]>
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 |
62e609ab | 8 | import contextlib |
c496ca96 | 9 | import datetime |
0c265486 | 10 | import email.header |
f8271158 | 11 | import email.utils |
f45c185f | 12 | import errno |
d77c3dfd | 13 | import gzip |
49fa4d9a N |
14 | import hashlib |
15 | import hmac | |
ac668111 | 16 | import html.entities |
17 | import html.parser | |
54007a45 | 18 | import http.client |
19 | import http.cookiejar | |
019a94f7 | 20 | import importlib.util |
b1f94422 | 21 | import inspect |
03f9daab | 22 | import io |
79a2e94e | 23 | import itertools |
f4bfd65f | 24 | import json |
d77c3dfd | 25 | import locale |
02dbf93f | 26 | import math |
f8271158 | 27 | import mimetypes |
347de493 | 28 | import operator |
d77c3dfd | 29 | import os |
c496ca96 | 30 | import platform |
773f291d | 31 | import random |
d77c3dfd | 32 | import re |
f8271158 | 33 | import shlex |
c496ca96 | 34 | import socket |
79a2e94e | 35 | import ssl |
ac668111 | 36 | import struct |
1c088fa8 | 37 | import subprocess |
d77c3dfd | 38 | import sys |
181c8655 | 39 | import tempfile |
c380cc28 | 40 | import time |
01951dda | 41 | import traceback |
64fa820c | 42 | import types |
989a01c2 | 43 | import unicodedata |
14f25df2 | 44 | import urllib.error |
f8271158 | 45 | import urllib.parse |
ac668111 | 46 | import urllib.request |
bcf89ce6 | 47 | import xml.etree.ElementTree |
d77c3dfd | 48 | import zlib |
d77c3dfd | 49 | |
6929b41a | 50 | from .compat import functools # isort: split |
8c25f81b | 51 | from .compat import ( |
36e6f62c | 52 | compat_etree_fromstring, |
51098426 | 53 | compat_expanduser, |
f8271158 | 54 | compat_HTMLParseError, |
efa97bdc | 55 | compat_os_name, |
702ccf2d | 56 | compat_shlex_quote, |
8c25f81b | 57 | ) |
ac668111 | 58 | from .dependencies import brotli, certifi, websockets, xattr |
f8271158 | 59 | from .socks import ProxyType, sockssocket |
71aff188 | 60 | |
4644ac55 | 61 | |
51fb4995 YCH |
62 | def register_socks_protocols(): |
63 | # "Register" SOCKS protocols | |
d5ae6bb5 YCH |
64 | # In Python < 2.6.5, urlsplit() suffers from bug https://bugs.python.org/issue7904 |
65 | # URLs with protocols not in urlparse.uses_netloc are not handled correctly | |
51fb4995 | 66 | for scheme in ('socks', 'socks4', 'socks4a', 'socks5'): |
14f25df2 | 67 | if scheme not in urllib.parse.uses_netloc: |
68 | urllib.parse.uses_netloc.append(scheme) | |
51fb4995 YCH |
69 | |
70 | ||
468e2e92 FV |
71 | # This is not clearly defined otherwise |
72 | compiled_regex_type = type(re.compile('')) | |
73 | ||
f7a147e3 S |
74 | |
75 | def random_user_agent(): | |
76 | _USER_AGENT_TPL = 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/%s Safari/537.36' | |
77 | _CHROME_VERSIONS = ( | |
19b4c74d | 78 | '90.0.4430.212', |
79 | '90.0.4430.24', | |
80 | '90.0.4430.70', | |
81 | '90.0.4430.72', | |
82 | '90.0.4430.85', | |
83 | '90.0.4430.93', | |
84 | '91.0.4472.101', | |
85 | '91.0.4472.106', | |
86 | '91.0.4472.114', | |
87 | '91.0.4472.124', | |
88 | '91.0.4472.164', | |
89 | '91.0.4472.19', | |
90 | '91.0.4472.77', | |
91 | '92.0.4515.107', | |
92 | '92.0.4515.115', | |
93 | '92.0.4515.131', | |
94 | '92.0.4515.159', | |
95 | '92.0.4515.43', | |
96 | '93.0.4556.0', | |
97 | '93.0.4577.15', | |
98 | '93.0.4577.63', | |
99 | '93.0.4577.82', | |
100 | '94.0.4606.41', | |
101 | '94.0.4606.54', | |
102 | '94.0.4606.61', | |
103 | '94.0.4606.71', | |
104 | '94.0.4606.81', | |
105 | '94.0.4606.85', | |
106 | '95.0.4638.17', | |
107 | '95.0.4638.50', | |
108 | '95.0.4638.54', | |
109 | '95.0.4638.69', | |
110 | '95.0.4638.74', | |
111 | '96.0.4664.18', | |
112 | '96.0.4664.45', | |
113 | '96.0.4664.55', | |
114 | '96.0.4664.93', | |
115 | '97.0.4692.20', | |
f7a147e3 S |
116 | ) |
117 | return _USER_AGENT_TPL % random.choice(_CHROME_VERSIONS) | |
118 | ||
119 | ||
4390d5ec | 120 | SUPPORTED_ENCODINGS = [ |
121 | 'gzip', 'deflate' | |
122 | ] | |
9b8ee23b | 123 | if brotli: |
4390d5ec | 124 | SUPPORTED_ENCODINGS.append('br') |
125 | ||
3e669f36 | 126 | std_headers = { |
f7a147e3 | 127 | 'User-Agent': random_user_agent(), |
59ae15a5 | 128 | 'Accept': 'text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8', |
59ae15a5 | 129 | 'Accept-Language': 'en-us,en;q=0.5', |
b1156c1e | 130 | 'Sec-Fetch-Mode': 'navigate', |
3e669f36 | 131 | } |
f427df17 | 132 | |
5f6a1245 | 133 | |
fb37eb25 S |
134 | USER_AGENTS = { |
135 | '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', | |
136 | } | |
137 | ||
138 | ||
bf42a990 | 139 | NO_DEFAULT = object() |
7b2c3f47 | 140 | IDENTITY = lambda x: x |
bf42a990 | 141 | |
7105440c YCH |
142 | ENGLISH_MONTH_NAMES = [ |
143 | 'January', 'February', 'March', 'April', 'May', 'June', | |
144 | 'July', 'August', 'September', 'October', 'November', 'December'] | |
145 | ||
f6717dec S |
146 | MONTH_NAMES = { |
147 | 'en': ENGLISH_MONTH_NAMES, | |
148 | 'fr': [ | |
3e4185c3 S |
149 | 'janvier', 'février', 'mars', 'avril', 'mai', 'juin', |
150 | 'juillet', 'août', 'septembre', 'octobre', 'novembre', 'décembre'], | |
f6717dec | 151 | } |
a942d6cb | 152 | |
8f53dc44 | 153 | # From https://github.com/python/cpython/blob/3.11/Lib/email/_parseaddr.py#L36-L42 |
154 | TIMEZONE_NAMES = { | |
155 | 'UT': 0, 'UTC': 0, 'GMT': 0, 'Z': 0, | |
156 | 'AST': -4, 'ADT': -3, # Atlantic (used in Canada) | |
157 | 'EST': -5, 'EDT': -4, # Eastern | |
158 | 'CST': -6, 'CDT': -5, # Central | |
159 | 'MST': -7, 'MDT': -6, # Mountain | |
160 | 'PST': -8, 'PDT': -7 # Pacific | |
161 | } | |
162 | ||
c587cbb7 | 163 | # needed for sanitizing filenames in restricted mode |
c8827027 | 164 | ACCENT_CHARS = dict(zip('ÂÃÄÀÁÅÆÇÈÉÊËÌÍÎÏÐÑÒÓÔÕÖŐØŒÙÚÛÜŰÝÞßàáâãäåæçèéêëìíîïðñòóôõöőøœùúûüűýþÿ', |
fd35d8cd JW |
165 | itertools.chain('AAAAAA', ['AE'], 'CEEEEIIIIDNOOOOOOO', ['OE'], 'UUUUUY', ['TH', 'ss'], |
166 | 'aaaaaa', ['ae'], 'ceeeeiiiionooooooo', ['oe'], 'uuuuuy', ['th'], 'y'))) | |
c587cbb7 | 167 | |
46f59e89 S |
168 | DATE_FORMATS = ( |
169 | '%d %B %Y', | |
170 | '%d %b %Y', | |
171 | '%B %d %Y', | |
cb655f34 S |
172 | '%B %dst %Y', |
173 | '%B %dnd %Y', | |
9d30c213 | 174 | '%B %drd %Y', |
cb655f34 | 175 | '%B %dth %Y', |
46f59e89 | 176 | '%b %d %Y', |
cb655f34 S |
177 | '%b %dst %Y', |
178 | '%b %dnd %Y', | |
9d30c213 | 179 | '%b %drd %Y', |
cb655f34 | 180 | '%b %dth %Y', |
46f59e89 S |
181 | '%b %dst %Y %I:%M', |
182 | '%b %dnd %Y %I:%M', | |
9d30c213 | 183 | '%b %drd %Y %I:%M', |
46f59e89 S |
184 | '%b %dth %Y %I:%M', |
185 | '%Y %m %d', | |
186 | '%Y-%m-%d', | |
bccdbd22 | 187 | '%Y.%m.%d.', |
46f59e89 | 188 | '%Y/%m/%d', |
81c13222 | 189 | '%Y/%m/%d %H:%M', |
46f59e89 | 190 | '%Y/%m/%d %H:%M:%S', |
1931a55e THD |
191 | '%Y%m%d%H%M', |
192 | '%Y%m%d%H%M%S', | |
4f3fa23e | 193 | '%Y%m%d', |
0c1c6f4b | 194 | '%Y-%m-%d %H:%M', |
46f59e89 S |
195 | '%Y-%m-%d %H:%M:%S', |
196 | '%Y-%m-%d %H:%M:%S.%f', | |
5014558a | 197 | '%Y-%m-%d %H:%M:%S:%f', |
46f59e89 S |
198 | '%d.%m.%Y %H:%M', |
199 | '%d.%m.%Y %H.%M', | |
200 | '%Y-%m-%dT%H:%M:%SZ', | |
201 | '%Y-%m-%dT%H:%M:%S.%fZ', | |
202 | '%Y-%m-%dT%H:%M:%S.%f0Z', | |
203 | '%Y-%m-%dT%H:%M:%S', | |
204 | '%Y-%m-%dT%H:%M:%S.%f', | |
205 | '%Y-%m-%dT%H:%M', | |
c6eed6b8 S |
206 | '%b %d %Y at %H:%M', |
207 | '%b %d %Y at %H:%M:%S', | |
b555ae9b S |
208 | '%B %d %Y at %H:%M', |
209 | '%B %d %Y at %H:%M:%S', | |
a63d9bd0 | 210 | '%H:%M %d-%b-%Y', |
46f59e89 S |
211 | ) |
212 | ||
213 | DATE_FORMATS_DAY_FIRST = list(DATE_FORMATS) | |
214 | DATE_FORMATS_DAY_FIRST.extend([ | |
215 | '%d-%m-%Y', | |
216 | '%d.%m.%Y', | |
217 | '%d.%m.%y', | |
218 | '%d/%m/%Y', | |
219 | '%d/%m/%y', | |
220 | '%d/%m/%Y %H:%M:%S', | |
47304e07 | 221 | '%d-%m-%Y %H:%M', |
46f59e89 S |
222 | ]) |
223 | ||
224 | DATE_FORMATS_MONTH_FIRST = list(DATE_FORMATS) | |
225 | DATE_FORMATS_MONTH_FIRST.extend([ | |
226 | '%m-%d-%Y', | |
227 | '%m.%d.%Y', | |
228 | '%m/%d/%Y', | |
229 | '%m/%d/%y', | |
230 | '%m/%d/%Y %H:%M:%S', | |
231 | ]) | |
232 | ||
06b3fe29 | 233 | PACKED_CODES_RE = r"}\('(.+)',(\d+),(\d+),'([^']+)'\.split\('\|'\)" |
ae61d108 | 234 | JSON_LD_RE = r'(?is)<script[^>]+type=(["\']?)application/ld\+json\1[^>]*>\s*(?P<json_ld>{.+?})\s*</script>' |
06b3fe29 | 235 | |
1d485a1a | 236 | NUMBER_RE = r'\d+(?:\.\d+)?' |
237 | ||
7105440c | 238 | |
0b9c08b4 | 239 | @functools.cache |
d77c3dfd | 240 | def preferredencoding(): |
59ae15a5 | 241 | """Get preferred encoding. |
d77c3dfd | 242 | |
59ae15a5 PH |
243 | Returns the best encoding scheme for the system, based on |
244 | locale.getpreferredencoding() and some further tweaks. | |
245 | """ | |
246 | try: | |
247 | pref = locale.getpreferredencoding() | |
28e614de | 248 | 'TEST'.encode(pref) |
70a1165b | 249 | except Exception: |
59ae15a5 | 250 | pref = 'UTF-8' |
bae611f2 | 251 | |
59ae15a5 | 252 | return pref |
d77c3dfd | 253 | |
f4bfd65f | 254 | |
181c8655 | 255 | def write_json_file(obj, fn): |
1394646a | 256 | """ Encode obj as JSON and write it to fn, atomically if possible """ |
181c8655 | 257 | |
cfb0511d | 258 | tf = tempfile.NamedTemporaryFile( |
259 | prefix=f'{os.path.basename(fn)}.', dir=os.path.dirname(fn), | |
260 | suffix='.tmp', delete=False, mode='w', encoding='utf-8') | |
181c8655 PH |
261 | |
262 | try: | |
263 | with tf: | |
45d86abe | 264 | json.dump(obj, tf, ensure_ascii=False) |
1394646a IK |
265 | if sys.platform == 'win32': |
266 | # Need to remove existing file on Windows, else os.rename raises | |
267 | # WindowsError or FileExistsError. | |
19a03940 | 268 | with contextlib.suppress(OSError): |
1394646a | 269 | os.unlink(fn) |
19a03940 | 270 | with contextlib.suppress(OSError): |
9cd5f54e R |
271 | mask = os.umask(0) |
272 | os.umask(mask) | |
273 | os.chmod(tf.name, 0o666 & ~mask) | |
181c8655 | 274 | os.rename(tf.name, fn) |
70a1165b | 275 | except Exception: |
19a03940 | 276 | with contextlib.suppress(OSError): |
181c8655 | 277 | os.remove(tf.name) |
181c8655 PH |
278 | raise |
279 | ||
280 | ||
cfb0511d | 281 | def find_xpath_attr(node, xpath, key, val=None): |
282 | """ Find the xpath xpath[@key=val] """ | |
283 | assert re.match(r'^[a-zA-Z_-]+$', key) | |
86e5f3ed | 284 | expr = xpath + ('[@%s]' % key if val is None else f"[@{key}='{val}']") |
cfb0511d | 285 | return node.find(expr) |
59ae56fa | 286 | |
d7e66d39 JMF |
287 | # On python2.6 the xml.etree.ElementTree.Element methods don't support |
288 | # the namespace parameter | |
5f6a1245 JW |
289 | |
290 | ||
d7e66d39 JMF |
291 | def xpath_with_ns(path, ns_map): |
292 | components = [c.split(':') for c in path.split('/')] | |
293 | replaced = [] | |
294 | for c in components: | |
295 | if len(c) == 1: | |
296 | replaced.append(c[0]) | |
297 | else: | |
298 | ns, tag = c | |
299 | replaced.append('{%s}%s' % (ns_map[ns], tag)) | |
300 | return '/'.join(replaced) | |
301 | ||
d77c3dfd | 302 | |
a41fb80c | 303 | def xpath_element(node, xpath, name=None, fatal=False, default=NO_DEFAULT): |
578c0745 | 304 | def _find_xpath(xpath): |
f9934b96 | 305 | return node.find(xpath) |
578c0745 | 306 | |
14f25df2 | 307 | if isinstance(xpath, str): |
578c0745 S |
308 | n = _find_xpath(xpath) |
309 | else: | |
310 | for xp in xpath: | |
311 | n = _find_xpath(xp) | |
312 | if n is not None: | |
313 | break | |
d74bebd5 | 314 | |
8e636da4 | 315 | if n is None: |
bf42a990 S |
316 | if default is not NO_DEFAULT: |
317 | return default | |
318 | elif fatal: | |
bf0ff932 PH |
319 | name = xpath if name is None else name |
320 | raise ExtractorError('Could not find XML element %s' % name) | |
321 | else: | |
322 | return None | |
a41fb80c S |
323 | return n |
324 | ||
325 | ||
326 | def xpath_text(node, xpath, name=None, fatal=False, default=NO_DEFAULT): | |
8e636da4 S |
327 | n = xpath_element(node, xpath, name, fatal=fatal, default=default) |
328 | if n is None or n == default: | |
329 | return n | |
330 | if n.text is None: | |
331 | if default is not NO_DEFAULT: | |
332 | return default | |
333 | elif fatal: | |
334 | name = xpath if name is None else name | |
335 | raise ExtractorError('Could not find XML element\'s text %s' % name) | |
336 | else: | |
337 | return None | |
338 | return n.text | |
a41fb80c S |
339 | |
340 | ||
341 | def xpath_attr(node, xpath, key, name=None, fatal=False, default=NO_DEFAULT): | |
342 | n = find_xpath_attr(node, xpath, key) | |
343 | if n is None: | |
344 | if default is not NO_DEFAULT: | |
345 | return default | |
346 | elif fatal: | |
86e5f3ed | 347 | name = f'{xpath}[@{key}]' if name is None else name |
a41fb80c S |
348 | raise ExtractorError('Could not find XML attribute %s' % name) |
349 | else: | |
350 | return None | |
351 | return n.attrib[key] | |
bf0ff932 PH |
352 | |
353 | ||
c487cf00 | 354 | def get_element_by_id(id, html, **kwargs): |
43e8fafd | 355 | """Return the content of the tag with the specified ID in the passed HTML document""" |
c487cf00 | 356 | return get_element_by_attribute('id', id, html, **kwargs) |
43e8fafd | 357 | |
12ea2f30 | 358 | |
c487cf00 | 359 | def get_element_html_by_id(id, html, **kwargs): |
6f32a0b5 | 360 | """Return the html of the tag with the specified ID in the passed HTML document""" |
c487cf00 | 361 | return get_element_html_by_attribute('id', id, html, **kwargs) |
6f32a0b5 ZM |
362 | |
363 | ||
84c237fb | 364 | def get_element_by_class(class_name, html): |
2af12ad9 TC |
365 | """Return the content of the first tag with the specified class in the passed HTML document""" |
366 | retval = get_elements_by_class(class_name, html) | |
367 | return retval[0] if retval else None | |
368 | ||
369 | ||
6f32a0b5 ZM |
370 | def get_element_html_by_class(class_name, html): |
371 | """Return the html of the first tag with the specified class in the passed HTML document""" | |
372 | retval = get_elements_html_by_class(class_name, html) | |
373 | return retval[0] if retval else None | |
374 | ||
375 | ||
c487cf00 | 376 | def get_element_by_attribute(attribute, value, html, **kwargs): |
377 | retval = get_elements_by_attribute(attribute, value, html, **kwargs) | |
2af12ad9 TC |
378 | return retval[0] if retval else None |
379 | ||
380 | ||
c487cf00 | 381 | def get_element_html_by_attribute(attribute, value, html, **kargs): |
382 | retval = get_elements_html_by_attribute(attribute, value, html, **kargs) | |
6f32a0b5 ZM |
383 | return retval[0] if retval else None |
384 | ||
385 | ||
c487cf00 | 386 | def get_elements_by_class(class_name, html, **kargs): |
2af12ad9 TC |
387 | """Return the content of all tags with the specified class in the passed HTML document as a list""" |
388 | return get_elements_by_attribute( | |
64fa820c | 389 | 'class', r'[^\'"]*(?<=[\'"\s])%s(?=[\'"\s])[^\'"]*' % re.escape(class_name), |
84c237fb YCH |
390 | html, escape_value=False) |
391 | ||
392 | ||
6f32a0b5 ZM |
393 | def get_elements_html_by_class(class_name, html): |
394 | """Return the html of all tags with the specified class in the passed HTML document as a list""" | |
395 | return get_elements_html_by_attribute( | |
64fa820c | 396 | 'class', r'[^\'"]*(?<=[\'"\s])%s(?=[\'"\s])[^\'"]*' % re.escape(class_name), |
6f32a0b5 ZM |
397 | html, escape_value=False) |
398 | ||
399 | ||
400 | def get_elements_by_attribute(*args, **kwargs): | |
43e8fafd | 401 | """Return the content of the tag with the specified attribute in the passed HTML document""" |
6f32a0b5 ZM |
402 | return [content for content, _ in get_elements_text_and_html_by_attribute(*args, **kwargs)] |
403 | ||
404 | ||
405 | def get_elements_html_by_attribute(*args, **kwargs): | |
406 | """Return the html of the tag with the specified attribute in the passed HTML document""" | |
407 | return [whole for _, whole in get_elements_text_and_html_by_attribute(*args, **kwargs)] | |
408 | ||
409 | ||
410 | def get_elements_text_and_html_by_attribute(attribute, value, html, escape_value=True): | |
411 | """ | |
412 | Return the text (content) and the html (whole) of the tag with the specified | |
413 | attribute in the passed HTML document | |
414 | """ | |
9e6dd238 | 415 | |
86e5f3ed | 416 | quote = '' if re.match(r'''[\s"'`=<>]''', value) else '?' |
0254f162 | 417 | |
84c237fb YCH |
418 | value = re.escape(value) if escape_value else value |
419 | ||
86e5f3ed | 420 | partial_element_re = rf'''(?x) |
6f32a0b5 | 421 | <(?P<tag>[a-zA-Z0-9:._-]+) |
0254f162 | 422 | (?:\s(?:[^>"']|"[^"]*"|'[^']*')*)? |
86e5f3ed | 423 | \s{re.escape(attribute)}\s*=\s*(?P<_q>['"]{quote})(?-x:{value})(?P=_q) |
424 | ''' | |
38285056 | 425 | |
0254f162 ZM |
426 | for m in re.finditer(partial_element_re, html): |
427 | content, whole = get_element_text_and_html_by_tag(m.group('tag'), html[m.start():]) | |
a921f407 | 428 | |
0254f162 ZM |
429 | yield ( |
430 | unescapeHTML(re.sub(r'^(?P<q>["\'])(?P<content>.*)(?P=q)$', r'\g<content>', content, flags=re.DOTALL)), | |
431 | whole | |
432 | ) | |
a921f407 | 433 | |
c5229f39 | 434 | |
ac668111 | 435 | class HTMLBreakOnClosingTagParser(html.parser.HTMLParser): |
6f32a0b5 ZM |
436 | """ |
437 | HTML parser which raises HTMLBreakOnClosingTagException upon reaching the | |
438 | closing tag for the first opening tag it has encountered, and can be used | |
439 | as a context manager | |
440 | """ | |
441 | ||
442 | class HTMLBreakOnClosingTagException(Exception): | |
443 | pass | |
444 | ||
445 | def __init__(self): | |
446 | self.tagstack = collections.deque() | |
ac668111 | 447 | html.parser.HTMLParser.__init__(self) |
6f32a0b5 ZM |
448 | |
449 | def __enter__(self): | |
450 | return self | |
451 | ||
452 | def __exit__(self, *_): | |
453 | self.close() | |
454 | ||
455 | def close(self): | |
456 | # handle_endtag does not return upon raising HTMLBreakOnClosingTagException, | |
457 | # so data remains buffered; we no longer have any interest in it, thus | |
458 | # override this method to discard it | |
459 | pass | |
460 | ||
461 | def handle_starttag(self, tag, _): | |
462 | self.tagstack.append(tag) | |
463 | ||
464 | def handle_endtag(self, tag): | |
465 | if not self.tagstack: | |
466 | raise compat_HTMLParseError('no tags in the stack') | |
467 | while self.tagstack: | |
468 | inner_tag = self.tagstack.pop() | |
469 | if inner_tag == tag: | |
470 | break | |
471 | else: | |
472 | raise compat_HTMLParseError(f'matching opening tag for closing {tag} tag not found') | |
473 | if not self.tagstack: | |
474 | raise self.HTMLBreakOnClosingTagException() | |
475 | ||
476 | ||
477 | def get_element_text_and_html_by_tag(tag, html): | |
478 | """ | |
