]>
Commit | Line | Data |
---|---|---|
6929b41a | 1 | import asyncio |
15dfb392 | 2 | import atexit |
1e399778 | 3 | import base64 |
5bc880b9 | 4 | import binascii |
912b38b4 | 5 | import calendar |
676eb3f2 | 6 | import codecs |
c380cc28 | 7 | import collections |
ab029d7e | 8 | import collections.abc |
62e609ab | 9 | import contextlib |
c496ca96 | 10 | import datetime |
0c265486 | 11 | import email.header |
f8271158 | 12 | import email.utils |
f45c185f | 13 | import errno |
d77c3dfd | 14 | import gzip |
49fa4d9a N |
15 | import hashlib |
16 | import hmac | |
ac668111 | 17 | import html.entities |
18 | import html.parser | |
54007a45 | 19 | import http.client |
20 | import http.cookiejar | |
019a94f7 | 21 | import importlib.util |
b1f94422 | 22 | import inspect |
03f9daab | 23 | import io |
79a2e94e | 24 | import itertools |
f4bfd65f | 25 | import json |
d77c3dfd | 26 | import locale |
02dbf93f | 27 | import math |
f8271158 | 28 | import mimetypes |
347de493 | 29 | import operator |
d77c3dfd | 30 | import os |
c496ca96 | 31 | import platform |
773f291d | 32 | import random |
d77c3dfd | 33 | import re |
f8271158 | 34 | import shlex |
c496ca96 | 35 | import socket |
79a2e94e | 36 | import ssl |
ac668111 | 37 | import struct |
1c088fa8 | 38 | import subprocess |
d77c3dfd | 39 | import sys |
181c8655 | 40 | import tempfile |
c380cc28 | 41 | import time |
01951dda | 42 | import traceback |
64fa820c | 43 | import types |
989a01c2 | 44 | import unicodedata |
14f25df2 | 45 | import urllib.error |
f8271158 | 46 | import urllib.parse |
ac668111 | 47 | import urllib.request |
bcf89ce6 | 48 | import xml.etree.ElementTree |
d77c3dfd | 49 | import zlib |
d77c3dfd | 50 | |
6929b41a | 51 | from .compat import functools # isort: split |
8c25f81b | 52 | from .compat import ( |
36e6f62c | 53 | compat_etree_fromstring, |
51098426 | 54 | compat_expanduser, |
f8271158 | 55 | compat_HTMLParseError, |
efa97bdc | 56 | compat_os_name, |
702ccf2d | 57 | compat_shlex_quote, |
8c25f81b | 58 | ) |
ac668111 | 59 | from .dependencies import brotli, certifi, websockets, xattr |
f8271158 | 60 | from .socks import ProxyType, sockssocket |
71aff188 | 61 | |
4644ac55 | 62 | |
51fb4995 YCH |
63 | def register_socks_protocols(): |
64 | # "Register" SOCKS protocols | |
d5ae6bb5 YCH |
65 | # In Python < 2.6.5, urlsplit() suffers from bug https://bugs.python.org/issue7904 |
66 | # URLs with protocols not in urlparse.uses_netloc are not handled correctly | |
51fb4995 | 67 | for scheme in ('socks', 'socks4', 'socks4a', 'socks5'): |
14f25df2 | 68 | if scheme not in urllib.parse.uses_netloc: |
69 | urllib.parse.uses_netloc.append(scheme) | |
51fb4995 YCH |
70 | |
71 | ||
468e2e92 FV |
72 | # This is not clearly defined otherwise |
73 | compiled_regex_type = type(re.compile('')) | |
74 | ||
f7a147e3 S |
75 | |
76 | def random_user_agent(): | |
77 | _USER_AGENT_TPL = 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/%s Safari/537.36' | |
78 | _CHROME_VERSIONS = ( | |
19b4c74d | 79 | '90.0.4430.212', |
80 | '90.0.4430.24', | |
81 | '90.0.4430.70', | |
82 | '90.0.4430.72', | |
83 | '90.0.4430.85', | |
84 | '90.0.4430.93', | |
85 | '91.0.4472.101', | |
86 | '91.0.4472.106', | |
87 | '91.0.4472.114', | |
88 | '91.0.4472.124', | |
89 | '91.0.4472.164', | |
90 | '91.0.4472.19', | |
91 | '91.0.4472.77', | |
92 | '92.0.4515.107', | |
93 | '92.0.4515.115', | |
94 | '92.0.4515.131', | |
95 | '92.0.4515.159', | |
96 | '92.0.4515.43', | |
97 | '93.0.4556.0', | |
98 | '93.0.4577.15', | |
99 | '93.0.4577.63', | |
100 | '93.0.4577.82', | |
101 | '94.0.4606.41', | |
102 | '94.0.4606.54', | |
103 | '94.0.4606.61', | |
104 | '94.0.4606.71', | |
105 | '94.0.4606.81', | |
106 | '94.0.4606.85', | |
107 | '95.0.4638.17', | |
108 | '95.0.4638.50', | |
109 | '95.0.4638.54', | |
110 | '95.0.4638.69', | |
111 | '95.0.4638.74', | |
112 | '96.0.4664.18', | |
113 | '96.0.4664.45', | |
114 | '96.0.4664.55', | |
115 | '96.0.4664.93', | |
116 | '97.0.4692.20', | |
f7a147e3 S |
117 | ) |
118 | return _USER_AGENT_TPL % random.choice(_CHROME_VERSIONS) | |
119 | ||
120 | ||
4390d5ec | 121 | SUPPORTED_ENCODINGS = [ |
122 | 'gzip', 'deflate' | |
123 | ] | |
9b8ee23b | 124 | if brotli: |
4390d5ec | 125 | SUPPORTED_ENCODINGS.append('br') |
126 | ||
3e669f36 | 127 | std_headers = { |
f7a147e3 | 128 | 'User-Agent': random_user_agent(), |
59ae15a5 | 129 | 'Accept': 'text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8', |
59ae15a5 | 130 | 'Accept-Language': 'en-us,en;q=0.5', |
b1156c1e | 131 | 'Sec-Fetch-Mode': 'navigate', |
3e669f36 | 132 | } |
f427df17 | 133 | |
5f6a1245 | 134 | |
fb37eb25 S |
135 | USER_AGENTS = { |
136 | 'Safari': 'Mozilla/5.0 (X11; Linux x86_64; rv:10.0) AppleWebKit/533.20.25 (KHTML, like Gecko) Version/5.0.4 Safari/533.20.27', | |
137 | } | |
138 | ||
139 | ||
bf42a990 | 140 | NO_DEFAULT = object() |
7b2c3f47 | 141 | IDENTITY = lambda x: x |
bf42a990 | 142 | |
7105440c YCH |
143 | ENGLISH_MONTH_NAMES = [ |
144 | 'January', 'February', 'March', 'April', 'May', 'June', | |
145 | 'July', 'August', 'September', 'October', 'November', 'December'] | |
146 | ||
f6717dec S |
147 | MONTH_NAMES = { |
148 | 'en': ENGLISH_MONTH_NAMES, | |
149 | 'fr': [ | |
3e4185c3 S |
150 | 'janvier', 'février', 'mars', 'avril', 'mai', 'juin', |
151 | 'juillet', 'août', 'septembre', 'octobre', 'novembre', 'décembre'], | |
f6717dec | 152 | } |
a942d6cb | 153 | |
8f53dc44 | 154 | # From https://github.com/python/cpython/blob/3.11/Lib/email/_parseaddr.py#L36-L42 |
155 | TIMEZONE_NAMES = { | |
156 | 'UT': 0, 'UTC': 0, 'GMT': 0, 'Z': 0, | |
157 | 'AST': -4, 'ADT': -3, # Atlantic (used in Canada) | |
158 | 'EST': -5, 'EDT': -4, # Eastern | |
159 | 'CST': -6, 'CDT': -5, # Central | |
160 | 'MST': -7, 'MDT': -6, # Mountain | |
161 | 'PST': -8, 'PDT': -7 # Pacific | |
162 | } | |
163 | ||
c587cbb7 | 164 | # needed for sanitizing filenames in restricted mode |
c8827027 | 165 | ACCENT_CHARS = dict(zip('ÂÃÄÀÁÅÆÇÈÉÊËÌÍÎÏÐÑÒÓÔÕÖŐØŒÙÚÛÜŰÝÞßàáâãäåæçèéêëìíîïðñòóôõöőøœùúûüűýþÿ', |
fd35d8cd JW |
166 | itertools.chain('AAAAAA', ['AE'], 'CEEEEIIIIDNOOOOOOO', ['OE'], 'UUUUUY', ['TH', 'ss'], |
167 | 'aaaaaa', ['ae'], 'ceeeeiiiionooooooo', ['oe'], 'uuuuuy', ['th'], 'y'))) | |
c587cbb7 | 168 | |
46f59e89 S |
169 | DATE_FORMATS = ( |
170 | '%d %B %Y', | |
171 | '%d %b %Y', | |
172 | '%B %d %Y', | |
cb655f34 S |
173 | '%B %dst %Y', |
174 | '%B %dnd %Y', | |
9d30c213 | 175 | '%B %drd %Y', |
cb655f34 | 176 | '%B %dth %Y', |
46f59e89 | 177 | '%b %d %Y', |
cb655f34 S |
178 | '%b %dst %Y', |
179 | '%b %dnd %Y', | |
9d30c213 | 180 | '%b %drd %Y', |
cb655f34 | 181 | '%b %dth %Y', |
46f59e89 S |
182 | '%b %dst %Y %I:%M', |
183 | '%b %dnd %Y %I:%M', | |
9d30c213 | 184 | '%b %drd %Y %I:%M', |
46f59e89 S |
185 | '%b %dth %Y %I:%M', |
186 | '%Y %m %d', | |
187 | '%Y-%m-%d', | |
bccdbd22 | 188 | '%Y.%m.%d.', |
46f59e89 | 189 | '%Y/%m/%d', |
81c13222 | 190 | '%Y/%m/%d %H:%M', |
46f59e89 | 191 | '%Y/%m/%d %H:%M:%S', |
1931a55e THD |
192 | '%Y%m%d%H%M', |
193 | '%Y%m%d%H%M%S', | |
4f3fa23e | 194 | '%Y%m%d', |
0c1c6f4b | 195 | '%Y-%m-%d %H:%M', |
46f59e89 S |
196 | '%Y-%m-%d %H:%M:%S', |
197 | '%Y-%m-%d %H:%M:%S.%f', | |
5014558a | 198 | '%Y-%m-%d %H:%M:%S:%f', |
46f59e89 S |
199 | '%d.%m.%Y %H:%M', |
200 | '%d.%m.%Y %H.%M', | |
201 | '%Y-%m-%dT%H:%M:%SZ', | |
202 | '%Y-%m-%dT%H:%M:%S.%fZ', | |
203 | '%Y-%m-%dT%H:%M:%S.%f0Z', | |
204 | '%Y-%m-%dT%H:%M:%S', | |
205 | '%Y-%m-%dT%H:%M:%S.%f', | |
206 | '%Y-%m-%dT%H:%M', | |
c6eed6b8 S |
207 | '%b %d %Y at %H:%M', |
208 | '%b %d %Y at %H:%M:%S', | |
b555ae9b S |
209 | '%B %d %Y at %H:%M', |
210 | '%B %d %Y at %H:%M:%S', | |
a63d9bd0 | 211 | '%H:%M %d-%b-%Y', |
46f59e89 S |
212 | ) |
213 | ||
214 | DATE_FORMATS_DAY_FIRST = list(DATE_FORMATS) | |
215 | DATE_FORMATS_DAY_FIRST.extend([ | |
216 | '%d-%m-%Y', | |
217 | '%d.%m.%Y', | |
218 | '%d.%m.%y', | |
219 | '%d/%m/%Y', | |
220 | '%d/%m/%y', | |
221 | '%d/%m/%Y %H:%M:%S', | |
47304e07 | 222 | '%d-%m-%Y %H:%M', |
46f59e89 S |
223 | ]) |
224 | ||
225 | DATE_FORMATS_MONTH_FIRST = list(DATE_FORMATS) | |
226 | DATE_FORMATS_MONTH_FIRST.extend([ | |
227 | '%m-%d-%Y', | |
228 | '%m.%d.%Y', | |
229 | '%m/%d/%Y', | |
230 | '%m/%d/%y', | |
231 | '%m/%d/%Y %H:%M:%S', | |
232 | ]) | |
233 | ||
06b3fe29 | 234 | PACKED_CODES_RE = r"}\('(.+)',(\d+),(\d+),'([^']+)'\.split\('\|'\)" |
0f60ba6e | 235 | JSON_LD_RE = r'(?is)<script[^>]+type=(["\']?)application/ld\+json\1[^>]*>\s*(?P<json_ld>{.+?}|\[.+?\])\s*</script>' |
06b3fe29 | 236 | |
1d485a1a | 237 | NUMBER_RE = r'\d+(?:\.\d+)?' |
238 | ||
7105440c | 239 | |
0b9c08b4 | 240 | @functools.cache |
d77c3dfd | 241 | def preferredencoding(): |
59ae15a5 | 242 | """Get preferred encoding. |
d77c3dfd | 243 | |
59ae15a5 PH |
244 | Returns the best encoding scheme for the system, based on |
245 | locale.getpreferredencoding() and some further tweaks. | |
246 | """ | |
247 | try: | |
248 | pref = locale.getpreferredencoding() | |
28e614de | 249 | 'TEST'.encode(pref) |
70a1165b | 250 | except Exception: |
59ae15a5 | 251 | pref = 'UTF-8' |
bae611f2 | 252 | |
59ae15a5 | 253 | return pref |
d77c3dfd | 254 | |
f4bfd65f | 255 | |
181c8655 | 256 | def write_json_file(obj, fn): |
1394646a | 257 | """ Encode obj as JSON and write it to fn, atomically if possible """ |
181c8655 | 258 | |
cfb0511d | 259 | tf = tempfile.NamedTemporaryFile( |
260 | prefix=f'{os.path.basename(fn)}.', dir=os.path.dirname(fn), | |
261 | suffix='.tmp', delete=False, mode='w', encoding='utf-8') | |
181c8655 PH |
262 | |
263 | try: | |
264 | with tf: | |
45d86abe | 265 | json.dump(obj, tf, ensure_ascii=False) |
1394646a IK |
266 | if sys.platform == 'win32': |
267 | # Need to remove existing file on Windows, else os.rename raises | |
268 | # WindowsError or FileExistsError. | |
19a03940 | 269 | with contextlib.suppress(OSError): |
1394646a | 270 | os.unlink(fn) |
19a03940 | 271 | with contextlib.suppress(OSError): |
9cd5f54e R |
272 | mask = os.umask(0) |
273 | os.umask(mask) | |
274 | os.chmod(tf.name, 0o666 & ~mask) | |
181c8655 | 275 | os.rename(tf.name, fn) |
70a1165b | 276 | except Exception: |
19a03940 | 277 | with contextlib.suppress(OSError): |
181c8655 | 278 | os.remove(tf.name) |
181c8655 PH |
279 | raise |
280 | ||
281 | ||
cfb0511d | 282 | def find_xpath_attr(node, xpath, key, val=None): |
283 | """ Find the xpath xpath[@key=val] """ | |
284 | assert re.match(r'^[a-zA-Z_-]+$', key) | |
86e5f3ed | 285 | expr = xpath + ('[@%s]' % key if val is None else f"[@{key}='{val}']") |
cfb0511d | 286 | return node.find(expr) |
59ae56fa | 287 | |
d7e66d39 JMF |
288 | # On python2.6 the xml.etree.ElementTree.Element methods don't support |
289 | # the namespace parameter | |
5f6a1245 JW |
290 | |
291 | ||
d7e66d39 JMF |
292 | def xpath_with_ns(path, ns_map): |
293 | components = [c.split(':') for c in path.split('/')] | |
294 | replaced = [] | |
295 | for c in components: | |
296 | if len(c) == 1: | |
297 | replaced.append(c[0]) | |
298 | else: | |
299 | ns, tag = c | |
300 | replaced.append('{%s}%s' % (ns_map[ns], tag)) | |
301 | return '/'.join(replaced) | |
302 | ||
d77c3dfd | 303 | |
a41fb80c | 304 | def xpath_element(node, xpath, name=None, fatal=False, default=NO_DEFAULT): |
578c0745 | 305 | def _find_xpath(xpath): |
f9934b96 | 306 | return node.find(xpath) |
578c0745 | 307 | |
14f25df2 | 308 | if isinstance(xpath, str): |
578c0745 S |
309 | n = _find_xpath(xpath) |
310 | else: | |
311 | for xp in xpath: | |
312 | n = _find_xpath(xp) | |
313 | if n is not None: | |
314 | break | |
d74bebd5 | 315 | |
8e636da4 | 316 | if n is None: |
bf42a990 S |
317 | if default is not NO_DEFAULT: |
318 | return default | |
319 | elif fatal: | |
bf0ff932 PH |
320 | name = xpath if name is None else name |
321 | raise ExtractorError('Could not find XML element %s' % name) | |
322 | else: | |
323 | return None | |
a41fb80c S |
324 | return n |
325 | ||
326 | ||
327 | def xpath_text(node, xpath, name=None, fatal=False, default=NO_DEFAULT): | |
8e636da4 S |
328 | n = xpath_element(node, xpath, name, fatal=fatal, default=default) |
329 | if n is None or n == default: | |
330 | return n | |
331 | if n.text is None: | |
332 | if default is not NO_DEFAULT: | |
333 | return default | |
334 | elif fatal: | |
335 | name = xpath if name is None else name | |
336 | raise ExtractorError('Could not find XML element\'s text %s' % name) | |
337 | else: | |
338 | return None | |
339 | return n.text | |
a41fb80c S |
340 | |
341 | ||
342 | def xpath_attr(node, xpath, key, name=None, fatal=False, default=NO_DEFAULT): | |
343 | n = find_xpath_attr(node, xpath, key) | |
344 | if n is None: | |
345 | if default is not NO_DEFAULT: | |
346 | return default | |
347 | elif fatal: | |
86e5f3ed | 348 | name = f'{xpath}[@{key}]' if name is None else name |
a41fb80c S |
349 | raise ExtractorError('Could not find XML attribute %s' % name) |
350 | else: | |
351 | return None | |
352 | return n.attrib[key] | |
bf0ff932 PH |
353 | |
354 | ||
c487cf00 | 355 | def get_element_by_id(id, html, **kwargs): |
43e8fafd | 356 | """Return the content of the tag with the specified ID in the passed HTML document""" |
c487cf00 | 357 | return get_element_by_attribute('id', id, html, **kwargs) |
43e8fafd | 358 | |
12ea2f30 | 359 | |
c487cf00 | 360 | def get_element_html_by_id(id, html, **kwargs): |
6f32a0b5 | 361 | """Return the html of the tag with the specified ID in the passed HTML document""" |
c487cf00 | 362 | return get_element_html_by_attribute('id', id, html, **kwargs) |
6f32a0b5 ZM |
363 | |
364 | ||
84c237fb | 365 | def get_element_by_class(class_name, html): |
2af12ad9 TC |
366 | """Return the content of the first tag with the specified class in the passed HTML document""" |
367 | retval = get_elements_by_class(class_name, html) | |
368 | return retval[0] if retval else None | |
369 | ||
370 | ||
6f32a0b5 ZM |
371 | def get_element_html_by_class(class_name, html): |
372 | """Return the html of the first tag with the specified class in the passed HTML document""" | |
373 | retval = get_elements_html_by_class(class_name, html) | |
374 | return retval[0] if retval else None | |
375 | ||
376 | ||
c487cf00 | 377 | def get_element_by_attribute(attribute, value, html, **kwargs): |
378 | retval = get_elements_by_attribute(attribute, value, html, **kwargs) | |
2af12ad9 TC |
379 | return retval[0] if retval else None |
380 | ||
381 | ||
c487cf00 | 382 | def get_element_html_by_attribute(attribute, value, html, **kargs): |
383 | retval = get_elements_html_by_attribute(attribute, value, html, **kargs) | |
6f32a0b5 ZM |
384 | return retval[0] if retval else None |
385 | ||
386 | ||
c487cf00 | 387 | def get_elements_by_class(class_name, html, **kargs): |
2af12ad9 TC |
388 | """Return the content of all tags with the specified class in the passed HTML document as a list""" |
389 | return get_elements_by_attribute( | |
64fa820c | 390 | 'class', r'[^\'"]*(?<=[\'"\s])%s(?=[\'"\s])[^\'"]*' % re.escape(class_name), |
84c237fb YCH |
391 | html, escape_value=False) |
392 | ||
393 | ||
6f32a0b5 ZM |
394 | def get_elements_html_by_class(class_name, html): |
395 | """Return the html of all tags with the specified class in the passed HTML document as a list""" | |
396 | return get_elements_html_by_attribute( | |
64fa820c | 397 | 'class', r'[^\'"]*(?<=[\'"\s])%s(?=[\'"\s])[^\'"]*' % re.escape(class_name), |
6f32a0b5 ZM |
398 | html, escape_value=False) |
399 | ||
400 | ||
401 | def get_elements_by_attribute(*args, **kwargs): | |
43e8fafd | 402 | """Return the content of the tag with the specified attribute in the passed HTML document""" |
6f32a0b5 ZM |
403 | return [content for content, _ in get_elements_text_and_html_by_attribute(*args, **kwargs)] |
404 | ||
405 | ||
406 | def get_elements_html_by_attribute(*args, **kwargs): | |
407 | """Return the html of the tag with the specified attribute in the passed HTML document""" | |
408 | return [whole for _, whole in get_elements_text_and_html_by_attribute(*args, **kwargs)] | |
409 | ||
410 | ||
411 | def get_elements_text_and_html_by_attribute(attribute, value, html, escape_value=True): | |
412 | """ | |
413 | Return the text (content) and the html (whole) of the tag with the specified | |
414 | attribute in the passed HTML document | |
415 | """ | |
9e6dd238 | 416 | |
86e5f3ed | 417 | quote = '' if re.match(r'''[\s"'`=<>]''', value) else '?' |
0254f162 | 418 | |
84c237fb YCH |
419 | value = re.escape(value) if escape_value else value |
420 | ||
86e5f3ed | 421 | partial_element_re = rf'''(?x) |
6f32a0b5 | 422 | <(?P<tag>[a-zA-Z0-9:._-]+) |
0254f162 | 423 | (?:\s(?:[^>"']|"[^"]*"|'[^']*')*)? |
86e5f3ed | 424 | \s{re.escape(attribute)}\s*=\s*(?P<_q>['"]{quote})(?-x:{value})(?P=_q) |
425 | ''' | |
38285056 | 426 | |
0254f162 ZM |
427 | for m in re.finditer(partial_element_re, html): |
428 | content, whole = get_element_text_and_html_by_tag(m.group('tag'), html[m.start():]) | |
a921f407 | 429 | |
0254f162 ZM |
430 | yield ( |
431 | unescapeHTML(re.sub(r'^(?P<q>["\'])(?P<content>.*)(?P=q)$', r'\g<content>', content, flags=re.DOTALL)), | |
432 | whole | |
433 | ) | |
a921f407 | 434 | |
c5229f39 | 435 | |
ac668111 | 436 | class HTMLBreakOnClosingTagParser(html.parser.HTMLParser): |
6f32a0b5 ZM |
437 | """ |
438 | HTML parser which raises HTMLBreakOnClosingTagException upon reaching the | |
439 | closing tag for the first opening tag it has encountered, and can be used | |
440 | as a context manager | |
441 | """ | |
442 | ||
443 | class HTMLBreakOnClosingTagException(Exception): | |
444 | pass | |
445 | ||
446 | def __init__(self): | |
447 | self.tagstack = collections.deque() | |
ac668111 | 448 | html.parser.HTMLParser.__init__(self) |
6f32a0b5 ZM |
449 | |
450 | def __enter__(self): | |
451 | return self | |
452 | ||
453 | def __exit__(self, *_): | |
454 | self.close() | |
455 | ||
456 | def close(self): | |
457 | # handle_endtag does not return upon raising HTMLBreakOnClosingTagException, | |
458 | # so data remains buffered; we no longer have any interest in it, thus | |
459 | # override this method to discard it | |
460 | pass | |
461 | ||
462 | def handle_starttag(self, tag, _): | |
463 | self.tagstack.append(tag) | |
464 | ||
465 | def handle_endtag(self, tag): | |
466 | if not self.tagstack: | |
467 | raise compat_HTMLParseError('no tags in the stack') | |
468 | while self.tagstack: | |
469 | inner_tag = self.tagstack.pop() | |
470 | if inner_tag == tag: | |
471 | break | |
472 | else: | |
473 | raise compat_HTMLParseError(f'matching opening tag for closing {tag} tag not found') | |
474 | if not self.tagstack: | |
475 | raise self.HTMLBreakOnClosingTagException() | |
476 | ||
477 | ||
478 | def get_element_text_and_html_by_tag(tag, html): | |
479 | """ | |
480 | For the first element with the specified tag in the passed HTML document | |
481 | return its' content (text) and the whole element (html) | |
482 | """ | |
483 | def find_or_raise(haystack, needle, exc): | |
484 | try: | |
485 | return haystack.index(needle) | |
486 | except ValueError: | |
487 | raise exc | |
488 | closing_tag = f'</{tag}>' | |
489 | whole_start = find_or_raise( | |
490 | html, f'<{tag}', compat_HTMLParseError(f'opening {tag} tag not found')) | |
491 | content_start = find_or_raise( | |
492 | html[whole_start:], '>', compat_HTMLParseError(f'malformed opening {tag} tag')) | |
493 | content_start += whole_start + 1 | |
494 | with HTMLBreakOnClosingTagParser() as parser: | |
495 | parser.feed(html[whole_start:content_start]) | |
496 | if not parser.tagstack or parser.tagstack[0] != tag: | |
497 | raise compat_HTMLParseError(f'parser did not match opening {tag} tag') | |
498 | offset = content_start | |
499 | while offset < len(html): | |
500 | next_closing_tag_start = find_or_raise( | |
501 | html[offset:], closing_tag, | |
502 | compat_HTMLParseError(f'closing {tag} tag not found')) | |
503 | next_closing_tag_end = next_closing_tag_start + len(closing_tag) | |
504 | try: | |
505 | parser.feed(html[offset:offset + next_closing_tag_end]) | |
506 | offset += next_closing_tag_end | |
507 | except HTMLBreakOnClosingTagParser.HTMLBreakOnClosingTagException: | |
508 | return html[content_start:offset + next_closing_tag_start], \ | |
509 | html[whole_start:offset + next_closing_tag_end] | |
510 | raise compat_HTMLParseError('unexpected end of html') | |
511 | ||
512 | ||
ac668111 | 513 | class HTMLAttributeParser(html.parser.HTMLParser): |
8bb56eee | 514 | """Trivial HTML parser to gather the attributes for a single element""" |
b6e0c7d2 | 515 | |
8bb56eee | 516 | def __init__(self): |
c5229f39 | 517 | self.attrs = {} |
ac668111 | 518 | html.parser.HTMLParser.__init__(self) |
8bb56eee BF |
519 | |
520 | def handle_starttag(self, tag, attrs): | |
521 | self.attrs = dict(attrs) | |
522 | ||
c5229f39 | 523 | |
ac668111 | 524 | class HTMLListAttrsParser(html.parser.HTMLParser): |
73673ccf FF |
525 | """HTML parser to gather the attributes for the elements of a list""" |
526 | ||
527 | def __init__(self): | |
ac668111 | 528 | html.parser.HTMLParser.__init__(self) |
73673ccf FF |
529 | self.items = [] |
530 | self._level = 0 | |
531 | ||
532 | def handle_starttag(self, tag, attrs): | |
533 | if tag == 'li' and self._level == 0: | |
534 | self.items.append(dict(attrs)) | |
535 | self._level += 1 | |
536 | ||
537 | def handle_endtag(self, tag): | |
538 | self._level -= 1 | |
539 | ||
540 | ||
8bb56eee BF |
541 | def extract_attributes(html_element): |
542 | """Given a string for an HTML element such as | |
543 | <el | |
544 | a="foo" B="bar" c="&98;az" d=boz | |
545 | empty= noval entity="&" | |
546 | sq='"' dq="'" | |
547 | > | |
548 | Decode and return a dictionary of attributes. | |
549 | { | |
550 | 'a': 'foo', 'b': 'bar', c: 'baz', d: 'boz', | |
551 | 'empty': '', 'noval': None, 'entity': '&', | |
552 | 'sq': '"', 'dq': '\'' | |
553 | }. | |
8bb56eee BF |
554 | """ |
555 | parser = HTMLAttributeParser() | |
19a03940 | 556 | with contextlib.suppress(compat_HTMLParseError): |
b4a3d461 S |
557 | parser.feed(html_element) |
558 | parser.close() | |
8bb56eee | 559 | return parser.attrs |
9e6dd238 | 560 | |
c5229f39 | 561 | |
73673ccf FF |
562 | def parse_list(webpage): |
563 | """Given a string for an series of HTML <li> elements, | |
564 | return a dictionary of their attributes""" | |
565 | parser = HTMLListAttrsParser() | |
566 | parser.feed(webpage) | |
567 | parser.close() | |
568 | return parser.items | |
569 | ||
570 | ||
9e6dd238 | 571 | def clean_html(html): |
59ae15a5 | 572 | """Clean an HTML snippet into a readable string""" |
dd622d7c PH |
573 | |
574 | if html is None: # Convenience for sanitizing descriptions etc. | |
575 | return html | |
576 | ||
49185227 | 577 | html = re.sub(r'\s+', ' ', html) |
578 | html = re.sub(r'(?u)\s?<\s?br\s?/?\s?>\s?', '\n', html) | |
579 | html = re.sub(r'(?u)<\s?/\s?p\s?>\s?<\s?p[^>]*>', '\n', html) | |
59ae15a5 PH |
580 | # Strip html tags |
581 | html = re.sub('<.*?>', '', html) | |
582 | # Replace html entities | |
583 | html = unescapeHTML(html) | |
7decf895 | 584 | return html.strip() |
9e6dd238 FV |
585 | |
586 | ||
b7c47b74 | 587 | class LenientJSONDecoder(json.JSONDecoder): |
588 | def __init__(self, *args, transform_source=None, ignore_extra=False, **kwargs): | |
589 | self.transform_source, self.ignore_extra = transform_source, ignore_extra | |
590 | super().__init__(*args, **kwargs) | |
591 | ||
592 | def decode(self, s): | |
593 | if self.transform_source: | |
594 | s = self.transform_source(s) | |
2fa669f7 | 595 | try: |
596 | if self.ignore_extra: | |
597 | return self.raw_decode(s.lstrip())[0] | |
598 | return super().decode(s) | |
599 | except json.JSONDecodeError as e: | |
600 | if e.pos is not None: | |
601 | raise type(e)(f'{e.msg} in {s[e.pos-10:e.pos+10]!r}', s, e.pos) | |
602 | raise | |
b7c47b74 | 603 | |
604 | ||
d77c3dfd | 605 | def sanitize_open(filename, open_mode): |
59ae15a5 PH |
606 | """Try to open the given filename, and slightly tweak it if this fails. |
607 | ||
608 | Attempts to open the given filename. If this fails, it tries to change | |
609 | the filename slightly, step by step, until it's either able to open it | |
610 | or it fails and raises a final exception, like the standard open() | |
611 | function. | |
612 | ||
613 | It returns the tuple (stream, definitive_file_name). | |
614 | """ | |
0edb3e33 | 615 | if filename == '-': |
616 | if sys.platform == 'win32': | |
617 | import msvcrt | |
be5c1ae8 | 618 | |
62b58c09 | 619 | # stdout may be any IO stream, e.g. when using contextlib.redirect_stdout |
daef7911 | 620 | with contextlib.suppress(io.UnsupportedOperation): |
621 | msvcrt.setmode(sys.stdout.fileno(), os.O_BINARY) | |
0edb3e33 | 622 | return (sys.stdout.buffer if hasattr(sys.stdout, 'buffer') else sys.stdout, filename) |
59ae15a5 | 623 | |
0edb3e33 | 624 | for attempt in range(2): |
625 | try: | |
626 | try: | |
89737671 | 627 | if sys.platform == 'win32': |
b506289f | 628 | # FIXME: An exclusive lock also locks the file from being read. |
629 | # Since windows locks are mandatory, don't lock the file on windows (for now). | |
630 | # Ref: https://github.com/yt-dlp/yt-dlp/issues/3124 | |
89737671 | 631 | raise LockingUnsupportedError() |
0edb3e33 | 632 | stream = locked_file(filename, open_mode, block=False).__enter__() |
8a82af35 | 633 | except OSError: |
0edb3e33 | 634 | stream = open(filename, open_mode) |
8a82af35 | 635 | return stream, filename |
86e5f3ed | 636 | except OSError as err: |
0edb3e33 | 637 | if attempt or err.errno in (errno.EACCES,): |
638 | raise | |
639 | old_filename, filename = filename, sanitize_path(filename) | |
640 | if old_filename == filename: | |
641 | raise | |
d77c3dfd FV |
642 | |
643 | ||
644 | def timeconvert(timestr): | |
59ae15a5 PH |
645 | """Convert RFC 2822 defined time string into system timestamp""" |
646 | timestamp = None | |
647 | timetuple = email.utils.parsedate_tz(timestr) | |
648 | if timetuple is not None: | |
649 | timestamp = email.utils.mktime_tz(timetuple) | |
650 | return timestamp | |
1c469a94 | 651 | |
5f6a1245 | 652 | |
5c3895ff | 653 | def sanitize_filename(s, restricted=False, is_id=NO_DEFAULT): |
59ae15a5 | 654 | """Sanitizes a string so it could be used as part of a filename. |
5c3895ff | 655 | @param restricted Use a stricter subset of allowed characters |
656 | @param is_id Whether this is an ID that should be kept unchanged if possible. | |
657 | If unset, yt-dlp's new sanitization rules are in effect | |
59ae15a5 | 658 | """ |
5c3895ff | 659 | if s == '': |
660 | return '' | |
661 | ||
59ae15a5 | 662 | def replace_insane(char): |
c587cbb7 AT |
663 | if restricted and char in ACCENT_CHARS: |
664 | return ACCENT_CHARS[char] | |
91dd88b9 | 665 | elif not restricted and char == '\n': |
5c3895ff | 666 | return '\0 ' |
989a01c2 | 667 | elif is_id is NO_DEFAULT and not restricted and char in '"*:<>?|/\\': |
668 | # Replace with their full-width unicode counterparts | |
669 | return {'/': '\u29F8', '\\': '\u29f9'}.get(char, chr(ord(char) + 0xfee0)) | |
91dd88b9 | 670 | elif char == '?' or ord(char) < 32 or ord(char) == 127: |
59ae15a5 PH |
671 | return '' |
672 | elif char == '"': | |
673 | return '' if restricted else '\'' | |
674 | elif char == ':': | |
5c3895ff | 675 | return '\0_\0-' if restricted else '\0 \0-' |
59ae15a5 | 676 | elif char in '\\/|*<>': |
5c3895ff | 677 | return '\0_' |
678 | if restricted and (char in '!&\'()[]{}$;`^,#' or char.isspace() or ord(char) > 127): | |
679 | return '\0_' | |
59ae15a5 PH |
680 | return char |
681 | ||
989a01c2 | 682 | if restricted and is_id is NO_DEFAULT: |
683 | s = unicodedata.normalize('NFKC', s) | |
5c3895ff | 684 | s = re.sub(r'[0-9]+(?::[0-9]+)+', lambda m: m.group(0).replace(':', '_'), s) # Handle timestamps |
28e614de | 685 | result = ''.join(map(replace_insane, s)) |
5c3895ff | 686 | if is_id is NO_DEFAULT: |
ae61d108 | 687 | result = re.sub(r'(\0.)(?:(?=\1)..)+', r'\1', result) # Remove repeated substitute chars |
688 | STRIP_RE = r'(?:\0.|[ _-])*' | |
5c3895ff | 689 | result = re.sub(f'^\0.{STRIP_RE}|{STRIP_RE}\0.$', '', result) # Remove substitute chars from start/end |
690 | result = result.replace('\0', '') or '_' | |
691 | ||
796173d0 PH |
692 | if not is_id: |
693 | while '__' in result: | |
694 | result = result.replace('__', '_') | |
695 | result = result.strip('_') | |
696 | # Common case of "Foreign band name - English song title" | |
697 | if restricted and result.startswith('-_'): | |
698 | result = result[2:] | |
5a42414b PH |
699 | if result.startswith('-'): |
700 | result = '_' + result[len('-'):] | |
a7440261 | 701 | result = result.lstrip('.') |
796173d0 PH |
702 | if not result: |
703 | result = '_' | |
59ae15a5 | 704 | return result |
d77c3dfd | 705 | |
5f6a1245 | 706 | |
c2934512 | 707 | def sanitize_path(s, force=False): |
a2aaf4db | 708 | """Sanitizes and normalizes path on Windows""" |
c2934512 | 709 | if sys.platform == 'win32': |
c4218ac3 | 710 | force = False |
c2934512 | 711 | drive_or_unc, _ = os.path.splitdrive(s) |
c2934512 | 712 | elif force: |
713 | drive_or_unc = '' | |
714 | else: | |
a2aaf4db | 715 | return s |
c2934512 | 716 | |
be531ef1 S |
717 | norm_path = os.path.normpath(remove_start(s, drive_or_unc)).split(os.path.sep) |
718 | if drive_or_unc: | |
a2aaf4db S |
719 | norm_path.pop(0) |
720 | sanitized_path = [ | |
ec85ded8 | 721 | path_part if path_part in ['.', '..'] else re.sub(r'(?:[/<>:"\|\\?\*]|[\s.]$)', '#', path_part) |
a2aaf4db | 722 | for path_part in norm_path] |
be531ef1 S |
723 | if drive_or_unc: |
724 | sanitized_path.insert(0, drive_or_unc + os.path.sep) | |
4abea8ca | 725 | elif force and s and s[0] == os.path.sep: |
c4218ac3 | 726 | sanitized_path.insert(0, os.path.sep) |
a2aaf4db S |
727 | return os.path.join(*sanitized_path) |
728 | ||
729 | ||
8f97a15d | 730 | def sanitize_url(url, *, scheme='http'): |
befa4708 S |
731 | # Prepend protocol-less URLs with `http:` scheme in order to mitigate |
732 | # the number of unwanted failures due to missing protocol | |
21633673 | 733 | if url is None: |
734 | return | |
735 | elif url.startswith('//'): | |
8f97a15d | 736 | return f'{scheme}:{url}' |
befa4708 S |
737 | # Fix some common typos seen so far |
738 | COMMON_TYPOS = ( | |
067aa17e | 739 | # https://github.com/ytdl-org/youtube-dl/issues/15649 |
befa4708 S |
740 | (r'^httpss://', r'https://'), |
741 | # https://bx1.be/lives/direct-tv/ | |
742 | (r'^rmtp([es]?)://', r'rtmp\1://'), | |
743 | ) | |
744 | for mistake, fixup in COMMON_TYPOS: | |
