]>
Commit | Line | Data |
---|---|---|
8222d8de | 1 | #!/usr/bin/env python |
dcdb292f | 2 | # coding: utf-8 |
8222d8de | 3 | |
6febd1c1 | 4 | from __future__ import absolute_import, unicode_literals |
8222d8de | 5 | |
26e63931 | 6 | import collections |
31bd3925 | 7 | import contextlib |
317f7ab6 | 8 | import copy |
9d2ecdbc | 9 | import datetime |
c1c9a79c | 10 | import errno |
31bd3925 | 11 | import fileinput |
8222d8de | 12 | import io |
b82f815f | 13 | import itertools |
8694c600 | 14 | import json |
62fec3b2 | 15 | import locale |
083c9df9 | 16 | import operator |
8222d8de | 17 | import os |
dca08720 | 18 | import platform |
8222d8de JMF |
19 | import re |
20 | import shutil | |
dca08720 | 21 | import subprocess |
8222d8de JMF |
22 | import socket |
23 | import sys | |
24 | import time | |
67134eab | 25 | import tokenize |
8222d8de | 26 | import traceback |
75822ca7 | 27 | import random |
8222d8de | 28 | |
961ea474 S |
29 | from string import ascii_letters |
30 | ||
8c25f81b | 31 | from .compat import ( |
82d8a8b6 | 32 | compat_basestring, |
dca08720 | 33 | compat_cookiejar, |
003c69a8 | 34 | compat_get_terminal_size, |
ce02ed60 | 35 | compat_http_client, |
4f026faf | 36 | compat_kwargs, |
d0d9ade4 | 37 | compat_numeric_types, |
e9c0cdd3 | 38 | compat_os_name, |
ce02ed60 | 39 | compat_str, |
67134eab | 40 | compat_tokenize_tokenize, |
ce02ed60 PH |
41 | compat_urllib_error, |
42 | compat_urllib_request, | |
8b172c2e | 43 | compat_urllib_request_DataHandler, |
8c25f81b PH |
44 | ) |
45 | from .utils import ( | |
eedb7ba5 S |
46 | age_restricted, |
47 | args_to_str, | |
ce02ed60 PH |
48 | ContentTooShortError, |
49 | date_from_str, | |
50 | DateRange, | |
acd69589 | 51 | DEFAULT_OUTTMPL, |
ce02ed60 | 52 | determine_ext, |
b5559424 | 53 | determine_protocol, |
ce02ed60 | 54 | DownloadError, |
c0384f22 | 55 | encode_compat_str, |
ce02ed60 | 56 | encodeFilename, |
9b9c5355 | 57 | error_to_compat_str, |
590bc6f6 | 58 | expand_path, |
ce02ed60 | 59 | ExtractorError, |
02dbf93f | 60 | format_bytes, |
525ef922 | 61 | formatSeconds, |
773f291d | 62 | GeoRestrictedError, |
c9969434 | 63 | int_or_none, |
773f291d | 64 | ISO3166Utils, |
ce02ed60 | 65 | locked_file, |
dca08720 | 66 | make_HTTPS_handler, |
ce02ed60 | 67 | MaxDownloadsReached, |
cd6fc19e | 68 | orderedSet, |
b7ab0590 | 69 | PagedList, |
083c9df9 | 70 | parse_filesize, |
91410c9b | 71 | PerRequestProxyHandler, |
dca08720 | 72 | platform_name, |
eedb7ba5 | 73 | PostProcessingError, |
ce02ed60 | 74 | preferredencoding, |
eedb7ba5 | 75 | prepend_extension, |
51fb4995 | 76 | register_socks_protocols, |
cfb56d1a | 77 | render_table, |
eedb7ba5 | 78 | replace_extension, |
ce02ed60 PH |
79 | SameFileError, |
80 | sanitize_filename, | |
1bb5c511 | 81 | sanitize_path, |
dcf77cf1 | 82 | sanitize_url, |
67dda517 | 83 | sanitized_Request, |
e5660ee6 | 84 | std_headers, |
1211bb6d | 85 | str_or_none, |
ce02ed60 | 86 | subtitles_filename, |
ce02ed60 | 87 | UnavailableVideoError, |
29eb5174 | 88 | url_basename, |
58b1f00d | 89 | version_tuple, |
ce02ed60 PH |
90 | write_json_file, |
91 | write_string, | |
1bab3437 | 92 | YoutubeDLCookieJar, |
6a3f4c3f | 93 | YoutubeDLCookieProcessor, |
dca08720 | 94 | YoutubeDLHandler, |
fca6dba8 | 95 | YoutubeDLRedirectHandler, |
ce02ed60 | 96 | ) |
a0e07d31 | 97 | from .cache import Cache |
e0986e31 | 98 | from .extractor import get_info_extractor, gen_extractor_classes, _LAZY_LOADER |
4c54b89e | 99 | from .extractor.openload import PhantomJSwrapper |
3bc2ddcc | 100 | from .downloader import get_suitable_downloader |
4c83c967 | 101 | from .downloader.rtmp import rtmpdump_version |
4f026faf | 102 | from .postprocessor import ( |
f17f8651 | 103 | FFmpegFixupM3u8PP, |
62cd676c | 104 | FFmpegFixupM4aPP, |
6271f1ca | 105 | FFmpegFixupStretchedPP, |
4f026faf PH |
106 | FFmpegMergerPP, |
107 | FFmpegPostProcessor, | |
57df9f53 | 108 | FFmpegSubtitlesConvertorPP, |
4f026faf PH |
109 | get_postprocessor, |
110 | ) | |
dca08720 | 111 | from .version import __version__ |
8222d8de | 112 | |
e9c0cdd3 YCH |
113 | if compat_os_name == 'nt': |
114 | import ctypes | |
115 | ||
2459b6e1 | 116 | |
8222d8de JMF |
117 | class YoutubeDL(object): |
118 | """YoutubeDL class. | |
119 | ||
120 | YoutubeDL objects are the ones responsible of downloading the | |
121 | actual video file and writing it to disk if the user has requested | |
122 | it, among some other tasks. In most cases there should be one per | |
123 | program. As, given a video URL, the downloader doesn't know how to | |
124 | extract all the needed information, task that InfoExtractors do, it | |
125 | has to pass the URL to one of them. | |
126 | ||
127 | For this, YoutubeDL objects have a method that allows | |
128 | InfoExtractors to be registered in a given order. When it is passed | |
129 | a URL, the YoutubeDL object handles it to the first InfoExtractor it | |
130 | finds that reports being able to handle it. The InfoExtractor extracts | |
131 | all the information about the video or videos the URL refers to, and | |
132 | YoutubeDL process the extracted information, possibly using a File | |
133 | Downloader to download the video. | |
134 | ||
135 | YoutubeDL objects accept a lot of parameters. In order not to saturate | |
136 | the object constructor with arguments, it receives a dictionary of | |
137 | options instead. These options are available through the params | |
138 | attribute for the InfoExtractors to use. The YoutubeDL also | |
139 | registers itself as the downloader in charge for the InfoExtractors | |
140 | that are added to it, so this is a "mutual registration". | |
141 | ||
142 | Available options: | |
143 | ||
144 | username: Username for authentication purposes. | |
145 | password: Password for authentication purposes. | |
180940e0 | 146 | videopassword: Password for accessing a video. |
1da50aa3 S |
147 | ap_mso: Adobe Pass multiple-system operator identifier. |
148 | ap_username: Multiple-system operator account username. | |
149 | ap_password: Multiple-system operator account password. | |
8222d8de JMF |
150 | usenetrc: Use netrc for authentication instead. |
151 | verbose: Print additional info to stdout. | |
152 | quiet: Do not print messages to stdout. | |
ad8915b7 | 153 | no_warnings: Do not print out anything for warnings. |
8222d8de JMF |
154 | forceurl: Force printing final URL. |
155 | forcetitle: Force printing title. | |
156 | forceid: Force printing ID. | |
157 | forcethumbnail: Force printing thumbnail URL. | |
158 | forcedescription: Force printing description. | |
159 | forcefilename: Force printing final filename. | |
525ef922 | 160 | forceduration: Force printing duration. |
8694c600 | 161 | forcejson: Force printing info_dict as JSON. |
63e0be34 PH |
162 | dump_single_json: Force printing the info_dict of the whole playlist |
163 | (or video) as a single JSON line. | |
8222d8de | 164 | simulate: Do not download the video files. |
eb8a4433 | 165 | format: Video format code. see "FORMAT SELECTION" for more details. |
166 | format_sort: How to sort the video formats. see "Sorting Formats" for more details. | |
167 | format_sort_force: Force the given format_sort. see "Sorting Formats" for more details. | |
909d24dd | 168 | allow_multiple_video_streams: Allow multiple video streams to be merged into a single file |
169 | allow_multiple_audio_streams: Allow multiple audio streams to be merged into a single file | |
8222d8de | 170 | outtmpl: Template for output names. |
bdc3fd2f U |
171 | restrictfilenames: Do not allow "&" and spaces in file names. |
172 | trim_file_name: Limit length of filename (extension excluded). | |
8222d8de | 173 | ignoreerrors: Do not stop on download errors. |
d22dec74 | 174 | force_generic_extractor: Force downloader to use the generic extractor |
8222d8de JMF |
175 | nooverwrites: Prevent overwriting files. |
176 | playliststart: Playlist item to start at. | |
177 | playlistend: Playlist item to end at. | |
c14e88f0 | 178 | playlist_items: Specific indices of playlist to download. |
ff815fe6 | 179 | playlistreverse: Download playlist items in reverse order. |
75822ca7 | 180 | playlistrandom: Download playlist items in random order. |
8222d8de JMF |
181 | matchtitle: Download only matching titles. |
182 | rejecttitle: Reject downloads for matching titles. | |
8bf9319e | 183 | logger: Log messages to a logging.Logger instance. |
8222d8de JMF |
184 | logtostderr: Log messages to stderr instead of stdout. |
185 | writedescription: Write the video description to a .description file | |
186 | writeinfojson: Write the video description to a .info.json file | |
1fb07d10 | 187 | writeannotations: Write the video annotations to a .annotations.xml file |
8222d8de | 188 | writethumbnail: Write the thumbnail image to a file |
ec82d85a | 189 | write_all_thumbnails: Write all thumbnail formats to files |
8222d8de | 190 | writesubtitles: Write the video subtitles to a file |
741dd8ea | 191 | writeautomaticsub: Write the automatically generated subtitles to a file |
8222d8de | 192 | allsubtitles: Downloads all the subtitles of the video |
0b7f3118 | 193 | (requires writesubtitles or writeautomaticsub) |
8222d8de | 194 | listsubtitles: Lists all available subtitles for the video |
a504ced0 | 195 | subtitlesformat: The format code for subtitles |
aa6a10c4 | 196 | subtitleslangs: List of languages of the subtitles to download |
8222d8de JMF |
197 | keepvideo: Keep the video file after post-processing |
198 | daterange: A DateRange object, download only if the upload_date is in the range. | |
199 | skip_download: Skip the actual download of the video file | |
c35f9e72 | 200 | cachedir: Location of the cache files in the filesystem. |
a0e07d31 | 201 | False to disable filesystem cache. |
47192f92 | 202 | noplaylist: Download single video instead of a playlist if in doubt. |
8dbe9899 PH |
203 | age_limit: An integer representing the user's age in years. |
204 | Unsuitable videos for the given age are skipped. | |
5fe18bdb PH |
205 | min_views: An integer representing the minimum view count the video |
206 | must have in order to not be skipped. | |
207 | Videos without view count information are always | |
208 | downloaded. None for no limit. | |
209 | max_views: An integer representing the maximum view count. | |
210 | Videos that are more popular than that are not | |
211 | downloaded. | |
212 | Videos without view count information are always | |
213 | downloaded. None for no limit. | |
214 | download_archive: File name of a file where all downloads are recorded. | |
c1c9a79c PH |
215 | Videos already present in the file are not downloaded |
216 | again. | |
ea6e0c2b | 217 | break_on_existing: Stop the download process after attempting to download a file that's |
218 | in the archive. | |
dca08720 | 219 | cookiefile: File name where cookies should be read from and dumped to. |
a1ee09e8 | 220 | nocheckcertificate:Do not verify SSL certificates |
7e8c0af0 PH |
221 | prefer_insecure: Use HTTP instead of HTTPS to retrieve information. |
222 | At the moment, this is only supported by YouTube. | |
a1ee09e8 | 223 | proxy: URL of the proxy server to use |
38cce791 | 224 | geo_verification_proxy: URL of the proxy to use for IP address verification |
504f20dd | 225 | on geo-restricted sites. |
e344693b | 226 | socket_timeout: Time to wait for unresponsive hosts, in seconds |
0783b09b PH |
227 | bidi_workaround: Work around buggy terminals without bidirectional text |
228 | support, using fridibi | |
a0ddb8a2 | 229 | debug_printtraffic:Print out sent and received HTTP traffic |
7b0817e8 | 230 | include_ads: Download ads as well |
04b4d394 PH |
231 | default_search: Prepend this string if an input url is not valid. |
232 | 'auto' for elaborate guessing | |
62fec3b2 | 233 | encoding: Use this encoding instead of the system-specified. |
e8ee972c | 234 | extract_flat: Do not resolve URLs, return the immediate result. |
057a5206 PH |
235 | Pass in 'in_playlist' to only show this behavior for |
236 | playlist items. | |
4f026faf | 237 | postprocessors: A list of dictionaries, each with an entry |
71b640cc | 238 | * key: The name of the postprocessor. See |
cefecac1 | 239 | youtube_dlc/postprocessor/__init__.py for a list. |
4f026faf PH |
240 | as well as any further keyword arguments for the |
241 | postprocessor. | |
71b640cc PH |
242 | progress_hooks: A list of functions that get called on download |
243 | progress, with a dictionary with the entries | |
5cda4eda | 244 | * status: One of "downloading", "error", or "finished". |
ee69b99a | 245 | Check this first and ignore unknown values. |
71b640cc | 246 | |
5cda4eda | 247 | If status is one of "downloading", or "finished", the |
ee69b99a PH |
248 | following properties may also be present: |
249 | * filename: The final filename (always present) | |
5cda4eda | 250 | * tmpfilename: The filename we're currently writing to |
71b640cc PH |
251 | * downloaded_bytes: Bytes on disk |
252 | * total_bytes: Size of the whole file, None if unknown | |
5cda4eda PH |
253 | * total_bytes_estimate: Guess of the eventual file size, |
254 | None if unavailable. | |
255 | * elapsed: The number of seconds since download started. | |
71b640cc PH |
256 | * eta: The estimated time in seconds, None if unknown |
257 | * speed: The download speed in bytes/second, None if | |
258 | unknown | |
5cda4eda PH |
259 | * fragment_index: The counter of the currently |
260 | downloaded video fragment. | |
261 | * fragment_count: The number of fragments (= individual | |
262 | files that will be merged) | |
71b640cc PH |
263 | |
264 | Progress hooks are guaranteed to be called at least once | |
265 | (with status "finished") if the download is successful. | |
45598f15 | 266 | merge_output_format: Extension to use when merging formats. |
6271f1ca PH |
267 | fixup: Automatically correct known faults of the file. |
268 | One of: | |
269 | - "never": do nothing | |
270 | - "warn": only emit a warning | |
271 | - "detect_or_warn": check whether we can do anything | |
62cd676c | 272 | about it, warn otherwise (default) |
504f20dd | 273 | source_address: Client-side IP address to bind to. |
6ec6cb4e | 274 | call_home: Boolean, true iff we are allowed to contact the |
cefecac1 | 275 | youtube-dlc servers for debugging. |
7aa589a5 S |
276 | sleep_interval: Number of seconds to sleep before each download when |
277 | used alone or a lower bound of a range for randomized | |
278 | sleep before each download (minimum possible number | |
279 | of seconds to sleep) when used along with | |
280 | max_sleep_interval. | |
281 | max_sleep_interval:Upper bound of a range for randomized sleep before each | |
282 | download (maximum possible number of seconds to sleep). | |
283 | Must only be used along with sleep_interval. | |
284 | Actual sleep time will be a random float from range | |
285 | [sleep_interval; max_sleep_interval]. | |
cfb56d1a PH |
286 | listformats: Print an overview of available video formats and exit. |
287 | list_thumbnails: Print a table of all thumbnails and exit. | |
347de493 PH |
288 | match_filter: A function that gets called with the info_dict of |
289 | every video. | |
290 | If it returns a message, the video is ignored. | |
291 | If it returns None, the video is downloaded. | |
292 | match_filter_func in utils.py is one example for this. | |