479 | For the first element with the specified tag in the passed HTML document | |
480 | return its' content (text) and the whole element (html) | |
481 | """ | |
482 | def find_or_raise(haystack, needle, exc): | |
483 | try: | |
484 | return haystack.index(needle) | |
485 | except ValueError: | |
486 | raise exc | |
487 | closing_tag = f'</{tag}>' | |
488 | whole_start = find_or_raise( | |
489 | html, f'<{tag}', compat_HTMLParseError(f'opening {tag} tag not found')) | |
490 | content_start = find_or_raise( | |
491 | html[whole_start:], '>', compat_HTMLParseError(f'malformed opening {tag} tag')) | |
492 | content_start += whole_start + 1 | |
493 | with HTMLBreakOnClosingTagParser() as parser: | |
494 | parser.feed(html[whole_start:content_start]) | |
495 | if not parser.tagstack or parser.tagstack[0] != tag: | |
496 | raise compat_HTMLParseError(f'parser did not match opening {tag} tag') | |
497 | offset = content_start | |
498 | while offset < len(html): | |
499 | next_closing_tag_start = find_or_raise( | |
500 | html[offset:], closing_tag, | |
501 | compat_HTMLParseError(f'closing {tag} tag not found')) | |
502 | next_closing_tag_end = next_closing_tag_start + len(closing_tag) | |
503 | try: | |
504 | parser.feed(html[offset:offset + next_closing_tag_end]) | |
505 | offset += next_closing_tag_end | |
506 | except HTMLBreakOnClosingTagParser.HTMLBreakOnClosingTagException: | |
507 | return html[content_start:offset + next_closing_tag_start], \ | |
508 | html[whole_start:offset + next_closing_tag_end] | |
509 | raise compat_HTMLParseError('unexpected end of html') | |
510 | ||
511 | ||
ac668111 | 512 | class HTMLAttributeParser(html.parser.HTMLParser): |
8bb56eee | 513 | """Trivial HTML parser to gather the attributes for a single element""" |
b6e0c7d2 | 514 | |
8bb56eee | 515 | def __init__(self): |
c5229f39 | 516 | self.attrs = {} |
ac668111 | 517 | html.parser.HTMLParser.__init__(self) |
8bb56eee BF |
518 | |
519 | def handle_starttag(self, tag, attrs): | |
520 | self.attrs = dict(attrs) | |
521 | ||
c5229f39 | 522 | |
ac668111 | 523 | class HTMLListAttrsParser(html.parser.HTMLParser): |
73673ccf FF |
524 | """HTML parser to gather the attributes for the elements of a list""" |
525 | ||
526 | def __init__(self): | |
ac668111 | 527 | html.parser.HTMLParser.__init__(self) |
73673ccf FF |
528 | self.items = [] |
529 | self._level = 0 | |
530 | ||
531 | def handle_starttag(self, tag, attrs): | |
532 | if tag == 'li' and self._level == 0: | |
533 | self.items.append(dict(attrs)) | |
534 | self._level += 1 | |
535 | ||
536 | def handle_endtag(self, tag): | |
537 | self._level -= 1 | |
538 | ||
539 | ||
8bb56eee BF |
540 | def extract_attributes(html_element): |
541 | """Given a string for an HTML element such as | |
542 | <el | |
543 | a="foo" B="bar" c="&98;az" d=boz | |
544 | empty= noval entity="&" | |
545 | sq='"' dq="'" | |
546 | > | |
547 | Decode and return a dictionary of attributes. | |
548 | { | |
549 | 'a': 'foo', 'b': 'bar', c: 'baz', d: 'boz', | |
550 | 'empty': '', 'noval': None, 'entity': '&', | |
551 | 'sq': '"', 'dq': '\'' | |
552 | }. | |
8bb56eee BF |
553 | """ |
554 | parser = HTMLAttributeParser() | |
19a03940 | 555 | with contextlib.suppress(compat_HTMLParseError): |
b4a3d461 S |
556 | parser.feed(html_element) |
557 | parser.close() | |
8bb56eee | 558 | return parser.attrs |
9e6dd238 | 559 | |
c5229f39 | 560 | |
73673ccf FF |
561 | def parse_list(webpage): |
562 | """Given a string for an series of HTML <li> elements, | |
563 | return a dictionary of their attributes""" | |
564 | parser = HTMLListAttrsParser() | |
565 | parser.feed(webpage) | |
566 | parser.close() | |
567 | return parser.items | |
568 | ||
569 | ||
9e6dd238 | 570 | def clean_html(html): |
59ae15a5 | 571 | """Clean an HTML snippet into a readable string""" |
dd622d7c PH |
572 | |
573 | if html is None: # Convenience for sanitizing descriptions etc. | |
574 | return html | |
575 | ||
49185227 | 576 | html = re.sub(r'\s+', ' ', html) |
577 | html = re.sub(r'(?u)\s?<\s?br\s?/?\s?>\s?', '\n', html) | |
578 | html = re.sub(r'(?u)<\s?/\s?p\s?>\s?<\s?p[^>]*>', '\n', html) | |
59ae15a5 PH |
579 | # Strip html tags |
580 | html = re.sub('<.*?>', '', html) | |
581 | # Replace html entities | |
582 | html = unescapeHTML(html) | |
7decf895 | 583 | return html.strip() |
9e6dd238 FV |
584 | |
585 | ||
b7c47b74 | 586 | class LenientJSONDecoder(json.JSONDecoder): |
587 | def __init__(self, *args, transform_source=None, ignore_extra=False, **kwargs): | |
588 | self.transform_source, self.ignore_extra = transform_source, ignore_extra | |
589 | super().__init__(*args, **kwargs) | |
590 | ||
591 | def decode(self, s): | |
592 | if self.transform_source: | |
593 | s = self.transform_source(s) | |
594 | if self.ignore_extra: | |
595 | return self.raw_decode(s.lstrip())[0] | |
596 | return super().decode(s) | |
597 | ||
598 | ||
d77c3dfd | 599 | def sanitize_open(filename, open_mode): |
59ae15a5 PH |
600 | """Try to open the given filename, and slightly tweak it if this fails. |
601 | ||
602 | Attempts to open the given filename. If this fails, it tries to change | |
603 | the filename slightly, step by step, until it's either able to open it | |
604 | or it fails and raises a final exception, like the standard open() | |
605 | function. | |
606 | ||
607 | It returns the tuple (stream, definitive_file_name). | |
608 | """ | |
0edb3e33 | 609 | if filename == '-': |
610 | if sys.platform == 'win32': | |
611 | import msvcrt | |
be5c1ae8 | 612 | |
62b58c09 | 613 | # stdout may be any IO stream, e.g. when using contextlib.redirect_stdout |
daef7911 | 614 | with contextlib.suppress(io.UnsupportedOperation): |
615 | msvcrt.setmode(sys.stdout.fileno(), os.O_BINARY) | |
0edb3e33 | 616 | return (sys.stdout.buffer if hasattr(sys.stdout, 'buffer') else sys.stdout, filename) |
59ae15a5 | 617 | |
0edb3e33 | 618 | for attempt in range(2): |
619 | try: | |
620 | try: | |
89737671 | 621 | if sys.platform == 'win32': |
b506289f | 622 | # FIXME: An exclusive lock also locks the file from being read. |
623 | # Since windows locks are mandatory, don't lock the file on windows (for now). | |
624 | # Ref: https://github.com/yt-dlp/yt-dlp/issues/3124 | |
89737671 | 625 | raise LockingUnsupportedError() |
0edb3e33 | 626 | stream = locked_file(filename, open_mode, block=False).__enter__() |
8a82af35 | 627 | except OSError: |
0edb3e33 | 628 | stream = open(filename, open_mode) |
8a82af35 | 629 | return stream, filename |
86e5f3ed | 630 | except OSError as err: |
0edb3e33 | 631 | if attempt or err.errno in (errno.EACCES,): |
632 | raise | |
633 | old_filename, filename = filename, sanitize_path(filename) | |
634 | if old_filename == filename: | |
635 | raise | |
d77c3dfd FV |
636 | |
637 | ||
638 | def timeconvert(timestr): | |
59ae15a5 PH |
639 | """Convert RFC 2822 defined time string into system timestamp""" |
640 | timestamp = None | |
641 | timetuple = email.utils.parsedate_tz(timestr) | |
642 | if timetuple is not None: | |
643 | timestamp = email.utils.mktime_tz(timetuple) | |
644 | return timestamp | |
1c469a94 | 645 | |
5f6a1245 | 646 | |
5c3895ff | 647 | def sanitize_filename(s, restricted=False, is_id=NO_DEFAULT): |
59ae15a5 | 648 | """Sanitizes a string so it could be used as part of a filename. |
5c3895ff | 649 | @param restricted Use a stricter subset of allowed characters |
650 | @param is_id Whether this is an ID that should be kept unchanged if possible. | |
651 | If unset, yt-dlp's new sanitization rules are in effect | |
59ae15a5 | 652 | """ |
5c3895ff | 653 | if s == '': |
654 | return '' | |
655 | ||
59ae15a5 | 656 | def replace_insane(char): |
c587cbb7 AT |
657 | if restricted and char in ACCENT_CHARS: |
658 | return ACCENT_CHARS[char] | |
91dd88b9 | 659 | elif not restricted and char == '\n': |
5c3895ff | 660 | return '\0 ' |
989a01c2 | 661 | elif is_id is NO_DEFAULT and not restricted and char in '"*:<>?|/\\': |
662 | # Replace with their full-width unicode counterparts | |
663 | return {'/': '\u29F8', '\\': '\u29f9'}.get(char, chr(ord(char) + 0xfee0)) | |
91dd88b9 | 664 | elif char == '?' or ord(char) < 32 or ord(char) == 127: |
59ae15a5 PH |
665 | return '' |
666 | elif char == '"': | |
667 | return '' if restricted else '\'' | |
668 | elif char == ':': | |
5c3895ff | 669 | return '\0_\0-' if restricted else '\0 \0-' |
59ae15a5 | 670 | elif char in '\\/|*<>': |
5c3895ff | 671 | return '\0_' |
672 | if restricted and (char in '!&\'()[]{}$;`^,#' or char.isspace() or ord(char) > 127): | |
673 | return '\0_' | |
59ae15a5 PH |
674 | return char |
675 | ||
989a01c2 | 676 | if restricted and is_id is NO_DEFAULT: |
677 | s = unicodedata.normalize('NFKC', s) | |
5c3895ff | 678 | s = re.sub(r'[0-9]+(?::[0-9]+)+', lambda m: m.group(0).replace(':', '_'), s) # Handle timestamps |
28e614de | 679 | result = ''.join(map(replace_insane, s)) |
5c3895ff | 680 | if is_id is NO_DEFAULT: |
ae61d108 | 681 | result = re.sub(r'(\0.)(?:(?=\1)..)+', r'\1', result) # Remove repeated substitute chars |
682 | STRIP_RE = r'(?:\0.|[ _-])*' | |
5c3895ff | 683 | result = re.sub(f'^\0.{STRIP_RE}|{STRIP_RE}\0.$', '', result) # Remove substitute chars from start/end |
684 | result = result.replace('\0', '') or '_' | |
685 | ||
796173d0 PH |
686 | if not is_id: |
687 | while '__' in result: | |
688 | result = result.replace('__', '_') | |
689 | result = result.strip('_') | |
690 | # Common case of "Foreign band name - English song title" | |
691 | if restricted and result.startswith('-_'): | |
692 | result = result[2:] | |
5a42414b PH |
693 | if result.startswith('-'): |
694 | result = '_' + result[len('-'):] | |
a7440261 | 695 | result = result.lstrip('.') |
796173d0 PH |
696 | if not result: |
697 | result = '_' | |
59ae15a5 | 698 | return result |
d77c3dfd | 699 | |
5f6a1245 | 700 | |
c2934512 | 701 | def sanitize_path(s, force=False): |
a2aaf4db | 702 | """Sanitizes and normalizes path on Windows""" |
c2934512 | 703 | if sys.platform == 'win32': |
c4218ac3 | 704 | force = False |
c2934512 | 705 | drive_or_unc, _ = os.path.splitdrive(s) |
c2934512 | 706 | elif force: |
707 | drive_or_unc = '' | |
708 | else: | |
a2aaf4db | 709 | return s |
c2934512 | 710 | |
be531ef1 S |
711 | norm_path = os.path.normpath(remove_start(s, drive_or_unc)).split(os.path.sep) |
712 | if drive_or_unc: | |
a2aaf4db S |
713 | norm_path.pop(0) |
714 | sanitized_path = [ | |
ec85ded8 | 715 | path_part if path_part in ['.', '..'] else re.sub(r'(?:[/<>:"\|\\?\*]|[\s.]$)', '#', path_part) |
a2aaf4db | 716 | for path_part in norm_path] |
be531ef1 S |
717 | if drive_or_unc: |
718 | sanitized_path.insert(0, drive_or_unc + os.path.sep) | |
4abea8ca | 719 | elif force and s and s[0] == os.path.sep: |
c4218ac3 | 720 | sanitized_path.insert(0, os.path.sep) |
a2aaf4db S |
721 | return os.path.join(*sanitized_path) |
722 | ||
723 | ||
8f97a15d | 724 | def sanitize_url(url, *, scheme='http'): |
befa4708 S |
725 | # Prepend protocol-less URLs with `http:` scheme in order to mitigate |
726 | # the number of unwanted failures due to missing protocol | |
21633673 | 727 | if url is None: |
728 | return | |
729 | elif url.startswith('//'): | |
8f97a15d | 730 | return f'{scheme}:{url}' |
befa4708 S |
731 | # Fix some common typos seen so far |
732 | COMMON_TYPOS = ( | |
067aa17e | 733 | # https://github.com/ytdl-org/youtube-dl/issues/15649 |
befa4708 S |
734 | (r'^httpss://', r'https://'), |
735 | # https://bx1.be/lives/direct-tv/ | |
736 | (r'^rmtp([es]?)://', r'rtmp\1://'), | |
737 | ) | |
738 | for mistake, fixup in COMMON_TYPOS: | |
739 | if re.match(mistake, url): | |
740 | return re.sub(mistake, fixup, url) | |
bc6b9bcd | 741 | return url |
17bcc626 S |
742 | |
743 | ||
5435dcf9 | 744 | def extract_basic_auth(url): |
14f25df2 | 745 | parts = urllib.parse.urlsplit(url) |
5435dcf9 HH |
746 | if parts.username is None: |
747 | return url, None | |
14f25df2 | 748 | url = urllib.parse.urlunsplit(parts._replace(netloc=( |
5435dcf9 HH |
749 | parts.hostname if parts.port is None |
750 | else '%s:%d' % (parts.hostname, parts.port)))) | |
751 | auth_payload = base64.b64encode( | |
0f06bcd7 | 752 | ('%s:%s' % (parts.username, parts.password or '')).encode()) |
753 | return url, f'Basic {auth_payload.decode()}' | |
5435dcf9 HH |
754 | |
755 | ||
67dda517 | 756 | def sanitized_Request(url, *args, **kwargs): |
bc6b9bcd | 757 | url, auth_header = extract_basic_auth(escape_url(sanitize_url(url))) |
5435dcf9 HH |
758 | if auth_header is not None: |
759 | headers = args[1] if len(args) >= 2 else kwargs.setdefault('headers', {}) | |
760 | headers['Authorization'] = auth_header | |
ac668111 | 761 | return urllib.request.Request(url, *args, **kwargs) |
67dda517 S |
762 | |
763 | ||
51098426 S |
764 | def expand_path(s): |
765 | """Expand shell variables and ~""" | |
766 | return os.path.expandvars(compat_expanduser(s)) | |
767 | ||
768 | ||
7e9a6125 | 769 | def orderedSet(iterable, *, lazy=False): |
770 | """Remove all duplicates from the input iterable""" | |
771 | def _iter(): | |
772 | seen = [] # Do not use set since the items can be unhashable | |
773 | for x in iterable: | |
774 | if x not in seen: | |
775 | seen.append(x) | |
776 | yield x | |
777 | ||
778 | return _iter() if lazy else list(_iter()) | |
d77c3dfd | 779 | |
912b38b4 | 780 | |
55b2f099 | 781 | def _htmlentity_transform(entity_with_semicolon): |
4e408e47 | 782 | """Transforms an HTML entity to a character.""" |
55b2f099 YCH |
783 | entity = entity_with_semicolon[:-1] |
784 | ||
4e408e47 | 785 | # Known non-numeric HTML entity |
ac668111 | 786 | if entity in html.entities.name2codepoint: |
787 | return chr(html.entities.name2codepoint[entity]) | |
4e408e47 | 788 | |
62b58c09 L |
789 | # TODO: HTML5 allows entities without a semicolon. |
790 | # E.g. 'Éric' should be decoded as 'Éric'. | |
ac668111 | 791 | if entity_with_semicolon in html.entities.html5: |
792 | return html.entities.html5[entity_with_semicolon] | |
55b2f099 | 793 | |
91757b0f | 794 | mobj = re.match(r'#(x[0-9a-fA-F]+|[0-9]+)', entity) |
4e408e47 PH |
795 | if mobj is not None: |
796 | numstr = mobj.group(1) | |
28e614de | 797 | if numstr.startswith('x'): |
4e408e47 | 798 | base = 16 |
28e614de | 799 | numstr = '0%s' % numstr |
4e408e47 PH |
800 | else: |
801 | base = 10 | |
067aa17e | 802 | # See https://github.com/ytdl-org/youtube-dl/issues/7518 |
19a03940 | 803 | with contextlib.suppress(ValueError): |
ac668111 | 804 | return chr(int(numstr, base)) |
4e408e47 PH |
805 | |
806 | # Unknown entity in name, return its literal representation | |
7a3f0c00 | 807 | return '&%s;' % entity |
4e408e47 PH |
808 | |
809 | ||
d77c3dfd | 810 | def unescapeHTML(s): |
912b38b4 PH |
811 | if s is None: |
812 | return None | |
19a03940 | 813 | assert isinstance(s, str) |
d77c3dfd | 814 | |
4e408e47 | 815 | return re.sub( |
95f3f7c2 | 816 | r'&([^&;]+;)', lambda m: _htmlentity_transform(m.group(1)), s) |
d77c3dfd | 817 | |
8bf48f23 | 818 | |
cdb19aa4 | 819 | def escapeHTML(text): |
820 | return ( | |
821 | text | |
822 | .replace('&', '&') | |
823 | .replace('<', '<') | |
824 | .replace('>', '>') | |
825 | .replace('"', '"') | |
826 | .replace("'", ''') | |
827 | ) | |
828 | ||
829 | ||
f5b1bca9 | 830 | def process_communicate_or_kill(p, *args, **kwargs): |
8a82af35 | 831 | write_string('DeprecationWarning: yt_dlp.utils.process_communicate_or_kill is deprecated ' |
832 | 'and may be removed in a future version. Use yt_dlp.utils.Popen.communicate_or_kill instead') | |
833 | return Popen.communicate_or_kill(p, *args, **kwargs) | |
f5b1bca9 | 834 | |
835 | ||
d3c93ec2 | 836 | class Popen(subprocess.Popen): |
837 | if sys.platform == 'win32': | |
838 | _startupinfo = subprocess.STARTUPINFO() | |
839 | _startupinfo.dwFlags |= subprocess.STARTF_USESHOWWINDOW | |
840 | else: | |
841 | _startupinfo = None | |
842 | ||
f0c9fb96 | 843 | def __init__(self, *args, text=False, **kwargs): |
844 | if text is True: | |
845 | kwargs['universal_newlines'] = True # For 3.6 compatibility | |
846 | kwargs.setdefault('encoding', 'utf-8') | |
847 | kwargs.setdefault('errors', 'replace') | |
86e5f3ed | 848 | super().__init__(*args, **kwargs, startupinfo=self._startupinfo) |
d3c93ec2 | 849 | |
850 | def communicate_or_kill(self, *args, **kwargs): | |
8a82af35 | 851 | try: |
852 | return self.communicate(*args, **kwargs) | |
853 | except BaseException: # Including KeyboardInterrupt | |
f0c9fb96 | 854 | self.kill(timeout=None) |
8a82af35 | 855 | raise |
d3c93ec2 | 856 | |
f0c9fb96 | 857 | def kill(self, *, timeout=0): |
858 | super().kill() | |
859 | if timeout != 0: | |
860 | self.wait(timeout=timeout) | |
861 | ||
862 | @classmethod | |
863 | def run(cls, *args, **kwargs): | |
864 | with cls(*args, **kwargs) as proc: | |
865 | stdout, stderr = proc.communicate_or_kill() | |
866 | return stdout or '', stderr or '', proc.returncode | |
867 | ||
d3c93ec2 | 868 | |
aa49acd1 S |
869 | def get_subprocess_encoding(): |
870 | if sys.platform == 'win32' and sys.getwindowsversion()[0] >= 5: | |
871 | # For subprocess calls, encode with locale encoding | |
872 | # Refer to http://stackoverflow.com/a/9951851/35070 | |
873 | encoding = preferredencoding() | |
874 | else: | |
875 | encoding = sys.getfilesystemencoding() | |
876 | if encoding is None: | |
877 | encoding = 'utf-8' | |
878 | return encoding | |
879 | ||
880 | ||
8bf48f23 | 881 | def encodeFilename(s, for_subprocess=False): |
19a03940 | 882 | assert isinstance(s, str) |
cfb0511d | 883 | return s |
aa49acd1 S |
884 | |
885 | ||
886 | def decodeFilename(b, for_subprocess=False): | |
cfb0511d | 887 | return b |
8bf48f23 | 888 | |
f07b74fc PH |
889 | |
890 | def encodeArgument(s): | |
cfb0511d | 891 | # Legacy code that uses byte strings |
892 | # Uncomment the following line after fixing all post processors | |
14f25df2 | 893 | # assert isinstance(s, str), 'Internal error: %r should be of type %r, is %r' % (s, str, type(s)) |
cfb0511d | 894 | return s if isinstance(s, str) else s.decode('ascii') |
f07b74fc PH |
895 | |
896 | ||
aa49acd1 | 897 | def decodeArgument(b): |
cfb0511d | 898 | return b |
aa49acd1 S |
899 | |
900 | ||
8271226a PH |
901 | def decodeOption(optval): |
902 | if optval is None: | |
903 | return optval | |
904 | if isinstance(optval, bytes): | |
905 | optval = optval.decode(preferredencoding()) | |
906 | ||