745 | if re.match(mistake, url): | |
746 | return re.sub(mistake, fixup, url) | |
bc6b9bcd | 747 | return url |
17bcc626 S |
748 | |
749 | ||
5435dcf9 | 750 | def extract_basic_auth(url): |
14f25df2 | 751 | parts = urllib.parse.urlsplit(url) |
5435dcf9 HH |
752 | if parts.username is None: |
753 | return url, None | |
14f25df2 | 754 | url = urllib.parse.urlunsplit(parts._replace(netloc=( |
5435dcf9 HH |
755 | parts.hostname if parts.port is None |
756 | else '%s:%d' % (parts.hostname, parts.port)))) | |
757 | auth_payload = base64.b64encode( | |
0f06bcd7 | 758 | ('%s:%s' % (parts.username, parts.password or '')).encode()) |
759 | return url, f'Basic {auth_payload.decode()}' | |
5435dcf9 HH |
760 | |
761 | ||
67dda517 | 762 | def sanitized_Request(url, *args, **kwargs): |
bc6b9bcd | 763 | url, auth_header = extract_basic_auth(escape_url(sanitize_url(url))) |
5435dcf9 HH |
764 | if auth_header is not None: |
765 | headers = args[1] if len(args) >= 2 else kwargs.setdefault('headers', {}) | |
766 | headers['Authorization'] = auth_header | |
ac668111 | 767 | return urllib.request.Request(url, *args, **kwargs) |
67dda517 S |
768 | |
769 | ||
51098426 | 770 | def expand_path(s): |
2fa669f7 | 771 | """Expand shell variables and ~""" |
51098426 S |
772 | return os.path.expandvars(compat_expanduser(s)) |
773 | ||
774 | ||
7e9a6125 | 775 | def orderedSet(iterable, *, lazy=False): |
776 | """Remove all duplicates from the input iterable""" | |
777 | def _iter(): | |
778 | seen = [] # Do not use set since the items can be unhashable | |
779 | for x in iterable: | |
780 | if x not in seen: | |
781 | seen.append(x) | |
782 | yield x | |
783 | ||
784 | return _iter() if lazy else list(_iter()) | |
d77c3dfd | 785 | |
912b38b4 | 786 | |
55b2f099 | 787 | def _htmlentity_transform(entity_with_semicolon): |
4e408e47 | 788 | """Transforms an HTML entity to a character.""" |
55b2f099 YCH |
789 | entity = entity_with_semicolon[:-1] |
790 | ||
4e408e47 | 791 | # Known non-numeric HTML entity |
ac668111 | 792 | if entity in html.entities.name2codepoint: |
793 | return chr(html.entities.name2codepoint[entity]) | |
4e408e47 | 794 | |
62b58c09 L |
795 | # TODO: HTML5 allows entities without a semicolon. |
796 | # E.g. 'Éric' should be decoded as 'Éric'. | |
ac668111 | 797 | if entity_with_semicolon in html.entities.html5: |
798 | return html.entities.html5[entity_with_semicolon] | |
55b2f099 | 799 | |
91757b0f | 800 | mobj = re.match(r'#(x[0-9a-fA-F]+|[0-9]+)', entity) |
4e408e47 PH |
801 | if mobj is not None: |
802 | numstr = mobj.group(1) | |
28e614de | 803 | if numstr.startswith('x'): |
4e408e47 | 804 | base = 16 |
28e614de | 805 | numstr = '0%s' % numstr |
4e408e47 PH |
806 | else: |
807 | base = 10 | |
067aa17e | 808 | # See https://github.com/ytdl-org/youtube-dl/issues/7518 |
19a03940 | 809 | with contextlib.suppress(ValueError): |
ac668111 | 810 | return chr(int(numstr, base)) |
4e408e47 PH |
811 | |
812 | # Unknown entity in name, return its literal representation | |
7a3f0c00 | 813 | return '&%s;' % entity |
4e408e47 PH |
814 | |
815 | ||
d77c3dfd | 816 | def unescapeHTML(s): |
912b38b4 PH |
817 | if s is None: |
818 | return None | |
19a03940 | 819 | assert isinstance(s, str) |
d77c3dfd | 820 | |
4e408e47 | 821 | return re.sub( |
95f3f7c2 | 822 | r'&([^&;]+;)', lambda m: _htmlentity_transform(m.group(1)), s) |
d77c3dfd | 823 | |
8bf48f23 | 824 | |
cdb19aa4 | 825 | def escapeHTML(text): |
826 | return ( | |
827 | text | |
828 | .replace('&', '&') | |
829 | .replace('<', '<') | |
830 | .replace('>', '>') | |
831 | .replace('"', '"') | |
832 | .replace("'", ''') | |
833 | ) | |
834 | ||
835 | ||
f5b1bca9 | 836 | def process_communicate_or_kill(p, *args, **kwargs): |
da4db748 | 837 | deprecation_warning(f'"{__name__}.process_communicate_or_kill" is deprecated and may be removed ' |
838 | f'in a future version. Use "{__name__}.Popen.communicate_or_kill" instead') | |
8a82af35 | 839 | return Popen.communicate_or_kill(p, *args, **kwargs) |
f5b1bca9 | 840 | |
841 | ||
d3c93ec2 | 842 | class Popen(subprocess.Popen): |
843 | if sys.platform == 'win32': | |
844 | _startupinfo = subprocess.STARTUPINFO() | |
845 | _startupinfo.dwFlags |= subprocess.STARTF_USESHOWWINDOW | |
846 | else: | |
847 | _startupinfo = None | |
848 | ||
82ea226c L |
849 | @staticmethod |
850 | def _fix_pyinstaller_ld_path(env): | |
851 | """Restore LD_LIBRARY_PATH when using PyInstaller | |
852 | Ref: https://github.com/pyinstaller/pyinstaller/blob/develop/doc/runtime-information.rst#ld_library_path--libpath-considerations | |
853 | https://github.com/yt-dlp/yt-dlp/issues/4573 | |
854 | """ | |
855 | if not hasattr(sys, '_MEIPASS'): | |
856 | return | |
857 | ||
858 | def _fix(key): | |
859 | orig = env.get(f'{key}_ORIG') | |
860 | if orig is None: | |
861 | env.pop(key, None) | |
862 | else: | |
863 | env[key] = orig | |
864 | ||
865 | _fix('LD_LIBRARY_PATH') # Linux | |
866 | _fix('DYLD_LIBRARY_PATH') # macOS | |
867 | ||
868 | def __init__(self, *args, env=None, text=False, **kwargs): | |
869 | if env is None: | |
870 | env = os.environ.copy() | |
871 | self._fix_pyinstaller_ld_path(env) | |
872 | ||
f0c9fb96 | 873 | if text is True: |
874 | kwargs['universal_newlines'] = True # For 3.6 compatibility | |
875 | kwargs.setdefault('encoding', 'utf-8') | |
876 | kwargs.setdefault('errors', 'replace') | |
82ea226c | 877 | super().__init__(*args, env=env, **kwargs, startupinfo=self._startupinfo) |
d3c93ec2 | 878 | |
879 | def communicate_or_kill(self, *args, **kwargs): | |
8a82af35 | 880 | try: |
881 | return self.communicate(*args, **kwargs) | |
882 | except BaseException: # Including KeyboardInterrupt | |
f0c9fb96 | 883 | self.kill(timeout=None) |
8a82af35 | 884 | raise |
d3c93ec2 | 885 | |
f0c9fb96 | 886 | def kill(self, *, timeout=0): |
887 | super().kill() | |
888 | if timeout != 0: | |
889 | self.wait(timeout=timeout) | |
890 | ||
891 | @classmethod | |
992dc6b4 | 892 | def run(cls, *args, timeout=None, **kwargs): |
f0c9fb96 | 893 | with cls(*args, **kwargs) as proc: |
914491b8 | 894 | default = '' if proc.text_mode else b'' |
992dc6b4 | 895 | stdout, stderr = proc.communicate_or_kill(timeout=timeout) |
914491b8 | 896 | return stdout or default, stderr or default, proc.returncode |
f0c9fb96 | 897 | |
d3c93ec2 | 898 | |
aa49acd1 S |
899 | def get_subprocess_encoding(): |
900 | if sys.platform == 'win32' and sys.getwindowsversion()[0] >= 5: | |
901 | # For subprocess calls, encode with locale encoding | |
902 | # Refer to http://stackoverflow.com/a/9951851/35070 | |
903 | encoding = preferredencoding() | |
904 | else: | |
905 | encoding = sys.getfilesystemencoding() | |
906 | if encoding is None: | |
907 | encoding = 'utf-8' | |
908 | return encoding | |
909 | ||
910 | ||
8bf48f23 | 911 | def encodeFilename(s, for_subprocess=False): |
19a03940 | 912 | assert isinstance(s, str) |
cfb0511d | 913 | return s |
aa49acd1 S |
914 | |
915 | ||
916 | def decodeFilename(b, for_subprocess=False): | |
cfb0511d | 917 | return b |
8bf48f23 | 918 | |
f07b74fc PH |
919 | |
920 | def encodeArgument(s): | |
cfb0511d | 921 | # Legacy code that uses byte strings |
922 | # Uncomment the following line after fixing all post processors | |
14f25df2 | 923 | # assert isinstance(s, str), 'Internal error: %r should be of type %r, is %r' % (s, str, type(s)) |
cfb0511d | 924 | return s if isinstance(s, str) else s.decode('ascii') |
f07b74fc PH |
925 | |
926 | ||
aa49acd1 | 927 | def decodeArgument(b): |
cfb0511d | 928 | return b |
aa49acd1 S |
929 | |
930 | ||
8271226a PH |
931 | def decodeOption(optval): |
932 | if optval is None: | |
933 | return optval | |
934 | if isinstance(optval, bytes): | |
935 | optval = optval.decode(preferredencoding()) | |
936 | ||
14f25df2 | 937 | assert isinstance(optval, str) |
8271226a | 938 | return optval |
1c256f70 | 939 | |
5f6a1245 | 940 | |
aa7785f8 | 941 | _timetuple = collections.namedtuple('Time', ('hours', 'minutes', 'seconds', 'milliseconds')) |
942 | ||
943 | ||
944 | def timetuple_from_msec(msec): | |
945 | secs, msec = divmod(msec, 1000) | |
946 | mins, secs = divmod(secs, 60) | |
947 | hrs, mins = divmod(mins, 60) | |
948 | return _timetuple(hrs, mins, secs, msec) | |
949 | ||
950 | ||
cdb19aa4 | 951 | def formatSeconds(secs, delim=':', msec=False): |
aa7785f8 | 952 | time = timetuple_from_msec(secs * 1000) |
953 | if time.hours: | |
954 | ret = '%d%s%02d%s%02d' % (time.hours, delim, time.minutes, delim, time.seconds) | |
955 | elif time.minutes: | |
956 | ret = '%d%s%02d' % (time.minutes, delim, time.seconds) | |
4539dd30 | 957 | else: |
aa7785f8 | 958 | ret = '%d' % time.seconds |
959 | return '%s.%03d' % (ret, time.milliseconds) if msec else ret | |
4539dd30 | 960 | |
a0ddb8a2 | 961 | |
77562778 | 962 | def _ssl_load_windows_store_certs(ssl_context, storename): |
963 | # Code adapted from _load_windows_store_certs in https://github.com/python/cpython/blob/main/Lib/ssl.py | |
964 | try: | |
965 | certs = [cert for cert, encoding, trust in ssl.enum_certificates(storename) | |
966 | if encoding == 'x509_asn' and ( | |
967 | trust is True or ssl.Purpose.SERVER_AUTH.oid in trust)] | |
968 | except PermissionError: | |
969 | return | |
970 | for cert in certs: | |
19a03940 | 971 | with contextlib.suppress(ssl.SSLError): |
77562778 | 972 | ssl_context.load_verify_locations(cadata=cert) |
a2366922 | 973 | |
77562778 | 974 | |
975 | def make_HTTPS_handler(params, **kwargs): | |
976 | opts_check_certificate = not params.get('nocheckcertificate') | |
977 | context = ssl.SSLContext(ssl.PROTOCOL_TLS_CLIENT) | |
978 | context.check_hostname = opts_check_certificate | |
f81c62a6 | 979 | if params.get('legacyserverconnect'): |
980 | context.options |= 4 # SSL_OP_LEGACY_SERVER_CONNECT | |
4f28b537 | 981 | # Allow use of weaker ciphers in Python 3.10+. See https://bugs.python.org/issue43998 |
982 | context.set_ciphers('DEFAULT') | |
8a82af35 | 983 | |
77562778 | 984 | context.verify_mode = ssl.CERT_REQUIRED if opts_check_certificate else ssl.CERT_NONE |
985 | if opts_check_certificate: | |
d5820461 | 986 | if has_certifi and 'no-certifi' not in params.get('compat_opts', []): |
987 | context.load_verify_locations(cafile=certifi.where()) | |
168bbc4f | 988 | else: |
989 | try: | |
990 | context.load_default_certs() | |
991 | # Work around the issue in load_default_certs when there are bad certificates. See: | |
992 | # https://github.com/yt-dlp/yt-dlp/issues/1060, | |
993 | # https://bugs.python.org/issue35665, https://bugs.python.org/issue45312 | |
994 | except ssl.SSLError: | |
995 | # enum_certificates is not present in mingw python. See https://github.com/yt-dlp/yt-dlp/issues/1151 | |
996 | if sys.platform == 'win32' and hasattr(ssl, 'enum_certificates'): | |
997 | for storename in ('CA', 'ROOT'): | |
998 | _ssl_load_windows_store_certs(context, storename) | |
999 | context.set_default_verify_paths() | |
8a82af35 | 1000 | |
bb58c9ed | 1001 | client_certfile = params.get('client_certificate') |
1002 | if client_certfile: | |
1003 | try: | |
1004 | context.load_cert_chain( | |
1005 | client_certfile, keyfile=params.get('client_certificate_key'), | |
1006 | password=params.get('client_certificate_password')) | |
1007 | except ssl.SSLError: | |
1008 | raise YoutubeDLError('Unable to load client certificate') | |
2c6dcb65 | 1009 | |
1010 | # Some servers may reject requests if ALPN extension is not sent. See: | |
1011 | # https://github.com/python/cpython/issues/85140 | |
1012 | # https://github.com/yt-dlp/yt-dlp/issues/3878 | |
1013 | with contextlib.suppress(NotImplementedError): | |
1014 | context.set_alpn_protocols(['http/1.1']) | |
1015 | ||
77562778 | 1016 | return YoutubeDLHTTPSHandler(params, context=context, **kwargs) |
ea6d901e | 1017 | |
732ea2f0 | 1018 | |
5873d4cc | 1019 | def bug_reports_message(before=';'): |
57e0f077 | 1020 | from .update import REPOSITORY |
1021 | ||
1022 | msg = (f'please report this issue on https://github.com/{REPOSITORY}/issues?q= , ' | |
1023 | 'filling out the appropriate issue template. Confirm you are on the latest version using yt-dlp -U') | |
5873d4cc F |
1024 | |
1025 | before = before.rstrip() | |
1026 | if not before or before.endswith(('.', '!', '?')): | |
1027 | msg = msg[0].title() + msg[1:] | |
1028 | ||
1029 | return (before + ' ' if before else '') + msg | |
08f2a92c JMF |
1030 | |
1031 | ||
bf5b9d85 PM |
1032 | class YoutubeDLError(Exception): |
1033 | """Base exception for YoutubeDL errors.""" | |
aa9369a2 | 1034 | msg = None |
1035 | ||
1036 | def __init__(self, msg=None): | |
1037 | if msg is not None: | |
1038 | self.msg = msg | |
1039 | elif self.msg is None: | |
1040 | self.msg = type(self).__name__ | |
1041 | super().__init__(self.msg) | |
bf5b9d85 PM |
1042 | |
1043 | ||
ac668111 | 1044 | network_exceptions = [urllib.error.URLError, http.client.HTTPException, socket.error] |
3158150c | 1045 | if hasattr(ssl, 'CertificateError'): |
1046 | network_exceptions.append(ssl.CertificateError) | |
1047 | network_exceptions = tuple(network_exceptions) | |
1048 | ||
1049 | ||
bf5b9d85 | 1050 | class ExtractorError(YoutubeDLError): |
1c256f70 | 1051 | """Error during info extraction.""" |
5f6a1245 | 1052 | |
1151c407 | 1053 | def __init__(self, msg, tb=None, expected=False, cause=None, video_id=None, ie=None): |
9a82b238 | 1054 | """ tb, if given, is the original traceback (so that it can be printed out). |
7a5c1cfe | 1055 | If expected is set, this is a normal error message and most likely not a bug in yt-dlp. |
9a82b238 | 1056 | """ |
3158150c | 1057 | if sys.exc_info()[0] in network_exceptions: |
9a82b238 | 1058 | expected = True |
d5979c5d | 1059 | |
7265a219 | 1060 | self.orig_msg = str(msg) |
1c256f70 | 1061 | self.traceback = tb |
1151c407 | 1062 | self.expected = expected |
2eabb802 | 1063 | self.cause = cause |
d11271dd | 1064 | self.video_id = video_id |
1151c407 | 1065 | self.ie = ie |
1066 | self.exc_info = sys.exc_info() # preserve original exception | |
5df14442 | 1067 | if isinstance(self.exc_info[1], ExtractorError): |
1068 | self.exc_info = self.exc_info[1].exc_info | |
1151c407 | 1069 | |
86e5f3ed | 1070 | super().__init__(''.join(( |
a70635b8 | 1071 | format_field(ie, None, '[%s] '), |
1072 | format_field(video_id, None, '%s: '), | |
7265a219 | 1073 | msg, |
a70635b8 | 1074 | format_field(cause, None, ' (caused by %r)'), |
1151c407 | 1075 | '' if expected else bug_reports_message()))) |
1c256f70 | 1076 | |
01951dda | 1077 | def format_traceback(self): |
497d2fab | 1078 | return join_nonempty( |
1079 | self.traceback and ''.join(traceback.format_tb(self.traceback)), | |
e491d06d | 1080 | self.cause and ''.join(traceback.format_exception(None, self.cause, self.cause.__traceback__)[1:]), |
497d2fab | 1081 | delim='\n') or None |
01951dda | 1082 | |
1c256f70 | 1083 | |
416c7fcb PH |
1084 | class UnsupportedError(ExtractorError): |
1085 | def __init__(self, url): | |
86e5f3ed | 1086 | super().__init__( |
416c7fcb PH |
1087 | 'Unsupported URL: %s' % url, expected=True) |
1088 | self.url = url | |
1089 | ||
1090 | ||
55b3e45b JMF |
1091 | class RegexNotFoundError(ExtractorError): |
1092 | """Error when a regex didn't match""" | |
1093 | pass | |
1094 | ||
1095 | ||
773f291d S |
1096 | class GeoRestrictedError(ExtractorError): |
1097 | """Geographic restriction Error exception. | |
1098 | ||
1099 | This exception may be thrown when a video is not available from your | |
1100 | geographic location due to geographic restrictions imposed by a website. | |
1101 | """ | |
b6e0c7d2 | 1102 | |
0db3bae8 | 1103 | def __init__(self, msg, countries=None, **kwargs): |
1104 | kwargs['expected'] = True | |
86e5f3ed | 1105 | super().__init__(msg, **kwargs) |
773f291d S |
1106 | self.countries = countries |
1107 | ||
1108 | ||
693f0600 | 1109 | class UserNotLive(ExtractorError): |
1110 | """Error when a channel/user is not live""" | |
1111 | ||
1112 | def __init__(self, msg=None, **kwargs): | |
1113 | kwargs['expected'] = True | |
1114 | super().__init__(msg or 'The channel is not currently live', **kwargs) | |
1115 | ||
1116 | ||
bf5b9d85 | 1117 | class DownloadError(YoutubeDLError): |
59ae15a5 | 1118 | """Download Error exception. |
d77c3dfd | 1119 | |
59ae15a5 PH |
1120 | This exception may be thrown by FileDownloader objects if they are not |
1121 | configured to continue on errors. They will contain the appropriate | |
1122 | error message. | |
1123 | """ | |
5f6a1245 | 1124 | |
8cc83b8d FV |
1125 | def __init__(self, msg, exc_info=None): |
1126 | """ exc_info, if given, is the original exception that caused the trouble (as returned by sys.exc_info()). """ | |
86e5f3ed | 1127 | super().__init__(msg) |
8cc83b8d | 1128 | self.exc_info = exc_info |
d77c3dfd FV |
1129 | |
1130 | ||
498f5606 | 1131 | class EntryNotInPlaylist(YoutubeDLError): |
1132 | """Entry not in playlist exception. | |
1133 | ||
1134 | This exception will be thrown by YoutubeDL when a requested entry | |
1135 | is not found in the playlist info_dict | |
1136 | """ | |
aa9369a2 | 1137 | msg = 'Entry not found in info' |
498f5606 | 1138 | |
1139 | ||
bf5b9d85 | 1140 | class SameFileError(YoutubeDLError): |
59ae15a5 | 1141 | """Same File exception. |
d77c3dfd | 1142 | |
59ae15a5 PH |
1143 | This exception will be thrown by FileDownloader objects if they detect |
1144 | multiple files would have to be downloaded to the same file on disk. | |
1145 | """ | |
aa9369a2 | 1146 | msg = 'Fixed output name but more than one file to download' |
1147 | ||
1148 | def __init__(self, filename=None): | |
1149 | if filename is not None: | |
1150 | self.msg += f': {filename}' | |
1151 | super().__init__(self.msg) | |
d77c3dfd FV |
1152 | |
1153 | ||
bf5b9d85 | 1154 | class PostProcessingError(YoutubeDLError): |
59ae15a5 | 1155 | """Post Processing exception. |
d77c3dfd | 1156 | |
59ae15a5 PH |
1157 | This exception may be raised by PostProcessor's .run() method to |
1158 | indicate an error in the postprocessing task. | |
1159 | """ | |
5f6a1245 | 1160 | |
5f6a1245 | 1161 | |
48f79687 | 1162 | class DownloadCancelled(YoutubeDLError): |
1163 | """ Exception raised when the download queue should be interrupted """ | |
1164 | msg = 'The download was cancelled' | |
8b0d7497 | 1165 | |
8b0d7497 | 1166 | |
48f79687 | 1167 | class ExistingVideoReached(DownloadCancelled): |
1168 | """ --break-on-existing triggered """ | |
1169 | msg = 'Encountered a video that is already in the archive, stopping due to --break-on-existing' | |
8b0d7497 | 1170 | |
48f79687 | 1171 | |
1172 | class RejectedVideoReached(DownloadCancelled): | |
1173 | """ --break-on-reject triggered """ | |
1174 | msg = 'Encountered a video that did not match filter, stopping due to --break-on-reject' | |
51d9739f | 1175 | |
1176 | ||
48f79687 | 1177 | class MaxDownloadsReached(DownloadCancelled): |
59ae15a5 | 1178 | """ --max-downloads limit has been reached. """ |
48f79687 | 1179 | msg = 'Maximum number of downloads reached, stopping due to --max-downloads' |
1180 | ||
1181 | ||
f2ebc5c7 | 1182 | class ReExtractInfo(YoutubeDLError): |
1183 | """ Video info needs to be re-extracted. """ | |
1184 | ||
1185 | def __init__(self, msg, expected=False): | |
1186 | super().__init__(msg) | |
1187 | self.expected = expected | |
1188 | ||
1189 | ||
1190 | class ThrottledDownload(ReExtractInfo): | |
48f79687 | 1191 | """ Download speed below --throttled-rate. """ |
aa9369a2 | 1192 | msg = 'The download speed is below throttle limit' |
d77c3dfd | 1193 | |
43b22906 | 1194 | def __init__(self): |
1195 | super().__init__(self.msg, expected=False) | |
f2ebc5c7 | 1196 | |
d77c3dfd | 1197 | |
bf5b9d85 | 1198 | class UnavailableVideoError(YoutubeDLError): |
59ae15a5 | 1199 | """Unavailable Format exception. |
d77c3dfd | 1200 | |
59ae15a5 PH |
1201 | This exception will be thrown when a video is requested |
1202 | in a format that is not available for that video. | |
1203 | """ | |
aa9369a2 | 1204 | msg = 'Unable to download video' |
1205 | ||
1206 | def __init__(self, err=None): | |
1207 | if err is not None: | |
1208 | self.msg += f': {err}' | |
1209 | super().__init__(self.msg) | |
d77c3dfd FV |
1210 | |
1211 | ||
bf5b9d85 | 1212 | class ContentTooShortError(YoutubeDLError): |
59ae15a5 | 1213 | """Content Too Short exception. |
d77c3dfd | 1214 | |
59ae15a5 PH |
1215 | This exception may be raised by FileDownloader objects when a file they |
1216 | download is too small for what the server announced first, indicating | |
1217 | the connection was probably interrupted. | |
1218 | """ | |
d77c3dfd | 1219 | |
59ae15a5 | 1220 | def __init__(self, downloaded, expected): |
86e5f3ed | 1221 | super().__init__(f'Downloaded {downloaded} bytes, expected {expected} bytes') |
2c7ed247 | 1222 | # Both in bytes |
59ae15a5 PH |
1223 | self.downloaded = downloaded |
1224 | self.expected = expected | |
d77c3dfd | 1225 | |
5f6a1245 | 1226 | |
bf5b9d85 | 1227 | class XAttrMetadataError(YoutubeDLError): |
efa97bdc | 1228 | def __init__(self, code=None, msg='Unknown error'): |
86e5f3ed | 1229 | super().__init__(msg) |
efa97bdc | 1230 | self.code = code |
bd264412 | 1231 | self.msg = msg |
efa97bdc YCH |
1232 | |
1233 | # Parsing code and msg | |
3089bc74 | 1234 | if (self.code in (errno.ENOSPC, errno.EDQUOT) |
a0566bbf | 1235 | or 'No space left' in self.msg or 'Disk quota exceeded' in self.msg): |
efa97bdc YCH |
1236 | self.reason = 'NO_SPACE' |
1237 | elif self.code == errno.E2BIG or 'Argument list too long' in self.msg: | |
1238 | self.reason = 'VALUE_TOO_LONG' | |
1239 | else: | |
1240 | self.reason = 'NOT_SUPPORTED' | |
1241 | ||
1242 | ||
bf5b9d85 | 1243 | class XAttrUnavailableError(YoutubeDLError): |
efa97bdc YCH |
1244 | pass |
1245 | ||
1246 | ||
c5a59d93 | 1247 | def _create_http_connection(ydl_handler, http_class, is_https, *args, **kwargs): |
f9934b96 | 1248 | hc = http_class(*args, **kwargs) |
be4a824d | 1249 | source_address = ydl_handler._params.get('source_address') |
8959018a | 1250 | |
be4a824d | 1251 | if source_address is not None: |
8959018a AU |
1252 | # This is to workaround _create_connection() from socket where it will try all |
1253 | # address data from getaddrinfo() including IPv6. This filters the result from | |
1254 | # getaddrinfo() based on the source_address value. | |
1255 | # This is based on the cpython socket.create_connection() function. | |
1256 | # https://github.com/python/cpython/blob/master/Lib/socket.py#L691 | |
1257 | def _create_connection(address, timeout=socket._GLOBAL_DEFAULT_TIMEOUT, source_address=None): | |
1258 | host, port = address | |
1259 | err = None | |
1260 | addrs = socket.getaddrinfo(host, port, 0, socket.SOCK_STREAM) | |
9e21e6d9 S |
1261 | af = socket.AF_INET if '.' in source_address[0] else socket.AF_INET6 |
1262 | ip_addrs = [addr for addr in addrs if addr[0] == af] | |
1263 | if addrs and not ip_addrs: | |
1264 | ip_version = 'v4' if af == socket.AF_INET else 'v6' | |
86e5f3ed | 1265 | raise OSError( |
9e21e6d9 S |
1266 | "No remote IP%s addresses available for connect, can't use '%s' as source address" |
1267 | % (ip_version, source_address[0])) | |
8959018a AU |
1268 | for res in ip_addrs: |
1269 | af, socktype, proto, canonname, sa = res | |
1270 | sock = None | |
1271 | try: | |
1272 | sock = socket.socket(af, socktype, proto) | |
1273 | if timeout is not socket._GLOBAL_DEFAULT_TIMEOUT: | |
1274 | sock.settimeout(timeout) | |
1275 | sock.bind(source_address) | |
1276 | sock.connect(sa) | |
1277 | err = None # Explicitly break reference cycle | |
1278 | return sock | |
86e5f3ed | 1279 | except OSError as _: |
8959018a AU |
1280 | err = _ |
1281 | if sock is not None: | |
1282 | sock.close() | |
1283 | if err is not None: | |
1284 | raise err | |
1285 | else: | |
86e5f3ed | 1286 | raise OSError('getaddrinfo returns an empty list') |
9e21e6d9 S |
1287 | if hasattr(hc, '_create_connection'): |
1288 | hc._create_connection = _create_connection | |
cfb0511d | 1289 | hc.source_address = (source_address, 0) |
be4a824d PH |
1290 | |
1291 | return hc | |
1292 | ||
1293 | ||
87f0e62d | 1294 | def handle_youtubedl_headers(headers): |
992fc9d6 YCH |
1295 | filtered_headers = headers |
1296 | ||
1297 | if 'Youtubedl-no-compression' in filtered_headers: | |
86e5f3ed | 1298 | filtered_headers = {k: v for k, v in filtered_headers.items() if k.lower() != 'accept-encoding'} |
87f0e62d | 1299 | del filtered_headers['Youtubedl-no-compression'] |
87f0e62d | 1300 | |
992fc9d6 | 1301 | return filtered_headers |
87f0e62d YCH |
1302 | |
1303 | ||
ac668111 | 1304 | class YoutubeDLHandler(urllib.request.HTTPHandler): |
59ae15a5 PH |
1305 | """Handler for HTTP requests and responses. |
1306 | ||
1307 | This class, when installed with an OpenerDirector, automatically adds | |
1308 | the standard headers to every HTTP request and handles gzipped and | |
1309 | deflated responses from web servers. If compression is to be avoided in | |
1310 | a particular request, the original request in the program code only has | |
0424ec30 | 1311 | to include the HTTP header "Youtubedl-no-compression", which will be |
59ae15a5 PH |
1312 | removed before making the real request. |
1313 | ||
1314 | Part of this code was copied from: | |
1315 | ||
1316 | http://techknack.net/python-urllib2-handlers/ | |
1317 | ||
1318 | Andrew Rowls, the author of that code, agreed to release it to the | |
1319 | public domain. | |
1320 | """ | |
1321 | ||
be4a824d | 1322 | def __init__(self, params, *args, **kwargs): |
ac668111 | 1323 | urllib.request.HTTPHandler.__init__(self, *args, **kwargs) |
be4a824d PH |
1324 | self._params = params |
1325 | ||
1326 | def http_open(self, req): | |
ac668111 | 1327 | conn_class = http.client.HTTPConnection |
71aff188 YCH |
1328 | |
1329 | socks_proxy = req.headers.get('Ytdl-socks-proxy') | |
1330 | if socks_proxy: | |
1331 | conn_class = make_socks_conn_class(conn_class, socks_proxy) | |
1332 | del req.headers['Ytdl-socks-proxy'] | |
1333 | ||
be4a824d | 1334 | return self.do_open(functools.partial( |
71aff188 | 1335 | _create_http_connection, self, conn_class, False), |
be4a824d PH |
1336 | req) |
1337 | ||
59ae15a5 PH |
1338 | @staticmethod |
1339 | def deflate(data): | |
fc2119f2 | 1340 | if not data: |
1341 | return data | |
59ae15a5 PH |
1342 | try: |
1343 | return zlib.decompress(data, -zlib.MAX_WBITS) | |
1344 | except zlib.error: | |
1345 | return zlib.decompress(data) | |
1346 | ||
4390d5ec | 1347 | @staticmethod |
1348 | def brotli(data): | |
1349 | if not data: | |
1350 | return data | |
9b8ee23b | 1351 | return brotli.decompress(data) |
4390d5ec | 1352 | |
acebc9cd | 1353 | def http_request(self, req): |
51f267d9 S |
1354 | # According to RFC 3986, URLs can not contain non-ASCII characters, however this is not |
1355 | # always respected by websites, some tend to give out URLs with non percent-encoded | |
1356 | # non-ASCII characters (see telemb.py, ard.py [#3412]) | |
1357 | # urllib chokes on URLs with non-ASCII characters (see http://bugs.python.org/issue3991) | |
1358 | # To work around aforementioned issue we will replace request's original URL with | |
1359 | # percent-encoded one | |
1360 | # Since redirects are also affected (e.g. http://www.southpark.de/alle-episoden/s18e09) | |
1361 | # the code of this workaround has been moved here from YoutubeDL.urlopen() | |
1362 | url = req.get_full_url() | |
1363 | url_escaped = escape_url(url) | |
1364 | ||
1365 | # Substitute URL if any change after escaping | |
1366 | if url != url_escaped: | |
15d260eb | 1367 | req = update_Request(req, url=url_escaped) |
51f267d9 | 1368 | |
8b7539d2 | 1369 | for h, v in self._params.get('http_headers', std_headers).items(): |
3d5f7a39 JK |
1370 | # Capitalize is needed because of Python bug 2275: http://bugs.python.org/issue2275 |
1371 | # The dict keys are capitalized because of this bug by urllib | |
1372 | if h.capitalize() not in req.headers: | |
33ac271b | 1373 | req.add_header(h, v) |
87f0e62d | 1374 | |
af14914b | 1375 | if 'Accept-encoding' not in req.headers: |
1376 | req.add_header('Accept-encoding', ', '.join(SUPPORTED_ENCODINGS)) | |
1377 | ||
87f0e62d | 1378 | req.headers = handle_youtubedl_headers(req.headers) |
989b4b2b | 1379 | |
379a4f16 | 1380 | return super().do_request_(req) |
59ae15a5 | 1381 | |
acebc9cd | 1382 | def http_response(self, req, resp): |
59ae15a5 PH |
1383 | old_resp = resp |
1384 | # gzip | |
1385 | if resp.headers.get('Content-encoding', '') == 'gzip': | |
aa3e9507 PH |
1386 | content = resp.read() |
1387 | gz = gzip.GzipFile(fileobj=io.BytesIO(content), mode='rb') | |
1388 | try: | |
1389 | uncompressed = io.BytesIO(gz.read()) | |
86e5f3ed | 1390 | except OSError as original_ioerror: |
aa3e9507 PH |
1391 | # There may be junk add the end of the file |
1392 | # See http://stackoverflow.com/q/4928560/35070 for details | |
1393 | for i in range(1, 1024): | |
1394 | try: | |
1395 | gz = gzip.GzipFile(fileobj=io.BytesIO(content[:-i]), mode='rb') | |
1396 | uncompressed = io.BytesIO(gz.read()) | |
86e5f3ed | 1397 | except OSError: |
aa3e9507 PH |
1398 | continue |
1399 | break | |
1400 | else: | |
1401 | raise original_ioerror | |
ac668111 | 1402 | resp = urllib.request.addinfourl(uncompressed, old_resp.headers, old_resp.url, old_resp.code) |
59ae15a5 | 1403 | resp.msg = old_resp.msg |
c047270c | 1404 | del resp.headers['Content-encoding'] |
59ae15a5 PH |
1405 | # deflate |
1406 | if resp.headers.get('Content-encoding', '') == 'deflate': | |
1407 | gz = io.BytesIO(self.deflate(resp.read())) | |
ac668111 | 1408 | resp = urllib.request.addinfourl(gz, old_resp.headers, old_resp.url, old_resp.code) |
59ae15a5 | 1409 | resp.msg = old_resp.msg |
c047270c | 1410 | del resp.headers['Content-encoding'] |
4390d5ec | 1411 | # brotli |
1412 | if resp.headers.get('Content-encoding', '') == 'br': | |
ac668111 | 1413 | resp = urllib.request.addinfourl( |
4390d5ec | 1414 | io.BytesIO(self.brotli(resp.read())), old_resp.headers, old_resp.url, old_resp.code) |
1415 | resp.msg = old_resp.msg | |
1416 | del resp.headers['Content-encoding'] | |
ad729172 | 1417 | # Percent-encode redirect URL of Location HTTP header to satisfy RFC 3986 (see |
067aa17e | 1418 | # https://github.com/ytdl-org/youtube-dl/issues/6457). |
5a4d9ddb S |
1419 | if 300 <= resp.code < 400: |
1420 | location = resp.headers.get('Location') | |
1421 | if location: | |
1422 | # As of RFC 2616 default charset is iso-8859-1 that is respected by python 3 | |
0f06bcd7 | 1423 | location = location.encode('iso-8859-1').decode() |