7e5db8c9 | 293 | no_color: Do not emit color codes in output. |
0a840f58 | 294 | geo_bypass: Bypass geographic restriction via faking X-Forwarded-For |
504f20dd | 295 | HTTP header |
0a840f58 | 296 | geo_bypass_country: |
773f291d S |
297 | Two-letter ISO 3166-2 country code that will be used for |
298 | explicit geographic restriction bypassing via faking | |
504f20dd | 299 | X-Forwarded-For HTTP header |
5f95927a S |
300 | geo_bypass_ip_block: |
301 | IP range in CIDR notation that will be used similarly to | |
504f20dd | 302 | geo_bypass_country |
71b640cc | 303 | |
85729c51 PH |
304 | The following options determine which downloader is picked: |
305 | external_downloader: Executable of the external downloader to call. | |
306 | None or unset for standard (built-in) downloader. | |
bf09af3a S |
307 | hls_prefer_native: Use the native HLS downloader instead of ffmpeg/avconv |
308 | if True, otherwise use ffmpeg/avconv if False, otherwise | |
309 | use downloader suggested by extractor if None. | |
fe7e0c98 | 310 | |
8222d8de | 311 | The following parameters are not used by YoutubeDL itself, they are used by |
cefecac1 | 312 | the downloader (see youtube_dlc/downloader/common.py): |
8222d8de | 313 | nopart, updatetime, buffersize, ratelimit, min_filesize, max_filesize, test, |
881e6a1f | 314 | noresizebuffer, retries, continuedl, noprogress, consoletitle, |
b54d4a5c S |
315 | xattr_set_filesize, external_downloader_args, hls_use_mpegts, |
316 | http_chunk_size. | |
76b1bd67 JMF |
317 | |
318 | The following options are used by the post processors: | |
d4a24f40 S |
319 | prefer_ffmpeg: If False, use avconv instead of ffmpeg if both are available, |
320 | otherwise prefer ffmpeg. | |
c0b7d117 S |
321 | ffmpeg_location: Location of the ffmpeg/avconv binary; either the path |
322 | to the binary or its containing directory. | |
f72b0a60 S |
323 | postprocessor_args: A list of additional command-line arguments for the |
324 | postprocessor. | |
3836b02c | 325 | |
3600fd59 S |
326 | The following options are used by the Youtube extractor: |
327 | youtube_include_dash_manifest: If True (default), DASH manifests and related | |
328 | data will be downloaded and processed by extractor. | |
329 | You can reduce network I/O by disabling it if you don't | |
330 | care about DASH. | |
8222d8de JMF |
331 | """ |
332 | ||
c9969434 S |
333 | _NUMERIC_FIELDS = set(( |
334 | 'width', 'height', 'tbr', 'abr', 'asr', 'vbr', 'fps', 'filesize', 'filesize_approx', | |
335 | 'timestamp', 'upload_year', 'upload_month', 'upload_day', | |
336 | 'duration', 'view_count', 'like_count', 'dislike_count', 'repost_count', | |
337 | 'average_rating', 'comment_count', 'age_limit', | |
338 | 'start_time', 'end_time', | |
339 | 'chapter_number', 'season_number', 'episode_number', | |
340 | 'track_number', 'disc_number', 'release_year', | |
341 | 'playlist_index', | |
342 | )) | |
343 | ||
8222d8de JMF |
344 | params = None |
345 | _ies = [] | |
346 | _pps = [] | |
347 | _download_retcode = None | |
348 | _num_downloads = None | |
349 | _screen_file = None | |
350 | ||
3511266b | 351 | def __init__(self, params=None, auto_init=True): |
8222d8de | 352 | """Create a FileDownloader object with the given options.""" |
e9f9a10f JMF |
353 | if params is None: |
354 | params = {} | |
8222d8de | 355 | self._ies = [] |
56c73665 | 356 | self._ies_instances = {} |
8222d8de | 357 | self._pps = [] |
933605d7 | 358 | self._progress_hooks = [] |
8222d8de JMF |
359 | self._download_retcode = 0 |
360 | self._num_downloads = 0 | |
361 | self._screen_file = [sys.stdout, sys.stderr][params.get('logtostderr', False)] | |
0783b09b | 362 | self._err_file = sys.stderr |
4abf617b S |
363 | self.params = { |
364 | # Default parameters | |
365 | 'nocheckcertificate': False, | |
366 | } | |
367 | self.params.update(params) | |
a0e07d31 | 368 | self.cache = Cache(self) |
a45e8619 | 369 | self.archive = set() |
ecdec191 JB |
370 | |
371 | """Preload the archive, if any is specified""" | |
372 | def preload_download_archive(self): | |
373 | fn = self.params.get('download_archive') | |
374 | if fn is None: | |
375 | return False | |
376 | try: | |
377 | with locked_file(fn, 'r', encoding='utf-8') as archive_file: | |
378 | for line in archive_file: | |
a45e8619 | 379 | self.archive.add(line.strip()) |
ecdec191 JB |
380 | except IOError as ioe: |
381 | if ioe.errno != errno.ENOENT: | |
382 | raise | |
1d74d8d9 | 383 | return False |
ecdec191 | 384 | return True |
34308b30 | 385 | |
be5df5ee S |
386 | def check_deprecated(param, option, suggestion): |
387 | if self.params.get(param) is not None: | |
388 | self.report_warning( | |
389 | '%s is deprecated. Use %s instead.' % (option, suggestion)) | |
390 | return True | |
391 | return False | |
392 | ||
1de7ea76 JB |
393 | if self.params.get('verbose'): |
394 | self.to_stdout('[debug] Loading archive file %r' % self.params.get('download_archive')) | |
395 | ||
ecdec191 JB |
396 | preload_download_archive(self) |
397 | ||
be5df5ee | 398 | if check_deprecated('cn_verification_proxy', '--cn-verification-proxy', '--geo-verification-proxy'): |
38cce791 YCH |
399 | if self.params.get('geo_verification_proxy') is None: |
400 | self.params['geo_verification_proxy'] = self.params['cn_verification_proxy'] | |
401 | ||
be5df5ee S |
402 | check_deprecated('autonumber_size', '--autonumber-size', 'output template with %(autonumber)0Nd, where N in the number of digits') |
403 | check_deprecated('autonumber', '--auto-number', '-o "%(autonumber)s-%(title)s.%(ext)s"') | |
404 | check_deprecated('usetitle', '--title', '-o "%(title)s-%(id)s.%(ext)s"') | |
405 | ||
0783b09b | 406 | if params.get('bidi_workaround', False): |
1c088fa8 PH |
407 | try: |
408 | import pty | |
409 | master, slave = pty.openpty() | |
003c69a8 | 410 | width = compat_get_terminal_size().columns |
1c088fa8 PH |
411 | if width is None: |
412 | width_args = [] | |
413 | else: | |
414 | width_args = ['-w', str(width)] | |
5d681e96 | 415 | sp_kwargs = dict( |
1c088fa8 PH |
416 | stdin=subprocess.PIPE, |
417 | stdout=slave, | |
418 | stderr=self._err_file) | |
5d681e96 PH |
419 | try: |
420 | self._output_process = subprocess.Popen( | |
421 | ['bidiv'] + width_args, **sp_kwargs | |
422 | ) | |
423 | except OSError: | |
5d681e96 PH |
424 | self._output_process = subprocess.Popen( |
425 | ['fribidi', '-c', 'UTF-8'] + width_args, **sp_kwargs) | |
426 | self._output_channel = os.fdopen(master, 'rb') | |
1c088fa8 | 427 | except OSError as ose: |
66e7ace1 | 428 | if ose.errno == errno.ENOENT: |
6febd1c1 | 429 | 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 $PATH.') |
1c088fa8 PH |
430 | else: |
431 | raise | |
0783b09b | 432 | |
3089bc74 S |
433 | if (sys.platform != 'win32' |
434 | and sys.getfilesystemencoding() in ['ascii', 'ANSI_X3.4-1968'] | |
435 | and not params.get('restrictfilenames', False)): | |
e9137224 | 436 | # Unicode filesystem API will throw errors (#1474, #13027) |
34308b30 | 437 | self.report_warning( |
6febd1c1 | 438 | 'Assuming --restrict-filenames since file system encoding ' |
1b725173 | 439 | 'cannot encode all characters. ' |
6febd1c1 | 440 | 'Set the LC_ALL environment variable to fix this.') |
4a98cdbf | 441 | self.params['restrictfilenames'] = True |
34308b30 | 442 | |
486dd09e PH |
443 | if isinstance(params.get('outtmpl'), bytes): |
444 | self.report_warning( | |
445 | 'Parameter outtmpl is bytes, but should be a unicode string. ' | |
446 | 'Put from __future__ import unicode_literals at the top of your code file or consider switching to Python 3.x.') | |
447 | ||
dca08720 PH |
448 | self._setup_opener() |
449 | ||
3511266b PH |
450 | if auto_init: |
451 | self.print_debug_header() | |
452 | self.add_default_info_extractors() | |
453 | ||
4f026faf PH |
454 | for pp_def_raw in self.params.get('postprocessors', []): |
455 | pp_class = get_postprocessor(pp_def_raw['key']) | |
456 | pp_def = dict(pp_def_raw) | |
457 | del pp_def['key'] | |
458 | pp = pp_class(self, **compat_kwargs(pp_def)) | |
459 | self.add_post_processor(pp) | |
460 | ||
71b640cc PH |
461 | for ph in self.params.get('progress_hooks', []): |
462 | self.add_progress_hook(ph) | |
463 | ||
51fb4995 YCH |
464 | register_socks_protocols() |
465 | ||
7d4111ed PH |
466 | def warn_if_short_id(self, argv): |
467 | # short YouTube ID starting with dash? | |
468 | idxs = [ | |
469 | i for i, a in enumerate(argv) | |
470 | if re.match(r'^-[0-9A-Za-z_-]{10}$', a)] | |
471 | if idxs: | |
472 | correct_argv = ( | |
cefecac1 | 473 | ['youtube-dlc'] |
3089bc74 S |
474 | + [a for i, a in enumerate(argv) if i not in idxs] |
475 | + ['--'] + [argv[i] for i in idxs] | |
7d4111ed PH |
476 | ) |
477 | self.report_warning( | |
478 | 'Long argument string detected. ' | |
479 | 'Use -- to separate parameters and URLs, like this:\n%s\n' % | |
480 | args_to_str(correct_argv)) | |
481 | ||
8222d8de JMF |
482 | def add_info_extractor(self, ie): |
483 | """Add an InfoExtractor object to the end of the list.""" | |
484 | self._ies.append(ie) | |
e52d7f85 JMF |
485 | if not isinstance(ie, type): |
486 | self._ies_instances[ie.ie_key()] = ie | |
487 | ie.set_downloader(self) | |
8222d8de | 488 | |
56c73665 JMF |
489 | def get_info_extractor(self, ie_key): |
490 | """ | |
491 | Get an instance of an IE with name ie_key, it will try to get one from | |
492 | the _ies list, if there's no instance it will create a new one and add | |
493 | it to the extractor list. | |
494 | """ | |
495 | ie = self._ies_instances.get(ie_key) | |
496 | if ie is None: | |
497 | ie = get_info_extractor(ie_key)() | |
498 | self.add_info_extractor(ie) | |
499 | return ie | |
500 | ||
023fa8c4 JMF |
501 | def add_default_info_extractors(self): |
502 | """ | |
503 | Add the InfoExtractors returned by gen_extractors to the end of the list | |
504 | """ | |
e52d7f85 | 505 | for ie in gen_extractor_classes(): |
023fa8c4 JMF |
506 | self.add_info_extractor(ie) |
507 | ||
8222d8de JMF |
508 | def add_post_processor(self, pp): |
509 | """Add a PostProcessor object to the end of the chain.""" | |
510 | self._pps.append(pp) | |
511 | pp.set_downloader(self) | |
512 | ||
933605d7 JMF |
513 | def add_progress_hook(self, ph): |
514 | """Add the progress hook (currently only for the file downloader)""" | |
515 | self._progress_hooks.append(ph) | |
8ab470f1 | 516 | |
1c088fa8 | 517 | def _bidi_workaround(self, message): |
5d681e96 | 518 | if not hasattr(self, '_output_channel'): |
1c088fa8 PH |
519 | return message |
520 | ||
5d681e96 | 521 | assert hasattr(self, '_output_process') |
11b85ce6 | 522 | assert isinstance(message, compat_str) |
6febd1c1 PH |
523 | line_count = message.count('\n') + 1 |
524 | self._output_process.stdin.write((message + '\n').encode('utf-8')) | |
5d681e96 | 525 | self._output_process.stdin.flush() |
6febd1c1 | 526 | res = ''.join(self._output_channel.readline().decode('utf-8') |
9e1a5b84 | 527 | for _ in range(line_count)) |
6febd1c1 | 528 | return res[:-len('\n')] |
1c088fa8 | 529 | |
8222d8de | 530 | def to_screen(self, message, skip_eol=False): |
0783b09b PH |
531 | """Print message to stdout if not in quiet mode.""" |
532 | return self.to_stdout(message, skip_eol, check_quiet=True) | |
533 | ||
734f90bb | 534 | def _write_string(self, s, out=None): |
b58ddb32 | 535 | write_string(s, out=out, encoding=self.params.get('encoding')) |
734f90bb | 536 | |
0783b09b | 537 | def to_stdout(self, message, skip_eol=False, check_quiet=False): |
8222d8de | 538 | """Print message to stdout if not in quiet mode.""" |
8bf9319e | 539 | if self.params.get('logger'): |
43afe285 | 540 | self.params['logger'].debug(message) |
0783b09b | 541 | elif not check_quiet or not self.params.get('quiet', False): |
1c088fa8 | 542 | message = self._bidi_workaround(message) |
6febd1c1 | 543 | terminator = ['\n', ''][skip_eol] |
8222d8de | 544 | output = message + terminator |
1c088fa8 | 545 | |
734f90bb | 546 | self._write_string(output, self._screen_file) |
8222d8de JMF |
547 | |
548 | def to_stderr(self, message): | |
549 | """Print message to stderr.""" | |
11b85ce6 | 550 | assert isinstance(message, compat_str) |
8bf9319e | 551 | if self.params.get('logger'): |
43afe285 IB |
552 | self.params['logger'].error(message) |
553 | else: | |
1c088fa8 | 554 | message = self._bidi_workaround(message) |
6febd1c1 | 555 | output = message + '\n' |
734f90bb | 556 | self._write_string(output, self._err_file) |
8222d8de | 557 | |
1e5b9a95 PH |
558 | def to_console_title(self, message): |
559 | if not self.params.get('consoletitle', False): | |
560 | return | |
4bede0d8 C |
561 | if compat_os_name == 'nt': |
562 | if ctypes.windll.kernel32.GetConsoleWindow(): | |
563 | # c_wchar_p() might not be necessary if `message` is | |
564 | # already of type unicode() | |
565 | ctypes.windll.kernel32.SetConsoleTitleW(ctypes.c_wchar_p(message)) | |
1e5b9a95 | 566 | elif 'TERM' in os.environ: |
734f90bb | 567 | self._write_string('\033]0;%s\007' % message, self._screen_file) |
1e5b9a95 | 568 | |
bdde425c PH |
569 | def save_console_title(self): |
570 | if not self.params.get('consoletitle', False): | |
571 | return | |
94c3442e S |
572 | if self.params.get('simulate', False): |
573 | return | |
4bede0d8 | 574 | if compat_os_name != 'nt' and 'TERM' in os.environ: |
efd6c574 | 575 | # Save the title on stack |
734f90bb | 576 | self._write_string('\033[22;0t', self._screen_file) |
bdde425c PH |
577 | |
578 | def restore_console_title(self): | |
579 | if not self.params.get('consoletitle', False): | |
580 | return | |
94c3442e S |
581 | if self.params.get('simulate', False): |
582 | return | |
4bede0d8 | 583 | if compat_os_name != 'nt' and 'TERM' in os.environ: |
efd6c574 | 584 | # Restore the title from stack |
734f90bb | 585 | self._write_string('\033[23;0t', self._screen_file) |
bdde425c PH |
586 | |
587 | def __enter__(self): | |
588 | self.save_console_title() | |
589 | return self | |
590 | ||
591 | def __exit__(self, *args): | |
592 | self.restore_console_title() | |
f89197d7 | 593 | |
dca08720 | 594 | if self.params.get('cookiefile') is not None: |
1bab3437 | 595 | self.cookiejar.save(ignore_discard=True, ignore_expires=True) |
bdde425c | 596 | |
8222d8de JMF |
597 | def trouble(self, message=None, tb=None): |
598 | """Determine action to take when a download problem appears. | |
599 | ||
600 | Depending on if the downloader has been configured to ignore | |
601 | download errors or not, this method may throw an exception or | |
602 | not when errors are found, after printing the message. | |
603 | ||
604 | tb, if given, is additional traceback information. | |
605 | """ | |
606 | if message is not None: | |
607 | self.to_stderr(message) | |
608 | if self.params.get('verbose'): | |
609 | if tb is None: | |
610 | if sys.exc_info()[0]: # if .trouble has been called from an except block | |
6febd1c1 | 611 | tb = '' |
8222d8de | 612 | if hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]: |
6febd1c1 | 613 | tb += ''.join(traceback.format_exception(*sys.exc_info()[1].exc_info)) |
c0384f22 | 614 | tb += encode_compat_str(traceback.format_exc()) |
8222d8de JMF |
615 | else: |
616 | tb_data = traceback.format_list(traceback.extract_stack()) | |
6febd1c1 | 617 | tb = ''.join(tb_data) |
8222d8de JMF |
618 | self.to_stderr(tb) |
619 | if not self.params.get('ignoreerrors', False): | |
620 | if sys.exc_info()[0] and hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]: | |