14f25df2 | 907 | assert isinstance(optval, str) |
8271226a | 908 | return optval |
1c256f70 | 909 | |
5f6a1245 | 910 | |
aa7785f8 | 911 | _timetuple = collections.namedtuple('Time', ('hours', 'minutes', 'seconds', 'milliseconds')) |
912 | ||
913 | ||
914 | def timetuple_from_msec(msec): | |
915 | secs, msec = divmod(msec, 1000) | |
916 | mins, secs = divmod(secs, 60) | |
917 | hrs, mins = divmod(mins, 60) | |
918 | return _timetuple(hrs, mins, secs, msec) | |
919 | ||
920 | ||
cdb19aa4 | 921 | def formatSeconds(secs, delim=':', msec=False): |
aa7785f8 | 922 | time = timetuple_from_msec(secs * 1000) |
923 | if time.hours: | |
924 | ret = '%d%s%02d%s%02d' % (time.hours, delim, time.minutes, delim, time.seconds) | |
925 | elif time.minutes: | |
926 | ret = '%d%s%02d' % (time.minutes, delim, time.seconds) | |
4539dd30 | 927 | else: |
aa7785f8 | 928 | ret = '%d' % time.seconds |
929 | return '%s.%03d' % (ret, time.milliseconds) if msec else ret | |
4539dd30 | 930 | |
a0ddb8a2 | 931 | |
77562778 | 932 | def _ssl_load_windows_store_certs(ssl_context, storename): |
933 | # Code adapted from _load_windows_store_certs in https://github.com/python/cpython/blob/main/Lib/ssl.py | |
934 | try: | |
935 | certs = [cert for cert, encoding, trust in ssl.enum_certificates(storename) | |
936 | if encoding == 'x509_asn' and ( | |
937 | trust is True or ssl.Purpose.SERVER_AUTH.oid in trust)] | |
938 | except PermissionError: | |
939 | return | |
940 | for cert in certs: | |
19a03940 | 941 | with contextlib.suppress(ssl.SSLError): |
77562778 | 942 | ssl_context.load_verify_locations(cadata=cert) |
a2366922 | 943 | |
77562778 | 944 | |
945 | def make_HTTPS_handler(params, **kwargs): | |
946 | opts_check_certificate = not params.get('nocheckcertificate') | |
947 | context = ssl.SSLContext(ssl.PROTOCOL_TLS_CLIENT) | |
948 | context.check_hostname = opts_check_certificate | |
f81c62a6 | 949 | if params.get('legacyserverconnect'): |
950 | context.options |= 4 # SSL_OP_LEGACY_SERVER_CONNECT | |
4f28b537 | 951 | # Allow use of weaker ciphers in Python 3.10+. See https://bugs.python.org/issue43998 |
952 | context.set_ciphers('DEFAULT') | |
8a82af35 | 953 | |
77562778 | 954 | context.verify_mode = ssl.CERT_REQUIRED if opts_check_certificate else ssl.CERT_NONE |
955 | if opts_check_certificate: | |
d5820461 | 956 | if has_certifi and 'no-certifi' not in params.get('compat_opts', []): |
957 | context.load_verify_locations(cafile=certifi.where()) | |
168bbc4f | 958 | else: |
959 | try: | |
960 | context.load_default_certs() | |
961 | # Work around the issue in load_default_certs when there are bad certificates. See: | |
962 | # https://github.com/yt-dlp/yt-dlp/issues/1060, | |
963 | # https://bugs.python.org/issue35665, https://bugs.python.org/issue45312 | |
964 | except ssl.SSLError: | |
965 | # enum_certificates is not present in mingw python. See https://github.com/yt-dlp/yt-dlp/issues/1151 | |
966 | if sys.platform == 'win32' and hasattr(ssl, 'enum_certificates'): | |
967 | for storename in ('CA', 'ROOT'): | |
968 | _ssl_load_windows_store_certs(context, storename) | |
969 | context.set_default_verify_paths() | |
8a82af35 | 970 | |
bb58c9ed | 971 | client_certfile = params.get('client_certificate') |
972 | if client_certfile: | |
973 | try: | |
974 | context.load_cert_chain( | |
975 | client_certfile, keyfile=params.get('client_certificate_key'), | |
976 | password=params.get('client_certificate_password')) | |
977 | except ssl.SSLError: | |
978 | raise YoutubeDLError('Unable to load client certificate') | |
2c6dcb65 | 979 | |
980 | # Some servers may reject requests if ALPN extension is not sent. See: | |
981 | # https://github.com/python/cpython/issues/85140 | |
982 | # https://github.com/yt-dlp/yt-dlp/issues/3878 | |
983 | with contextlib.suppress(NotImplementedError): | |
984 | context.set_alpn_protocols(['http/1.1']) | |
985 | ||
77562778 | 986 | return YoutubeDLHTTPSHandler(params, context=context, **kwargs) |
ea6d901e | 987 | |
732ea2f0 | 988 | |
5873d4cc | 989 | def bug_reports_message(before=';'): |
57e0f077 | 990 | from .update import REPOSITORY |
991 | ||
992 | msg = (f'please report this issue on https://github.com/{REPOSITORY}/issues?q= , ' | |
993 | 'filling out the appropriate issue template. Confirm you are on the latest version using yt-dlp -U') | |
5873d4cc F |
994 | |
995 | before = before.rstrip() | |
996 | if not before or before.endswith(('.', '!', '?')): | |
997 | msg = msg[0].title() + msg[1:] | |
998 | ||
999 | return (before + ' ' if before else '') + msg | |
08f2a92c JMF |
1000 | |
1001 | ||
bf5b9d85 PM |
1002 | class YoutubeDLError(Exception): |
1003 | """Base exception for YoutubeDL errors.""" | |
aa9369a2 | 1004 | msg = None |
1005 | ||
1006 | def __init__(self, msg=None): | |
1007 | if msg is not None: | |
1008 | self.msg = msg | |
1009 | elif self.msg is None: | |
1010 | self.msg = type(self).__name__ | |
1011 | super().__init__(self.msg) | |
bf5b9d85 PM |
1012 | |
1013 | ||
ac668111 | 1014 | network_exceptions = [urllib.error.URLError, http.client.HTTPException, socket.error] |
3158150c | 1015 | if hasattr(ssl, 'CertificateError'): |
1016 | network_exceptions.append(ssl.CertificateError) | |
1017 | network_exceptions = tuple(network_exceptions) | |
1018 | ||
1019 | ||
bf5b9d85 | 1020 | class ExtractorError(YoutubeDLError): |
1c256f70 | 1021 | """Error during info extraction.""" |
5f6a1245 | 1022 | |
1151c407 | 1023 | def __init__(self, msg, tb=None, expected=False, cause=None, video_id=None, ie=None): |
9a82b238 | 1024 | """ tb, if given, is the original traceback (so that it can be printed out). |
7a5c1cfe | 1025 | If expected is set, this is a normal error message and most likely not a bug in yt-dlp. |
9a82b238 | 1026 | """ |
3158150c | 1027 | if sys.exc_info()[0] in network_exceptions: |
9a82b238 | 1028 | expected = True |
d5979c5d | 1029 | |
7265a219 | 1030 | self.orig_msg = str(msg) |
1c256f70 | 1031 | self.traceback = tb |
1151c407 | 1032 | self.expected = expected |
2eabb802 | 1033 | self.cause = cause |
d11271dd | 1034 | self.video_id = video_id |
1151c407 | 1035 | self.ie = ie |
1036 | self.exc_info = sys.exc_info() # preserve original exception | |
5df14442 | 1037 | if isinstance(self.exc_info[1], ExtractorError): |
1038 | self.exc_info = self.exc_info[1].exc_info | |
1151c407 | 1039 | |
86e5f3ed | 1040 | super().__init__(''.join(( |
a70635b8 | 1041 | format_field(ie, None, '[%s] '), |
1042 | format_field(video_id, None, '%s: '), | |
7265a219 | 1043 | msg, |
a70635b8 | 1044 | format_field(cause, None, ' (caused by %r)'), |
1151c407 | 1045 | '' if expected else bug_reports_message()))) |
1c256f70 | 1046 | |
01951dda | 1047 | def format_traceback(self): |
497d2fab | 1048 | return join_nonempty( |
1049 | self.traceback and ''.join(traceback.format_tb(self.traceback)), | |
e491d06d | 1050 | self.cause and ''.join(traceback.format_exception(None, self.cause, self.cause.__traceback__)[1:]), |
497d2fab | 1051 | delim='\n') or None |
01951dda | 1052 | |
1c256f70 | 1053 | |
416c7fcb PH |
1054 | class UnsupportedError(ExtractorError): |
1055 | def __init__(self, url): | |
86e5f3ed | 1056 | super().__init__( |
416c7fcb PH |
1057 | 'Unsupported URL: %s' % url, expected=True) |
1058 | self.url = url | |
1059 | ||
1060 | ||
55b3e45b JMF |
1061 | class RegexNotFoundError(ExtractorError): |
1062 | """Error when a regex didn't match""" | |
1063 | pass | |
1064 | ||
1065 | ||
773f291d S |
1066 | class GeoRestrictedError(ExtractorError): |
1067 | """Geographic restriction Error exception. | |
1068 | ||
1069 | This exception may be thrown when a video is not available from your | |
1070 | geographic location due to geographic restrictions imposed by a website. | |
1071 | """ | |
b6e0c7d2 | 1072 | |
0db3bae8 | 1073 | def __init__(self, msg, countries=None, **kwargs): |
1074 | kwargs['expected'] = True | |
86e5f3ed | 1075 | super().__init__(msg, **kwargs) |
773f291d S |
1076 | self.countries = countries |
1077 | ||
1078 | ||
693f0600 | 1079 | class UserNotLive(ExtractorError): |
1080 | """Error when a channel/user is not live""" | |
1081 | ||
1082 | def __init__(self, msg=None, **kwargs): | |
1083 | kwargs['expected'] = True | |
1084 | super().__init__(msg or 'The channel is not currently live', **kwargs) | |
1085 | ||
1086 | ||
bf5b9d85 | 1087 | class DownloadError(YoutubeDLError): |
59ae15a5 | 1088 | """Download Error exception. |
d77c3dfd | 1089 | |
59ae15a5 PH |
1090 | This exception may be thrown by FileDownloader objects if they are not |
1091 | configured to continue on errors. They will contain the appropriate | |
1092 | error message. | |
1093 | """ | |
5f6a1245 | 1094 | |
8cc83b8d FV |
1095 | def __init__(self, msg, exc_info=None): |
1096 | """ exc_info, if given, is the original exception that caused the trouble (as returned by sys.exc_info()). """ | |
86e5f3ed | 1097 | super().__init__(msg) |
8cc83b8d | 1098 | self.exc_info = exc_info |
d77c3dfd FV |
1099 | |
1100 | ||
498f5606 | 1101 | class EntryNotInPlaylist(YoutubeDLError): |
1102 | """Entry not in playlist exception. | |
1103 | ||
1104 | This exception will be thrown by YoutubeDL when a requested entry | |
1105 | is not found in the playlist info_dict | |
1106 | """ | |
aa9369a2 | 1107 | msg = 'Entry not found in info' |
498f5606 | 1108 | |
1109 | ||
bf5b9d85 | 1110 | class SameFileError(YoutubeDLError): |
59ae15a5 | 1111 | """Same File exception. |
d77c3dfd | 1112 | |
59ae15a5 PH |
1113 | This exception will be thrown by FileDownloader objects if they detect |
1114 | multiple files would have to be downloaded to the same file on disk. | |
1115 | """ | |
aa9369a2 | 1116 | msg = 'Fixed output name but more than one file to download' |
1117 | ||
1118 | def __init__(self, filename=None): | |
1119 | if filename is not None: | |
1120 | self.msg += f': {filename}' | |
1121 | super().__init__(self.msg) | |
d77c3dfd FV |
1122 | |
1123 | ||
bf5b9d85 | 1124 | class PostProcessingError(YoutubeDLError): |
59ae15a5 | 1125 | """Post Processing exception. |
d77c3dfd | 1126 | |
59ae15a5 PH |
1127 | This exception may be raised by PostProcessor's .run() method to |
1128 | indicate an error in the postprocessing task. | |
1129 | """ | |
5f6a1245 | 1130 | |
5f6a1245 | 1131 | |
48f79687 | 1132 | class DownloadCancelled(YoutubeDLError): |
1133 | """ Exception raised when the download queue should be interrupted """ | |
1134 | msg = 'The download was cancelled' | |
8b0d7497 | 1135 | |
8b0d7497 | 1136 | |
48f79687 | 1137 | class ExistingVideoReached(DownloadCancelled): |
1138 | """ --break-on-existing triggered """ | |
1139 | msg = 'Encountered a video that is already in the archive, stopping due to --break-on-existing' | |
8b0d7497 | 1140 | |
48f79687 | 1141 | |
1142 | class RejectedVideoReached(DownloadCancelled): | |
1143 | """ --break-on-reject triggered """ | |
1144 | msg = 'Encountered a video that did not match filter, stopping due to --break-on-reject' | |
51d9739f | 1145 | |
1146 | ||
48f79687 | 1147 | class MaxDownloadsReached(DownloadCancelled): |
59ae15a5 | 1148 | """ --max-downloads limit has been reached. """ |
48f79687 | 1149 | msg = 'Maximum number of downloads reached, stopping due to --max-downloads' |
1150 | ||
1151 | ||
f2ebc5c7 | 1152 | class ReExtractInfo(YoutubeDLError): |
1153 | """ Video info needs to be re-extracted. """ | |
1154 | ||
1155 | def __init__(self, msg, expected=False): | |
1156 | super().__init__(msg) | |
1157 | self.expected = expected | |
1158 | ||
1159 | ||
1160 | class ThrottledDownload(ReExtractInfo): | |
48f79687 | 1161 | """ Download speed below --throttled-rate. """ |
aa9369a2 | 1162 | msg = 'The download speed is below throttle limit' |
d77c3dfd | 1163 | |
43b22906 | 1164 | def __init__(self): |
1165 | super().__init__(self.msg, expected=False) | |
f2ebc5c7 | 1166 | |
d77c3dfd | 1167 | |
bf5b9d85 | 1168 | class UnavailableVideoError(YoutubeDLError): |
59ae15a5 | 1169 | """Unavailable Format exception. |
d77c3dfd | 1170 | |
59ae15a5 PH |
1171 | This exception will be thrown when a video is requested |
1172 | in a format that is not available for that video. | |
1173 | """ | |
aa9369a2 | 1174 | msg = 'Unable to download video' |
1175 | ||
1176 | def __init__(self, err=None): | |
1177 | if err is not None: | |
1178 | self.msg += f': {err}' | |
1179 | super().__init__(self.msg) | |
d77c3dfd FV |
1180 | |
1181 | ||
bf5b9d85 | 1182 | class ContentTooShortError(YoutubeDLError): |
59ae15a5 | 1183 | """Content Too Short exception. |
d77c3dfd | 1184 | |
59ae15a5 PH |
1185 | This exception may be raised by FileDownloader objects when a file they |
1186 | download is too small for what the server announced first, indicating | |
1187 | the connection was probably interrupted. | |
1188 | """ | |
d77c3dfd | 1189 | |
59ae15a5 | 1190 | def __init__(self, downloaded, expected): |
86e5f3ed | 1191 | super().__init__(f'Downloaded {downloaded} bytes, expected {expected} bytes') |
2c7ed247 | 1192 | # Both in bytes |
59ae15a5 PH |
1193 | self.downloaded = downloaded |
1194 | self.expected = expected | |
d77c3dfd | 1195 | |
5f6a1245 | 1196 | |
bf5b9d85 | 1197 | class XAttrMetadataError(YoutubeDLError): |
efa97bdc | 1198 | def __init__(self, code=None, msg='Unknown error'): |
86e5f3ed | 1199 | super().__init__(msg) |
efa97bdc | 1200 | self.code = code |
bd264412 | 1201 | self.msg = msg |
efa97bdc YCH |
1202 | |
1203 | # Parsing code and msg | |
3089bc74 | 1204 | if (self.code in (errno.ENOSPC, errno.EDQUOT) |
a0566bbf | 1205 | or 'No space left' in self.msg or 'Disk quota exceeded' in self.msg): |
efa97bdc YCH |
1206 | self.reason = 'NO_SPACE' |
1207 | elif self.code == errno.E2BIG or 'Argument list too long' in self.msg: | |
1208 | self.reason = 'VALUE_TOO_LONG' | |
1209 | else: | |
1210 | self.reason = 'NOT_SUPPORTED' | |
1211 | ||
1212 | ||
bf5b9d85 | 1213 | class XAttrUnavailableError(YoutubeDLError): |
efa97bdc YCH |
1214 | pass |
1215 | ||
1216 | ||
c5a59d93 | 1217 | def _create_http_connection(ydl_handler, http_class, is_https, *args, **kwargs): |
f9934b96 | 1218 | hc = http_class(*args, **kwargs) |
be4a824d | 1219 | source_address = ydl_handler._params.get('source_address') |
8959018a | 1220 | |
be4a824d | 1221 | if source_address is not None: |
8959018a AU |
1222 | # This is to workaround _create_connection() from socket where it will try all |
1223 | # address data from getaddrinfo() including IPv6. This filters the result from | |
1224 | # getaddrinfo() based on the source_address value. | |
1225 | # This is based on the cpython socket.create_connection() function. | |
1226 | # https://github.com/python/cpython/blob/master/Lib/socket.py#L691 | |
1227 | def _create_connection(address, timeout=socket._GLOBAL_DEFAULT_TIMEOUT, source_address=None): | |
1228 | host, port = address | |
1229 | err = None | |
1230 | addrs = socket.getaddrinfo(host, port, 0, socket.SOCK_STREAM) | |
9e21e6d9 S |
1231 | af = socket.AF_INET if '.' in source_address[0] else socket.AF_INET6 |
1232 | ip_addrs = [addr for addr in addrs if addr[0] == af] | |
1233 | if addrs and not ip_addrs: | |
1234 | ip_version = 'v4' if af == socket.AF_INET else 'v6' | |
86e5f3ed | 1235 | raise OSError( |
9e21e6d9 S |
1236 | "No remote IP%s addresses available for connect, can't use '%s' as source address" |
1237 | % (ip_version, source_address[0])) | |
8959018a AU |
1238 | for res in ip_addrs: |
1239 | af, socktype, proto, canonname, sa = res | |
1240 | sock = None | |
1241 | try: | |
1242 | sock = socket.socket(af, socktype, proto) | |
1243 | if timeout is not socket._GLOBAL_DEFAULT_TIMEOUT: | |
1244 | sock.settimeout(timeout) | |
1245 | sock.bind(source_address) | |
1246 | sock.connect(sa) | |
1247 | err = None # Explicitly break reference cycle | |
1248 | return sock | |
86e5f3ed | 1249 | except OSError as _: |
8959018a AU |
1250 | err = _ |
1251 | if sock is not None: | |
1252 | sock.close() | |
1253 | if err is not None: | |
1254 | raise err | |
1255 | else: | |
86e5f3ed | 1256 | raise OSError('getaddrinfo returns an empty list') |
9e21e6d9 S |
1257 | if hasattr(hc, '_create_connection'): |
1258 | hc._create_connection = _create_connection | |
cfb0511d | 1259 | hc.source_address = (source_address, 0) |
be4a824d PH |
1260 | |
1261 | return hc | |
1262 | ||
1263 | ||
87f0e62d | 1264 | def handle_youtubedl_headers(headers): |
992fc9d6 YCH |
1265 | filtered_headers = headers |
1266 | ||
1267 | if 'Youtubedl-no-compression' in filtered_headers: | |
86e5f3ed | 1268 | filtered_headers = {k: v for k, v in filtered_headers.items() if k.lower() != 'accept-encoding'} |
87f0e62d | 1269 | del filtered_headers['Youtubedl-no-compression'] |
87f0e62d | 1270 | |
992fc9d6 | 1271 | return filtered_headers |
87f0e62d YCH |
1272 | |
1273 | ||
ac668111 | 1274 | class YoutubeDLHandler(urllib.request.HTTPHandler): |
59ae15a5 PH |
1275 | """Handler for HTTP requests and responses. |
1276 | ||
1277 | This class, when installed with an OpenerDirector, automatically adds | |
1278 | the standard headers to every HTTP request and handles gzipped and | |
1279 | deflated responses from web servers. If compression is to be avoided in | |
1280 | a particular request, the original request in the program code only has | |
0424ec30 | 1281 | to include the HTTP header "Youtubedl-no-compression", which will be |
59ae15a5 PH |
1282 | removed before making the real request. |
1283 | ||
1284 | Part of this code was copied from: | |
1285 | ||
1286 | http://techknack.net/python-urllib2-handlers/ | |
1287 | ||
1288 | Andrew Rowls, the author of that code, agreed to release it to the | |
1289 | public domain. | |
1290 | """ | |
1291 | ||
be4a824d | 1292 | def __init__(self, params, *args, **kwargs): |
ac668111 | 1293 | urllib.request.HTTPHandler.__init__(self, *args, **kwargs) |
be4a824d PH |
1294 | self._params = params |
1295 | ||
1296 | def http_open(self, req): | |
ac668111 | 1297 | conn_class = http.client.HTTPConnection |
71aff188 YCH |
1298 | |
1299 | socks_proxy = req.headers.get('Ytdl-socks-proxy') | |
1300 | if socks_proxy: | |
1301 | conn_class = make_socks_conn_class(conn_class, socks_proxy) | |
1302 | del req.headers['Ytdl-socks-proxy'] | |
1303 | ||
be4a824d | 1304 | return self.do_open(functools.partial( |
71aff188 | 1305 | _create_http_connection, self, conn_class, False), |
be4a824d PH |
1306 | req) |
1307 | ||
59ae15a5 PH |
1308 | @staticmethod |
1309 | def deflate(data): | |
