') | |
0202b52a | 2121 | files_to_move = {} |
8222d8de JMF |
2122 | |
2123 | # Forced printings | |
0202b52a | 2124 | self.__forced_printings(info_dict, full_filename, incomplete=False) |
8222d8de | 2125 | |
8222d8de | 2126 | if self.params.get('simulate', False): |
2d30509f | 2127 | if self.params.get('force_write_download_archive', False): |
2128 | self.record_download_archive(info_dict) | |
2129 | ||
2130 | # Do nothing else if in simulate mode | |
8222d8de JMF |
2131 | return |
2132 | ||
de6000d9 | 2133 | if full_filename is None: |
8222d8de JMF |
2134 | return |
2135 | ||
e92caff5 | 2136 | if not self._ensure_dir_exists(encodeFilename(full_filename)): |
0202b52a | 2137 | return |
e92caff5 | 2138 | if not self._ensure_dir_exists(encodeFilename(temp_filename)): |
8222d8de JMF |
2139 | return |
2140 | ||
2141 | if self.params.get('writedescription', False): | |
de6000d9 | 2142 | descfn = self.prepare_filename(info_dict, 'description') |
e92caff5 | 2143 | if not self._ensure_dir_exists(encodeFilename(descfn)): |
0202b52a | 2144 | return |
0c3d0f51 | 2145 | if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(descfn)): |
6febd1c1 | 2146 | self.to_screen('[info] Video description is already present') |
f00fd51d JMF |
2147 | elif info_dict.get('description') is None: |
2148 | self.report_warning('There\'s no description to write.') | |
7b6fefc9 PH |
2149 | else: |
2150 | try: | |
6febd1c1 | 2151 | self.to_screen('[info] Writing video description to: ' + descfn) |
7b6fefc9 PH |
2152 | with io.open(encodeFilename(descfn), 'w', encoding='utf-8') as descfile: |
2153 | descfile.write(info_dict['description']) | |
7b6fefc9 | 2154 | except (OSError, IOError): |
6febd1c1 | 2155 | self.report_error('Cannot write description file ' + descfn) |
7b6fefc9 | 2156 | return |
8222d8de | 2157 | |
1fb07d10 | 2158 | if self.params.get('writeannotations', False): |
de6000d9 | 2159 | annofn = self.prepare_filename(info_dict, 'annotation') |
e92caff5 | 2160 | if not self._ensure_dir_exists(encodeFilename(annofn)): |
0202b52a | 2161 | return |
0c3d0f51 | 2162 | if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(annofn)): |
6febd1c1 | 2163 | self.to_screen('[info] Video annotations are already present') |
ffddb112 RA |
2164 | elif not info_dict.get('annotations'): |
2165 | self.report_warning('There are no annotations to write.') | |
7b6fefc9 PH |
2166 | else: |
2167 | try: | |
6febd1c1 | 2168 | self.to_screen('[info] Writing video annotations to: ' + annofn) |
7b6fefc9 PH |
2169 | with io.open(encodeFilename(annofn), 'w', encoding='utf-8') as annofile: |
2170 | annofile.write(info_dict['annotations']) | |
2171 | except (KeyError, TypeError): | |
6febd1c1 | 2172 | self.report_warning('There are no annotations to write.') |
7b6fefc9 | 2173 | except (OSError, IOError): |
6febd1c1 | 2174 | self.report_error('Cannot write annotations file: ' + annofn) |
7b6fefc9 | 2175 | return |
1fb07d10 | 2176 | |
9f448fcb | 2177 | def dl(name, info, subtitle=False): |
98b69821 | 2178 | fd = get_suitable_downloader(info, self.params)(self, self.params) |
2179 | for ph in self._progress_hooks: | |
2180 | fd.add_progress_hook(ph) | |
2181 | if self.params.get('verbose'): | |
29f7c58a | 2182 | self.to_screen('[debug] Invoking downloader on %r' % info.get('url')) |
46906886 DA |
2183 | new_info = dict(info) |
2184 | if new_info.get('http_headers') is None: | |
2185 | new_info['http_headers'] = self._calc_headers(new_info) | |
2186 | return fd.download(name, new_info, subtitle) | |
98b69821 | 2187 | |
c4a91be7 | 2188 | subtitles_are_requested = any([self.params.get('writesubtitles', False), |
0b7f3118 | 2189 | self.params.get('writeautomaticsub')]) |
c4a91be7 | 2190 | |
c84dd8a9 | 2191 | if subtitles_are_requested and info_dict.get('requested_subtitles'): |
8222d8de JMF |
2192 | # subtitles download errors are already managed as troubles in relevant IE |
2193 | # that way it will silently go on when used with unsupporting IE | |
c84dd8a9 | 2194 | subtitles = info_dict['requested_subtitles'] |
fa57af1e | 2195 | # ie = self.get_info_extractor(info_dict['extractor_key']) |
a504ced0 JMF |
2196 | for sub_lang, sub_info in subtitles.items(): |
2197 | sub_format = sub_info['ext'] | |
56d868db | 2198 | sub_filename = subtitles_filename(temp_filename, sub_lang, sub_format, info_dict.get('ext')) |
2199 | sub_filename_final = subtitles_filename( | |
2200 | self.prepare_filename(info_dict, 'subtitle'), sub_lang, sub_format, info_dict.get('ext')) | |
0c3d0f51 | 2201 | if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(sub_filename)): |
5ff1bc0c | 2202 | self.to_screen('[info] Video subtitle %s.%s is already present' % (sub_lang, sub_format)) |
dcf64d43 | 2203 | sub_info['filepath'] = sub_filename |
0202b52a | 2204 | files_to_move[sub_filename] = sub_filename_final |
a504ced0 | 2205 | else: |
0c9df79e | 2206 | self.to_screen('[info] Writing video subtitles to: ' + sub_filename) |
5ff1bc0c RA |
2207 | if sub_info.get('data') is not None: |
2208 | try: | |
2209 | # Use newline='' to prevent conversion of newline characters | |
067aa17e | 2210 | # See https://github.com/ytdl-org/youtube-dl/issues/10268 |
5ff1bc0c RA |
2211 | with io.open(encodeFilename(sub_filename), 'w', encoding='utf-8', newline='') as subfile: |
2212 | subfile.write(sub_info['data']) | |
dcf64d43 | 2213 | sub_info['filepath'] = sub_filename |
0202b52a | 2214 | files_to_move[sub_filename] = sub_filename_final |
5ff1bc0c RA |
2215 | except (OSError, IOError): |
2216 | self.report_error('Cannot write subtitles file ' + sub_filename) | |
2217 | return | |
7b6fefc9 | 2218 | else: |