621 | exc_info = sys.exc_info()[1].exc_info | |
622 | else: | |
623 | exc_info = sys.exc_info() | |
624 | raise DownloadError(message, exc_info) | |
625 | self._download_retcode = 1 | |
626 | ||
627 | def report_warning(self, message): | |
628 | ''' | |
629 | Print the message to stderr, it will be prefixed with 'WARNING:' | |
630 | If stderr is a tty file the 'WARNING:' will be colored | |
631 | ''' | |
6d07ce01 JMF |
632 | if self.params.get('logger') is not None: |
633 | self.params['logger'].warning(message) | |
8222d8de | 634 | else: |
ad8915b7 PH |
635 | if self.params.get('no_warnings'): |
636 | return | |
e9c0cdd3 | 637 | if not self.params.get('no_color') and self._err_file.isatty() and compat_os_name != 'nt': |
6d07ce01 JMF |
638 | _msg_header = '\033[0;33mWARNING:\033[0m' |
639 | else: | |
640 | _msg_header = 'WARNING:' | |
641 | warning_message = '%s %s' % (_msg_header, message) | |
642 | self.to_stderr(warning_message) | |
8222d8de JMF |
643 | |
644 | def report_error(self, message, tb=None): | |
645 | ''' | |
646 | Do the same as trouble, but prefixes the message with 'ERROR:', colored | |
647 | in red if stderr is a tty file. | |
648 | ''' | |
e9c0cdd3 | 649 | if not self.params.get('no_color') and self._err_file.isatty() and compat_os_name != 'nt': |
6febd1c1 | 650 | _msg_header = '\033[0;31mERROR:\033[0m' |
8222d8de | 651 | else: |
6febd1c1 PH |
652 | _msg_header = 'ERROR:' |
653 | error_message = '%s %s' % (_msg_header, message) | |
8222d8de JMF |
654 | self.trouble(error_message, tb) |
655 | ||
8222d8de JMF |
656 | def report_file_already_downloaded(self, file_name): |
657 | """Report file has already been fully downloaded.""" | |
658 | try: | |
6febd1c1 | 659 | self.to_screen('[download] %s has already been downloaded' % file_name) |
ce02ed60 | 660 | except UnicodeEncodeError: |
6febd1c1 | 661 | self.to_screen('[download] The file has already been downloaded') |
8222d8de | 662 | |
8222d8de JMF |
663 | def prepare_filename(self, info_dict): |
664 | """Generate the output filename.""" | |
665 | try: | |
666 | template_dict = dict(info_dict) | |
667 | ||
668 | template_dict['epoch'] = int(time.time()) | |
669 | autonumber_size = self.params.get('autonumber_size') | |
670 | if autonumber_size is None: | |
671 | autonumber_size = 5 | |
89db639d | 672 | template_dict['autonumber'] = self.params.get('autonumber_start', 1) - 1 + self._num_downloads |
17b75c0d PH |
673 | if template_dict.get('resolution') is None: |
674 | if template_dict.get('width') and template_dict.get('height'): | |
675 | template_dict['resolution'] = '%dx%d' % (template_dict['width'], template_dict['height']) | |
676 | elif template_dict.get('height'): | |
805ef3c6 | 677 | template_dict['resolution'] = '%sp' % template_dict['height'] |
17b75c0d | 678 | elif template_dict.get('width'): |
51ce9117 | 679 | template_dict['resolution'] = '%dx?' % template_dict['width'] |
8222d8de | 680 | |
586a91b6 | 681 | sanitize = lambda k, v: sanitize_filename( |
45598aab | 682 | compat_str(v), |
1bb5c511 | 683 | restricted=self.params.get('restrictfilenames'), |
40df485f | 684 | is_id=(k == 'id' or k.endswith('_id'))) |
d0d9ade4 | 685 | template_dict = dict((k, v if isinstance(v, compat_numeric_types) else sanitize(k, v)) |
45598aab | 686 | for k, v in template_dict.items() |
f0e14fdd | 687 | if v is not None and not isinstance(v, (list, tuple, dict))) |
6febd1c1 | 688 | template_dict = collections.defaultdict(lambda: 'NA', template_dict) |
8222d8de | 689 | |
b3613d36 | 690 | outtmpl = self.params.get('outtmpl', DEFAULT_OUTTMPL) |
d0d9ade4 | 691 | |
89db639d S |
692 | # For fields playlist_index and autonumber convert all occurrences |
693 | # of %(field)s to %(field)0Nd for backward compatibility | |
694 | field_size_compat_map = { | |
695 | 'playlist_index': len(str(template_dict['n_entries'])), | |
696 | 'autonumber': autonumber_size, | |
697 | } | |
698 | FIELD_SIZE_COMPAT_RE = r'(?<!%)%\((?P<field>autonumber|playlist_index)\)s' | |
699 | mobj = re.search(FIELD_SIZE_COMPAT_RE, outtmpl) | |
700 | if mobj: | |
701 | outtmpl = re.sub( | |
702 | FIELD_SIZE_COMPAT_RE, | |
703 | r'%%(\1)0%dd' % field_size_compat_map[mobj.group('field')], | |
704 | outtmpl) | |
705 | ||
d0d9ade4 S |
706 | # Missing numeric fields used together with integer presentation types |
707 | # in format specification will break the argument substitution since | |
708 | # string 'NA' is returned for missing fields. We will patch output | |
709 | # template for missing fields to meet string presentation type. | |
c9969434 | 710 | for numeric_field in self._NUMERIC_FIELDS: |
d0d9ade4 S |
711 | if numeric_field not in template_dict: |
712 | # As of [1] format syntax is: | |
713 | # %[mapping_key][conversion_flags][minimum_width][.precision][length_modifier]type | |
714 | # 1. https://docs.python.org/2/library/stdtypes.html#string-formatting | |
715 | FORMAT_RE = r'''(?x) | |
716 | (?<!%) | |
717 | % | |
718 | \({0}\) # mapping key | |
719 | (?:[#0\-+ ]+)? # conversion flags (optional) | |
720 | (?:\d+)? # minimum field width (optional) | |
721 | (?:\.\d+)? # precision (optional) | |
722 | [hlL]? # length modifier (optional) | |
723 | [diouxXeEfFgGcrs%] # conversion type | |
724 | ''' | |
725 | outtmpl = re.sub( | |
726 | FORMAT_RE.format(numeric_field), | |
727 | r'%({0})s'.format(numeric_field), outtmpl) | |
728 | ||
15da37c7 S |
729 | # expand_path translates '%%' into '%' and '$$' into '$' |
730 | # correspondingly that is not what we want since we need to keep | |
731 | # '%%' intact for template dict substitution step. Working around | |
732 | # with boundary-alike separator hack. | |
961ea474 | 733 | sep = ''.join([random.choice(ascii_letters) for _ in range(32)]) |
15da37c7 S |
734 | outtmpl = outtmpl.replace('%%', '%{0}%'.format(sep)).replace('$$', '${0}$'.format(sep)) |
735 | ||
736 | # outtmpl should be expand_path'ed before template dict substitution | |
737 | # because meta fields may contain env variables we don't want to | |
738 | # be expanded. For example, for outtmpl "%(title)s.%(ext)s" and | |
739 | # title "Hello $PATH", we don't want `$PATH` to be expanded. | |
740 | filename = expand_path(outtmpl).replace(sep, '') % template_dict | |
741 | ||
bdc3fd2f U |
742 | # https://github.com/blackjack4494/youtube-dlc/issues/85 |
743 | trim_file_name = self.params.get('trim_file_name', False) | |
744 | if trim_file_name: | |
745 | fn_groups = filename.rsplit('.') | |
746 | ext = fn_groups[-1] | |
747 | sub_ext = '' | |
748 | if len(fn_groups) > 2: | |
749 | sub_ext = fn_groups[-2] | |
750 | filename = '.'.join(filter(None, [fn_groups[0][:trim_file_name], sub_ext, ext])) | |
751 | ||
3a0d2f52 S |
752 | # Temporary fix for #4787 |
753 | # 'Treat' all problem characters by passing filename through preferredencoding | |
754 | # to workaround encoding issues with subprocess on python2 @ Windows | |
755 | if sys.version_info < (3, 0) and sys.platform == 'win32': | |
756 | filename = encodeFilename(filename, True).decode(preferredencoding()) | |
b3613d36 | 757 | return sanitize_path(filename) |
8222d8de | 758 | except ValueError as err: |
6febd1c1 | 759 | self.report_error('Error in output template: ' + str(err) + ' (encoding: ' + repr(preferredencoding()) + ')') |
8222d8de JMF |
760 | return None |
761 | ||
442c37b7 | 762 | def _match_entry(self, info_dict, incomplete): |
ecdec191 | 763 | """ Returns None if the file should be downloaded """ |
8222d8de | 764 | |
6febd1c1 | 765 | video_title = info_dict.get('title', info_dict.get('id', 'video')) |
7012b23c PH |
766 | if 'title' in info_dict: |
767 | # This can happen when we're just evaluating the playlist | |
768 | title = info_dict['title'] | |
769 | matchtitle = self.params.get('matchtitle', False) | |
770 | if matchtitle: | |
771 | if not re.search(matchtitle, title, re.IGNORECASE): | |
6febd1c1 | 772 | return '"' + title + '" title did not match pattern "' + matchtitle + '"' |
7012b23c PH |
773 | rejecttitle = self.params.get('rejecttitle', False) |
774 | if rejecttitle: | |
775 | if re.search(rejecttitle, title, re.IGNORECASE): | |
6febd1c1 | 776 | return '"' + title + '" title matched reject pattern "' + rejecttitle + '"' |
d800609c | 777 | date = info_dict.get('upload_date') |
8222d8de JMF |
778 | if date is not None: |
779 | dateRange = self.params.get('daterange', DateRange()) | |
780 | if date not in dateRange: | |
6febd1c1 | 781 | return '%s upload date is not in range %s' % (date_from_str(date).isoformat(), dateRange) |
d800609c | 782 | view_count = info_dict.get('view_count') |
5fe18bdb PH |
783 | if view_count is not None: |
784 | min_views = self.params.get('min_views') | |
785 | if min_views is not None and view_count < min_views: | |
6febd1c1 | 786 | return 'Skipping %s, because it has not reached minimum view count (%d/%d)' % (video_title, view_count, min_views) |
5fe18bdb PH |
787 | max_views = self.params.get('max_views') |
788 | if max_views is not None and view_count > max_views: | |
6febd1c1 | 789 | return 'Skipping %s, because it has exceeded the maximum view count (%d/%d)' % (video_title, view_count, max_views) |
05900629 | 790 | if age_restricted(info_dict.get('age_limit'), self.params.get('age_limit')): |
347de493 | 791 | return 'Skipping "%s" because it is age restricted' % video_title |
c1c9a79c | 792 | if self.in_download_archive(info_dict): |
6febd1c1 | 793 | return '%s has already been recorded in archive' % video_title |
347de493 | 794 | |
442c37b7 PH |
795 | if not incomplete: |
796 | match_filter = self.params.get('match_filter') | |
797 | if match_filter is not None: | |
798 | ret = match_filter(info_dict) | |
799 | if ret is not None: | |
800 | return ret | |
347de493 | 801 | |
8222d8de | 802 | return None |
fe7e0c98 | 803 | |
b6c45014 JMF |
804 | @staticmethod |
805 | def add_extra_info(info_dict, extra_info): | |
806 | '''Set the keys from extra_info in info dict if they are missing''' | |
807 | for key, value in extra_info.items(): | |
808 | info_dict.setdefault(key, value) | |
809 | ||
0704d222 | 810 | def extract_info(self, url, download=True, ie_key=None, info_dict=None, extra_info={}, |
61aa5ba3 | 811 | process=True, force_generic_extractor=False): |
8222d8de JMF |
812 | ''' |
813 | Returns a list with a dictionary for each video we find. | |
814 | If 'download', also downloads the videos. | |
815 | extra_info is a dict containing the extra values to add to each result | |
613b2d9d | 816 | ''' |
fe7e0c98 | 817 | |
61aa5ba3 | 818 | if not ie_key and force_generic_extractor: |
d22dec74 S |
819 | ie_key = 'Generic' |
820 | ||
8222d8de | 821 | if ie_key: |
56c73665 | 822 | ies = [self.get_info_extractor(ie_key)] |
8222d8de JMF |
823 | else: |
824 | ies = self._ies | |
825 | ||
826 | for ie in ies: | |
827 | if not ie.suitable(url): | |
828 | continue | |
829 | ||
9a68de12 | 830 | ie_key = ie.ie_key() |
831 | ie = self.get_info_extractor(ie_key) | |
8222d8de | 832 | if not ie.working(): |
6febd1c1 PH |
833 | self.report_warning('The program functionality for this site has been marked as broken, ' |
834 | 'and will probably not work.') | |
8222d8de JMF |
835 | |
836 | try: | |
a0566bbf | 837 | temp_id = ie.extract_id(url) if callable(getattr(ie, 'extract_id', None)) else ie._match_id(url) |
838 | except (AssertionError, IndexError, AttributeError): | |
839 | temp_id = None | |
840 | if temp_id is not None and self.in_download_archive({'id': temp_id, 'ie_key': ie_key}): | |
841 | self.to_screen("[%s] %s: has already been recorded in archive" % ( | |
842 | ie_key, temp_id)) | |
843 | break | |
844 | ||
845 | return self.__extract_info(url, ie, download, extra_info, process, info_dict) | |
846 | ||
847 | else: | |
848 | self.report_error('no suitable InfoExtractor for URL %s' % url) | |
849 | ||
850 | def __handle_extraction_exceptions(func): | |
851 | def wrapper(self, *args, **kwargs): | |
852 | try: | |
853 | return func(self, *args, **kwargs) | |
773f291d S |
854 | except GeoRestrictedError as e: |
855 | msg = e.msg | |
856 | if e.countries: | |
857 | msg += '\nThis video is available in %s.' % ', '.join( | |
858 | map(ISO3166Utils.short2full, e.countries)) | |
859 | msg += '\nYou might want to use a VPN or a proxy server (with --proxy) to workaround.' | |
860 | self.report_error(msg) | |
fb043a6e | 861 | except ExtractorError as e: # An error we somewhat expected |
2c74e6fa | 862 | self.report_error(compat_str(e), e.format_traceback()) |
d3e5bbf4 PH |
863 | except MaxDownloadsReached: |
864 | raise | |
8222d8de JMF |
865 | except Exception as e: |
866 | if self.params.get('ignoreerrors', False): | |
9b9c5355 | 867 | self.report_error(error_to_compat_str(e), tb=encode_compat_str(traceback.format_exc())) |
8222d8de JMF |
868 | else: |
869 | raise | |
a0566bbf | 870 | return wrapper |
871 | ||
872 | @__handle_extraction_exceptions | |
873 | def __extract_info(self, url, ie, download, extra_info, process, info_dict): | |
874 | ie_result = ie.extract(url) | |
875 | if ie_result is None: # Finished already (backwards compatibility; listformats and friends should be moved here) | |
876 | return | |
877 | if isinstance(ie_result, list): | |
878 | # Backwards compatibility: old IE result format | |
879 | ie_result = { | |
880 | '_type': 'compat_list', | |
881 | 'entries': ie_result, | |
882 | } | |
883 | if info_dict: | |
884 | if info_dict.get('id'): | |
885 | ie_result['id'] = info_dict['id'] | |
886 | if info_dict.get('title'): | |
887 | ie_result['title'] = info_dict['title'] | |
888 | self.add_default_extra_info(ie_result, ie, url) | |
889 | if process: | |
890 | return self.process_ie_result(ie_result, download, extra_info) | |
8222d8de | 891 | else: |
a0566bbf | 892 | return ie_result |
fe7e0c98 | 893 | |
ea38e55f PH |
894 | def add_default_extra_info(self, ie_result, ie, url): |
895 | self.add_extra_info(ie_result, { | |
896 | 'extractor': ie.IE_NAME, | |
897 | 'webpage_url': url, | |
898 | 'webpage_url_basename': url_basename(url), | |
899 | 'extractor_key': ie.ie_key(), | |
900 | }) | |
901 | ||
8222d8de JMF |
902 | def process_ie_result(self, ie_result, download=True, extra_info={}): |
903 | """ | |
904 | Take the result of the ie(may be modified) and resolve all unresolved | |
905 | references (URLs, playlist items). | |
906 | ||
907 | It will also download the videos if 'download'. | |
908 | Returns the resolved ie_result. | |
909 | """ | |
e8ee972c PH |
910 | result_type = ie_result.get('_type', 'video') |
911 | ||
057a5206 | 912 | if result_type in ('url', 'url_transparent'): |
134c6ea8 | 913 | ie_result['url'] = sanitize_url(ie_result['url']) |
057a5206 | 914 | extract_flat = self.params.get('extract_flat', False) |
3089bc74 S |
915 | if ((extract_flat == 'in_playlist' and 'playlist' in extra_info) |
916 | or extract_flat is True): | |
d06daf23 S |
917 | self.__forced_printings( |
918 | ie_result, self.prepare_filename(ie_result), | |
919 | incomplete=True) | |
e8ee972c PH |
920 | return ie_result |
921 | ||
8222d8de | 922 | if result_type == 'video': |
b6c45014 | 923 | self.add_extra_info(ie_result, extra_info) |
feee2ecf | 924 | return self.process_video_result(ie_result, download=download) |
8222d8de JMF |
925 | elif result_type == 'url': |
926 | # We have to add extra_info to the results because it may be | |
927 | # contained in a playlist | |