fc2119f2 | 1310 | if not data: |
1311 | return data | |
59ae15a5 PH |
1312 | try: |
1313 | return zlib.decompress(data, -zlib.MAX_WBITS) | |
1314 | except zlib.error: | |
1315 | return zlib.decompress(data) | |
1316 | ||
4390d5ec | 1317 | @staticmethod |
1318 | def brotli(data): | |
1319 | if not data: | |
1320 | return data | |
9b8ee23b | 1321 | return brotli.decompress(data) |
4390d5ec | 1322 | |
acebc9cd | 1323 | def http_request(self, req): |
51f267d9 S |
1324 | # According to RFC 3986, URLs can not contain non-ASCII characters, however this is not |
1325 | # always respected by websites, some tend to give out URLs with non percent-encoded | |
1326 | # non-ASCII characters (see telemb.py, ard.py [#3412]) | |
1327 | # urllib chokes on URLs with non-ASCII characters (see http://bugs.python.org/issue3991) | |
1328 | # To work around aforementioned issue we will replace request's original URL with | |
1329 | # percent-encoded one | |
1330 | # Since redirects are also affected (e.g. http://www.southpark.de/alle-episoden/s18e09) | |
1331 | # the code of this workaround has been moved here from YoutubeDL.urlopen() | |
1332 | url = req.get_full_url() | |
1333 | url_escaped = escape_url(url) | |
1334 | ||
1335 | # Substitute URL if any change after escaping | |
1336 | if url != url_escaped: | |
15d260eb | 1337 | req = update_Request(req, url=url_escaped) |
51f267d9 | 1338 | |
8b7539d2 | 1339 | for h, v in self._params.get('http_headers', std_headers).items(): |
3d5f7a39 JK |
1340 | # Capitalize is needed because of Python bug 2275: http://bugs.python.org/issue2275 |
1341 | # The dict keys are capitalized because of this bug by urllib | |
1342 | if h.capitalize() not in req.headers: | |
33ac271b | 1343 | req.add_header(h, v) |
87f0e62d | 1344 | |
af14914b | 1345 | if 'Accept-encoding' not in req.headers: |
1346 | req.add_header('Accept-encoding', ', '.join(SUPPORTED_ENCODINGS)) | |
1347 | ||
87f0e62d | 1348 | req.headers = handle_youtubedl_headers(req.headers) |
989b4b2b | 1349 | |
379a4f16 | 1350 | return super().do_request_(req) |
59ae15a5 | 1351 | |
acebc9cd | 1352 | def http_response(self, req, resp): |
59ae15a5 PH |
1353 | old_resp = resp |
1354 | # gzip | |
1355 | if resp.headers.get('Content-encoding', '') == 'gzip': | |
aa3e9507 PH |
1356 | content = resp.read() |
1357 | gz = gzip.GzipFile(fileobj=io.BytesIO(content), mode='rb') | |
1358 | try: | |
1359 | uncompressed = io.BytesIO(gz.read()) | |
86e5f3ed | 1360 | except OSError as original_ioerror: |
aa3e9507 PH |
1361 | # There may be junk add the end of the file |
1362 | # See http://stackoverflow.com/q/4928560/35070 for details | |
1363 | for i in range(1, 1024): | |
1364 | try: | |
1365 | gz = gzip.GzipFile(fileobj=io.BytesIO(content[:-i]), mode='rb') | |
1366 | uncompressed = io.BytesIO(gz.read()) | |
86e5f3ed | 1367 | except OSError: |
aa3e9507 PH |
1368 | continue |
1369 | break | |
1370 | else: | |
1371 | raise original_ioerror | |
ac668111 | 1372 | resp = urllib.request.addinfourl(uncompressed, old_resp.headers, old_resp.url, old_resp.code) |
59ae15a5 | 1373 | resp.msg = old_resp.msg |
c047270c | 1374 | del resp.headers['Content-encoding'] |
59ae15a5 PH |
1375 | # deflate |
1376 | if resp.headers.get('Content-encoding', '') == 'deflate': | |
1377 | gz = io.BytesIO(self.deflate(resp.read())) | |
ac668111 | 1378 | resp = urllib.request.addinfourl(gz, old_resp.headers, old_resp.url, old_resp.code) |
59ae15a5 | 1379 | resp.msg = old_resp.msg |
c047270c | 1380 | del resp.headers['Content-encoding'] |
4390d5ec | 1381 | # brotli |
1382 | if resp.headers.get('Content-encoding', '') == 'br': | |
ac668111 | 1383 | resp = urllib.request.addinfourl( |
4390d5ec | 1384 | io.BytesIO(self.brotli(resp.read())), old_resp.headers, old_resp.url, old_resp.code) |
1385 | resp.msg = old_resp.msg | |
1386 | del resp.headers['Content-encoding'] | |
ad729172 | 1387 | # Percent-encode redirect URL of Location HTTP header to satisfy RFC 3986 (see |
067aa17e | 1388 | # https://github.com/ytdl-org/youtube-dl/issues/6457). |
5a4d9ddb S |
1389 | if 300 <= resp.code < 400: |
1390 | location = resp.headers.get('Location') | |
1391 | if location: | |
1392 | # As of RFC 2616 default charset is iso-8859-1 that is respected by python 3 | |
0f06bcd7 | 1393 | location = location.encode('iso-8859-1').decode() |
5a4d9ddb S |
1394 | location_escaped = escape_url(location) |
1395 | if location != location_escaped: | |
1396 | del resp.headers['Location'] | |
1397 | resp.headers['Location'] = location_escaped | |
59ae15a5 | 1398 | return resp |
0f8d03f8 | 1399 | |
acebc9cd PH |
1400 | https_request = http_request |
1401 | https_response = http_response | |
bf50b038 | 1402 | |
5de90176 | 1403 | |
71aff188 YCH |
1404 | def make_socks_conn_class(base_class, socks_proxy): |
1405 | assert issubclass(base_class, ( | |
ac668111 | 1406 | http.client.HTTPConnection, http.client.HTTPSConnection)) |
71aff188 | 1407 | |
14f25df2 | 1408 | url_components = urllib.parse.urlparse(socks_proxy) |
71aff188 YCH |
1409 | if url_components.scheme.lower() == 'socks5': |
1410 | socks_type = ProxyType.SOCKS5 | |
1411 | elif url_components.scheme.lower() in ('socks', 'socks4'): | |
1412 | socks_type = ProxyType.SOCKS4 | |
51fb4995 YCH |
1413 | elif url_components.scheme.lower() == 'socks4a': |
1414 | socks_type = ProxyType.SOCKS4A | |
71aff188 | 1415 | |
cdd94c2e YCH |
1416 | def unquote_if_non_empty(s): |
1417 | if not s: | |
1418 | return s | |
ac668111 | 1419 | return urllib.parse.unquote_plus(s) |
cdd94c2e | 1420 | |
71aff188 YCH |
1421 | proxy_args = ( |
1422 | socks_type, | |
1423 | url_components.hostname, url_components.port or 1080, | |
1424 | True, # Remote DNS | |
cdd94c2e YCH |
1425 | unquote_if_non_empty(url_components.username), |
1426 | unquote_if_non_empty(url_components.password), | |
71aff188 YCH |
1427 | ) |
1428 | ||
1429 | class SocksConnection(base_class): | |
1430 | def connect(self): | |
1431 | self.sock = sockssocket() | |
1432 | self.sock.setproxy(*proxy_args) | |
19a03940 | 1433 | if isinstance(self.timeout, (int, float)): |
71aff188 YCH |
1434 | self.sock.settimeout(self.timeout) |
1435 | self.sock.connect((self.host, self.port)) | |
1436 | ||
ac668111 | 1437 | if isinstance(self, http.client.HTTPSConnection): |
71aff188 YCH |
1438 | if hasattr(self, '_context'): # Python > 2.6 |
1439 | self.sock = self._context.wrap_socket( | |
1440 | self.sock, server_hostname=self.host) | |
1441 | else: | |
1442 | self.sock = ssl.wrap_socket(self.sock) | |
1443 | ||
1444 | return SocksConnection | |
1445 | ||
1446 | ||
ac668111 | 1447 | class YoutubeDLHTTPSHandler(urllib.request.HTTPSHandler): |
be4a824d | 1448 | def __init__(self, params, https_conn_class=None, *args, **kwargs): |
ac668111 | 1449 | urllib.request.HTTPSHandler.__init__(self, *args, **kwargs) |
1450 | self._https_conn_class = https_conn_class or http.client.HTTPSConnection | |
be4a824d PH |
1451 | self._params = params |
1452 | ||
1453 | def https_open(self, req): | |
4f264c02 | 1454 | kwargs = {} |
71aff188 YCH |
1455 | conn_class = self._https_conn_class |
1456 | ||
4f264c02 JMF |
1457 | if hasattr(self, '_context'): # python > 2.6 |
1458 | kwargs['context'] = self._context | |
1459 | if hasattr(self, '_check_hostname'): # python 3.x | |
1460 | kwargs['check_hostname'] = self._check_hostname | |
71aff188 YCH |
1461 | |
1462 | socks_proxy = req.headers.get('Ytdl-socks-proxy') | |
1463 | if socks_proxy: | |
1464 | conn_class = make_socks_conn_class(conn_class, socks_proxy) | |
1465 | del req.headers['Ytdl-socks-proxy'] | |
1466 | ||
4f28b537 | 1467 | try: |
1468 | return self.do_open( | |
1469 | functools.partial(_create_http_connection, self, conn_class, True), req, **kwargs) | |
1470 | except urllib.error.URLError as e: | |
1471 | if (isinstance(e.reason, ssl.SSLError) | |
1472 | and getattr(e.reason, 'reason', None) == 'SSLV3_ALERT_HANDSHAKE_FAILURE'): | |
1473 | raise YoutubeDLError('SSLV3_ALERT_HANDSHAKE_FAILURE: Try using --legacy-server-connect') | |
1474 | raise | |
be4a824d PH |
1475 | |
1476 | ||
ac668111 | 1477 | class YoutubeDLCookieJar(http.cookiejar.MozillaCookieJar): |
f1a8511f S |
1478 | """ |
1479 | See [1] for cookie file format. | |
1480 | ||
1481 | 1. https://curl.haxx.se/docs/http-cookies.html | |
1482 | """ | |
e7e62441 | 1483 | _HTTPONLY_PREFIX = '#HttpOnly_' |
c380cc28 S |
1484 | _ENTRY_LEN = 7 |
1485 | _HEADER = '''# Netscape HTTP Cookie File | |
7a5c1cfe | 1486 | # This file is generated by yt-dlp. Do not edit. |
c380cc28 S |
1487 | |
1488 | ''' | |
1489 | _CookieFileEntry = collections.namedtuple( | |
1490 | 'CookieFileEntry', | |
1491 | ('domain_name', 'include_subdomains', 'path', 'https_only', 'expires_at', 'name', 'value')) | |
e7e62441 | 1492 | |
d76fa1f3 | 1493 | def __init__(self, filename=None, *args, **kwargs): |
1494 | super().__init__(None, *args, **kwargs) | |
1495 | if self.is_path(filename): | |
1496 | filename = os.fspath(filename) | |
1497 | self.filename = filename | |
1498 | ||
24146491 | 1499 | @staticmethod |
1500 | def _true_or_false(cndn): | |
1501 | return 'TRUE' if cndn else 'FALSE' | |
1502 | ||
d76fa1f3 | 1503 | @staticmethod |
1504 | def is_path(file): | |
1505 | return isinstance(file, (str, bytes, os.PathLike)) | |
1506 | ||
1507 | @contextlib.contextmanager | |
1508 | def open(self, file, *, write=False): | |
1509 | if self.is_path(file): | |
1510 | with open(file, 'w' if write else 'r', encoding='utf-8') as f: | |
1511 | yield f | |
1512 | else: | |
1513 | if write: | |
1514 | file.truncate(0) | |
1515 | yield file | |
1516 | ||
24146491 | 1517 | def _really_save(self, f, ignore_discard=False, ignore_expires=False): |
1518 | now = time.time() | |
1519 | for cookie in self: | |
1520 | if (not ignore_discard and cookie.discard | |
1521 | or not ignore_expires and cookie.is_expired(now)): | |
1522 | continue | |
1523 | name, value = cookie.name, cookie.value | |
1524 | if value is None: | |
1525 | # cookies.txt regards 'Set-Cookie: foo' as a cookie | |
1526 | # with no name, whereas http.cookiejar regards it as a | |
1527 | # cookie with no value. | |
1528 | name, value = '', name | |
1529 | f.write('%s\n' % '\t'.join(( | |
1530 | cookie.domain, | |
1531 | self._true_or_false(cookie.domain.startswith('.')), | |
1532 | cookie.path, | |
1533 | self._true_or_false(cookie.secure), | |
1534 | str_or_none(cookie.expires, default=''), | |
1535 | name, value | |
1536 | ))) | |
1537 | ||
1538 | def save(self, filename=None, *args, **kwargs): | |
c380cc28 S |
1539 | """ |
1540 | Save cookies to a file. | |
24146491 | 1541 | Code is taken from CPython 3.6 |
1542 | https://github.com/python/cpython/blob/8d999cbf4adea053be6dbb612b9844635c4dfb8e/Lib/http/cookiejar.py#L2091-L2117 """ | |
c380cc28 | 1543 | |
c380cc28 S |
1544 | if filename is None: |
1545 | if self.filename is not None: | |
1546 | filename = self.filename | |
1547 | else: | |
ac668111 | 1548 | raise ValueError(http.cookiejar.MISSING_FILENAME_TEXT) |
c380cc28 | 1549 | |
24146491 | 1550 | # Store session cookies with `expires` set to 0 instead of an empty string |
1bab3437 S |
1551 | for cookie in self: |
1552 | if cookie.expires is None: | |
1553 | cookie.expires = 0 | |
c380cc28 | 1554 | |
d76fa1f3 | 1555 | with self.open(filename, write=True) as f: |
c380cc28 | 1556 | f.write(self._HEADER) |
24146491 | 1557 | self._really_save(f, *args, **kwargs) |
1bab3437 S |
1558 | |
1559 | def load(self, filename=None, ignore_discard=False, ignore_expires=False): | |
e7e62441 | 1560 | """Load cookies from a file.""" |
1561 | if filename is None: | |
1562 | if self.filename is not None: | |
1563 | filename = self.filename | |
1564 | else: | |
ac668111 | 1565 | raise ValueError(http.cookiejar.MISSING_FILENAME_TEXT) |
e7e62441 | 1566 | |
c380cc28 S |
1567 | def prepare_line(line): |
1568 | if line.startswith(self._HTTPONLY_PREFIX): | |
1569 | line = line[len(self._HTTPONLY_PREFIX):] | |
1570 | # comments and empty lines are fine | |
1571 | if line.startswith('#') or not line.strip(): | |
1572 | return line | |
1573 | cookie_list = line.split('\t') | |
1574 | if len(cookie_list) != self._ENTRY_LEN: | |
ac668111 | 1575 | raise http.cookiejar.LoadError('invalid length %d' % len(cookie_list)) |
c380cc28 S |
1576 | cookie = self._CookieFileEntry(*cookie_list) |
1577 | if cookie.expires_at and not cookie.expires_at.isdigit(): | |
ac668111 | 1578 | raise http.cookiejar.LoadError('invalid expires at %s' % cookie.expires_at) |
c380cc28 S |
1579 | return line |
1580 | ||
e7e62441 | 1581 | cf = io.StringIO() |
d76fa1f3 | 1582 | with self.open(filename) as f: |
e7e62441 | 1583 | for line in f: |
c380cc28 S |
1584 | try: |
1585 | cf.write(prepare_line(line)) | |
ac668111 | 1586 | except http.cookiejar.LoadError as e: |
94aa0644 | 1587 | if f'{line.strip()} '[0] in '[{"': |
ac668111 | 1588 | raise http.cookiejar.LoadError( |
94aa0644 L |
1589 | 'Cookies file must be Netscape formatted, not JSON. See ' |
1590 | 'https://github.com/ytdl-org/youtube-dl#how-do-i-pass-cookies-to-youtube-dl') | |
19a03940 | 1591 | write_string(f'WARNING: skipping cookie file entry due to {e}: {line!r}\n') |
c380cc28 | 1592 | continue |
e7e62441 | 1593 | cf.seek(0) |
1594 | self._really_load(cf, filename, ignore_discard, ignore_expires) | |
1bab3437 S |
1595 | # Session cookies are denoted by either `expires` field set to |
1596 | # an empty string or 0. MozillaCookieJar only recognizes the former | |
1597 | # (see [1]). So we need force the latter to be recognized as session | |
1598 | # cookies on our own. | |
1599 | # Session cookies may be important for cookies-based authentication, | |
1600 | # e.g. usually, when user does not check 'Remember me' check box while | |
1601 | # logging in on a site, some important cookies are stored as session | |
1602 | # cookies so that not recognizing them will result in failed login. | |
1603 | # 1. https://bugs.python.org/issue17164 | |
1604 | for cookie in self: | |
1605 | # Treat `expires=0` cookies as session cookies | |
1606 | if cookie.expires == 0: | |
1607 | cookie.expires = None | |
1608 | cookie.discard = True | |
1609 | ||
1610 | ||
ac668111 | 1611 | class YoutubeDLCookieProcessor(urllib.request.HTTPCookieProcessor): |
a6420bf5 | 1612 | def __init__(self, cookiejar=None): |
ac668111 | 1613 | urllib.request.HTTPCookieProcessor.__init__(self, cookiejar) |
a6420bf5 S |
1614 | |
1615 | def http_response(self, request, response): | |
ac668111 | 1616 | return urllib.request.HTTPCookieProcessor.http_response(self, request, response) |
a6420bf5 | 1617 | |
ac668111 | 1618 | https_request = urllib.request.HTTPCookieProcessor.http_request |
a6420bf5 S |
1619 | https_response = http_response |
1620 | ||
1621 | ||
ac668111 | 1622 | class YoutubeDLRedirectHandler(urllib.request.HTTPRedirectHandler): |
201c1459 | 1623 | """YoutubeDL redirect handler |
1624 | ||
1625 | The code is based on HTTPRedirectHandler implementation from CPython [1]. | |
1626 | ||
1627 | This redirect handler solves two issues: | |
1628 | - ensures redirect URL is always unicode under python 2 | |
1629 | - introduces support for experimental HTTP response status code | |
1630 | 308 Permanent Redirect [2] used by some sites [3] | |
1631 | ||
1632 | 1. https://github.com/python/cpython/blob/master/Lib/urllib/request.py | |
1633 | 2. https://developer.mozilla.org/en-US/docs/Web/HTTP/Status/308 | |
1634 | 3. https://github.com/ytdl-org/youtube-dl/issues/28768 | |
1635 | """ | |
1636 | ||
ac668111 | 1637 | http_error_301 = http_error_303 = http_error_307 = http_error_308 = urllib.request.HTTPRedirectHandler.http_error_302 |
201c1459 | 1638 | |
1639 | def redirect_request(self, req, fp, code, msg, headers, newurl): | |
1640 | """Return a Request or None in response to a redirect. | |
1641 | ||
1642 | This is called by the http_error_30x methods when a | |
1643 | redirection response is received. If a redirection should | |
1644 | take place, return a new Request to allow http_error_30x to | |
1645 | perform the redirect. Otherwise, raise HTTPError if no-one | |
1646 | else should try to handle this url. Return None if you can't | |
1647 | but another Handler might. | |
1648 | """ | |
1649 | m = req.get_method() | |
1650 | if (not (code in (301, 302, 303, 307, 308) and m in ("GET", "HEAD") | |
1651 | or code in (301, 302, 303) and m == "POST")): | |
14f25df2 | 1652 | raise urllib.error.HTTPError(req.full_url, code, msg, headers, fp) |
201c1459 | 1653 | # Strictly (according to RFC 2616), 301 or 302 in response to |
1654 | # a POST MUST NOT cause a redirection without confirmation | |
1655 | # from the user (of urllib.request, in this case). In practice, | |
1656 | # essentially all clients do redirect in this case, so we do | |
1657 | # the same. | |
1658 | ||
201c1459 | 1659 | # Be conciliant with URIs containing a space. This is mainly |
1660 | # redundant with the more complete encoding done in http_error_302(), | |
1661 | # but it is kept for compatibility with other callers. | |
1662 | newurl = newurl.replace(' ', '%20') | |
1663 | ||
1664 | CONTENT_HEADERS = ("content-length", "content-type") | |
1665 | # NB: don't use dict comprehension for python 2.6 compatibility | |
86e5f3ed | 1666 | newheaders = {k: v for k, v in req.headers.items() if k.lower() not in CONTENT_HEADERS} |
afac4caa | 1667 | |
1668 | # A 303 must either use GET or HEAD for subsequent request | |
1669 | # https://datatracker.ietf.org/doc/html/rfc7231#section-6.4.4 | |
1670 | if code == 303 and m != 'HEAD': | |
1671 | m = 'GET' | |
1672 | # 301 and 302 redirects are commonly turned into a GET from a POST | |
1673 | # for subsequent requests by browsers, so we'll do the same. | |
1674 | # https://datatracker.ietf.org/doc/html/rfc7231#section-6.4.2 | |
1675 | # https://datatracker.ietf.org/doc/html/rfc7231#section-6.4.3 | |
1676 | if code in (301, 302) and m == 'POST': | |
1677 | m = 'GET' | |
1678 | ||
ac668111 | 1679 | return urllib.request.Request( |
201c1459 | 1680 | newurl, headers=newheaders, origin_req_host=req.origin_req_host, |
afac4caa | 1681 | unverifiable=True, method=m) |
fca6dba8 S |
1682 | |
1683 | ||
46f59e89 S |
1684 | def extract_timezone(date_str): |
1685 | m = re.search( | |
f137e4c2 | 1686 | r'''(?x) |
1687 | ^.{8,}? # >=8 char non-TZ prefix, if present | |
1688 | (?P<tz>Z| # just the UTC Z, or | |
1689 | (?:(?<=.\b\d{4}|\b\d{2}:\d\d)| # preceded by 4 digits or hh:mm or | |