5ff1bc0c | 2219 | try: |
dcf64d43 | 2220 | dl(sub_filename, sub_info.copy(), subtitle=True) |
2221 | sub_info['filepath'] = sub_filename | |
0202b52a | 2222 | files_to_move[sub_filename] = sub_filename_final |
0c9df79e | 2223 | except (ExtractorError, IOError, OSError, ValueError, compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: |
5ff1bc0c RA |
2224 | self.report_warning('Unable to download subtitle for "%s": %s' % |
2225 | (sub_lang, error_to_compat_str(err))) | |
2226 | continue | |
8222d8de | 2227 | |
8222d8de | 2228 | if self.params.get('writeinfojson', False): |
de6000d9 | 2229 | infofn = self.prepare_filename(info_dict, 'infojson') |
e92caff5 | 2230 | if not self._ensure_dir_exists(encodeFilename(infofn)): |
0202b52a | 2231 | return |
0c3d0f51 | 2232 | if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(infofn)): |
66c935fb | 2233 | self.to_screen('[info] Video metadata is already present') |
7b6fefc9 | 2234 | else: |
66c935fb | 2235 | self.to_screen('[info] Writing video metadata as JSON to: ' + infofn) |
7b6fefc9 | 2236 | try: |
75d43ca0 | 2237 | write_json_file(self.filter_requested_info(info_dict, self.params.get('clean_infojson', True)), infofn) |
7b6fefc9 | 2238 | except (OSError, IOError): |
66c935fb | 2239 | self.report_error('Cannot write video metadata to JSON file ' + infofn) |
7b6fefc9 | 2240 | return |
de6000d9 | 2241 | info_dict['__infojson_filename'] = infofn |
8222d8de | 2242 | |
56d868db | 2243 | for thumb_ext in self._write_thumbnails(info_dict, temp_filename): |
2244 | thumb_filename_temp = replace_extension(temp_filename, thumb_ext, info_dict.get('ext')) | |
2245 | thumb_filename = replace_extension( | |
2246 | self.prepare_filename(info_dict, 'thumbnail'), thumb_ext, info_dict.get('ext')) | |
dcf64d43 | 2247 | files_to_move[thumb_filename_temp] = thumb_filename |
8222d8de | 2248 | |
732044af | 2249 | # Write internet shortcut files |
2250 | url_link = webloc_link = desktop_link = False | |
2251 | if self.params.get('writelink', False): | |
2252 | if sys.platform == "darwin": # macOS. | |
2253 | webloc_link = True | |
2254 | elif sys.platform.startswith("linux"): | |
2255 | desktop_link = True | |
2256 | else: # if sys.platform in ['win32', 'cygwin']: | |
2257 | url_link = True | |
2258 | if self.params.get('writeurllink', False): | |
2259 | url_link = True | |
2260 | if self.params.get('writewebloclink', False): | |
2261 | webloc_link = True | |
2262 | if self.params.get('writedesktoplink', False): | |
2263 | desktop_link = True | |
2264 | ||
2265 | if url_link or webloc_link or desktop_link: | |
2266 | if 'webpage_url' not in info_dict: | |
2267 | self.report_error('Cannot write internet shortcut file because the "webpage_url" field is missing in the media information') | |
2268 | return | |
2269 | ascii_url = iri_to_uri(info_dict['webpage_url']) | |
2270 | ||
2271 | def _write_link_file(extension, template, newline, embed_filename): | |
0202b52a | 2272 | linkfn = replace_extension(full_filename, extension, info_dict.get('ext')) |
10e3742e | 2273 | if self.params.get('overwrites', True) and os.path.exists(encodeFilename(linkfn)): |
732044af | 2274 | self.to_screen('[info] Internet shortcut is already present') |
2275 | else: | |
2276 | try: | |
2277 | self.to_screen('[info] Writing internet shortcut to: ' + linkfn) | |
2278 | with io.open(encodeFilename(to_high_limit_path(linkfn)), 'w', encoding='utf-8', newline=newline) as linkfile: | |
2279 | template_vars = {'url': ascii_url} | |
2280 | if embed_filename: | |
2281 | template_vars['filename'] = linkfn[:-(len(extension) + 1)] | |
2282 | linkfile.write(template % template_vars) | |
2283 | except (OSError, IOError): | |
2284 | self.report_error('Cannot write internet shortcut ' + linkfn) | |
2285 | return False | |
2286 | return True | |
2287 | ||
2288 | if url_link: | |
2289 | if not _write_link_file('url', DOT_URL_LINK_TEMPLATE, '\r\n', embed_filename=False): | |
2290 | return | |
2291 | if webloc_link: | |
2292 | if not _write_link_file('webloc', DOT_WEBLOC_LINK_TEMPLATE, '\n', embed_filename=False): | |
2293 | return | |
2294 | if desktop_link: | |
2295 | if not _write_link_file('desktop', DOT_DESKTOP_LINK_TEMPLATE, '\n', embed_filename=True): | |
2296 | return | |
2297 | ||
56d868db | 2298 | try: |
2299 | info_dict, files_to_move = self.pre_process(info_dict, 'before_dl', files_to_move) | |
2300 | except PostProcessingError as err: | |
2301 | self.report_error('Preprocessing: %s' % str(err)) | |
2302 | return | |
2303 | ||
732044af | 2304 | must_record_download_archive = False |
56d868db | 2305 | if self.params.get('skip_download', False): |
2306 | info_dict['filepath'] = temp_filename | |
2307 | info_dict['__finaldir'] = os.path.dirname(os.path.abspath(encodeFilename(full_filename))) | |
2308 | info_dict['__files_to_move'] = files_to_move | |
2309 | info_dict = self.run_pp(MoveFilesAfterDownloadPP(self, False), info_dict) | |
2310 | else: | |
2311 | # Download | |
4340deca | 2312 | try: |
0202b52a | 2313 | |
6b591b29 | 2314 | def existing_file(*filepaths): |
2315 | ext = info_dict.get('ext') | |
2316 | final_ext = self.params.get('final_ext', ext) | |
2317 | existing_files = [] | |
2318 | for file in orderedSet(filepaths): | |
2319 | if final_ext != ext: | |
2320 | converted = replace_extension(file, final_ext, ext) | |
2321 | if os.path.exists(encodeFilename(converted)): | |
2322 | existing_files.append(converted) | |
2323 | if os.path.exists(encodeFilename(file)): | |
2324 | existing_files.append(file) | |
2325 | ||
2326 | if not existing_files or self.params.get('overwrites', False): | |