928 | return self.extract_info(ie_result['url'], | |
0704d222 | 929 | download, info_dict=ie_result, |
8222d8de JMF |
930 | ie_key=ie_result.get('ie_key'), |
931 | extra_info=extra_info) | |
7fc3fa05 PH |
932 | elif result_type == 'url_transparent': |
933 | # Use the information from the embedding page | |
934 | info = self.extract_info( | |
935 | ie_result['url'], ie_key=ie_result.get('ie_key'), | |
936 | extra_info=extra_info, download=False, process=False) | |
937 | ||
1640eb09 S |
938 | # extract_info may return None when ignoreerrors is enabled and |
939 | # extraction failed with an error, don't crash and return early | |
940 | # in this case | |
941 | if not info: | |
942 | return info | |
943 | ||
412c617d PH |
944 | force_properties = dict( |
945 | (k, v) for k, v in ie_result.items() if v is not None) | |
0396806f | 946 | for f in ('_type', 'url', 'id', 'extractor', 'extractor_key', 'ie_key'): |
412c617d PH |
947 | if f in force_properties: |
948 | del force_properties[f] | |
949 | new_result = info.copy() | |
950 | new_result.update(force_properties) | |
7fc3fa05 | 951 | |
0563f7ac S |
952 | # Extracted info may not be a video result (i.e. |
953 | # info.get('_type', 'video') != video) but rather an url or | |
954 | # url_transparent. In such cases outer metadata (from ie_result) | |
955 | # should be propagated to inner one (info). For this to happen | |
956 | # _type of info should be overridden with url_transparent. This | |
067aa17e | 957 | # fixes issue from https://github.com/ytdl-org/youtube-dl/pull/11163. |
0563f7ac S |
958 | if new_result.get('_type') == 'url': |
959 | new_result['_type'] = 'url_transparent' | |
7fc3fa05 PH |
960 | |
961 | return self.process_ie_result( | |
962 | new_result, download=download, extra_info=extra_info) | |
40fcba5e | 963 | elif result_type in ('playlist', 'multi_video'): |
8222d8de | 964 | # We process each entry in the playlist |
d800609c | 965 | playlist = ie_result.get('title') or ie_result.get('id') |
6febd1c1 | 966 | self.to_screen('[download] Downloading playlist: %s' % playlist) |
8222d8de JMF |
967 | |
968 | playlist_results = [] | |
969 | ||
8222d8de | 970 | playliststart = self.params.get('playliststart', 1) - 1 |
d800609c | 971 | playlistend = self.params.get('playlistend') |
a19fd00c | 972 | # For backwards compatibility, interpret -1 as whole list |
8222d8de | 973 | if playlistend == -1: |
a19fd00c | 974 | playlistend = None |
8222d8de | 975 | |
d800609c | 976 | playlistitems_str = self.params.get('playlist_items') |
c14e88f0 PH |
977 | playlistitems = None |
978 | if playlistitems_str is not None: | |
979 | def iter_playlistitems(format): | |
980 | for string_segment in format.split(','): | |
981 | if '-' in string_segment: | |
982 | start, end = string_segment.split('-') | |
983 | for item in range(int(start), int(end) + 1): | |
984 | yield int(item) | |
985 | else: | |
986 | yield int(string_segment) | |
cd6fc19e | 987 | playlistitems = orderedSet(iter_playlistitems(playlistitems_str)) |
c14e88f0 | 988 | |
b82f815f | 989 | ie_entries = ie_result['entries'] |
7e85e872 S |
990 | |
991 | def make_playlistitems_entries(list_ie_entries): | |
992 | num_entries = len(list_ie_entries) | |
993 | return [ | |
994 | list_ie_entries[i - 1] for i in playlistitems | |
995 | if -num_entries <= i - 1 < num_entries] | |
996 | ||
997 | def report_download(num_entries): | |
998 | self.to_screen( | |
999 | '[%s] playlist %s: Downloading %d videos' % | |
1000 | (ie_result['extractor'], playlist, num_entries)) | |
1001 | ||
b82f815f PH |
1002 | if isinstance(ie_entries, list): |
1003 | n_all_entries = len(ie_entries) | |
c14e88f0 | 1004 | if playlistitems: |
7e85e872 | 1005 | entries = make_playlistitems_entries(ie_entries) |
c14e88f0 PH |
1006 | else: |
1007 | entries = ie_entries[playliststart:playlistend] | |
b7ab0590 PH |
1008 | n_entries = len(entries) |
1009 | self.to_screen( | |
611c1dd9 | 1010 | '[%s] playlist %s: Collected %d video ids (downloading %d of them)' % |
b7ab0590 | 1011 | (ie_result['extractor'], playlist, n_all_entries, n_entries)) |
b82f815f | 1012 | elif isinstance(ie_entries, PagedList): |
c14e88f0 PH |
1013 | if playlistitems: |
1014 | entries = [] | |
1015 | for item in playlistitems: | |
1016 | entries.extend(ie_entries.getslice( | |
1017 | item - 1, item | |
1018 | )) | |
1019 | else: | |
1020 | entries = ie_entries.getslice( | |
1021 | playliststart, playlistend) | |
b7ab0590 | 1022 | n_entries = len(entries) |
7e85e872 | 1023 | report_download(n_entries) |
b82f815f | 1024 | else: # iterable |
c14e88f0 | 1025 | if playlistitems: |
5871ebac S |
1026 | entries = make_playlistitems_entries(list(itertools.islice( |
1027 | ie_entries, 0, max(playlistitems)))) | |
c14e88f0 PH |
1028 | else: |
1029 | entries = list(itertools.islice( | |
1030 | ie_entries, playliststart, playlistend)) | |
b82f815f | 1031 | n_entries = len(entries) |
7e85e872 | 1032 | report_download(n_entries) |
8222d8de | 1033 | |
ff815fe6 MS |
1034 | if self.params.get('playlistreverse', False): |
1035 | entries = entries[::-1] | |
1036 | ||
75822ca7 TC |
1037 | if self.params.get('playlistrandom', False): |
1038 | random.shuffle(entries) | |
1039 | ||
0016b84e S |
1040 | x_forwarded_for = ie_result.get('__x_forwarded_for_ip') |
1041 | ||
fe7e0c98 | 1042 | for i, entry in enumerate(entries, 1): |
734ea11e | 1043 | self.to_screen('[download] Downloading video %s of %s' % (i, n_entries)) |
0016b84e S |
1044 | # This __x_forwarded_for_ip thing is a bit ugly but requires |
1045 | # minimal changes | |
1046 | if x_forwarded_for: | |
1047 | entry['__x_forwarded_for_ip'] = x_forwarded_for | |
8222d8de | 1048 | extra = { |
c6b4132a | 1049 | 'n_entries': n_entries, |
fe7e0c98 | 1050 | 'playlist': playlist, |
a1cf99d0 PH |
1051 | 'playlist_id': ie_result.get('id'), |
1052 | 'playlist_title': ie_result.get('title'), | |
3961c6cb S |
1053 | 'playlist_uploader': ie_result.get('uploader'), |
1054 | 'playlist_uploader_id': ie_result.get('uploader_id'), | |
de1121d7 | 1055 | 'playlist_index': playlistitems[i - 1] if playlistitems else i + playliststart, |
b6c45014 | 1056 | 'extractor': ie_result['extractor'], |
9103bbc5 | 1057 | 'webpage_url': ie_result['webpage_url'], |
29eb5174 | 1058 | 'webpage_url_basename': url_basename(ie_result['webpage_url']), |
be97abc2 | 1059 | 'extractor_key': ie_result['extractor_key'], |
fe7e0c98 | 1060 | } |
7012b23c | 1061 | |
442c37b7 | 1062 | reason = self._match_entry(entry, incomplete=True) |
7012b23c | 1063 | if reason is not None: |
ea6e0c2b | 1064 | if reason.endswith('has already been recorded in the archive') and self.params.get('break_on_existing'): |
1065 | print('[download] tried downloading a file that\'s already in the archive, stopping since --break-on-existing is set.') | |
1066 | break | |
1067 | else: | |
1068 | self.to_screen('[download] ' + reason) | |
1069 | continue | |
7012b23c | 1070 | |
a0566bbf | 1071 | entry_result = self.__process_iterable_entry(entry, download, extra) |
1072 | # TODO: skip failed (empty) entries? | |
8222d8de JMF |
1073 | playlist_results.append(entry_result) |
1074 | ie_result['entries'] = playlist_results | |
371c3b79 | 1075 | self.to_screen('[download] Finished downloading playlist: %s' % playlist) |
8222d8de JMF |
1076 | return ie_result |
1077 | elif result_type == 'compat_list': | |
c9bf4114 PH |
1078 | self.report_warning( |
1079 | 'Extractor %s returned a compat_list result. ' | |
1080 | 'It needs to be updated.' % ie_result.get('extractor')) | |
5f6a1245 | 1081 | |
8222d8de | 1082 | def _fixup(r): |
9e1a5b84 JW |
1083 | self.add_extra_info( |
1084 | r, | |
9103bbc5 JMF |
1085 | { |
1086 | 'extractor': ie_result['extractor'], | |
1087 | 'webpage_url': ie_result['webpage_url'], | |
29eb5174 | 1088 | 'webpage_url_basename': url_basename(ie_result['webpage_url']), |
be97abc2 | 1089 | 'extractor_key': ie_result['extractor_key'], |
9e1a5b84 JW |
1090 | } |
1091 | ) | |
8222d8de JMF |
1092 | return r |
1093 | ie_result['entries'] = [ | |
b6c45014 | 1094 | self.process_ie_result(_fixup(r), download, extra_info) |
8222d8de JMF |
1095 | for r in ie_result['entries'] |
1096 | ] | |
1097 | return ie_result | |
1098 | else: | |
1099 | raise Exception('Invalid result type: %s' % result_type) | |
1100 | ||
a0566bbf | 1101 | @__handle_extraction_exceptions |
1102 | def __process_iterable_entry(self, entry, download, extra_info): | |
1103 | return self.process_ie_result( | |
1104 | entry, download=download, extra_info=extra_info) | |
1105 | ||
67134eab JMF |
1106 | def _build_format_filter(self, filter_spec): |
1107 | " Returns a function to filter the formats according to the filter_spec " | |
083c9df9 PH |
1108 | |
1109 | OPERATORS = { | |
1110 | '<': operator.lt, | |
1111 | '<=': operator.le, | |
1112 | '>': operator.gt, | |
1113 | '>=': operator.ge, | |
1114 | '=': operator.eq, | |
1115 | '!=': operator.ne, | |
1116 | } | |
67134eab | 1117 | operator_rex = re.compile(r'''(?x)\s* |
a03a3c80 | 1118 | (?P<key>width|height|tbr|abr|vbr|asr|filesize|filesize_approx|fps) |
083c9df9 PH |
1119 | \s*(?P<op>%s)(?P<none_inclusive>\s*\?)?\s* |
1120 | (?P<value>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?) | |
67134eab | 1121 | $ |
083c9df9 | 1122 | ''' % '|'.join(map(re.escape, OPERATORS.keys()))) |
67134eab | 1123 | m = operator_rex.search(filter_spec) |
9ddb6925 S |
1124 | if m: |
1125 | try: | |
1126 | comparison_value = int(m.group('value')) | |
1127 | except ValueError: | |
1128 | comparison_value = parse_filesize(m.group('value')) | |
1129 | if comparison_value is None: | |
1130 | comparison_value = parse_filesize(m.group('value') + 'B') | |
1131 | if comparison_value is None: | |
1132 | raise ValueError( | |
1133 | 'Invalid value %r in format specification %r' % ( | |
67134eab | 1134 | m.group('value'), filter_spec)) |
9ddb6925 S |
1135 | op = OPERATORS[m.group('op')] |
1136 | ||
083c9df9 | 1137 | if not m: |
9ddb6925 S |
1138 | STR_OPERATORS = { |
1139 | '=': operator.eq, | |
10d33b34 YCH |
1140 | '^=': lambda attr, value: attr.startswith(value), |
1141 | '$=': lambda attr, value: attr.endswith(value), | |
1142 | '*=': lambda attr, value: value in attr, | |
9ddb6925 | 1143 | } |
67134eab | 1144 | str_operator_rex = re.compile(r'''(?x) |
d5aacf9a | 1145 | \s*(?P<key>ext|acodec|vcodec|container|protocol|format_id) |
2cc779f4 | 1146 | \s*(?P<negation>!\s*)?(?P<op>%s)(?P<none_inclusive>\s*\?)? |
b0df5223 | 1147 | \s*(?P<value>[a-zA-Z0-9._-]+) |
67134eab | 1148 | \s*$ |
9ddb6925 | 1149 | ''' % '|'.join(map(re.escape, STR_OPERATORS.keys()))) |
67134eab | 1150 | m = str_operator_rex.search(filter_spec) |
9ddb6925 S |
1151 | if m: |
1152 | comparison_value = m.group('value') | |
2cc779f4 S |
1153 | str_op = STR_OPERATORS[m.group('op')] |
1154 | if m.group('negation'): | |
e118a879 | 1155 | op = lambda attr, value: not str_op(attr, value) |
2cc779f4 S |
1156 | else: |
1157 | op = str_op | |
083c9df9 | 1158 | |
9ddb6925 | 1159 | if not m: |
67134eab | 1160 | raise ValueError('Invalid filter specification %r' % filter_spec) |
083c9df9 PH |
1161 | |
1162 | def _filter(f): | |
1163 | actual_value = f.get(m.group('key')) | |
1164 | if actual_value is None: | |
1165 | return m.group('none_inclusive') | |
1166 | return op(actual_value, comparison_value) | |
67134eab JMF |
1167 | return _filter |
1168 | ||
0017d9ad | 1169 | def _default_format_spec(self, info_dict, download=True): |
0017d9ad | 1170 | |
af0f7428 S |
1171 | def can_merge(): |
1172 | merger = FFmpegMergerPP(self) | |
1173 | return merger.available and merger.can_merge() | |
1174 | ||
1175 | def prefer_best(): | |
0017d9ad | 1176 | if self.params.get('simulate', False): |
af0f7428 | 1177 | return False |
0017d9ad | 1178 | if not download: |
0017d9ad | 1179 | return False |
af0f7428 S |
1180 | if self.params.get('outtmpl', DEFAULT_OUTTMPL) == '-': |
1181 | return True | |
0017d9ad | 1182 | if info_dict.get('is_live'): |
af0f7428 S |
1183 | return True |
1184 | if not can_merge(): | |
1185 | return True | |
1186 | return False | |
1187 | ||
1188 | req_format_list = ['bestvideo+bestaudio', 'best'] | |
1189 | if prefer_best(): | |
1190 | req_format_list.reverse() | |
0017d9ad S |
1191 | return '/'.join(req_format_list) |
1192 | ||
67134eab JMF |
1193 | def build_format_selector(self, format_spec): |
1194 | def syntax_error(note, start): | |
1195 | message = ( | |
1196 | 'Invalid format specification: ' | |
1197 | '{0}\n\t{1}\n\t{2}^'.format(note, format_spec, ' ' * start[1])) | |
1198 | return SyntaxError(message) | |
1199 | ||
1200 | PICKFIRST = 'PICKFIRST' | |
1201 | MERGE = 'MERGE' | |
1202 | SINGLE = 'SINGLE' | |
0130afb7 | 1203 | GROUP = 'GROUP' |
67134eab JMF |
1204 | FormatSelector = collections.namedtuple('FormatSelector', ['type', 'selector', 'filters']) |
1205 | ||
909d24dd | 1206 | allow_multiple_streams = {'audio': self.params.get('allow_multiple_audio_streams', True), |
1207 | 'video': self.params.get('allow_multiple_video_streams', True)} | |
1208 | ||
67134eab JMF |
1209 | def _parse_filter(tokens): |
1210 | filter_parts = [] | |
1211 | for type, string, start, _, _ in tokens: | |
1212 | if type == tokenize.OP and string == ']': | |
1213 | return ''.join(filter_parts) | |
1214 | else: | |
1215 | filter_parts.append(string) | |
1216 | ||
232541df | 1217 | def _remove_unused_ops(tokens): |
17cc1534 | 1218 | # Remove operators that we don't use and join them with the surrounding strings |
232541df JMF |
1219 | # for example: 'mp4' '-' 'baseline' '-' '16x9' is converted to 'mp4-baseline-16x9' |
1220 | ALLOWED_OPS = ('/', '+', ',', '(', ')') | |
1221 | last_string, last_start, last_end, last_line = None, None, None, None | |
1222 | for type, string, start, end, line in tokens: | |
1223 | if type == tokenize.OP and string == '[': | |
1224 | if last_string: | |
1225 | yield tokenize.NAME, last_string, last_start, last_end, last_line | |
1226 | last_string = None | |
1227 | yield type, string, start, end, line | |
1228 | # everything inside brackets will be handled by _parse_filter | |
1229 | for type, string, start, end, line in tokens: | |
1230 | yield type, string, start, end, line | |
1231 | if type == tokenize.OP and string == ']': | |
1232 | break | |
1233 | elif type == tokenize.OP and string in ALLOWED_OPS: | |
1234 | if last_string: | |
1235 | yield tokenize.NAME, last_string, last_start, last_end, last_line | |
1236 | last_string = None | |
1237 | yield type, string, start, end, line | |
1238 | elif type in [tokenize.NAME, tokenize.NUMBER, tokenize.OP]: | |
1239 | if not last_string: | |
1240 | last_string = string | |
1241 | last_start = start | |
1242 | last_end = end | |
1243 | else: | |
1244 | last_string += string | |
1245 | if last_string: | |
1246 | yield tokenize.NAME, last_string, last_start, last_end, last_line | |
1247 | ||
cf2ac6df | 1248 | def _parse_format_selection(tokens, inside_merge=False, inside_choice=False, inside_group=False): |
67134eab JMF |
1249 | selectors = [] |
1250 | current_selector = None | |
1251 | for type, string, start, _, _ in tokens: | |
1252 | # ENCODING is only defined in python 3.x | |
1253 | if type == getattr(tokenize, 'ENCODING', None): | |
1254 | continue | |
1255 | elif type in [tokenize.NAME, tokenize.NUMBER]: | |
1256 | current_selector = FormatSelector(SINGLE, string, []) | |
1257 | elif type == tokenize.OP: | |
cf2ac6df JMF |
1258 | if string == ')': |