1690 | (?<!.\b[a-zA-Z]{3}|[a-zA-Z]{4}|..\b\d\d)) # not preceded by 3 alpha word or >= 4 alpha or 2 digits | |
1691 | [ ]? # optional space | |
1692 | (?P<sign>\+|-) # +/- | |
1693 | (?P<hours>[0-9]{2}):?(?P<minutes>[0-9]{2}) # hh[:]mm | |
1694 | $) | |
1695 | ''', date_str) | |
46f59e89 | 1696 | if not m: |
8f53dc44 | 1697 | m = re.search(r'\d{1,2}:\d{1,2}(?:\.\d+)?(?P<tz>\s*[A-Z]+)$', date_str) |
1698 | timezone = TIMEZONE_NAMES.get(m and m.group('tz').strip()) | |
1699 | if timezone is not None: | |
1700 | date_str = date_str[:-len(m.group('tz'))] | |
1701 | timezone = datetime.timedelta(hours=timezone or 0) | |
46f59e89 S |
1702 | else: |
1703 | date_str = date_str[:-len(m.group('tz'))] | |
1704 | if not m.group('sign'): | |
1705 | timezone = datetime.timedelta() | |
1706 | else: | |
1707 | sign = 1 if m.group('sign') == '+' else -1 | |
1708 | timezone = datetime.timedelta( | |
1709 | hours=sign * int(m.group('hours')), | |
1710 | minutes=sign * int(m.group('minutes'))) | |
1711 | return timezone, date_str | |
1712 | ||
1713 | ||
08b38d54 | 1714 | def parse_iso8601(date_str, delimiter='T', timezone=None): |
912b38b4 PH |
1715 | """ Return a UNIX timestamp from the given date """ |
1716 | ||
1717 | if date_str is None: | |
1718 | return None | |
1719 | ||
52c3a6e4 S |
1720 | date_str = re.sub(r'\.[0-9]+', '', date_str) |
1721 | ||
08b38d54 | 1722 | if timezone is None: |
46f59e89 S |
1723 | timezone, date_str = extract_timezone(date_str) |
1724 | ||
19a03940 | 1725 | with contextlib.suppress(ValueError): |
86e5f3ed | 1726 | date_format = f'%Y-%m-%d{delimiter}%H:%M:%S' |
52c3a6e4 S |
1727 | dt = datetime.datetime.strptime(date_str, date_format) - timezone |
1728 | return calendar.timegm(dt.timetuple()) | |
912b38b4 PH |
1729 | |
1730 | ||
46f59e89 S |
1731 | def date_formats(day_first=True): |
1732 | return DATE_FORMATS_DAY_FIRST if day_first else DATE_FORMATS_MONTH_FIRST | |
1733 | ||
1734 | ||
42bdd9d0 | 1735 | def unified_strdate(date_str, day_first=True): |
bf50b038 | 1736 | """Return a string with the date in the format YYYYMMDD""" |
64e7ad60 PH |
1737 | |
1738 | if date_str is None: | |
1739 | return None | |
bf50b038 | 1740 | upload_date = None |
5f6a1245 | 1741 | # Replace commas |
026fcc04 | 1742 | date_str = date_str.replace(',', ' ') |
42bdd9d0 | 1743 | # Remove AM/PM + timezone |
9bb8e0a3 | 1744 | date_str = re.sub(r'(?i)\s*(?:AM|PM)(?:\s+[A-Z]+)?', '', date_str) |
46f59e89 | 1745 | _, date_str = extract_timezone(date_str) |
42bdd9d0 | 1746 | |
46f59e89 | 1747 | for expression in date_formats(day_first): |
19a03940 | 1748 | with contextlib.suppress(ValueError): |
bf50b038 | 1749 | upload_date = datetime.datetime.strptime(date_str, expression).strftime('%Y%m%d') |
42393ce2 PH |
1750 | if upload_date is None: |
1751 | timetuple = email.utils.parsedate_tz(date_str) | |
1752 | if timetuple: | |
19a03940 | 1753 | with contextlib.suppress(ValueError): |
c6b9cf05 | 1754 | upload_date = datetime.datetime(*timetuple[:6]).strftime('%Y%m%d') |
6a750402 | 1755 | if upload_date is not None: |
14f25df2 | 1756 | return str(upload_date) |
bf50b038 | 1757 | |
5f6a1245 | 1758 | |
46f59e89 S |
1759 | def unified_timestamp(date_str, day_first=True): |
1760 | if date_str is None: | |
1761 | return None | |
1762 | ||
8f53dc44 | 1763 | date_str = re.sub(r'\s+', ' ', re.sub( |
1764 | r'(?i)[,|]|(mon|tues?|wed(nes)?|thu(rs)?|fri|sat(ur)?)(day)?', '', date_str)) | |
46f59e89 | 1765 | |
7dc2a74e | 1766 | pm_delta = 12 if re.search(r'(?i)PM', date_str) else 0 |
46f59e89 S |
1767 | timezone, date_str = extract_timezone(date_str) |
1768 | ||
1769 | # Remove AM/PM + timezone | |
1770 | date_str = re.sub(r'(?i)\s*(?:AM|PM)(?:\s+[A-Z]+)?', '', date_str) | |
1771 | ||
deef3195 S |
1772 | # Remove unrecognized timezones from ISO 8601 alike timestamps |
1773 | m = re.search(r'\d{1,2}:\d{1,2}(?:\.\d+)?(?P<tz>\s*[A-Z]+)$', date_str) | |
1774 | if m: | |
1775 | date_str = date_str[:-len(m.group('tz'))] | |
1776 | ||
f226880c PH |
1777 | # Python only supports microseconds, so remove nanoseconds |
1778 | m = re.search(r'^([0-9]{4,}-[0-9]{1,2}-[0-9]{1,2}T[0-9]{1,2}:[0-9]{1,2}:[0-9]{1,2}\.[0-9]{6})[0-9]+$', date_str) | |
1779 | if m: | |
1780 | date_str = m.group(1) | |
1781 | ||
46f59e89 | 1782 | for expression in date_formats(day_first): |
19a03940 | 1783 | with contextlib.suppress(ValueError): |
7dc2a74e | 1784 | dt = datetime.datetime.strptime(date_str, expression) - timezone + datetime.timedelta(hours=pm_delta) |
46f59e89 | 1785 | return calendar.timegm(dt.timetuple()) |
8f53dc44 | 1786 | |
46f59e89 S |
1787 | timetuple = email.utils.parsedate_tz(date_str) |
1788 | if timetuple: | |
8f53dc44 | 1789 | return calendar.timegm(timetuple) + pm_delta * 3600 - timezone.total_seconds() |
46f59e89 S |
1790 | |
1791 | ||
28e614de | 1792 | def determine_ext(url, default_ext='unknown_video'): |
85750f89 | 1793 | if url is None or '.' not in url: |
f4776371 | 1794 | return default_ext |
9cb9a5df | 1795 | guess = url.partition('?')[0].rpartition('.')[2] |
73e79f2a PH |
1796 | if re.match(r'^[A-Za-z0-9]+$', guess): |
1797 | return guess | |
a7aaa398 S |
1798 | # Try extract ext from URLs like http://example.com/foo/bar.mp4/?download |
1799 | elif guess.rstrip('/') in KNOWN_EXTENSIONS: | |
9cb9a5df | 1800 | return guess.rstrip('/') |
73e79f2a | 1801 | else: |
cbdbb766 | 1802 | return default_ext |
73e79f2a | 1803 | |
5f6a1245 | 1804 | |
824fa511 S |
1805 | def subtitles_filename(filename, sub_lang, sub_format, expected_real_ext=None): |
1806 | return replace_extension(filename, sub_lang + '.' + sub_format, expected_real_ext) | |
d4051a8e | 1807 | |
5f6a1245 | 1808 | |
9e62f283 | 1809 | def datetime_from_str(date_str, precision='auto', format='%Y%m%d'): |
3d38b2d6 | 1810 | R""" |
1811 | Return a datetime object from a string. | |
1812 | Supported format: | |
1813 | (now|today|yesterday|DATE)([+-]\d+(microsecond|second|minute|hour|day|week|month|year)s?)? | |
1814 | ||
1815 | @param format strftime format of DATE | |
1816 | @param precision Round the datetime object: auto|microsecond|second|minute|hour|day | |
1817 | auto: round to the unit provided in date_str (if applicable). | |
9e62f283 | 1818 | """ |
1819 | auto_precision = False | |
1820 | if precision == 'auto': | |
1821 | auto_precision = True | |
1822 | precision = 'microsecond' | |
396a76f7 | 1823 | today = datetime_round(datetime.datetime.utcnow(), precision) |
f8795e10 | 1824 | if date_str in ('now', 'today'): |
37254abc | 1825 | return today |
f8795e10 PH |
1826 | if date_str == 'yesterday': |
1827 | return today - datetime.timedelta(days=1) | |
9e62f283 | 1828 | match = re.match( |
3d38b2d6 | 1829 | r'(?P<start>.+)(?P<sign>[+-])(?P<time>\d+)(?P<unit>microsecond|second|minute|hour|day|week|month|year)s?', |
9e62f283 | 1830 | date_str) |
37254abc | 1831 | if match is not None: |
9e62f283 | 1832 | start_time = datetime_from_str(match.group('start'), precision, format) |
1833 | time = int(match.group('time')) * (-1 if match.group('sign') == '-' else 1) | |
37254abc | 1834 | unit = match.group('unit') |
9e62f283 | 1835 | if unit == 'month' or unit == 'year': |
1836 | new_date = datetime_add_months(start_time, time * 12 if unit == 'year' else time) | |
37254abc | 1837 | unit = 'day' |
9e62f283 | 1838 | else: |
1839 | if unit == 'week': | |
1840 | unit = 'day' | |
1841 | time *= 7 | |
1842 | delta = datetime.timedelta(**{unit + 's': time}) | |
1843 | new_date = start_time + delta | |
1844 | if auto_precision: | |
1845 | return datetime_round(new_date, unit) | |
1846 | return new_date | |
1847 | ||
1848 | return datetime_round(datetime.datetime.strptime(date_str, format), precision) | |
1849 | ||
1850 | ||
d49f8db3 | 1851 | def date_from_str(date_str, format='%Y%m%d', strict=False): |
3d38b2d6 | 1852 | R""" |
1853 | Return a date object from a string using datetime_from_str | |
9e62f283 | 1854 | |
3d38b2d6 | 1855 | @param strict Restrict allowed patterns to "YYYYMMDD" and |
1856 | (now|today|yesterday)(-\d+(day|week|month|year)s?)? | |
9e62f283 | 1857 | """ |
3d38b2d6 | 1858 | if strict and not re.fullmatch(r'\d{8}|(now|today|yesterday)(-\d+(day|week|month|year)s?)?', date_str): |
1859 | raise ValueError(f'Invalid date format "{date_str}"') | |
9e62f283 | 1860 | return datetime_from_str(date_str, precision='microsecond', format=format).date() |
1861 | ||
1862 | ||
1863 | def datetime_add_months(dt, months): | |
1864 | """Increment/Decrement a datetime object by months.""" | |
1865 | month = dt.month + months - 1 | |
1866 | year = dt.year + month // 12 | |
1867 | month = month % 12 + 1 | |
1868 | day = min(dt.day, calendar.monthrange(year, month)[1]) | |
1869 | return dt.replace(year, month, day) | |
1870 | ||
1871 | ||
1872 | def datetime_round(dt, precision='day'): | |
1873 | """ | |
1874 | Round a datetime object's time to a specific precision | |
1875 | """ | |
1876 | if precision == 'microsecond': | |
1877 | return dt | |
1878 | ||
1879 | unit_seconds = { | |
1880 | 'day': 86400, | |
1881 | 'hour': 3600, | |
1882 | 'minute': 60, | |
1883 | 'second': 1, | |
1884 | } | |
1885 | roundto = lambda x, n: ((x + n / 2) // n) * n | |
1886 | timestamp = calendar.timegm(dt.timetuple()) | |
1887 | return datetime.datetime.utcfromtimestamp(roundto(timestamp, unit_seconds[precision])) | |
5f6a1245 JW |
1888 | |
1889 | ||
e63fc1be | 1890 | def hyphenate_date(date_str): |
1891 | """ | |
1892 | Convert a date in 'YYYYMMDD' format to 'YYYY-MM-DD' format""" | |
1893 | match = re.match(r'^(\d\d\d\d)(\d\d)(\d\d)$', date_str) | |
1894 | if match is not None: | |
1895 | return '-'.join(match.groups()) | |
1896 | else: | |
1897 | return date_str | |
1898 | ||
5f6a1245 | 1899 | |
86e5f3ed | 1900 | class DateRange: |
bd558525 | 1901 | """Represents a time interval between two dates""" |
5f6a1245 | 1902 | |
bd558525 JMF |
1903 | def __init__(self, start=None, end=None): |
1904 | """start and end must be strings in the format accepted by date""" | |
1905 | if start is not None: | |
d49f8db3 | 1906 | self.start = date_from_str(start, strict=True) |
bd558525 JMF |
1907 | else: |
1908 | self.start = datetime.datetime.min.date() | |
1909 | if end is not None: | |
d49f8db3 | 1910 | self.end = date_from_str(end, strict=True) |
bd558525 JMF |
1911 | else: |
1912 | self.end = datetime.datetime.max.date() | |
37254abc | 1913 | if self.start > self.end: |
bd558525 | 1914 | raise ValueError('Date range: "%s" , the start date must be before the end date' % self) |
5f6a1245 | 1915 | |
bd558525 JMF |
1916 | @classmethod |
1917 | def day(cls, day): | |
1918 | """Returns a range that only contains the given day""" | |
5f6a1245 JW |
1919 | return cls(day, day) |
1920 | ||
bd558525 JMF |
1921 | def __contains__(self, date): |
1922 | """Check if the date is in the range""" | |
37254abc JMF |
1923 | if not isinstance(date, datetime.date): |
1924 | date = date_from_str(date) | |
1925 | return self.start <= date <= self.end | |
5f6a1245 | 1926 | |
bd558525 | 1927 | def __str__(self): |
86e5f3ed | 1928 | return f'{self.start.isoformat()} - {self.end.isoformat()}' |
c496ca96 | 1929 | |
f2df4071 | 1930 | def __eq__(self, other): |
1931 | return (isinstance(other, DateRange) | |
1932 | and self.start == other.start and self.end == other.end) | |
1933 | ||
c496ca96 PH |
1934 | |
1935 | def platform_name(): | |
14f25df2 | 1936 | """ Returns the platform name as a str """ |
b1f94422 | 1937 | write_string('DeprecationWarning: yt_dlp.utils.platform_name is deprecated, use platform.platform instead') |
1938 | return platform.platform() | |
c496ca96 | 1939 | |
b1f94422 | 1940 | |
1941 | @functools.cache | |
1942 | def system_identifier(): | |
1943 | python_implementation = platform.python_implementation() | |
1944 | if python_implementation == 'PyPy' and hasattr(sys, 'pypy_version_info'): | |
1945 | python_implementation += ' version %d.%d.%d' % sys.pypy_version_info[:3] | |
1946 | ||
1947 | return 'Python %s (%s %s) - %s %s' % ( | |
1948 | platform.python_version(), | |
1949 | python_implementation, | |
1950 | platform.architecture()[0], | |
1951 | platform.platform(), | |
1952 | format_field(join_nonempty(*platform.libc_ver(), delim=' '), None, '(%s)'), | |
1953 | ) | |
c257baff PH |
1954 | |
1955 | ||
0b9c08b4 | 1956 | @functools.cache |
49fa4d9a | 1957 | def get_windows_version(): |
8a82af35 | 1958 | ''' Get Windows version. returns () if it's not running on Windows ''' |
49fa4d9a N |
1959 | if compat_os_name == 'nt': |
1960 | return version_tuple(platform.win32_ver()[1]) | |
1961 | else: | |
8a82af35 | 1962 | return () |
49fa4d9a N |
1963 | |
1964 | ||
734f90bb | 1965 | def write_string(s, out=None, encoding=None): |
19a03940 | 1966 | assert isinstance(s, str) |
1967 | out = out or sys.stderr | |
7459e3a2 | 1968 | |
fe1daad3 | 1969 | if compat_os_name == 'nt' and supports_terminal_sequences(out): |
3fe75fdc | 1970 | s = re.sub(r'([\r\n]+)', r' \1', s) |
59f943cd | 1971 | |
8a82af35 | 1972 | enc, buffer = None, out |
cfb0511d | 1973 | if 'b' in getattr(out, 'mode', ''): |
c487cf00 | 1974 | enc = encoding or preferredencoding() |
104aa738 | 1975 | elif hasattr(out, 'buffer'): |
8a82af35 | 1976 | buffer = out.buffer |
104aa738 | 1977 | enc = encoding or getattr(out, 'encoding', None) or preferredencoding() |
c487cf00 | 1978 | |
8a82af35 | 1979 | buffer.write(s.encode(enc, 'ignore') if enc else s) |
7459e3a2 PH |
1980 | out.flush() |
1981 | ||
1982 | ||
48ea9cea PH |
1983 | def bytes_to_intlist(bs): |
1984 | if not bs: | |
1985 | return [] | |
1986 | if isinstance(bs[0], int): # Python 3 | |
1987 | return list(bs) | |
1988 | else: | |
1989 | return [ord(c) for c in bs] | |
1990 | ||
c257baff | 1991 | |
cba892fa | 1992 | def intlist_to_bytes(xs): |
1993 | if not xs: | |
1994 | return b'' | |
ac668111 | 1995 | return struct.pack('%dB' % len(xs), *xs) |
c38b1e77 PH |
1996 | |
1997 | ||
8a82af35 | 1998 | class LockingUnsupportedError(OSError): |
1890fc63 | 1999 | msg = 'File locking is not supported' |
0edb3e33 | 2000 | |
2001 | def __init__(self): | |
2002 | super().__init__(self.msg) | |
2003 | ||
2004 | ||
c1c9a79c PH |
2005 | # Cross-platform file locking |
2006 | if sys.platform == 'win32': | |
fe0918bb | 2007 | import ctypes |
c1c9a79c PH |
2008 | import ctypes.wintypes |
2009 | import msvcrt | |
2010 | ||
2011 | class OVERLAPPED(ctypes.Structure): | |
2012 | _fields_ = [ | |
2013 | ('Internal', ctypes.wintypes.LPVOID), | |
2014 | ('InternalHigh', ctypes.wintypes.LPVOID), | |
2015 | ('Offset', ctypes.wintypes.DWORD), | |
2016 | ('OffsetHigh', ctypes.wintypes.DWORD), | |
2017 | ('hEvent', ctypes.wintypes.HANDLE), | |
2018 | ] | |
2019 | ||
2020 | kernel32 = ctypes.windll.kernel32 | |
2021 | LockFileEx = kernel32.LockFileEx | |
2022 | LockFileEx.argtypes = [ | |
2023 | ctypes.wintypes.HANDLE, # hFile | |
2024 | ctypes.wintypes.DWORD, # dwFlags | |
2025 | ctypes.wintypes.DWORD, # dwReserved | |
2026 | ctypes.wintypes.DWORD, # nNumberOfBytesToLockLow | |
2027 | ctypes.wintypes.DWORD, # nNumberOfBytesToLockHigh | |
2028 | ctypes.POINTER(OVERLAPPED) # Overlapped | |
2029 | ] | |
2030 | LockFileEx.restype = ctypes.wintypes.BOOL | |
2031 | UnlockFileEx = kernel32.UnlockFileEx | |
2032 | UnlockFileEx.argtypes = [ | |
2033 | ctypes.wintypes.HANDLE, # hFile | |
2034 | ctypes.wintypes.DWORD, # dwReserved | |
2035 | ctypes.wintypes.DWORD, # nNumberOfBytesToLockLow | |
2036 | ctypes.wintypes.DWORD, # nNumberOfBytesToLockHigh | |
2037 | ctypes.POINTER(OVERLAPPED) # Overlapped | |
2038 | ] | |
2039 | UnlockFileEx.restype = ctypes.wintypes.BOOL | |
2040 | whole_low = 0xffffffff | |
2041 | whole_high = 0x7fffffff | |
2042 | ||
747c0bd1 | 2043 | def _lock_file(f, exclusive, block): |
c1c9a79c PH |
2044 | overlapped = OVERLAPPED() |
2045 | overlapped.Offset = 0 | |
2046 | overlapped.OffsetHigh = 0 | |
2047 | overlapped.hEvent = 0 | |
2048 | f._lock_file_overlapped_p = ctypes.pointer(overlapped) | |
747c0bd1 | 2049 | |
2050 | if not LockFileEx(msvcrt.get_osfhandle(f.fileno()), | |
2051 | (0x2 if exclusive else 0x0) | (0x0 if block else 0x1), | |
2052 | 0, whole_low, whole_high, f._lock_file_overlapped_p): | |
2cb19820 | 2053 | # NB: No argument form of "ctypes.FormatError" does not work on PyPy |
2054 | raise BlockingIOError(f'Locking file failed: {ctypes.FormatError(ctypes.GetLastError())!r}') | |
c1c9a79c PH |
2055 | |
2056 | def _unlock_file(f): | |
2057 | assert f._lock_file_overlapped_p | |
2058 | handle = msvcrt.get_osfhandle(f.fileno()) | |
747c0bd1 | 2059 | if not UnlockFileEx(handle, 0, whole_low, whole_high, f._lock_file_overlapped_p): |
c1c9a79c PH |
2060 | raise OSError('Unlocking file failed: %r' % ctypes.FormatError()) |
2061 | ||
2062 | else: | |
399a76e6 YCH |
2063 | try: |
2064 | import fcntl | |
c1c9a79c | 2065 | |
a3125791 | 2066 | def _lock_file(f, exclusive, block): |
b63837bc | 2067 | flags = fcntl.LOCK_EX if exclusive else fcntl.LOCK_SH |
2068 | if not block: | |
2069 | flags |= fcntl.LOCK_NB | |
acea8d7c | 2070 | try: |
b63837bc | 2071 | fcntl.flock(f, flags) |
acea8d7c JK |
2072 | except BlockingIOError: |
2073 | raise | |
2074 | except OSError: # AOSP does not have flock() | |
b63837bc | 2075 | fcntl.lockf(f, flags) |
c1c9a79c | 2076 | |
399a76e6 | 2077 | def _unlock_file(f): |
acea8d7c JK |
2078 | try: |
2079 | fcntl.flock(f, fcntl.LOCK_UN) | |
2080 | except OSError: | |
2081 | fcntl.lockf(f, fcntl.LOCK_UN) | |
a3125791 | 2082 | |
399a76e6 | 2083 | except ImportError: |
399a76e6 | 2084 | |
a3125791 | 2085 | def _lock_file(f, exclusive, block): |
0edb3e33 | 2086 | raise LockingUnsupportedError() |
399a76e6 YCH |
2087 | |
2088 | def _unlock_file(f): | |
0edb3e33 | 2089 | raise LockingUnsupportedError() |
c1c9a79c PH |
2090 | |
2091 | ||
86e5f3ed | 2092 | class locked_file: |
0edb3e33 | 2093 | locked = False |
747c0bd1 | 2094 | |
a3125791 | 2095 | def __init__(self, filename, mode, block=True, encoding=None): |
fcfa8853 JK |
2096 | if mode not in {'r', 'rb', 'a', 'ab', 'w', 'wb'}: |
2097 | raise NotImplementedError(mode) | |
2098 | self.mode, self.block = mode, block | |
2099 | ||
2100 | writable = any(f in mode for f in 'wax+') | |
2101 | readable = any(f in mode for f in 'r+') | |
2102 | flags = functools.reduce(operator.ior, ( | |
2103 | getattr(os, 'O_CLOEXEC', 0), # UNIX only | |
2104 | getattr(os, 'O_BINARY', 0), # Windows only | |