5a4d9ddb S |
1424 | location_escaped = escape_url(location) |
1425 | if location != location_escaped: | |
1426 | del resp.headers['Location'] | |
1427 | resp.headers['Location'] = location_escaped | |
59ae15a5 | 1428 | return resp |
0f8d03f8 | 1429 | |
acebc9cd PH |
1430 | https_request = http_request |
1431 | https_response = http_response | |
bf50b038 | 1432 | |
5de90176 | 1433 | |
71aff188 YCH |
1434 | def make_socks_conn_class(base_class, socks_proxy): |
1435 | assert issubclass(base_class, ( | |
ac668111 | 1436 | http.client.HTTPConnection, http.client.HTTPSConnection)) |
71aff188 | 1437 | |
14f25df2 | 1438 | url_components = urllib.parse.urlparse(socks_proxy) |
71aff188 YCH |
1439 | if url_components.scheme.lower() == 'socks5': |
1440 | socks_type = ProxyType.SOCKS5 | |
1441 | elif url_components.scheme.lower() in ('socks', 'socks4'): | |
1442 | socks_type = ProxyType.SOCKS4 | |
51fb4995 YCH |
1443 | elif url_components.scheme.lower() == 'socks4a': |
1444 | socks_type = ProxyType.SOCKS4A | |
71aff188 | 1445 | |
cdd94c2e YCH |
1446 | def unquote_if_non_empty(s): |
1447 | if not s: | |
1448 | return s | |
ac668111 | 1449 | return urllib.parse.unquote_plus(s) |
cdd94c2e | 1450 | |
71aff188 YCH |
1451 | proxy_args = ( |
1452 | socks_type, | |
1453 | url_components.hostname, url_components.port or 1080, | |
1454 | True, # Remote DNS | |
cdd94c2e YCH |
1455 | unquote_if_non_empty(url_components.username), |
1456 | unquote_if_non_empty(url_components.password), | |
71aff188 YCH |
1457 | ) |
1458 | ||
1459 | class SocksConnection(base_class): | |
1460 | def connect(self): | |
1461 | self.sock = sockssocket() | |
1462 | self.sock.setproxy(*proxy_args) | |
19a03940 | 1463 | if isinstance(self.timeout, (int, float)): |
71aff188 YCH |
1464 | self.sock.settimeout(self.timeout) |
1465 | self.sock.connect((self.host, self.port)) | |
1466 | ||
ac668111 | 1467 | if isinstance(self, http.client.HTTPSConnection): |
71aff188 YCH |
1468 | if hasattr(self, '_context'): # Python > 2.6 |
1469 | self.sock = self._context.wrap_socket( | |
1470 | self.sock, server_hostname=self.host) | |
1471 | else: | |
1472 | self.sock = ssl.wrap_socket(self.sock) | |
1473 | ||
1474 | return SocksConnection | |
1475 | ||
1476 | ||
ac668111 | 1477 | class YoutubeDLHTTPSHandler(urllib.request.HTTPSHandler): |
be4a824d | 1478 | def __init__(self, params, https_conn_class=None, *args, **kwargs): |
ac668111 | 1479 | urllib.request.HTTPSHandler.__init__(self, *args, **kwargs) |
1480 | self._https_conn_class = https_conn_class or http.client.HTTPSConnection | |
be4a824d PH |
1481 | self._params = params |
1482 | ||
1483 | def https_open(self, req): | |
4f264c02 | 1484 | kwargs = {} |
71aff188 YCH |
1485 | conn_class = self._https_conn_class |
1486 | ||
4f264c02 JMF |
1487 | if hasattr(self, '_context'): # python > 2.6 |
1488 | kwargs['context'] = self._context | |
1489 | if hasattr(self, '_check_hostname'): # python 3.x | |
1490 | kwargs['check_hostname'] = self._check_hostname | |
71aff188 YCH |
1491 | |
1492 | socks_proxy = req.headers.get('Ytdl-socks-proxy') | |
1493 | if socks_proxy: | |
1494 | conn_class = make_socks_conn_class(conn_class, socks_proxy) | |
1495 | del req.headers['Ytdl-socks-proxy'] | |
1496 | ||
4f28b537 | 1497 | try: |
1498 | return self.do_open( | |
1499 | functools.partial(_create_http_connection, self, conn_class, True), req, **kwargs) | |
1500 | except urllib.error.URLError as e: | |
1501 | if (isinstance(e.reason, ssl.SSLError) | |
1502 | and getattr(e.reason, 'reason', None) == 'SSLV3_ALERT_HANDSHAKE_FAILURE'): | |
1503 | raise YoutubeDLError('SSLV3_ALERT_HANDSHAKE_FAILURE: Try using --legacy-server-connect') | |
1504 | raise | |
be4a824d PH |
1505 | |
1506 | ||
941e881e | 1507 | def is_path_like(f): |
1508 | return isinstance(f, (str, bytes, os.PathLike)) | |
1509 | ||
1510 | ||
ac668111 | 1511 | class YoutubeDLCookieJar(http.cookiejar.MozillaCookieJar): |
f1a8511f S |
1512 | """ |
1513 | See [1] for cookie file format. | |
1514 | ||
1515 | 1. https://curl.haxx.se/docs/http-cookies.html | |
1516 | """ | |
e7e62441 | 1517 | _HTTPONLY_PREFIX = '#HttpOnly_' |
c380cc28 S |
1518 | _ENTRY_LEN = 7 |
1519 | _HEADER = '''# Netscape HTTP Cookie File | |
7a5c1cfe | 1520 | # This file is generated by yt-dlp. Do not edit. |
c380cc28 S |
1521 | |
1522 | ''' | |
1523 | _CookieFileEntry = collections.namedtuple( | |
1524 | 'CookieFileEntry', | |
1525 | ('domain_name', 'include_subdomains', 'path', 'https_only', 'expires_at', 'name', 'value')) | |
e7e62441 | 1526 | |
d76fa1f3 | 1527 | def __init__(self, filename=None, *args, **kwargs): |
1528 | super().__init__(None, *args, **kwargs) | |
941e881e | 1529 | if is_path_like(filename): |
d76fa1f3 | 1530 | filename = os.fspath(filename) |
1531 | self.filename = filename | |
1532 | ||
24146491 | 1533 | @staticmethod |
1534 | def _true_or_false(cndn): | |
1535 | return 'TRUE' if cndn else 'FALSE' | |
1536 | ||
d76fa1f3 | 1537 | @contextlib.contextmanager |
1538 | def open(self, file, *, write=False): | |
941e881e | 1539 | if is_path_like(file): |
d76fa1f3 | 1540 | with open(file, 'w' if write else 'r', encoding='utf-8') as f: |
1541 | yield f | |
1542 | else: | |
1543 | if write: | |
1544 | file.truncate(0) | |
1545 | yield file | |
1546 | ||
24146491 | 1547 | def _really_save(self, f, ignore_discard=False, ignore_expires=False): |
1548 | now = time.time() | |
1549 | for cookie in self: | |
1550 | if (not ignore_discard and cookie.discard | |
1551 | or not ignore_expires and cookie.is_expired(now)): | |
1552 | continue | |
1553 | name, value = cookie.name, cookie.value | |
1554 | if value is None: | |
1555 | # cookies.txt regards 'Set-Cookie: foo' as a cookie | |
1556 | # with no name, whereas http.cookiejar regards it as a | |
1557 | # cookie with no value. | |
1558 | name, value = '', name | |
1559 | f.write('%s\n' % '\t'.join(( | |
1560 | cookie.domain, | |
1561 | self._true_or_false(cookie.domain.startswith('.')), | |
1562 | cookie.path, | |
1563 | self._true_or_false(cookie.secure), | |
1564 | str_or_none(cookie.expires, default=''), | |
1565 | name, value | |
1566 | ))) | |
1567 | ||
1568 | def save(self, filename=None, *args, **kwargs): | |
c380cc28 S |
1569 | """ |
1570 | Save cookies to a file. | |
24146491 | 1571 | Code is taken from CPython 3.6 |
1572 | https://github.com/python/cpython/blob/8d999cbf4adea053be6dbb612b9844635c4dfb8e/Lib/http/cookiejar.py#L2091-L2117 """ | |
c380cc28 | 1573 | |
c380cc28 S |
1574 | if filename is None: |
1575 | if self.filename is not None: | |
1576 | filename = self.filename | |
1577 | else: | |
ac668111 | 1578 | raise ValueError(http.cookiejar.MISSING_FILENAME_TEXT) |
c380cc28 | 1579 | |
24146491 | 1580 | # Store session cookies with `expires` set to 0 instead of an empty string |
1bab3437 S |
1581 | for cookie in self: |
1582 | if cookie.expires is None: | |
1583 | cookie.expires = 0 | |
c380cc28 | 1584 | |
d76fa1f3 | 1585 | with self.open(filename, write=True) as f: |
c380cc28 | 1586 | f.write(self._HEADER) |
24146491 | 1587 | self._really_save(f, *args, **kwargs) |
1bab3437 S |
1588 | |
1589 | def load(self, filename=None, ignore_discard=False, ignore_expires=False): | |
e7e62441 | 1590 | """Load cookies from a file.""" |
1591 | if filename is None: | |
1592 | if self.filename is not None: | |
1593 | filename = self.filename | |
1594 | else: | |
ac668111 | 1595 | raise ValueError(http.cookiejar.MISSING_FILENAME_TEXT) |
e7e62441 | 1596 | |
c380cc28 S |
1597 | def prepare_line(line): |
1598 | if line.startswith(self._HTTPONLY_PREFIX): | |
1599 | line = line[len(self._HTTPONLY_PREFIX):] | |
1600 | # comments and empty lines are fine | |
1601 | if line.startswith('#') or not line.strip(): | |
1602 | return line | |
1603 | cookie_list = line.split('\t') | |
1604 | if len(cookie_list) != self._ENTRY_LEN: | |
ac668111 | 1605 | raise http.cookiejar.LoadError('invalid length %d' % len(cookie_list)) |
c380cc28 S |
1606 | cookie = self._CookieFileEntry(*cookie_list) |
1607 | if cookie.expires_at and not cookie.expires_at.isdigit(): | |
ac668111 | 1608 | raise http.cookiejar.LoadError('invalid expires at %s' % cookie.expires_at) |
c380cc28 S |
1609 | return line |
1610 | ||
e7e62441 | 1611 | cf = io.StringIO() |
d76fa1f3 | 1612 | with self.open(filename) as f: |
e7e62441 | 1613 | for line in f: |
c380cc28 S |
1614 | try: |
1615 | cf.write(prepare_line(line)) | |
ac668111 | 1616 | except http.cookiejar.LoadError as e: |
94aa0644 | 1617 | if f'{line.strip()} '[0] in '[{"': |
ac668111 | 1618 | raise http.cookiejar.LoadError( |
94aa0644 | 1619 | 'Cookies file must be Netscape formatted, not JSON. See ' |
17ffed18 | 1620 | 'https://github.com/yt-dlp/yt-dlp/wiki/FAQ#how-do-i-pass-cookies-to-yt-dlp') |
19a03940 | 1621 | write_string(f'WARNING: skipping cookie file entry due to {e}: {line!r}\n') |
c380cc28 | 1622 | continue |
e7e62441 | 1623 | cf.seek(0) |
1624 | self._really_load(cf, filename, ignore_discard, ignore_expires) | |
1bab3437 S |
1625 | # Session cookies are denoted by either `expires` field set to |
1626 | # an empty string or 0. MozillaCookieJar only recognizes the former | |
1627 | # (see [1]). So we need force the latter to be recognized as session | |
1628 | # cookies on our own. | |
1629 | # Session cookies may be important for cookies-based authentication, | |
1630 | # e.g. usually, when user does not check 'Remember me' check box while | |
1631 | # logging in on a site, some important cookies are stored as session | |
1632 | # cookies so that not recognizing them will result in failed login. | |
1633 | # 1. https://bugs.python.org/issue17164 | |
1634 | for cookie in self: | |
1635 | # Treat `expires=0` cookies as session cookies | |
1636 | if cookie.expires == 0: | |
1637 | cookie.expires = None | |
1638 | cookie.discard = True | |
1639 | ||
1640 | ||
ac668111 | 1641 | class YoutubeDLCookieProcessor(urllib.request.HTTPCookieProcessor): |
a6420bf5 | 1642 | def __init__(self, cookiejar=None): |
ac668111 | 1643 | urllib.request.HTTPCookieProcessor.__init__(self, cookiejar) |
a6420bf5 S |
1644 | |
1645 | def http_response(self, request, response): | |
ac668111 | 1646 | return urllib.request.HTTPCookieProcessor.http_response(self, request, response) |
a6420bf5 | 1647 | |
ac668111 | 1648 | https_request = urllib.request.HTTPCookieProcessor.http_request |
a6420bf5 S |
1649 | https_response = http_response |
1650 | ||
1651 | ||
ac668111 | 1652 | class YoutubeDLRedirectHandler(urllib.request.HTTPRedirectHandler): |
201c1459 | 1653 | """YoutubeDL redirect handler |
1654 | ||
1655 | The code is based on HTTPRedirectHandler implementation from CPython [1]. | |
1656 | ||
1657 | This redirect handler solves two issues: | |
1658 | - ensures redirect URL is always unicode under python 2 | |
1659 | - introduces support for experimental HTTP response status code | |
1660 | 308 Permanent Redirect [2] used by some sites [3] | |
1661 | ||
1662 | 1. https://github.com/python/cpython/blob/master/Lib/urllib/request.py | |
1663 | 2. https://developer.mozilla.org/en-US/docs/Web/HTTP/Status/308 | |
1664 | 3. https://github.com/ytdl-org/youtube-dl/issues/28768 | |
1665 | """ | |
1666 | ||
ac668111 | 1667 | http_error_301 = http_error_303 = http_error_307 = http_error_308 = urllib.request.HTTPRedirectHandler.http_error_302 |
201c1459 | 1668 | |
1669 | def redirect_request(self, req, fp, code, msg, headers, newurl): | |
1670 | """Return a Request or None in response to a redirect. | |
1671 | ||
1672 | This is called by the http_error_30x methods when a | |
1673 | redirection response is received. If a redirection should | |
1674 | take place, return a new Request to allow http_error_30x to | |
1675 | perform the redirect. Otherwise, raise HTTPError if no-one | |
1676 | else should try to handle this url. Return None if you can't | |
1677 | but another Handler might. | |
1678 | """ | |
1679 | m = req.get_method() | |
1680 | if (not (code in (301, 302, 303, 307, 308) and m in ("GET", "HEAD") | |
1681 | or code in (301, 302, 303) and m == "POST")): | |
14f25df2 | 1682 | raise urllib.error.HTTPError(req.full_url, code, msg, headers, fp) |
201c1459 | 1683 | # Strictly (according to RFC 2616), 301 or 302 in response to |
1684 | # a POST MUST NOT cause a redirection without confirmation | |
1685 | # from the user (of urllib.request, in this case). In practice, | |
1686 | # essentially all clients do redirect in this case, so we do | |
1687 | # the same. | |
1688 | ||
201c1459 | 1689 | # Be conciliant with URIs containing a space. This is mainly |
1690 | # redundant with the more complete encoding done in http_error_302(), | |
1691 | # but it is kept for compatibility with other callers. | |
1692 | newurl = newurl.replace(' ', '%20') | |
1693 | ||
1694 | CONTENT_HEADERS = ("content-length", "content-type") | |
1695 | # NB: don't use dict comprehension for python 2.6 compatibility | |
86e5f3ed | 1696 | newheaders = {k: v for k, v in req.headers.items() if k.lower() not in CONTENT_HEADERS} |
afac4caa | 1697 | |
1698 | # A 303 must either use GET or HEAD for subsequent request | |
1699 | # https://datatracker.ietf.org/doc/html/rfc7231#section-6.4.4 | |
1700 | if code == 303 and m != 'HEAD': | |
1701 | m = 'GET' | |
1702 | # 301 and 302 redirects are commonly turned into a GET from a POST | |
1703 | # for subsequent requests by browsers, so we'll do the same. | |
1704 | # https://datatracker.ietf.org/doc/html/rfc7231#section-6.4.2 | |
1705 | # https://datatracker.ietf.org/doc/html/rfc7231#section-6.4.3 | |
1706 | if code in (301, 302) and m == 'POST': | |
1707 | m = 'GET' | |
1708 | ||
ac668111 | 1709 | return urllib.request.Request( |
201c1459 | 1710 | newurl, headers=newheaders, origin_req_host=req.origin_req_host, |
afac4caa | 1711 | unverifiable=True, method=m) |
fca6dba8 S |
1712 | |
1713 | ||
46f59e89 S |
1714 | def extract_timezone(date_str): |
1715 | m = re.search( | |
f137e4c2 | 1716 | r'''(?x) |
1717 | ^.{8,}? # >=8 char non-TZ prefix, if present | |
1718 | (?P<tz>Z| # just the UTC Z, or | |
1719 | (?:(?<=.\b\d{4}|\b\d{2}:\d\d)| # preceded by 4 digits or hh:mm or | |
1720 | (?<!.\b[a-zA-Z]{3}|[a-zA-Z]{4}|..\b\d\d)) # not preceded by 3 alpha word or >= 4 alpha or 2 digits | |
1721 | [ ]? # optional space | |
1722 | (?P<sign>\+|-) # +/- | |
1723 | (?P<hours>[0-9]{2}):?(?P<minutes>[0-9]{2}) # hh[:]mm | |
1724 | $) | |
1725 | ''', date_str) | |
46f59e89 | 1726 | if not m: |
8f53dc44 | 1727 | m = re.search(r'\d{1,2}:\d{1,2}(?:\.\d+)?(?P<tz>\s*[A-Z]+)$', date_str) |
1728 | timezone = TIMEZONE_NAMES.get(m and m.group('tz').strip()) | |
1729 | if timezone is not None: | |
1730 | date_str = date_str[:-len(m.group('tz'))] | |
1731 | timezone = datetime.timedelta(hours=timezone or 0) | |
46f59e89 S |
1732 | else: |
1733 | date_str = date_str[:-len(m.group('tz'))] | |
1734 | if not m.group('sign'): | |
1735 | timezone = datetime.timedelta() | |
1736 | else: | |
1737 | sign = 1 if m.group('sign') == '+' else -1 | |
1738 | timezone = datetime.timedelta( | |
1739 | hours=sign * int(m.group('hours')), | |
1740 | minutes=sign * int(m.group('minutes'))) | |
1741 | return timezone, date_str | |
1742 | ||
1743 | ||
08b38d54 | 1744 | def parse_iso8601(date_str, delimiter='T', timezone=None): |
912b38b4 PH |
1745 | """ Return a UNIX timestamp from the given date """ |
1746 | ||
1747 | if date_str is None: | |
1748 | return None | |
1749 | ||
52c3a6e4 S |
1750 | date_str = re.sub(r'\.[0-9]+', '', date_str) |
1751 | ||
08b38d54 | 1752 | if timezone is None: |
46f59e89 S |
1753 | timezone, date_str = extract_timezone(date_str) |
1754 | ||
19a03940 | 1755 | with contextlib.suppress(ValueError): |
86e5f3ed | 1756 | date_format = f'%Y-%m-%d{delimiter}%H:%M:%S' |
52c3a6e4 S |
1757 | dt = datetime.datetime.strptime(date_str, date_format) - timezone |
1758 | return calendar.timegm(dt.timetuple()) | |
912b38b4 PH |
1759 | |
1760 | ||
46f59e89 S |
1761 | def date_formats(day_first=True): |
1762 | return DATE_FORMATS_DAY_FIRST if day_first else DATE_FORMATS_MONTH_FIRST | |
1763 | ||
1764 | ||
42bdd9d0 | 1765 | def unified_strdate(date_str, day_first=True): |
bf50b038 | 1766 | """Return a string with the date in the format YYYYMMDD""" |
64e7ad60 PH |
1767 | |
1768 | if date_str is None: | |
1769 | return None | |
bf50b038 | 1770 | upload_date = None |
5f6a1245 | 1771 | # Replace commas |
026fcc04 | 1772 | date_str = date_str.replace(',', ' ') |
42bdd9d0 | 1773 | # Remove AM/PM + timezone |
9bb8e0a3 | 1774 | date_str = re.sub(r'(?i)\s*(?:AM|PM)(?:\s+[A-Z]+)?', '', date_str) |
46f59e89 | 1775 | _, date_str = extract_timezone(date_str) |
42bdd9d0 | 1776 | |
46f59e89 | 1777 | for expression in date_formats(day_first): |
19a03940 | 1778 | with contextlib.suppress(ValueError): |
bf50b038 | 1779 | upload_date = datetime.datetime.strptime(date_str, expression).strftime('%Y%m%d') |
42393ce2 PH |
1780 | if upload_date is None: |
1781 | timetuple = email.utils.parsedate_tz(date_str) | |
1782 | if timetuple: | |
19a03940 | 1783 | with contextlib.suppress(ValueError): |
c6b9cf05 | 1784 | upload_date = datetime.datetime(*timetuple[:6]).strftime('%Y%m%d') |
6a750402 | 1785 | if upload_date is not None: |
14f25df2 | 1786 | return str(upload_date) |
bf50b038 | 1787 | |
5f6a1245 | 1788 | |
46f59e89 S |
1789 | def unified_timestamp(date_str, day_first=True): |
1790 | if date_str is None: | |
1791 | return None | |
1792 | ||
8f53dc44 | 1793 | date_str = re.sub(r'\s+', ' ', re.sub( |
1794 | r'(?i)[,|]|(mon|tues?|wed(nes)?|thu(rs)?|fri|sat(ur)?)(day)?', '', date_str)) | |
46f59e89 | 1795 | |
7dc2a74e | 1796 | pm_delta = 12 if re.search(r'(?i)PM', date_str) else 0 |
46f59e89 S |
1797 | timezone, date_str = extract_timezone(date_str) |
1798 | ||
1799 | # Remove AM/PM + timezone | |
1800 | date_str = re.sub(r'(?i)\s*(?:AM|PM)(?:\s+[A-Z]+)?', '', date_str) | |
1801 | ||
deef3195 S |
1802 | # Remove unrecognized timezones from ISO 8601 alike timestamps |
1803 | m = re.search(r'\d{1,2}:\d{1,2}(?:\.\d+)?(?P<tz>\s*[A-Z]+)$', date_str) | |
1804 | if m: | |
1805 | date_str = date_str[:-len(m.group('tz'))] | |
1806 | ||
f226880c PH |
1807 | # Python only supports microseconds, so remove nanoseconds |
1808 | m = re.search(r'^([0-9]{4,}-[0-9]{1,2}-[0-9]{1,2}T[0-9]{1,2}:[0-9]{1,2}:[0-9]{1,2}\.[0-9]{6})[0-9]+$', date_str) | |
1809 | if m: | |
1810 | date_str = m.group(1) | |
1811 | ||
46f59e89 | 1812 | for expression in date_formats(day_first): |
19a03940 | 1813 | with contextlib.suppress(ValueError): |
7dc2a74e | 1814 | dt = datetime.datetime.strptime(date_str, expression) - timezone + datetime.timedelta(hours=pm_delta) |
46f59e89 | 1815 | return calendar.timegm(dt.timetuple()) |
8f53dc44 | 1816 | |
46f59e89 S |
1817 | timetuple = email.utils.parsedate_tz(date_str) |
1818 | if timetuple: | |
8f53dc44 | 1819 | return calendar.timegm(timetuple) + pm_delta * 3600 - timezone.total_seconds() |
46f59e89 S |
1820 | |
1821 | ||
28e614de | 1822 | def determine_ext(url, default_ext='unknown_video'): |
85750f89 | 1823 | if url is None or '.' not in url: |
f4776371 | 1824 | return default_ext |
9cb9a5df | 1825 | guess = url.partition('?')[0].rpartition('.')[2] |
73e79f2a PH |
1826 | if re.match(r'^[A-Za-z0-9]+$', guess): |
1827 | return guess | |
a7aaa398 S |
1828 | # Try extract ext from URLs like http://example.com/foo/bar.mp4/?download |
1829 | elif guess.rstrip('/') in KNOWN_EXTENSIONS: | |
9cb9a5df | 1830 | return guess.rstrip('/') |
73e79f2a | 1831 | else: |
cbdbb766 | 1832 | return default_ext |
73e79f2a | 1833 | |
5f6a1245 | 1834 | |
824fa511 S |
1835 | def subtitles_filename(filename, sub_lang, sub_format, expected_real_ext=None): |
1836 | return replace_extension(filename, sub_lang + '.' + sub_format, expected_real_ext) | |
d4051a8e | 1837 | |
5f6a1245 | 1838 | |
9e62f283 | 1839 | def datetime_from_str(date_str, precision='auto', format='%Y%m%d'): |
3d38b2d6 | 1840 | R""" |
1841 | Return a datetime object from a string. | |
1842 | Supported format: | |
1843 | (now|today|yesterday|DATE)([+-]\d+(microsecond|second|minute|hour|day|week|month|year)s?)? | |
1844 | ||
1845 | @param format strftime format of DATE | |
1846 | @param precision Round the datetime object: auto|microsecond|second|minute|hour|day | |
1847 | auto: round to the unit provided in date_str (if applicable). | |
9e62f283 | 1848 | """ |
1849 | auto_precision = False | |
1850 | if precision == 'auto': | |
1851 | auto_precision = True | |
1852 | precision = 'microsecond' | |
396a76f7 | 1853 | today = datetime_round(datetime.datetime.utcnow(), precision) |
f8795e10 | 1854 | if date_str in ('now', 'today'): |
37254abc | 1855 | return today |
f8795e10 PH |
1856 | if date_str == 'yesterday': |
1857 | return today - datetime.timedelta(days=1) | |
9e62f283 | 1858 | match = re.match( |
3d38b2d6 | 1859 | r'(?P<start>.+)(?P<sign>[+-])(?P<time>\d+)(?P<unit>microsecond|second|minute|hour|day|week|month|year)s?', |
9e62f283 | 1860 | date_str) |
37254abc | 1861 | if match is not None: |
9e62f283 | 1862 | start_time = datetime_from_str(match.group('start'), precision, format) |
1863 | time = int(match.group('time')) * (-1 if match.group('sign') == '-' else 1) | |
37254abc | 1864 | unit = match.group('unit') |
9e62f283 | 1865 | if unit == 'month' or unit == 'year': |
1866 | new_date = datetime_add_months(start_time, time * 12 if unit == 'year' else time) | |
37254abc | 1867 | unit = 'day' |
9e62f283 | 1868 | else: |
1869 | if unit == 'week': | |
1870 | unit = 'day' | |
1871 | time *= 7 | |
1872 | delta = datetime.timedelta(**{unit + 's': time}) | |
1873 | new_date = start_time + delta | |
1874 | if auto_precision: | |
1875 | return datetime_round(new_date, unit) | |
1876 | return new_date | |
1877 | ||
1878 | return datetime_round(datetime.datetime.strptime(date_str, format), precision) | |
1879 | ||
1880 | ||
d49f8db3 | 1881 | def date_from_str(date_str, format='%Y%m%d', strict=False): |
3d38b2d6 | 1882 | R""" |
1883 | Return a date object from a string using datetime_from_str | |
9e62f283 | 1884 | |
3d38b2d6 | 1885 | @param strict Restrict allowed patterns to "YYYYMMDD" and |
1886 | (now|today|yesterday)(-\d+(day|week|month|year)s?)? | |
9e62f283 | 1887 | """ |
3d38b2d6 | 1888 | if strict and not re.fullmatch(r'\d{8}|(now|today|yesterday)(-\d+(day|week|month|year)s?)?', date_str): |
1889 | raise ValueError(f'Invalid date format "{date_str}"') | |
9e62f283 | 1890 | return datetime_from_str(date_str, precision='microsecond', format=format).date() |
1891 | ||
1892 | ||
1893 | def datetime_add_months(dt, months): | |
1894 | """Increment/Decrement a datetime object by months.""" | |
1895 | month = dt.month + months - 1 | |
1896 | year = dt.year + month // 12 | |
1897 | month = month % 12 + 1 | |
1898 | day = min(dt.day, calendar.monthrange(year, month)[1]) | |
1899 | return dt.replace(year, month, day) | |
1900 | ||
1901 | ||
1902 | def datetime_round(dt, precision='day'): | |
1903 | """ | |
1904 | Round a datetime object's time to a specific precision | |
1905 | """ | |
1906 | if precision == 'microsecond': | |
1907 | return dt | |
1908 | ||
1909 | unit_seconds = { | |
1910 | 'day': 86400, | |
1911 | 'hour': 3600, | |
1912 | 'minute': 60, | |
1913 | 'second': 1, | |
1914 | } | |
1915 | roundto = lambda x, n: ((x + n / 2) // n) * n | |
1916 | timestamp = calendar.timegm(dt.timetuple()) | |
1917 | return datetime.datetime.utcfromtimestamp(roundto(timestamp, unit_seconds[precision])) | |
5f6a1245 JW |
1918 | |
1919 | ||
e63fc1be | 1920 | def hyphenate_date(date_str): |
1921 | """ | |
1922 | Convert a date in 'YYYYMMDD' format to 'YYYY-MM-DD' format""" | |
1923 | match = re.match(r'^(\d\d\d\d)(\d\d)(\d\d)$', date_str) | |
1924 | if match is not None: | |
1925 | return '-'.join(match.groups()) | |
1926 | else: | |
1927 | return date_str | |
1928 | ||
5f6a1245 | 1929 | |
86e5f3ed | 1930 | class DateRange: |
bd558525 | 1931 | """Represents a time interval between two dates""" |
5f6a1245 | 1932 | |
bd558525 JMF |
1933 | def __init__(self, start=None, end=None): |
1934 | """start and end must be strings in the format accepted by date""" | |
1935 | if start is not None: | |
d49f8db3 | 1936 | self.start = date_from_str(start, strict=True) |
bd558525 JMF |
1937 | else: |
1938 | self.start = datetime.datetime.min.date() | |
1939 | if end is not None: | |
d49f8db3 | 1940 | self.end = date_from_str(end, strict=True) |
bd558525 JMF |
1941 | else: |
1942 | self.end = datetime.datetime.max.date() | |
37254abc | 1943 | if self.start > self.end: |
bd558525 | 1944 | raise ValueError('Date range: "%s" , the start date must be before the end date' % self) |
5f6a1245 | 1945 | |
bd558525 JMF |
1946 | @classmethod |
1947 | def day(cls, day): | |
1948 | """Returns a range that only contains the given day""" | |
5f6a1245 JW |
1949 | return cls(day, day) |
1950 | ||
bd558525 JMF |
1951 | def __contains__(self, date): |
1952 | """Check if the date is in the range""" | |
37254abc JMF |
1953 | if not isinstance(date, datetime.date): |
1954 | date = date_from_str(date) | |
1955 | return self.start <= date <= self.end | |
5f6a1245 | 1956 | |
bd558525 | 1957 | def __str__(self): |
86e5f3ed | 1958 | return f'{self.start.isoformat()} - {self.end.isoformat()}' |
c496ca96 | 1959 | |
f2df4071 | 1960 | def __eq__(self, other): |
1961 | return (isinstance(other, DateRange) | |
1962 | and self.start == other.start and self.end == other.end) | |
1963 | ||
c496ca96 PH |
1964 | |
1965 | def platform_name(): | |
14f25df2 | 1966 | """ Returns the platform name as a str """ |
da4db748 | 1967 | deprecation_warning(f'"{__name__}.platform_name" is deprecated, use "platform.platform" instead') |
b1f94422 | 1968 | return platform.platform() |
c496ca96 | 1969 | |
b1f94422 | 1970 | |
1971 | @functools.cache | |
1972 | def system_identifier(): | |
1973 | python_implementation = platform.python_implementation() | |
1974 | if python_implementation == 'PyPy' and hasattr(sys, 'pypy_version_info'): | |
1975 | python_implementation += ' version %d.%d.%d' % sys.pypy_version_info[:3] | |
dab284f8 | 1976 | libc_ver = [] |
1977 | with contextlib.suppress(OSError): # We may not have access to the executable | |
1978 | libc_ver = platform.libc_ver() | |
b1f94422 | 1979 | |
1980 | return 'Python %s (%s %s) - %s %s' % ( | |
1981 | platform.python_version(), | |
1982 | python_implementation, | |
1983 | platform.architecture()[0], | |
1984 | platform.platform(), | |
dab284f8 | 1985 | format_field(join_nonempty(*libc_ver, delim=' '), None, '(%s)'), |
b1f94422 | 1986 | ) |
c257baff PH |
1987 | |
1988 | ||
0b9c08b4 | 1989 | @functools.cache |
49fa4d9a | 1990 | def get_windows_version(): |
8a82af35 | 1991 | ''' Get Windows version. returns () if it's not running on Windows ''' |
49fa4d9a N |
1992 | if compat_os_name == 'nt': |
1993 | return version_tuple(platform.win32_ver()[1]) | |
1994 | else: | |
8a82af35 | 1995 | return () |
49fa4d9a N |
1996 | |
1997 | ||
734f90bb | 1998 | def write_string(s, out=None, encoding=None): |
19a03940 | 1999 | assert isinstance(s, str) |
2000 | out = out or sys.stderr | |
7459e3a2 | 2001 | |
fe1daad3 | 2002 | if compat_os_name == 'nt' and supports_terminal_sequences(out): |
3fe75fdc | 2003 | s = re.sub(r'([\r\n]+)', r' \1', s) |
59f943cd | 2004 | |
8a82af35 | 2005 | enc, buffer = None, out |
cfb0511d | 2006 | if 'b' in getattr(out, 'mode', ''): |
c487cf00 | 2007 | enc = encoding or preferredencoding() |
104aa738 | 2008 | elif hasattr(out, 'buffer'): |
8a82af35 | 2009 | buffer = out.buffer |
104aa738 | 2010 | enc = encoding or getattr(out, 'encoding', None) or preferredencoding() |
c487cf00 | 2011 | |
8a82af35 | 2012 | buffer.write(s.encode(enc, 'ignore') if enc else s) |
7459e3a2 PH |
2013 | out.flush() |
2014 | ||
2015 | ||
da4db748 | 2016 | def deprecation_warning(msg, *, printer=None, stacklevel=0, **kwargs): |
2017 | from . import _IN_CLI | |
2018 | if _IN_CLI: | |
2019 | if msg in deprecation_warning._cache: | |
2020 | return | |
2021 | deprecation_warning._cache.add(msg) | |
2022 | if printer: | |
2023 | return printer(f'{msg}{bug_reports_message()}', **kwargs) | |
2024 | return write_string(f'ERROR: {msg}{bug_reports_message()}\n', **kwargs) | |
2025 | else: | |
2026 | import warnings | |
2027 | warnings.warn(DeprecationWarning(msg), stacklevel=stacklevel + 3) | |
2028 | ||
2029 | ||
2030 | deprecation_warning._cache = set() | |
2031 | ||
2032 | ||
48ea9cea PH |
2033 | def bytes_to_intlist(bs): |
2034 | if not bs: | |
2035 | return [] | |
2036 | if isinstance(bs[0], int): # Python 3 | |
2037 | return list(bs) | |
2038 | else: | |
2039 | return [ord(c) for c in bs] | |
2040 | ||
c257baff | 2041 | |
cba892fa | 2042 | def intlist_to_bytes(xs): |
2043 | if not xs: | |
2044 | return b'' | |
ac668111 | 2045 | return struct.pack('%dB' % len(xs), *xs) |
c38b1e77 PH |
2046 | |
2047 | ||
8a82af35 | 2048 | class LockingUnsupportedError(OSError): |
1890fc63 | 2049 | msg = 'File locking is not supported' |
0edb3e33 | 2050 | |
2051 | def __init__(self): | |
2052 | super().__init__(self.msg) | |
2053 | ||
2054 | ||
c1c9a79c PH |
2055 | # Cross-platform file locking |
2056 | if sys.platform == 'win32': | |
fe0918bb | 2057 | import ctypes |
c1c9a79c PH |
2058 | import ctypes.wintypes |
2059 | import msvcrt | |
2060 | ||
2061 | class OVERLAPPED(ctypes.Structure): | |
2062 | _fields_ = [ | |
2063 | ('Internal', ctypes.wintypes.LPVOID), | |
2064 | ('InternalHigh', ctypes.wintypes.LPVOID), | |
2065 | ('Offset', ctypes.wintypes.DWORD), | |
2066 | ('OffsetHigh', ctypes.wintypes.DWORD), | |
2067 | ('hEvent', ctypes.wintypes.HANDLE), | |
2068 | ] | |
2069 | ||
2070 | kernel32 = ctypes.windll.kernel32 | |
2071 | LockFileEx = kernel32.LockFileEx | |
2072 | LockFileEx.argtypes = [ | |
2073 | ctypes.wintypes.HANDLE, # hFile | |
2074 | ctypes.wintypes.DWORD, # dwFlags | |
2075 | ctypes.wintypes.DWORD, # dwReserved | |
2076 | ctypes.wintypes.DWORD, # nNumberOfBytesToLockLow | |
2077 | ctypes.wintypes.DWORD, # nNumberOfBytesToLockHigh | |
2078 | ctypes.POINTER(OVERLAPPED) # Overlapped | |
2079 | ] | |
2080 | LockFileEx.restype = ctypes.wintypes.BOOL | |
2081 | UnlockFileEx = kernel32.UnlockFileEx | |
2082 | UnlockFileEx.argtypes = [ | |
2083 | ctypes.wintypes.HANDLE, # hFile | |
2084 | ctypes.wintypes.DWORD, # dwReserved | |
2085 | ctypes.wintypes.DWORD, # nNumberOfBytesToLockLow | |
2086 | ctypes.wintypes.DWORD, # nNumberOfBytesToLockHigh | |
2087 | ctypes.POINTER(OVERLAPPED) # Overlapped | |
2088 | ] | |
2089 | UnlockFileEx.restype = ctypes.wintypes.BOOL | |