2327 | for file in orderedSet(existing_files): | |
2328 | self.report_file_delete(file) | |
2329 | os.remove(encodeFilename(file)) | |
2330 | return None | |
2331 | ||
2332 | self.report_file_already_downloaded(existing_files[0]) | |
2333 | info_dict['ext'] = os.path.splitext(existing_files[0])[1][1:] | |
2334 | return existing_files[0] | |
0202b52a | 2335 | |
2336 | success = True | |
4340deca P |
2337 | if info_dict.get('requested_formats') is not None: |
2338 | downloaded = [] | |
d47aeb22 | 2339 | merger = FFmpegMergerPP(self) |
63ad4d43 | 2340 | if self.params.get('allow_unplayable_formats'): |
2341 | self.report_warning( | |
2342 | 'You have requested merging of multiple formats ' | |
2343 | 'while also allowing unplayable formats to be downloaded. ' | |
2344 | 'The formats won\'t be merged to prevent data corruption.') | |
2345 | elif not merger.available: | |
2346 | self.report_warning( | |
2347 | 'You have requested merging of multiple formats but ffmpeg is not installed. ' | |
2348 | 'The formats won\'t be merged.') | |
81cd954a S |
2349 | |
2350 | def compatible_formats(formats): | |
d03cfdce | 2351 | # TODO: some formats actually allow this (mkv, webm, ogg, mp4), but not all of them. |
2352 | video_formats = [format for format in formats if format.get('vcodec') != 'none'] | |
2353 | audio_formats = [format for format in formats if format.get('acodec') != 'none'] | |
2354 | if len(video_formats) > 2 or len(audio_formats) > 2: | |
2355 | return False | |
2356 | ||
81cd954a | 2357 | # Check extension |
d03cfdce | 2358 | exts = set(format.get('ext') for format in formats) |
2359 | COMPATIBLE_EXTS = ( | |
2360 | set(('mp3', 'mp4', 'm4a', 'm4p', 'm4b', 'm4r', 'm4v', 'ismv', 'isma')), | |
2361 | set(('webm',)), | |
2362 | ) | |
2363 | for ext_sets in COMPATIBLE_EXTS: | |
2364 | if ext_sets.issuperset(exts): | |
2365 | return True | |
81cd954a S |
2366 | # TODO: Check acodec/vcodec |
2367 | return False | |
2368 | ||
2369 | requested_formats = info_dict['requested_formats'] | |
0202b52a | 2370 | old_ext = info_dict['ext'] |
4d971a16 | 2371 | if self.params.get('merge_output_format') is None: |
2372 | if not compatible_formats(requested_formats): | |
2373 | info_dict['ext'] = 'mkv' | |
2374 | self.report_warning( | |
2375 | 'Requested formats are incompatible for merge and will be merged into mkv.') | |
2376 | if (info_dict['ext'] == 'webm' | |
2377 | and self.params.get('writethumbnail', False) | |
2378 | and info_dict.get('thumbnails')): | |
2379 | info_dict['ext'] = 'mkv' | |
2380 | self.report_warning( | |
2381 | 'webm doesn\'t support embedding a thumbnail, mkv will be used.') | |
0202b52a | 2382 | |
2383 | def correct_ext(filename): | |
2384 | filename_real_ext = os.path.splitext(filename)[1][1:] | |
2385 | filename_wo_ext = ( | |
2386 | os.path.splitext(filename)[0] | |
2387 | if filename_real_ext == old_ext | |
2388 | else filename) | |
2389 | return '%s.%s' % (filename_wo_ext, info_dict['ext']) | |
2390 | ||
38c6902b | 2391 | # Ensure filename always has a correct extension for successful merge |
0202b52a | 2392 | full_filename = correct_ext(full_filename) |
2393 | temp_filename = correct_ext(temp_filename) | |
2394 | dl_filename = existing_file(full_filename, temp_filename) | |
1ea24129 | 2395 | info_dict['__real_download'] = False |
0202b52a | 2396 | if dl_filename is None: |
81cd954a | 2397 | for f in requested_formats: |
5b5fbc08 JMF |
2398 | new_info = dict(info_dict) |
2399 | new_info.update(f) | |
c5c9bf0c | 2400 | fname = prepend_extension( |
de6000d9 | 2401 | self.prepare_filename(new_info, 'temp'), |
c5c9bf0c | 2402 | 'f%s' % f['format_id'], new_info['ext']) |
e92caff5 | 2403 | if not self._ensure_dir_exists(fname): |
c5c9bf0c | 2404 | return |
5b5fbc08 | 2405 | downloaded.append(fname) |
a9e7f546 | 2406 | partial_success, real_download = dl(fname, new_info) |
1ea24129 | 2407 | info_dict['__real_download'] = info_dict['__real_download'] or real_download |
5b5fbc08 | 2408 | success = success and partial_success |
63ad4d43 | 2409 | if merger.available and not self.params.get('allow_unplayable_formats'): |
efabc161 | 2410 | info_dict['__postprocessors'].append(merger) |
1ea24129 | 2411 | info_dict['__files_to_merge'] = downloaded |
2412 | # Even if there were no downloads, it is being merged only now | |
2413 | info_dict['__real_download'] = True | |
42bb0c59 | 2414 | else: |
2415 | for file in downloaded: | |
2416 | files_to_move[file] = None | |
4340deca P |
2417 | else: |
2418 | # Just a single file | |
0202b52a | 2419 | dl_filename = existing_file(full_filename, temp_filename) |
2420 | if dl_filename is None: | |
2421 | success, real_download = dl(temp_filename, info_dict) | |
2422 | info_dict['__real_download'] = real_download | |
2423 | ||
0202b52a | 2424 | dl_filename = dl_filename or temp_filename |
c571435f | 2425 | info_dict['__finaldir'] = os.path.dirname(os.path.abspath(encodeFilename(full_filename))) |
0202b52a | 2426 | |
4340deca | 2427 | except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: |
7960b056 | 2428 | self.report_error('unable to download video data: %s' % error_to_compat_str(err)) |
4340deca P |
2429 | return |
2430 | except (OSError, IOError) as err: | |
2431 | raise UnavailableVideoError(err) | |
2432 | except (ContentTooShortError, ) as err: | |
2433 | self.report_error('content too short (expected %s bytes and served %s)' % (err.expected, err.downloaded)) | |
2434 | return | |
8222d8de | 2435 | |
de6000d9 | 2436 | if success and full_filename != '-': |
6271f1ca | 2437 | # Fixup content |
62cd676c PH |
2438 | fixup_policy = self.params.get('fixup') |
2439 | if fixup_policy is None: | |