1259 | if not inside_group: | |
1260 | # ')' will be handled by the parentheses group | |
1261 | tokens.restore_last_token() | |
67134eab | 1262 | break |
cf2ac6df | 1263 | elif inside_merge and string in ['/', ',']: |
0130afb7 JMF |
1264 | tokens.restore_last_token() |
1265 | break | |
cf2ac6df JMF |
1266 | elif inside_choice and string == ',': |
1267 | tokens.restore_last_token() | |
1268 | break | |
1269 | elif string == ',': | |
0a31a350 JMF |
1270 | if not current_selector: |
1271 | raise syntax_error('"," must follow a format selector', start) | |
67134eab JMF |
1272 | selectors.append(current_selector) |
1273 | current_selector = None | |
1274 | elif string == '/': | |
d96d604e JMF |
1275 | if not current_selector: |
1276 | raise syntax_error('"/" must follow a format selector', start) | |
67134eab | 1277 | first_choice = current_selector |
cf2ac6df | 1278 | second_choice = _parse_format_selection(tokens, inside_choice=True) |
f5f4a27a | 1279 | current_selector = FormatSelector(PICKFIRST, (first_choice, second_choice), []) |
67134eab JMF |
1280 | elif string == '[': |
1281 | if not current_selector: | |
1282 | current_selector = FormatSelector(SINGLE, 'best', []) | |
1283 | format_filter = _parse_filter(tokens) | |
1284 | current_selector.filters.append(format_filter) | |
0130afb7 JMF |
1285 | elif string == '(': |
1286 | if current_selector: | |
1287 | raise syntax_error('Unexpected "("', start) | |
cf2ac6df JMF |
1288 | group = _parse_format_selection(tokens, inside_group=True) |
1289 | current_selector = FormatSelector(GROUP, group, []) | |
67134eab | 1290 | elif string == '+': |
d03cfdce | 1291 | if not current_selector: |
1292 | raise syntax_error('Unexpected "+"', start) | |
1293 | selector_1 = current_selector | |
1294 | selector_2 = _parse_format_selection(tokens, inside_merge=True) | |
1295 | if not selector_2: | |
1296 | raise syntax_error('Expected a selector', start) | |
1297 | current_selector = FormatSelector(MERGE, (selector_1, selector_2), []) | |
67134eab JMF |
1298 | else: |
1299 | raise syntax_error('Operator not recognized: "{0}"'.format(string), start) | |
1300 | elif type == tokenize.ENDMARKER: | |
1301 | break | |
1302 | if current_selector: | |
1303 | selectors.append(current_selector) | |
1304 | return selectors | |
1305 | ||
1306 | def _build_selector_function(selector): | |
909d24dd | 1307 | if isinstance(selector, list): # , |
67134eab JMF |
1308 | fs = [_build_selector_function(s) for s in selector] |
1309 | ||
317f7ab6 | 1310 | def selector_function(ctx): |
67134eab | 1311 | for f in fs: |
317f7ab6 | 1312 | for format in f(ctx): |
67134eab JMF |
1313 | yield format |
1314 | return selector_function | |
909d24dd | 1315 | |
1316 | elif selector.type == GROUP: # () | |
0130afb7 | 1317 | selector_function = _build_selector_function(selector.selector) |
909d24dd | 1318 | |
1319 | elif selector.type == PICKFIRST: # / | |
67134eab JMF |
1320 | fs = [_build_selector_function(s) for s in selector.selector] |
1321 | ||
317f7ab6 | 1322 | def selector_function(ctx): |
67134eab | 1323 | for f in fs: |
317f7ab6 | 1324 | picked_formats = list(f(ctx)) |
67134eab JMF |
1325 | if picked_formats: |
1326 | return picked_formats | |
1327 | return [] | |
67134eab | 1328 | |
909d24dd | 1329 | elif selector.type == SINGLE: # atom |
1330 | format_spec = selector.selector if selector.selector is not None else 'best' | |
1331 | ||
1332 | if format_spec == 'all': | |
1333 | def selector_function(ctx): | |
1334 | formats = list(ctx['formats']) | |
1335 | if formats: | |
1336 | for f in formats: | |
1337 | yield f | |
1338 | ||
1339 | else: | |
1340 | format_fallback = False | |
1341 | format_spec_obj = re.match(r'(best|worst|b|w)(video|audio|v|a)?(\*)?$', format_spec) | |
1342 | if format_spec_obj is not None: | |
1343 | format_idx = 0 if format_spec_obj.group(1)[0] == 'w' else -1 | |
1344 | format_type = format_spec_obj.group(2)[0] if format_spec_obj.group(2) else False | |
1345 | not_format_type = 'v' if format_type == 'a' else 'a' | |
1346 | format_modified = format_spec_obj.group(3) is not None | |
1347 | ||
1348 | format_fallback = not format_type and not format_modified # for b, w | |
1349 | filter_f = ((lambda f: f.get(format_type + 'codec') != 'none') | |
1350 | if format_type and format_modified # bv*, ba*, wv*, wa* | |
1351 | else (lambda f: f.get(not_format_type + 'codec') == 'none') | |
1352 | if format_type # bv, ba, wv, wa | |
1353 | else (lambda f: f.get('vcodec') != 'none' and f.get('acodec') != 'none') | |
1354 | if not format_modified # b, w | |
1355 | else None) # b*, w* | |
67134eab | 1356 | else: |
909d24dd | 1357 | format_idx = -1 |
1358 | filter_f = ((lambda f: f.get('ext') == format_spec) | |
1359 | if format_spec in ['mp4', 'flv', 'webm', '3gp', 'm4a', 'mp3', 'ogg', 'aac', 'wav'] # extension | |
1360 | else (lambda f: f.get('format_id') == format_spec)) # id | |
1361 | ||
1362 | def selector_function(ctx): | |
1363 | formats = list(ctx['formats']) | |
1364 | if not formats: | |
1365 | return | |
1366 | matches = list(filter(filter_f, formats)) if filter_f is not None else formats | |
67134eab | 1367 | if matches: |
909d24dd | 1368 | yield matches[format_idx] |
1369 | elif format_fallback == 'force' or (format_fallback and ctx['incomplete_formats']): | |
1370 | # for extractors with incomplete formats (audio only (soundcloud) | |
1371 | # or video only (imgur)) best/worst will fallback to | |
1372 | # best/worst {video,audio}-only format | |
1373 | yield formats[format_idx] | |
1374 | ||
1375 | elif selector.type == MERGE: # + | |
d03cfdce | 1376 | def _merge(formats_pair): |
1377 | format_1, format_2 = formats_pair | |
1378 | ||
1379 | formats_info = [] | |
1380 | formats_info.extend(format_1.get('requested_formats', (format_1,))) | |
1381 | formats_info.extend(format_2.get('requested_formats', (format_2,))) | |
1382 | ||
909d24dd | 1383 | if not allow_multiple_streams['video'] or not allow_multiple_streams['audio']: |
1384 | get_no_more = {"video": False, "audio": False} | |
1385 | for (i, fmt_info) in enumerate(formats_info): | |
1386 | for aud_vid in ["audio", "video"]: | |
1387 | if not allow_multiple_streams[aud_vid] and fmt_info.get(aud_vid[0] + 'codec') != 'none': | |
1388 | if get_no_more[aud_vid]: | |
1389 | formats_info.pop(i) | |
1390 | get_no_more[aud_vid] = True | |
1391 | ||
1392 | if len(formats_info) == 1: | |
1393 | return formats_info[0] | |
1394 | ||
d03cfdce | 1395 | video_fmts = [fmt_info for fmt_info in formats_info if fmt_info.get('vcodec') != 'none'] |
1396 | audio_fmts = [fmt_info for fmt_info in formats_info if fmt_info.get('acodec') != 'none'] | |
1397 | ||
1398 | the_only_video = video_fmts[0] if len(video_fmts) == 1 else None | |
1399 | the_only_audio = audio_fmts[0] if len(audio_fmts) == 1 else None | |
1400 | ||
1401 | output_ext = self.params.get('merge_output_format') | |
1402 | if not output_ext: | |
1403 | if the_only_video: | |
1404 | output_ext = the_only_video['ext'] | |
1405 | elif the_only_audio and not video_fmts: | |
1406 | output_ext = the_only_audio['ext'] | |
1407 | else: | |
1408 | output_ext = 'mkv' | |
1409 | ||
1410 | new_dict = { | |
67134eab | 1411 | 'requested_formats': formats_info, |
d03cfdce | 1412 | 'format': '+'.join(fmt_info.get('format') for fmt_info in formats_info), |
1413 | 'format_id': '+'.join(fmt_info.get('format_id') for fmt_info in formats_info), | |
67134eab JMF |
1414 | 'ext': output_ext, |
1415 | } | |
d03cfdce | 1416 | |
1417 | if the_only_video: | |
1418 | new_dict.update({ | |
1419 | 'width': the_only_video.get('width'), | |
1420 | 'height': the_only_video.get('height'), | |
1421 | 'resolution': the_only_video.get('resolution'), | |
1422 | 'fps': the_only_video.get('fps'), | |
1423 | 'vcodec': the_only_video.get('vcodec'), | |
1424 | 'vbr': the_only_video.get('vbr'), | |
1425 | 'stretched_ratio': the_only_video.get('stretched_ratio'), | |
1426 | }) | |
1427 | ||
1428 | if the_only_audio: | |
1429 | new_dict.update({ | |
1430 | 'acodec': the_only_audio.get('acodec'), | |
1431 | 'abr': the_only_audio.get('abr'), | |
1432 | }) | |
1433 | ||
1434 | return new_dict | |
1435 | ||
1436 | selector_1, selector_2 = map(_build_selector_function, selector.selector) | |
083c9df9 | 1437 | |
317f7ab6 S |
1438 | def selector_function(ctx): |
1439 | for pair in itertools.product( | |
d03cfdce | 1440 | selector_1(copy.deepcopy(ctx)), selector_2(copy.deepcopy(ctx))): |
67134eab | 1441 | yield _merge(pair) |
083c9df9 | 1442 | |
67134eab | 1443 | filters = [self._build_format_filter(f) for f in selector.filters] |
083c9df9 | 1444 | |
317f7ab6 S |
1445 | def final_selector(ctx): |
1446 | ctx_copy = copy.deepcopy(ctx) | |
67134eab | 1447 | for _filter in filters: |
317f7ab6 S |
1448 | ctx_copy['formats'] = list(filter(_filter, ctx_copy['formats'])) |
1449 | return selector_function(ctx_copy) | |
67134eab | 1450 | return final_selector |
083c9df9 | 1451 | |
67134eab | 1452 | stream = io.BytesIO(format_spec.encode('utf-8')) |
0130afb7 | 1453 | try: |
232541df | 1454 | tokens = list(_remove_unused_ops(compat_tokenize_tokenize(stream.readline))) |
0130afb7 JMF |
1455 | except tokenize.TokenError: |
1456 | raise syntax_error('Missing closing/opening brackets or parenthesis', (0, len(format_spec))) | |
1457 | ||
1458 | class TokenIterator(object): | |
1459 | def __init__(self, tokens): | |
1460 | self.tokens = tokens | |
1461 | self.counter = 0 | |
1462 | ||
1463 | def __iter__(self): | |
1464 | return self | |
1465 | ||
1466 | def __next__(self): | |
1467 | if self.counter >= len(self.tokens): | |
1468 | raise StopIteration() | |
1469 | value = self.tokens[self.counter] | |
1470 | self.counter += 1 | |
1471 | return value | |
1472 | ||
1473 | next = __next__ | |
1474 | ||
1475 | def restore_last_token(self): | |
1476 | self.counter -= 1 | |
1477 | ||
1478 | parsed_selector = _parse_format_selection(iter(TokenIterator(tokens))) | |
67134eab | 1479 | return _build_selector_function(parsed_selector) |
a9c58ad9 | 1480 | |
e5660ee6 JMF |
1481 | def _calc_headers(self, info_dict): |
1482 | res = std_headers.copy() | |
1483 | ||
1484 | add_headers = info_dict.get('http_headers') | |
1485 | if add_headers: | |
1486 | res.update(add_headers) | |
1487 | ||
1488 | cookies = self._calc_cookies(info_dict) | |
1489 | if cookies: | |
1490 | res['Cookie'] = cookies | |
1491 | ||
0016b84e S |
1492 | if 'X-Forwarded-For' not in res: |
1493 | x_forwarded_for_ip = info_dict.get('__x_forwarded_for_ip') | |
1494 | if x_forwarded_for_ip: | |
1495 | res['X-Forwarded-For'] = x_forwarded_for_ip | |
1496 | ||
e5660ee6 JMF |
1497 | return res |
1498 | ||
1499 | def _calc_cookies(self, info_dict): | |
5c2266df | 1500 | pr = sanitized_Request(info_dict['url']) |
e5660ee6 | 1501 | self.cookiejar.add_cookie_header(pr) |
662435f7 | 1502 | return pr.get_header('Cookie') |
e5660ee6 | 1503 | |
dd82ffea JMF |
1504 | def process_video_result(self, info_dict, download=True): |
1505 | assert info_dict.get('_type', 'video') == 'video' | |
1506 | ||
bec1fad2 PH |
1507 | if 'id' not in info_dict: |
1508 | raise ExtractorError('Missing "id" field in extractor result') | |
1509 | if 'title' not in info_dict: | |
1510 | raise ExtractorError('Missing "title" field in extractor result') | |
1511 | ||
c9969434 S |
1512 | def report_force_conversion(field, field_not, conversion): |
1513 | self.report_warning( | |
1514 | '"%s" field is not %s - forcing %s conversion, there is an error in extractor' | |
1515 | % (field, field_not, conversion)) | |
1516 | ||
1517 | def sanitize_string_field(info, string_field): | |
1518 | field = info.get(string_field) | |
1519 | if field is None or isinstance(field, compat_str): | |
1520 | return | |
1521 | report_force_conversion(string_field, 'a string', 'string') | |
1522 | info[string_field] = compat_str(field) | |
1523 | ||
1524 | def sanitize_numeric_fields(info): | |
1525 | for numeric_field in self._NUMERIC_FIELDS: | |
1526 | field = info.get(numeric_field) | |
1527 | if field is None or isinstance(field, compat_numeric_types): | |
1528 | continue | |
1529 | report_force_conversion(numeric_field, 'numeric', 'int') | |
1530 | info[numeric_field] = int_or_none(field) | |
1531 | ||
1532 | sanitize_string_field(info_dict, 'id') | |
1533 | sanitize_numeric_fields(info_dict) | |
be6217b2 | 1534 | |
dd82ffea JMF |
1535 | if 'playlist' not in info_dict: |
1536 | # It isn't part of a playlist | |
1537 | info_dict['playlist'] = None | |
1538 | info_dict['playlist_index'] = None | |
1539 | ||
d5519808 | 1540 | thumbnails = info_dict.get('thumbnails') |
cfb56d1a PH |
1541 | if thumbnails is None: |
1542 | thumbnail = info_dict.get('thumbnail') | |
1543 | if thumbnail: | |
a7a14d95 | 1544 | info_dict['thumbnails'] = thumbnails = [{'url': thumbnail}] |
d5519808 | 1545 | if thumbnails: |
be6d7229 | 1546 | thumbnails.sort(key=lambda t: ( |
d37708fc RA |
1547 | t.get('preference') if t.get('preference') is not None else -1, |
1548 | t.get('width') if t.get('width') is not None else -1, | |
1549 | t.get('height') if t.get('height') is not None else -1, | |
1550 | t.get('id') if t.get('id') is not None else '', t.get('url'))) | |
f6c24009 | 1551 | for i, t in enumerate(thumbnails): |
dcf77cf1 | 1552 | t['url'] = sanitize_url(t['url']) |
9603e8a7 | 1553 | if t.get('width') and t.get('height'): |
d5519808 | 1554 | t['resolution'] = '%dx%d' % (t['width'], t['height']) |
f6c24009 PH |
1555 | if t.get('id') is None: |
1556 | t['id'] = '%d' % i | |
d5519808 | 1557 | |
b7b72db9 | 1558 | if self.params.get('list_thumbnails'): |
1559 | self.list_thumbnails(info_dict) | |
1560 | return | |
1561 | ||
536a55da S |
1562 | thumbnail = info_dict.get('thumbnail') |
1563 | if thumbnail: | |
1564 | info_dict['thumbnail'] = sanitize_url(thumbnail) | |
1565 | elif thumbnails: | |
d5519808 PH |
1566 | info_dict['thumbnail'] = thumbnails[-1]['url'] |
1567 | ||
c9ae7b95 | 1568 | if 'display_id' not in info_dict and 'id' in info_dict: |
0afef30b PH |
1569 | info_dict['display_id'] = info_dict['id'] |
1570 | ||
955c4514 | 1571 | if info_dict.get('upload_date') is None and info_dict.get('timestamp') is not None: |
a55e36f4 S |
1572 | # Working around out-of-range timestamp values (e.g. negative ones on Windows, |
1573 | # see http://bugs.python.org/issue1646728) | |
1574 | try: | |
1575 | upload_date = datetime.datetime.utcfromtimestamp(info_dict['timestamp']) | |
1576 | info_dict['upload_date'] = upload_date.strftime('%Y%m%d') | |
1577 | except (ValueError, OverflowError, OSError): | |
1578 | pass | |
9d2ecdbc | 1579 | |
33d2fc2f S |
1580 | # Auto generate title fields corresponding to the *_number fields when missing |
1581 | # in order to always have clean titles. This is very common for TV series. | |
1582 | for field in ('chapter', 'season', 'episode'): | |
1583 | if info_dict.get('%s_number' % field) is not None and not info_dict.get(field): | |
1584 | info_dict[field] = '%s %d' % (field.capitalize(), info_dict['%s_number' % field]) | |
1585 | ||
05108a49 S |
1586 | for cc_kind in ('subtitles', 'automatic_captions'): |
1587 | cc = info_dict.get(cc_kind) | |
1588 | if cc: | |
1589 | for _, subtitle in cc.items(): | |
1590 | for subtitle_format in subtitle: | |
1591 | if subtitle_format.get('url'): | |
1592 | subtitle_format['url'] = sanitize_url(subtitle_format['url']) | |
1593 | if subtitle_format.get('ext') is None: | |