2105 | getattr(os, 'O_NOINHERIT', 0), # Windows only | |
2106 | os.O_CREAT if writable else 0, # O_TRUNC only after locking | |
2107 | os.O_APPEND if 'a' in mode else 0, | |
2108 | os.O_EXCL if 'x' in mode else 0, | |
2109 | os.O_RDONLY if not writable else os.O_RDWR if readable else os.O_WRONLY, | |
2110 | )) | |
2111 | ||
98804d03 | 2112 | self.f = os.fdopen(os.open(filename, flags, 0o666), mode, encoding=encoding) |
c1c9a79c PH |
2113 | |
2114 | def __enter__(self): | |
a3125791 | 2115 | exclusive = 'r' not in self.mode |
c1c9a79c | 2116 | try: |
a3125791 | 2117 | _lock_file(self.f, exclusive, self.block) |
0edb3e33 | 2118 | self.locked = True |
86e5f3ed | 2119 | except OSError: |
c1c9a79c PH |
2120 | self.f.close() |
2121 | raise | |
fcfa8853 | 2122 | if 'w' in self.mode: |
131e14dc JK |
2123 | try: |
2124 | self.f.truncate() | |
2125 | except OSError as e: | |
1890fc63 | 2126 | if e.errno not in ( |
2127 | errno.ESPIPE, # Illegal seek - expected for FIFO | |
2128 | errno.EINVAL, # Invalid argument - expected for /dev/null | |
2129 | ): | |
2130 | raise | |
c1c9a79c PH |
2131 | return self |
2132 | ||
0edb3e33 | 2133 | def unlock(self): |
2134 | if not self.locked: | |
2135 | return | |
c1c9a79c | 2136 | try: |
0edb3e33 | 2137 | _unlock_file(self.f) |
c1c9a79c | 2138 | finally: |
0edb3e33 | 2139 | self.locked = False |
c1c9a79c | 2140 | |
0edb3e33 | 2141 | def __exit__(self, *_): |
2142 | try: | |
2143 | self.unlock() | |
2144 | finally: | |
2145 | self.f.close() | |
4eb7f1d1 | 2146 | |
0edb3e33 | 2147 | open = __enter__ |
2148 | close = __exit__ | |
a3125791 | 2149 | |
0edb3e33 | 2150 | def __getattr__(self, attr): |
2151 | return getattr(self.f, attr) | |
a3125791 | 2152 | |
0edb3e33 | 2153 | def __iter__(self): |
2154 | return iter(self.f) | |
a3125791 | 2155 | |
4eb7f1d1 | 2156 | |
0b9c08b4 | 2157 | @functools.cache |
4644ac55 S |
2158 | def get_filesystem_encoding(): |
2159 | encoding = sys.getfilesystemencoding() | |
2160 | return encoding if encoding is not None else 'utf-8' | |
2161 | ||
2162 | ||
4eb7f1d1 | 2163 | def shell_quote(args): |
a6a173c2 | 2164 | quoted_args = [] |
4644ac55 | 2165 | encoding = get_filesystem_encoding() |
a6a173c2 JMF |
2166 | for a in args: |
2167 | if isinstance(a, bytes): | |
2168 | # We may get a filename encoded with 'encodeFilename' | |
2169 | a = a.decode(encoding) | |
aefce8e6 | 2170 | quoted_args.append(compat_shlex_quote(a)) |
28e614de | 2171 | return ' '.join(quoted_args) |
9d4660ca PH |
2172 | |
2173 | ||
2174 | def smuggle_url(url, data): | |
2175 | """ Pass additional data in a URL for internal use. """ | |
2176 | ||
81953d1a RA |
2177 | url, idata = unsmuggle_url(url, {}) |
2178 | data.update(idata) | |
14f25df2 | 2179 | sdata = urllib.parse.urlencode( |
28e614de PH |
2180 | {'__youtubedl_smuggle': json.dumps(data)}) |
2181 | return url + '#' + sdata | |
9d4660ca PH |
2182 | |
2183 | ||
79f82953 | 2184 | def unsmuggle_url(smug_url, default=None): |
83e865a3 | 2185 | if '#__youtubedl_smuggle' not in smug_url: |
79f82953 | 2186 | return smug_url, default |
28e614de | 2187 | url, _, sdata = smug_url.rpartition('#') |
14f25df2 | 2188 | jsond = urllib.parse.parse_qs(sdata)['__youtubedl_smuggle'][0] |
9d4660ca PH |
2189 | data = json.loads(jsond) |
2190 | return url, data | |
02dbf93f PH |
2191 | |
2192 | ||
e0fd9573 | 2193 | def format_decimal_suffix(num, fmt='%d%s', *, factor=1000): |
2194 | """ Formats numbers with decimal sufixes like K, M, etc """ | |
2195 | num, factor = float_or_none(num), float(factor) | |
4c3f8c3f | 2196 | if num is None or num < 0: |
e0fd9573 | 2197 | return None |
eeb2a770 | 2198 | POSSIBLE_SUFFIXES = 'kMGTPEZY' |
2199 | exponent = 0 if num == 0 else min(int(math.log(num, factor)), len(POSSIBLE_SUFFIXES)) | |
2200 | suffix = ['', *POSSIBLE_SUFFIXES][exponent] | |
abbeeebc | 2201 | if factor == 1024: |
2202 | suffix = {'k': 'Ki', '': ''}.get(suffix, f'{suffix}i') | |
e0fd9573 | 2203 | converted = num / (factor ** exponent) |
abbeeebc | 2204 | return fmt % (converted, suffix) |
e0fd9573 | 2205 | |
2206 | ||
02dbf93f | 2207 | def format_bytes(bytes): |
f02d24d8 | 2208 | return format_decimal_suffix(bytes, '%.2f%sB', factor=1024) or 'N/A' |
f53c966a | 2209 | |
1c088fa8 | 2210 | |
fb47597b S |
2211 | def lookup_unit_table(unit_table, s): |
2212 | units_re = '|'.join(re.escape(u) for u in unit_table) | |
2213 | m = re.match( | |
782b1b5b | 2214 | r'(?P<num>[0-9]+(?:[,.][0-9]*)?)\s*(?P<unit>%s)\b' % units_re, s) |
fb47597b S |
2215 | if not m: |
2216 | return None | |
2217 | num_str = m.group('num').replace(',', '.') | |
2218 | mult = unit_table[m.group('unit')] | |
2219 | return int(float(num_str) * mult) | |
2220 | ||
2221 | ||
be64b5b0 PH |
2222 | def parse_filesize(s): |
2223 | if s is None: | |
2224 | return None | |
2225 | ||
dfb1b146 | 2226 | # The lower-case forms are of course incorrect and unofficial, |
be64b5b0 PH |
2227 | # but we support those too |
2228 | _UNIT_TABLE = { | |
2229 | 'B': 1, | |
2230 | 'b': 1, | |
70852b47 | 2231 | 'bytes': 1, |
be64b5b0 PH |
2232 | 'KiB': 1024, |
2233 | 'KB': 1000, | |
2234 | 'kB': 1024, | |
2235 | 'Kb': 1000, | |
13585d76 | 2236 | 'kb': 1000, |
70852b47 YCH |
2237 | 'kilobytes': 1000, |
2238 | 'kibibytes': 1024, | |
be64b5b0 PH |
2239 | 'MiB': 1024 ** 2, |
2240 | 'MB': 1000 ** 2, | |
2241 | 'mB': 1024 ** 2, | |
2242 | 'Mb': 1000 ** 2, | |
13585d76 | 2243 | 'mb': 1000 ** 2, |
70852b47 YCH |
2244 | 'megabytes': 1000 ** 2, |
2245 | 'mebibytes': 1024 ** 2, | |
be64b5b0 PH |
2246 | 'GiB': 1024 ** 3, |
2247 | 'GB': 1000 ** 3, | |
2248 | 'gB': 1024 ** 3, | |
2249 | 'Gb': 1000 ** 3, | |
13585d76 | 2250 | 'gb': 1000 ** 3, |
70852b47 YCH |
2251 | 'gigabytes': 1000 ** 3, |
2252 | 'gibibytes': 1024 ** 3, | |
be64b5b0 PH |
2253 | 'TiB': 1024 ** 4, |
2254 | 'TB': 1000 ** 4, | |
2255 | 'tB': 1024 ** 4, | |
2256 | 'Tb': 1000 ** 4, | |
13585d76 | 2257 | 'tb': 1000 ** 4, |
70852b47 YCH |
2258 | 'terabytes': 1000 ** 4, |
2259 | 'tebibytes': 1024 ** 4, | |
be64b5b0 PH |
2260 | 'PiB': 1024 ** 5, |
2261 | 'PB': 1000 ** 5, | |
2262 | 'pB': 1024 ** 5, | |
2263 | 'Pb': 1000 ** 5, | |
13585d76 | 2264 | 'pb': 1000 ** 5, |
70852b47 YCH |
2265 | 'petabytes': 1000 ** 5, |
2266 | 'pebibytes': 1024 ** 5, | |
be64b5b0 PH |
2267 | 'EiB': 1024 ** 6, |
2268 | 'EB': 1000 ** 6, | |
2269 | 'eB': 1024 ** 6, | |
2270 | 'Eb': 1000 ** 6, | |
13585d76 | 2271 | 'eb': 1000 ** 6, |
70852b47 YCH |
2272 | 'exabytes': 1000 ** 6, |
2273 | 'exbibytes': 1024 ** 6, | |
be64b5b0 PH |
2274 | 'ZiB': 1024 ** 7, |
2275 | 'ZB': 1000 ** 7, | |
2276 | 'zB': 1024 ** 7, | |
2277 | 'Zb': 1000 ** 7, | |
13585d76 | 2278 | 'zb': 1000 ** 7, |
70852b47 YCH |
2279 | 'zettabytes': 1000 ** 7, |
2280 | 'zebibytes': 1024 ** 7, | |
be64b5b0 PH |
2281 | 'YiB': 1024 ** 8, |
2282 | 'YB': 1000 ** 8, | |
2283 | 'yB': 1024 ** 8, | |
2284 | 'Yb': 1000 ** 8, | |
13585d76 | 2285 | 'yb': 1000 ** 8, |
70852b47 YCH |
2286 | 'yottabytes': 1000 ** 8, |
2287 | 'yobibytes': 1024 ** 8, | |
be64b5b0 PH |
2288 | } |
2289 | ||
fb47597b S |
2290 | return lookup_unit_table(_UNIT_TABLE, s) |
2291 | ||
2292 | ||
2293 | def parse_count(s): | |
2294 | if s is None: | |
be64b5b0 PH |
2295 | return None |
2296 | ||
352d5da8 | 2297 | s = re.sub(r'^[^\d]+\s', '', s).strip() |
fb47597b S |
2298 | |
2299 | if re.match(r'^[\d,.]+$', s): | |
2300 | return str_to_int(s) | |
2301 | ||
2302 | _UNIT_TABLE = { | |
2303 | 'k': 1000, | |
2304 | 'K': 1000, | |
2305 | 'm': 1000 ** 2, | |
2306 | 'M': 1000 ** 2, | |
2307 | 'kk': 1000 ** 2, | |
2308 | 'KK': 1000 ** 2, | |
352d5da8 | 2309 | 'b': 1000 ** 3, |
2310 | 'B': 1000 ** 3, | |
fb47597b | 2311 | } |
be64b5b0 | 2312 | |
352d5da8 | 2313 | ret = lookup_unit_table(_UNIT_TABLE, s) |
2314 | if ret is not None: | |
2315 | return ret | |
2316 | ||
2317 | mobj = re.match(r'([\d,.]+)(?:$|\s)', s) | |
2318 | if mobj: | |
2319 | return str_to_int(mobj.group(1)) | |
be64b5b0 | 2320 | |
2f7ae819 | 2321 | |
5d45484c | 2322 | def parse_resolution(s, *, lenient=False): |
b871d7e9 S |
2323 | if s is None: |
2324 | return {} | |
2325 | ||
5d45484c LNO |
2326 | if lenient: |
2327 | mobj = re.search(r'(?P<w>\d+)\s*[xX×,]\s*(?P<h>\d+)', s) | |
2328 | else: | |
2329 | mobj = re.search(r'(?<![a-zA-Z0-9])(?P<w>\d+)\s*[xX×,]\s*(?P<h>\d+)(?![a-zA-Z0-9])', s) | |
b871d7e9 S |
2330 | if mobj: |
2331 | return { | |
2332 | 'width': int(mobj.group('w')), | |
2333 | 'height': int(mobj.group('h')), | |
2334 | } | |
2335 | ||
17ec8bcf | 2336 | mobj = re.search(r'(?<![a-zA-Z0-9])(\d+)[pPiI](?![a-zA-Z0-9])', s) |
b871d7e9 S |
2337 | if mobj: |
2338 | return {'height': int(mobj.group(1))} | |
2339 | ||
2340 | mobj = re.search(r'\b([48])[kK]\b', s) | |
2341 | if mobj: | |
2342 | return {'height': int(mobj.group(1)) * 540} | |
2343 | ||
2344 | return {} | |
2345 | ||
2346 | ||
0dc41787 | 2347 | def parse_bitrate(s): |
14f25df2 | 2348 | if not isinstance(s, str): |
0dc41787 S |
2349 | return |
2350 | mobj = re.search(r'\b(\d+)\s*kbps', s) | |
2351 | if mobj: | |
2352 | return int(mobj.group(1)) | |
2353 | ||
2354 | ||
a942d6cb | 2355 | def month_by_name(name, lang='en'): |
caefb1de PH |
2356 | """ Return the number of a month by (locale-independently) English name """ |
2357 | ||
f6717dec | 2358 | month_names = MONTH_NAMES.get(lang, MONTH_NAMES['en']) |
a942d6cb | 2359 | |
caefb1de | 2360 | try: |
f6717dec | 2361 | return month_names.index(name) + 1 |
7105440c YCH |
2362 | except ValueError: |
2363 | return None | |
2364 | ||
2365 | ||
2366 | def month_by_abbreviation(abbrev): | |
2367 | """ Return the number of a month by (locale-independently) English | |
2368 | abbreviations """ | |
2369 | ||
2370 | try: | |
2371 | return [s[:3] for s in ENGLISH_MONTH_NAMES].index(abbrev) + 1 | |
caefb1de PH |
2372 | except ValueError: |
2373 | return None | |
18258362 JMF |
2374 | |
2375 | ||
5aafe895 | 2376 | def fix_xml_ampersands(xml_str): |
18258362 | 2377 | """Replace all the '&' by '&' in XML""" |
5aafe895 PH |
2378 | return re.sub( |
2379 | r'&(?!amp;|lt;|gt;|apos;|quot;|#x[0-9a-fA-F]{,4};|#[0-9]{,4};)', | |
28e614de | 2380 | '&', |
5aafe895 | 2381 | xml_str) |
e3946f98 PH |
2382 | |
2383 | ||
2384 | def setproctitle(title): | |
14f25df2 | 2385 | assert isinstance(title, str) |
c1c05c67 | 2386 | |
fe0918bb | 2387 | # Workaround for https://github.com/yt-dlp/yt-dlp/issues/4541 |
2388 | try: | |
2389 | import ctypes | |
2390 | except ImportError: | |
c1c05c67 YCH |
2391 | return |
2392 | ||
e3946f98 | 2393 | try: |
611c1dd9 | 2394 | libc = ctypes.cdll.LoadLibrary('libc.so.6') |
e3946f98 PH |
2395 | except OSError: |
2396 | return | |
2f49bcd6 RC |
2397 | except TypeError: |
2398 | # LoadLibrary in Windows Python 2.7.13 only expects | |
2399 | # a bytestring, but since unicode_literals turns | |
2400 | # every string into a unicode string, it fails. | |
2401 | return | |
0f06bcd7 | 2402 | title_bytes = title.encode() |
6eefe533 PH |
2403 | buf = ctypes.create_string_buffer(len(title_bytes)) |
2404 | buf.value = title_bytes | |
e3946f98 | 2405 | try: |
6eefe533 | 2406 | libc.prctl(15, buf, 0, 0, 0) |
e3946f98 PH |
2407 | except AttributeError: |
2408 | return # Strange libc, just skip this | |
d7dda168 PH |
2409 | |
2410 | ||
2411 | def remove_start(s, start): | |
46bc9b7d | 2412 | return s[len(start):] if s is not None and s.startswith(start) else s |
29eb5174 PH |
2413 | |
2414 | ||
2b9faf55 | 2415 | def remove_end(s, end): |
46bc9b7d | 2416 | return s[:-len(end)] if s is not None and s.endswith(end) else s |
2b9faf55 PH |
2417 | |
2418 | ||
31b2051e S |
2419 | def remove_quotes(s): |
2420 | if s is None or len(s) < 2: | |
2421 | return s | |
2422 | for quote in ('"', "'", ): | |
2423 | if s[0] == quote and s[-1] == quote: | |
2424 | return s[1:-1] | |
2425 | return s | |
2426 | ||
2427 | ||
b6e0c7d2 | 2428 | def get_domain(url): |
ebf99aaf | 2429 | """ |
2430 | This implementation is inconsistent, but is kept for compatibility. | |
2431 | Use this only for "webpage_url_domain" | |
2432 | """ | |
2433 | return remove_start(urllib.parse.urlparse(url).netloc, 'www.') or None | |
b6e0c7d2 U |
2434 | |
2435 | ||
29eb5174 | 2436 | def url_basename(url): |
14f25df2 | 2437 | path = urllib.parse.urlparse(url).path |
28e614de | 2438 | return path.strip('/').split('/')[-1] |
aa94a6d3 PH |
2439 | |
2440 | ||
02dc0a36 S |
2441 | def base_url(url): |
2442 | return re.match(r'https?://[^?#&]+/', url).group() | |
2443 | ||
2444 | ||
e34c3361 | 2445 | def urljoin(base, path): |
4b5de77b | 2446 | if isinstance(path, bytes): |
0f06bcd7 | 2447 | path = path.decode() |
14f25df2 | 2448 | if not isinstance(path, str) or not path: |
e34c3361 | 2449 | return None |
fad4ceb5 | 2450 | if re.match(r'^(?:[a-zA-Z][a-zA-Z0-9+-.]*:)?//', path): |
e34c3361 | 2451 | return path |
4b5de77b | 2452 | if isinstance(base, bytes): |
0f06bcd7 | 2453 | base = base.decode() |
14f25df2 | 2454 | if not isinstance(base, str) or not re.match( |
4b5de77b | 2455 | r'^(?:https?:)?//', base): |
e34c3361 | 2456 | return None |
14f25df2 | 2457 | return urllib.parse.urljoin(base, path) |
e34c3361 S |
2458 | |
2459 | ||
ac668111 | 2460 | class HEADRequest(urllib.request.Request): |
aa94a6d3 | 2461 | def get_method(self): |
611c1dd9 | 2462 | return 'HEAD' |
7217e148 PH |
2463 | |
2464 | ||
ac668111 | 2465 | class PUTRequest(urllib.request.Request): |
95cf60e8 S |
2466 | def get_method(self): |
2467 | return 'PUT' | |
2468 | ||
2469 | ||
9732d77e | 2470 | def int_or_none(v, scale=1, default=None, get_attr=None, invscale=1): |
9e907ebd | 2471 | if get_attr and v is not None: |
2472 | v = getattr(v, get_attr, None) | |
1812afb7 S |
2473 | try: |
2474 | return int(v) * invscale // scale | |
31c49255 | 2475 | except (ValueError, TypeError, OverflowError): |
af98f8ff | 2476 | return default |
9732d77e | 2477 | |
9572013d | 2478 | |
40a90862 | 2479 | def str_or_none(v, default=None): |
14f25df2 | 2480 | return default if v is None else str(v) |
40a90862 | 2481 | |
9732d77e PH |
2482 | |
2483 | def str_to_int(int_str): | |
48d4681e | 2484 | """ A more relaxed version of int_or_none """ |
f9934b96 | 2485 | if isinstance(int_str, int): |
348c6bf1 | 2486 | return int_str |
14f25df2 | 2487 | elif isinstance(int_str, str): |
42db58ec S |
2488 | int_str = re.sub(r'[,\.\+]', '', int_str) |
2489 | return int_or_none(int_str) | |
608d11f5 PH |
2490 | |
2491 | ||
9732d77e | 2492 | def float_or_none(v, scale=1, invscale=1, default=None): |
caf80631 S |
2493 | if v is None: |
2494 | return default | |
2495 | try: | |
2496 | return float(v) * invscale / scale | |
5e1271c5 | 2497 | except (ValueError, TypeError): |
caf80631 | 2498 | return default |
43f775e4 PH |
2499 | |
2500 | ||
c7e327c4 S |
2501 | def bool_or_none(v, default=None): |
2502 | return v if isinstance(v, bool) else default | |
2503 | ||
2504 | ||
53cd37ba | 2505 | def strip_or_none(v, default=None): |
14f25df2 | 2506 | return v.strip() if isinstance(v, str) else default |
b72b4431 S |
2507 | |
2508 | ||
af03000a | 2509 | def url_or_none(url): |
14f25df2 | 2510 | if not url or not isinstance(url, str): |
af03000a S |
2511 | return None |
2512 | url = url.strip() | |
29f7c58a | 2513 | return url if re.match(r'^(?:(?:https?|rt(?:m(?:pt?[es]?|fp)|sp[su]?)|mms|ftps?):)?//', url) else None |
af03000a S |
2514 | |
2515 | ||
3e9b66d7 | 2516 | def request_to_url(req): |
ac668111 | 2517 | if isinstance(req, urllib.request.Request): |
3e9b66d7 LNO |
2518 | return req.get_full_url() |
2519 | else: | |
2520 | return req | |
2521 | ||
2522 | ||
e29663c6 | 2523 | def strftime_or_none(timestamp, date_format, default=None): |
2524 | datetime_object = None | |
2525 | try: | |
f9934b96 | 2526 | if isinstance(timestamp, (int, float)): # unix timestamp |
e29663c6 | 2527 | datetime_object = datetime.datetime.utcfromtimestamp(timestamp) |
14f25df2 | 2528 | elif isinstance(timestamp, str): # assume YYYYMMDD |
e29663c6 | 2529 | datetime_object = datetime.datetime.strptime(timestamp, '%Y%m%d') |
2530 | return datetime_object.strftime(date_format) | |
2531 | except (ValueError, TypeError, AttributeError): | |
2532 | return default | |
2533 | ||
2534 | ||
608d11f5 | 2535 | def parse_duration(s): |
f9934b96 | 2536 | if not isinstance(s, str): |
608d11f5 | 2537 | return None |
ca7b3246 | 2538 | s = s.strip() |
38d79fd1 | 2539 | if not s: |
2540 | return None | |
ca7b3246 | 2541 | |
acaff495 | 2542 | days, hours, mins, secs, ms = [None] * 5 |
8bd1c00b | 2543 | m = re.match(r'''(?x) |
2544 | (?P<before_secs> | |
2545 | (?:(?:(?P<days>[0-9]+):)?(?P<hours>[0-9]+):)?(?P<mins>[0-9]+):)? | |
2546 | (?P<secs>(?(before_secs)[0-9]{1,2}|[0-9]+)) | |
2547 | (?P<ms>[.:][0-9]+)?Z?$ | |
2548 | ''', s) | |
acaff495 | 2549 | if m: |
8bd1c00b | 2550 | days, hours, mins, secs, ms = m.group('days', 'hours', 'mins', 'secs', 'ms') |
acaff495 | 2551 | else: |
2552 | m = re.match( | |
056653bb S |
2553 | r'''(?ix)(?:P? |
2554 | (?: | |
1c1b2f96 | 2555 | [0-9]+\s*y(?:ears?)?,?\s* |
056653bb S |
2556 | )? |
2557 | (?: | |
1c1b2f96 | 2558 | [0-9]+\s*m(?:onths?)?,?\s* |
056653bb S |
2559 | )? |
2560 | (?: | |
1c1b2f96 | 2561 | [0-9]+\s*w(?:eeks?)?,?\s* |
056653bb | 2562 | )? |
8f4b58d7 | 2563 | (?: |
1c1b2f96 | 2564 | (?P<days>[0-9]+)\s*d(?:ays?)?,?\s* |
8f4b58d7 | 2565 | )? |
056653bb | 2566 | T)? |
acaff495 | 2567 | (?: |
1c1b2f96 | 2568 | (?P<hours>[0-9]+)\s*h(?:ours?)?,?\s* |
acaff495 | 2569 | )? |
2570 | (?: | |
1c1b2f96 | 2571 | (?P<mins>[0-9]+)\s*m(?:in(?:ute)?s?)?,?\s* |
acaff495 | 2572 | )? |
2573 | (?: | |
2574 | (?P<secs>[0-9]+)(?P<ms>\.[0-9]+)?\s*s(?:ec(?:ond)?s?)?\s* | |
15846398 | 2575 | )?Z?$''', s) |
acaff495 | 2576 | if m: |
2577 | days, hours, mins, secs, ms = m.groups() | |
2578 | else: | |
15846398 | 2579 | m = re.match(r'(?i)(?:(?P<hours>[0-9.]+)\s*(?:hours?)|(?P<mins>[0-9.]+)\s*(?:mins?\.?|minutes?)\s*)Z?$', s) |
acaff495 | 2580 | if m: |
2581 | hours, mins = m.groups() | |
2582 | else: | |
2583 | return None | |
2584 | ||
acaff495 | 2585 | if ms: |
19a03940 | 2586 | ms = ms.replace(':', '.') |
2587 | return sum(float(part or 0) * mult for part, mult in ( | |
2588 | (days, 86400), (hours, 3600), (mins, 60), (secs, 1), (ms, 1))) | |
91d7d0b3 JMF |
2589 | |
2590 | ||
e65e4c88 | 2591 | def prepend_extension(filename, ext, expected_real_ext=None): |
5f6a1245 | 2592 | name, real_ext = os.path.splitext(filename) |
e65e4c88 | 2593 | return ( |
86e5f3ed | 2594 | f'{name}.{ext}{real_ext}' |