2440 | fixup_policy = 'detect_or_warn' | |
2441 | ||
e4172ac9 | 2442 | INSTALL_FFMPEG_MESSAGE = 'Install ffmpeg to fix this automatically.' |
d1e4a464 | 2443 | |
6271f1ca PH |
2444 | stretched_ratio = info_dict.get('stretched_ratio') |
2445 | if stretched_ratio is not None and stretched_ratio != 1: | |
6271f1ca PH |
2446 | if fixup_policy == 'warn': |
2447 | self.report_warning('%s: Non-uniform pixel ratio (%s)' % ( | |
2448 | info_dict['id'], stretched_ratio)) | |
2449 | elif fixup_policy == 'detect_or_warn': | |
2450 | stretched_pp = FFmpegFixupStretchedPP(self) | |
2451 | if stretched_pp.available: | |
6271f1ca PH |
2452 | info_dict['__postprocessors'].append(stretched_pp) |
2453 | else: | |
2454 | self.report_warning( | |
d1e4a464 S |
2455 | '%s: Non-uniform pixel ratio (%s). %s' |
2456 | % (info_dict['id'], stretched_ratio, INSTALL_FFMPEG_MESSAGE)) | |
6271f1ca | 2457 | else: |
62cd676c PH |
2458 | assert fixup_policy in ('ignore', 'never') |
2459 | ||
3089bc74 | 2460 | if (info_dict.get('requested_formats') is None |
6b591b29 | 2461 | and info_dict.get('container') == 'm4a_dash' |
2462 | and info_dict.get('ext') == 'm4a'): | |
62cd676c | 2463 | if fixup_policy == 'warn': |
d1e4a464 S |
2464 | self.report_warning( |
2465 | '%s: writing DASH m4a. ' | |
2466 | 'Only some players support this container.' | |
2467 | % info_dict['id']) | |
62cd676c PH |
2468 | elif fixup_policy == 'detect_or_warn': |
2469 | fixup_pp = FFmpegFixupM4aPP(self) | |
2470 | if fixup_pp.available: | |
62cd676c PH |
2471 | info_dict['__postprocessors'].append(fixup_pp) |
2472 | else: | |
2473 | self.report_warning( | |
d1e4a464 S |
2474 | '%s: writing DASH m4a. ' |
2475 | 'Only some players support this container. %s' | |
2476 | % (info_dict['id'], INSTALL_FFMPEG_MESSAGE)) | |
62cd676c PH |
2477 | else: |
2478 | assert fixup_policy in ('ignore', 'never') | |
6271f1ca | 2479 | |
0a473f2f | 2480 | if ('protocol' in info_dict |
2481 | and get_suitable_downloader(info_dict, self.params).__name__ == 'HlsFD'): | |
f17f8651 | 2482 | if fixup_policy == 'warn': |
a02682fd | 2483 | self.report_warning('%s: malformed AAC bitstream detected.' % ( |
f17f8651 | 2484 | info_dict['id'])) |
2485 | elif fixup_policy == 'detect_or_warn': | |
2486 | fixup_pp = FFmpegFixupM3u8PP(self) | |
2487 | if fixup_pp.available: | |
f17f8651 | 2488 | info_dict['__postprocessors'].append(fixup_pp) |
2489 | else: | |
2490 | self.report_warning( | |
a02682fd | 2491 | '%s: malformed AAC bitstream detected. %s' |
d1e4a464 | 2492 | % (info_dict['id'], INSTALL_FFMPEG_MESSAGE)) |
f17f8651 | 2493 | else: |
2494 | assert fixup_policy in ('ignore', 'never') | |
2495 | ||
8222d8de | 2496 | try: |
23c1a667 | 2497 | info_dict = self.post_process(dl_filename, info_dict, files_to_move) |
af819c21 | 2498 | except PostProcessingError as err: |
2499 | self.report_error('Postprocessing: %s' % str(err)) | |
8222d8de | 2500 | return |
ab8e5e51 AM |
2501 | try: |
2502 | for ph in self._post_hooks: | |
23c1a667 | 2503 | ph(info_dict['filepath']) |
ab8e5e51 AM |
2504 | except Exception as err: |
2505 | self.report_error('post hooks: %s' % str(err)) | |
2506 | return | |
2d30509f | 2507 | must_record_download_archive = True |
2508 | ||
2509 | if must_record_download_archive or self.params.get('force_write_download_archive', False): | |
2510 | self.record_download_archive(info_dict) | |
c3e6ffba | 2511 | max_downloads = self.params.get('max_downloads') |
2512 | if max_downloads is not None and self._num_downloads >= int(max_downloads): | |
2513 | raise MaxDownloadsReached() | |
8222d8de JMF |
2514 | |
2515 | def download(self, url_list): | |
2516 | """Download a given list of URLs.""" | |
de6000d9 | 2517 | outtmpl = self.outtmpl_dict['default'] |
3089bc74 S |
2518 | if (len(url_list) > 1 |
2519 | and outtmpl != '-' | |
2520 | and '%' not in outtmpl | |
2521 | and self.params.get('max_downloads') != 1): | |
acd69589 | 2522 | raise SameFileError(outtmpl) |
8222d8de JMF |
2523 | |
2524 | for url in url_list: | |
2525 | try: | |
5f6a1245 | 2526 | # It also downloads the videos |
61aa5ba3 S |
2527 | res = self.extract_info( |
2528 | url, force_generic_extractor=self.params.get('force_generic_extractor', False)) | |
8222d8de | 2529 | except UnavailableVideoError: |
6febd1c1 | 2530 | self.report_error('unable to download video') |
8222d8de | 2531 | except MaxDownloadsReached: |
8b0d7497 | 2532 | self.to_screen('[info] Maximum number of downloaded files reached') |
2533 | raise | |
2534 | except ExistingVideoReached: | |
d83cb531 | 2535 | self.to_screen('[info] Encountered a file that is already in the archive, stopping due to --break-on-existing') |
8b0d7497 | 2536 | raise |
2537 | except RejectedVideoReached: | |
d83cb531 | 2538 | self.to_screen('[info] Encountered a file that did not match filter, stopping due to --break-on-reject') |
8222d8de | 2539 | raise |
63e0be34 PH |
2540 | else: |
2541 | if self.params.get('dump_single_json', False): | |
277d6ff5 | 2542 | self.post_extract(res) |
75d43ca0 | 2543 | self.to_stdout(json.dumps(res, default=repr)) |
8222d8de JMF |
2544 | |
2545 | return self._download_retcode | |
2546 | ||
1dcc4c0c | 2547 | def download_with_info_file(self, info_filename): |
31bd3925 JMF |
2548 | with contextlib.closing(fileinput.FileInput( |
2549 | [info_filename], mode='r', | |
2550 | openhook=fileinput.hook_encoded('utf-8'))) as f: | |
2551 | # FileInput doesn't have a read method, we can't call json.load | |
498f5606 | 2552 | info = self.filter_requested_info(json.loads('\n'.join(f)), self.params.get('clean_infojson', True)) |
d4943898 JMF |
2553 | try: |