1594 | subtitle_format['ext'] = determine_ext(subtitle_format['url']).lower() | |
1595 | ||
1596 | automatic_captions = info_dict.get('automatic_captions') | |
4bba3716 | 1597 | subtitles = info_dict.get('subtitles') |
4bba3716 | 1598 | |
a504ced0 | 1599 | if self.params.get('listsubtitles', False): |
360e1ca5 | 1600 | if 'automatic_captions' in info_dict: |
05108a49 S |
1601 | self.list_subtitles( |
1602 | info_dict['id'], automatic_captions, 'automatic captions') | |
4bba3716 | 1603 | self.list_subtitles(info_dict['id'], subtitles, 'subtitles') |
a504ced0 | 1604 | return |
05108a49 | 1605 | |
360e1ca5 | 1606 | info_dict['requested_subtitles'] = self.process_subtitles( |
05108a49 | 1607 | info_dict['id'], subtitles, automatic_captions) |
a504ced0 | 1608 | |
dd82ffea JMF |
1609 | # We now pick which formats have to be downloaded |
1610 | if info_dict.get('formats') is None: | |
1611 | # There's only one format available | |
1612 | formats = [info_dict] | |
1613 | else: | |
1614 | formats = info_dict['formats'] | |
1615 | ||
db95dc13 PH |
1616 | if not formats: |
1617 | raise ExtractorError('No video formats found!') | |
1618 | ||
73af5cc8 S |
1619 | def is_wellformed(f): |
1620 | url = f.get('url') | |
a5ac0c47 | 1621 | if not url: |
73af5cc8 S |
1622 | self.report_warning( |
1623 | '"url" field is missing or empty - skipping format, ' | |
1624 | 'there is an error in extractor') | |
a5ac0c47 S |
1625 | return False |
1626 | if isinstance(url, bytes): | |
1627 | sanitize_string_field(f, 'url') | |
1628 | return True | |
73af5cc8 S |
1629 | |
1630 | # Filter out malformed formats for better extraction robustness | |
1631 | formats = list(filter(is_wellformed, formats)) | |
1632 | ||
181c7053 S |
1633 | formats_dict = {} |
1634 | ||
dd82ffea | 1635 | # We check that all the formats have the format and format_id fields |
db95dc13 | 1636 | for i, format in enumerate(formats): |
c9969434 S |
1637 | sanitize_string_field(format, 'format_id') |
1638 | sanitize_numeric_fields(format) | |
dcf77cf1 | 1639 | format['url'] = sanitize_url(format['url']) |
e74e3b63 | 1640 | if not format.get('format_id'): |
8016c922 | 1641 | format['format_id'] = compat_str(i) |
e2effb08 S |
1642 | else: |
1643 | # Sanitize format_id from characters used in format selector expression | |
ec85ded8 | 1644 | format['format_id'] = re.sub(r'[\s,/+\[\]()]', '_', format['format_id']) |
181c7053 S |
1645 | format_id = format['format_id'] |
1646 | if format_id not in formats_dict: | |
1647 | formats_dict[format_id] = [] | |
1648 | formats_dict[format_id].append(format) | |
1649 | ||
1650 | # Make sure all formats have unique format_id | |
1651 | for format_id, ambiguous_formats in formats_dict.items(): | |
1652 | if len(ambiguous_formats) > 1: | |
1653 | for i, format in enumerate(ambiguous_formats): | |
1654 | format['format_id'] = '%s-%d' % (format_id, i) | |
1655 | ||
1656 | for i, format in enumerate(formats): | |
8c51aa65 | 1657 | if format.get('format') is None: |
6febd1c1 | 1658 | format['format'] = '{id} - {res}{note}'.format( |
8c51aa65 JMF |
1659 | id=format['format_id'], |
1660 | res=self.format_resolution(format), | |
6febd1c1 | 1661 | note=' ({0})'.format(format['format_note']) if format.get('format_note') is not None else '', |
8c51aa65 | 1662 | ) |
c1002e96 | 1663 | # Automatically determine file extension if missing |
5b1d8575 | 1664 | if format.get('ext') is None: |
cce929ea | 1665 | format['ext'] = determine_ext(format['url']).lower() |
b5559424 S |
1666 | # Automatically determine protocol if missing (useful for format |
1667 | # selection purposes) | |
6f0be937 | 1668 | if format.get('protocol') is None: |
b5559424 | 1669 | format['protocol'] = determine_protocol(format) |
e5660ee6 JMF |
1670 | # Add HTTP headers, so that external programs can use them from the |
1671 | # json output | |
1672 | full_format_info = info_dict.copy() | |
1673 | full_format_info.update(format) | |
1674 | format['http_headers'] = self._calc_headers(full_format_info) | |
0016b84e S |
1675 | # Remove private housekeeping stuff |
1676 | if '__x_forwarded_for_ip' in info_dict: | |
1677 | del info_dict['__x_forwarded_for_ip'] | |
dd82ffea | 1678 | |
4bcc7bd1 | 1679 | # TODO Central sorting goes here |
99e206d5 | 1680 | |
f89197d7 | 1681 | if formats[0] is not info_dict: |
b3d9ef88 JMF |
1682 | # only set the 'formats' fields if the original info_dict list them |
1683 | # otherwise we end up with a circular reference, the first (and unique) | |
f89197d7 | 1684 | # element in the 'formats' field in info_dict is info_dict itself, |
dfb1b146 | 1685 | # which can't be exported to json |
b3d9ef88 | 1686 | info_dict['formats'] = formats |
cfb56d1a | 1687 | if self.params.get('listformats'): |
bfaae0a7 | 1688 | self.list_formats(info_dict) |
1689 | return | |
1690 | ||
de3ef3ed | 1691 | req_format = self.params.get('format') |
a9c58ad9 | 1692 | if req_format is None: |
0017d9ad S |
1693 | req_format = self._default_format_spec(info_dict, download=download) |
1694 | if self.params.get('verbose'): | |
1695 | self.to_stdout('[debug] Default format spec: %s' % req_format) | |
1696 | ||
5acfa126 | 1697 | format_selector = self.build_format_selector(req_format) |
317f7ab6 S |
1698 | |
1699 | # While in format selection we may need to have an access to the original | |
1700 | # format set in order to calculate some metrics or do some processing. | |
1701 | # For now we need to be able to guess whether original formats provided | |
1702 | # by extractor are incomplete or not (i.e. whether extractor provides only | |
1703 | # video-only or audio-only formats) for proper formats selection for | |
1704 | # extractors with such incomplete formats (see | |
067aa17e | 1705 | # https://github.com/ytdl-org/youtube-dl/pull/5556). |
317f7ab6 S |
1706 | # Since formats may be filtered during format selection and may not match |
1707 | # the original formats the results may be incorrect. Thus original formats | |
1708 | # or pre-calculated metrics should be passed to format selection routines | |
1709 | # as well. | |
1710 | # We will pass a context object containing all necessary additional data | |
1711 | # instead of just formats. | |
1712 | # This fixes incorrect format selection issue (see | |
067aa17e | 1713 | # https://github.com/ytdl-org/youtube-dl/issues/10083). |
2e221ca3 | 1714 | incomplete_formats = ( |
317f7ab6 | 1715 | # All formats are video-only or |
3089bc74 | 1716 | all(f.get('vcodec') != 'none' and f.get('acodec') == 'none' for f in formats) |
317f7ab6 | 1717 | # all formats are audio-only |
3089bc74 | 1718 | or all(f.get('vcodec') == 'none' and f.get('acodec') != 'none' for f in formats)) |
317f7ab6 S |
1719 | |
1720 | ctx = { | |
1721 | 'formats': formats, | |
1722 | 'incomplete_formats': incomplete_formats, | |
1723 | } | |
1724 | ||
1725 | formats_to_download = list(format_selector(ctx)) | |
dd82ffea | 1726 | if not formats_to_download: |
6febd1c1 | 1727 | raise ExtractorError('requested format not available', |
78a3a9f8 | 1728 | expected=True) |
dd82ffea JMF |
1729 | |
1730 | if download: | |
909d24dd | 1731 | self.to_screen('[info] Downloading format(s) %s' % ", ".join([f['format_id'] for f in formats_to_download])) |
dd82ffea | 1732 | if len(formats_to_download) > 1: |
6febd1c1 | 1733 | self.to_screen('[info] %s: downloading video in %s formats' % (info_dict['id'], len(formats_to_download))) |
dd82ffea JMF |
1734 | for format in formats_to_download: |
1735 | new_info = dict(info_dict) | |
1736 | new_info.update(format) | |
1737 | self.process_info(new_info) | |
1738 | # We update the info dict with the best quality format (backwards compatibility) | |
1739 | info_dict.update(formats_to_download[-1]) | |
1740 | return info_dict | |
1741 | ||
98c70d6f | 1742 | def process_subtitles(self, video_id, normal_subtitles, automatic_captions): |
a504ced0 | 1743 | """Select the requested subtitles and their format""" |
98c70d6f JMF |
1744 | available_subs = {} |
1745 | if normal_subtitles and self.params.get('writesubtitles'): | |
1746 | available_subs.update(normal_subtitles) | |
1747 | if automatic_captions and self.params.get('writeautomaticsub'): | |
1748 | for lang, cap_info in automatic_captions.items(): | |
360e1ca5 JMF |
1749 | if lang not in available_subs: |
1750 | available_subs[lang] = cap_info | |
1751 | ||
4d171848 JMF |
1752 | if (not self.params.get('writesubtitles') and not |
1753 | self.params.get('writeautomaticsub') or not | |
1754 | available_subs): | |
1755 | return None | |
a504ced0 JMF |
1756 | |
1757 | if self.params.get('allsubtitles', False): | |
1758 | requested_langs = available_subs.keys() | |
1759 | else: | |
1760 | if self.params.get('subtitleslangs', False): | |
1761 | requested_langs = self.params.get('subtitleslangs') | |
1762 | elif 'en' in available_subs: | |
1763 | requested_langs = ['en'] | |
1764 | else: | |
1765 | requested_langs = [list(available_subs.keys())[0]] | |
1766 | ||
1767 | formats_query = self.params.get('subtitlesformat', 'best') | |
1768 | formats_preference = formats_query.split('/') if formats_query else [] | |
1769 | subs = {} | |
1770 | for lang in requested_langs: | |
1771 | formats = available_subs.get(lang) | |
1772 | if formats is None: | |
1773 | self.report_warning('%s subtitles not available for %s' % (lang, video_id)) | |
1774 | continue | |
a504ced0 JMF |
1775 | for ext in formats_preference: |
1776 | if ext == 'best': | |
1777 | f = formats[-1] | |
1778 | break | |
1779 | matches = list(filter(lambda f: f['ext'] == ext, formats)) | |
1780 | if matches: | |
1781 | f = matches[-1] | |
1782 | break | |
1783 | else: | |
1784 | f = formats[-1] | |
1785 | self.report_warning( | |
1786 | 'No subtitle format found matching "%s" for language %s, ' | |
1787 | 'using %s' % (formats_query, lang, f['ext'])) | |
1788 | subs[lang] = f | |
1789 | return subs | |
1790 | ||
d06daf23 S |
1791 | def __forced_printings(self, info_dict, filename, incomplete): |
1792 | def print_mandatory(field): | |
1793 | if (self.params.get('force%s' % field, False) | |
1794 | and (not incomplete or info_dict.get(field) is not None)): | |
1795 | self.to_stdout(info_dict[field]) | |
1796 | ||
1797 | def print_optional(field): | |
1798 | if (self.params.get('force%s' % field, False) | |
1799 | and info_dict.get(field) is not None): | |
1800 | self.to_stdout(info_dict[field]) | |
1801 | ||
1802 | print_mandatory('title') | |
1803 | print_mandatory('id') | |
1804 | if self.params.get('forceurl', False) and not incomplete: | |
1805 | if info_dict.get('requested_formats') is not None: | |
1806 | for f in info_dict['requested_formats']: | |
1807 | self.to_stdout(f['url'] + f.get('play_path', '')) | |
1808 | else: | |
1809 | # For RTMP URLs, also include the playpath | |
1810 | self.to_stdout(info_dict['url'] + info_dict.get('play_path', '')) | |
1811 | print_optional('thumbnail') | |
1812 | print_optional('description') | |
1813 | if self.params.get('forcefilename', False) and filename is not None: | |
1814 | self.to_stdout(filename) | |
1815 | if self.params.get('forceduration', False) and info_dict.get('duration') is not None: | |
1816 | self.to_stdout(formatSeconds(info_dict['duration'])) | |
1817 | print_mandatory('format') | |
1818 | if self.params.get('forcejson', False): | |
1819 | self.to_stdout(json.dumps(info_dict)) | |
1820 | ||
8222d8de JMF |
1821 | def process_info(self, info_dict): |
1822 | """Process a single resolved IE result.""" | |
1823 | ||
1824 | assert info_dict.get('_type', 'video') == 'video' | |
fd288278 PH |
1825 | |
1826 | max_downloads = self.params.get('max_downloads') | |
1827 | if max_downloads is not None: | |
1828 | if self._num_downloads >= int(max_downloads): | |
1829 | raise MaxDownloadsReached() | |
8222d8de | 1830 | |
d06daf23 | 1831 | # TODO: backward compatibility, to be removed |
8222d8de | 1832 | info_dict['fulltitle'] = info_dict['title'] |
8222d8de | 1833 | |
11b85ce6 | 1834 | if 'format' not in info_dict: |
8222d8de JMF |
1835 | info_dict['format'] = info_dict['ext'] |
1836 | ||
442c37b7 | 1837 | reason = self._match_entry(info_dict, incomplete=False) |
8222d8de | 1838 | if reason is not None: |
6febd1c1 | 1839 | self.to_screen('[download] ' + reason) |
8222d8de JMF |
1840 | return |
1841 | ||
fd288278 | 1842 | self._num_downloads += 1 |
8222d8de | 1843 | |
e72c7e41 | 1844 | info_dict['_filename'] = filename = self.prepare_filename(info_dict) |
8222d8de JMF |
1845 | |
1846 | # Forced printings | |
d06daf23 | 1847 | self.__forced_printings(info_dict, filename, incomplete=False) |
8222d8de JMF |
1848 | |
1849 | # Do nothing else if in simulate mode | |
1850 | if self.params.get('simulate', False): | |
1851 | return | |
1852 | ||
1853 | if filename is None: | |
1854 | return | |
1855 | ||
c5c9bf0c S |
1856 | def ensure_dir_exists(path): |
1857 | try: | |
1858 | dn = os.path.dirname(path) | |
1859 | if dn and not os.path.exists(dn): | |
1860 | os.makedirs(dn) | |
1861 | return True | |
1862 | except (OSError, IOError) as err: | |
1863 | self.report_error('unable to create directory ' + error_to_compat_str(err)) | |
1864 | return False | |
1865 | ||
1866 | if not ensure_dir_exists(sanitize_path(encodeFilename(filename))): | |
8222d8de JMF |
1867 | return |
1868 | ||
1869 | if self.params.get('writedescription', False): | |
2699da80 | 1870 | descfn = replace_extension(filename, 'description', info_dict.get('ext')) |
7b6fefc9 | 1871 | if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(descfn)): |
6febd1c1 | 1872 | self.to_screen('[info] Video description is already present') |
f00fd51d JMF |
1873 | elif info_dict.get('description') is None: |
1874 | self.report_warning('There\'s no description to write.') | |
7b6fefc9 PH |
1875 | else: |
1876 | try: | |
6febd1c1 | 1877 | self.to_screen('[info] Writing video description to: ' + descfn) |
7b6fefc9 PH |
1878 | with io.open(encodeFilename(descfn), 'w', encoding='utf-8') as descfile: |
1879 | descfile.write(info_dict['description']) | |
7b6fefc9 | 1880 | except (OSError, IOError): |
6febd1c1 | 1881 | self.report_error('Cannot write description file ' + descfn) |
7b6fefc9 | 1882 | return |
8222d8de | 1883 | |
1fb07d10 | 1884 | if self.params.get('writeannotations', False): |
98727e12 | 1885 | annofn = replace_extension(filename, 'annotations.xml', info_dict.get('ext')) |
7b6fefc9 | 1886 | if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(annofn)): |
6febd1c1 | 1887 | self.to_screen('[info] Video annotations are already present') |
ffddb112 RA |
1888 | elif not info_dict.get('annotations'): |
1889 | self.report_warning('There are no annotations to write.') | |
7b6fefc9 PH |
1890 | else: |
1891 | try: | |
6febd1c1 | 1892 | self.to_screen('[info] Writing video annotations to: ' + annofn) |
7b6fefc9 PH |
1893 | with io.open(encodeFilename(annofn), 'w', encoding='utf-8') as annofile: |
1894 | annofile.write(info_dict['annotations']) | |
1895 | except (KeyError, TypeError): | |
6febd1c1 | 1896 | self.report_warning('There are no annotations to write.') |
7b6fefc9 | 1897 | except (OSError, IOError): |
6febd1c1 | 1898 | self.report_error('Cannot write annotations file: ' + annofn) |
7b6fefc9 | 1899 | return |
1fb07d10 | 1900 | |
9f448fcb | 1901 | def dl(name, info, subtitle=False): |
98b69821 | 1902 | fd = get_suitable_downloader(info, self.params)(self, self.params) |