2090 | whole_low = 0xffffffff | |
2091 | whole_high = 0x7fffffff | |
2092 | ||
747c0bd1 | 2093 | def _lock_file(f, exclusive, block): |
c1c9a79c PH |
2094 | overlapped = OVERLAPPED() |
2095 | overlapped.Offset = 0 | |
2096 | overlapped.OffsetHigh = 0 | |
2097 | overlapped.hEvent = 0 | |
2098 | f._lock_file_overlapped_p = ctypes.pointer(overlapped) | |
747c0bd1 | 2099 | |
2100 | if not LockFileEx(msvcrt.get_osfhandle(f.fileno()), | |
2101 | (0x2 if exclusive else 0x0) | (0x0 if block else 0x1), | |
2102 | 0, whole_low, whole_high, f._lock_file_overlapped_p): | |
2cb19820 | 2103 | # NB: No argument form of "ctypes.FormatError" does not work on PyPy |
2104 | raise BlockingIOError(f'Locking file failed: {ctypes.FormatError(ctypes.GetLastError())!r}') | |
c1c9a79c PH |
2105 | |
2106 | def _unlock_file(f): | |
2107 | assert f._lock_file_overlapped_p | |
2108 | handle = msvcrt.get_osfhandle(f.fileno()) | |
747c0bd1 | 2109 | if not UnlockFileEx(handle, 0, whole_low, whole_high, f._lock_file_overlapped_p): |
c1c9a79c PH |
2110 | raise OSError('Unlocking file failed: %r' % ctypes.FormatError()) |
2111 | ||
2112 | else: | |
399a76e6 YCH |
2113 | try: |
2114 | import fcntl | |
c1c9a79c | 2115 | |
a3125791 | 2116 | def _lock_file(f, exclusive, block): |
b63837bc | 2117 | flags = fcntl.LOCK_EX if exclusive else fcntl.LOCK_SH |
2118 | if not block: | |
2119 | flags |= fcntl.LOCK_NB | |
acea8d7c | 2120 | try: |
b63837bc | 2121 | fcntl.flock(f, flags) |
acea8d7c JK |
2122 | except BlockingIOError: |
2123 | raise | |
2124 | except OSError: # AOSP does not have flock() | |
b63837bc | 2125 | fcntl.lockf(f, flags) |
c1c9a79c | 2126 | |
399a76e6 | 2127 | def _unlock_file(f): |
acea8d7c JK |
2128 | try: |
2129 | fcntl.flock(f, fcntl.LOCK_UN) | |
2130 | except OSError: | |
2131 | fcntl.lockf(f, fcntl.LOCK_UN) | |
a3125791 | 2132 | |
399a76e6 | 2133 | except ImportError: |
399a76e6 | 2134 | |
a3125791 | 2135 | def _lock_file(f, exclusive, block): |
0edb3e33 | 2136 | raise LockingUnsupportedError() |
399a76e6 YCH |
2137 | |
2138 | def _unlock_file(f): | |
0edb3e33 | 2139 | raise LockingUnsupportedError() |
c1c9a79c PH |
2140 | |
2141 | ||
86e5f3ed | 2142 | class locked_file: |
0edb3e33 | 2143 | locked = False |
747c0bd1 | 2144 | |
a3125791 | 2145 | def __init__(self, filename, mode, block=True, encoding=None): |
fcfa8853 JK |
2146 | if mode not in {'r', 'rb', 'a', 'ab', 'w', 'wb'}: |
2147 | raise NotImplementedError(mode) | |
2148 | self.mode, self.block = mode, block | |
2149 | ||
2150 | writable = any(f in mode for f in 'wax+') | |
2151 | readable = any(f in mode for f in 'r+') | |
2152 | flags = functools.reduce(operator.ior, ( | |
2153 | getattr(os, 'O_CLOEXEC', 0), # UNIX only | |
2154 | getattr(os, 'O_BINARY', 0), # Windows only | |
2155 | getattr(os, 'O_NOINHERIT', 0), # Windows only | |
2156 | os.O_CREAT if writable else 0, # O_TRUNC only after locking | |
2157 | os.O_APPEND if 'a' in mode else 0, | |
2158 | os.O_EXCL if 'x' in mode else 0, | |
2159 | os.O_RDONLY if not writable else os.O_RDWR if readable else os.O_WRONLY, | |
2160 | )) | |
2161 | ||
98804d03 | 2162 | self.f = os.fdopen(os.open(filename, flags, 0o666), mode, encoding=encoding) |
c1c9a79c PH |
2163 | |
2164 | def __enter__(self): | |
a3125791 | 2165 | exclusive = 'r' not in self.mode |
c1c9a79c | 2166 | try: |
a3125791 | 2167 | _lock_file(self.f, exclusive, self.block) |
0edb3e33 | 2168 | self.locked = True |
86e5f3ed | 2169 | except OSError: |
c1c9a79c PH |
2170 | self.f.close() |
2171 | raise | |
fcfa8853 | 2172 | if 'w' in self.mode: |
131e14dc JK |
2173 | try: |
2174 | self.f.truncate() | |
2175 | except OSError as e: | |
1890fc63 | 2176 | if e.errno not in ( |
2177 | errno.ESPIPE, # Illegal seek - expected for FIFO | |
2178 | errno.EINVAL, # Invalid argument - expected for /dev/null | |
2179 | ): | |
2180 | raise | |
c1c9a79c PH |
2181 | return self |
2182 | ||
0edb3e33 | 2183 | def unlock(self): |
2184 | if not self.locked: | |
2185 | return | |
c1c9a79c | 2186 | try: |
0edb3e33 | 2187 | _unlock_file(self.f) |
c1c9a79c | 2188 | finally: |
0edb3e33 | 2189 | self.locked = False |
c1c9a79c | 2190 | |
0edb3e33 | 2191 | def __exit__(self, *_): |
2192 | try: | |
2193 | self.unlock() | |
2194 | finally: | |
2195 | self.f.close() | |
4eb7f1d1 | 2196 | |
0edb3e33 | 2197 | open = __enter__ |
2198 | close = __exit__ | |
a3125791 | 2199 | |
0edb3e33 | 2200 | def __getattr__(self, attr): |
2201 | return getattr(self.f, attr) | |
a3125791 | 2202 | |
0edb3e33 | 2203 | def __iter__(self): |
2204 | return iter(self.f) | |
a3125791 | 2205 | |
4eb7f1d1 | 2206 | |
0b9c08b4 | 2207 | @functools.cache |
4644ac55 S |
2208 | def get_filesystem_encoding(): |
2209 | encoding = sys.getfilesystemencoding() | |
2210 | return encoding if encoding is not None else 'utf-8' | |
2211 | ||
2212 | ||
4eb7f1d1 | 2213 | def shell_quote(args): |
a6a173c2 | 2214 | quoted_args = [] |
4644ac55 | 2215 | encoding = get_filesystem_encoding() |
a6a173c2 JMF |
2216 | for a in args: |
2217 | if isinstance(a, bytes): | |
2218 | # We may get a filename encoded with 'encodeFilename' | |
2219 | a = a.decode(encoding) | |
aefce8e6 | 2220 | quoted_args.append(compat_shlex_quote(a)) |
28e614de | 2221 | return ' '.join(quoted_args) |
9d4660ca PH |
2222 | |
2223 | ||
2224 | def smuggle_url(url, data): | |
2225 | """ Pass additional data in a URL for internal use. """ | |
2226 | ||
81953d1a RA |
2227 | url, idata = unsmuggle_url(url, {}) |
2228 | data.update(idata) | |
14f25df2 | 2229 | sdata = urllib.parse.urlencode( |
28e614de PH |
2230 | {'__youtubedl_smuggle': json.dumps(data)}) |
2231 | return url + '#' + sdata | |
9d4660ca PH |
2232 | |
2233 | ||
79f82953 | 2234 | def unsmuggle_url(smug_url, default=None): |
83e865a3 | 2235 | if '#__youtubedl_smuggle' not in smug_url: |
79f82953 | 2236 | return smug_url, default |
28e614de | 2237 | url, _, sdata = smug_url.rpartition('#') |
14f25df2 | 2238 | jsond = urllib.parse.parse_qs(sdata)['__youtubedl_smuggle'][0] |
9d4660ca PH |
2239 | data = json.loads(jsond) |
2240 | return url, data | |
02dbf93f PH |
2241 | |
2242 | ||
e0fd9573 | 2243 | def format_decimal_suffix(num, fmt='%d%s', *, factor=1000): |
2244 | """ Formats numbers with decimal sufixes like K, M, etc """ | |
2245 | num, factor = float_or_none(num), float(factor) | |
4c3f8c3f | 2246 | if num is None or num < 0: |
e0fd9573 | 2247 | return None |
eeb2a770 | 2248 | POSSIBLE_SUFFIXES = 'kMGTPEZY' |
2249 | exponent = 0 if num == 0 else min(int(math.log(num, factor)), len(POSSIBLE_SUFFIXES)) | |
2250 | suffix = ['', *POSSIBLE_SUFFIXES][exponent] | |
abbeeebc | 2251 | if factor == 1024: |
2252 | suffix = {'k': 'Ki', '': ''}.get(suffix, f'{suffix}i') | |
e0fd9573 | 2253 | converted = num / (factor ** exponent) |
abbeeebc | 2254 | return fmt % (converted, suffix) |
e0fd9573 | 2255 | |
2256 | ||
02dbf93f | 2257 | def format_bytes(bytes): |
f02d24d8 | 2258 | return format_decimal_suffix(bytes, '%.2f%sB', factor=1024) or 'N/A' |
f53c966a | 2259 | |
1c088fa8 | 2260 | |
fb47597b S |
2261 | def lookup_unit_table(unit_table, s): |
2262 | units_re = '|'.join(re.escape(u) for u in unit_table) | |
2263 | m = re.match( | |
782b1b5b | 2264 | r'(?P<num>[0-9]+(?:[,.][0-9]*)?)\s*(?P<unit>%s)\b' % units_re, s) |
fb47597b S |
2265 | if not m: |
2266 | return None | |
2267 | num_str = m.group('num').replace(',', '.') | |
2268 | mult = unit_table[m.group('unit')] | |
2269 | return int(float(num_str) * mult) | |
2270 | ||
2271 | ||
be64b5b0 PH |
2272 | def parse_filesize(s): |
2273 | if s is None: | |
2274 | return None | |
2275 | ||
dfb1b146 | 2276 | # The lower-case forms are of course incorrect and unofficial, |
be64b5b0 PH |
2277 | # but we support those too |
2278 | _UNIT_TABLE = { | |
2279 | 'B': 1, | |
2280 | 'b': 1, | |
70852b47 | 2281 | 'bytes': 1, |
be64b5b0 PH |
2282 | 'KiB': 1024, |
2283 | 'KB': 1000, | |
2284 | 'kB': 1024, | |
2285 | 'Kb': 1000, | |
13585d76 | 2286 | 'kb': 1000, |
70852b47 YCH |
2287 | 'kilobytes': 1000, |
2288 | 'kibibytes': 1024, | |
be64b5b0 PH |
2289 | 'MiB': 1024 ** 2, |
2290 | 'MB': 1000 ** 2, | |
2291 | 'mB': 1024 ** 2, | |
2292 | 'Mb': 1000 ** 2, | |
13585d76 | 2293 | 'mb': 1000 ** 2, |
70852b47 YCH |
2294 | 'megabytes': 1000 ** 2, |
2295 | 'mebibytes': 1024 ** 2, | |
be64b5b0 PH |
2296 | 'GiB': 1024 ** 3, |
2297 | 'GB': 1000 ** 3, | |
2298 | 'gB': 1024 ** 3, | |
2299 | 'Gb': 1000 ** 3, | |
13585d76 | 2300 | 'gb': 1000 ** 3, |
70852b47 YCH |
2301 | 'gigabytes': 1000 ** 3, |
2302 | 'gibibytes': 1024 ** 3, | |
be64b5b0 PH |
2303 | 'TiB': 1024 ** 4, |
2304 | 'TB': 1000 ** 4, | |
2305 | 'tB': 1024 ** 4, | |
2306 | 'Tb': 1000 ** 4, | |
13585d76 | 2307 | 'tb': 1000 ** 4, |
70852b47 YCH |
2308 | 'terabytes': 1000 ** 4, |
2309 | 'tebibytes': 1024 ** 4, | |
be64b5b0 PH |
2310 | 'PiB': 1024 ** 5, |
2311 | 'PB': 1000 ** 5, | |
2312 | 'pB': 1024 ** 5, | |
2313 | 'Pb': 1000 ** 5, | |
13585d76 | 2314 | 'pb': 1000 ** 5, |
70852b47 YCH |
2315 | 'petabytes': 1000 ** 5, |
2316 | 'pebibytes': 1024 ** 5, | |
be64b5b0 PH |
2317 | 'EiB': 1024 ** 6, |
2318 | 'EB': 1000 ** 6, | |
2319 | 'eB': 1024 ** 6, | |
2320 | 'Eb': 1000 ** 6, | |
13585d76 | 2321 | 'eb': 1000 ** 6, |
70852b47 YCH |
2322 | 'exabytes': 1000 ** 6, |
2323 | 'exbibytes': 1024 ** 6, | |
be64b5b0 PH |
2324 | 'ZiB': 1024 ** 7, |
2325 | 'ZB': 1000 ** 7, | |
2326 | 'zB': 1024 ** 7, | |
2327 | 'Zb': 1000 ** 7, | |
13585d76 | 2328 | 'zb': 1000 ** 7, |
70852b47 YCH |
2329 | 'zettabytes': 1000 ** 7, |
2330 | 'zebibytes': 1024 ** 7, | |
be64b5b0 PH |
2331 | 'YiB': 1024 ** 8, |
2332 | 'YB': 1000 ** 8, | |
2333 | 'yB': 1024 ** 8, | |
2334 | 'Yb': 1000 ** 8, | |
13585d76 | 2335 | 'yb': 1000 ** 8, |
70852b47 YCH |
2336 | 'yottabytes': 1000 ** 8, |
2337 | 'yobibytes': 1024 ** 8, | |
be64b5b0 PH |
2338 | } |
2339 | ||
fb47597b S |
2340 | return lookup_unit_table(_UNIT_TABLE, s) |
2341 | ||
2342 | ||
2343 | def parse_count(s): | |
2344 | if s is None: | |
be64b5b0 PH |
2345 | return None |
2346 | ||
352d5da8 | 2347 | s = re.sub(r'^[^\d]+\s', '', s).strip() |
fb47597b S |
2348 | |
2349 | if re.match(r'^[\d,.]+$', s): | |
2350 | return str_to_int(s) | |
2351 | ||
2352 | _UNIT_TABLE = { | |
2353 | 'k': 1000, | |
2354 | 'K': 1000, | |
2355 | 'm': 1000 ** 2, | |
2356 | 'M': 1000 ** 2, | |
2357 | 'kk': 1000 ** 2, | |
2358 | 'KK': 1000 ** 2, | |
352d5da8 | 2359 | 'b': 1000 ** 3, |
2360 | 'B': 1000 ** 3, | |
fb47597b | 2361 | } |
be64b5b0 | 2362 | |
352d5da8 | 2363 | ret = lookup_unit_table(_UNIT_TABLE, s) |
2364 | if ret is not None: | |
2365 | return ret | |
2366 | ||
2367 | mobj = re.match(r'([\d,.]+)(?:$|\s)', s) | |
2368 | if mobj: | |
2369 | return str_to_int(mobj.group(1)) | |
be64b5b0 | 2370 | |
2f7ae819 | 2371 | |
5d45484c | 2372 | def parse_resolution(s, *, lenient=False): |
b871d7e9 S |
2373 | if s is None: |
2374 | return {} | |
2375 | ||
5d45484c LNO |
2376 | if lenient: |
2377 | mobj = re.search(r'(?P<w>\d+)\s*[xX×,]\s*(?P<h>\d+)', s) | |
2378 | else: | |
2379 | mobj = re.search(r'(?<![a-zA-Z0-9])(?P<w>\d+)\s*[xX×,]\s*(?P<h>\d+)(?![a-zA-Z0-9])', s) | |
b871d7e9 S |
2380 | if mobj: |
2381 | return { | |
2382 | 'width': int(mobj.group('w')), | |
2383 | 'height': int(mobj.group('h')), | |
2384 | } | |
2385 | ||
17ec8bcf | 2386 | mobj = re.search(r'(?<![a-zA-Z0-9])(\d+)[pPiI](?![a-zA-Z0-9])', s) |
b871d7e9 S |
2387 | if mobj: |
2388 | return {'height': int(mobj.group(1))} | |
2389 | ||
2390 | mobj = re.search(r'\b([48])[kK]\b', s) | |
2391 | if mobj: | |
2392 | return {'height': int(mobj.group(1)) * 540} | |
2393 | ||
2394 | return {} | |
2395 | ||
2396 | ||
0dc41787 | 2397 | def parse_bitrate(s): |
14f25df2 | 2398 | if not isinstance(s, str): |
0dc41787 S |
2399 | return |
2400 | mobj = re.search(r'\b(\d+)\s*kbps', s) | |
2401 | if mobj: | |
2402 | return int(mobj.group(1)) | |
2403 | ||
2404 | ||
a942d6cb | 2405 | def month_by_name(name, lang='en'): |
caefb1de PH |
2406 | """ Return the number of a month by (locale-independently) English name """ |
2407 | ||
f6717dec | 2408 | month_names = MONTH_NAMES.get(lang, MONTH_NAMES['en']) |
a942d6cb | 2409 | |
caefb1de | 2410 | try: |
f6717dec | 2411 | return month_names.index(name) + 1 |
7105440c YCH |
2412 | except ValueError: |
2413 | return None | |
2414 | ||
2415 | ||
2416 | def month_by_abbreviation(abbrev): | |
2417 | """ Return the number of a month by (locale-independently) English | |
2418 | abbreviations """ | |
2419 | ||
2420 | try: | |
2421 | return [s[:3] for s in ENGLISH_MONTH_NAMES].index(abbrev) + 1 | |
caefb1de PH |
2422 | except ValueError: |
2423 | return None | |
18258362 JMF |
2424 | |
2425 | ||
5aafe895 | 2426 | def fix_xml_ampersands(xml_str): |
18258362 | 2427 | """Replace all the '&' by '&' in XML""" |
5aafe895 PH |
2428 | return re.sub( |
2429 | r'&(?!amp;|lt;|gt;|apos;|quot;|#x[0-9a-fA-F]{,4};|#[0-9]{,4};)', | |
28e614de | 2430 | '&', |
5aafe895 | 2431 | xml_str) |
e3946f98 PH |
2432 | |
2433 | ||
2434 | def setproctitle(title): | |
14f25df2 | 2435 | assert isinstance(title, str) |
c1c05c67 | 2436 | |
fe0918bb | 2437 | # Workaround for https://github.com/yt-dlp/yt-dlp/issues/4541 |
2438 | try: | |
2439 | import ctypes | |
2440 | except ImportError: | |
c1c05c67 YCH |
2441 | return |
2442 | ||
e3946f98 | 2443 | try: |
611c1dd9 | 2444 | libc = ctypes.cdll.LoadLibrary('libc.so.6') |
e3946f98 PH |
2445 | except OSError: |
2446 | return | |
2f49bcd6 RC |
2447 | except TypeError: |
2448 | # LoadLibrary in Windows Python 2.7.13 only expects | |
2449 | # a bytestring, but since unicode_literals turns | |
2450 | # every string into a unicode string, it fails. | |
2451 | return | |
0f06bcd7 | 2452 | title_bytes = title.encode() |
6eefe533 PH |
2453 | buf = ctypes.create_string_buffer(len(title_bytes)) |
2454 | buf.value = title_bytes | |
e3946f98 | 2455 | try: |
6eefe533 | 2456 | libc.prctl(15, buf, 0, 0, 0) |
e3946f98 PH |
2457 | except AttributeError: |
2458 | return # Strange libc, just skip this | |
d7dda168 PH |
2459 | |
2460 | ||
2461 | def remove_start(s, start): | |
46bc9b7d | 2462 | return s[len(start):] if s is not None and s.startswith(start) else s |
29eb5174 PH |
2463 | |
2464 | ||
2b9faf55 | 2465 | def remove_end(s, end): |
46bc9b7d | 2466 | return s[:-len(end)] if s is not None and s.endswith(end) else s |
2b9faf55 PH |
2467 | |
2468 | ||
31b2051e S |
2469 | def remove_quotes(s): |
2470 | if s is None or len(s) < 2: | |
2471 | return s | |
2472 | for quote in ('"', "'", ): | |
2473 | if s[0] == quote and s[-1] == quote: | |
2474 | return s[1:-1] | |
2475 | return s | |
2476 | ||
2477 | ||
b6e0c7d2 | 2478 | def get_domain(url): |
ebf99aaf | 2479 | """ |
2480 | This implementation is inconsistent, but is kept for compatibility. | |
2481 | Use this only for "webpage_url_domain" | |
2482 | """ | |
2483 | return remove_start(urllib.parse.urlparse(url).netloc, 'www.') or None | |
b6e0c7d2 U |
2484 | |
2485 | ||
29eb5174 | 2486 | def url_basename(url): |
14f25df2 | 2487 | path = urllib.parse.urlparse(url).path |
28e614de | 2488 | return path.strip('/').split('/')[-1] |
aa94a6d3 PH |
2489 | |
2490 | ||
02dc0a36 | 2491 | def base_url(url): |
7657ec7e | 2492 | return re.match(r'https?://[^?#]+/', url).group() |
02dc0a36 S |
2493 | |
2494 | ||
e34c3361 | 2495 | def urljoin(base, path): |
4b5de77b | 2496 | if isinstance(path, bytes): |
0f06bcd7 | 2497 | path = path.decode() |
14f25df2 | 2498 | if not isinstance(path, str) or not path: |
e34c3361 | 2499 | return None |
fad4ceb5 | 2500 | if re.match(r'^(?:[a-zA-Z][a-zA-Z0-9+-.]*:)?//', path): |
e34c3361 | 2501 | return path |
4b5de77b | 2502 | if isinstance(base, bytes): |
0f06bcd7 | 2503 | base = base.decode() |
14f25df2 | 2504 | if not isinstance(base, str) or not re.match( |
4b5de77b | 2505 | r'^(?:https?:)?//', base): |
e34c3361 | 2506 | return None |
14f25df2 | 2507 | return urllib.parse.urljoin(base, path) |
e34c3361 S |
2508 | |
2509 | ||
ac668111 | 2510 | class HEADRequest(urllib.request.Request): |
aa94a6d3 | 2511 | def get_method(self): |
611c1dd9 | 2512 | return 'HEAD' |
7217e148 PH |
2513 | |
2514 | ||
ac668111 | 2515 | class PUTRequest(urllib.request.Request): |
95cf60e8 S |
2516 | def get_method(self): |
2517 | return 'PUT' | |
2518 | ||
2519 | ||
9732d77e | 2520 | def int_or_none(v, scale=1, default=None, get_attr=None, invscale=1): |
9e907ebd | 2521 | if get_attr and v is not None: |
2522 | v = getattr(v, get_attr, None) | |
1812afb7 S |
2523 | try: |
2524 | return int(v) * invscale // scale | |
31c49255 | 2525 | except (ValueError, TypeError, OverflowError): |
af98f8ff | 2526 | return default |
9732d77e | 2527 | |
9572013d | 2528 | |
40a90862 | 2529 | def str_or_none(v, default=None): |
14f25df2 | 2530 | return default if v is None else str(v) |
40a90862 | 2531 | |
9732d77e PH |
2532 | |
2533 | def str_to_int(int_str): | |
48d4681e | 2534 | """ A more relaxed version of int_or_none """ |
f9934b96 | 2535 | if isinstance(int_str, int): |
348c6bf1 | 2536 | return int_str |
14f25df2 | 2537 | elif isinstance(int_str, str): |
42db58ec S |
2538 | int_str = re.sub(r'[,\.\+]', '', int_str) |
2539 | return int_or_none(int_str) | |
608d11f5 PH |
2540 | |
2541 | ||
9732d77e | 2542 | def float_or_none(v, scale=1, invscale=1, default=None): |
caf80631 S |
2543 | if v is None: |
2544 | return default | |
2545 | try: | |
2546 | return float(v) * invscale / scale | |
5e1271c5 | 2547 | except (ValueError, TypeError): |
caf80631 | 2548 | return default |
43f775e4 PH |
2549 | |
2550 | ||
c7e327c4 S |
2551 | def bool_or_none(v, default=None): |
2552 | return v if isinstance(v, bool) else default | |
2553 | ||
2554 | ||
53cd37ba | 2555 | def strip_or_none(v, default=None): |
14f25df2 | 2556 | return v.strip() if isinstance(v, str) else default |
b72b4431 S |
2557 | |
2558 | ||
af03000a | 2559 | def url_or_none(url): |
14f25df2 | 2560 | if not url or not isinstance(url, str): |
af03000a S |
2561 | return None |
2562 | url = url.strip() | |
29f7c58a | 2563 | return url if re.match(r'^(?:(?:https?|rt(?:m(?:pt?[es]?|fp)|sp[su]?)|mms|ftps?):)?//', url) else None |
af03000a S |
2564 | |
2565 | ||
3e9b66d7 | 2566 | def request_to_url(req): |
ac668111 | 2567 | if isinstance(req, urllib.request.Request): |
3e9b66d7 LNO |
2568 | return req.get_full_url() |
2569 | else: | |
2570 | return req | |
2571 | ||
2572 | ||
e29663c6 | 2573 | def strftime_or_none(timestamp, date_format, default=None): |
2574 | datetime_object = None | |
2575 | try: | |
f9934b96 | 2576 | if isinstance(timestamp, (int, float)): # unix timestamp |
e29663c6 | 2577 | datetime_object = datetime.datetime.utcfromtimestamp(timestamp) |
14f25df2 | 2578 | elif isinstance(timestamp, str): # assume YYYYMMDD |
e29663c6 | 2579 | datetime_object = datetime.datetime.strptime(timestamp, '%Y%m%d') |
9665f15a | 2580 | date_format = re.sub( # Support %s on windows |
2581 | r'(?<!%)(%%)*%s', rf'\g<1>{int(datetime_object.timestamp())}', date_format) | |
e29663c6 | 2582 | return datetime_object.strftime(date_format) |
2583 | except (ValueError, TypeError, AttributeError): | |
2584 | return default | |
2585 | ||
2586 | ||
608d11f5 | 2587 | def parse_duration(s): |
f9934b96 | 2588 | if not isinstance(s, str): |
608d11f5 | 2589 | return None |
ca7b3246 | 2590 | s = s.strip() |
38d79fd1 | 2591 | if not s: |
2592 | return None | |
ca7b3246 | 2593 | |
acaff495 | 2594 | days, hours, mins, secs, ms = [None] * 5 |
8bd1c00b | 2595 | m = re.match(r'''(?x) |
2596 | (?P<before_secs> | |
2597 | (?:(?:(?P<days>[0-9]+):)?(?P<hours>[0-9]+):)?(?P<mins>[0-9]+):)? | |
2598 | (?P<secs>(?(before_secs)[0-9]{1,2}|[0-9]+)) | |
2599 | (?P<ms>[.:][0-9]+)?Z?$ | |
2600 | ''', s) | |
acaff495 | 2601 | if m: |
8bd1c00b | 2602 | days, hours, mins, secs, ms = m.group('days', 'hours', 'mins', 'secs', 'ms') |
acaff495 | 2603 | else: |
2604 | m = re.match( | |
056653bb S |
2605 | r'''(?ix)(?:P? |
2606 | (?: | |
1c1b2f96 | 2607 | [0-9]+\s*y(?:ears?)?,?\s* |
056653bb S |
2608 | )? |
2609 | (?: | |
1c1b2f96 | 2610 | [0-9]+\s*m(?:onths?)?,?\s* |
056653bb S |
2611 | )? |
2612 | (?: | |
1c1b2f96 | 2613 | [0-9]+\s*w(?:eeks?)?,?\s* |
056653bb | 2614 | )? |
8f4b58d7 | 2615 | (?: |
1c1b2f96 | 2616 | (?P<days>[0-9]+)\s*d(?:ays?)?,?\s* |
8f4b58d7 | 2617 | )? |
056653bb | 2618 | T)? |
acaff495 | 2619 | (?: |
1c1b2f96 | 2620 | (?P<hours>[0-9]+)\s*h(?:ours?)?,?\s* |
acaff495 | 2621 | )? |
2622 | (?: | |
1c1b2f96 | 2623 | (?P<mins>[0-9]+)\s*m(?:in(?:ute)?s?)?,?\s* |
acaff495 | 2624 | )? |
2625 | (?: | |
2626 | (?P<secs>[0-9]+)(?P<ms>\.[0-9]+)?\s*s(?:ec(?:ond)?s?)?\s* | |
15846398 | 2627 | )?Z?$''', s) |
acaff495 | 2628 | if m: |
2629 | days, hours, mins, secs, ms = m.groups() | |
2630 | else: | |
15846398 | 2631 | m = re.match(r'(?i)(?:(?P<hours>[0-9.]+)\s*(?:hours?)|(?P<mins>[0-9.]+)\s*(?:mins?\.?|minutes?)\s*)Z?$', s) |
acaff495 | 2632 | if m: |
2633 | hours, mins = m.groups() | |
2634 | else: | |
2635 | return None | |
2636 | ||
acaff495 | 2637 | if ms: |
19a03940 | 2638 | ms = ms.replace(':', '.') |
2639 | return sum(float(part or 0) * mult for part, mult in ( | |
2640 | (days, 86400), (hours, 3600), (mins, 60), (secs, 1), (ms, 1))) | |
91d7d0b3 JMF |
2641 | |
2642 | ||
e65e4c88 | 2643 | def prepend_extension(filename, ext, expected_real_ext=None): |
5f6a1245 | 2644 | name, real_ext = os.path.splitext(filename) |
e65e4c88 | 2645 | return ( |
86e5f3ed | 2646 | f'{name}.{ext}{real_ext}' |
e65e4c88 | 2647 | if not expected_real_ext or real_ext[1:] == expected_real_ext |
86e5f3ed | 2648 | else f'{filename}.{ext}') |
d70ad093 PH |
2649 | |
2650 | ||
b3ed15b7 S |
2651 | def replace_extension(filename, ext, expected_real_ext=None): |
2652 | name, real_ext = os.path.splitext(filename) | |
86e5f3ed | 2653 | return '{}.{}'.format( |
b3ed15b7 S |
2654 | name if not expected_real_ext or real_ext[1:] == expected_real_ext else filename, |
2655 | ext) | |
2656 | ||
2657 | ||
d70ad093 PH |
2658 | def check_executable(exe, args=[]): |
2659 | """ Checks if the given binary is installed somewhere in PATH, and returns its name. | |
2660 | args can be a list of arguments for a short output (like -version) """ | |
2661 | try: | |
f0c9fb96 | 2662 | Popen.run([exe] + args, stdout=subprocess.PIPE, stderr=subprocess.PIPE) |
d70ad093 PH |
2663 | except OSError: |
2664 | return False | |
2665 | return exe | |
b7ab0590 PH |
2666 | |
2667 | ||
8a7f68d0 | 2668 | def _get_exe_version_output(exe, args, *, to_screen=None): |
2669 | if to_screen: | |
2670 | to_screen(f'Checking exe version: {shell_quote([exe] + args)}') | |
95807118 | 2671 | try: |
b64d04c1 | 2672 | # STDIN should be redirected too. On UNIX-like systems, ffmpeg triggers |
7a5c1cfe | 2673 | # SIGTTOU if yt-dlp is run in the background. |
067aa17e | 2674 | # See https://github.com/ytdl-org/youtube-dl/issues/955#issuecomment-209789656 |
f0c9fb96 | 2675 | stdout, _, _ = Popen.run([encodeArgument(exe)] + args, text=True, |
2676 | stdin=subprocess.PIPE, stdout=subprocess.PIPE, stderr=subprocess.STDOUT) | |
95807118 PH |
2677 | except OSError: |
2678 | return False | |
f0c9fb96 | 2679 | return stdout |
cae97f65 PH |
2680 | |
2681 | ||
2682 | def detect_exe_version(output, version_re=None, unrecognized='present'): | |
14f25df2 | 2683 | assert isinstance(output, str) |
cae97f65 PH |
2684 | if version_re is None: |
2685 | version_re = r'version\s+([-0-9._a-zA-Z]+)' | |
2686 | m = re.search(version_re, output) | |
95807118 PH |
2687 | if m: |
2688 | return m.group(1) | |
2689 | else: | |
2690 | return unrecognized | |
2691 | ||
2692 | ||
9af98e17 | 2693 | def get_exe_version(exe, args=['--version'], |
2694 | version_re=None, unrecognized='present'): | |
2695 | """ Returns the version of the specified executable, | |
2696 | or False if the executable is not present """ | |
2697 | out = _get_exe_version_output(exe, args) | |
2698 | return detect_exe_version(out, version_re, unrecognized) if out else False | |
2699 | ||
2700 | ||
7e88d7d7 | 2701 | def frange(start=0, stop=None, step=1): |
2702 | """Float range""" | |
2703 | if stop is None: | |
2704 | start, stop = 0, start | |
2705 | sign = [-1, 1][step > 0] if step else 0 | |
2706 | while sign * start < sign * stop: | |
2707 | yield start | |
2708 | start += step | |
2709 | ||
2710 | ||
cb89cfc1 | 2711 | class LazyList(collections.abc.Sequence): |
0f06bcd7 | 2712 | """Lazy immutable list from an iterable |
2713 | Note that slices of a LazyList are lists and not LazyList""" | |
483336e7 | 2714 | |
8e5fecc8 | 2715 | class IndexError(IndexError): |
2716 | pass | |
2717 | ||
282f5709 | 2718 | def __init__(self, iterable, *, reverse=False, _cache=None): |
0f06bcd7 | 2719 | self._iterable = iter(iterable) |
2720 | self._cache = [] if _cache is None else _cache | |
2721 | self._reversed = reverse | |
483336e7 | 2722 | |
2723 | def __iter__(self): | |
0f06bcd7 | 2724 | if self._reversed: |
28419ca2 | 2725 | # We need to consume the entire iterable to iterate in reverse |
981052c9 | 2726 | yield from self.exhaust() |
28419ca2 | 2727 | return |
0f06bcd7 | 2728 | yield from self._cache |
2729 | for item in self._iterable: | |
2730 | self._cache.append(item) | |
483336e7 | 2731 | yield item |
2732 | ||
0f06bcd7 | 2733 | def _exhaust(self): |
2734 | self._cache.extend(self._iterable) | |
2735 | self._iterable = [] # Discard the emptied iterable to make it pickle-able | |
2736 | return self._cache | |
28419ca2 | 2737 | |
981052c9 | 2738 | def exhaust(self): |
0f06bcd7 | 2739 | """Evaluate the entire iterable""" |
2740 | return self._exhaust()[::-1 if self._reversed else 1] | |
981052c9 | 2741 | |
28419ca2 | 2742 | @staticmethod |
0f06bcd7 | 2743 | def _reverse_index(x): |
f2df4071 | 2744 | return None if x is None else ~x |
483336e7 | 2745 | |
2746 | def __getitem__(self, idx): | |
2747 | if isinstance(idx, slice): | |
0f06bcd7 | 2748 | if self._reversed: |
2749 | idx = slice(self._reverse_index(idx.start), self._reverse_index(idx.stop), -(idx.step or 1)) | |
e0f2b4b4 | 2750 | start, stop, step = idx.start, idx.stop, idx.step or 1 |
483336e7 | 2751 | elif isinstance(idx, int): |
0f06bcd7 | 2752 | if self._reversed: |
2753 | idx = self._reverse_index(idx) | |
e0f2b4b4 | 2754 | start, stop, step = idx, idx, 0 |
483336e7 | 2755 | else: |
2756 | raise TypeError('indices must be integers or slices') | |
e0f2b4b4 | 2757 | if ((start or 0) < 0 or (stop or 0) < 0 |
2758 | or (start is None and step < 0) | |
2759 | or (stop is None and step > 0)): | |
483336e7 | 2760 | # We need to consume the entire iterable to be able to slice from the end |
2761 | # Obviously, never use this with infinite iterables | |
0f06bcd7 | 2762 | self._exhaust() |
8e5fecc8 | 2763 | try: |
0f06bcd7 | 2764 | return self._cache[idx] |
8e5fecc8 | 2765 | except IndexError as e: |
2766 | raise self.IndexError(e) from e | |
0f06bcd7 | 2767 | n = max(start or 0, stop or 0) - len(self._cache) + 1 |
28419ca2 | 2768 | if n > 0: |
0f06bcd7 | 2769 | self._cache.extend(itertools.islice(self._iterable, n)) |
8e5fecc8 | 2770 | try: |
0f06bcd7 | 2771 | return self._cache[idx] |
8e5fecc8 | 2772 | except IndexError as e: |
2773 | raise self.IndexError(e) from e | |
483336e7 | 2774 | |
2775 | def __bool__(self): | |
2776 | try: | |
0f06bcd7 | 2777 | self[-1] if self._reversed else self[0] |
8e5fecc8 | 2778 | except self.IndexError: |
483336e7 | 2779 | return False |
2780 | return True | |
2781 | ||
2782 | def __len__(self): | |
0f06bcd7 | 2783 | self._exhaust() |
2784 | return len(self._cache) | |
483336e7 | 2785 | |
282f5709 | 2786 | def __reversed__(self): |
0f06bcd7 | 2787 | return type(self)(self._iterable, reverse=not self._reversed, _cache=self._cache) |
282f5709 | 2788 | |
2789 | def __copy__(self): | |
0f06bcd7 | 2790 | return type(self)(self._iterable, reverse=self._reversed, _cache=self._cache) |
282f5709 | 2791 | |
28419ca2 | 2792 | def __repr__(self): |
2793 | # repr and str should mimic a list. So we exhaust the iterable | |
2794 | return repr(self.exhaust()) | |
2795 | ||
2796 | def __str__(self): | |
2797 | return repr(self.exhaust()) | |
2798 | ||
483336e7 | 2799 | |
7be9ccff | 2800 | class PagedList: |
c07a39ae | 2801 | |
2802 | class IndexError(IndexError): | |
2803 | pass | |
2804 | ||
dd26ced1 PH |
2805 | def __len__(self): |
2806 | # This is only useful for tests | |
2807 | return len(self.getslice()) | |
2808 | ||
7be9ccff | 2809 | def __init__(self, pagefunc, pagesize, use_cache=True): |
2810 | self._pagefunc = pagefunc | |
2811 | self._pagesize = pagesize | |
f1d13090 | 2812 | self._pagecount = float('inf') |
7be9ccff | 2813 | self._use_cache = use_cache |
2814 | self._cache = {} | |
2815 | ||
2816 | def getpage(self, pagenum): | |
d8cf8d97 | 2817 | page_results = self._cache.get(pagenum) |
2818 | if page_results is None: | |
f1d13090 | 2819 | page_results = [] if pagenum > self._pagecount else list(self._pagefunc(pagenum)) |
7be9ccff | 2820 | if self._use_cache: |
2821 | self._cache[pagenum] = page_results | |
2822 | return page_results | |
2823 | ||
2824 | def getslice(self, start=0, end=None): | |
2825 | return list(self._getslice(start, end)) | |
2826 | ||
2827 | def _getslice(self, start, end): | |
55575225 | 2828 | raise NotImplementedError('This method must be implemented by subclasses') |
2829 | ||
2830 | def __getitem__(self, idx): | |
f1d13090 | 2831 | assert self._use_cache, 'Indexing PagedList requires cache' |
55575225 | 2832 | if not isinstance(idx, int) or idx < 0: |
2833 | raise TypeError('indices must be non-negative integers') | |
2834 | entries = self.getslice(idx, idx + 1) | |
d8cf8d97 | 2835 | if not entries: |
c07a39ae | 2836 | raise self.IndexError() |
d8cf8d97 | 2837 | return entries[0] |
55575225 | 2838 | |
9c44d242 PH |
2839 | |
2840 | class OnDemandPagedList(PagedList): | |
a44ca5a4 | 2841 | """Download pages until a page with less than maximum results""" |
86e5f3ed | 2842 | |
7be9ccff | 2843 | def _getslice(self, start, end): |
b7ab0590 PH |
2844 | for pagenum in itertools.count(start // self._pagesize): |
2845 | firstid = pagenum * self._pagesize | |
2846 | nextfirstid = pagenum * self._pagesize + self._pagesize | |
2847 | if start >= nextfirstid: | |
2848 | continue | |
2849 | ||
b7ab0590 PH |
2850 | startv = ( |
2851 | start % self._pagesize | |
2852 | if firstid <= start < nextfirstid | |
2853 | else 0) | |
b7ab0590 PH |
2854 | endv = ( |
2855 | ((end - 1) % self._pagesize) + 1 | |
2856 | if (end is not None and firstid <= end <= nextfirstid) | |
2857 | else None) | |
2858 | ||
f1d13090 | 2859 | try: |
2860 | page_results = self.getpage(pagenum) | |
2861 | except Exception: | |