2554 | self.process_ie_result(info, download=True) | |
498f5606 | 2555 | except (DownloadError, EntryNotInPlaylist): |
d4943898 JMF |
2556 | webpage_url = info.get('webpage_url') |
2557 | if webpage_url is not None: | |
6febd1c1 | 2558 | self.report_warning('The info failed to download, trying with "%s"' % webpage_url) |
d4943898 JMF |
2559 | return self.download([webpage_url]) |
2560 | else: | |
2561 | raise | |
2562 | return self._download_retcode | |
1dcc4c0c | 2563 | |
cb202fd2 | 2564 | @staticmethod |
75d43ca0 | 2565 | def filter_requested_info(info_dict, actually_filter=True): |
2566 | if not actually_filter: | |
394dcd44 | 2567 | info_dict['epoch'] = int(time.time()) |
75d43ca0 | 2568 | return info_dict |
5226731e | 2569 | exceptions = { |
498f5606 | 2570 | 'remove': ['requested_formats', 'requested_subtitles', 'requested_entries', 'filepath', 'entries'], |
5226731e | 2571 | 'keep': ['_type'], |
2572 | } | |
2573 | keep_key = lambda k: k in exceptions['keep'] or not (k.startswith('_') or k in exceptions['remove']) | |
2574 | filter_fn = lambda obj: ( | |
a515a78d | 2575 | list(map(filter_fn, obj)) if isinstance(obj, (list, tuple)) |
2576 | else obj if not isinstance(obj, dict) | |
2577 | else dict((k, filter_fn(v)) for k, v in obj.items() if keep_key(k))) | |
5226731e | 2578 | return filter_fn(info_dict) |
cb202fd2 | 2579 | |
dcf64d43 | 2580 | def run_pp(self, pp, infodict): |
5bfa4862 | 2581 | files_to_delete = [] |
dcf64d43 | 2582 | if '__files_to_move' not in infodict: |
2583 | infodict['__files_to_move'] = {} | |
af819c21 | 2584 | files_to_delete, infodict = pp.run(infodict) |
5bfa4862 | 2585 | if not files_to_delete: |
dcf64d43 | 2586 | return infodict |
5bfa4862 | 2587 | |
2588 | if self.params.get('keepvideo', False): | |
2589 | for f in files_to_delete: | |
dcf64d43 | 2590 | infodict['__files_to_move'].setdefault(f, '') |
5bfa4862 | 2591 | else: |
2592 | for old_filename in set(files_to_delete): | |
2593 | self.to_screen('Deleting original file %s (pass -k to keep)' % old_filename) | |
2594 | try: | |
2595 | os.remove(encodeFilename(old_filename)) | |
2596 | except (IOError, OSError): | |
2597 | self.report_warning('Unable to remove downloaded original file') | |
dcf64d43 | 2598 | if old_filename in infodict['__files_to_move']: |
2599 | del infodict['__files_to_move'][old_filename] | |
2600 | return infodict | |
5bfa4862 | 2601 | |
277d6ff5 | 2602 | @staticmethod |
2603 | def post_extract(info_dict): | |
2604 | def actual_post_extract(info_dict): | |
2605 | if info_dict.get('_type') in ('playlist', 'multi_video'): | |
2606 | for video_dict in info_dict.get('entries', {}): | |
b050d210 | 2607 | actual_post_extract(video_dict or {}) |
277d6ff5 | 2608 | return |
2609 | ||
2610 | if '__post_extractor' not in info_dict: | |
2611 | return | |
2612 | post_extractor = info_dict['__post_extractor'] | |
2613 | if post_extractor: | |
2614 | info_dict.update(post_extractor().items()) | |
2615 | del info_dict['__post_extractor'] | |
2616 | return | |
2617 | ||
b050d210 | 2618 | actual_post_extract(info_dict or {}) |
277d6ff5 | 2619 | |
56d868db | 2620 | def pre_process(self, ie_info, key='pre_process', files_to_move=None): |
5bfa4862 | 2621 | info = dict(ie_info) |
56d868db | 2622 | info['__files_to_move'] = files_to_move or {} |
2623 | for pp in self._pps[key]: | |
dcf64d43 | 2624 | info = self.run_pp(pp, info) |
56d868db | 2625 | return info, info.pop('__files_to_move', None) |
5bfa4862 | 2626 | |
dcf64d43 | 2627 | def post_process(self, filename, ie_info, files_to_move=None): |
8222d8de JMF |
2628 | """Run all the postprocessors on the given file.""" |
2629 | info = dict(ie_info) | |
2630 | info['filepath'] = filename | |
dcf64d43 | 2631 | info['__files_to_move'] = files_to_move or {} |
0202b52a | 2632 | |
56d868db | 2633 | for pp in ie_info.get('__postprocessors', []) + self._pps['post_process']: |
dcf64d43 | 2634 | info = self.run_pp(pp, info) |
2635 | info = self.run_pp(MoveFilesAfterDownloadPP(self), info) | |
2636 | del info['__files_to_move'] | |
56d868db | 2637 | for pp in self._pps['after_move']: |
dcf64d43 | 2638 | info = self.run_pp(pp, info) |
23c1a667 | 2639 | return info |
c1c9a79c | 2640 | |
5db07df6 | 2641 | def _make_archive_id(self, info_dict): |
e9fef7ee S |
2642 | video_id = info_dict.get('id') |
2643 | if not video_id: | |
2644 | return | |
5db07df6 PH |
2645 | # Future-proof against any change in case |
2646 | # and backwards compatibility with prior versions | |
e9fef7ee | 2647 | extractor = info_dict.get('extractor_key') or info_dict.get('ie_key') # key in a playlist |
7012b23c | 2648 | if extractor is None: |
1211bb6d S |
2649 | url = str_or_none(info_dict.get('url')) |
2650 | if not url: | |
2651 | return | |
e9fef7ee S |
2652 | # Try to find matching extractor for the URL and take its ie_key |
2653 | for ie in self._ies: | |
1211bb6d | 2654 | if ie.suitable(url): |
e9fef7ee S |
2655 | extractor = ie.ie_key() |
2656 | break | |
2657 | else: | |
2658 | return | |
d0757229 | 2659 | return '%s %s' % (extractor.lower(), video_id) |
5db07df6 PH |
2660 | |
2661 | def in_download_archive(self, info_dict): | |
2662 | fn = self.params.get('download_archive') | |
2663 | if fn is None: | |
2664 | return False | |
2665 | ||
2666 | vid_id = self._make_archive_id(info_dict) | |
e9fef7ee | 2667 | if not vid_id: |
7012b23c | 2668 | return False # Incomplete video information |
5db07df6 | 2669 | |
a45e8619 | 2670 | return vid_id in self.archive |
c1c9a79c PH |
2671 | |
2672 | def record_download_archive(self, info_dict): | |
2673 | fn = self.params.get('download_archive') | |
