]> jfr.im git - yt-dlp.git/blob - youtube_dl/__init__.py
Bugfix: Allow colons in custom HTTP header values.
[yt-dlp.git] / youtube_dl / __init__.py
1 #!/usr/bin/env python
2 # -*- coding: utf-8 -*-
3
4 from __future__ import unicode_literals
5
6 __license__ = 'Public Domain'
7
8 import codecs
9 import io
10 import os
11 import random
12 import sys
13
14
15 from .options import (
16 parseOpts,
17 )
18 from .compat import (
19 compat_expanduser,
20 compat_getpass,
21 compat_print,
22 compat_shlex_split,
23 workaround_optparse_bug9161,
24 )
25 from .utils import (
26 DateRange,
27 decodeOption,
28 DEFAULT_OUTTMPL,
29 DownloadError,
30 match_filter_func,
31 MaxDownloadsReached,
32 preferredencoding,
33 read_batch_urls,
34 SameFileError,
35 setproctitle,
36 std_headers,
37 write_string,
38 )
39 from .update import update_self
40 from .downloader import (
41 FileDownloader,
42 )
43 from .extractor import gen_extractors, list_extractors
44 from .YoutubeDL import YoutubeDL
45
46
47 def _real_main(argv=None):
48 # Compatibility fixes for Windows
49 if sys.platform == 'win32':
50 # https://github.com/rg3/youtube-dl/issues/820
51 codecs.register(lambda name: codecs.lookup('utf-8') if name == 'cp65001' else None)
52
53 workaround_optparse_bug9161()
54
55 setproctitle('youtube-dl')
56
57 parser, opts, args = parseOpts(argv)
58
59 # Set user agent
60 if opts.user_agent is not None:
61 std_headers['User-Agent'] = opts.user_agent
62
63 # Set referer
64 if opts.referer is not None:
65 std_headers['Referer'] = opts.referer
66
67 # Custom HTTP headers
68 if opts.headers is not None:
69 for h in opts.headers:
70 if h.find(':', 1) < 0:
71 parser.error('wrong header formatting, it should be key:value, not "%s"' % h)
72 key, value = h.split(':', 1)
73 if opts.verbose:
74 write_string('[debug] Adding header from command line option %s:%s\n' % (key, value))
75 std_headers[key] = value
76
77 # Dump user agent
78 if opts.dump_user_agent:
79 compat_print(std_headers['User-Agent'])
80 sys.exit(0)
81
82 # Batch file verification
83 batch_urls = []
84 if opts.batchfile is not None:
85 try:
86 if opts.batchfile == '-':
87 batchfd = sys.stdin
88 else:
89 batchfd = io.open(
90 compat_expanduser(opts.batchfile),
91 'r', encoding='utf-8', errors='ignore')
92 batch_urls = read_batch_urls(batchfd)
93 if opts.verbose:
94 write_string('[debug] Batch file urls: ' + repr(batch_urls) + '\n')
95 except IOError:
96 sys.exit('ERROR: batch file could not be read')
97 all_urls = batch_urls + args
98 all_urls = [url.strip() for url in all_urls]
99 _enc = preferredencoding()
100 all_urls = [url.decode(_enc, 'ignore') if isinstance(url, bytes) else url for url in all_urls]
101
102 if opts.list_extractors:
103 for ie in list_extractors(opts.age_limit):
104 compat_print(ie.IE_NAME + (' (CURRENTLY BROKEN)' if not ie._WORKING else ''))
105 matchedUrls = [url for url in all_urls if ie.suitable(url)]
106 for mu in matchedUrls:
107 compat_print(' ' + mu)
108 sys.exit(0)
109 if opts.list_extractor_descriptions:
110 for ie in list_extractors(opts.age_limit):
111 if not ie._WORKING:
112 continue
113 desc = getattr(ie, 'IE_DESC', ie.IE_NAME)
114 if desc is False:
115 continue
116 if hasattr(ie, 'SEARCH_KEY'):
117 _SEARCHES = ('cute kittens', 'slithering pythons', 'falling cat', 'angry poodle', 'purple fish', 'running tortoise', 'sleeping bunny', 'burping cow')
118 _COUNTS = ('', '5', '10', 'all')
119 desc += ' (Example: "%s%s:%s" )' % (ie.SEARCH_KEY, random.choice(_COUNTS), random.choice(_SEARCHES))