e65e4c88 | 2595 | if not expected_real_ext or real_ext[1:] == expected_real_ext |
86e5f3ed | 2596 | else f'{filename}.{ext}') |
d70ad093 PH |
2597 | |
2598 | ||
b3ed15b7 S |
2599 | def replace_extension(filename, ext, expected_real_ext=None): |
2600 | name, real_ext = os.path.splitext(filename) | |
86e5f3ed | 2601 | return '{}.{}'.format( |
b3ed15b7 S |
2602 | name if not expected_real_ext or real_ext[1:] == expected_real_ext else filename, |
2603 | ext) | |
2604 | ||
2605 | ||
d70ad093 PH |
2606 | def check_executable(exe, args=[]): |
2607 | """ Checks if the given binary is installed somewhere in PATH, and returns its name. | |
2608 | args can be a list of arguments for a short output (like -version) """ | |
2609 | try: | |
f0c9fb96 | 2610 | Popen.run([exe] + args, stdout=subprocess.PIPE, stderr=subprocess.PIPE) |
d70ad093 PH |
2611 | except OSError: |
2612 | return False | |
2613 | return exe | |
b7ab0590 PH |
2614 | |
2615 | ||
8a7f68d0 | 2616 | def _get_exe_version_output(exe, args, *, to_screen=None): |
2617 | if to_screen: | |
2618 | to_screen(f'Checking exe version: {shell_quote([exe] + args)}') | |
95807118 | 2619 | try: |
b64d04c1 | 2620 | # STDIN should be redirected too. On UNIX-like systems, ffmpeg triggers |
7a5c1cfe | 2621 | # SIGTTOU if yt-dlp is run in the background. |
067aa17e | 2622 | # See https://github.com/ytdl-org/youtube-dl/issues/955#issuecomment-209789656 |
f0c9fb96 | 2623 | stdout, _, _ = Popen.run([encodeArgument(exe)] + args, text=True, |
2624 | stdin=subprocess.PIPE, stdout=subprocess.PIPE, stderr=subprocess.STDOUT) | |
95807118 PH |
2625 | except OSError: |
2626 | return False | |
f0c9fb96 | 2627 | return stdout |
cae97f65 PH |
2628 | |
2629 | ||
2630 | def detect_exe_version(output, version_re=None, unrecognized='present'): | |
14f25df2 | 2631 | assert isinstance(output, str) |
cae97f65 PH |
2632 | if version_re is None: |
2633 | version_re = r'version\s+([-0-9._a-zA-Z]+)' | |
2634 | m = re.search(version_re, output) | |
95807118 PH |
2635 | if m: |
2636 | return m.group(1) | |
2637 | else: | |
2638 | return unrecognized | |
2639 | ||
2640 | ||
9af98e17 | 2641 | def get_exe_version(exe, args=['--version'], |
2642 | version_re=None, unrecognized='present'): | |
2643 | """ Returns the version of the specified executable, | |
2644 | or False if the executable is not present """ | |
2645 | out = _get_exe_version_output(exe, args) | |
2646 | return detect_exe_version(out, version_re, unrecognized) if out else False | |
2647 | ||
2648 | ||
7e88d7d7 | 2649 | def frange(start=0, stop=None, step=1): |
2650 | """Float range""" | |
2651 | if stop is None: | |
2652 | start, stop = 0, start | |
2653 | sign = [-1, 1][step > 0] if step else 0 | |
2654 | while sign * start < sign * stop: | |
2655 | yield start | |
2656 | start += step | |
2657 | ||
2658 | ||
cb89cfc1 | 2659 | class LazyList(collections.abc.Sequence): |
0f06bcd7 | 2660 | """Lazy immutable list from an iterable |
2661 | Note that slices of a LazyList are lists and not LazyList""" | |
483336e7 | 2662 | |
8e5fecc8 | 2663 | class IndexError(IndexError): |
2664 | pass | |
2665 | ||
282f5709 | 2666 | def __init__(self, iterable, *, reverse=False, _cache=None): |
0f06bcd7 | 2667 | self._iterable = iter(iterable) |
2668 | self._cache = [] if _cache is None else _cache | |
2669 | self._reversed = reverse | |
483336e7 | 2670 | |
2671 | def __iter__(self): | |
0f06bcd7 | 2672 | if self._reversed: |
28419ca2 | 2673 | # We need to consume the entire iterable to iterate in reverse |
981052c9 | 2674 | yield from self.exhaust() |
28419ca2 | 2675 | return |
0f06bcd7 | 2676 | yield from self._cache |
2677 | for item in self._iterable: | |
2678 | self._cache.append(item) | |
483336e7 | 2679 | yield item |
2680 | ||
0f06bcd7 | 2681 | def _exhaust(self): |
2682 | self._cache.extend(self._iterable) | |
2683 | self._iterable = [] # Discard the emptied iterable to make it pickle-able | |
2684 | return self._cache | |
28419ca2 | 2685 | |
981052c9 | 2686 | def exhaust(self): |
0f06bcd7 | 2687 | """Evaluate the entire iterable""" |
2688 | return self._exhaust()[::-1 if self._reversed else 1] | |
981052c9 | 2689 | |
28419ca2 | 2690 | @staticmethod |
0f06bcd7 | 2691 | def _reverse_index(x): |
f2df4071 | 2692 | return None if x is None else ~x |
483336e7 | 2693 | |
2694 | def __getitem__(self, idx): | |
2695 | if isinstance(idx, slice): | |
0f06bcd7 | 2696 | if self._reversed: |
2697 | idx = slice(self._reverse_index(idx.start), self._reverse_index(idx.stop), -(idx.step or 1)) | |
e0f2b4b4 | 2698 | start, stop, step = idx.start, idx.stop, idx.step or 1 |
483336e7 | 2699 | elif isinstance(idx, int): |
0f06bcd7 | 2700 | if self._reversed: |
2701 | idx = self._reverse_index(idx) | |
e0f2b4b4 | 2702 | start, stop, step = idx, idx, 0 |
483336e7 | 2703 | else: |
2704 | raise TypeError('indices must be integers or slices') | |
e0f2b4b4 | 2705 | if ((start or 0) < 0 or (stop or 0) < 0 |
2706 | or (start is None and step < 0) | |
2707 | or (stop is None and step > 0)): | |
483336e7 | 2708 | # We need to consume the entire iterable to be able to slice from the end |
2709 | # Obviously, never use this with infinite iterables | |
0f06bcd7 | 2710 | self._exhaust() |
8e5fecc8 | 2711 | try: |
0f06bcd7 | 2712 | return self._cache[idx] |
8e5fecc8 | 2713 | except IndexError as e: |
2714 | raise self.IndexError(e) from e | |
0f06bcd7 | 2715 | n = max(start or 0, stop or 0) - len(self._cache) + 1 |
28419ca2 | 2716 | if n > 0: |
0f06bcd7 | 2717 | self._cache.extend(itertools.islice(self._iterable, n)) |
8e5fecc8 | 2718 | try: |
0f06bcd7 | 2719 | return self._cache[idx] |
8e5fecc8 | 2720 | except IndexError as e: |
2721 | raise self.IndexError(e) from e | |
483336e7 | 2722 | |
2723 | def __bool__(self): | |
2724 | try: | |
0f06bcd7 | 2725 | self[-1] if self._reversed else self[0] |
8e5fecc8 | 2726 | except self.IndexError: |
483336e7 | 2727 | return False |
2728 | return True | |
2729 | ||
2730 | def __len__(self): | |
0f06bcd7 | 2731 | self._exhaust() |
2732 | return len(self._cache) | |
483336e7 | 2733 | |
282f5709 | 2734 | def __reversed__(self): |
0f06bcd7 | 2735 | return type(self)(self._iterable, reverse=not self._reversed, _cache=self._cache) |
282f5709 | 2736 | |
2737 | def __copy__(self): | |
0f06bcd7 | 2738 | return type(self)(self._iterable, reverse=self._reversed, _cache=self._cache) |
282f5709 | 2739 | |
28419ca2 | 2740 | def __repr__(self): |
2741 | # repr and str should mimic a list. So we exhaust the iterable | |
2742 | return repr(self.exhaust()) | |
2743 | ||
2744 | def __str__(self): | |
2745 | return repr(self.exhaust()) | |
2746 | ||
483336e7 | 2747 | |
7be9ccff | 2748 | class PagedList: |
c07a39ae | 2749 | |
2750 | class IndexError(IndexError): | |
2751 | pass | |
2752 | ||
dd26ced1 PH |
2753 | def __len__(self): |
2754 | # This is only useful for tests | |
2755 | return len(self.getslice()) | |
2756 | ||
7be9ccff | 2757 | def __init__(self, pagefunc, pagesize, use_cache=True): |
2758 | self._pagefunc = pagefunc | |
2759 | self._pagesize = pagesize | |
f1d13090 | 2760 | self._pagecount = float('inf') |
7be9ccff | 2761 | self._use_cache = use_cache |
2762 | self._cache = {} | |
2763 | ||
2764 | def getpage(self, pagenum): | |
d8cf8d97 | 2765 | page_results = self._cache.get(pagenum) |
2766 | if page_results is None: | |
f1d13090 | 2767 | page_results = [] if pagenum > self._pagecount else list(self._pagefunc(pagenum)) |
7be9ccff | 2768 | if self._use_cache: |
2769 | self._cache[pagenum] = page_results | |
2770 | return page_results | |
2771 | ||
2772 | def getslice(self, start=0, end=None): | |
2773 | return list(self._getslice(start, end)) | |
2774 | ||
2775 | def _getslice(self, start, end): | |
55575225 | 2776 | raise NotImplementedError('This method must be implemented by subclasses') |
2777 | ||
2778 | def __getitem__(self, idx): | |
f1d13090 | 2779 | assert self._use_cache, 'Indexing PagedList requires cache' |
55575225 | 2780 | if not isinstance(idx, int) or idx < 0: |
2781 | raise TypeError('indices must be non-negative integers') | |
2782 | entries = self.getslice(idx, idx + 1) | |
d8cf8d97 | 2783 | if not entries: |
c07a39ae | 2784 | raise self.IndexError() |
d8cf8d97 | 2785 | return entries[0] |
55575225 | 2786 | |
9c44d242 PH |
2787 | |
2788 | class OnDemandPagedList(PagedList): | |
a44ca5a4 | 2789 | """Download pages until a page with less than maximum results""" |
86e5f3ed | 2790 | |
7be9ccff | 2791 | def _getslice(self, start, end): |
b7ab0590 PH |
2792 | for pagenum in itertools.count(start // self._pagesize): |
2793 | firstid = pagenum * self._pagesize | |
2794 | nextfirstid = pagenum * self._pagesize + self._pagesize | |
2795 | if start >= nextfirstid: | |
2796 | continue | |
2797 | ||
b7ab0590 PH |
2798 | startv = ( |
2799 | start % self._pagesize | |
2800 | if firstid <= start < nextfirstid | |
2801 | else 0) | |
b7ab0590 PH |
2802 | endv = ( |
2803 | ((end - 1) % self._pagesize) + 1 | |
2804 | if (end is not None and firstid <= end <= nextfirstid) | |
2805 | else None) | |
2806 | ||
f1d13090 | 2807 | try: |
2808 | page_results = self.getpage(pagenum) | |
2809 | except Exception: | |
2810 | self._pagecount = pagenum - 1 | |
2811 | raise | |
b7ab0590 PH |
2812 | if startv != 0 or endv is not None: |
2813 | page_results = page_results[startv:endv] | |
7be9ccff | 2814 | yield from page_results |
b7ab0590 PH |
2815 | |
2816 | # A little optimization - if current page is not "full", ie. does | |
2817 | # not contain page_size videos then we can assume that this page | |
2818 | # is the last one - there are no more ids on further pages - | |
2819 | # i.e. no need to query again. | |
2820 | if len(page_results) + startv < self._pagesize: | |
2821 | break | |
2822 | ||
2823 | # If we got the whole page, but the next page is not interesting, | |
2824 | # break out early as well | |
2825 | if end == nextfirstid: | |
2826 | break | |
81c2f20b PH |
2827 | |
2828 | ||
9c44d242 | 2829 | class InAdvancePagedList(PagedList): |
a44ca5a4 | 2830 | """PagedList with total number of pages known in advance""" |
86e5f3ed | 2831 | |
9c44d242 | 2832 | def __init__(self, pagefunc, pagecount, pagesize): |
7be9ccff | 2833 | PagedList.__init__(self, pagefunc, pagesize, True) |
f1d13090 | 2834 | self._pagecount = pagecount |
9c44d242 | 2835 | |
7be9ccff | 2836 | def _getslice(self, start, end): |
9c44d242 | 2837 | start_page = start // self._pagesize |
d37707bd | 2838 | end_page = self._pagecount if end is None else min(self._pagecount, end // self._pagesize + 1) |
9c44d242 PH |
2839 | skip_elems = start - start_page * self._pagesize |
2840 | only_more = None if end is None else end - start | |
2841 | for pagenum in range(start_page, end_page): | |
7be9ccff | 2842 | page_results = self.getpage(pagenum) |
9c44d242 | 2843 | if skip_elems: |
7be9ccff | 2844 | page_results = page_results[skip_elems:] |
9c44d242 PH |
2845 | skip_elems = None |
2846 | if only_more is not None: | |
7be9ccff | 2847 | if len(page_results) < only_more: |
2848 | only_more -= len(page_results) | |
9c44d242 | 2849 | else: |
7be9ccff | 2850 | yield from page_results[:only_more] |
9c44d242 | 2851 | break |
7be9ccff | 2852 | yield from page_results |
9c44d242 PH |
2853 | |
2854 | ||
7e88d7d7 | 2855 | class PlaylistEntries: |
2856 | MissingEntry = object() | |
2857 | is_exhausted = False | |
2858 | ||
2859 | def __init__(self, ydl, info_dict): | |
7e9a6125 | 2860 | self.ydl = ydl |
2861 | ||
2862 | # _entries must be assigned now since infodict can change during iteration | |
2863 | entries = info_dict.get('entries') | |
2864 | if entries is None: | |
2865 | raise EntryNotInPlaylist('There are no entries') | |
2866 | elif isinstance(entries, list): | |
2867 | self.is_exhausted = True | |
2868 | ||
2869 | requested_entries = info_dict.get('requested_entries') | |
2870 | self.is_incomplete = bool(requested_entries) | |
2871 | if self.is_incomplete: | |
2872 | assert self.is_exhausted | |
2873 | self._entries = [self.MissingEntry] * max(requested_entries) | |
2874 | for i, entry in zip(requested_entries, entries): | |
2875 | self._entries[i - 1] = entry | |
2876 | elif isinstance(entries, (list, PagedList, LazyList)): | |
2877 | self._entries = entries | |
2878 | else: | |
2879 | self._entries = LazyList(entries) | |
7e88d7d7 | 2880 | |
2881 | PLAYLIST_ITEMS_RE = re.compile(r'''(?x) | |
2882 | (?P<start>[+-]?\d+)? | |
2883 | (?P<range>[:-] | |
2884 | (?P<end>[+-]?\d+|inf(?:inite)?)? | |
2885 | (?::(?P<step>[+-]?\d+))? | |
2886 | )?''') | |
2887 | ||
2888 | @classmethod | |
2889 | def parse_playlist_items(cls, string): | |
2890 | for segment in string.split(','): | |
2891 | if not segment: | |
2892 | raise ValueError('There is two or more consecutive commas') | |
2893 | mobj = cls.PLAYLIST_ITEMS_RE.fullmatch(segment) | |
2894 | if not mobj: | |
2895 | raise ValueError(f'{segment!r} is not a valid specification') | |
2896 | start, end, step, has_range = mobj.group('start', 'end', 'step', 'range') | |
2897 | if int_or_none(step) == 0: | |
2898 | raise ValueError(f'Step in {segment!r} cannot be zero') | |
2899 | yield slice(int_or_none(start), float_or_none(end), int_or_none(step)) if has_range else int(start) | |
2900 | ||
2901 | def get_requested_items(self): | |
2902 | playlist_items = self.ydl.params.get('playlist_items') | |
2903 | playlist_start = self.ydl.params.get('playliststart', 1) | |
2904 | playlist_end = self.ydl.params.get('playlistend') | |
2905 | # For backwards compatibility, interpret -1 as whole list | |
2906 | if playlist_end in (-1, None): | |
2907 | playlist_end = '' | |
2908 | if not playlist_items: | |
2909 | playlist_items = f'{playlist_start}:{playlist_end}' | |
2910 | elif playlist_start != 1 or playlist_end: | |
2911 | self.ydl.report_warning('Ignoring playliststart and playlistend because playlistitems was given', only_once=True) | |
2912 | ||
2913 | for index in self.parse_playlist_items(playlist_items): | |
2914 | for i, entry in self[index]: | |
2915 | yield i, entry | |
1ac4fd80 | 2916 | if not entry: |
2917 | continue | |
7e88d7d7 | 2918 | try: |
2919 | # TODO: Add auto-generated fields | |
2920 | self.ydl._match_entry(entry, incomplete=True, silent=True) | |
2921 | except (ExistingVideoReached, RejectedVideoReached): | |
2922 | return | |
2923 | ||
7e9a6125 | 2924 | def get_full_count(self): |
2925 | if self.is_exhausted and not self.is_incomplete: | |
7e88d7d7 | 2926 | return len(self) |
2927 | elif isinstance(self._entries, InAdvancePagedList): | |
2928 | if self._entries._pagesize == 1: | |
2929 | return self._entries._pagecount | |
2930 | ||
7e88d7d7 | 2931 | @functools.cached_property |
2932 | def _getter(self): | |
2933 | if isinstance(self._entries, list): | |
2934 | def get_entry(i): | |
2935 | try: | |
2936 | entry = self._entries[i] | |
2937 | except IndexError: | |
2938 | entry = self.MissingEntry | |
2939 | if not self.is_incomplete: | |
2940 | raise self.IndexError() | |
2941 | if entry is self.MissingEntry: | |
2942 | raise EntryNotInPlaylist(f'Entry {i} cannot be found') | |
2943 | return entry | |
2944 | else: | |
2945 | def get_entry(i): | |
2946 | try: | |
2947 | return type(self.ydl)._handle_extraction_exceptions(lambda _, i: self._entries[i])(self.ydl, i) | |
2948 | except (LazyList.IndexError, PagedList.IndexError): | |
2949 | raise self.IndexError() | |
2950 | return get_entry | |
2951 | ||
2952 | def __getitem__(self, idx): | |
2953 | if isinstance(idx, int): | |
2954 | idx = slice(idx, idx) | |
2955 | ||
2956 | # NB: PlaylistEntries[1:10] => (0, 1, ... 9) | |
2957 | step = 1 if idx.step is None else idx.step | |
2958 | if idx.start is None: | |
2959 | start = 0 if step > 0 else len(self) - 1 | |
2960 | else: | |
2961 | start = idx.start - 1 if idx.start >= 0 else len(self) + idx.start | |
2962 | ||
2963 | # NB: Do not call len(self) when idx == [:] | |
2964 | if idx.stop is None: | |
2965 | stop = 0 if step < 0 else float('inf') | |
2966 | else: | |
2967 | stop = idx.stop - 1 if idx.stop >= 0 else len(self) + idx.stop | |
2968 | stop += [-1, 1][step > 0] | |
2969 | ||
2970 | for i in frange(start, stop, step): | |
2971 | if i < 0: | |
2972 | continue | |
2973 | try: | |
7e9a6125 | 2974 | entry = self._getter(i) |
2975 | except self.IndexError: | |
2976 | self.is_exhausted = True | |
2977 | if step > 0: | |
7e88d7d7 | 2978 | break |
7e9a6125 | 2979 | continue |
7e88d7d7 | 2980 | yield i + 1, entry |
2981 | ||
2982 | def __len__(self): | |
2983 | return len(tuple(self[:])) | |
2984 | ||
2985 | class IndexError(IndexError): | |
2986 | pass | |
2987 | ||
2988 | ||
81c2f20b | 2989 | def uppercase_escape(s): |
676eb3f2 | 2990 | unicode_escape = codecs.getdecoder('unicode_escape') |
81c2f20b | 2991 | return re.sub( |
a612753d | 2992 | r'\\U[0-9a-fA-F]{8}', |
676eb3f2 PH |
2993 | lambda m: unicode_escape(m.group(0))[0], |
2994 | s) | |
0fe2ff78 YCH |
2995 | |
2996 | ||
2997 | def lowercase_escape(s): | |
2998 | unicode_escape = codecs.getdecoder('unicode_escape') | |
2999 | return re.sub( | |
3000 | r'\\u[0-9a-fA-F]{4}', | |
3001 | lambda m: unicode_escape(m.group(0))[0], | |
3002 | s) | |
b53466e1 | 3003 | |
d05cfe06 S |
3004 | |
3005 | def escape_rfc3986(s): | |
3006 | """Escape non-ASCII characters as suggested by RFC 3986""" | |
f9934b96 | 3007 | return urllib.parse.quote(s, b"%/;:@&=+$,!~*'()?#[]") |
d05cfe06 S |
3008 | |
3009 | ||
3010 | def escape_url(url): | |
3011 | """Escape URL as suggested by RFC 3986""" | |
14f25df2 | 3012 | url_parsed = urllib.parse.urlparse(url) |
d05cfe06 | 3013 | return url_parsed._replace( |
efbed08d | 3014 | netloc=url_parsed.netloc.encode('idna').decode('ascii'), |
d05cfe06 S |
3015 | path=escape_rfc3986(url_parsed.path), |
3016 | params=escape_rfc3986(url_parsed.params), | |
3017 | query=escape_rfc3986(url_parsed.query), | |
3018 | fragment=escape_rfc3986(url_parsed.fragment) | |
3019 | ).geturl() | |
3020 | ||
62e609ab | 3021 | |
4dfbf869 | 3022 | def parse_qs(url): |
14f25df2 | 3023 | return urllib.parse.parse_qs(urllib.parse.urlparse(url).query) |
4dfbf869 | 3024 | |
3025 | ||
62e609ab PH |
3026 | def read_batch_urls(batch_fd): |
3027 | def fixup(url): | |
14f25df2 | 3028 | if not isinstance(url, str): |
62e609ab | 3029 | url = url.decode('utf-8', 'replace') |
8c04f0be | 3030 | BOM_UTF8 = ('\xef\xbb\xbf', '\ufeff') |
3031 | for bom in BOM_UTF8: | |
3032 | if url.startswith(bom): | |
3033 | url = url[len(bom):] | |