2862 | self._pagecount = pagenum - 1 | |
2863 | raise | |
b7ab0590 PH |
2864 | if startv != 0 or endv is not None: |
2865 | page_results = page_results[startv:endv] | |
7be9ccff | 2866 | yield from page_results |
b7ab0590 PH |
2867 | |
2868 | # A little optimization - if current page is not "full", ie. does | |
2869 | # not contain page_size videos then we can assume that this page | |
2870 | # is the last one - there are no more ids on further pages - | |
2871 | # i.e. no need to query again. | |
2872 | if len(page_results) + startv < self._pagesize: | |
2873 | break | |
2874 | ||
2875 | # If we got the whole page, but the next page is not interesting, | |
2876 | # break out early as well | |
2877 | if end == nextfirstid: | |
2878 | break | |
81c2f20b PH |
2879 | |
2880 | ||
9c44d242 | 2881 | class InAdvancePagedList(PagedList): |
a44ca5a4 | 2882 | """PagedList with total number of pages known in advance""" |
86e5f3ed | 2883 | |
9c44d242 | 2884 | def __init__(self, pagefunc, pagecount, pagesize): |
7be9ccff | 2885 | PagedList.__init__(self, pagefunc, pagesize, True) |
f1d13090 | 2886 | self._pagecount = pagecount |
9c44d242 | 2887 | |
7be9ccff | 2888 | def _getslice(self, start, end): |
9c44d242 | 2889 | start_page = start // self._pagesize |
d37707bd | 2890 | end_page = self._pagecount if end is None else min(self._pagecount, end // self._pagesize + 1) |
9c44d242 PH |
2891 | skip_elems = start - start_page * self._pagesize |
2892 | only_more = None if end is None else end - start | |
2893 | for pagenum in range(start_page, end_page): | |
7be9ccff | 2894 | page_results = self.getpage(pagenum) |
9c44d242 | 2895 | if skip_elems: |
7be9ccff | 2896 | page_results = page_results[skip_elems:] |
9c44d242 PH |
2897 | skip_elems = None |
2898 | if only_more is not None: | |
7be9ccff | 2899 | if len(page_results) < only_more: |
2900 | only_more -= len(page_results) | |
9c44d242 | 2901 | else: |
7be9ccff | 2902 | yield from page_results[:only_more] |
9c44d242 | 2903 | break |
7be9ccff | 2904 | yield from page_results |
9c44d242 PH |
2905 | |
2906 | ||
7e88d7d7 | 2907 | class PlaylistEntries: |
2908 | MissingEntry = object() | |
2909 | is_exhausted = False | |
2910 | ||
2911 | def __init__(self, ydl, info_dict): | |
7e9a6125 | 2912 | self.ydl = ydl |
2913 | ||
2914 | # _entries must be assigned now since infodict can change during iteration | |
2915 | entries = info_dict.get('entries') | |
2916 | if entries is None: | |
2917 | raise EntryNotInPlaylist('There are no entries') | |
2918 | elif isinstance(entries, list): | |
2919 | self.is_exhausted = True | |
2920 | ||
2921 | requested_entries = info_dict.get('requested_entries') | |
2922 | self.is_incomplete = bool(requested_entries) | |
2923 | if self.is_incomplete: | |
2924 | assert self.is_exhausted | |
2925 | self._entries = [self.MissingEntry] * max(requested_entries) | |
2926 | for i, entry in zip(requested_entries, entries): | |
2927 | self._entries[i - 1] = entry | |
2928 | elif isinstance(entries, (list, PagedList, LazyList)): | |
2929 | self._entries = entries | |
2930 | else: | |
2931 | self._entries = LazyList(entries) | |
7e88d7d7 | 2932 | |
2933 | PLAYLIST_ITEMS_RE = re.compile(r'''(?x) | |
2934 | (?P<start>[+-]?\d+)? | |
2935 | (?P<range>[:-] | |
2936 | (?P<end>[+-]?\d+|inf(?:inite)?)? | |
2937 | (?::(?P<step>[+-]?\d+))? | |
2938 | )?''') | |
2939 | ||
2940 | @classmethod | |
2941 | def parse_playlist_items(cls, string): | |
2942 | for segment in string.split(','): | |
2943 | if not segment: | |
2944 | raise ValueError('There is two or more consecutive commas') | |
2945 | mobj = cls.PLAYLIST_ITEMS_RE.fullmatch(segment) | |
2946 | if not mobj: | |
2947 | raise ValueError(f'{segment!r} is not a valid specification') | |
2948 | start, end, step, has_range = mobj.group('start', 'end', 'step', 'range') | |
2949 | if int_or_none(step) == 0: | |
2950 | raise ValueError(f'Step in {segment!r} cannot be zero') | |
2951 | yield slice(int_or_none(start), float_or_none(end), int_or_none(step)) if has_range else int(start) | |
2952 | ||
2953 | def get_requested_items(self): | |
2954 | playlist_items = self.ydl.params.get('playlist_items') | |
2955 | playlist_start = self.ydl.params.get('playliststart', 1) | |
2956 | playlist_end = self.ydl.params.get('playlistend') | |
2957 | # For backwards compatibility, interpret -1 as whole list | |
2958 | if playlist_end in (-1, None): | |
2959 | playlist_end = '' | |
2960 | if not playlist_items: | |
2961 | playlist_items = f'{playlist_start}:{playlist_end}' | |
2962 | elif playlist_start != 1 or playlist_end: | |
2963 | self.ydl.report_warning('Ignoring playliststart and playlistend because playlistitems was given', only_once=True) | |
2964 | ||
2965 | for index in self.parse_playlist_items(playlist_items): | |
2966 | for i, entry in self[index]: | |
2967 | yield i, entry | |
1ac4fd80 | 2968 | if not entry: |
2969 | continue | |
7e88d7d7 | 2970 | try: |
2971 | # TODO: Add auto-generated fields | |
2972 | self.ydl._match_entry(entry, incomplete=True, silent=True) | |
2973 | except (ExistingVideoReached, RejectedVideoReached): | |
2974 | return | |
2975 | ||
7e9a6125 | 2976 | def get_full_count(self): |
2977 | if self.is_exhausted and not self.is_incomplete: | |
7e88d7d7 | 2978 | return len(self) |
2979 | elif isinstance(self._entries, InAdvancePagedList): | |
2980 | if self._entries._pagesize == 1: | |
2981 | return self._entries._pagecount | |
2982 | ||
7e88d7d7 | 2983 | @functools.cached_property |
2984 | def _getter(self): | |
2985 | if isinstance(self._entries, list): | |
2986 | def get_entry(i): | |
2987 | try: | |
2988 | entry = self._entries[i] | |
2989 | except IndexError: | |
2990 | entry = self.MissingEntry | |
2991 | if not self.is_incomplete: | |
2992 | raise self.IndexError() | |
2993 | if entry is self.MissingEntry: | |
2994 | raise EntryNotInPlaylist(f'Entry {i} cannot be found') | |
2995 | return entry | |
2996 | else: | |
2997 | def get_entry(i): | |
2998 | try: | |
2999 | return type(self.ydl)._handle_extraction_exceptions(lambda _, i: self._entries[i])(self.ydl, i) | |
3000 | except (LazyList.IndexError, PagedList.IndexError): | |
3001 | raise self.IndexError() | |
3002 | return get_entry | |
3003 | ||
3004 | def __getitem__(self, idx): | |
3005 | if isinstance(idx, int): | |
3006 | idx = slice(idx, idx) | |
3007 | ||
3008 | # NB: PlaylistEntries[1:10] => (0, 1, ... 9) | |
3009 | step = 1 if idx.step is None else idx.step | |
3010 | if idx.start is None: | |
3011 | start = 0 if step > 0 else len(self) - 1 | |
3012 | else: | |
3013 | start = idx.start - 1 if idx.start >= 0 else len(self) + idx.start | |
3014 | ||
3015 | # NB: Do not call len(self) when idx == [:] | |
3016 | if idx.stop is None: | |
3017 | stop = 0 if step < 0 else float('inf') | |
3018 | else: | |
3019 | stop = idx.stop - 1 if idx.stop >= 0 else len(self) + idx.stop | |
3020 | stop += [-1, 1][step > 0] | |
3021 | ||
3022 | for i in frange(start, stop, step): | |
3023 | if i < 0: | |
3024 | continue | |
3025 | try: | |
7e9a6125 | 3026 | entry = self._getter(i) |
3027 | except self.IndexError: | |
3028 | self.is_exhausted = True | |
3029 | if step > 0: | |
7e88d7d7 | 3030 | break |
7e9a6125 | 3031 | continue |
7e88d7d7 | 3032 | yield i + 1, entry |
3033 | ||
3034 | def __len__(self): | |
3035 | return len(tuple(self[:])) | |
3036 | ||
3037 | class IndexError(IndexError): | |
3038 | pass | |
3039 | ||
3040 | ||
81c2f20b | 3041 | def uppercase_escape(s): |
676eb3f2 | 3042 | unicode_escape = codecs.getdecoder('unicode_escape') |
81c2f20b | 3043 | return re.sub( |
a612753d | 3044 | r'\\U[0-9a-fA-F]{8}', |
676eb3f2 PH |
3045 | lambda m: unicode_escape(m.group(0))[0], |
3046 | s) | |
0fe2ff78 YCH |
3047 | |
3048 | ||
3049 | def lowercase_escape(s): | |
3050 | unicode_escape = codecs.getdecoder('unicode_escape') | |
3051 | return re.sub( | |
3052 | r'\\u[0-9a-fA-F]{4}', | |
3053 | lambda m: unicode_escape(m.group(0))[0], | |
3054 | s) | |
b53466e1 | 3055 | |
d05cfe06 S |
3056 | |
3057 | def escape_rfc3986(s): | |
3058 | """Escape non-ASCII characters as suggested by RFC 3986""" | |
f9934b96 | 3059 | return urllib.parse.quote(s, b"%/;:@&=+$,!~*'()?#[]") |
d05cfe06 S |
3060 | |
3061 | ||
3062 | def escape_url(url): | |
3063 | """Escape URL as suggested by RFC 3986""" | |
14f25df2 | 3064 | url_parsed = urllib.parse.urlparse(url) |
d05cfe06 | 3065 | return url_parsed._replace( |
efbed08d | 3066 | netloc=url_parsed.netloc.encode('idna').decode('ascii'), |
d05cfe06 S |
3067 | path=escape_rfc3986(url_parsed.path), |
3068 | params=escape_rfc3986(url_parsed.params), | |
3069 | query=escape_rfc3986(url_parsed.query), | |
3070 | fragment=escape_rfc3986(url_parsed.fragment) | |
3071 | ).geturl() | |
3072 | ||
62e609ab | 3073 | |
4dfbf869 | 3074 | def parse_qs(url): |
14f25df2 | 3075 | return urllib.parse.parse_qs(urllib.parse.urlparse(url).query) |
4dfbf869 | 3076 | |
3077 | ||
62e609ab PH |
3078 | def read_batch_urls(batch_fd): |
3079 | def fixup(url): | |
14f25df2 | 3080 | if not isinstance(url, str): |
62e609ab | 3081 | url = url.decode('utf-8', 'replace') |
8c04f0be | 3082 | BOM_UTF8 = ('\xef\xbb\xbf', '\ufeff') |
3083 | for bom in BOM_UTF8: | |
3084 | if url.startswith(bom): | |
3085 | url = url[len(bom):] | |
3086 | url = url.lstrip() | |
3087 | if not url or url.startswith(('#', ';', ']')): | |
62e609ab | 3088 | return False |
8c04f0be | 3089 | # "#" cannot be stripped out since it is part of the URI |
962ffcf8 | 3090 | # However, it can be safely stripped out if following a whitespace |
8c04f0be | 3091 | return re.split(r'\s#', url, 1)[0].rstrip() |
62e609ab PH |
3092 | |
3093 | with contextlib.closing(batch_fd) as fd: | |
3094 | return [url for url in map(fixup, fd) if url] | |
b74fa8cd JMF |
3095 | |
3096 | ||
3097 | def urlencode_postdata(*args, **kargs): | |
14f25df2 | 3098 | return urllib.parse.urlencode(*args, **kargs).encode('ascii') |
bcf89ce6 PH |
3099 | |
3100 | ||
38f9ef31 | 3101 | def update_url_query(url, query): |
cacd9966 YCH |
3102 | if not query: |
3103 | return url | |
14f25df2 | 3104 | parsed_url = urllib.parse.urlparse(url) |
3105 | qs = urllib.parse.parse_qs(parsed_url.query) | |
38f9ef31 | 3106 | qs.update(query) |
14f25df2 | 3107 | return urllib.parse.urlunparse(parsed_url._replace( |
3108 | query=urllib.parse.urlencode(qs, True))) | |
16392824 | 3109 | |
8e60dc75 | 3110 | |
c043c246 | 3111 | def update_Request(req, url=None, data=None, headers=None, query=None): |
ed0291d1 | 3112 | req_headers = req.headers.copy() |
c043c246 | 3113 | req_headers.update(headers or {}) |
ed0291d1 S |
3114 | req_data = data or req.data |
3115 | req_url = update_url_query(url or req.get_full_url(), query) | |
95cf60e8 S |
3116 | req_get_method = req.get_method() |
3117 | if req_get_method == 'HEAD': | |
3118 | req_type = HEADRequest | |
3119 | elif req_get_method == 'PUT': | |
3120 | req_type = PUTRequest | |
3121 | else: | |
ac668111 | 3122 | req_type = urllib.request.Request |
ed0291d1 S |
3123 | new_req = req_type( |
3124 | req_url, data=req_data, headers=req_headers, | |
3125 | origin_req_host=req.origin_req_host, unverifiable=req.unverifiable) | |
3126 | if hasattr(req, 'timeout'): | |
3127 | new_req.timeout = req.timeout | |
3128 | return new_req | |
3129 | ||
3130 | ||
10c87c15 | 3131 | def _multipart_encode_impl(data, boundary): |
0c265486 YCH |
3132 | content_type = 'multipart/form-data; boundary=%s' % boundary |
3133 | ||
3134 | out = b'' | |
3135 | for k, v in data.items(): | |
3136 | out += b'--' + boundary.encode('ascii') + b'\r\n' | |
14f25df2 | 3137 | if isinstance(k, str): |
0f06bcd7 | 3138 | k = k.encode() |
14f25df2 | 3139 | if isinstance(v, str): |
0f06bcd7 | 3140 | v = v.encode() |
0c265486 YCH |
3141 | # RFC 2047 requires non-ASCII field names to be encoded, while RFC 7578 |
3142 | # suggests sending UTF-8 directly. Firefox sends UTF-8, too | |
b2ad479d | 3143 | content = b'Content-Disposition: form-data; name="' + k + b'"\r\n\r\n' + v + b'\r\n' |
0c265486 YCH |
3144 | if boundary.encode('ascii') in content: |
3145 | raise ValueError('Boundary overlaps with data') | |
3146 | out += content | |
3147 | ||
3148 | out += b'--' + boundary.encode('ascii') + b'--\r\n' | |
3149 | ||
3150 | return out, content_type | |
3151 | ||
3152 | ||
3153 | def multipart_encode(data, boundary=None): | |
3154 | ''' | |
3155 | Encode a dict to RFC 7578-compliant form-data | |
3156 | ||
3157 | data: | |
3158 | A dict where keys and values can be either Unicode or bytes-like | |
3159 | objects. | |
3160 | boundary: | |
3161 | If specified a Unicode object, it's used as the boundary. Otherwise | |
3162 | a random boundary is generated. | |
3163 | ||
3164 | Reference: https://tools.ietf.org/html/rfc7578 | |
3165 | ''' | |
3166 | has_specified_boundary = boundary is not None | |
3167 | ||
3168 | while True: | |
3169 | if boundary is None: | |
3170 | boundary = '---------------' + str(random.randrange(0x0fffffff, 0xffffffff)) | |
3171 | ||
3172 | try: | |
10c87c15 | 3173 | out, content_type = _multipart_encode_impl(data, boundary) |
0c265486 YCH |
3174 | break |
3175 | except ValueError: | |
3176 | if has_specified_boundary: | |
3177 | raise | |
3178 | boundary = None | |
3179 | ||
3180 | return out, content_type | |
3181 | ||
3182 | ||
304ad45a | 3183 | def variadic(x, allowed_types=(str, bytes, dict)): |
3184 | return x if isinstance(x, collections.abc.Iterable) and not isinstance(x, allowed_types) else (x,) | |
3185 | ||
3186 | ||
86296ad2 | 3187 | def dict_get(d, key_or_keys, default=None, skip_false_values=True): |
a44ca5a4 | 3188 | for val in map(d.get, variadic(key_or_keys)): |
3189 | if val is not None and (val or not skip_false_values): | |
3190 | return val | |
3191 | return default | |
cbecc9b9 S |
3192 | |
3193 | ||
c4f60dd7 | 3194 | def try_call(*funcs, expected_type=None, args=[], kwargs={}): |
3195 | for f in funcs: | |
a32a9a7e | 3196 | try: |
c4f60dd7 | 3197 | val = f(*args, **kwargs) |
ab029d7e | 3198 | except (AttributeError, KeyError, TypeError, IndexError, ValueError, ZeroDivisionError): |
a32a9a7e S |
3199 | pass |
3200 | else: | |
c4f60dd7 | 3201 | if expected_type is None or isinstance(val, expected_type): |
3202 | return val | |
3203 | ||
3204 | ||
3205 | def try_get(src, getter, expected_type=None): | |
3206 | return try_call(*variadic(getter), args=(src,), expected_type=expected_type) | |
329ca3be S |
3207 | |
3208 | ||
90137ca4 | 3209 | def filter_dict(dct, cndn=lambda _, v: v is not None): |
3210 | return {k: v for k, v in dct.items() if cndn(k, v)} | |
3211 | ||
3212 | ||
6cc62232 S |
3213 | def merge_dicts(*dicts): |
3214 | merged = {} | |
3215 | for a_dict in dicts: | |
3216 | for k, v in a_dict.items(): | |
90137ca4 | 3217 | if (v is not None and k not in merged |
3218 | or isinstance(v, str) and merged[k] == ''): | |
6cc62232 S |
3219 | merged[k] = v |
3220 | return merged | |
3221 | ||
3222 | ||
8e60dc75 | 3223 | def encode_compat_str(string, encoding=preferredencoding(), errors='strict'): |
14f25df2 | 3224 | return string if isinstance(string, str) else str(string, encoding, errors) |
8e60dc75 | 3225 | |
16392824 | 3226 | |
a1a530b0 PH |
3227 | US_RATINGS = { |
3228 | 'G': 0, | |
3229 | 'PG': 10, | |
3230 | 'PG-13': 13, | |
3231 | 'R': 16, | |
3232 | 'NC': 18, | |
3233 | } | |
fac55558 PH |
3234 | |
3235 | ||
a8795327 | 3236 | TV_PARENTAL_GUIDELINES = { |
5a16c9d9 RA |
3237 | 'TV-Y': 0, |
3238 | 'TV-Y7': 7, | |
3239 | 'TV-G': 0, | |
3240 | 'TV-PG': 0, | |
3241 | 'TV-14': 14, | |
3242 | 'TV-MA': 17, | |
a8795327 S |
3243 | } |
3244 | ||
3245 | ||
146c80e2 | 3246 | def parse_age_limit(s): |
19a03940 | 3247 | # isinstance(False, int) is True. So type() must be used instead |
c487cf00 | 3248 | if type(s) is int: # noqa: E721 |
a8795327 | 3249 | return s if 0 <= s <= 21 else None |
19a03940 | 3250 | elif not isinstance(s, str): |
d838b1bd | 3251 | return None |
146c80e2 | 3252 | m = re.match(r'^(?P<age>\d{1,2})\+?$', s) |
a8795327 S |
3253 | if m: |
3254 | return int(m.group('age')) | |
5c5fae6d | 3255 | s = s.upper() |
a8795327 S |
3256 | if s in US_RATINGS: |
3257 | return US_RATINGS[s] | |
5a16c9d9 | 3258 | m = re.match(r'^TV[_-]?(%s)$' % '|'.join(k[3:] for k in TV_PARENTAL_GUIDELINES), s) |
b8361187 | 3259 | if m: |
5a16c9d9 | 3260 | return TV_PARENTAL_GUIDELINES['TV-' + m.group(1)] |
b8361187 | 3261 | return None |
146c80e2 S |
3262 | |
3263 | ||
fac55558 | 3264 | def strip_jsonp(code): |
609a61e3 | 3265 | return re.sub( |
5552c9eb | 3266 | r'''(?sx)^ |
e9c671d5 | 3267 | (?:window\.)?(?P<func_name>[a-zA-Z0-9_.$]*) |
5552c9eb YCH |
3268 | (?:\s*&&\s*(?P=func_name))? |
3269 | \s*\(\s*(?P<callback_data>.*)\);? | |
3270 | \s*?(?://[^\n]*)*$''', | |
3271 | r'\g<callback_data>', code) | |
478c2c61 PH |
3272 | |
3273 | ||
8f53dc44 | 3274 | def js_to_json(code, vars={}, *, strict=False): |
5c610515 | 3275 | # vars is a dict of var, val pairs to substitute |
c843e685 | 3276 | COMMENT_RE = r'/\*(?:(?!\*/).)*?\*/|//[^\n]*\n' |
86e5f3ed | 3277 | SKIP_RE = fr'\s*(?:{COMMENT_RE})?\s*' |
4195096e | 3278 | INTEGER_TABLE = ( |
86e5f3ed | 3279 | (fr'(?s)^(0[xX][0-9a-fA-F]+){SKIP_RE}:?$', 16), |
3280 | (fr'(?s)^(0+[0-7]+){SKIP_RE}:?$', 8), | |
4195096e S |
3281 | ) |
3282 | ||
e05f6939 | 3283 | def fix_kv(m): |
e7b6d122 PH |
3284 | v = m.group(0) |
3285 | if v in ('true', 'false', 'null'): | |
3286 | return v | |
421ddcb8 C |
3287 | elif v in ('undefined', 'void 0'): |
3288 | return 'null' | |
8bdd16b4 | 3289 | elif v.startswith('/*') or v.startswith('//') or v.startswith('!') or v == ',': |
bd1e4844 | 3290 | return "" |
3291 | ||
3292 | if v[0] in ("'", '"'): | |
3293 | v = re.sub(r'(?s)\\.|"', lambda m: { | |
e7b6d122 | 3294 | '"': '\\"', |
bd1e4844 | 3295 | "\\'": "'", |
3296 | '\\\n': '', | |
3297 | '\\x': '\\u00', | |
3298 | }.get(m.group(0), m.group(0)), v[1:-1]) | |
8bdd16b4 | 3299 | else: |
3300 | for regex, base in INTEGER_TABLE: | |
3301 | im = re.match(regex, v) | |
3302 | if im: | |
3303 | i = int(im.group(1), base) | |
3304 | return '"%d":' % i if v.endswith(':') else '%d' % i | |
89ac4a19 | 3305 | |
5c610515 | 3306 | if v in vars: |
f55523cf | 3307 | return json.dumps(vars[v]) |
8f53dc44 | 3308 | if strict: |
3309 | raise ValueError(f'Unknown value: {v}') | |
5c610515 | 3310 | |
e7b6d122 | 3311 | return '"%s"' % v |
e05f6939 | 3312 | |
8072ef2b | 3313 | def create_map(mobj): |
3314 | return json.dumps(dict(json.loads(js_to_json(mobj.group(1) or '[]', vars=vars)))) | |
3315 | ||
8072ef2b | 3316 | code = re.sub(r'new Map\((\[.*?\])?\)', create_map, code) |
8f53dc44 | 3317 | if not strict: |
3318 | code = re.sub(r'new Date\((".+")\)', r'\g<1>', code) | |
f55523cf | 3319 | code = re.sub(r'new \w+\((.*?)\)', lambda m: json.dumps(m.group(0)), code) |
febff4c1 | 3320 | |
bd1e4844 | 3321 | return re.sub(r'''(?sx) |
3322 | "(?:[^"\\]*(?:\\\\|\\['"nurtbfx/\n]))*[^"\\]*"| | |
3323 | '(?:[^'\\]*(?:\\\\|\\['"nurtbfx/\n]))*[^'\\]*'| | |
4195096e | 3324 | {comment}|,(?={skip}[\]}}])| |
421ddcb8 | 3325 | void\s0|(?:(?<![0-9])[eE]|[a-df-zA-DF-Z_$])[.a-zA-Z_$0-9]*| |
4195096e | 3326 | \b(?:0[xX][0-9a-fA-F]+|0+[0-7]+)(?:{skip}:)?| |
8bdd16b4 | 3327 | [0-9]+(?={skip}:)| |
3328 | !+ | |
4195096e | 3329 | '''.format(comment=COMMENT_RE, skip=SKIP_RE), fix_kv, code) |
e05f6939 PH |
3330 | |
3331 | ||
478c2c61 PH |
3332 | def qualities(quality_ids): |
3333 | """ Get a numeric quality value out of a list of possible values """ | |
3334 | def q(qid): | |
3335 | try: | |
3336 | return quality_ids.index(qid) | |
3337 | except ValueError: | |
3338 | return -1 | |
3339 | return q | |
3340 | ||
acd69589 | 3341 | |
8aa0e7cd | 3342 | POSTPROCESS_WHEN = ('pre_process', 'after_filter', 'before_dl', 'post_process', 'after_move', 'after_video', 'playlist') |
1e43a6f7 | 3343 | |
3344 | ||
de6000d9 | 3345 | DEFAULT_OUTTMPL = { |
3346 | 'default': '%(title)s [%(id)s].%(ext)s', | |
72755351 | 3347 | 'chapter': '%(title)s - %(section_number)03d %(section_title)s [%(id)s].%(ext)s', |
de6000d9 | 3348 | } |
3349 | OUTTMPL_TYPES = { | |
72755351 | 3350 | 'chapter': None, |
de6000d9 | 3351 | 'subtitle': None, |
3352 | 'thumbnail': None, | |
3353 | 'description': 'description', | |
3354 | 'annotation': 'annotations.xml', | |
3355 | 'infojson': 'info.json', | |
08438d2c | 3356 | 'link': None, |
3b603dbd | 3357 | 'pl_video': None, |
5112f26a | 3358 | 'pl_thumbnail': None, |
de6000d9 | 3359 | 'pl_description': 'description', |
3360 | 'pl_infojson': 'info.json', | |
3361 | } | |
0a871f68 | 3362 | |
143db31d | 3363 | # As of [1] format syntax is: |
3364 | # %[mapping_key][conversion_flags][minimum_width][.precision][length_modifier]type | |
3365 | # 1. https://docs.python.org/2/library/stdtypes.html#string-formatting | |
901130bb | 3366 | STR_FORMAT_RE_TMPL = r'''(?x) |
3367 | (?<!%)(?P<prefix>(?:%%)*) | |
143db31d | 3368 | % |
524e2e4f | 3369 | (?P<has_key>\((?P<key>{0})\))? |
752cda38 | 3370 | (?P<format> |
524e2e4f | 3371 | (?P<conversion>[#0\-+ ]+)? |
3372 | (?P<min_width>\d+)? | |
3373 | (?P<precision>\.\d+)? | |
3374 | (?P<len_mod>[hlL])? # unused in python | |
901130bb | 3375 | {1} # conversion type |
752cda38 | 3376 | ) |
143db31d | 3377 | ''' |
3378 | ||
7d1eb38a | 3379 | |
901130bb | 3380 | STR_FORMAT_TYPES = 'diouxXeEfFgGcrs' |
a020a0dc | 3381 | |
7d1eb38a | 3382 | |
a020a0dc PH |
3383 | def limit_length(s, length): |
3384 | """ Add ellipses to overly long strings """ | |
3385 | if s is None: | |
3386 | return None | |
3387 | ELLIPSES = '...' | |
3388 | if len(s) > length: | |
3389 | return s[:length - len(ELLIPSES)] + ELLIPSES | |
3390 | return s | |
48844745 PH |
3391 | |
3392 | ||
3393 | def version_tuple(v): | |
5f9b8394 | 3394 | return tuple(int(e) for e in re.split(r'[-.]', v)) |
48844745 PH |
3395 | |
3396 | ||
3397 | def is_outdated_version(version, limit, assume_new=True): | |
3398 | if not version: | |
3399 | return not assume_new | |
3400 | try: | |
3401 | return version_tuple(version) < version_tuple(limit) | |
3402 | except ValueError: | |
3403 | return not assume_new | |
732ea2f0 PH |
3404 | |
3405 | ||
3406 | def ytdl_is_updateable(): | |
7a5c1cfe | 3407 | """ Returns if yt-dlp can be updated with -U """ |
735d865e | 3408 | |
5d535b4a | 3409 | from .update import is_non_updateable |
732ea2f0 | 3410 | |
5d535b4a | 3411 | return not is_non_updateable() |
7d4111ed PH |
3412 | |
3413 | ||
3414 | def args_to_str(args): | |
3415 | # Get a short string representation for a subprocess command | |
702ccf2d | 3416 | return ' '.join(compat_shlex_quote(a) for a in args) |
2ccd1b10 PH |
3417 | |
3418 | ||
9b9c5355 | 3419 | def error_to_compat_str(err): |
cfb0511d | 3420 | return str(err) |
fdae2358 S |
3421 | |
3422 | ||
a44ca5a4 | 3423 | def error_to_str(err): |
3424 | return f'{type(err).__name__}: {err}' | |
3425 | ||
3426 | ||
c460bdd5 | 3427 | def mimetype2ext(mt): |
eb9ee194 S |
3428 | if mt is None: |
3429 | return None | |
3430 | ||
9359f3d4 F |
3431 | mt, _, params = mt.partition(';') |
3432 | mt = mt.strip() | |
3433 | ||
3434 | FULL_MAP = { | |
765ac263 | 3435 | 'audio/mp4': 'm4a', |
6c33d24b YCH |
3436 | # Per RFC 3003, audio/mpeg can be .mp1, .mp2 or .mp3. Here use .mp3 as |
3437 | # it's the most popular one | |
3438 | 'audio/mpeg': 'mp3', | |
ba39289d | 3439 | 'audio/x-wav': 'wav', |
9359f3d4 F |
3440 | 'audio/wav': 'wav', |
3441 | 'audio/wave': 'wav', | |
3442 | } | |
3443 | ||
3444 | ext = FULL_MAP.get(mt) | |
765ac263 JMF |
3445 | if ext is not None: |
3446 | return ext | |
3447 | ||
9359f3d4 | 3448 | SUBTYPE_MAP = { |
f6861ec9 | 3449 | '3gpp': '3gp', |
cafcf657 | 3450 | 'smptett+xml': 'tt', |
cafcf657 | 3451 | 'ttaf+xml': 'dfxp', |
a0d8d704 | 3452 | 'ttml+xml': 'ttml', |
f6861ec9 | 3453 | 'x-flv': 'flv', |
a0d8d704 | 3454 | 'x-mp4-fragmented': 'mp4', |
d4f05d47 | 3455 | 'x-ms-sami': 'sami', |
a0d8d704 | 3456 | 'x-ms-wmv': 'wmv', |
b4173f15 RA |
3457 | 'mpegurl': 'm3u8', |
3458 | 'x-mpegurl': 'm3u8', | |
3459 | 'vnd.apple.mpegurl': 'm3u8', | |
3460 | 'dash+xml': 'mpd', | |
b4173f15 | 3461 | 'f4m+xml': 'f4m', |
f164b971 | 3462 | 'hds+xml': 'f4m', |
e910fe2f | 3463 | 'vnd.ms-sstr+xml': 'ism', |
c2b2c7e1 | 3464 | 'quicktime': 'mov', |
98ce1a3f | 3465 | 'mp2t': 'ts', |
39e7107d | 3466 | 'x-wav': 'wav', |
9359f3d4 F |
3467 | 'filmstrip+json': 'fs', |
3468 | 'svg+xml': 'svg', | |
3469 | } | |
3470 | ||
3471 | _, _, subtype = mt.rpartition('/') | |
3472 | ext = SUBTYPE_MAP.get(subtype.lower()) | |
3473 | if ext is not None: | |
3474 | return ext | |
3475 | ||
3476 | SUFFIX_MAP = { | |
3477 | 'json': 'json', | |
3478 | 'xml': 'xml', | |
3479 | 'zip': 'zip', | |
3480 | 'gzip': 'gz', | |
3481 | } | |
3482 | ||
3483 | _, _, suffix = subtype.partition('+') | |
3484 | ext = SUFFIX_MAP.get(suffix) | |
3485 | if ext is not None: | |
3486 | return ext | |
3487 | ||
3488 | return subtype.replace('+', '.') | |
c460bdd5 PH |
3489 | |
3490 | ||
2814f12b THD |
3491 | def ext2mimetype(ext_or_url): |
3492 | if not ext_or_url: | |
3493 | return None | |
3494 | if '.' not in ext_or_url: | |
3495 | ext_or_url = f'file.{ext_or_url}' | |
3496 | return mimetypes.guess_type(ext_or_url)[0] | |
3497 | ||
3498 | ||
4f3c5e06 | 3499 | def parse_codecs(codecs_str): |
3500 | # http://tools.ietf.org/html/rfc6381 | |
3501 | if not codecs_str: | |
3502 | return {} | |
a0566bbf | 3503 | split_codecs = list(filter(None, map( |
dbf5416a | 3504 | str.strip, codecs_str.strip().strip(',').split(',')))) |
3fe75fdc | 3505 | vcodec, acodec, scodec, hdr = None, None, None, None |
a0566bbf | 3506 | for full_codec in split_codecs: |
d816f61f | 3507 | parts = re.sub(r'0+(?=\d)', '', full_codec).split('.') |
3508 | if parts[0] in ('avc1', 'avc2', 'avc3', 'avc4', 'vp9', 'vp8', 'hev1', 'hev2', | |
3509 | 'h263', 'h264', 'mp4v', 'hvc1', 'av1', 'theora', 'dvh1', 'dvhe'): | |
3510 | if vcodec: | |
3511 | continue | |
3512 | vcodec = full_codec | |
3513 | if parts[0] in ('dvh1', 'dvhe'): | |
3514 | hdr = 'DV' | |
3515 | elif parts[0] == 'av1' and traverse_obj(parts, 3) == '10': | |
3516 | hdr = 'HDR10' | |
3517 | elif parts[:2] == ['vp9', '2']: | |
3518 | hdr = 'HDR10' | |
3519 | elif parts[0] in ('flac', 'mp4a', 'opus', 'vorbis', 'mp3', 'aac', | |
3520 | 'ac-3', 'ec-3', 'eac3', 'dtsc', 'dtse', 'dtsh', 'dtsl'): | |
3521 | acodec = acodec or full_codec | |
3522 | elif parts[0] in ('stpp', 'wvtt'): | |
3523 | scodec = scodec or full_codec | |
4f3c5e06 | 3524 | else: |
19a03940 | 3525 | write_string(f'WARNING: Unknown codec {full_codec}\n') |
3fe75fdc | 3526 | if vcodec or acodec or scodec: |
4f3c5e06 | 3527 | return { |
3528 | 'vcodec': vcodec or 'none', | |
3529 | 'acodec': acodec or 'none', | |
176f1866 | 3530 | 'dynamic_range': hdr, |
3fe75fdc | 3531 | **({'scodec': scodec} if scodec is not None else {}), |
4f3c5e06 | 3532 | } |
b69fd25c | 3533 | elif len(split_codecs) == 2: |
3534 | return { | |
3535 | 'vcodec': split_codecs[0], | |
3536 | 'acodec': split_codecs[1], | |
3537 | } | |
4f3c5e06 | 3538 | return {} |
3539 | ||
3540 | ||
fc61aff4 LL |
3541 | def get_compatible_ext(*, vcodecs, acodecs, vexts, aexts, preferences=None): |
3542 | assert len(vcodecs) == len(vexts) and len(acodecs) == len(aexts) | |
3543 | ||
3544 | allow_mkv = not preferences or 'mkv' in preferences | |
3545 | ||
3546 | if allow_mkv and max(len(acodecs), len(vcodecs)) > 1: | |
3547 | return 'mkv' # TODO: any other format allows this? | |
3548 | ||
3549 | # TODO: All codecs supported by parse_codecs isn't handled here | |
3550 | COMPATIBLE_CODECS = { | |
3551 | 'mp4': { | |
3552 | 'av1', 'hevc', 'avc1', 'mp4a', # fourcc (m3u8, mpd) | |
81b6102d | 3553 | 'h264', 'aacl', 'ec-3', # Set in ISM |
fc61aff4 LL |
3554 | }, |
3555 | 'webm': { | |
3556 | 'av1', 'vp9', 'vp8', 'opus', 'vrbs', | |
3557 | 'vp9x', 'vp8x', # in the webm spec | |
3558 | }, | |
3559 | } | |
3560 | ||
8f84770a | 3561 | sanitize_codec = functools.partial(try_get, getter=lambda x: x[0].split('.')[0].replace('0', '')) |
3562 | vcodec, acodec = sanitize_codec(vcodecs), sanitize_codec(acodecs) | |
fc61aff4 LL |
3563 | |
3564 | for ext in preferences or COMPATIBLE_CODECS.keys(): | |
3565 | codec_set = COMPATIBLE_CODECS.get(ext, set()) | |
3566 | if ext == 'mkv' or codec_set.issuperset((vcodec, acodec)): | |
3567 | return ext | |
3568 | ||
3569 | COMPATIBLE_EXTS = ( | |
3570 | {'mp3', 'mp4', 'm4a', 'm4p', 'm4b', 'm4r', 'm4v', 'ismv', 'isma', 'mov'}, | |
3571 | {'webm'}, | |
3572 | ) | |
3573 | for ext in preferences or vexts: | |
3574 | current_exts = {ext, *vexts, *aexts} | |
3575 | if ext == 'mkv' or current_exts == {ext} or any( | |
3576 | ext_sets.issuperset(current_exts) for ext_sets in COMPATIBLE_EXTS): | |
3577 | return ext | |
3578 | return 'mkv' if allow_mkv else preferences[-1] | |
3579 | ||
3580 | ||
2ccd1b10 | 3581 | def urlhandle_detect_ext(url_handle): |
79298173 | 3582 | getheader = url_handle.headers.get |
2ccd1b10 | 3583 | |
b55ee18f PH |
3584 | cd = getheader('Content-Disposition') |
3585 | if cd: | |
3586 | m = re.match(r'attachment;\s*filename="(?P<filename>[^"]+)"', cd) | |
3587 | if m: | |
3588 | e = determine_ext(m.group('filename'), default_ext=None) | |
3589 | if e: | |
3590 | return e | |
3591 | ||
c460bdd5 | 3592 | return mimetype2ext(getheader('Content-Type')) |
05900629 PH |
3593 | |
3594 | ||
1e399778 YCH |
3595 | def encode_data_uri(data, mime_type): |
3596 | return 'data:%s;base64,%s' % (mime_type, base64.b64encode(data).decode('ascii')) | |
3597 | ||
3598 | ||
05900629 | 3599 | def age_restricted(content_limit, age_limit): |
6ec6cb4e | 3600 | """ Returns True iff the content should be blocked """ |
05900629 PH |
3601 | |
3602 | if age_limit is None: # No limit set | |
3603 | return False | |
3604 | if content_limit is None: | |
3605 | return False # Content available for everyone | |
3606 | return age_limit < content_limit | |
61ca9a80 PH |
3607 | |
3608 | ||
88f60feb | 3609 | # List of known byte-order-marks (BOM) |
a904a7f8 L |
3610 | BOMS = [ |
3611 | (b'\xef\xbb\xbf', 'utf-8'), | |
3612 | (b'\x00\x00\xfe\xff', 'utf-32-be'), | |
3613 | (b'\xff\xfe\x00\x00', 'utf-32-le'), | |
3614 | (b'\xff\xfe', 'utf-16-le'), | |
3615 | (b'\xfe\xff', 'utf-16-be'), | |
3616 | ] | |
a904a7f8 L |
3617 | |
3618 | ||
61ca9a80 PH |
3619 | def is_html(first_bytes): |
3620 | """ Detect whether a file contains HTML by examining its first bytes. """ | |
3621 | ||
80e8493e | 3622 | encoding = 'utf-8' |
61ca9a80 | 3623 | for bom, enc in BOMS: |
80e8493e | 3624 | while first_bytes.startswith(bom): |
3625 | encoding, first_bytes = enc, first_bytes[len(bom):] | |