1903 | for ph in self._progress_hooks: | |
1904 | fd.add_progress_hook(ph) | |
1905 | if self.params.get('verbose'): | |
1906 | self.to_stdout('[debug] Invoking downloader on %r' % info.get('url')) | |
9f448fcb | 1907 | return fd.download(name, info, subtitle) |
98b69821 | 1908 | |
c4a91be7 | 1909 | subtitles_are_requested = any([self.params.get('writesubtitles', False), |
0b7f3118 | 1910 | self.params.get('writeautomaticsub')]) |
c4a91be7 | 1911 | |
c84dd8a9 | 1912 | if subtitles_are_requested and info_dict.get('requested_subtitles'): |
8222d8de JMF |
1913 | # subtitles download errors are already managed as troubles in relevant IE |
1914 | # that way it will silently go on when used with unsupporting IE | |
c84dd8a9 | 1915 | subtitles = info_dict['requested_subtitles'] |
fa57af1e | 1916 | # ie = self.get_info_extractor(info_dict['extractor_key']) |
a504ced0 JMF |
1917 | for sub_lang, sub_info in subtitles.items(): |
1918 | sub_format = sub_info['ext'] | |
824fa511 | 1919 | sub_filename = subtitles_filename(filename, sub_lang, sub_format, info_dict.get('ext')) |
5ff1bc0c RA |
1920 | if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(sub_filename)): |
1921 | self.to_screen('[info] Video subtitle %s.%s is already present' % (sub_lang, sub_format)) | |
a504ced0 | 1922 | else: |
0c9df79e | 1923 | self.to_screen('[info] Writing video subtitles to: ' + sub_filename) |
5ff1bc0c RA |
1924 | if sub_info.get('data') is not None: |
1925 | try: | |
1926 | # Use newline='' to prevent conversion of newline characters | |
067aa17e | 1927 | # See https://github.com/ytdl-org/youtube-dl/issues/10268 |
5ff1bc0c RA |
1928 | with io.open(encodeFilename(sub_filename), 'w', encoding='utf-8', newline='') as subfile: |
1929 | subfile.write(sub_info['data']) | |
1930 | except (OSError, IOError): | |
1931 | self.report_error('Cannot write subtitles file ' + sub_filename) | |
1932 | return | |
7b6fefc9 | 1933 | else: |
5ff1bc0c | 1934 | try: |
9f448fcb U |
1935 | dl(sub_filename, sub_info, subtitle=True) |
1936 | ''' | |
0c9df79e U |
1937 | if self.params.get('sleep_interval_subtitles', False): |
1938 | dl(sub_filename, sub_info) | |
1939 | else: | |
1940 | sub_data = ie._request_webpage( | |
1941 | sub_info['url'], info_dict['id'], note=False).read() | |
1942 | with io.open(encodeFilename(sub_filename), 'wb') as subfile: | |
1943 | subfile.write(sub_data) | |
9f448fcb | 1944 | ''' |
0c9df79e | 1945 | except (ExtractorError, IOError, OSError, ValueError, compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: |
5ff1bc0c RA |
1946 | self.report_warning('Unable to download subtitle for "%s": %s' % |
1947 | (sub_lang, error_to_compat_str(err))) | |
1948 | continue | |
8222d8de | 1949 | |
57df9f53 U |
1950 | if self.params.get('skip_download', False): |
1951 | if self.params.get('convertsubtitles', False): | |
1952 | subconv = FFmpegSubtitlesConvertorPP(self, format=self.params.get('convertsubtitles')) | |
1953 | filename_real_ext = os.path.splitext(filename)[1][1:] | |
1954 | filename_wo_ext = ( | |
1955 | os.path.splitext(filename)[0] | |
1956 | if filename_real_ext == info_dict['ext'] | |
1957 | else filename) | |
1958 | afilename = '%s.%s' % (filename_wo_ext, self.params.get('convertsubtitles')) | |
1959 | if subconv.available: | |
1960 | info_dict.setdefault('__postprocessors', []) | |
1961 | # info_dict['__postprocessors'].append(subconv) | |
1962 | if os.path.exists(encodeFilename(afilename)): | |
f791b419 U |
1963 | self.to_screen( |
1964 | '[download] %s has already been downloaded and ' | |
1965 | 'converted' % afilename) | |
57df9f53 U |
1966 | else: |
1967 | try: | |
1968 | self.post_process(filename, info_dict) | |
1969 | except (PostProcessingError) as err: | |
1970 | self.report_error('postprocessing: %s' % str(err)) | |
1971 | return | |
1972 | ||
8222d8de | 1973 | if self.params.get('writeinfojson', False): |
b29e0000 | 1974 | infofn = replace_extension(filename, 'info.json', info_dict.get('ext')) |
7b6fefc9 | 1975 | if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(infofn)): |
6febd1c1 | 1976 | self.to_screen('[info] Video description metadata is already present') |
7b6fefc9 | 1977 | else: |
6febd1c1 | 1978 | self.to_screen('[info] Writing video description metadata as JSON to: ' + infofn) |
7b6fefc9 | 1979 | try: |
cb202fd2 | 1980 | write_json_file(self.filter_requested_info(info_dict), infofn) |
7b6fefc9 | 1981 | except (OSError, IOError): |
6febd1c1 | 1982 | self.report_error('Cannot write metadata to JSON file ' + infofn) |
7b6fefc9 | 1983 | return |
8222d8de | 1984 | |
ec82d85a | 1985 | self._write_thumbnails(info_dict, filename) |
8222d8de JMF |
1986 | |
1987 | if not self.params.get('skip_download', False): | |
4340deca | 1988 | try: |
4340deca P |
1989 | if info_dict.get('requested_formats') is not None: |
1990 | downloaded = [] | |
1991 | success = True | |
d47aeb22 | 1992 | merger = FFmpegMergerPP(self) |
f740fae2 | 1993 | if not merger.available: |
4340deca P |
1994 | postprocessors = [] |
1995 | self.report_warning('You have requested multiple ' | |
1996 | 'formats but ffmpeg or avconv are not installed.' | |
4a5a898a | 1997 | ' The formats won\'t be merged.') |
6350728b | 1998 | else: |
4340deca | 1999 | postprocessors = [merger] |
81cd954a S |
2000 | |
2001 | def compatible_formats(formats): | |
d03cfdce | 2002 | # TODO: some formats actually allow this (mkv, webm, ogg, mp4), but not all of them. |
2003 | video_formats = [format for format in formats if format.get('vcodec') != 'none'] | |
2004 | audio_formats = [format for format in formats if format.get('acodec') != 'none'] | |
2005 | if len(video_formats) > 2 or len(audio_formats) > 2: | |
2006 | return False | |
2007 | ||
81cd954a | 2008 | # Check extension |
d03cfdce | 2009 | exts = set(format.get('ext') for format in formats) |
2010 | COMPATIBLE_EXTS = ( | |
2011 | set(('mp3', 'mp4', 'm4a', 'm4p', 'm4b', 'm4r', 'm4v', 'ismv', 'isma')), | |
2012 | set(('webm',)), | |
2013 | ) | |
2014 | for ext_sets in COMPATIBLE_EXTS: | |
2015 | if ext_sets.issuperset(exts): | |
2016 | return True | |
81cd954a S |
2017 | # TODO: Check acodec/vcodec |
2018 | return False | |
2019 | ||
38c6902b S |
2020 | filename_real_ext = os.path.splitext(filename)[1][1:] |
2021 | filename_wo_ext = ( | |
2022 | os.path.splitext(filename)[0] | |
2023 | if filename_real_ext == info_dict['ext'] | |
2024 | else filename) | |
81cd954a | 2025 | requested_formats = info_dict['requested_formats'] |
c0dea0a7 | 2026 | if self.params.get('merge_output_format') is None and not compatible_formats(requested_formats): |
38c6902b | 2027 | info_dict['ext'] = 'mkv' |
4a5a898a S |
2028 | self.report_warning( |
2029 | 'Requested formats are incompatible for merge and will be merged into mkv.') | |
38c6902b S |
2030 | # Ensure filename always has a correct extension for successful merge |
2031 | filename = '%s.%s' % (filename_wo_ext, info_dict['ext']) | |
5b5fbc08 JMF |
2032 | if os.path.exists(encodeFilename(filename)): |
2033 | self.to_screen( | |
2034 | '[download] %s has already been downloaded and ' | |
2035 | 'merged' % filename) | |
2036 | else: | |
81cd954a | 2037 | for f in requested_formats: |
5b5fbc08 JMF |
2038 | new_info = dict(info_dict) |
2039 | new_info.update(f) | |
c5c9bf0c S |
2040 | fname = prepend_extension( |
2041 | self.prepare_filename(new_info), | |
2042 | 'f%s' % f['format_id'], new_info['ext']) | |
2043 | if not ensure_dir_exists(fname): | |
2044 | return | |
5b5fbc08 JMF |
2045 | downloaded.append(fname) |
2046 | partial_success = dl(fname, new_info) | |
2047 | success = success and partial_success | |
2048 | info_dict['__postprocessors'] = postprocessors | |
2049 | info_dict['__files_to_merge'] = downloaded | |
4340deca P |
2050 | else: |
2051 | # Just a single file | |
2052 | success = dl(filename, info_dict) | |
2053 | except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: | |
7960b056 | 2054 | self.report_error('unable to download video data: %s' % error_to_compat_str(err)) |
4340deca P |
2055 | return |
2056 | except (OSError, IOError) as err: | |
2057 | raise UnavailableVideoError(err) | |
2058 | except (ContentTooShortError, ) as err: | |
2059 | self.report_error('content too short (expected %s bytes and served %s)' % (err.expected, err.downloaded)) | |
2060 | return | |
8222d8de | 2061 | |
e38cafe9 | 2062 | if success and filename != '-': |
6271f1ca | 2063 | # Fixup content |
62cd676c PH |
2064 | fixup_policy = self.params.get('fixup') |
2065 | if fixup_policy is None: | |
2066 | fixup_policy = 'detect_or_warn' | |
2067 | ||
d1e4a464 S |
2068 | INSTALL_FFMPEG_MESSAGE = 'Install ffmpeg or avconv to fix this automatically.' |
2069 | ||
6271f1ca PH |
2070 | stretched_ratio = info_dict.get('stretched_ratio') |
2071 | if stretched_ratio is not None and stretched_ratio != 1: | |
6271f1ca PH |
2072 | if fixup_policy == 'warn': |
2073 | self.report_warning('%s: Non-uniform pixel ratio (%s)' % ( | |
2074 | info_dict['id'], stretched_ratio)) | |
2075 | elif fixup_policy == 'detect_or_warn': | |
2076 | stretched_pp = FFmpegFixupStretchedPP(self) | |
2077 | if stretched_pp.available: | |
2078 | info_dict.setdefault('__postprocessors', []) | |
2079 | info_dict['__postprocessors'].append(stretched_pp) | |
2080 | else: | |
2081 | self.report_warning( | |
d1e4a464 S |
2082 | '%s: Non-uniform pixel ratio (%s). %s' |
2083 | % (info_dict['id'], stretched_ratio, INSTALL_FFMPEG_MESSAGE)) | |
6271f1ca | 2084 | else: |
62cd676c PH |
2085 | assert fixup_policy in ('ignore', 'never') |
2086 | ||
3089bc74 S |
2087 | if (info_dict.get('requested_formats') is None |
2088 | and info_dict.get('container') == 'm4a_dash'): | |
62cd676c | 2089 | if fixup_policy == 'warn': |
d1e4a464 S |
2090 | self.report_warning( |
2091 | '%s: writing DASH m4a. ' | |
2092 | 'Only some players support this container.' | |
2093 | % info_dict['id']) | |
62cd676c PH |
2094 | elif fixup_policy == 'detect_or_warn': |
2095 | fixup_pp = FFmpegFixupM4aPP(self) | |
2096 | if fixup_pp.available: | |
2097 | info_dict.setdefault('__postprocessors', []) | |
2098 | info_dict['__postprocessors'].append(fixup_pp) | |
2099 | else: | |
2100 | self.report_warning( | |
d1e4a464 S |
2101 | '%s: writing DASH m4a. ' |
2102 | 'Only some players support this container. %s' | |
2103 | % (info_dict['id'], INSTALL_FFMPEG_MESSAGE)) | |
62cd676c PH |
2104 | else: |
2105 | assert fixup_policy in ('ignore', 'never') | |
6271f1ca | 2106 | |
3089bc74 S |
2107 | if (info_dict.get('protocol') == 'm3u8_native' |
2108 | or info_dict.get('protocol') == 'm3u8' | |
2109 | and self.params.get('hls_prefer_native')): | |
f17f8651 | 2110 | if fixup_policy == 'warn': |
a02682fd | 2111 | self.report_warning('%s: malformed AAC bitstream detected.' % ( |
f17f8651 | 2112 | info_dict['id'])) |
2113 | elif fixup_policy == 'detect_or_warn': | |
2114 | fixup_pp = FFmpegFixupM3u8PP(self) | |
2115 | if fixup_pp.available: | |
2116 | info_dict.setdefault('__postprocessors', []) | |
2117 | info_dict['__postprocessors'].append(fixup_pp) | |
2118 | else: | |
2119 | self.report_warning( | |
a02682fd | 2120 | '%s: malformed AAC bitstream detected. %s' |
d1e4a464 | 2121 | % (info_dict['id'], INSTALL_FFMPEG_MESSAGE)) |
f17f8651 | 2122 | else: |
2123 | assert fixup_policy in ('ignore', 'never') | |
2124 | ||
8222d8de JMF |
2125 | try: |
2126 | self.post_process(filename, info_dict) | |
2127 | except (PostProcessingError) as err: | |
6febd1c1 | 2128 | self.report_error('postprocessing: %s' % str(err)) |
8222d8de | 2129 | return |
cd58dc3e | 2130 | self.record_download_archive(info_dict) |
8222d8de JMF |
2131 | |
2132 | def download(self, url_list): | |
2133 | """Download a given list of URLs.""" | |
acd69589 | 2134 | outtmpl = self.params.get('outtmpl', DEFAULT_OUTTMPL) |
3089bc74 S |
2135 | if (len(url_list) > 1 |
2136 | and outtmpl != '-' | |
2137 | and '%' not in outtmpl | |
2138 | and self.params.get('max_downloads') != 1): | |
acd69589 | 2139 | raise SameFileError(outtmpl) |
8222d8de JMF |
2140 | |
2141 | for url in url_list: | |
2142 | try: | |
5f6a1245 | 2143 | # It also downloads the videos |
61aa5ba3 S |
2144 | res = self.extract_info( |
2145 | url, force_generic_extractor=self.params.get('force_generic_extractor', False)) | |
8222d8de | 2146 | except UnavailableVideoError: |
6febd1c1 | 2147 | self.report_error('unable to download video') |
8222d8de | 2148 | except MaxDownloadsReached: |
6febd1c1 | 2149 | self.to_screen('[info] Maximum number of downloaded files reached.') |
8222d8de | 2150 | raise |
63e0be34 PH |
2151 | else: |
2152 | if self.params.get('dump_single_json', False): | |
2153 | self.to_stdout(json.dumps(res)) | |
8222d8de JMF |
2154 | |
2155 | return self._download_retcode | |
2156 | ||
1dcc4c0c | 2157 | def download_with_info_file(self, info_filename): |
31bd3925 JMF |
2158 | with contextlib.closing(fileinput.FileInput( |
2159 | [info_filename], mode='r', | |
2160 | openhook=fileinput.hook_encoded('utf-8'))) as f: | |
2161 | # FileInput doesn't have a read method, we can't call json.load | |
cb202fd2 | 2162 | info = self.filter_requested_info(json.loads('\n'.join(f))) |
d4943898 JMF |
2163 | try: |
2164 | self.process_ie_result(info, download=True) | |
2165 | except DownloadError: | |
2166 | webpage_url = info.get('webpage_url') | |
2167 | if webpage_url is not None: | |
6febd1c1 | 2168 | self.report_warning('The info failed to download, trying with "%s"' % webpage_url) |
d4943898 JMF |
2169 | return self.download([webpage_url]) |
2170 | else: | |
2171 | raise | |
2172 | return self._download_retcode | |
1dcc4c0c | 2173 | |
cb202fd2 S |
2174 | @staticmethod |
2175 | def filter_requested_info(info_dict): | |
2176 | return dict( | |
2177 | (k, v) for k, v in info_dict.items() | |
2178 | if k not in ['requested_formats', 'requested_subtitles']) | |
2179 | ||
8222d8de JMF |
2180 | def post_process(self, filename, ie_info): |
2181 | """Run all the postprocessors on the given file.""" | |
2182 | info = dict(ie_info) | |
2183 | info['filepath'] = filename | |
6350728b JMF |
2184 | pps_chain = [] |
2185 | if ie_info.get('__postprocessors') is not None: | |
2186 | pps_chain.extend(ie_info['__postprocessors']) | |
2187 | pps_chain.extend(self._pps) | |
2188 | for pp in pps_chain: | |
71646e46 | 2189 | files_to_delete = [] |
8222d8de | 2190 | try: |
592e97e8 | 2191 | files_to_delete, info = pp.run(info) |
8222d8de | 2192 | except PostProcessingError as e: |
bbcbf4d4 | 2193 | self.report_error(e.msg) |
592e97e8 | 2194 | if files_to_delete and not self.params.get('keepvideo', False): |
d03cfdce | 2195 | for old_filename in set(files_to_delete): |
f3ff1a36 | 2196 | self.to_screen('Deleting original file %s (pass -k to keep)' % old_filename) |
592e97e8 JMF |
2197 | try: |
2198 | os.remove(encodeFilename(old_filename)) | |
2199 | except (IOError, OSError): | |
2200 | self.report_warning('Unable to remove downloaded original file') | |
c1c9a79c | 2201 | |
5db07df6 | 2202 | def _make_archive_id(self, info_dict): |
e9fef7ee S |
2203 | video_id = info_dict.get('id') |
2204 | if not video_id: | |
2205 | return | |
5db07df6 PH |
2206 | # Future-proof against any change in case |
2207 | # and backwards compatibility with prior versions | |
e9fef7ee | 2208 | extractor = info_dict.get('extractor_key') or info_dict.get('ie_key') # key in a playlist |
7012b23c | 2209 | if extractor is None: |
1211bb6d S |