120 compat_print(desc)
121 sys.exit(0)
122
123 # Conflicting, missing and erroneous options
124 if opts.usenetrc and (opts.username is not None or opts.password is not None):
125 parser.error('using .netrc conflicts with giving username/password')
126 if opts.password is not None and opts.username is None:
127 parser.error('account username missing\n')
128 if opts.outtmpl is not None and (opts.usetitle or opts.autonumber or opts.useid):
129 parser.error('using output template conflicts with using title, video ID or auto number')
130 if opts.usetitle and opts.useid:
131 parser.error('using title conflicts with using video ID')
132 if opts.username is not None and opts.password is None:
133 opts.password = compat_getpass('Type account password and press [Return]: ')
134 if opts.ratelimit is not None:
135 numeric_limit = FileDownloader.parse_bytes(opts.ratelimit)
136 if numeric_limit is None:
137 parser.error('invalid rate limit specified')
138 opts.ratelimit = numeric_limit
139 if opts.min_filesize is not None:
140 numeric_limit = FileDownloader.parse_bytes(opts.min_filesize)
141 if numeric_limit is None:
142 parser.error('invalid min_filesize specified')
143 opts.min_filesize = numeric_limit
144 if opts.max_filesize is not None:
145 numeric_limit = FileDownloader.parse_bytes(opts.max_filesize)
146 if numeric_limit is None:
147 parser.error('invalid max_filesize specified')
148 opts.max_filesize = numeric_limit
149
150 def parse_retries(retries):
151 if retries in ('inf', 'infinite'):
152 parsed_retries = float('inf')
153 else:
154 try:
155 parsed_retries = int(retries)
156 except (TypeError, ValueError):
157 parser.error('invalid retry count specified')
158 return parsed_retries
159 if opts.retries is not None:
160 opts.retries = parse_retries(opts.retries)
161 if opts.fragment_retries is not None:
162 opts.fragment_retries = parse_retries(opts.fragment_retries)
163 if opts.buffersize is not None:
164 numeric_buffersize = FileDownloader.parse_bytes(opts.buffersize)
165 if numeric_buffersize is None:
166 parser.error('invalid buffer size specified')
167 opts.buffersize = numeric_buffersize
168 if opts.playliststart <= 0:
169 raise ValueError('Playlist start must be positive')
170 if opts.playlistend not in (-1, None) and opts.playlistend < opts.playliststart:
171 raise ValueError('Playlist end must be greater than playlist start')
172 if opts.extractaudio:
173 if opts.audioformat not in ['best', 'aac', 'mp3', 'm4a', 'opus', 'vorbis', 'wav']:
174 parser.error('invalid audio format specified')
175 if opts.audioquality:
176 opts.audioquality = opts.audioquality.strip('k').strip('K')
177 if not opts.audioquality.isdigit():
178 parser.error('invalid audio quality specified')
179 if opts.recodevideo is not None:
180 if opts.recodevideo not in ['mp4', 'flv', 'webm', 'ogg', 'mkv', 'avi']:
181 parser.error('invalid video recode format specified')
182 if opts.convertsubtitles is not None:
183 if opts.convertsubtitles not in ['srt', 'vtt', 'ass']:
184 parser.error('invalid subtitle format specified')
185
186 if opts.date is not None:
187 date = DateRange.day(opts.date)
188 else:
189 date = DateRange(opts.dateafter, opts.datebefore)
190
191 # Do not download videos when there are audio-only formats
192 if opts.extractaudio and not opts.keepvideo and opts.format is None:
193 opts.format = 'bestaudio/best'
194
195 # --all-sub automatically sets --write-sub if --write-auto-sub is not given
196 # this was the old behaviour if only --all-sub was given.