61ca9a80 | 3626 | |
80e8493e | 3627 | return re.match(r'^\s*<', first_bytes.decode(encoding, 'replace')) |
a055469f PH |
3628 | |
3629 | ||
3630 | def determine_protocol(info_dict): | |
3631 | protocol = info_dict.get('protocol') | |
3632 | if protocol is not None: | |
3633 | return protocol | |
3634 | ||
7de837a5 | 3635 | url = sanitize_url(info_dict['url']) |
a055469f PH |
3636 | if url.startswith('rtmp'): |
3637 | return 'rtmp' | |
3638 | elif url.startswith('mms'): | |
3639 | return 'mms' | |
3640 | elif url.startswith('rtsp'): | |
3641 | return 'rtsp' | |
3642 | ||
3643 | ext = determine_ext(url) | |
3644 | if ext == 'm3u8': | |
deae7c17 | 3645 | return 'm3u8' if info_dict.get('is_live') else 'm3u8_native' |
a055469f PH |
3646 | elif ext == 'f4m': |
3647 | return 'f4m' | |
3648 | ||
14f25df2 | 3649 | return urllib.parse.urlparse(url).scheme |
cfb56d1a PH |
3650 | |
3651 | ||
c5e3f849 | 3652 | def render_table(header_row, data, delim=False, extra_gap=0, hide_empty=False): |
3653 | """ Render a list of rows, each as a list of values. | |
3654 | Text after a \t will be right aligned """ | |
ec11a9f4 | 3655 | def width(string): |
c5e3f849 | 3656 | return len(remove_terminal_sequences(string).replace('\t', '')) |
76d321f6 | 3657 | |
3658 | def get_max_lens(table): | |
ec11a9f4 | 3659 | return [max(width(str(v)) for v in col) for col in zip(*table)] |
76d321f6 | 3660 | |
3661 | def filter_using_list(row, filterArray): | |
d16df59d | 3662 | return [col for take, col in itertools.zip_longest(filterArray, row, fillvalue=True) if take] |
76d321f6 | 3663 | |
d16df59d | 3664 | max_lens = get_max_lens(data) if hide_empty else [] |
3665 | header_row = filter_using_list(header_row, max_lens) | |
3666 | data = [filter_using_list(row, max_lens) for row in data] | |
76d321f6 | 3667 | |
cfb56d1a | 3668 | table = [header_row] + data |
76d321f6 | 3669 | max_lens = get_max_lens(table) |
c5e3f849 | 3670 | extra_gap += 1 |
76d321f6 | 3671 | if delim: |
c5e3f849 | 3672 | table = [header_row, [delim * (ml + extra_gap) for ml in max_lens]] + data |
1ed7953a | 3673 | table[1][-1] = table[1][-1][:-extra_gap * len(delim)] # Remove extra_gap from end of delimiter |
ec11a9f4 | 3674 | for row in table: |
3675 | for pos, text in enumerate(map(str, row)): | |
c5e3f849 | 3676 | if '\t' in text: |
3677 | row[pos] = text.replace('\t', ' ' * (max_lens[pos] - width(text))) + ' ' * extra_gap | |
3678 | else: | |
3679 | row[pos] = text + ' ' * (max_lens[pos] - width(text) + extra_gap) | |
3680 | ret = '\n'.join(''.join(row).rstrip() for row in table) | |
ec11a9f4 | 3681 | return ret |
347de493 PH |
3682 | |
3683 | ||
8f18aca8 | 3684 | def _match_one(filter_part, dct, incomplete): |
77b87f05 | 3685 | # TODO: Generalize code with YoutubeDL._build_format_filter |
a047eeb6 | 3686 | STRING_OPERATORS = { |
3687 | '*=': operator.contains, | |
3688 | '^=': lambda attr, value: attr.startswith(value), | |
3689 | '$=': lambda attr, value: attr.endswith(value), | |
3690 | '~=': lambda attr, value: re.search(value, attr), | |
3691 | } | |
347de493 | 3692 | COMPARISON_OPERATORS = { |
a047eeb6 | 3693 | **STRING_OPERATORS, |
3694 | '<=': operator.le, # "<=" must be defined above "<" | |
347de493 | 3695 | '<': operator.lt, |
347de493 | 3696 | '>=': operator.ge, |
a047eeb6 | 3697 | '>': operator.gt, |
347de493 | 3698 | '=': operator.eq, |
347de493 | 3699 | } |
a047eeb6 | 3700 | |
6db9c4d5 | 3701 | if isinstance(incomplete, bool): |
3702 | is_incomplete = lambda _: incomplete | |
3703 | else: | |
3704 | is_incomplete = lambda k: k in incomplete | |
3705 | ||
64fa820c | 3706 | operator_rex = re.compile(r'''(?x) |
347de493 | 3707 | (?P<key>[a-z_]+) |
77b87f05 | 3708 | \s*(?P<negation>!\s*)?(?P<op>%s)(?P<none_inclusive>\s*\?)?\s* |
347de493 | 3709 | (?: |
a047eeb6 | 3710 | (?P<quote>["\'])(?P<quotedstrval>.+?)(?P=quote)| |
3711 | (?P<strval>.+?) | |
347de493 | 3712 | ) |
347de493 | 3713 | ''' % '|'.join(map(re.escape, COMPARISON_OPERATORS.keys()))) |
64fa820c | 3714 | m = operator_rex.fullmatch(filter_part.strip()) |
347de493 | 3715 | if m: |
18f96d12 | 3716 | m = m.groupdict() |
3717 | unnegated_op = COMPARISON_OPERATORS[m['op']] | |
3718 | if m['negation']: | |
77b87f05 MT |
3719 | op = lambda attr, value: not unnegated_op(attr, value) |
3720 | else: | |
3721 | op = unnegated_op | |
18f96d12 | 3722 | comparison_value = m['quotedstrval'] or m['strval'] or m['intval'] |
3723 | if m['quote']: | |
3724 | comparison_value = comparison_value.replace(r'\%s' % m['quote'], m['quote']) | |
3725 | actual_value = dct.get(m['key']) | |
3726 | numeric_comparison = None | |
f9934b96 | 3727 | if isinstance(actual_value, (int, float)): |
e5a088dc S |
3728 | # If the original field is a string and matching comparisonvalue is |
3729 | # a number we should respect the origin of the original field | |
3730 | # and process comparison value as a string (see | |
18f96d12 | 3731 | # https://github.com/ytdl-org/youtube-dl/issues/11082) |
347de493 | 3732 | try: |
18f96d12 | 3733 | numeric_comparison = int(comparison_value) |
347de493 | 3734 | except ValueError: |
18f96d12 | 3735 | numeric_comparison = parse_filesize(comparison_value) |
3736 | if numeric_comparison is None: | |
3737 | numeric_comparison = parse_filesize(f'{comparison_value}B') | |
3738 | if numeric_comparison is None: | |
3739 | numeric_comparison = parse_duration(comparison_value) | |
3740 | if numeric_comparison is not None and m['op'] in STRING_OPERATORS: | |
3741 | raise ValueError('Operator %s only supports string values!' % m['op']) | |
347de493 | 3742 | if actual_value is None: |
6db9c4d5 | 3743 | return is_incomplete(m['key']) or m['none_inclusive'] |
18f96d12 | 3744 | return op(actual_value, comparison_value if numeric_comparison is None else numeric_comparison) |
347de493 PH |
3745 | |
3746 | UNARY_OPERATORS = { | |
1cc47c66 S |
3747 | '': lambda v: (v is True) if isinstance(v, bool) else (v is not None), |
3748 | '!': lambda v: (v is False) if isinstance(v, bool) else (v is None), | |
347de493 | 3749 | } |
64fa820c | 3750 | operator_rex = re.compile(r'''(?x) |
347de493 | 3751 | (?P<op>%s)\s*(?P<key>[a-z_]+) |
347de493 | 3752 | ''' % '|'.join(map(re.escape, UNARY_OPERATORS.keys()))) |
64fa820c | 3753 | m = operator_rex.fullmatch(filter_part.strip()) |
347de493 PH |
3754 | if m: |
3755 | op = UNARY_OPERATORS[m.group('op')] | |
3756 | actual_value = dct.get(m.group('key')) | |
6db9c4d5 | 3757 | if is_incomplete(m.group('key')) and actual_value is None: |
8f18aca8 | 3758 | return True |
347de493 PH |
3759 | return op(actual_value) |
3760 | ||
3761 | raise ValueError('Invalid filter part %r' % filter_part) | |
3762 | ||
3763 | ||
8f18aca8 | 3764 | def match_str(filter_str, dct, incomplete=False): |
6db9c4d5 | 3765 | """ Filter a dictionary with a simple string syntax. |
3766 | @returns Whether the filter passes | |
3767 | @param incomplete Set of keys that is expected to be missing from dct. | |
3768 | Can be True/False to indicate all/none of the keys may be missing. | |
3769 | All conditions on incomplete keys pass if the key is missing | |
8f18aca8 | 3770 | """ |
347de493 | 3771 | return all( |
8f18aca8 | 3772 | _match_one(filter_part.replace(r'\&', '&'), dct, incomplete) |
a047eeb6 | 3773 | for filter_part in re.split(r'(?<!\\)&', filter_str)) |
347de493 PH |
3774 | |
3775 | ||
b1a7cd05 | 3776 | def match_filter_func(filters): |
3777 | if not filters: | |
d1b5f70b | 3778 | return None |
492272fe | 3779 | filters = set(variadic(filters)) |
d1b5f70b | 3780 | |
492272fe | 3781 | interactive = '-' in filters |
3782 | if interactive: | |
3783 | filters.remove('-') | |
3784 | ||
3785 | def _match_func(info_dict, incomplete=False): | |
3786 | if not filters or any(match_str(f, info_dict, incomplete) for f in filters): | |
3787 | return NO_DEFAULT if interactive and not incomplete else None | |
347de493 | 3788 | else: |
3bec830a | 3789 | video_title = info_dict.get('title') or info_dict.get('id') or 'entry' |
b1a7cd05 | 3790 | filter_str = ') | ('.join(map(str.strip, filters)) |
3791 | return f'{video_title} does not pass filter ({filter_str}), skipping ..' | |
347de493 | 3792 | return _match_func |
91410c9b PH |
3793 | |
3794 | ||
f2df4071 | 3795 | class download_range_func: |
3796 | def __init__(self, chapters, ranges): | |
3797 | self.chapters, self.ranges = chapters, ranges | |
3798 | ||
3799 | def __call__(self, info_dict, ydl): | |
0500ee3d | 3800 | if not self.ranges and not self.chapters: |
3801 | yield {} | |
3802 | ||
5ec1b6b7 | 3803 | warning = ('There are no chapters matching the regex' if info_dict.get('chapters') |
56ba69e4 | 3804 | else 'Cannot match chapters since chapter information is unavailable') |
f2df4071 | 3805 | for regex in self.chapters or []: |
5ec1b6b7 | 3806 | for i, chapter in enumerate(info_dict.get('chapters') or []): |
3807 | if re.search(regex, chapter['title']): | |
3808 | warning = None | |
3809 | yield {**chapter, 'index': i} | |
f2df4071 | 3810 | if self.chapters and warning: |
5ec1b6b7 | 3811 | ydl.to_screen(f'[info] {info_dict["id"]}: {warning}') |
3812 | ||
f2df4071 | 3813 | yield from ({'start_time': start, 'end_time': end} for start, end in self.ranges or []) |
5ec1b6b7 | 3814 | |
f2df4071 | 3815 | def __eq__(self, other): |
3816 | return (isinstance(other, download_range_func) | |
3817 | and self.chapters == other.chapters and self.ranges == other.ranges) | |
5ec1b6b7 | 3818 | |
3819 | ||
bf6427d2 YCH |
3820 | def parse_dfxp_time_expr(time_expr): |
3821 | if not time_expr: | |
d631d5f9 | 3822 | return |
bf6427d2 | 3823 | |
1d485a1a | 3824 | mobj = re.match(rf'^(?P<time_offset>{NUMBER_RE})s?$', time_expr) |
bf6427d2 YCH |
3825 | if mobj: |
3826 | return float(mobj.group('time_offset')) | |
3827 | ||
db2fe38b | 3828 | mobj = re.match(r'^(\d+):(\d\d):(\d\d(?:(?:\.|:)\d+)?)$', time_expr) |
bf6427d2 | 3829 | if mobj: |
db2fe38b | 3830 | return 3600 * int(mobj.group(1)) + 60 * int(mobj.group(2)) + float(mobj.group(3).replace(':', '.')) |
bf6427d2 YCH |
3831 | |
3832 | ||
c1c924ab | 3833 | def srt_subtitles_timecode(seconds): |
aa7785f8 | 3834 | return '%02d:%02d:%02d,%03d' % timetuple_from_msec(seconds * 1000) |
3835 | ||
3836 | ||
3837 | def ass_subtitles_timecode(seconds): | |
3838 | time = timetuple_from_msec(seconds * 1000) | |
3839 | return '%01d:%02d:%02d.%02d' % (*time[:-1], time.milliseconds / 10) | |
bf6427d2 YCH |
3840 | |
3841 | ||
3842 | def dfxp2srt(dfxp_data): | |
3869028f YCH |
3843 | ''' |
3844 | @param dfxp_data A bytes-like object containing DFXP data | |
3845 | @returns A unicode object containing converted SRT data | |
3846 | ''' | |
5b995f71 | 3847 | LEGACY_NAMESPACES = ( |
3869028f YCH |
3848 | (b'http://www.w3.org/ns/ttml', [ |
3849 | b'http://www.w3.org/2004/11/ttaf1', | |
3850 | b'http://www.w3.org/2006/04/ttaf1', | |
3851 | b'http://www.w3.org/2006/10/ttaf1', | |
5b995f71 | 3852 | ]), |
3869028f YCH |
3853 | (b'http://www.w3.org/ns/ttml#styling', [ |
3854 | b'http://www.w3.org/ns/ttml#style', | |
5b995f71 RA |
3855 | ]), |
3856 | ) | |
3857 | ||
3858 | SUPPORTED_STYLING = [ | |
3859 | 'color', | |
3860 | 'fontFamily', | |
3861 | 'fontSize', | |
3862 | 'fontStyle', | |
3863 | 'fontWeight', | |
3864 | 'textDecoration' | |
3865 | ] | |
3866 | ||
4e335771 | 3867 | _x = functools.partial(xpath_with_ns, ns_map={ |
261f4730 | 3868 | 'xml': 'http://www.w3.org/XML/1998/namespace', |
4e335771 | 3869 | 'ttml': 'http://www.w3.org/ns/ttml', |
5b995f71 | 3870 | 'tts': 'http://www.w3.org/ns/ttml#styling', |
4e335771 | 3871 | }) |
bf6427d2 | 3872 | |
5b995f71 RA |
3873 | styles = {} |
3874 | default_style = {} | |
3875 | ||
86e5f3ed | 3876 | class TTMLPElementParser: |
5b995f71 RA |
3877 | _out = '' |
3878 | _unclosed_elements = [] | |
3879 | _applied_styles = [] | |
bf6427d2 | 3880 | |
2b14cb56 | 3881 | def start(self, tag, attrib): |
5b995f71 RA |
3882 | if tag in (_x('ttml:br'), 'br'): |
3883 | self._out += '\n' | |
3884 | else: | |
3885 | unclosed_elements = [] | |
3886 | style = {} | |
3887 | element_style_id = attrib.get('style') | |
3888 | if default_style: | |
3889 | style.update(default_style) | |
3890 | if element_style_id: | |
3891 | style.update(styles.get(element_style_id, {})) | |
3892 | for prop in SUPPORTED_STYLING: | |
3893 | prop_val = attrib.get(_x('tts:' + prop)) | |
3894 | if prop_val: | |
3895 | style[prop] = prop_val | |
3896 | if style: | |
3897 | font = '' | |
3898 | for k, v in sorted(style.items()): | |
3899 | if self._applied_styles and self._applied_styles[-1].get(k) == v: | |
3900 | continue | |
3901 | if k == 'color': | |
3902 | font += ' color="%s"' % v | |
3903 | elif k == 'fontSize': | |
3904 | font += ' size="%s"' % v | |
3905 | elif k == 'fontFamily': | |
3906 | font += ' face="%s"' % v | |
3907 | elif k == 'fontWeight' and v == 'bold': | |
3908 | self._out += '<b>' | |
3909 | unclosed_elements.append('b') | |
3910 | elif k == 'fontStyle' and v == 'italic': | |
3911 | self._out += '<i>' | |
3912 | unclosed_elements.append('i') | |
3913 | elif k == 'textDecoration' and v == 'underline': | |
3914 | self._out += '<u>' | |
3915 | unclosed_elements.append('u') | |
3916 | if font: | |
3917 | self._out += '<font' + font + '>' | |
3918 | unclosed_elements.append('font') | |
3919 | applied_style = {} | |
3920 | if self._applied_styles: | |
3921 | applied_style.update(self._applied_styles[-1]) | |
3922 | applied_style.update(style) | |
3923 | self._applied_styles.append(applied_style) | |
3924 | self._unclosed_elements.append(unclosed_elements) | |
bf6427d2 | 3925 | |
2b14cb56 | 3926 | def end(self, tag): |
5b995f71 RA |
3927 | if tag not in (_x('ttml:br'), 'br'): |
3928 | unclosed_elements = self._unclosed_elements.pop() | |
3929 | for element in reversed(unclosed_elements): | |
3930 | self._out += '</%s>' % element | |
3931 | if unclosed_elements and self._applied_styles: | |
3932 | self._applied_styles.pop() | |
bf6427d2 | 3933 | |
2b14cb56 | 3934 | def data(self, data): |
5b995f71 | 3935 | self._out += data |
2b14cb56 | 3936 | |
3937 | def close(self): | |
5b995f71 | 3938 | return self._out.strip() |
2b14cb56 | 3939 | |
3940 | def parse_node(node): | |
3941 | target = TTMLPElementParser() | |
3942 | parser = xml.etree.ElementTree.XMLParser(target=target) | |
3943 | parser.feed(xml.etree.ElementTree.tostring(node)) | |
3944 | return parser.close() | |
bf6427d2 | 3945 | |
5b995f71 RA |
3946 | for k, v in LEGACY_NAMESPACES: |
3947 | for ns in v: | |
3948 | dfxp_data = dfxp_data.replace(ns, k) | |
3949 | ||
3869028f | 3950 | dfxp = compat_etree_fromstring(dfxp_data) |
bf6427d2 | 3951 | out = [] |
5b995f71 | 3952 | paras = dfxp.findall(_x('.//ttml:p')) or dfxp.findall('.//p') |
1b0427e6 YCH |
3953 | |
3954 | if not paras: | |
3955 | raise ValueError('Invalid dfxp/TTML subtitle') | |
bf6427d2 | 3956 | |
5b995f71 RA |
3957 | repeat = False |
3958 | while True: | |
3959 | for style in dfxp.findall(_x('.//ttml:style')): | |
261f4730 RA |
3960 | style_id = style.get('id') or style.get(_x('xml:id')) |
3961 | if not style_id: | |
3962 | continue | |
5b995f71 RA |
3963 | parent_style_id = style.get('style') |
3964 | if parent_style_id: | |
3965 | if parent_style_id not in styles: | |
3966 | repeat = True | |
3967 | continue | |
3968 | styles[style_id] = styles[parent_style_id].copy() | |
3969 | for prop in SUPPORTED_STYLING: | |
3970 | prop_val = style.get(_x('tts:' + prop)) | |
3971 | if prop_val: | |
3972 | styles.setdefault(style_id, {})[prop] = prop_val | |
3973 | if repeat: | |
3974 | repeat = False | |
3975 | else: | |
3976 | break | |
3977 | ||
3978 | for p in ('body', 'div'): | |
3979 | ele = xpath_element(dfxp, [_x('.//ttml:' + p), './/' + p]) | |
3980 | if ele is None: | |
3981 | continue | |
3982 | style = styles.get(ele.get('style')) | |
3983 | if not style: | |
3984 | continue | |
3985 | default_style.update(style) | |
3986 | ||
bf6427d2 | 3987 | for para, index in zip(paras, itertools.count(1)): |
d631d5f9 | 3988 | begin_time = parse_dfxp_time_expr(para.attrib.get('begin')) |
7dff0363 | 3989 | end_time = parse_dfxp_time_expr(para.attrib.get('end')) |
d631d5f9 YCH |
3990 | dur = parse_dfxp_time_expr(para.attrib.get('dur')) |
3991 | if begin_time is None: | |
3992 | continue | |
7dff0363 | 3993 | if not end_time: |
d631d5f9 YCH |
3994 | if not dur: |
3995 | continue | |
3996 | end_time = begin_time + dur | |
bf6427d2 YCH |
3997 | out.append('%d\n%s --> %s\n%s\n\n' % ( |
3998 | index, | |
c1c924ab YCH |
3999 | srt_subtitles_timecode(begin_time), |
4000 | srt_subtitles_timecode(end_time), | |
bf6427d2 YCH |
4001 | parse_node(para))) |
4002 | ||
4003 | return ''.join(out) | |
4004 | ||
4005 | ||
c487cf00 | 4006 | def cli_option(params, command_option, param, separator=None): |
66e289ba | 4007 | param = params.get(param) |
c487cf00 | 4008 | return ([] if param is None |
4009 | else [command_option, str(param)] if separator is None | |
4010 | else [f'{command_option}{separator}{param}']) | |
66e289ba S |
4011 | |
4012 | ||
4013 | def cli_bool_option(params, command_option, param, true_value='true', false_value='false', separator=None): | |
4014 | param = params.get(param) | |
c487cf00 | 4015 | assert param in (True, False, None) |
4016 | return cli_option({True: true_value, False: false_value}, command_option, param, separator) | |
66e289ba S |
4017 | |
4018 | ||
4019 | def cli_valueless_option(params, command_option, param, expected_value=True): | |
c487cf00 | 4020 | return [command_option] if params.get(param) == expected_value else [] |
66e289ba S |
4021 | |
4022 | ||
e92caff5 | 4023 | def cli_configuration_args(argdict, keys, default=[], use_compat=True): |
eab9b2bc | 4024 | if isinstance(argdict, (list, tuple)): # for backward compatibility |
e92caff5 | 4025 | if use_compat: |
5b1ecbb3 | 4026 | return argdict |
4027 | else: | |
4028 | argdict = None | |
eab9b2bc | 4029 | if argdict is None: |
5b1ecbb3 | 4030 | return default |
eab9b2bc | 4031 | assert isinstance(argdict, dict) |
4032 | ||
e92caff5 | 4033 | assert isinstance(keys, (list, tuple)) |
4034 | for key_list in keys: | |
e92caff5 | 4035 | arg_list = list(filter( |
4036 | lambda x: x is not None, | |
6606817a | 4037 | [argdict.get(key.lower()) for key in variadic(key_list)])) |
e92caff5 | 4038 | if arg_list: |
4039 | return [arg for args in arg_list for arg in args] | |
4040 | return default | |
66e289ba | 4041 | |
6251555f | 4042 | |
330690a2 | 4043 | def _configuration_args(main_key, argdict, exe, keys=None, default=[], use_compat=True): |
4044 | main_key, exe = main_key.lower(), exe.lower() | |
4045 | root_key = exe if main_key == exe else f'{main_key}+{exe}' | |
4046 | keys = [f'{root_key}{k}' for k in (keys or [''])] | |
4047 | if root_key in keys: | |
4048 | if main_key != exe: | |
4049 | keys.append((main_key, exe)) | |
4050 | keys.append('default') | |
4051 | else: | |
4052 | use_compat = False | |
4053 | return cli_configuration_args(argdict, keys, default, use_compat) | |
4054 | ||
66e289ba | 4055 | |
86e5f3ed | 4056 | class ISO639Utils: |
39672624 YCH |
4057 | # See http://www.loc.gov/standards/iso639-2/ISO-639-2_utf-8.txt |
4058 | _lang_map = { | |
4059 | 'aa': 'aar', | |
4060 | 'ab': 'abk', | |
4061 | 'ae': 'ave', | |
4062 | 'af': 'afr', | |
4063 | 'ak': 'aka', | |
4064 | 'am': 'amh', | |
4065 | 'an': 'arg', | |
4066 | 'ar': 'ara', | |
4067 | 'as': 'asm', | |
4068 | 'av': 'ava', | |
4069 | 'ay': 'aym', | |
4070 | 'az': 'aze', | |
4071 | 'ba': 'bak', | |
4072 | 'be': 'bel', | |
4073 | 'bg': 'bul', | |
4074 | 'bh': 'bih', | |
4075 | 'bi': 'bis', | |
4076 | 'bm': 'bam', | |
4077 | 'bn': 'ben', | |
4078 | 'bo': 'bod', | |
4079 | 'br': 'bre', | |
4080 | 'bs': 'bos', | |
4081 | 'ca': 'cat', | |
4082 | 'ce': 'che', | |
4083 | 'ch': 'cha', | |
4084 | 'co': 'cos', | |
4085 | 'cr': 'cre', | |
4086 | 'cs': 'ces', | |
4087 | 'cu': 'chu', | |
4088 | 'cv': 'chv', | |
4089 | 'cy': 'cym', | |
4090 | 'da': 'dan', | |
4091 | 'de': 'deu', | |
4092 | 'dv': 'div', | |
4093 | 'dz': 'dzo', | |
4094 | 'ee': 'ewe', | |
4095 | 'el': 'ell', | |
4096 | 'en': 'eng', | |
4097 | 'eo': 'epo', | |
4098 | 'es': 'spa', | |
4099 | 'et': 'est', | |
4100 | 'eu': 'eus', | |
4101 | 'fa': 'fas', | |
4102 | 'ff': 'ful', | |
4103 | 'fi': 'fin', | |
4104 | 'fj': 'fij', | |
4105 | 'fo': 'fao', | |
4106 | 'fr': 'fra', | |
4107 | 'fy': 'fry', | |
4108 | 'ga': 'gle', | |
4109 | 'gd': 'gla', | |
4110 | 'gl': 'glg', | |
4111 | 'gn': 'grn', | |
4112 | 'gu': 'guj', | |
4113 | 'gv': 'glv', | |
4114 | 'ha': 'hau', | |
4115 | 'he': 'heb', | |
b7acc835 | 4116 | 'iw': 'heb', # Replaced by he in 1989 revision |
39672624 YCH |
4117 | 'hi': 'hin', |
4118 | 'ho': 'hmo', | |
4119 | 'hr': 'hrv', | |
4120 | 'ht': 'hat', | |
4121 | 'hu': 'hun', | |
4122 | 'hy': 'hye', | |
4123 | 'hz': 'her', | |
4124 | 'ia': 'ina', | |
4125 | 'id': 'ind', | |
b7acc835 | 4126 | 'in': 'ind', # Replaced by id in 1989 revision |
39672624 YCH |
4127 | 'ie': 'ile', |
4128 | 'ig': 'ibo', | |
4129 | 'ii': 'iii', | |
4130 | 'ik': 'ipk', | |
4131 | 'io': 'ido', | |
4132 | 'is': 'isl', | |
4133 | 'it': 'ita', | |
4134 | 'iu': 'iku', | |
4135 | 'ja': 'jpn', | |
4136 | 'jv': 'jav', | |
4137 | 'ka': 'kat', | |
4138 | 'kg': 'kon', | |
4139 | 'ki': 'kik', | |
4140 | 'kj': 'kua', | |
4141 | 'kk': 'kaz', | |
4142 | 'kl': 'kal', | |
4143 | 'km': 'khm', | |
4144 | 'kn': 'kan', | |
4145 | 'ko': 'kor', | |
4146 | 'kr': 'kau', | |
4147 | 'ks': 'kas', | |
4148 | 'ku': 'kur', | |
4149 | 'kv': 'kom', | |
4150 | 'kw': 'cor', | |
4151 | 'ky': 'kir', | |
4152 | 'la': 'lat', | |
4153 | 'lb': 'ltz', | |
4154 | 'lg': 'lug', | |
4155 | 'li': 'lim', | |
4156 | 'ln': 'lin', | |
4157 | 'lo': 'lao', | |
4158 | 'lt': 'lit', | |
4159 | 'lu': 'lub', | |
4160 | 'lv': 'lav', | |
4161 | 'mg': 'mlg', | |
4162 | 'mh': 'mah', | |
4163 | 'mi': 'mri', | |
4164 | 'mk': 'mkd', | |
4165 | 'ml': 'mal', | |
4166 | 'mn': 'mon', | |
4167 | 'mr': 'mar', | |
4168 | 'ms': 'msa', | |
4169 | 'mt': 'mlt', | |
4170 | 'my': 'mya', | |
4171 | 'na': 'nau', | |
4172 | 'nb': 'nob', | |
4173 | 'nd': 'nde', | |
4174 | 'ne': 'nep', | |
4175 | 'ng': 'ndo', | |
4176 | 'nl': 'nld', | |
4177 | 'nn': 'nno', | |
4178 | 'no': 'nor', | |
4179 | 'nr': 'nbl', | |
4180 | 'nv': 'nav', | |
4181 | 'ny': 'nya', | |
4182 | 'oc': 'oci', | |
4183 | 'oj': 'oji', | |
4184 | 'om': 'orm', | |
4185 | 'or': 'ori', | |
4186 | 'os': 'oss', | |
4187 | 'pa': 'pan', | |
4188 | 'pi': 'pli', | |
4189 | 'pl': 'pol', | |
4190 | 'ps': 'pus', | |
4191 | 'pt': 'por', | |
4192 | 'qu': 'que', | |
4193 | 'rm': 'roh', | |
4194 | 'rn': 'run', | |
4195 | 'ro': 'ron', | |
4196 | 'ru': 'rus', | |
4197 | 'rw': 'kin', | |
4198 | 'sa': 'san', | |
4199 | 'sc': 'srd', | |
4200 | 'sd': 'snd', | |
4201 | 'se': 'sme', | |
4202 | 'sg': 'sag', | |
4203 | 'si': 'sin', | |
4204 | 'sk': 'slk', | |
4205 | 'sl': 'slv', | |
4206 | 'sm': 'smo', | |
4207 | 'sn': 'sna', | |
4208 | 'so': 'som', | |
4209 | 'sq': 'sqi', | |
4210 | 'sr': 'srp', | |
4211 | 'ss': 'ssw', | |
4212 | 'st': 'sot', | |
4213 | 'su': 'sun', | |
4214 | 'sv': 'swe', | |
4215 | 'sw': 'swa', | |
4216 | 'ta': 'tam', | |
4217 | 'te': 'tel', | |
4218 | 'tg': 'tgk', | |
4219 | 'th': 'tha', | |
4220 | 'ti': 'tir', | |
4221 | 'tk': 'tuk', | |
4222 | 'tl': 'tgl', | |
4223 | 'tn': 'tsn', | |
4224 | 'to': 'ton', | |
4225 | 'tr': 'tur', | |
4226 | 'ts': 'tso', | |
4227 | 'tt': 'tat', | |
4228 | 'tw': 'twi', | |
4229 | 'ty': 'tah', | |
4230 | 'ug': 'uig', | |
4231 | 'uk': 'ukr', | |
4232 | 'ur': 'urd', | |
4233 | 'uz': 'uzb', | |
4234 | 've': 'ven', | |
4235 | 'vi': 'vie', | |
4236 | 'vo': 'vol', | |
4237 | 'wa': 'wln', | |
4238 | 'wo': 'wol', | |
4239 | 'xh': 'xho', | |
4240 | 'yi': 'yid', | |
e9a50fba | 4241 | 'ji': 'yid', # Replaced by yi in 1989 revision |
39672624 YCH |
4242 | 'yo': 'yor', |
4243 | 'za': 'zha', | |
4244 | 'zh': 'zho', | |
4245 | 'zu': 'zul', | |
4246 | } | |
4247 | ||
4248 | @classmethod | |
4249 | def short2long(cls, code): | |
4250 | """Convert language code from ISO 639-1 to ISO 639-2/T""" | |
4251 | return cls._lang_map.get(code[:2]) | |
4252 | ||
4253 | @classmethod | |
4254 | def long2short(cls, code): | |
4255 | """Convert language code from ISO 639-2/T to ISO 639-1""" | |
4256 | for short_name, long_name in cls._lang_map.items(): | |
4257 | if long_name == code: | |
4258 | return short_name | |
4259 | ||
4260 | ||
86e5f3ed | 4261 | class ISO3166Utils: |
4eb10f66 YCH |
4262 | # From http://data.okfn.org/data/core/country-list |
4263 | _country_map = { | |
4264 | 'AF': 'Afghanistan', | |
4265 | 'AX': 'Åland Islands', | |
4266 | 'AL': 'Albania', | |
4267 | 'DZ': 'Algeria', | |
4268 | 'AS': 'American Samoa', | |
4269 | 'AD': 'Andorra', | |
4270 | 'AO': 'Angola', | |
4271 | 'AI': 'Anguilla', | |
4272 | 'AQ': 'Antarctica', | |
4273 | 'AG': 'Antigua and Barbuda', | |
4274 | 'AR': 'Argentina', | |
4275 | 'AM': 'Armenia', | |
4276 | 'AW': 'Aruba', | |
4277 | 'AU': 'Australia', | |
4278 | 'AT': 'Austria', | |
4279 | 'AZ': 'Azerbaijan', | |
4280 | 'BS': 'Bahamas', | |
4281 | 'BH': 'Bahrain', | |
4282 | 'BD': 'Bangladesh', | |
4283 | 'BB': 'Barbados', | |
4284 | 'BY': 'Belarus', | |
4285 | 'BE': 'Belgium', | |
4286 | 'BZ': 'Belize', | |
4287 | 'BJ': 'Benin', | |
4288 | 'BM': 'Bermuda', | |
4289 | 'BT': 'Bhutan', | |
4290 | 'BO': 'Bolivia, Plurinational State of', | |
4291 | 'BQ': 'Bonaire, Sint Eustatius and Saba', | |
4292 | 'BA': 'Bosnia and Herzegovina', | |
4293 | 'BW': 'Botswana', | |
4294 | 'BV': 'Bouvet Island', | |
4295 | 'BR': 'Brazil', | |
4296 | 'IO': 'British Indian Ocean Territory', | |
4297 | 'BN': 'Brunei Darussalam', | |
4298 | 'BG': 'Bulgaria', | |
4299 | 'BF': 'Burkina Faso', | |
4300 | 'BI': 'Burundi', | |
4301 | 'KH': 'Cambodia', | |
4302 | 'CM': 'Cameroon', | |
4303 | 'CA': 'Canada', | |
4304 | 'CV': 'Cape Verde', | |
4305 | 'KY': 'Cayman Islands', | |
4306 | 'CF': 'Central African Republic', | |
4307 | 'TD': 'Chad', | |
4308 | 'CL': 'Chile', | |
4309 | 'CN': 'China', | |
4310 | 'CX': 'Christmas Island', | |
4311 | 'CC': 'Cocos (Keeling) Islands', | |
4312 | 'CO': 'Colombia', | |
4313 | 'KM': 'Comoros', | |
4314 | 'CG': 'Congo', | |
4315 | 'CD': 'Congo, the Democratic Republic of the', | |
4316 | 'CK': 'Cook Islands', | |
4317 | 'CR': 'Costa Rica', | |
4318 | 'CI': 'Côte d\'Ivoire', | |
4319 | 'HR': 'Croatia', | |
4320 | 'CU': 'Cuba', | |
4321 | 'CW': 'Curaçao', | |
4322 | 'CY': 'Cyprus', | |
4323 | 'CZ': 'Czech Republic', | |
4324 | 'DK': 'Denmark', | |
4325 | 'DJ': 'Djibouti', | |
4326 | 'DM': 'Dominica', | |
4327 | 'DO': 'Dominican Republic', | |
4328 | 'EC': 'Ecuador', | |
4329 | 'EG': 'Egypt', | |
4330 | 'SV': 'El Salvador', | |
4331 | 'GQ': 'Equatorial Guinea', | |
4332 | 'ER': 'Eritrea', | |
4333 | 'EE': 'Estonia', | |
4334 | 'ET': 'Ethiopia', | |
4335 | 'FK': 'Falkland Islands (Malvinas)', | |
4336 | 'FO': 'Faroe Islands', | |
4337 | 'FJ': 'Fiji', | |
4338 | 'FI': 'Finland', | |
4339 | 'FR': 'France', | |
4340 | 'GF': 'French Guiana', | |
4341 | 'PF': 'French Polynesia', | |
4342 | 'TF': 'French Southern Territories', | |
4343 | 'GA': 'Gabon', | |
4344 | 'GM': 'Gambia', | |
4345 | 'GE': 'Georgia', | |
4346 | 'DE': 'Germany', | |
4347 | 'GH': 'Ghana', | |
4348 | 'GI': 'Gibraltar', | |
4349 | 'GR': 'Greece', | |
4350 | 'GL': 'Greenland', | |
4351 | 'GD': 'Grenada', | |
4352 | 'GP': 'Guadeloupe', | |
4353 | 'GU': 'Guam', | |
4354 | 'GT': 'Guatemala', | |
4355 | 'GG': 'Guernsey', | |
4356 | 'GN': 'Guinea', | |
4357 | 'GW': 'Guinea-Bissau', | |
4358 | 'GY': 'Guyana', | |
4359 | 'HT': 'Haiti', | |
4360 | 'HM': 'Heard Island and McDonald Islands', | |
4361 | 'VA': 'Holy See (Vatican City State)', | |
4362 | 'HN': 'Honduras', | |
4363 | 'HK': 'Hong Kong', | |
4364 | 'HU': 'Hungary', | |
4365 | 'IS': 'Iceland', | |
4366 | 'IN': 'India', | |
4367 | 'ID': 'Indonesia', | |
4368 | 'IR': 'Iran, Islamic Republic of', | |
4369 | 'IQ': 'Iraq', | |
4370 | 'IE': 'Ireland', | |
4371 | 'IM': 'Isle of Man', | |
4372 | 'IL': 'Israel', | |
4373 | 'IT': 'Italy', | |
4374 | 'JM': 'Jamaica', | |
4375 | 'JP': 'Japan', | |
4376 | 'JE': 'Jersey', | |
4377 | 'JO': 'Jordan', | |
4378 | 'KZ': 'Kazakhstan', | |
4379 | 'KE': 'Kenya', | |
4380 | 'KI': 'Kiribati', | |
4381 | 'KP': 'Korea, Democratic People\'s Republic of', | |
4382 | 'KR': 'Korea, Republic of', | |
4383 | 'KW': 'Kuwait', | |
4384 | 'KG': 'Kyrgyzstan', | |
4385 | 'LA': 'Lao People\'s Democratic Republic', | |
4386 | 'LV': 'Latvia', | |
4387 | 'LB': 'Lebanon', | |
4388 | 'LS': 'Lesotho', | |
4389 | 'LR': 'Liberia', | |
4390 | 'LY': 'Libya', | |
4391 | 'LI': 'Liechtenstein', | |
4392 | 'LT': 'Lithuania', | |
4393 | 'LU': 'Luxembourg', | |
4394 | 'MO': 'Macao', | |
4395 | 'MK': 'Macedonia, the Former Yugoslav Republic of', | |
4396 | 'MG': 'Madagascar', | |
4397 | 'MW': 'Malawi', | |
4398 | 'MY': 'Malaysia', | |
4399 | 'MV': 'Maldives', | |
4400 | 'ML': 'Mali', | |
4401 | 'MT': 'Malta', | |
4402 | 'MH': 'Marshall Islands', | |
4403 | 'MQ': 'Martinique', | |
4404 | 'MR': 'Mauritania', | |
4405 | 'MU': 'Mauritius', | |
4406 | 'YT': 'Mayotte', | |
4407 | 'MX': 'Mexico', | |
4408 | 'FM': 'Micronesia, Federated States of', | |
4409 | 'MD': 'Moldova, Republic of', | |
4410 | 'MC': 'Monaco', | |
4411 | 'MN': 'Mongolia', | |
4412 | 'ME': 'Montenegro', | |
4413 | 'MS': 'Montserrat', | |
4414 | 'MA': 'Morocco', | |
4415 | 'MZ': 'Mozambique', | |
4416 | 'MM': 'Myanmar', | |
4417 | 'NA': 'Namibia', | |
4418 | 'NR': 'Nauru', | |
4419 | 'NP': 'Nepal', | |
4420 | 'NL': 'Netherlands', | |
4421 | 'NC': 'New Caledonia', | |
4422 | 'NZ': 'New Zealand', | |
4423 | 'NI': 'Nicaragua', | |
4424 | 'NE': 'Niger', | |
4425 | 'NG': 'Nigeria', | |
4426 | 'NU': 'Niue', | |
4427 | 'NF': 'Norfolk Island', | |
4428 | 'MP': 'Northern Mariana Islands', | |
4429 | 'NO': 'Norway', | |
4430 | 'OM': 'Oman', | |
4431 | 'PK': 'Pakistan', | |
4432 | 'PW': 'Palau', | |
4433 | 'PS': 'Palestine, State of', | |
4434 | 'PA': 'Panama', | |
4435 | 'PG': 'Papua New Guinea', | |
4436 | 'PY': 'Paraguay', | |
4437 | 'PE': 'Peru', | |
4438 | 'PH': 'Philippines', | |
4439 | 'PN': 'Pitcairn', | |
4440 | 'PL': 'Poland', | |
4441 | 'PT': 'Portugal', | |
4442 | 'PR': 'Puerto Rico', | |
4443 | 'QA': 'Qatar', | |
4444 | 'RE': 'Réunion', | |
4445 | 'RO': 'Romania', | |
4446 | 'RU': 'Russian Federation', | |
4447 | 'RW': 'Rwanda', | |
4448 | 'BL': 'Saint Barthélemy', | |
4449 | 'SH': 'Saint Helena, Ascension and Tristan da Cunha', | |
4450 | 'KN': 'Saint Kitts and Nevis', | |
4451 | 'LC': 'Saint Lucia', | |
4452 | 'MF': 'Saint Martin (French part)', | |
4453 | 'PM': 'Saint Pierre and Miquelon', | |