2210 | url = str_or_none(info_dict.get('url')) |
2211 | if not url: | |
2212 | return | |
e9fef7ee S |
2213 | # Try to find matching extractor for the URL and take its ie_key |
2214 | for ie in self._ies: | |
1211bb6d | 2215 | if ie.suitable(url): |
e9fef7ee S |
2216 | extractor = ie.ie_key() |
2217 | break | |
2218 | else: | |
2219 | return | |
2220 | return extractor.lower() + ' ' + video_id | |
5db07df6 PH |
2221 | |
2222 | def in_download_archive(self, info_dict): | |
2223 | fn = self.params.get('download_archive') | |
2224 | if fn is None: | |
2225 | return False | |
2226 | ||
2227 | vid_id = self._make_archive_id(info_dict) | |
e9fef7ee | 2228 | if not vid_id: |
7012b23c | 2229 | return False # Incomplete video information |
5db07df6 | 2230 | |
a45e8619 | 2231 | return vid_id in self.archive |
c1c9a79c PH |
2232 | |
2233 | def record_download_archive(self, info_dict): | |
2234 | fn = self.params.get('download_archive') | |
2235 | if fn is None: | |
2236 | return | |
5db07df6 PH |
2237 | vid_id = self._make_archive_id(info_dict) |
2238 | assert vid_id | |
c1c9a79c | 2239 | with locked_file(fn, 'a', encoding='utf-8') as archive_file: |
6febd1c1 | 2240 | archive_file.write(vid_id + '\n') |
a45e8619 | 2241 | self.archive.add(vid_id) |
dd82ffea | 2242 | |
8c51aa65 | 2243 | @staticmethod |
8abeeb94 | 2244 | def format_resolution(format, default='unknown'): |
fb04e403 PH |
2245 | if format.get('vcodec') == 'none': |
2246 | return 'audio only' | |
f49d89ee PH |
2247 | if format.get('resolution') is not None: |
2248 | return format['resolution'] | |
8c51aa65 JMF |
2249 | if format.get('height') is not None: |
2250 | if format.get('width') is not None: | |
6febd1c1 | 2251 | res = '%sx%s' % (format['width'], format['height']) |
8c51aa65 | 2252 | else: |
6febd1c1 | 2253 | res = '%sp' % format['height'] |
f49d89ee | 2254 | elif format.get('width') is not None: |
388ae76b | 2255 | res = '%dx?' % format['width'] |
8c51aa65 | 2256 | else: |
8abeeb94 | 2257 | res = default |
8c51aa65 JMF |
2258 | return res |
2259 | ||
c57f7757 PH |
2260 | def _format_note(self, fdict): |
2261 | res = '' | |
2262 | if fdict.get('ext') in ['f4f', 'f4m']: | |
2263 | res += '(unsupported) ' | |
32f90364 PH |
2264 | if fdict.get('language'): |
2265 | if res: | |
2266 | res += ' ' | |
9016d76f | 2267 | res += '[%s] ' % fdict['language'] |
c57f7757 PH |
2268 | if fdict.get('format_note') is not None: |
2269 | res += fdict['format_note'] + ' ' | |
2270 | if fdict.get('tbr') is not None: | |
2271 | res += '%4dk ' % fdict['tbr'] | |
2272 | if fdict.get('container') is not None: | |
2273 | if res: | |
2274 | res += ', ' | |
2275 | res += '%s container' % fdict['container'] | |
3089bc74 S |
2276 | if (fdict.get('vcodec') is not None |
2277 | and fdict.get('vcodec') != 'none'): | |
c57f7757 PH |
2278 | if res: |
2279 | res += ', ' | |
2280 | res += fdict['vcodec'] | |
91c7271a | 2281 | if fdict.get('vbr') is not None: |
c57f7757 PH |
2282 | res += '@' |
2283 | elif fdict.get('vbr') is not None and fdict.get('abr') is not None: | |
2284 | res += 'video@' | |
2285 | if fdict.get('vbr') is not None: | |
2286 | res += '%4dk' % fdict['vbr'] | |
fbb21cf5 | 2287 | if fdict.get('fps') is not None: |
5d583bdf S |
2288 | if res: |
2289 | res += ', ' | |
2290 | res += '%sfps' % fdict['fps'] | |
c57f7757 PH |
2291 | if fdict.get('acodec') is not None: |
2292 | if res: | |
2293 | res += ', ' | |
2294 | if fdict['acodec'] == 'none': | |
2295 | res += 'video only' | |
2296 | else: | |
2297 | res += '%-5s' % fdict['acodec'] | |
2298 | elif fdict.get('abr') is not None: | |
2299 | if res: | |
2300 | res += ', ' | |
2301 | res += 'audio' | |
2302 | if fdict.get('abr') is not None: | |
2303 | res += '@%3dk' % fdict['abr'] | |
2304 | if fdict.get('asr') is not None: | |
2305 | res += ' (%5dHz)' % fdict['asr'] | |
2306 | if fdict.get('filesize') is not None: | |
2307 | if res: | |
2308 | res += ', ' | |
2309 | res += format_bytes(fdict['filesize']) | |
9732d77e PH |
2310 | elif fdict.get('filesize_approx') is not None: |
2311 | if res: | |
2312 | res += ', ' | |
2313 | res += '~' + format_bytes(fdict['filesize_approx']) | |
c57f7757 | 2314 | return res |
91c7271a | 2315 | |
c57f7757 | 2316 | def list_formats(self, info_dict): |
94badb25 | 2317 | formats = info_dict.get('formats', [info_dict]) |
b81a359e PH |
2318 | table = [ |
2319 | [f['format_id'], f['ext'], self.format_resolution(f), self._format_note(f)] | |
2320 | for f in formats | |
e65566a9 | 2321 | if f.get('preference') is None or f['preference'] >= -1000] |
eb8a4433 | 2322 | # if len(formats) > 1: |
909d24dd | 2323 | # table[-1][-1] += (' ' if table[-1][-1] else '') + '(best*)' |
57dd9a8f | 2324 | |
b81a359e | 2325 | header_line = ['format code', 'extension', 'resolution', 'note'] |
cfb56d1a | 2326 | self.to_screen( |
b81a359e PH |
2327 | '[info] Available formats for %s:\n%s' % |
2328 | (info_dict['id'], render_table(header_line, table))) | |
cfb56d1a PH |
2329 | |
2330 | def list_thumbnails(self, info_dict): | |
2331 | thumbnails = info_dict.get('thumbnails') | |
2332 | if not thumbnails: | |
b7b72db9 | 2333 | self.to_screen('[info] No thumbnails present for %s' % info_dict['id']) |
2334 | return | |
cfb56d1a PH |
2335 | |
2336 | self.to_screen( | |
2337 | '[info] Thumbnails for %s:' % info_dict['id']) | |
2338 | self.to_screen(render_table( | |
2339 | ['ID', 'width', 'height', 'URL'], | |
2340 | [[t['id'], t.get('width', 'unknown'), t.get('height', 'unknown'), t['url']] for t in thumbnails])) | |
dca08720 | 2341 | |
360e1ca5 | 2342 | def list_subtitles(self, video_id, subtitles, name='subtitles'): |
a504ced0 | 2343 | if not subtitles: |
360e1ca5 | 2344 | self.to_screen('%s has no %s' % (video_id, name)) |
a504ced0 | 2345 | return |
a504ced0 | 2346 | self.to_screen( |
edab9dbf JMF |
2347 | 'Available %s for %s:' % (name, video_id)) |
2348 | self.to_screen(render_table( | |
2349 | ['Language', 'formats'], | |
2350 | [[lang, ', '.join(f['ext'] for f in reversed(formats))] | |
2351 | for lang, formats in subtitles.items()])) | |
a504ced0 | 2352 | |
dca08720 PH |
2353 | def urlopen(self, req): |
2354 | """ Start an HTTP download """ | |
82d8a8b6 | 2355 | if isinstance(req, compat_basestring): |
67dda517 | 2356 | req = sanitized_Request(req) |
19a41fc6 | 2357 | return self._opener.open(req, timeout=self._socket_timeout) |
dca08720 PH |
2358 | |
2359 | def print_debug_header(self): | |
2360 | if not self.params.get('verbose'): | |
2361 | return | |
62fec3b2 | 2362 | |
4192b51c | 2363 | if type('') is not compat_str: |
067aa17e | 2364 | # Python 2.6 on SLES11 SP1 (https://github.com/ytdl-org/youtube-dl/issues/3326) |
4192b51c PH |
2365 | self.report_warning( |
2366 | 'Your Python is broken! Update to a newer and supported version') | |
2367 | ||
c6afed48 PH |
2368 | stdout_encoding = getattr( |
2369 | sys.stdout, 'encoding', 'missing (%s)' % type(sys.stdout).__name__) | |
b0472057 | 2370 | encoding_str = ( |
734f90bb PH |
2371 | '[debug] Encodings: locale %s, fs %s, out %s, pref %s\n' % ( |
2372 | locale.getpreferredencoding(), | |
2373 | sys.getfilesystemencoding(), | |
c6afed48 | 2374 | stdout_encoding, |
b0472057 | 2375 | self.get_encoding())) |
4192b51c | 2376 | write_string(encoding_str, encoding=None) |
734f90bb | 2377 | |
cefecac1 | 2378 | self._write_string('[debug] youtube-dlc version ' + __version__ + '\n') |
e0986e31 JMF |
2379 | if _LAZY_LOADER: |
2380 | self._write_string('[debug] Lazy loading extractors enabled' + '\n') | |
dca08720 PH |
2381 | try: |
2382 | sp = subprocess.Popen( | |
2383 | ['git', 'rev-parse', '--short', 'HEAD'], | |
2384 | stdout=subprocess.PIPE, stderr=subprocess.PIPE, | |
2385 | cwd=os.path.dirname(os.path.abspath(__file__))) | |
2386 | out, err = sp.communicate() | |
2387 | out = out.decode().strip() | |
2388 | if re.match('[0-9a-f]+', out): | |
734f90bb | 2389 | self._write_string('[debug] Git HEAD: ' + out + '\n') |
70a1165b | 2390 | except Exception: |
dca08720 PH |
2391 | try: |
2392 | sys.exc_clear() | |
70a1165b | 2393 | except Exception: |
dca08720 | 2394 | pass |
b300cda4 S |
2395 | |
2396 | def python_implementation(): | |
2397 | impl_name = platform.python_implementation() | |
2398 | if impl_name == 'PyPy' and hasattr(sys, 'pypy_version_info'): | |
2399 | return impl_name + ' version %d.%d.%d' % sys.pypy_version_info[:3] | |
2400 | return impl_name | |
2401 | ||
2402 | self._write_string('[debug] Python version %s (%s) - %s\n' % ( | |
2403 | platform.python_version(), python_implementation(), | |
2404 | platform_name())) | |
d28b5171 | 2405 | |
73fac4e9 | 2406 | exe_versions = FFmpegPostProcessor.get_versions(self) |
4c83c967 | 2407 | exe_versions['rtmpdump'] = rtmpdump_version() |
feee8d32 | 2408 | exe_versions['phantomjs'] = PhantomJSwrapper._version() |
d28b5171 PH |
2409 | exe_str = ', '.join( |
2410 | '%s %s' % (exe, v) | |
2411 | for exe, v in sorted(exe_versions.items()) | |
2412 | if v | |
2413 | ) | |
2414 | if not exe_str: | |
2415 | exe_str = 'none' | |
2416 | self._write_string('[debug] exe versions: %s\n' % exe_str) | |
dca08720 PH |
2417 | |
2418 | proxy_map = {} | |
2419 | for handler in self._opener.handlers: | |
2420 | if hasattr(handler, 'proxies'): | |
2421 | proxy_map.update(handler.proxies) | |
734f90bb | 2422 | self._write_string('[debug] Proxy map: ' + compat_str(proxy_map) + '\n') |
dca08720 | 2423 | |
58b1f00d PH |
2424 | if self.params.get('call_home', False): |
2425 | ipaddr = self.urlopen('https://yt-dl.org/ip').read().decode('utf-8') | |
2426 | self._write_string('[debug] Public IP address: %s\n' % ipaddr) | |
2427 | latest_version = self.urlopen( | |
2428 | 'https://yt-dl.org/latest/version').read().decode('utf-8') | |
2429 | if version_tuple(latest_version) > version_tuple(__version__): | |
2430 | self.report_warning( | |
2431 | 'You are using an outdated version (newest version: %s)! ' | |
2432 | 'See https://yt-dl.org/update if you need help updating.' % | |
2433 | latest_version) | |
2434 | ||
e344693b | 2435 | def _setup_opener(self): |
6ad14cab | 2436 | timeout_val = self.params.get('socket_timeout') |
19a41fc6 | 2437 | self._socket_timeout = 600 if timeout_val is None else float(timeout_val) |
6ad14cab | 2438 | |
dca08720 PH |
2439 | opts_cookiefile = self.params.get('cookiefile') |
2440 | opts_proxy = self.params.get('proxy') | |
2441 | ||
2442 | if opts_cookiefile is None: | |
2443 | self.cookiejar = compat_cookiejar.CookieJar() | |
2444 | else: | |
590bc6f6 | 2445 | opts_cookiefile = expand_path(opts_cookiefile) |
1bab3437 | 2446 | self.cookiejar = YoutubeDLCookieJar(opts_cookiefile) |
dca08720 | 2447 | if os.access(opts_cookiefile, os.R_OK): |
1d88b3e6 | 2448 | self.cookiejar.load(ignore_discard=True, ignore_expires=True) |
dca08720 | 2449 | |
6a3f4c3f | 2450 | cookie_processor = YoutubeDLCookieProcessor(self.cookiejar) |
dca08720 PH |
2451 | if opts_proxy is not None: |
2452 | if opts_proxy == '': | |
2453 | proxies = {} | |
2454 | else: | |
2455 | proxies = {'http': opts_proxy, 'https': opts_proxy} | |
2456 | else: | |
2457 | proxies = compat_urllib_request.getproxies() | |
067aa17e | 2458 | # Set HTTPS proxy to HTTP one if given (https://github.com/ytdl-org/youtube-dl/issues/805) |
dca08720 PH |
2459 | if 'http' in proxies and 'https' not in proxies: |
2460 | proxies['https'] = proxies['http'] | |
91410c9b | 2461 | proxy_handler = PerRequestProxyHandler(proxies) |
a0ddb8a2 PH |
2462 | |
2463 | debuglevel = 1 if self.params.get('debug_printtraffic') else 0 | |
be4a824d PH |
2464 | https_handler = make_HTTPS_handler(self.params, debuglevel=debuglevel) |
2465 | ydlh = YoutubeDLHandler(self.params, debuglevel=debuglevel) | |
fca6dba8 | 2466 | redirect_handler = YoutubeDLRedirectHandler() |
8b172c2e | 2467 | data_handler = compat_urllib_request_DataHandler() |
6240b0a2 JMF |
2468 | |
2469 | # When passing our own FileHandler instance, build_opener won't add the | |
2470 | # default FileHandler and allows us to disable the file protocol, which | |
2471 | # can be used for malicious purposes (see | |
067aa17e | 2472 | # https://github.com/ytdl-org/youtube-dl/issues/8227) |
6240b0a2 JMF |
2473 | file_handler = compat_urllib_request.FileHandler() |
2474 | ||
2475 | def file_open(*args, **kwargs): | |
cefecac1 | 2476 | raise compat_urllib_error.URLError('file:// scheme is explicitly disabled in youtube-dlc for security reasons') |
6240b0a2 JMF |
2477 | file_handler.file_open = file_open |
2478 | ||
2479 | opener = compat_urllib_request.build_opener( | |
fca6dba8 | 2480 | proxy_handler, https_handler, cookie_processor, ydlh, redirect_handler, data_handler, file_handler) |
2461f79d | 2481 | |
dca08720 PH |
2482 | # Delete the default user-agent header, which would otherwise apply in |
2483 | # cases where our custom HTTP handler doesn't come into play | |
067aa17e | 2484 | # (See https://github.com/ytdl-org/youtube-dl/issues/1309 for details) |
dca08720 PH |
2485 | opener.addheaders = [] |
2486 | self._opener = opener | |
62fec3b2 PH |
2487 | |
2488 | def encode(self, s): | |
2489 | if isinstance(s, bytes): | |
2490 | return s # Already encoded | |
2491 | ||
2492 | try: | |
2493 | return s.encode(self.get_encoding()) | |
2494 | except UnicodeEncodeError as err: | |
2495 | err.reason = err.reason + '. Check your system encoding configuration or use the --encoding option.' | |
2496 | raise | |
2497 | ||
2498 | def get_encoding(self): | |
2499 | encoding = self.params.get('encoding') | |
2500 | if encoding is None: | |
2501 | encoding = preferredencoding() | |
2502 | return encoding | |
ec82d85a PH |
2503 | |
2504 | def _write_thumbnails(self, info_dict, filename): | |
2505 | if self.params.get('writethumbnail', False): | |
2506 | thumbnails = info_dict.get('thumbnails') | |
2507 | if thumbnails: | |
2508 | thumbnails = [thumbnails[-1]] | |
2509 | elif self.params.get('write_all_thumbnails', False): | |
2510 | thumbnails = info_dict.get('thumbnails') | |
2511 | else: | |
2512 | return | |
2513 | ||
2514 | if not thumbnails: | |
2515 | # No thumbnails present, so return immediately | |
2516 | return | |
2517 | ||
2518 | for t in thumbnails: | |
2519 | thumb_ext = determine_ext(t['url'], 'jpg') | |
2520 | suffix = '_%s' % t['id'] if len(thumbnails) > 1 else '' | |
2521 | thumb_display_id = '%s ' % t['id'] if len(thumbnails) > 1 else '' | |
82245a6d | 2522 | t['filename'] = thumb_filename = os.path.splitext(filename)[0] + suffix + '.' + thumb_ext |
ec82d85a PH |
2523 | |
2524 | if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(thumb_filename)): | |
2525 | self.to_screen('[%s] %s: Thumbnail %sis already present' % | |
2526 | (info_dict['extractor'], info_dict['id'], thumb_display_id)) | |
2527 | else: | |
2528 | self.to_screen('[%s] %s: Downloading thumbnail %s...' % | |
2529 | (info_dict['extractor'], info_dict['id'], thumb_display_id)) | |
2530 | try: | |
2531 | uf = self.urlopen(t['url']) | |
d3d89c32 | 2532 | with open(encodeFilename(thumb_filename), 'wb') as thumbf: |
ec82d85a PH |
2533 | shutil.copyfileobj(uf, thumbf) |
2534 | self.to_screen('[%s] %s: Writing thumbnail %sto: %s' % | |
2535 | (info_dict['extractor'], info_dict['id'], thumb_display_id, thumb_filename)) | |
2536 | except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: | |
2537 | self.report_warning('Unable to download thumbnail "%s": %s' % | |
9b9c5355 | 2538 | (t['url'], error_to_compat_str(err))) |