2674 | if fn is None: | |
2675 | return | |
5db07df6 PH |
2676 | vid_id = self._make_archive_id(info_dict) |
2677 | assert vid_id | |
c1c9a79c | 2678 | with locked_file(fn, 'a', encoding='utf-8') as archive_file: |
6febd1c1 | 2679 | archive_file.write(vid_id + '\n') |
a45e8619 | 2680 | self.archive.add(vid_id) |
dd82ffea | 2681 | |
8c51aa65 | 2682 | @staticmethod |
8abeeb94 | 2683 | def format_resolution(format, default='unknown'): |
fb04e403 PH |
2684 | if format.get('vcodec') == 'none': |
2685 | return 'audio only' | |
f49d89ee PH |
2686 | if format.get('resolution') is not None: |
2687 | return format['resolution'] | |
35615307 DA |
2688 | if format.get('width') and format.get('height'): |
2689 | res = '%dx%d' % (format['width'], format['height']) | |
2690 | elif format.get('height'): | |
2691 | res = '%sp' % format['height'] | |
2692 | elif format.get('width'): | |
388ae76b | 2693 | res = '%dx?' % format['width'] |
8c51aa65 | 2694 | else: |
8abeeb94 | 2695 | res = default |
8c51aa65 JMF |
2696 | return res |
2697 | ||
c57f7757 PH |
2698 | def _format_note(self, fdict): |
2699 | res = '' | |
2700 | if fdict.get('ext') in ['f4f', 'f4m']: | |
2701 | res += '(unsupported) ' | |
32f90364 PH |
2702 | if fdict.get('language'): |
2703 | if res: | |
2704 | res += ' ' | |
9016d76f | 2705 | res += '[%s] ' % fdict['language'] |
c57f7757 PH |
2706 | if fdict.get('format_note') is not None: |
2707 | res += fdict['format_note'] + ' ' | |
2708 | if fdict.get('tbr') is not None: | |
2709 | res += '%4dk ' % fdict['tbr'] | |
2710 | if fdict.get('container') is not None: | |
2711 | if res: | |
2712 | res += ', ' | |
2713 | res += '%s container' % fdict['container'] | |
3089bc74 S |
2714 | if (fdict.get('vcodec') is not None |
2715 | and fdict.get('vcodec') != 'none'): | |
c57f7757 PH |
2716 | if res: |
2717 | res += ', ' | |
2718 | res += fdict['vcodec'] | |
91c7271a | 2719 | if fdict.get('vbr') is not None: |
c57f7757 PH |
2720 | res += '@' |
2721 | elif fdict.get('vbr') is not None and fdict.get('abr') is not None: | |
2722 | res += 'video@' | |
2723 | if fdict.get('vbr') is not None: | |
2724 | res += '%4dk' % fdict['vbr'] | |
fbb21cf5 | 2725 | if fdict.get('fps') is not None: |
5d583bdf S |
2726 | if res: |
2727 | res += ', ' | |
2728 | res += '%sfps' % fdict['fps'] | |
c57f7757 PH |
2729 | if fdict.get('acodec') is not None: |
2730 | if res: | |
2731 | res += ', ' | |
2732 | if fdict['acodec'] == 'none': | |
2733 | res += 'video only' | |
2734 | else: | |
2735 | res += '%-5s' % fdict['acodec'] | |
2736 | elif fdict.get('abr') is not None: | |
2737 | if res: | |
2738 | res += ', ' | |
2739 | res += 'audio' | |
2740 | if fdict.get('abr') is not None: | |
2741 | res += '@%3dk' % fdict['abr'] | |
2742 | if fdict.get('asr') is not None: | |
2743 | res += ' (%5dHz)' % fdict['asr'] | |
2744 | if fdict.get('filesize') is not None: | |
2745 | if res: | |
2746 | res += ', ' | |
2747 | res += format_bytes(fdict['filesize']) | |
9732d77e PH |
2748 | elif fdict.get('filesize_approx') is not None: |
2749 | if res: | |
2750 | res += ', ' | |
2751 | res += '~' + format_bytes(fdict['filesize_approx']) | |
c57f7757 | 2752 | return res |
91c7271a | 2753 | |
76d321f6 | 2754 | def _format_note_table(self, f): |
2755 | def join_fields(*vargs): | |
2756 | return ', '.join((val for val in vargs if val != '')) | |
2757 | ||
2758 | return join_fields( | |
2759 | 'UNSUPPORTED' if f.get('ext') in ('f4f', 'f4m') else '', | |
2760 | format_field(f, 'language', '[%s]'), | |
2761 | format_field(f, 'format_note'), | |
2762 | format_field(f, 'container', ignore=(None, f.get('ext'))), | |
2763 | format_field(f, 'asr', '%5dHz')) | |
2764 | ||
c57f7757 | 2765 | def list_formats(self, info_dict): |
94badb25 | 2766 | formats = info_dict.get('formats', [info_dict]) |
76d321f6 | 2767 | new_format = self.params.get('listformats_table', False) |
2768 | if new_format: | |
2769 | table = [ | |
2770 | [ | |
2771 | format_field(f, 'format_id'), | |
2772 | format_field(f, 'ext'), | |
2773 | self.format_resolution(f), | |
2774 | format_field(f, 'fps', '%d'), | |
2775 | '|', | |
2776 | format_field(f, 'filesize', ' %s', func=format_bytes) + format_field(f, 'filesize_approx', '~%s', func=format_bytes), | |
2777 | format_field(f, 'tbr', '%4dk'), | |
52a8a1e1 | 2778 | shorten_protocol_name(f.get('protocol', '').replace("native", "n")), |
76d321f6 | 2779 | '|', |
2780 | format_field(f, 'vcodec', default='unknown').replace('none', ''), | |
2781 | format_field(f, 'vbr', '%4dk'), | |
2782 | format_field(f, 'acodec', default='unknown').replace('none', ''), | |
2783 | format_field(f, 'abr', '%3dk'), | |
2784 | format_field(f, 'asr', '%5dHz'), | |
2785 | self._format_note_table(f)] | |
2786 | for f in formats | |
2787 | if f.get('preference') is None or f['preference'] >= -1000] | |
2788 | header_line = ['ID', 'EXT', 'RESOLUTION', 'FPS', '|', ' FILESIZE', ' TBR', 'PROTO', | |
2789 | '|', 'VCODEC', ' VBR', 'ACODEC', ' ABR', ' ASR', 'NOTE'] | |
2790 | else: | |
2791 | table = [ | |
2792 | [ | |
2793 | format_field(f, 'format_id'), | |
2794 | format_field(f, 'ext'), | |
2795 | self.format_resolution(f), | |
2796 | self._format_note(f)] | |
2797 | for f in formats | |
2798 | if f.get('preference') is None or f['preference'] >= -1000] | |
2799 | header_line = ['format code', 'extension', 'resolution', 'note'] | |
57dd9a8f | 2800 | |
cfb56d1a | 2801 | self.to_screen( |
76d321f6 | 2802 | '[info] Available formats for %s:\n%s' % (info_dict['id'], render_table( |
2803 | header_line, | |
2804 | table, | |
2805 | delim=new_format, | |
2806 | extraGap=(0 if new_format else 1), | |