4454 | 'VC': 'Saint Vincent and the Grenadines', | |
4455 | 'WS': 'Samoa', | |
4456 | 'SM': 'San Marino', | |
4457 | 'ST': 'Sao Tome and Principe', | |
4458 | 'SA': 'Saudi Arabia', | |
4459 | 'SN': 'Senegal', | |
4460 | 'RS': 'Serbia', | |
4461 | 'SC': 'Seychelles', | |
4462 | 'SL': 'Sierra Leone', | |
4463 | 'SG': 'Singapore', | |
4464 | 'SX': 'Sint Maarten (Dutch part)', | |
4465 | 'SK': 'Slovakia', | |
4466 | 'SI': 'Slovenia', | |
4467 | 'SB': 'Solomon Islands', | |
4468 | 'SO': 'Somalia', | |
4469 | 'ZA': 'South Africa', | |
4470 | 'GS': 'South Georgia and the South Sandwich Islands', | |
4471 | 'SS': 'South Sudan', | |
4472 | 'ES': 'Spain', | |
4473 | 'LK': 'Sri Lanka', | |
4474 | 'SD': 'Sudan', | |
4475 | 'SR': 'Suriname', | |
4476 | 'SJ': 'Svalbard and Jan Mayen', | |
4477 | 'SZ': 'Swaziland', | |
4478 | 'SE': 'Sweden', | |
4479 | 'CH': 'Switzerland', | |
4480 | 'SY': 'Syrian Arab Republic', | |
4481 | 'TW': 'Taiwan, Province of China', | |
4482 | 'TJ': 'Tajikistan', | |
4483 | 'TZ': 'Tanzania, United Republic of', | |
4484 | 'TH': 'Thailand', | |
4485 | 'TL': 'Timor-Leste', | |
4486 | 'TG': 'Togo', | |
4487 | 'TK': 'Tokelau', | |
4488 | 'TO': 'Tonga', | |
4489 | 'TT': 'Trinidad and Tobago', | |
4490 | 'TN': 'Tunisia', | |
4491 | 'TR': 'Turkey', | |
4492 | 'TM': 'Turkmenistan', | |
4493 | 'TC': 'Turks and Caicos Islands', | |
4494 | 'TV': 'Tuvalu', | |
4495 | 'UG': 'Uganda', | |
4496 | 'UA': 'Ukraine', | |
4497 | 'AE': 'United Arab Emirates', | |
4498 | 'GB': 'United Kingdom', | |
4499 | 'US': 'United States', | |
4500 | 'UM': 'United States Minor Outlying Islands', | |
4501 | 'UY': 'Uruguay', | |
4502 | 'UZ': 'Uzbekistan', | |
4503 | 'VU': 'Vanuatu', | |
4504 | 'VE': 'Venezuela, Bolivarian Republic of', | |
4505 | 'VN': 'Viet Nam', | |
4506 | 'VG': 'Virgin Islands, British', | |
4507 | 'VI': 'Virgin Islands, U.S.', | |
4508 | 'WF': 'Wallis and Futuna', | |
4509 | 'EH': 'Western Sahara', | |
4510 | 'YE': 'Yemen', | |
4511 | 'ZM': 'Zambia', | |
4512 | 'ZW': 'Zimbabwe', | |
2f97cc61 | 4513 | # Not ISO 3166 codes, but used for IP blocks |
4514 | 'AP': 'Asia/Pacific Region', | |
4515 | 'EU': 'Europe', | |
4eb10f66 YCH |
4516 | } |
4517 | ||
4518 | @classmethod | |
4519 | def short2full(cls, code): | |
4520 | """Convert an ISO 3166-2 country code to the corresponding full name""" | |
4521 | return cls._country_map.get(code.upper()) | |
4522 | ||
4523 | ||
86e5f3ed | 4524 | class GeoUtils: |
773f291d S |
4525 | # Major IPv4 address blocks per country |
4526 | _country_ip_map = { | |
53896ca5 | 4527 | 'AD': '46.172.224.0/19', |
773f291d S |
4528 | 'AE': '94.200.0.0/13', |
4529 | 'AF': '149.54.0.0/17', | |
4530 | 'AG': '209.59.64.0/18', | |
4531 | 'AI': '204.14.248.0/21', | |
4532 | 'AL': '46.99.0.0/16', | |
4533 | 'AM': '46.70.0.0/15', | |
4534 | 'AO': '105.168.0.0/13', | |
53896ca5 S |
4535 | 'AP': '182.50.184.0/21', |
4536 | 'AQ': '23.154.160.0/24', | |
773f291d S |
4537 | 'AR': '181.0.0.0/12', |
4538 | 'AS': '202.70.112.0/20', | |
53896ca5 | 4539 | 'AT': '77.116.0.0/14', |
773f291d S |
4540 | 'AU': '1.128.0.0/11', |
4541 | 'AW': '181.41.0.0/18', | |
53896ca5 S |
4542 | 'AX': '185.217.4.0/22', |
4543 | 'AZ': '5.197.0.0/16', | |
773f291d S |
4544 | 'BA': '31.176.128.0/17', |
4545 | 'BB': '65.48.128.0/17', | |
4546 | 'BD': '114.130.0.0/16', | |
4547 | 'BE': '57.0.0.0/8', | |
53896ca5 | 4548 | 'BF': '102.178.0.0/15', |
773f291d S |
4549 | 'BG': '95.42.0.0/15', |
4550 | 'BH': '37.131.0.0/17', | |
4551 | 'BI': '154.117.192.0/18', | |
4552 | 'BJ': '137.255.0.0/16', | |
53896ca5 | 4553 | 'BL': '185.212.72.0/23', |
773f291d S |
4554 | 'BM': '196.12.64.0/18', |
4555 | 'BN': '156.31.0.0/16', | |
4556 | 'BO': '161.56.0.0/16', | |
4557 | 'BQ': '161.0.80.0/20', | |
53896ca5 | 4558 | 'BR': '191.128.0.0/12', |
773f291d S |
4559 | 'BS': '24.51.64.0/18', |
4560 | 'BT': '119.2.96.0/19', | |
4561 | 'BW': '168.167.0.0/16', | |
4562 | 'BY': '178.120.0.0/13', | |
4563 | 'BZ': '179.42.192.0/18', | |
4564 | 'CA': '99.224.0.0/11', | |
4565 | 'CD': '41.243.0.0/16', | |
53896ca5 S |
4566 | 'CF': '197.242.176.0/21', |
4567 | 'CG': '160.113.0.0/16', | |
773f291d | 4568 | 'CH': '85.0.0.0/13', |
53896ca5 | 4569 | 'CI': '102.136.0.0/14', |
773f291d S |
4570 | 'CK': '202.65.32.0/19', |
4571 | 'CL': '152.172.0.0/14', | |
53896ca5 | 4572 | 'CM': '102.244.0.0/14', |
773f291d S |
4573 | 'CN': '36.128.0.0/10', |
4574 | 'CO': '181.240.0.0/12', | |
4575 | 'CR': '201.192.0.0/12', | |
4576 | 'CU': '152.206.0.0/15', | |
4577 | 'CV': '165.90.96.0/19', | |
4578 | 'CW': '190.88.128.0/17', | |
53896ca5 | 4579 | 'CY': '31.153.0.0/16', |
773f291d S |
4580 | 'CZ': '88.100.0.0/14', |
4581 | 'DE': '53.0.0.0/8', | |
4582 | 'DJ': '197.241.0.0/17', | |
4583 | 'DK': '87.48.0.0/12', | |
4584 | 'DM': '192.243.48.0/20', | |
4585 | 'DO': '152.166.0.0/15', | |
4586 | 'DZ': '41.96.0.0/12', | |
4587 | 'EC': '186.68.0.0/15', | |
4588 | 'EE': '90.190.0.0/15', | |
4589 | 'EG': '156.160.0.0/11', | |
4590 | 'ER': '196.200.96.0/20', | |
4591 | 'ES': '88.0.0.0/11', | |
4592 | 'ET': '196.188.0.0/14', | |
4593 | 'EU': '2.16.0.0/13', | |
4594 | 'FI': '91.152.0.0/13', | |
4595 | 'FJ': '144.120.0.0/16', | |
53896ca5 | 4596 | 'FK': '80.73.208.0/21', |
773f291d S |
4597 | 'FM': '119.252.112.0/20', |
4598 | 'FO': '88.85.32.0/19', | |
4599 | 'FR': '90.0.0.0/9', | |
4600 | 'GA': '41.158.0.0/15', | |
4601 | 'GB': '25.0.0.0/8', | |
4602 | 'GD': '74.122.88.0/21', | |
4603 | 'GE': '31.146.0.0/16', | |
4604 | 'GF': '161.22.64.0/18', | |
4605 | 'GG': '62.68.160.0/19', | |
53896ca5 S |
4606 | 'GH': '154.160.0.0/12', |
4607 | 'GI': '95.164.0.0/16', | |
773f291d S |
4608 | 'GL': '88.83.0.0/19', |
4609 | 'GM': '160.182.0.0/15', | |
4610 | 'GN': '197.149.192.0/18', | |
4611 | 'GP': '104.250.0.0/19', | |
4612 | 'GQ': '105.235.224.0/20', | |
4613 | 'GR': '94.64.0.0/13', | |
4614 | 'GT': '168.234.0.0/16', | |
4615 | 'GU': '168.123.0.0/16', | |
4616 | 'GW': '197.214.80.0/20', | |
4617 | 'GY': '181.41.64.0/18', | |
4618 | 'HK': '113.252.0.0/14', | |
4619 | 'HN': '181.210.0.0/16', | |
4620 | 'HR': '93.136.0.0/13', | |
4621 | 'HT': '148.102.128.0/17', | |
4622 | 'HU': '84.0.0.0/14', | |
4623 | 'ID': '39.192.0.0/10', | |
4624 | 'IE': '87.32.0.0/12', | |
4625 | 'IL': '79.176.0.0/13', | |
4626 | 'IM': '5.62.80.0/20', | |
4627 | 'IN': '117.192.0.0/10', | |
4628 | 'IO': '203.83.48.0/21', | |
4629 | 'IQ': '37.236.0.0/14', | |
4630 | 'IR': '2.176.0.0/12', | |
4631 | 'IS': '82.221.0.0/16', | |
4632 | 'IT': '79.0.0.0/10', | |
4633 | 'JE': '87.244.64.0/18', | |
4634 | 'JM': '72.27.0.0/17', | |
4635 | 'JO': '176.29.0.0/16', | |
53896ca5 | 4636 | 'JP': '133.0.0.0/8', |
773f291d S |
4637 | 'KE': '105.48.0.0/12', |
4638 | 'KG': '158.181.128.0/17', | |
4639 | 'KH': '36.37.128.0/17', | |
4640 | 'KI': '103.25.140.0/22', | |
4641 | 'KM': '197.255.224.0/20', | |
53896ca5 | 4642 | 'KN': '198.167.192.0/19', |
773f291d S |
4643 | 'KP': '175.45.176.0/22', |
4644 | 'KR': '175.192.0.0/10', | |
4645 | 'KW': '37.36.0.0/14', | |
4646 | 'KY': '64.96.0.0/15', | |
4647 | 'KZ': '2.72.0.0/13', | |
4648 | 'LA': '115.84.64.0/18', | |
4649 | 'LB': '178.135.0.0/16', | |
53896ca5 | 4650 | 'LC': '24.92.144.0/20', |
773f291d S |
4651 | 'LI': '82.117.0.0/19', |
4652 | 'LK': '112.134.0.0/15', | |
53896ca5 | 4653 | 'LR': '102.183.0.0/16', |
773f291d S |
4654 | 'LS': '129.232.0.0/17', |
4655 | 'LT': '78.56.0.0/13', | |
4656 | 'LU': '188.42.0.0/16', | |
4657 | 'LV': '46.109.0.0/16', | |
4658 | 'LY': '41.252.0.0/14', | |
4659 | 'MA': '105.128.0.0/11', | |
4660 | 'MC': '88.209.64.0/18', | |
4661 | 'MD': '37.246.0.0/16', | |
4662 | 'ME': '178.175.0.0/17', | |
4663 | 'MF': '74.112.232.0/21', | |
4664 | 'MG': '154.126.0.0/17', | |
4665 | 'MH': '117.103.88.0/21', | |
4666 | 'MK': '77.28.0.0/15', | |
4667 | 'ML': '154.118.128.0/18', | |
4668 | 'MM': '37.111.0.0/17', | |
4669 | 'MN': '49.0.128.0/17', | |
4670 | 'MO': '60.246.0.0/16', | |
4671 | 'MP': '202.88.64.0/20', | |
4672 | 'MQ': '109.203.224.0/19', | |
4673 | 'MR': '41.188.64.0/18', | |
4674 | 'MS': '208.90.112.0/22', | |
4675 | 'MT': '46.11.0.0/16', | |
4676 | 'MU': '105.16.0.0/12', | |
4677 | 'MV': '27.114.128.0/18', | |
53896ca5 | 4678 | 'MW': '102.70.0.0/15', |
773f291d S |
4679 | 'MX': '187.192.0.0/11', |
4680 | 'MY': '175.136.0.0/13', | |
4681 | 'MZ': '197.218.0.0/15', | |
4682 | 'NA': '41.182.0.0/16', | |
4683 | 'NC': '101.101.0.0/18', | |
4684 | 'NE': '197.214.0.0/18', | |
4685 | 'NF': '203.17.240.0/22', | |
4686 | 'NG': '105.112.0.0/12', | |
4687 | 'NI': '186.76.0.0/15', | |
4688 | 'NL': '145.96.0.0/11', | |
4689 | 'NO': '84.208.0.0/13', | |
4690 | 'NP': '36.252.0.0/15', | |
4691 | 'NR': '203.98.224.0/19', | |
4692 | 'NU': '49.156.48.0/22', | |
4693 | 'NZ': '49.224.0.0/14', | |
4694 | 'OM': '5.36.0.0/15', | |
4695 | 'PA': '186.72.0.0/15', | |
4696 | 'PE': '186.160.0.0/14', | |
4697 | 'PF': '123.50.64.0/18', | |
4698 | 'PG': '124.240.192.0/19', | |
4699 | 'PH': '49.144.0.0/13', | |
4700 | 'PK': '39.32.0.0/11', | |
4701 | 'PL': '83.0.0.0/11', | |
4702 | 'PM': '70.36.0.0/20', | |
4703 | 'PR': '66.50.0.0/16', | |
4704 | 'PS': '188.161.0.0/16', | |
4705 | 'PT': '85.240.0.0/13', | |
4706 | 'PW': '202.124.224.0/20', | |
4707 | 'PY': '181.120.0.0/14', | |
4708 | 'QA': '37.210.0.0/15', | |
53896ca5 | 4709 | 'RE': '102.35.0.0/16', |
773f291d | 4710 | 'RO': '79.112.0.0/13', |
53896ca5 | 4711 | 'RS': '93.86.0.0/15', |
773f291d | 4712 | 'RU': '5.136.0.0/13', |
53896ca5 | 4713 | 'RW': '41.186.0.0/16', |
773f291d S |
4714 | 'SA': '188.48.0.0/13', |
4715 | 'SB': '202.1.160.0/19', | |
4716 | 'SC': '154.192.0.0/11', | |
53896ca5 | 4717 | 'SD': '102.120.0.0/13', |
773f291d | 4718 | 'SE': '78.64.0.0/12', |
53896ca5 | 4719 | 'SG': '8.128.0.0/10', |
773f291d S |
4720 | 'SI': '188.196.0.0/14', |
4721 | 'SK': '78.98.0.0/15', | |
53896ca5 | 4722 | 'SL': '102.143.0.0/17', |
773f291d S |
4723 | 'SM': '89.186.32.0/19', |
4724 | 'SN': '41.82.0.0/15', | |
53896ca5 | 4725 | 'SO': '154.115.192.0/18', |
773f291d S |
4726 | 'SR': '186.179.128.0/17', |
4727 | 'SS': '105.235.208.0/21', | |
4728 | 'ST': '197.159.160.0/19', | |
4729 | 'SV': '168.243.0.0/16', | |
4730 | 'SX': '190.102.0.0/20', | |
4731 | 'SY': '5.0.0.0/16', | |
4732 | 'SZ': '41.84.224.0/19', | |
4733 | 'TC': '65.255.48.0/20', | |
4734 | 'TD': '154.68.128.0/19', | |
4735 | 'TG': '196.168.0.0/14', | |
4736 | 'TH': '171.96.0.0/13', | |
4737 | 'TJ': '85.9.128.0/18', | |
4738 | 'TK': '27.96.24.0/21', | |
4739 | 'TL': '180.189.160.0/20', | |
4740 | 'TM': '95.85.96.0/19', | |
4741 | 'TN': '197.0.0.0/11', | |
4742 | 'TO': '175.176.144.0/21', | |
4743 | 'TR': '78.160.0.0/11', | |
4744 | 'TT': '186.44.0.0/15', | |
4745 | 'TV': '202.2.96.0/19', | |
4746 | 'TW': '120.96.0.0/11', | |
4747 | 'TZ': '156.156.0.0/14', | |
53896ca5 S |
4748 | 'UA': '37.52.0.0/14', |
4749 | 'UG': '102.80.0.0/13', | |
4750 | 'US': '6.0.0.0/8', | |
773f291d | 4751 | 'UY': '167.56.0.0/13', |
53896ca5 | 4752 | 'UZ': '84.54.64.0/18', |
773f291d | 4753 | 'VA': '212.77.0.0/19', |
53896ca5 | 4754 | 'VC': '207.191.240.0/21', |
773f291d | 4755 | 'VE': '186.88.0.0/13', |
53896ca5 | 4756 | 'VG': '66.81.192.0/20', |
773f291d S |
4757 | 'VI': '146.226.0.0/16', |
4758 | 'VN': '14.160.0.0/11', | |
4759 | 'VU': '202.80.32.0/20', | |
4760 | 'WF': '117.20.32.0/21', | |
4761 | 'WS': '202.4.32.0/19', | |
4762 | 'YE': '134.35.0.0/16', | |
4763 | 'YT': '41.242.116.0/22', | |
4764 | 'ZA': '41.0.0.0/11', | |
53896ca5 S |
4765 | 'ZM': '102.144.0.0/13', |
4766 | 'ZW': '102.177.192.0/18', | |
773f291d S |
4767 | } |
4768 | ||
4769 | @classmethod | |
5f95927a S |
4770 | def random_ipv4(cls, code_or_block): |
4771 | if len(code_or_block) == 2: | |
4772 | block = cls._country_ip_map.get(code_or_block.upper()) | |
4773 | if not block: | |
4774 | return None | |
4775 | else: | |
4776 | block = code_or_block | |
773f291d | 4777 | addr, preflen = block.split('/') |
ac668111 | 4778 | addr_min = struct.unpack('!L', socket.inet_aton(addr))[0] |
773f291d | 4779 | addr_max = addr_min | (0xffffffff >> int(preflen)) |
14f25df2 | 4780 | return str(socket.inet_ntoa( |
ac668111 | 4781 | struct.pack('!L', random.randint(addr_min, addr_max)))) |
773f291d S |
4782 | |
4783 | ||
ac668111 | 4784 | class PerRequestProxyHandler(urllib.request.ProxyHandler): |
2461f79d PH |
4785 | def __init__(self, proxies=None): |
4786 | # Set default handlers | |
4787 | for type in ('http', 'https'): | |
4788 | setattr(self, '%s_open' % type, | |
4789 | lambda r, proxy='__noproxy__', type=type, meth=self.proxy_open: | |
4790 | meth(r, proxy, type)) | |
ac668111 | 4791 | urllib.request.ProxyHandler.__init__(self, proxies) |
2461f79d | 4792 | |
91410c9b | 4793 | def proxy_open(self, req, proxy, type): |
2461f79d | 4794 | req_proxy = req.headers.get('Ytdl-request-proxy') |
91410c9b PH |
4795 | if req_proxy is not None: |
4796 | proxy = req_proxy | |
2461f79d PH |
4797 | del req.headers['Ytdl-request-proxy'] |
4798 | ||
4799 | if proxy == '__noproxy__': | |
4800 | return None # No Proxy | |
14f25df2 | 4801 | if urllib.parse.urlparse(proxy).scheme.lower() in ('socks', 'socks4', 'socks4a', 'socks5'): |
71aff188 | 4802 | req.add_header('Ytdl-socks-proxy', proxy) |
7a5c1cfe | 4803 | # yt-dlp's http/https handlers do wrapping the socket with socks |
71aff188 | 4804 | return None |
ac668111 | 4805 | return urllib.request.ProxyHandler.proxy_open( |
91410c9b | 4806 | self, req, proxy, type) |
5bc880b9 YCH |
4807 | |
4808 | ||
0a5445dd YCH |
4809 | # Both long_to_bytes and bytes_to_long are adapted from PyCrypto, which is |
4810 | # released into Public Domain | |
4811 | # https://github.com/dlitz/pycrypto/blob/master/lib/Crypto/Util/number.py#L387 | |
4812 | ||
4813 | def long_to_bytes(n, blocksize=0): | |
4814 | """long_to_bytes(n:long, blocksize:int) : string | |
4815 | Convert a long integer to a byte string. | |
4816 | ||
4817 | If optional blocksize is given and greater than zero, pad the front of the | |
4818 | byte string with binary zeros so that the length is a multiple of | |
4819 | blocksize. | |
4820 | """ | |
4821 | # after much testing, this algorithm was deemed to be the fastest | |
4822 | s = b'' | |
4823 | n = int(n) | |
4824 | while n > 0: | |
ac668111 | 4825 | s = struct.pack('>I', n & 0xffffffff) + s |
0a5445dd YCH |
4826 | n = n >> 32 |
4827 | # strip off leading zeros | |
4828 | for i in range(len(s)): | |
4829 | if s[i] != b'\000'[0]: | |
4830 | break | |
4831 | else: | |
4832 | # only happens when n == 0 | |
4833 | s = b'\000' | |
4834 | i = 0 | |
4835 | s = s[i:] | |
4836 | # add back some pad bytes. this could be done more efficiently w.r.t. the | |
4837 | # de-padding being done above, but sigh... | |
4838 | if blocksize > 0 and len(s) % blocksize: | |
4839 | s = (blocksize - len(s) % blocksize) * b'\000' + s | |
4840 | return s | |
4841 | ||
4842 | ||
4843 | def bytes_to_long(s): | |
4844 | """bytes_to_long(string) : long | |
4845 | Convert a byte string to a long integer. | |
4846 | ||
4847 | This is (essentially) the inverse of long_to_bytes(). | |
4848 | """ | |
4849 | acc = 0 | |
4850 | length = len(s) | |
4851 | if length % 4: | |
4852 | extra = (4 - length % 4) | |
4853 | s = b'\000' * extra + s | |
4854 | length = length + extra | |
4855 | for i in range(0, length, 4): | |
ac668111 | 4856 | acc = (acc << 32) + struct.unpack('>I', s[i:i + 4])[0] |
0a5445dd YCH |
4857 | return acc |
4858 | ||
4859 | ||
5bc880b9 YCH |
4860 | def ohdave_rsa_encrypt(data, exponent, modulus): |
4861 | ''' | |
4862 | Implement OHDave's RSA algorithm. See http://www.ohdave.com/rsa/ | |
4863 | ||
4864 | Input: | |
4865 | data: data to encrypt, bytes-like object | |
4866 | exponent, modulus: parameter e and N of RSA algorithm, both integer | |
4867 | Output: hex string of encrypted data | |
4868 | ||
4869 | Limitation: supports one block encryption only | |
4870 | ''' | |
4871 | ||
4872 | payload = int(binascii.hexlify(data[::-1]), 16) | |
4873 | encrypted = pow(payload, exponent, modulus) | |
4874 | return '%x' % encrypted | |
81bdc8fd YCH |
4875 | |
4876 | ||
f48409c7 YCH |
4877 | def pkcs1pad(data, length): |
4878 | """ | |
4879 | Padding input data with PKCS#1 scheme | |
4880 | ||
4881 | @param {int[]} data input data | |
4882 | @param {int} length target length | |
4883 | @returns {int[]} padded data | |
4884 | """ | |
4885 | if len(data) > length - 11: | |
4886 | raise ValueError('Input data too long for PKCS#1 padding') | |
4887 | ||
4888 | pseudo_random = [random.randint(0, 254) for _ in range(length - len(data) - 3)] | |
4889 | return [0, 2] + pseudo_random + [0] + data | |
4890 | ||
4891 | ||
7b2c3f47 | 4892 | def _base_n_table(n, table): |
4893 | if not table and not n: | |
4894 | raise ValueError('Either table or n must be specified') | |
612f2be5 | 4895 | table = (table or '0123456789abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ')[:n] |
4896 | ||
44f14eb4 | 4897 | if n and n != len(table): |
612f2be5 | 4898 | raise ValueError(f'base {n} exceeds table length {len(table)}') |
4899 | return table | |
59f898b7 | 4900 | |
5eb6bdce | 4901 | |
7b2c3f47 | 4902 | def encode_base_n(num, n=None, table=None): |
4903 | """Convert given int to a base-n string""" | |
612f2be5 | 4904 | table = _base_n_table(n, table) |
7b2c3f47 | 4905 | if not num: |
5eb6bdce YCH |
4906 | return table[0] |
4907 | ||
7b2c3f47 | 4908 | result, base = '', len(table) |
81bdc8fd | 4909 | while num: |
7b2c3f47 | 4910 | result = table[num % base] + result |
612f2be5 | 4911 | num = num // base |
7b2c3f47 | 4912 | return result |
4913 | ||
4914 | ||
4915 | def decode_base_n(string, n=None, table=None): | |
4916 | """Convert given base-n string to int""" | |
4917 | table = {char: index for index, char in enumerate(_base_n_table(n, table))} | |
4918 | result, base = 0, len(table) | |
4919 | for char in string: | |
4920 | result = result * base + table[char] | |
4921 | return result | |
4922 | ||
4923 | ||
4924 | def decode_base(value, digits): | |
da4db748 | 4925 | deprecation_warning(f'{__name__}.decode_base is deprecated and may be removed ' |
4926 | f'in a future version. Use {__name__}.decode_base_n instead') | |
7b2c3f47 | 4927 | return decode_base_n(value, table=digits) |
f52354a8 YCH |
4928 | |
4929 | ||
4930 | def decode_packed_codes(code): | |
06b3fe29 | 4931 | mobj = re.search(PACKED_CODES_RE, code) |
a0566bbf | 4932 | obfuscated_code, base, count, symbols = mobj.groups() |
f52354a8 YCH |
4933 | base = int(base) |
4934 | count = int(count) | |
4935 | symbols = symbols.split('|') | |
4936 | symbol_table = {} | |
4937 | ||
4938 | while count: | |
4939 | count -= 1 | |
5eb6bdce | 4940 | base_n_count = encode_base_n(count, base) |
f52354a8 YCH |
4941 | symbol_table[base_n_count] = symbols[count] or base_n_count |
4942 | ||
4943 | return re.sub( | |
4944 | r'\b(\w+)\b', lambda mobj: symbol_table[mobj.group(0)], | |
a0566bbf | 4945 | obfuscated_code) |
e154c651 | 4946 | |
4947 | ||
1ced2221 S |
4948 | def caesar(s, alphabet, shift): |
4949 | if shift == 0: | |
4950 | return s | |
4951 | l = len(alphabet) | |
4952 | return ''.join( | |
4953 | alphabet[(alphabet.index(c) + shift) % l] if c in alphabet else c | |
4954 | for c in s) | |
4955 | ||
4956 | ||
4957 | def rot47(s): | |
4958 | return caesar(s, r'''!"#$%&'()*+,-./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{|}~''', 47) | |
4959 | ||
4960 | ||
e154c651 | 4961 | def parse_m3u8_attributes(attrib): |
4962 | info = {} | |
4963 | for (key, val) in re.findall(r'(?P<key>[A-Z0-9-]+)=(?P<val>"[^"]+"|[^",]+)(?:,|$)', attrib): | |
4964 | if val.startswith('"'): | |
4965 | val = val[1:-1] | |
4966 | info[key] = val | |
4967 | return info | |
1143535d YCH |
4968 | |
4969 | ||
4970 | def urshift(val, n): | |
4971 | return val >> n if val >= 0 else (val + 0x100000000) >> n | |
d3f8e038 YCH |
4972 | |
4973 | ||
4974 | # Based on png2str() written by @gdkchan and improved by @yokrysty | |
067aa17e | 4975 | # Originally posted at https://github.com/ytdl-org/youtube-dl/issues/9706 |
d3f8e038 YCH |
4976 | def decode_png(png_data): |
4977 | # Reference: https://www.w3.org/TR/PNG/ | |
4978 | header = png_data[8:] | |
4979 | ||
4980 | if png_data[:8] != b'\x89PNG\x0d\x0a\x1a\x0a' or header[4:8] != b'IHDR': | |
86e5f3ed | 4981 | raise OSError('Not a valid PNG file.') |
d3f8e038 YCH |
4982 | |
4983 | int_map = {1: '>B', 2: '>H', 4: '>I'} | |
ac668111 | 4984 | unpack_integer = lambda x: struct.unpack(int_map[len(x)], x)[0] |
d3f8e038 YCH |
4985 | |
4986 | chunks = [] | |
4987 | ||
4988 | while header: | |
4989 | length = unpack_integer(header[:4]) | |
4990 | header = header[4:] | |
4991 | ||
4992 | chunk_type = header[:4] | |
4993 | header = header[4:] | |
4994 | ||
4995 | chunk_data = header[:length] | |
4996 | header = header[length:] | |
4997 | ||
4998 | header = header[4:] # Skip CRC | |
4999 | ||
5000 | chunks.append({ | |
5001 | 'type': chunk_type, | |
5002 | 'length': length, | |
5003 | 'data': chunk_data | |
5004 | }) | |
5005 | ||
5006 | ihdr = chunks[0]['data'] | |
5007 | ||
5008 | width = unpack_integer(ihdr[:4]) | |
5009 | height = unpack_integer(ihdr[4:8]) | |
5010 | ||
5011 | idat = b'' | |
5012 | ||
5013 | for chunk in chunks: | |
5014 | if chunk['type'] == b'IDAT': | |
5015 | idat += chunk['data'] | |
5016 | ||
5017 | if not idat: | |
86e5f3ed | 5018 | raise OSError('Unable to read PNG data.') |
d3f8e038 YCH |
5019 | |
5020 | decompressed_data = bytearray(zlib.decompress(idat)) | |
5021 | ||
5022 | stride = width * 3 | |
5023 | pixels = [] | |
5024 | ||
5025 | def _get_pixel(idx): | |
5026 | x = idx % stride | |
5027 | y = idx // stride | |
5028 | return pixels[y][x] | |
5029 | ||
5030 | for y in range(height): | |
5031 | basePos = y * (1 + stride) | |
5032 | filter_type = decompressed_data[basePos] | |
5033 | ||
5034 | current_row = [] | |
5035 | ||
5036 | pixels.append(current_row) | |
5037 | ||
5038 | for x in range(stride): | |
5039 | color = decompressed_data[1 + basePos + x] | |
5040 | basex = y * stride + x | |
5041 | left = 0 | |
5042 | up = 0 | |
5043 | ||
5044 | if x > 2: | |
5045 | left = _get_pixel(basex - 3) | |
5046 | if y > 0: | |
5047 | up = _get_pixel(basex - stride) | |
5048 | ||
5049 | if filter_type == 1: # Sub | |
5050 | color = (color + left) & 0xff | |
5051 | elif filter_type == 2: # Up | |
5052 | color = (color + up) & 0xff | |
5053 | elif filter_type == 3: # Average | |
5054 | color = (color + ((left + up) >> 1)) & 0xff | |
5055 | elif filter_type == 4: # Paeth | |
5056 | a = left | |
5057 | b = up | |
5058 | c = 0 | |
5059 | ||
5060 | if x > 2 and y > 0: | |
5061 | c = _get_pixel(basex - stride - 3) | |
5062 | ||
5063 | p = a + b - c | |
5064 | ||
5065 | pa = abs(p - a) | |
5066 | pb = abs(p - b) | |
5067 | pc = abs(p - c) | |
5068 | ||
5069 | if pa <= pb and pa <= pc: | |
5070 | color = (color + a) & 0xff | |
5071 | elif pb <= pc: | |
5072 | color = (color + b) & 0xff | |
5073 | else: | |
5074 | color = (color + c) & 0xff | |
5075 | ||
5076 | current_row.append(color) | |
5077 | ||
5078 | return width, height, pixels | |
efa97bdc YCH |
5079 | |
5080 | ||
5081 | def write_xattr(path, key, value): | |
6f7563be | 5082 | # Windows: Write xattrs to NTFS Alternate Data Streams: |
5083 | # http://en.wikipedia.org/wiki/NTFS#Alternate_data_streams_.28ADS.29 | |
5084 | if compat_os_name == 'nt': | |
5085 | assert ':' not in key | |
5086 | assert os.path.exists(path) | |
efa97bdc YCH |
5087 | |
5088 | try: | |
6f7563be | 5089 | with open(f'{path}:{key}', 'wb') as f: |
5090 | f.write(value) | |
86e5f3ed | 5091 | except OSError as e: |
efa97bdc | 5092 | raise XAttrMetadataError(e.errno, e.strerror) |
6f7563be | 5093 | return |
efa97bdc | 5094 | |
6f7563be | 5095 | # UNIX Method 1. Use xattrs/pyxattrs modules |
efa97bdc | 5096 | |
6f7563be | 5097 | setxattr = None |
5098 | if getattr(xattr, '_yt_dlp__identifier', None) == 'pyxattr': | |
5099 | # Unicode arguments are not supported in pyxattr until version 0.5.0 | |
5100 | # See https://github.com/ytdl-org/youtube-dl/issues/5498 | |
5101 | if version_tuple(xattr.__version__) >= (0, 5, 0): | |
5102 | setxattr = xattr.set | |
5103 | elif xattr: | |
5104 | setxattr = xattr.setxattr | |
efa97bdc | 5105 | |
6f7563be | 5106 | if setxattr: |
5107 | try: | |
5108 | setxattr(path, key, value) | |
5109 | except OSError as e: | |
5110 | raise XAttrMetadataError(e.errno, e.strerror) | |
5111 | return | |
efa97bdc | 5112 | |
6f7563be | 5113 | # UNIX Method 2. Use setfattr/xattr executables |
5114 | exe = ('setfattr' if check_executable('setfattr', ['--version']) | |
5115 | else 'xattr' if check_executable('xattr', ['-h']) else None) | |
5116 | if not exe: | |
5117 | raise XAttrUnavailableError( | |
5118 | 'Couldn\'t find a tool to set the xattrs. Install either the python "xattr" or "pyxattr" modules or the ' | |
5119 | + ('"xattr" binary' if sys.platform != 'linux' else 'GNU "attr" package (which contains the "setfattr" tool)')) | |
efa97bdc | 5120 | |
0f06bcd7 | 5121 | value = value.decode() |
6f7563be | 5122 | try: |
f0c9fb96 | 5123 | _, stderr, returncode = Popen.run( |
6f7563be | 5124 | [exe, '-w', key, value, path] if exe == 'xattr' else [exe, '-n', key, '-v', value, path], |
e121e3ce | 5125 | text=True, stdout=subprocess.PIPE, stderr=subprocess.PIPE, stdin=subprocess.PIPE) |
6f7563be | 5126 | except OSError as e: |
5127 | raise XAttrMetadataError(e.errno, e.strerror) | |
f0c9fb96 | 5128 | if returncode: |
5129 | raise XAttrMetadataError(returncode, stderr) | |
0c265486 YCH |
5130 | |
5131 | ||
5132 | def random_birthday(year_field, month_field, day_field): | |
aa374bc7 AS |
5133 | start_date = datetime.date(1950, 1, 1) |
5134 | end_date = datetime.date(1995, 12, 31) | |
5135 | offset = random.randint(0, (end_date - start_date).days) | |
5136 | random_date = start_date + datetime.timedelta(offset) | |
0c265486 | 5137 | return { |
aa374bc7 AS |
5138 | year_field: str(random_date.year), |
5139 | month_field: str(random_date.month), | |
5140 | day_field: str(random_date.day), | |
0c265486 | 5141 | } |
732044af | 5142 | |
c76eb41b | 5143 | |
732044af | 5144 | # Templates for internet shortcut files, which are plain text files. |
e5a998f3 | 5145 | DOT_URL_LINK_TEMPLATE = '''\ |
732044af | 5146 | [InternetShortcut] |
5147 | URL=%(url)s | |
e5a998f3 | 5148 | ''' |
732044af | 5149 | |
e5a998f3 | 5150 | DOT_WEBLOC_LINK_TEMPLATE = '''\ |
732044af | 5151 | <?xml version="1.0" encoding="UTF-8"?> |
5152 | <!DOCTYPE plist PUBLIC "-//Apple//DTD PLIST 1.0//EN" "http://www.apple.com/DTDs/PropertyList-1.0.dtd"> | |
5153 | <plist version="1.0"> | |
5154 | <dict> | |
5155 | \t<key>URL</key> | |
5156 | \t<string>%(url)s</string> | |
5157 | </dict> | |
5158 | </plist> | |
e5a998f3 | 5159 | ''' |
732044af | 5160 | |
e5a998f3 | 5161 | DOT_DESKTOP_LINK_TEMPLATE = '''\ |
732044af | 5162 | [Desktop Entry] |
5163 | Encoding=UTF-8 | |
5164 | Name=%(filename)s | |
5165 | Type=Link | |
5166 | URL=%(url)s | |
5167 | Icon=text-html | |
e5a998f3 | 5168 | ''' |
732044af | 5169 | |
08438d2c | 5170 | LINK_TEMPLATES = { |
5171 | 'url': DOT_URL_LINK_TEMPLATE, | |
5172 | 'desktop': DOT_DESKTOP_LINK_TEMPLATE, | |
5173 | 'webloc': DOT_WEBLOC_LINK_TEMPLATE, | |
5174 | } | |
5175 | ||
732044af | 5176 | |
5177 | def iri_to_uri(iri): | |
5178 | """ | |
5179 | Converts an IRI (Internationalized Resource Identifier, allowing Unicode characters) to a URI (Uniform Resource Identifier, ASCII-only). | |
5180 | ||
5181 | The function doesn't add an additional layer of escaping; e.g., it doesn't escape `%3C` as `%253C`. Instead, it percent-escapes characters with an underlying UTF-8 encoding *besides* those already escaped, leaving the URI intact. | |
5182 | """ | |
5183 | ||
14f25df2 | 5184 | iri_parts = urllib.parse.urlparse(iri) |
732044af | 5185 | |
5186 | if '[' in iri_parts.netloc: | |
5187 | raise ValueError('IPv6 URIs are not, yet, supported.') | |
5188 | # Querying `.netloc`, when there's only one bracket, also raises a ValueError. | |
5189 | ||
5190 | # The `safe` argument values, that the following code uses, contain the characters that should not be percent-encoded. Everything else but letters, digits and '_.-' will be percent-encoded with an underlying UTF-8 encoding. Everything already percent-encoded will be left as is. | |
5191 | ||
5192 | net_location = '' | |
5193 | if iri_parts.username: | |
f9934b96 | 5194 | net_location += urllib.parse.quote(iri_parts.username, safe=r"!$%&'()*+,~") |
732044af | 5195 | if iri_parts.password is not None: |
f9934b96 | 5196 | net_location += ':' + urllib.parse.quote(iri_parts.password, safe=r"!$%&'()*+,~") |
732044af | 5197 | net_location += '@' |
5198 | ||
0f06bcd7 | 5199 | net_location += iri_parts.hostname.encode('idna').decode() # Punycode for Unicode hostnames. |
732044af | 5200 | # The 'idna' encoding produces ASCII text. |
5201 | if iri_parts.port is not None and iri_parts.port != 80: | |
5202 | net_location += ':' + str(iri_parts.port) | |
5203 | ||
f9934b96 | 5204 | return urllib.parse.urlunparse( |
732044af | 5205 | (iri_parts.scheme, |
5206 | net_location, | |
5207 | ||
f9934b96 | 5208 | urllib.parse.quote_plus(iri_parts.path, safe=r"!$%&'()*+,/:;=@|~"), |
732044af | 5209 | |
5210 | # Unsure about the `safe` argument, since this is a legacy way of handling parameters. | |
f9934b96 | 5211 | urllib.parse.quote_plus(iri_parts.params, safe=r"!$%&'()*+,/:;=@|~"), |
732044af | 5212 | |
5213 | # Not totally sure about the `safe` argument, since the source does not explicitly mention the query URI component. | |
f9934b96 | 5214 | urllib.parse.quote_plus(iri_parts.query, safe=r"!$%&'()*+,/:;=?@{|}~"), |
732044af | 5215 | |
f9934b96 | 5216 | urllib.parse.quote_plus(iri_parts.fragment, safe=r"!#$%&'()*+,/:;=?@{|}~"))) |
732044af | 5217 | |
5218 | # Source for `safe` arguments: https://url.spec.whatwg.org/#percent-encoded-bytes. | |
5219 | ||
5220 | ||
5221 | def to_high_limit_path(path): | |
5222 | if sys.platform in ['win32', 'cygwin']: | |
5223 | # Work around MAX_PATH limitation on Windows. The maximum allowed length for the individual path segments may still be quite limited. | |
e5a998f3 | 5224 | return '\\\\?\\' + os.path.abspath(path) |
732044af | 5225 | |
5226 | return path | |
76d321f6 | 5227 | |
c76eb41b | 5228 | |
7b2c3f47 | 5229 | def format_field(obj, field=None, template='%s', ignore=NO_DEFAULT, default='', func=IDENTITY): |
e0ddbd02 | 5230 | val = traverse_obj(obj, *variadic(field)) |
7b2c3f47 | 5231 | if (not val and val != 0) if ignore is NO_DEFAULT else val in variadic(ignore): |
e0ddbd02 | 5232 | return default |