197 if opts.allsubtitles and not opts.writeautomaticsub:
198 opts.writesubtitles = True
199
200 outtmpl = ((opts.outtmpl is not None and opts.outtmpl) or
201 (opts.format == '-1' and opts.usetitle and '%(title)s-%(id)s-%(format)s.%(ext)s') or
202 (opts.format == '-1' and '%(id)s-%(format)s.%(ext)s') or
203 (opts.usetitle and opts.autonumber and '%(autonumber)s-%(title)s-%(id)s.%(ext)s') or
204 (opts.usetitle and '%(title)s-%(id)s.%(ext)s') or
205 (opts.useid and '%(id)s.%(ext)s') or
206 (opts.autonumber and '%(autonumber)s-%(id)s.%(ext)s') or
207 DEFAULT_OUTTMPL)
208 if not os.path.splitext(outtmpl)[1] and opts.extractaudio:
209 parser.error('Cannot download a video and extract audio into the same'
210 ' file! Use "{0}.%(ext)s" instead of "{0}" as the output'
211 ' template'.format(outtmpl))
212
213 any_getting = opts.geturl or opts.gettitle or opts.getid or opts.getthumbnail or opts.getdescription or opts.getfilename or opts.getformat or opts.getduration or opts.dumpjson or opts.dump_single_json
214 any_printing = opts.print_json
215 download_archive_fn = compat_expanduser(opts.download_archive) if opts.download_archive is not None else opts.download_archive
216
217 # PostProcessors
218 postprocessors = []
219 # Add the metadata pp first, the other pps will copy it
220 if opts.metafromtitle:
221 postprocessors.append({
222 'key': 'MetadataFromTitle',
223 'titleformat': opts.metafromtitle
224 })
225 if opts.addmetadata:
226 postprocessors.append({'key': 'FFmpegMetadata'})
227 if opts.extractaudio:
228 postprocessors.append({
229 'key': 'FFmpegExtractAudio',
230 'preferredcodec': opts.audioformat,
231 'preferredquality': opts.audioquality,
232 'nopostoverwrites': opts.nopostoverwrites,
233 })
234 if opts.recodevideo:
235 postprocessors.append({
236 'key': 'FFmpegVideoConvertor',
237 'preferedformat': opts.recodevideo,
238 })
239 if opts.convertsubtitles:
240 postprocessors.append({
241 'key': 'FFmpegSubtitlesConvertor',
242 'format': opts.convertsubtitles,
243 })
244 if opts.embedsubtitles:
245 postprocessors.append({
246 'key': 'FFmpegEmbedSubtitle',
247 })
248 if opts.xattrs:
249 postprocessors.append({'key': 'XAttrMetadata'})
250 if opts.embedthumbnail:
251 already_have_thumbnail = opts.writethumbnail or opts.write_all_thumbnails
252 postprocessors.append({
253 'key': 'EmbedThumbnail',
254 'already_have_thumbnail': already_have_thumbnail
255 })
256 if not already_have_thumbnail:
257 opts.writethumbnail = True
258 # Please keep ExecAfterDownload towards the bottom as it allows the user to modify the final file in any way.
259 # So if the user is able to remove the file before your postprocessor runs it might cause a few problems.
260 if opts.exec_cmd:
261 postprocessors.append({
262 'key': 'ExecAfterDownload',
263 'exec_cmd': opts.exec_cmd,
264 })
265 if opts.xattr_set_filesize:
266 try:
267 import xattr
268 xattr # Confuse flake8
269 except ImportError:
270 parser.error('setting filesize xattr requested but python-xattr is not available')
271 external_downloader_args = None
272 if opts.external_downloader_args:
273 external_downloader_args = compat_shlex_split(opts.external_downloader_args)
274 postprocessor_args = None
275 if opts.postprocessor_args:
276 postprocessor_args = compat_shlex_split(opts.postprocessor_args)
277 match_filter = (
278 None if opts.match_filter is None
279 else match_filter_func(opts.match_filter))
280
281 ydl_opts = {
282 'usenetrc': opts.usenetrc,
283 'username': opts.username,
284 'password': opts.password,
285 'twofactor': opts.twofactor,
286 'videopassword': opts.videopassword,
287 'quiet': (opts.quiet or any_getting or any_printing),
288 'no_warnings': opts.no_warnings,
289 'forceurl': opts.geturl,
290 'forcetitle': opts.gettitle,
291 'forceid': opts.getid,
292 'forcethumbnail': opts.getthumbnail,
293 'forcedescription': opts.getdescription,
294 'forceduration': opts.getduration,
295 'forcefilename': opts.getfilename,
296 'forceformat': opts.getformat,
297 'forcejson': opts.dumpjson or opts.print_json,
298 'dump_single_json': opts.dump_single_json,
299 'simulate': opts.simulate or any_getting,
300 'skip_download': opts.skip_download,
301 'format': opts.format,
302 'listformats': opts.listformats,
303 'outtmpl': outtmpl,
304 'autonumber_size': opts.autonumber_size,
305 'restrictfilenames': opts.restrictfilenames,
306 'ignoreerrors': opts.ignoreerrors,
307 'force_generic_extractor': opts.force_generic_extractor,
308 'ratelimit': opts.ratelimit,