2807 | hideEmpty=new_format))) | |
cfb56d1a PH |
2808 | |
2809 | def list_thumbnails(self, info_dict): | |
2810 | thumbnails = info_dict.get('thumbnails') | |
2811 | if not thumbnails: | |
b7b72db9 | 2812 | self.to_screen('[info] No thumbnails present for %s' % info_dict['id']) |
2813 | return | |
cfb56d1a PH |
2814 | |
2815 | self.to_screen( | |
2816 | '[info] Thumbnails for %s:' % info_dict['id']) | |
2817 | self.to_screen(render_table( | |
2818 | ['ID', 'width', 'height', 'URL'], | |
2819 | [[t['id'], t.get('width', 'unknown'), t.get('height', 'unknown'), t['url']] for t in thumbnails])) | |
dca08720 | 2820 | |
360e1ca5 | 2821 | def list_subtitles(self, video_id, subtitles, name='subtitles'): |
a504ced0 | 2822 | if not subtitles: |
360e1ca5 | 2823 | self.to_screen('%s has no %s' % (video_id, name)) |
a504ced0 | 2824 | return |
a504ced0 | 2825 | self.to_screen( |
edab9dbf JMF |
2826 | 'Available %s for %s:' % (name, video_id)) |
2827 | self.to_screen(render_table( | |
2828 | ['Language', 'formats'], | |
2829 | [[lang, ', '.join(f['ext'] for f in reversed(formats))] | |
2830 | for lang, formats in subtitles.items()])) | |
a504ced0 | 2831 | |
dca08720 PH |
2832 | def urlopen(self, req): |
2833 | """ Start an HTTP download """ | |
82d8a8b6 | 2834 | if isinstance(req, compat_basestring): |
67dda517 | 2835 | req = sanitized_Request(req) |
19a41fc6 | 2836 | return self._opener.open(req, timeout=self._socket_timeout) |
dca08720 PH |
2837 | |
2838 | def print_debug_header(self): | |
2839 | if not self.params.get('verbose'): | |
2840 | return | |
62fec3b2 | 2841 | |
4192b51c | 2842 | if type('') is not compat_str: |
067aa17e | 2843 | # Python 2.6 on SLES11 SP1 (https://github.com/ytdl-org/youtube-dl/issues/3326) |
4192b51c PH |
2844 | self.report_warning( |
2845 | 'Your Python is broken! Update to a newer and supported version') | |
2846 | ||
c6afed48 PH |
2847 | stdout_encoding = getattr( |
2848 | sys.stdout, 'encoding', 'missing (%s)' % type(sys.stdout).__name__) | |
b0472057 | 2849 | encoding_str = ( |
734f90bb PH |
2850 | '[debug] Encodings: locale %s, fs %s, out %s, pref %s\n' % ( |
2851 | locale.getpreferredencoding(), | |
2852 | sys.getfilesystemencoding(), | |
c6afed48 | 2853 | stdout_encoding, |
b0472057 | 2854 | self.get_encoding())) |
4192b51c | 2855 | write_string(encoding_str, encoding=None) |
734f90bb | 2856 | |
e5813e53 | 2857 | source = ( |
2858 | '(exe)' if hasattr(sys, 'frozen') | |
2859 | else '(zip)' if isinstance(globals().get('__loader__'), zipimporter) | |
2860 | else '(source)' if os.path.basename(sys.argv[0]) == '__main__.py' | |
2861 | else '') | |
2862 | self._write_string('[debug] yt-dlp version %s %s\n' % (__version__, source)) | |
e0986e31 | 2863 | if _LAZY_LOADER: |
f74980cb | 2864 | self._write_string('[debug] Lazy loading extractors enabled\n') |
2865 | if _PLUGIN_CLASSES: | |
2866 | self._write_string( | |
2867 | '[debug] Plugin Extractors: %s\n' % [ie.ie_key() for ie in _PLUGIN_CLASSES]) | |
dca08720 PH |
2868 | try: |
2869 | sp = subprocess.Popen( | |
2870 | ['git', 'rev-parse', '--short', 'HEAD'], | |
2871 | stdout=subprocess.PIPE, stderr=subprocess.PIPE, | |
2872 | cwd=os.path.dirname(os.path.abspath(__file__))) | |
f5b1bca9 | 2873 | out, err = process_communicate_or_kill(sp) |
dca08720 PH |
2874 | out = out.decode().strip() |
2875 | if re.match('[0-9a-f]+', out): | |
f74980cb | 2876 | self._write_string('[debug] Git HEAD: %s\n' % out) |
70a1165b | 2877 | except Exception: |
dca08720 PH |
2878 | try: |
2879 | sys.exc_clear() | |
70a1165b | 2880 | except Exception: |
dca08720 | 2881 | pass |
b300cda4 S |
2882 | |
2883 | def python_implementation(): | |
2884 | impl_name = platform.python_implementation() | |
2885 | if impl_name == 'PyPy' and hasattr(sys, 'pypy_version_info'): | |
2886 | return impl_name + ' version %d.%d.%d' % sys.pypy_version_info[:3] | |
2887 | return impl_name | |
2888 | ||
e5813e53 | 2889 | self._write_string('[debug] Python version %s (%s %s) - %s\n' % ( |
2890 | platform.python_version(), | |
2891 | python_implementation(), | |
2892 | platform.architecture()[0], | |
b300cda4 | 2893 | platform_name())) |
d28b5171 | 2894 | |
73fac4e9 | 2895 | exe_versions = FFmpegPostProcessor.get_versions(self) |
4c83c967 | 2896 | exe_versions['rtmpdump'] = rtmpdump_version() |
feee8d32 | 2897 | exe_versions['phantomjs'] = PhantomJSwrapper._version() |
d28b5171 PH |
2898 | exe_str = ', '.join( |
2899 | '%s %s' % (exe, v) | |
2900 | for exe, v in sorted(exe_versions.items()) | |
2901 | if v | |
2902 | ) | |
2903 | if not exe_str: | |
2904 | exe_str = 'none' | |
2905 | self._write_string('[debug] exe versions: %s\n' % exe_str) | |
dca08720 PH |
2906 | |
2907 | proxy_map = {} | |
2908 | for handler in self._opener.handlers: | |
2909 | if hasattr(handler, 'proxies'): | |
2910 | proxy_map.update(handler.proxies) | |
734f90bb | 2911 | self._write_string('[debug] Proxy map: ' + compat_str(proxy_map) + '\n') |
dca08720 | 2912 | |
58b1f00d PH |
2913 | if self.params.get('call_home', False): |
2914 | ipaddr = self.urlopen('https://yt-dl.org/ip').read().decode('utf-8') | |
2915 | self._write_string('[debug] Public IP address: %s\n' % ipaddr) | |
f5546c0b | 2916 | return |
58b1f00d PH |
2917 | latest_version = self.urlopen( |
2918 | 'https://yt-dl.org/latest/version').read().decode('utf-8') | |
2919 | if version_tuple(latest_version) > version_tuple(__version__): | |
2920 | self.report_warning( | |
2921 | 'You are using an outdated version (newest version: %s)! ' | |
2922 | 'See https://yt-dl.org/update if you need help updating.' % | |
2923 | latest_version) | |
2924 | ||