7b2c3f47 | 5233 | return template % func(val) |
00dd0cd5 | 5234 | |
5235 | ||
5236 | def clean_podcast_url(url): | |
5237 | return re.sub(r'''(?x) | |
5238 | (?: | |
5239 | (?: | |
5240 | chtbl\.com/track| | |
5241 | media\.blubrry\.com| # https://create.blubrry.com/resources/podcast-media-download-statistics/getting-started/ | |
5242 | play\.podtrac\.com | |
5243 | )/[^/]+| | |
5244 | (?:dts|www)\.podtrac\.com/(?:pts/)?redirect\.[0-9a-z]{3,4}| # http://analytics.podtrac.com/how-to-measure | |
5245 | flex\.acast\.com| | |
5246 | pd(?: | |
5247 | cn\.co| # https://podcorn.com/analytics-prefix/ | |
5248 | st\.fm # https://podsights.com/docs/ | |
5249 | )/e | |
5250 | )/''', '', url) | |
ffcb8191 THD |
5251 | |
5252 | ||
5253 | _HEX_TABLE = '0123456789abcdef' | |
5254 | ||
5255 | ||
5256 | def random_uuidv4(): | |
5257 | return re.sub(r'[xy]', lambda x: _HEX_TABLE[random.randint(0, 15)], 'xxxxxxxx-xxxx-4xxx-yxxx-xxxxxxxxxxxx') | |
0202b52a | 5258 | |
5259 | ||
5260 | def make_dir(path, to_screen=None): | |
5261 | try: | |
5262 | dn = os.path.dirname(path) | |
5263 | if dn and not os.path.exists(dn): | |
5264 | os.makedirs(dn) | |
5265 | return True | |
86e5f3ed | 5266 | except OSError as err: |
0202b52a | 5267 | if callable(to_screen) is not None: |
5268 | to_screen('unable to create directory ' + error_to_compat_str(err)) | |
5269 | return False | |
f74980cb | 5270 | |
5271 | ||
5272 | def get_executable_path(): | |
b5899f4f | 5273 | from .update import _get_variant_and_executable_path |
c487cf00 | 5274 | |
b5899f4f | 5275 | return os.path.dirname(os.path.abspath(_get_variant_and_executable_path()[1])) |
f74980cb | 5276 | |
5277 | ||
2f567473 | 5278 | def load_plugins(name, suffix, namespace): |
3ae5e797 | 5279 | classes = {} |
19a03940 | 5280 | with contextlib.suppress(FileNotFoundError): |
019a94f7 ÁS |
5281 | plugins_spec = importlib.util.spec_from_file_location( |
5282 | name, os.path.join(get_executable_path(), 'ytdlp_plugins', name, '__init__.py')) | |
5283 | plugins = importlib.util.module_from_spec(plugins_spec) | |
5284 | sys.modules[plugins_spec.name] = plugins | |
5285 | plugins_spec.loader.exec_module(plugins) | |
f74980cb | 5286 | for name in dir(plugins): |
2f567473 | 5287 | if name in namespace: |
5288 | continue | |
5289 | if not name.endswith(suffix): | |
f74980cb | 5290 | continue |
5291 | klass = getattr(plugins, name) | |
3ae5e797 | 5292 | classes[name] = namespace[name] = klass |
f74980cb | 5293 | return classes |
06167fbb | 5294 | |
5295 | ||
325ebc17 | 5296 | def traverse_obj( |
f99bbfc9 | 5297 | obj, *paths, default=NO_DEFAULT, expected_type=None, get_all=True, |
325ebc17 | 5298 | casesense=True, is_user_input=False, traverse_string=False): |
ab029d7e SS |
5299 | """ |
5300 | Safely traverse nested `dict`s and `Sequence`s | |
5301 | ||
5302 | >>> obj = [{}, {"key": "value"}] | |
5303 | >>> traverse_obj(obj, (1, "key")) | |
5304 | "value" | |
5305 | ||
5306 | Each of the provided `paths` is tested and the first producing a valid result will be returned. | |
f99bbfc9 | 5307 | The next path will also be tested if the path branched but no results could be found. |
7b0127e1 | 5308 | Supported values for traversal are `Mapping`, `Sequence` and `re.Match`. |
ab029d7e SS |
5309 | A value of None is treated as the absence of a value. |
5310 | ||
5311 | The paths will be wrapped in `variadic`, so that `'key'` is conveniently the same as `('key', )`. | |
5312 | ||
5313 | The keys in the path can be one of: | |
5314 | - `None`: Return the current object. | |
7b0127e1 | 5315 | - `str`/`int`: Return `obj[key]`. For `re.Match, return `obj.group(key)`. |
ab029d7e SS |
5316 | - `slice`: Branch out and return all values in `obj[key]`. |
5317 | - `Ellipsis`: Branch out and return a list of all values. | |
5318 | - `tuple`/`list`: Branch out and return a list of all matching values. | |
5319 | Read as: `[traverse_obj(obj, branch) for branch in branches]`. | |
5320 | - `function`: Branch out and return values filtered by the function. | |
5321 | Read as: `[value for key, value in obj if function(key, value)]`. | |
5322 | For `Sequence`s, `key` is the index of the value. | |
5323 | - `dict` Transform the current object and return a matching dict. | |
5324 | Read as: `{key: traverse_obj(obj, path) for key, path in dct.items()}`. | |
5325 | ||
7b0127e1 | 5326 | `tuple`, `list`, and `dict` all support nested paths and branches. |
ab029d7e SS |
5327 | |
5328 | @params paths Paths which to traverse by. | |
5329 | @param default Value to return if the paths do not match. | |
5330 | @param expected_type If a `type`, only accept final values of this type. | |
5331 | If any other callable, try to call the function on each result. | |
5332 | @param get_all If `False`, return the first matching result, otherwise all matching ones. | |
5333 | @param casesense If `False`, consider string dictionary keys as case insensitive. | |
5334 | ||
5335 | The following are only meant to be used by YoutubeDL.prepare_outtmpl and are not part of the API | |
5336 | ||
5337 | @param is_user_input Whether the keys are generated from user input. | |
5338 | If `True` strings get converted to `int`/`slice` if needed. | |
5339 | @param traverse_string Whether to traverse into objects as strings. | |
5340 | If `True`, any non-compatible object will first be | |
5341 | converted into a string and then traversed into. | |
5342 | ||
5343 | ||
5344 | @returns The result of the object traversal. | |
5345 | If successful, `get_all=True`, and the path branches at least once, | |
5346 | then a list of results is returned instead. | |
f99bbfc9 | 5347 | A list is always returned if the last path branches and no `default` is given. |
ab029d7e SS |
5348 | """ |
5349 | is_sequence = lambda x: isinstance(x, collections.abc.Sequence) and not isinstance(x, (str, bytes)) | |
5350 | casefold = lambda k: k.casefold() if isinstance(k, str) else k | |
325ebc17 | 5351 | |
352d63fd | 5352 | if isinstance(expected_type, type): |
5353 | type_test = lambda val: val if isinstance(val, expected_type) else None | |
352d63fd | 5354 | else: |
ab029d7e SS |
5355 | type_test = lambda val: try_call(expected_type or IDENTITY, args=(val,)) |
5356 | ||
5357 | def apply_key(key, obj): | |
5358 | if obj is None: | |
5359 | return | |
5360 | ||
5361 | elif key is None: | |
5362 | yield obj | |
5363 | ||
5364 | elif isinstance(key, (list, tuple)): | |
5365 | for branch in key: | |
5366 | _, result = apply_path(obj, branch) | |
5367 | yield from result | |
5368 | ||
5369 | elif key is ...: | |
5370 | if isinstance(obj, collections.abc.Mapping): | |
5371 | yield from obj.values() | |
5372 | elif is_sequence(obj): | |
5373 | yield from obj | |
7b0127e1 SS |
5374 | elif isinstance(obj, re.Match): |
5375 | yield from obj.groups() | |
ab029d7e SS |
5376 | elif traverse_string: |
5377 | yield from str(obj) | |
5378 | ||
5379 | elif callable(key): | |
5380 | if is_sequence(obj): | |
5381 | iter_obj = enumerate(obj) | |
5382 | elif isinstance(obj, collections.abc.Mapping): | |
5383 | iter_obj = obj.items() | |
7b0127e1 SS |
5384 | elif isinstance(obj, re.Match): |
5385 | iter_obj = enumerate((obj.group(), *obj.groups())) | |
ab029d7e SS |
5386 | elif traverse_string: |
5387 | iter_obj = enumerate(str(obj)) | |
352d63fd | 5388 | else: |
ab029d7e SS |
5389 | return |
5390 | yield from (v for k, v in iter_obj if try_call(key, args=(k, v))) | |
5391 | ||
5392 | elif isinstance(key, dict): | |
5393 | iter_obj = ((k, _traverse_obj(obj, v)) for k, v in key.items()) | |
5394 | yield {k: v if v is not None else default for k, v in iter_obj | |
f99bbfc9 | 5395 | if v is not None or default is not NO_DEFAULT} |
ab029d7e | 5396 | |
7b0127e1 | 5397 | elif isinstance(obj, collections.abc.Mapping): |
ab029d7e SS |
5398 | yield (obj.get(key) if casesense or (key in obj) |
5399 | else next((v for k, v in obj.items() if casefold(k) == key), None)) | |
5400 | ||
7b0127e1 SS |
5401 | elif isinstance(obj, re.Match): |
5402 | if isinstance(key, int) or casesense: | |
5403 | with contextlib.suppress(IndexError): | |
5404 | yield obj.group(key) | |
5405 | return | |
5406 | ||
5407 | if not isinstance(key, str): | |
5408 | return | |
5409 | ||
5410 | yield next((v for k, v in obj.groupdict().items() if casefold(k) == key), None) | |
5411 | ||
ab029d7e SS |
5412 | else: |
5413 | if is_user_input: | |
5414 | key = (int_or_none(key) if ':' not in key | |
5415 | else slice(*map(int_or_none, key.split(':')))) | |
5416 | ||
5417 | if not isinstance(key, (int, slice)): | |
5418 | return | |
5419 | ||
5420 | if not is_sequence(obj): | |
5421 | if not traverse_string: | |
5422 | return | |
5423 | obj = str(obj) | |
5424 | ||
5425 | with contextlib.suppress(IndexError): | |
5426 | yield obj[key] | |
5427 | ||
5428 | def apply_path(start_obj, path): | |
5429 | objs = (start_obj,) | |
5430 | has_branched = False | |
5431 | ||
5432 | for key in variadic(path): | |
5433 | if is_user_input and key == ':': | |
5434 | key = ... | |
5435 | ||
5436 | if not casesense and isinstance(key, str): | |
5437 | key = key.casefold() | |
5438 | ||
5439 | if key is ... or isinstance(key, (list, tuple)) or callable(key): | |
5440 | has_branched = True | |
5441 | ||
5442 | key_func = functools.partial(apply_key, key) | |
5443 | objs = itertools.chain.from_iterable(map(key_func, objs)) | |
5444 | ||
5445 | return has_branched, objs | |
5446 | ||
f99bbfc9 | 5447 | def _traverse_obj(obj, path, use_list=True): |
ab029d7e SS |
5448 | has_branched, results = apply_path(obj, path) |
5449 | results = LazyList(x for x in map(type_test, results) if x is not None) | |
ab029d7e | 5450 | |
f99bbfc9 SS |
5451 | if get_all and has_branched: |
5452 | return results.exhaust() if results or use_list else None | |
5453 | ||
5454 | return results[0] if results else None | |
5455 | ||
5456 | for index, path in enumerate(paths, 1): | |
5457 | use_list = default is NO_DEFAULT and index == len(paths) | |
5458 | result = _traverse_obj(obj, path, use_list) | |
ab029d7e SS |
5459 | if result is not None: |
5460 | return result | |
5461 | ||
f99bbfc9 | 5462 | return None if default is NO_DEFAULT else default |
324ad820 | 5463 | |
5464 | ||
5465 | def traverse_dict(dictn, keys, casesense=True): | |
da4db748 | 5466 | deprecation_warning(f'"{__name__}.traverse_dict" is deprecated and may be removed ' |
5467 | f'in a future version. Use "{__name__}.traverse_obj" instead') | |
ee8dd27a | 5468 | return traverse_obj(dictn, keys, casesense=casesense, is_user_input=True, traverse_string=True) |
6606817a | 5469 | |
5470 | ||
ff91cf74 | 5471 | def get_first(obj, keys, **kwargs): |
5472 | return traverse_obj(obj, (..., *variadic(keys)), **kwargs, get_all=False) | |
5473 | ||
5474 | ||
3e9b66d7 LNO |
5475 | def time_seconds(**kwargs): |
5476 | t = datetime.datetime.now(datetime.timezone(datetime.timedelta(**kwargs))) | |
5477 | return t.timestamp() | |
5478 | ||
5479 | ||
49fa4d9a N |
5480 | # create a JSON Web Signature (jws) with HS256 algorithm |
5481 | # the resulting format is in JWS Compact Serialization | |
5482 | # implemented following JWT https://www.rfc-editor.org/rfc/rfc7519.html | |
5483 | # implemented following JWS https://www.rfc-editor.org/rfc/rfc7515.html | |
5484 | def jwt_encode_hs256(payload_data, key, headers={}): | |
5485 | header_data = { | |
5486 | 'alg': 'HS256', | |
5487 | 'typ': 'JWT', | |
5488 | } | |
5489 | if headers: | |
5490 | header_data.update(headers) | |
0f06bcd7 | 5491 | header_b64 = base64.b64encode(json.dumps(header_data).encode()) |
5492 | payload_b64 = base64.b64encode(json.dumps(payload_data).encode()) | |
5493 | h = hmac.new(key.encode(), header_b64 + b'.' + payload_b64, hashlib.sha256) | |
49fa4d9a N |
5494 | signature_b64 = base64.b64encode(h.digest()) |
5495 | token = header_b64 + b'.' + payload_b64 + b'.' + signature_b64 | |
5496 | return token | |
819e0531 | 5497 | |
5498 | ||
16b0d7e6 | 5499 | # can be extended in future to verify the signature and parse header and return the algorithm used if it's not HS256 |
5500 | def jwt_decode_hs256(jwt): | |
5501 | header_b64, payload_b64, signature_b64 = jwt.split('.') | |
5502 | payload_data = json.loads(base64.urlsafe_b64decode(payload_b64)) | |
5503 | return payload_data | |
5504 | ||
5505 | ||
53973b4d | 5506 | WINDOWS_VT_MODE = False if compat_os_name == 'nt' else None |
5507 | ||
5508 | ||
7a32c70d | 5509 | @functools.cache |
819e0531 | 5510 | def supports_terminal_sequences(stream): |
5511 | if compat_os_name == 'nt': | |
8a82af35 | 5512 | if not WINDOWS_VT_MODE: |
819e0531 | 5513 | return False |
5514 | elif not os.getenv('TERM'): | |
5515 | return False | |
5516 | try: | |
5517 | return stream.isatty() | |
5518 | except BaseException: | |
5519 | return False | |
5520 | ||
5521 | ||
53973b4d | 5522 | def windows_enable_vt_mode(): # TODO: Do this the proper way https://bugs.python.org/issue30075 |
8a82af35 | 5523 | if get_windows_version() < (10, 0, 10586): |
53973b4d | 5524 | return |
5525 | global WINDOWS_VT_MODE | |
53973b4d | 5526 | try: |
f0c9fb96 | 5527 | Popen.run('', shell=True) |
53973b4d | 5528 | except Exception: |
5529 | return | |
5530 | ||
5531 | WINDOWS_VT_MODE = True | |
5532 | supports_terminal_sequences.cache_clear() | |
5533 | ||
5534 | ||
ec11a9f4 | 5535 | _terminal_sequences_re = re.compile('\033\\[[^m]+m') |
5536 | ||
5537 | ||
5538 | def remove_terminal_sequences(string): | |
5539 | return _terminal_sequences_re.sub('', string) | |
5540 | ||
5541 | ||
5542 | def number_of_digits(number): | |
5543 | return len('%d' % number) | |
34921b43 | 5544 | |
5545 | ||
5546 | def join_nonempty(*values, delim='-', from_dict=None): | |
5547 | if from_dict is not None: | |
7b2c3f47 | 5548 | values = (traverse_obj(from_dict, variadic(v)) for v in values) |
34921b43 | 5549 | return delim.join(map(str, filter(None, values))) |
06e57990 | 5550 | |
5551 | ||
27231526 ZM |
5552 | def scale_thumbnails_to_max_format_width(formats, thumbnails, url_width_re): |
5553 | """ | |
5554 | Find the largest format dimensions in terms of video width and, for each thumbnail: | |
5555 | * Modify the URL: Match the width with the provided regex and replace with the former width | |
5556 | * Update dimensions | |
5557 | ||
5558 | This function is useful with video services that scale the provided thumbnails on demand | |
5559 | """ | |
5560 | _keys = ('width', 'height') | |
5561 | max_dimensions = max( | |
86e5f3ed | 5562 | (tuple(format.get(k) or 0 for k in _keys) for format in formats), |
27231526 ZM |
5563 | default=(0, 0)) |
5564 | if not max_dimensions[0]: | |
5565 | return thumbnails | |
5566 | return [ | |
5567 | merge_dicts( | |
5568 | {'url': re.sub(url_width_re, str(max_dimensions[0]), thumbnail['url'])}, | |
5569 | dict(zip(_keys, max_dimensions)), thumbnail) | |
5570 | for thumbnail in thumbnails | |
5571 | ] | |
5572 | ||
5573 | ||
93c8410d LNO |
5574 | def parse_http_range(range): |
5575 | """ Parse value of "Range" or "Content-Range" HTTP header into tuple. """ | |
5576 | if not range: | |
5577 | return None, None, None | |
5578 | crg = re.search(r'bytes[ =](\d+)-(\d+)?(?:/(\d+))?', range) | |
5579 | if not crg: | |
5580 | return None, None, None | |
5581 | return int(crg.group(1)), int_or_none(crg.group(2)), int_or_none(crg.group(3)) | |
5582 | ||
5583 | ||
6b9e832d | 5584 | def read_stdin(what): |
5585 | eof = 'Ctrl+Z' if compat_os_name == 'nt' else 'Ctrl+D' | |
5586 | write_string(f'Reading {what} from STDIN - EOF ({eof}) to end:\n') | |
5587 | return sys.stdin | |
5588 | ||
5589 | ||
a904a7f8 L |
5590 | def determine_file_encoding(data): |
5591 | """ | |
88f60feb | 5592 | Detect the text encoding used |
a904a7f8 L |
5593 | @returns (encoding, bytes to skip) |
5594 | """ | |
5595 | ||
88f60feb | 5596 | # BOM marks are given priority over declarations |
a904a7f8 | 5597 | for bom, enc in BOMS: |
a904a7f8 L |
5598 | if data.startswith(bom): |
5599 | return enc, len(bom) | |
5600 | ||
88f60feb | 5601 | # Strip off all null bytes to match even when UTF-16 or UTF-32 is used. |
5602 | # We ignore the endianness to get a good enough match | |
a904a7f8 | 5603 | data = data.replace(b'\0', b'') |
88f60feb | 5604 | mobj = re.match(rb'(?m)^#\s*coding\s*:\s*(\S+)\s*$', data) |
5605 | return mobj.group(1).decode() if mobj else None, 0 | |
a904a7f8 L |
5606 | |
5607 | ||
06e57990 | 5608 | class Config: |
5609 | own_args = None | |
9e491463 | 5610 | parsed_args = None |
06e57990 | 5611 | filename = None |
5612 | __initialized = False | |
5613 | ||
5614 | def __init__(self, parser, label=None): | |
9e491463 | 5615 | self.parser, self.label = parser, label |
06e57990 | 5616 | self._loaded_paths, self.configs = set(), [] |
5617 | ||
5618 | def init(self, args=None, filename=None): | |
5619 | assert not self.__initialized | |
284a60c5 | 5620 | self.own_args, self.filename = args, filename |
5621 | return self.load_configs() | |
5622 | ||
5623 | def load_configs(self): | |
65662dff | 5624 | directory = '' |
284a60c5 | 5625 | if self.filename: |
5626 | location = os.path.realpath(self.filename) | |
65662dff | 5627 | directory = os.path.dirname(location) |
06e57990 | 5628 | if location in self._loaded_paths: |
5629 | return False | |
5630 | self._loaded_paths.add(location) | |
5631 | ||
284a60c5 | 5632 | self.__initialized = True |
5633 | opts, _ = self.parser.parse_known_args(self.own_args) | |
5634 | self.parsed_args = self.own_args | |
9e491463 | 5635 | for location in opts.config_locations or []: |
6b9e832d | 5636 | if location == '-': |
1060f82f | 5637 | if location in self._loaded_paths: |
5638 | continue | |
5639 | self._loaded_paths.add(location) | |
6b9e832d | 5640 | self.append_config(shlex.split(read_stdin('options'), comments=True), label='stdin') |
5641 | continue | |
65662dff | 5642 | location = os.path.join(directory, expand_path(location)) |
06e57990 | 5643 | if os.path.isdir(location): |
5644 | location = os.path.join(location, 'yt-dlp.conf') | |
5645 | if not os.path.exists(location): | |
9e491463 | 5646 | self.parser.error(f'config location {location} does not exist') |
06e57990 | 5647 | self.append_config(self.read_file(location), location) |
5648 | return True | |
5649 | ||
5650 | def __str__(self): | |
5651 | label = join_nonempty( | |
5652 | self.label, 'config', f'"{self.filename}"' if self.filename else '', | |
5653 | delim=' ') | |
5654 | return join_nonempty( | |
5655 | self.own_args is not None and f'{label[0].upper()}{label[1:]}: {self.hide_login_info(self.own_args)}', | |
5656 | *(f'\n{c}'.replace('\n', '\n| ')[1:] for c in self.configs), | |
5657 | delim='\n') | |
5658 | ||
7a32c70d | 5659 | @staticmethod |
06e57990 | 5660 | def read_file(filename, default=[]): |
5661 | try: | |
a904a7f8 | 5662 | optionf = open(filename, 'rb') |
86e5f3ed | 5663 | except OSError: |
06e57990 | 5664 | return default # silently skip if file is not present |
a904a7f8 L |
5665 | try: |
5666 | enc, skip = determine_file_encoding(optionf.read(512)) | |
5667 | optionf.seek(skip, io.SEEK_SET) | |
5668 | except OSError: | |
5669 | enc = None # silently skip read errors | |
06e57990 | 5670 | try: |
5671 | # FIXME: https://github.com/ytdl-org/youtube-dl/commit/dfe5fa49aed02cf36ba9f743b11b0903554b5e56 | |
a904a7f8 | 5672 | contents = optionf.read().decode(enc or preferredencoding()) |
f9934b96 | 5673 | res = shlex.split(contents, comments=True) |
44a6fcff | 5674 | except Exception as err: |
5675 | raise ValueError(f'Unable to parse "{filename}": {err}') | |
06e57990 | 5676 | finally: |
5677 | optionf.close() | |
5678 | return res | |
5679 | ||
7a32c70d | 5680 | @staticmethod |
06e57990 | 5681 | def hide_login_info(opts): |
86e5f3ed | 5682 | PRIVATE_OPTS = {'-p', '--password', '-u', '--username', '--video-password', '--ap-password', '--ap-username'} |
06e57990 | 5683 | eqre = re.compile('^(?P<key>' + ('|'.join(re.escape(po) for po in PRIVATE_OPTS)) + ')=.+$') |
5684 | ||
5685 | def _scrub_eq(o): | |
5686 | m = eqre.match(o) | |
5687 | if m: | |
5688 | return m.group('key') + '=PRIVATE' | |
5689 | else: | |
5690 | return o | |
5691 | ||
5692 | opts = list(map(_scrub_eq, opts)) | |
5693 | for idx, opt in enumerate(opts): | |
5694 | if opt in PRIVATE_OPTS and idx + 1 < len(opts): | |
5695 | opts[idx + 1] = 'PRIVATE' | |
5696 | return opts | |
5697 | ||
5698 | def append_config(self, *args, label=None): | |
9e491463 | 5699 | config = type(self)(self.parser, label) |
06e57990 | 5700 | config._loaded_paths = self._loaded_paths |
5701 | if config.init(*args): | |
5702 | self.configs.append(config) | |
5703 | ||
7a32c70d | 5704 | @property |
06e57990 | 5705 | def all_args(self): |
5706 | for config in reversed(self.configs): | |
5707 | yield from config.all_args | |
9e491463 | 5708 | yield from self.parsed_args or [] |
5709 | ||
5710 | def parse_known_args(self, **kwargs): | |
5711 | return self.parser.parse_known_args(self.all_args, **kwargs) | |
06e57990 | 5712 | |
5713 | def parse_args(self): | |
9e491463 | 5714 | return self.parser.parse_args(self.all_args) |
da42679b LNO |
5715 | |
5716 | ||
5717 | class WebSocketsWrapper(): | |
5718 | """Wraps websockets module to use in non-async scopes""" | |
abfecb7b | 5719 | pool = None |
da42679b | 5720 | |
3cea3edd | 5721 | def __init__(self, url, headers=None, connect=True): |
059bc4db | 5722 | self.loop = asyncio.new_event_loop() |
9cd08050 | 5723 | # XXX: "loop" is deprecated |
5724 | self.conn = websockets.connect( | |
5725 | url, extra_headers=headers, ping_interval=None, | |
5726 | close_timeout=float('inf'), loop=self.loop, ping_timeout=float('inf')) | |
3cea3edd LNO |
5727 | if connect: |
5728 | self.__enter__() | |
15dfb392 | 5729 | atexit.register(self.__exit__, None, None, None) |
da42679b LNO |
5730 | |
5731 | def __enter__(self): | |
3cea3edd | 5732 | if not self.pool: |
9cd08050 | 5733 | self.pool = self.run_with_loop(self.conn.__aenter__(), self.loop) |
da42679b LNO |
5734 | return self |
5735 | ||
5736 | def send(self, *args): | |
5737 | self.run_with_loop(self.pool.send(*args), self.loop) | |
5738 | ||
5739 | def recv(self, *args): | |
5740 | return self.run_with_loop(self.pool.recv(*args), self.loop) | |
5741 | ||
5742 | def __exit__(self, type, value, traceback): | |
5743 | try: | |
5744 | return self.run_with_loop(self.conn.__aexit__(type, value, traceback), self.loop) | |
5745 | finally: | |
5746 | self.loop.close() | |
15dfb392 | 5747 | self._cancel_all_tasks(self.loop) |
da42679b LNO |
5748 | |
5749 | # taken from https://github.com/python/cpython/blob/3.9/Lib/asyncio/runners.py with modifications | |
5750 | # for contributors: If there's any new library using asyncio needs to be run in non-async, move these function out of this class | |
7a32c70d | 5751 | @staticmethod |
da42679b | 5752 | def run_with_loop(main, loop): |
059bc4db | 5753 | if not asyncio.iscoroutine(main): |
da42679b LNO |
5754 | raise ValueError(f'a coroutine was expected, got {main!r}') |
5755 | ||
5756 | try: | |
5757 | return loop.run_until_complete(main) | |
5758 | finally: | |
5759 | loop.run_until_complete(loop.shutdown_asyncgens()) | |
5760 | if hasattr(loop, 'shutdown_default_executor'): | |
5761 | loop.run_until_complete(loop.shutdown_default_executor()) | |
5762 | ||
7a32c70d | 5763 | @staticmethod |
da42679b | 5764 | def _cancel_all_tasks(loop): |
059bc4db | 5765 | to_cancel = asyncio.all_tasks(loop) |
da42679b LNO |
5766 | |
5767 | if not to_cancel: | |
5768 | return | |
5769 | ||
5770 | for task in to_cancel: | |
5771 | task.cancel() | |
5772 | ||
9cd08050 | 5773 | # XXX: "loop" is removed in python 3.10+ |
da42679b | 5774 | loop.run_until_complete( |
059bc4db | 5775 | asyncio.gather(*to_cancel, loop=loop, return_exceptions=True)) |
da42679b LNO |
5776 | |
5777 | for task in to_cancel: | |
5778 | if task.cancelled(): | |
5779 | continue | |
5780 | if task.exception() is not None: | |
5781 | loop.call_exception_handler({ | |
5782 | 'message': 'unhandled exception during asyncio.run() shutdown', | |
5783 | 'exception': task.exception(), | |
5784 | 'task': task, | |
5785 | }) | |
5786 | ||
5787 | ||
8b7539d2 | 5788 | def merge_headers(*dicts): |
08d30158 | 5789 | """Merge dicts of http headers case insensitively, prioritizing the latter ones""" |
76aa9913 | 5790 | return {k.title(): v for k, v in itertools.chain.from_iterable(map(dict.items, dicts))} |
28787f16 | 5791 | |
5792 | ||
b1f94422 | 5793 | def cached_method(f): |
5794 | """Cache a method""" | |
5795 | signature = inspect.signature(f) | |
5796 | ||
7a32c70d | 5797 | @functools.wraps(f) |
b1f94422 | 5798 | def wrapper(self, *args, **kwargs): |
5799 | bound_args = signature.bind(self, *args, **kwargs) | |
5800 | bound_args.apply_defaults() | |
5801 | key = tuple(bound_args.arguments.values()) | |
5802 | ||
5803 | if not hasattr(self, '__cached_method__cache'): | |
5804 | self.__cached_method__cache = {} | |
5805 | cache = self.__cached_method__cache.setdefault(f.__name__, {}) | |
5806 | if key not in cache: | |
5807 | cache[key] = f(self, *args, **kwargs) | |
5808 | return cache[key] | |
5809 | return wrapper | |
5810 | ||
5811 | ||
28787f16 | 5812 | class classproperty: |
b1f94422 | 5813 | """property access for class methods""" |
c487cf00 | 5814 | |
5815 | def __init__(self, func): | |
5816 | functools.update_wrapper(self, func) | |
5817 | self.func = func | |
28787f16 | 5818 | |
5819 | def __get__(self, _, cls): | |
c487cf00 | 5820 | return self.func(cls) |
19a03940 | 5821 | |
5822 | ||
64fa820c | 5823 | class Namespace(types.SimpleNamespace): |
591bb9d3 | 5824 | """Immutable namespace""" |
591bb9d3 | 5825 | |
7896214c | 5826 | def __iter__(self): |
64fa820c | 5827 | return iter(self.__dict__.values()) |
7896214c | 5828 | |
7a32c70d | 5829 | @property |
64fa820c | 5830 | def items_(self): |
5831 | return self.__dict__.items() | |
9b8ee23b | 5832 | |
5833 | ||
8dc59305 | 5834 | MEDIA_EXTENSIONS = Namespace( |
5835 | common_video=('avi', 'flv', 'mkv', 'mov', 'mp4', 'webm'), | |
5836 | video=('3g2', '3gp', 'f4v', 'mk3d', 'divx', 'mpg', 'ogv', 'm4v', 'wmv'), | |
5837 | common_audio=('aiff', 'alac', 'flac', 'm4a', 'mka', 'mp3', 'ogg', 'opus', 'wav'), | |
5838 | audio=('aac', 'ape', 'asf', 'f4a', 'f4b', 'm4b', 'm4p', 'm4r', 'oga', 'ogx', 'spx', 'vorbis', 'wma'), | |
5839 | thumbnails=('jpg', 'png', 'webp'), | |
5840 | storyboards=('mhtml', ), | |
5841 | subtitles=('srt', 'vtt', 'ass', 'lrc'), | |
5842 | manifests=('f4f', 'f4m', 'm3u8', 'smil', 'mpd'), | |
5843 | ) | |
5844 | MEDIA_EXTENSIONS.video += MEDIA_EXTENSIONS.common_video | |
5845 | MEDIA_EXTENSIONS.audio += MEDIA_EXTENSIONS.common_audio | |
5846 | ||
5847 | KNOWN_EXTENSIONS = (*MEDIA_EXTENSIONS.video, *MEDIA_EXTENSIONS.audio, *MEDIA_EXTENSIONS.manifests) | |
5848 | ||
5849 | ||
be5c1ae8 | 5850 | class RetryManager: |
5851 | """Usage: | |
5852 | for retry in RetryManager(...): | |
5853 | try: | |
5854 | ... | |
5855 | except SomeException as err: | |
5856 | retry.error = err | |
5857 | continue | |
5858 | """ | |
5859 | attempt, _error = 0, None | |
5860 | ||
5861 | def __init__(self, _retries, _error_callback, **kwargs): | |
5862 | self.retries = _retries or 0 | |
5863 | self.error_callback = functools.partial(_error_callback, **kwargs) | |
5864 | ||
5865 | def _should_retry(self): | |
5866 | return self._error is not NO_DEFAULT and self.attempt <= self.retries | |
5867 | ||
7a32c70d | 5868 | @property |
be5c1ae8 | 5869 | def error(self): |
5870 | if self._error is NO_DEFAULT: | |
5871 | return None | |
5872 | return self._error | |
5873 | ||
7a32c70d | 5874 | @error.setter |
be5c1ae8 | 5875 | def error(self, value): |
5876 | self._error = value | |
5877 | ||
5878 | def __iter__(self): | |
5879 | while self._should_retry(): | |
5880 | self.error = NO_DEFAULT | |
5881 | self.attempt += 1 | |
5882 | yield self | |
5883 | if self.error: | |
5884 | self.error_callback(self.error, self.attempt, self.retries) | |
5885 | ||
7a32c70d | 5886 | @staticmethod |
be5c1ae8 | 5887 | def report_retry(e, count, retries, *, sleep_func, info, warn, error=None, suffix=None): |
5888 | """Utility function for reporting retries""" | |
5889 | if count > retries: | |
5890 | if error: | |
5891 | return error(f'{e}. Giving up after {count - 1} retries') if count > 1 else error(str(e)) | |
5892 | raise e | |
5893 | ||
5894 | if not count: | |
5895 | return warn(e) | |
5896 | elif isinstance(e, ExtractorError): | |
3ce29336 | 5897 | e = remove_end(str_or_none(e.cause) or e.orig_msg, '.') |
be5c1ae8 | 5898 | warn(f'{e}. Retrying{format_field(suffix, None, " %s")} ({count}/{retries})...') |
5899 | ||
5900 | delay = float_or_none(sleep_func(n=count - 1)) if callable(sleep_func) else sleep_func | |
5901 | if delay: | |
5902 | info(f'Sleeping {delay:.2f} seconds ...') | |
5903 | time.sleep(delay) | |
5904 | ||
5905 | ||
0647d925 | 5906 | def make_archive_id(ie, video_id): |
5907 | ie_key = ie if isinstance(ie, str) else ie.ie_key() | |
5908 | return f'{ie_key.lower()} {video_id}' | |
5909 | ||
5910 | ||
a1c5bd82 | 5911 | def truncate_string(s, left, right=0): |
5912 | assert left > 3 and right >= 0 | |
5913 | if s is None or len(s) <= left + right: | |
5914 | return s | |
5915 | return f'{s[:left-3]}...{s[-right:]}' | |
5916 | ||
5917 | ||
5314b521 | 5918 | def orderedSet_from_options(options, alias_dict, *, use_regex=False, start=None): |
5919 | assert 'all' in alias_dict, '"all" alias is required' | |
5920 | requested = list(start or []) | |
5921 | for val in options: | |
5922 | discard = val.startswith('-') | |
5923 | if discard: | |
5924 | val = val[1:] | |
5925 | ||
5926 | if val in alias_dict: | |
5927 | val = alias_dict[val] if not discard else [ | |
5928 | i[1:] if i.startswith('-') else f'-{i}' for i in alias_dict[val]] | |
5929 | # NB: Do not allow regex in aliases for performance | |
5930 | requested = orderedSet_from_options(val, alias_dict, start=requested) | |
5931 | continue | |
5932 | ||
5933 | current = (filter(re.compile(val, re.I).fullmatch, alias_dict['all']) if use_regex | |
5934 | else [val] if val in alias_dict['all'] else None) | |
5935 | if current is None: | |
5936 | raise ValueError(val) | |
5937 | ||
5938 | if discard: | |
5939 | for item in current: | |
5940 | while item in requested: | |
5941 | requested.remove(item) | |
5942 | else: | |
5943 | requested.extend(current) | |
5944 | ||
5945 | return orderedSet(requested) | |
5946 | ||
5947 | ||
9b8ee23b | 5948 | # Deprecated |
5949 | has_certifi = bool(certifi) | |
5950 | has_websockets = bool(websockets) |