309 'nooverwrites': opts.nooverwrites,
310 'retries': opts.retries,
311 'fragment_retries': opts.fragment_retries,
312 'buffersize': opts.buffersize,
313 'noresizebuffer': opts.noresizebuffer,
314 'continuedl': opts.continue_dl,
315 'noprogress': opts.noprogress,
316 'progress_with_newline': opts.progress_with_newline,
317 'playliststart': opts.playliststart,
318 'playlistend': opts.playlistend,
319 'playlistreverse': opts.playlist_reverse,
320 'noplaylist': opts.noplaylist,
321 'logtostderr': opts.outtmpl == '-',
322 'consoletitle': opts.consoletitle,
323 'nopart': opts.nopart,
324 'updatetime': opts.updatetime,
325 'writedescription': opts.writedescription,
326 'writeannotations': opts.writeannotations,
327 'writeinfojson': opts.writeinfojson,
328 'writethumbnail': opts.writethumbnail,
329 'write_all_thumbnails': opts.write_all_thumbnails,
330 'writesubtitles': opts.writesubtitles,
331 'writeautomaticsub': opts.writeautomaticsub,
332 'allsubtitles': opts.allsubtitles,
333 'listsubtitles': opts.listsubtitles,
334 'subtitlesformat': opts.subtitlesformat,
335 'subtitleslangs': opts.subtitleslangs,
336 'matchtitle': decodeOption(opts.matchtitle),
337 'rejecttitle': decodeOption(opts.rejecttitle),
338 'max_downloads': opts.max_downloads,
339 'prefer_free_formats': opts.prefer_free_formats,
340 'verbose': opts.verbose,
341 'dump_intermediate_pages': opts.dump_intermediate_pages,
342 'write_pages': opts.write_pages,
343 'test': opts.test,
344 'keepvideo': opts.keepvideo,
345 'min_filesize': opts.min_filesize,
346 'max_filesize': opts.max_filesize,
347 'min_views': opts.min_views,
348 'max_views': opts.max_views,
349 'daterange': date,
350 'cachedir': opts.cachedir,
351 'youtube_print_sig_code': opts.youtube_print_sig_code,
352 'age_limit': opts.age_limit,
353 'download_archive': download_archive_fn,
354 'cookiefile': opts.cookiefile,
355 'nocheckcertificate': opts.no_check_certificate,
356 'prefer_insecure': opts.prefer_insecure,
357 'proxy': opts.proxy,
358 'socket_timeout': opts.socket_timeout,
359 'bidi_workaround': opts.bidi_workaround,
360 'debug_printtraffic': opts.debug_printtraffic,
361 'prefer_ffmpeg': opts.prefer_ffmpeg,
362 'include_ads': opts.include_ads,
363 'default_search': opts.default_search,
364 'youtube_include_dash_manifest': opts.youtube_include_dash_manifest,
365 'encoding': opts.encoding,
366 'extract_flat': opts.extract_flat,
367 'mark_watched': opts.mark_watched,
368 'merge_output_format': opts.merge_output_format,
369 'postprocessors': postprocessors,
370 'fixup': opts.fixup,
371 'source_address': opts.source_address,
372 'call_home': opts.call_home,
373 'sleep_interval': opts.sleep_interval,
374 'external_downloader': opts.external_downloader,
375 'list_thumbnails': opts.list_thumbnails,
376 'playlist_items': opts.playlist_items,
377 'xattr_set_filesize': opts.xattr_set_filesize,
378 'match_filter': match_filter,
379 'no_color': opts.no_color,
380 'ffmpeg_location': opts.ffmpeg_location,
381 'hls_prefer_native': opts.hls_prefer_native,
382 'hls_use_mpegts': opts.hls_use_mpegts,
383 'external_downloader_args': external_downloader_args,
384 'postprocessor_args': postprocessor_args,
385 'cn_verification_proxy': opts.cn_verification_proxy,
386 }
387
388 with YoutubeDL(ydl_opts) as ydl:
389 # Update version
390 if opts.update_self:
391 update_self(ydl.to_screen, opts.verbose, ydl._opener)
392
393 # Remove cache dir
394 if opts.rm_cachedir:
395 ydl.cache.remove()
396
397 # Maybe do nothing
398 if (len(all_urls) < 1) and (opts.load_info_filename is None):
399 if opts.update_self or opts.rm_cachedir:
400 sys.exit()
401
402 ydl.warn_if_short_id(sys.argv[1:] if argv is None else argv)
403 parser.error(
404 'You must provide at least one URL.\n'
405 'Type youtube-dl --help to see a list of all options.')
406
407 try:
408 if opts.load_info_filename is not None:
409 retcode = ydl.download_with_info_file(compat_expanduser(opts.load_info_filename))
410 else:
411 retcode = ydl.download(all_urls)
412 except MaxDownloadsReached:
413 ydl.to_screen('--max-download limit reached, aborting.')
414 retcode = 101
415
416 sys.exit(retcode)
417
418
419 def main(argv=None):
420 try:
421 _real_main(argv)
422 except DownloadError:
423 sys.exit(1)
424 except SameFileError:
425 sys.exit('ERROR: fixed output name but more than one file to download')
426 except KeyboardInterrupt:
427 sys.exit('\nERROR: Interrupted by user')
428
429 __all__ = ['main', 'YoutubeDL', 'gen_extractors', 'list_extractors']