e344693b | 2925 | def _setup_opener(self): |
6ad14cab | 2926 | timeout_val = self.params.get('socket_timeout') |
19a41fc6 | 2927 | self._socket_timeout = 600 if timeout_val is None else float(timeout_val) |
6ad14cab | 2928 | |
dca08720 PH |
2929 | opts_cookiefile = self.params.get('cookiefile') |
2930 | opts_proxy = self.params.get('proxy') | |
2931 | ||
2932 | if opts_cookiefile is None: | |
2933 | self.cookiejar = compat_cookiejar.CookieJar() | |
2934 | else: | |
590bc6f6 | 2935 | opts_cookiefile = expand_path(opts_cookiefile) |
1bab3437 | 2936 | self.cookiejar = YoutubeDLCookieJar(opts_cookiefile) |
dca08720 | 2937 | if os.access(opts_cookiefile, os.R_OK): |
1d88b3e6 | 2938 | self.cookiejar.load(ignore_discard=True, ignore_expires=True) |
dca08720 | 2939 | |
6a3f4c3f | 2940 | cookie_processor = YoutubeDLCookieProcessor(self.cookiejar) |
dca08720 PH |
2941 | if opts_proxy is not None: |
2942 | if opts_proxy == '': | |
2943 | proxies = {} | |
2944 | else: | |
2945 | proxies = {'http': opts_proxy, 'https': opts_proxy} | |
2946 | else: | |
2947 | proxies = compat_urllib_request.getproxies() | |
067aa17e | 2948 | # Set HTTPS proxy to HTTP one if given (https://github.com/ytdl-org/youtube-dl/issues/805) |
dca08720 PH |
2949 | if 'http' in proxies and 'https' not in proxies: |
2950 | proxies['https'] = proxies['http'] | |
91410c9b | 2951 | proxy_handler = PerRequestProxyHandler(proxies) |
a0ddb8a2 PH |
2952 | |
2953 | debuglevel = 1 if self.params.get('debug_printtraffic') else 0 | |
be4a824d PH |
2954 | https_handler = make_HTTPS_handler(self.params, debuglevel=debuglevel) |
2955 | ydlh = YoutubeDLHandler(self.params, debuglevel=debuglevel) | |
fca6dba8 | 2956 | redirect_handler = YoutubeDLRedirectHandler() |
8b172c2e | 2957 | data_handler = compat_urllib_request_DataHandler() |
6240b0a2 JMF |
2958 | |
2959 | # When passing our own FileHandler instance, build_opener won't add the | |
2960 | # default FileHandler and allows us to disable the file protocol, which | |
2961 | # can be used for malicious purposes (see | |
067aa17e | 2962 | # https://github.com/ytdl-org/youtube-dl/issues/8227) |
6240b0a2 JMF |
2963 | file_handler = compat_urllib_request.FileHandler() |
2964 | ||
2965 | def file_open(*args, **kwargs): | |
7a5c1cfe | 2966 | raise compat_urllib_error.URLError('file:// scheme is explicitly disabled in yt-dlp for security reasons') |
6240b0a2 JMF |
2967 | file_handler.file_open = file_open |
2968 | ||
2969 | opener = compat_urllib_request.build_opener( | |
fca6dba8 | 2970 | proxy_handler, https_handler, cookie_processor, ydlh, redirect_handler, data_handler, file_handler) |
2461f79d | 2971 | |
dca08720 PH |
2972 | # Delete the default user-agent header, which would otherwise apply in |
2973 | # cases where our custom HTTP handler doesn't come into play | |
067aa17e | 2974 | # (See https://github.com/ytdl-org/youtube-dl/issues/1309 for details) |
dca08720 PH |
2975 | opener.addheaders = [] |
2976 | self._opener = opener | |
62fec3b2 PH |
2977 | |
2978 | def encode(self, s): | |
2979 | if isinstance(s, bytes): | |
2980 | return s # Already encoded | |
2981 | ||
2982 | try: | |
2983 | return s.encode(self.get_encoding()) | |
2984 | except UnicodeEncodeError as err: | |
2985 | err.reason = err.reason + '. Check your system encoding configuration or use the --encoding option.' | |
2986 | raise | |
2987 | ||
2988 | def get_encoding(self): | |
2989 | encoding = self.params.get('encoding') | |
2990 | if encoding is None: | |
2991 | encoding = preferredencoding() | |
2992 | return encoding | |
ec82d85a | 2993 | |
de6000d9 | 2994 | def _write_thumbnails(self, info_dict, filename): # return the extensions |
6c4fd172 | 2995 | write_all = self.params.get('write_all_thumbnails', False) |
2996 | thumbnails = [] | |
2997 | if write_all or self.params.get('writethumbnail', False): | |
0202b52a | 2998 | thumbnails = info_dict.get('thumbnails') or [] |
6c4fd172 | 2999 | multiple = write_all and len(thumbnails) > 1 |
ec82d85a | 3000 | |
0202b52a | 3001 | ret = [] |
6c4fd172 | 3002 | for t in thumbnails[::1 if write_all else -1]: |
ec82d85a | 3003 | thumb_ext = determine_ext(t['url'], 'jpg') |
6c4fd172 | 3004 | suffix = '%s.' % t['id'] if multiple else '' |
3005 | thumb_display_id = '%s ' % t['id'] if multiple else '' | |
dcf64d43 | 3006 | t['filepath'] = thumb_filename = replace_extension(filename, suffix + thumb_ext, info_dict.get('ext')) |
ec82d85a | 3007 | |
0c3d0f51 | 3008 | if not self.params.get('overwrites', True) and os.path.exists(encodeFilename(thumb_filename)): |
de6000d9 | 3009 | ret.append(suffix + thumb_ext) |
ec82d85a PH |
3010 | self.to_screen('[%s] %s: Thumbnail %sis already present' % |
3011 | (info_dict['extractor'], info_dict['id'], thumb_display_id)) | |
3012 | else: | |
5ef7d9bd | 3013 | self.to_screen('[%s] %s: Downloading thumbnail %s ...' % |
ec82d85a PH |
3014 | (info_dict['extractor'], info_dict['id'], thumb_display_id)) |
3015 | try: | |
3016 | uf = self.urlopen(t['url']) | |
d3d89c32 | 3017 | with open(encodeFilename(thumb_filename), 'wb') as thumbf: |
ec82d85a | 3018 | shutil.copyfileobj(uf, thumbf) |
de6000d9 | 3019 | ret.append(suffix + thumb_ext) |
ec82d85a PH |
3020 | self.to_screen('[%s] %s: Writing thumbnail %sto: %s' % |
3021 | (info_dict['extractor'], info_dict['id'], thumb_display_id, thumb_filename)) | |
3022 | except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: | |
3023 | self.report_warning('Unable to download thumbnail "%s": %s' % | |
9b9c5355 | 3024 | (t['url'], error_to_compat_str(err))) |
6c4fd172 | 3025 | if ret and not write_all: |
3026 | break | |
0202b52a | 3027 | return ret |