3034 | url = url.lstrip() | |
3035 | if not url or url.startswith(('#', ';', ']')): | |
62e609ab | 3036 | return False |
8c04f0be | 3037 | # "#" cannot be stripped out since it is part of the URI |
962ffcf8 | 3038 | # However, it can be safely stripped out if following a whitespace |
8c04f0be | 3039 | return re.split(r'\s#', url, 1)[0].rstrip() |
62e609ab PH |
3040 | |
3041 | with contextlib.closing(batch_fd) as fd: | |
3042 | return [url for url in map(fixup, fd) if url] | |
b74fa8cd JMF |
3043 | |
3044 | ||
3045 | def urlencode_postdata(*args, **kargs): | |
14f25df2 | 3046 | return urllib.parse.urlencode(*args, **kargs).encode('ascii') |
bcf89ce6 PH |
3047 | |
3048 | ||
38f9ef31 | 3049 | def update_url_query(url, query): |
cacd9966 YCH |
3050 | if not query: |
3051 | return url | |
14f25df2 | 3052 | parsed_url = urllib.parse.urlparse(url) |
3053 | qs = urllib.parse.parse_qs(parsed_url.query) | |
38f9ef31 | 3054 | qs.update(query) |
14f25df2 | 3055 | return urllib.parse.urlunparse(parsed_url._replace( |
3056 | query=urllib.parse.urlencode(qs, True))) | |
16392824 | 3057 | |
8e60dc75 | 3058 | |
c043c246 | 3059 | def update_Request(req, url=None, data=None, headers=None, query=None): |
ed0291d1 | 3060 | req_headers = req.headers.copy() |
c043c246 | 3061 | req_headers.update(headers or {}) |
ed0291d1 S |
3062 | req_data = data or req.data |
3063 | req_url = update_url_query(url or req.get_full_url(), query) | |
95cf60e8 S |
3064 | req_get_method = req.get_method() |
3065 | if req_get_method == 'HEAD': | |
3066 | req_type = HEADRequest | |
3067 | elif req_get_method == 'PUT': | |
3068 | req_type = PUTRequest | |
3069 | else: | |
ac668111 | 3070 | req_type = urllib.request.Request |
ed0291d1 S |
3071 | new_req = req_type( |
3072 | req_url, data=req_data, headers=req_headers, | |
3073 | origin_req_host=req.origin_req_host, unverifiable=req.unverifiable) | |
3074 | if hasattr(req, 'timeout'): | |
3075 | new_req.timeout = req.timeout | |
3076 | return new_req | |
3077 | ||
3078 | ||
10c87c15 | 3079 | def _multipart_encode_impl(data, boundary): |
0c265486 YCH |
3080 | content_type = 'multipart/form-data; boundary=%s' % boundary |
3081 | ||
3082 | out = b'' | |
3083 | for k, v in data.items(): | |
3084 | out += b'--' + boundary.encode('ascii') + b'\r\n' | |
14f25df2 | 3085 | if isinstance(k, str): |
0f06bcd7 | 3086 | k = k.encode() |
14f25df2 | 3087 | if isinstance(v, str): |
0f06bcd7 | 3088 | v = v.encode() |
0c265486 YCH |
3089 | # RFC 2047 requires non-ASCII field names to be encoded, while RFC 7578 |
3090 | # suggests sending UTF-8 directly. Firefox sends UTF-8, too | |
b2ad479d | 3091 | content = b'Content-Disposition: form-data; name="' + k + b'"\r\n\r\n' + v + b'\r\n' |
0c265486 YCH |
3092 | if boundary.encode('ascii') in content: |
3093 | raise ValueError('Boundary overlaps with data') | |
3094 | out += content | |
3095 | ||
3096 | out += b'--' + boundary.encode('ascii') + b'--\r\n' | |
3097 | ||
3098 | return out, content_type | |
3099 | ||
3100 | ||
3101 | def multipart_encode(data, boundary=None): | |
3102 | ''' | |
3103 | Encode a dict to RFC 7578-compliant form-data | |
3104 | ||
3105 | data: | |
3106 | A dict where keys and values can be either Unicode or bytes-like | |
3107 | objects. | |
3108 | boundary: | |
3109 | If specified a Unicode object, it's used as the boundary. Otherwise | |
3110 | a random boundary is generated. | |
3111 | ||
3112 | Reference: https://tools.ietf.org/html/rfc7578 | |
3113 | ''' | |
3114 | has_specified_boundary = boundary is not None | |
3115 | ||
3116 | while True: | |
3117 | if boundary is None: | |
3118 | boundary = '---------------' + str(random.randrange(0x0fffffff, 0xffffffff)) | |
3119 | ||
3120 | try: | |
10c87c15 | 3121 | out, content_type = _multipart_encode_impl(data, boundary) |
0c265486 YCH |
3122 | break |
3123 | except ValueError: | |
3124 | if has_specified_boundary: | |
3125 | raise | |
3126 | boundary = None | |
3127 | ||
3128 | return out, content_type | |
3129 | ||
3130 | ||
86296ad2 | 3131 | def dict_get(d, key_or_keys, default=None, skip_false_values=True): |
a44ca5a4 | 3132 | for val in map(d.get, variadic(key_or_keys)): |
3133 | if val is not None and (val or not skip_false_values): | |
3134 | return val | |
3135 | return default | |
cbecc9b9 S |
3136 | |
3137 | ||
c4f60dd7 | 3138 | def try_call(*funcs, expected_type=None, args=[], kwargs={}): |
3139 | for f in funcs: | |
a32a9a7e | 3140 | try: |
c4f60dd7 | 3141 | val = f(*args, **kwargs) |
3142 | except (AttributeError, KeyError, TypeError, IndexError, ZeroDivisionError): | |
a32a9a7e S |
3143 | pass |
3144 | else: | |
c4f60dd7 | 3145 | if expected_type is None or isinstance(val, expected_type): |
3146 | return val | |
3147 | ||
3148 | ||
3149 | def try_get(src, getter, expected_type=None): | |
3150 | return try_call(*variadic(getter), args=(src,), expected_type=expected_type) | |
329ca3be S |
3151 | |
3152 | ||
90137ca4 | 3153 | def filter_dict(dct, cndn=lambda _, v: v is not None): |
3154 | return {k: v for k, v in dct.items() if cndn(k, v)} | |
3155 | ||
3156 | ||
6cc62232 S |
3157 | def merge_dicts(*dicts): |
3158 | merged = {} | |
3159 | for a_dict in dicts: | |
3160 | for k, v in a_dict.items(): | |
90137ca4 | 3161 | if (v is not None and k not in merged |
3162 | or isinstance(v, str) and merged[k] == ''): | |
6cc62232 S |
3163 | merged[k] = v |
3164 | return merged | |
3165 | ||
3166 | ||
8e60dc75 | 3167 | def encode_compat_str(string, encoding=preferredencoding(), errors='strict'): |
14f25df2 | 3168 | return string if isinstance(string, str) else str(string, encoding, errors) |
8e60dc75 | 3169 | |
16392824 | 3170 | |
a1a530b0 PH |
3171 | US_RATINGS = { |
3172 | 'G': 0, | |
3173 | 'PG': 10, | |
3174 | 'PG-13': 13, | |
3175 | 'R': 16, | |
3176 | 'NC': 18, | |
3177 | } | |
fac55558 PH |
3178 | |
3179 | ||
a8795327 | 3180 | TV_PARENTAL_GUIDELINES = { |
5a16c9d9 RA |
3181 | 'TV-Y': 0, |
3182 | 'TV-Y7': 7, | |
3183 | 'TV-G': 0, | |
3184 | 'TV-PG': 0, | |
3185 | 'TV-14': 14, | |
3186 | 'TV-MA': 17, | |
a8795327 S |
3187 | } |
3188 | ||
3189 | ||
146c80e2 | 3190 | def parse_age_limit(s): |
19a03940 | 3191 | # isinstance(False, int) is True. So type() must be used instead |
c487cf00 | 3192 | if type(s) is int: # noqa: E721 |
a8795327 | 3193 | return s if 0 <= s <= 21 else None |
19a03940 | 3194 | elif not isinstance(s, str): |
d838b1bd | 3195 | return None |
146c80e2 | 3196 | m = re.match(r'^(?P<age>\d{1,2})\+?$', s) |
a8795327 S |
3197 | if m: |
3198 | return int(m.group('age')) | |
5c5fae6d | 3199 | s = s.upper() |
a8795327 S |
3200 | if s in US_RATINGS: |
3201 | return US_RATINGS[s] | |
5a16c9d9 | 3202 | m = re.match(r'^TV[_-]?(%s)$' % '|'.join(k[3:] for k in TV_PARENTAL_GUIDELINES), s) |
b8361187 | 3203 | if m: |
5a16c9d9 | 3204 | return TV_PARENTAL_GUIDELINES['TV-' + m.group(1)] |
b8361187 | 3205 | return None |
146c80e2 S |
3206 | |
3207 | ||
fac55558 | 3208 | def strip_jsonp(code): |
609a61e3 | 3209 | return re.sub( |
5552c9eb | 3210 | r'''(?sx)^ |
e9c671d5 | 3211 | (?:window\.)?(?P<func_name>[a-zA-Z0-9_.$]*) |
5552c9eb YCH |
3212 | (?:\s*&&\s*(?P=func_name))? |
3213 | \s*\(\s*(?P<callback_data>.*)\);? | |
3214 | \s*?(?://[^\n]*)*$''', | |
3215 | r'\g<callback_data>', code) | |
478c2c61 PH |
3216 | |
3217 | ||
8f53dc44 | 3218 | def js_to_json(code, vars={}, *, strict=False): |
5c610515 | 3219 | # vars is a dict of var, val pairs to substitute |
c843e685 | 3220 | COMMENT_RE = r'/\*(?:(?!\*/).)*?\*/|//[^\n]*\n' |
86e5f3ed | 3221 | SKIP_RE = fr'\s*(?:{COMMENT_RE})?\s*' |
4195096e | 3222 | INTEGER_TABLE = ( |
86e5f3ed | 3223 | (fr'(?s)^(0[xX][0-9a-fA-F]+){SKIP_RE}:?$', 16), |
3224 | (fr'(?s)^(0+[0-7]+){SKIP_RE}:?$', 8), | |
4195096e S |
3225 | ) |
3226 | ||
e05f6939 | 3227 | def fix_kv(m): |
e7b6d122 PH |
3228 | v = m.group(0) |
3229 | if v in ('true', 'false', 'null'): | |
3230 | return v | |
421ddcb8 C |
3231 | elif v in ('undefined', 'void 0'): |
3232 | return 'null' | |
8bdd16b4 | 3233 | elif v.startswith('/*') or v.startswith('//') or v.startswith('!') or v == ',': |
bd1e4844 | 3234 | return "" |
3235 | ||
3236 | if v[0] in ("'", '"'): | |
3237 | v = re.sub(r'(?s)\\.|"', lambda m: { | |
e7b6d122 | 3238 | '"': '\\"', |
bd1e4844 | 3239 | "\\'": "'", |
3240 | '\\\n': '', | |
3241 | '\\x': '\\u00', | |
3242 | }.get(m.group(0), m.group(0)), v[1:-1]) | |
8bdd16b4 | 3243 | else: |
3244 | for regex, base in INTEGER_TABLE: | |
3245 | im = re.match(regex, v) | |
3246 | if im: | |
3247 | i = int(im.group(1), base) | |
3248 | return '"%d":' % i if v.endswith(':') else '%d' % i | |
89ac4a19 | 3249 | |
5c610515 | 3250 | if v in vars: |
3251 | return vars[v] | |
8f53dc44 | 3252 | if strict: |
3253 | raise ValueError(f'Unknown value: {v}') | |
5c610515 | 3254 | |
e7b6d122 | 3255 | return '"%s"' % v |
e05f6939 | 3256 | |
8072ef2b | 3257 | def create_map(mobj): |
3258 | return json.dumps(dict(json.loads(js_to_json(mobj.group(1) or '[]', vars=vars)))) | |
3259 | ||
8072ef2b | 3260 | code = re.sub(r'new Map\((\[.*?\])?\)', create_map, code) |
8f53dc44 | 3261 | if not strict: |
3262 | code = re.sub(r'new Date\((".+")\)', r'\g<1>', code) | |
febff4c1 | 3263 | |
bd1e4844 | 3264 | return re.sub(r'''(?sx) |
3265 | "(?:[^"\\]*(?:\\\\|\\['"nurtbfx/\n]))*[^"\\]*"| | |
3266 | '(?:[^'\\]*(?:\\\\|\\['"nurtbfx/\n]))*[^'\\]*'| | |
4195096e | 3267 | {comment}|,(?={skip}[\]}}])| |
421ddcb8 | 3268 | void\s0|(?:(?<![0-9])[eE]|[a-df-zA-DF-Z_$])[.a-zA-Z_$0-9]*| |
4195096e | 3269 | \b(?:0[xX][0-9a-fA-F]+|0+[0-7]+)(?:{skip}:)?| |
8bdd16b4 | 3270 | [0-9]+(?={skip}:)| |
3271 | !+ | |
4195096e | 3272 | '''.format(comment=COMMENT_RE, skip=SKIP_RE), fix_kv, code) |
e05f6939 PH |
3273 | |
3274 | ||
478c2c61 PH |
3275 | def qualities(quality_ids): |
3276 | """ Get a numeric quality value out of a list of possible values """ | |
3277 | def q(qid): | |
3278 | try: | |
3279 | return quality_ids.index(qid) | |
3280 | except ValueError: | |
3281 | return -1 | |
3282 | return q | |
3283 | ||
acd69589 | 3284 | |
8aa0e7cd | 3285 | POSTPROCESS_WHEN = ('pre_process', 'after_filter', 'before_dl', 'post_process', 'after_move', 'after_video', 'playlist') |
1e43a6f7 | 3286 | |
3287 | ||
de6000d9 | 3288 | DEFAULT_OUTTMPL = { |
3289 | 'default': '%(title)s [%(id)s].%(ext)s', | |
72755351 | 3290 | 'chapter': '%(title)s - %(section_number)03d %(section_title)s [%(id)s].%(ext)s', |
de6000d9 | 3291 | } |
3292 | OUTTMPL_TYPES = { | |
72755351 | 3293 | 'chapter': None, |
de6000d9 | 3294 | 'subtitle': None, |
3295 | 'thumbnail': None, | |
3296 | 'description': 'description', | |
3297 | 'annotation': 'annotations.xml', | |
3298 | 'infojson': 'info.json', | |
08438d2c | 3299 | 'link': None, |
3b603dbd | 3300 | 'pl_video': None, |
5112f26a | 3301 | 'pl_thumbnail': None, |
de6000d9 | 3302 | 'pl_description': 'description', |
3303 | 'pl_infojson': 'info.json', | |
3304 | } | |
0a871f68 | 3305 | |
143db31d | 3306 | # As of [1] format syntax is: |
3307 | # %[mapping_key][conversion_flags][minimum_width][.precision][length_modifier]type | |
3308 | # 1. https://docs.python.org/2/library/stdtypes.html#string-formatting | |
901130bb | 3309 | STR_FORMAT_RE_TMPL = r'''(?x) |
3310 | (?<!%)(?P<prefix>(?:%%)*) | |
143db31d | 3311 | % |
524e2e4f | 3312 | (?P<has_key>\((?P<key>{0})\))? |
752cda38 | 3313 | (?P<format> |
524e2e4f | 3314 | (?P<conversion>[#0\-+ ]+)? |
3315 | (?P<min_width>\d+)? | |
3316 | (?P<precision>\.\d+)? | |
3317 | (?P<len_mod>[hlL])? # unused in python | |
901130bb | 3318 | {1} # conversion type |
752cda38 | 3319 | ) |
143db31d | 3320 | ''' |
3321 | ||
7d1eb38a | 3322 | |
901130bb | 3323 | STR_FORMAT_TYPES = 'diouxXeEfFgGcrs' |
a020a0dc | 3324 | |
7d1eb38a | 3325 | |
a020a0dc PH |
3326 | def limit_length(s, length): |
3327 | """ Add ellipses to overly long strings """ | |
3328 | if s is None: | |
3329 | return None | |
3330 | ELLIPSES = '...' | |
3331 | if len(s) > length: | |
3332 | return s[:length - len(ELLIPSES)] + ELLIPSES | |
3333 | return s | |
48844745 PH |
3334 | |
3335 | ||
3336 | def version_tuple(v): | |
5f9b8394 | 3337 | return tuple(int(e) for e in re.split(r'[-.]', v)) |
48844745 PH |
3338 | |
3339 | ||
3340 | def is_outdated_version(version, limit, assume_new=True): | |
3341 | if not version: | |
3342 | return not assume_new | |
3343 | try: | |
3344 | return version_tuple(version) < version_tuple(limit) | |
3345 | except ValueError: | |
3346 | return not assume_new | |
732ea2f0 PH |
3347 | |
3348 | ||
3349 | def ytdl_is_updateable(): | |
7a5c1cfe | 3350 | """ Returns if yt-dlp can be updated with -U """ |
735d865e | 3351 | |
5d535b4a | 3352 | from .update import is_non_updateable |
732ea2f0 | 3353 | |
5d535b4a | 3354 | return not is_non_updateable() |
7d4111ed PH |
3355 | |
3356 | ||
3357 | def args_to_str(args): | |
3358 | # Get a short string representation for a subprocess command | |
702ccf2d | 3359 | return ' '.join(compat_shlex_quote(a) for a in args) |
2ccd1b10 PH |
3360 | |
3361 | ||
9b9c5355 | 3362 | def error_to_compat_str(err): |
cfb0511d | 3363 | return str(err) |
fdae2358 S |
3364 | |
3365 | ||
a44ca5a4 | 3366 | def error_to_str(err): |
3367 | return f'{type(err).__name__}: {err}' | |
3368 | ||
3369 | ||
c460bdd5 | 3370 | def mimetype2ext(mt): |
eb9ee194 S |
3371 | if mt is None: |
3372 | return None | |
3373 | ||
9359f3d4 F |
3374 | mt, _, params = mt.partition(';') |
3375 | mt = mt.strip() | |
3376 | ||
3377 | FULL_MAP = { | |
765ac263 | 3378 | 'audio/mp4': 'm4a', |
6c33d24b YCH |
3379 | # Per RFC 3003, audio/mpeg can be .mp1, .mp2 or .mp3. Here use .mp3 as |
3380 | # it's the most popular one | |
3381 | 'audio/mpeg': 'mp3', | |
ba39289d | 3382 | 'audio/x-wav': 'wav', |
9359f3d4 F |
3383 | 'audio/wav': 'wav', |
3384 | 'audio/wave': 'wav', | |
3385 | } | |
3386 | ||
3387 | ext = FULL_MAP.get(mt) | |
765ac263 JMF |
3388 | if ext is not None: |
3389 | return ext | |
3390 | ||
9359f3d4 | 3391 | SUBTYPE_MAP = { |
f6861ec9 | 3392 | '3gpp': '3gp', |
cafcf657 | 3393 | 'smptett+xml': 'tt', |
cafcf657 | 3394 | 'ttaf+xml': 'dfxp', |
a0d8d704 | 3395 | 'ttml+xml': 'ttml', |
f6861ec9 | 3396 | 'x-flv': 'flv', |
a0d8d704 | 3397 | 'x-mp4-fragmented': 'mp4', |
d4f05d47 | 3398 | 'x-ms-sami': 'sami', |
a0d8d704 | 3399 | 'x-ms-wmv': 'wmv', |
b4173f15 RA |
3400 | 'mpegurl': 'm3u8', |
3401 | 'x-mpegurl': 'm3u8', | |
3402 | 'vnd.apple.mpegurl': 'm3u8', | |
3403 | 'dash+xml': 'mpd', | |
b4173f15 | 3404 | 'f4m+xml': 'f4m', |
f164b971 | 3405 | 'hds+xml': 'f4m', |
e910fe2f | 3406 | 'vnd.ms-sstr+xml': 'ism', |
c2b2c7e1 | 3407 | 'quicktime': 'mov', |
98ce1a3f | 3408 | 'mp2t': 'ts', |
39e7107d | 3409 | 'x-wav': 'wav', |
9359f3d4 F |
3410 | 'filmstrip+json': 'fs', |
3411 | 'svg+xml': 'svg', | |
3412 | } | |
3413 | ||
3414 | _, _, subtype = mt.rpartition('/') | |
3415 | ext = SUBTYPE_MAP.get(subtype.lower()) | |
3416 | if ext is not None: | |
3417 | return ext | |
3418 | ||
3419 | SUFFIX_MAP = { | |
3420 | 'json': 'json', | |
3421 | 'xml': 'xml', | |
3422 | 'zip': 'zip', | |
3423 | 'gzip': 'gz', | |
3424 | } | |
3425 | ||
3426 | _, _, suffix = subtype.partition('+') | |
3427 | ext = SUFFIX_MAP.get(suffix) | |
3428 | if ext is not None: | |
3429 | return ext | |
3430 | ||
3431 | return subtype.replace('+', '.') | |
c460bdd5 PH |
3432 | |
3433 | ||
2814f12b THD |
3434 | def ext2mimetype(ext_or_url): |
3435 | if not ext_or_url: | |
3436 | return None | |
3437 | if '.' not in ext_or_url: | |
3438 | ext_or_url = f'file.{ext_or_url}' | |
3439 | return mimetypes.guess_type(ext_or_url)[0] | |
3440 | ||
3441 | ||
4f3c5e06 | 3442 | def parse_codecs(codecs_str): |
3443 | # http://tools.ietf.org/html/rfc6381 | |
3444 | if not codecs_str: | |
3445 | return {} | |
a0566bbf | 3446 | split_codecs = list(filter(None, map( |
dbf5416a | 3447 | str.strip, codecs_str.strip().strip(',').split(',')))) |
3fe75fdc | 3448 | vcodec, acodec, scodec, hdr = None, None, None, None |
a0566bbf | 3449 | for full_codec in split_codecs: |
d816f61f | 3450 | parts = re.sub(r'0+(?=\d)', '', full_codec).split('.') |
3451 | if parts[0] in ('avc1', 'avc2', 'avc3', 'avc4', 'vp9', 'vp8', 'hev1', 'hev2', | |
3452 | 'h263', 'h264', 'mp4v', 'hvc1', 'av1', 'theora', 'dvh1', 'dvhe'): | |
3453 | if vcodec: | |
3454 | continue | |
3455 | vcodec = full_codec | |
3456 | if parts[0] in ('dvh1', 'dvhe'): | |
3457 | hdr = 'DV' | |
3458 | elif parts[0] == 'av1' and traverse_obj(parts, 3) == '10': | |
3459 | hdr = 'HDR10' | |
3460 | elif parts[:2] == ['vp9', '2']: | |
3461 | hdr = 'HDR10' | |
3462 | elif parts[0] in ('flac', 'mp4a', 'opus', 'vorbis', 'mp3', 'aac', | |
3463 | 'ac-3', 'ec-3', 'eac3', 'dtsc', 'dtse', 'dtsh', 'dtsl'): | |
3464 | acodec = acodec or full_codec | |
3465 | elif parts[0] in ('stpp', 'wvtt'): | |
3466 | scodec = scodec or full_codec | |
4f3c5e06 | 3467 | else: |
19a03940 | 3468 | write_string(f'WARNING: Unknown codec {full_codec}\n') |
3fe75fdc | 3469 | if vcodec or acodec or scodec: |
4f3c5e06 | 3470 | return { |
3471 | 'vcodec': vcodec or 'none', | |
3472 | 'acodec': acodec or 'none', | |
176f1866 | 3473 | 'dynamic_range': hdr, |
3fe75fdc | 3474 | **({'scodec': scodec} if scodec is not None else {}), |
4f3c5e06 | 3475 | } |
b69fd25c | 3476 | elif len(split_codecs) == 2: |
3477 | return { | |
3478 | 'vcodec': split_codecs[0], | |
3479 | 'acodec': split_codecs[1], | |
3480 | } | |
4f3c5e06 | 3481 | return {} |
3482 | ||
3483 | ||
fc61aff4 LL |
3484 | def get_compatible_ext(*, vcodecs, acodecs, vexts, aexts, preferences=None): |
3485 | assert len(vcodecs) == len(vexts) and len(acodecs) == len(aexts) | |
3486 | ||
3487 | allow_mkv = not preferences or 'mkv' in preferences | |
3488 | ||
3489 | if allow_mkv and max(len(acodecs), len(vcodecs)) > 1: | |
3490 | return 'mkv' # TODO: any other format allows this? | |
3491 | ||
3492 | # TODO: All codecs supported by parse_codecs isn't handled here | |
3493 | COMPATIBLE_CODECS = { | |
3494 | 'mp4': { | |
3495 | 'av1', 'hevc', 'avc1', 'mp4a', # fourcc (m3u8, mpd) | |
3496 | 'h264', 'aacl', # Set in ISM | |
3497 | }, | |
3498 | 'webm': { | |
3499 | 'av1', 'vp9', 'vp8', 'opus', 'vrbs', | |
3500 | 'vp9x', 'vp8x', # in the webm spec | |
3501 | }, | |
3502 | } | |
3503 | ||
8f84770a | 3504 | sanitize_codec = functools.partial(try_get, getter=lambda x: x[0].split('.')[0].replace('0', '')) |
3505 | vcodec, acodec = sanitize_codec(vcodecs), sanitize_codec(acodecs) | |
fc61aff4 LL |
3506 | |
3507 | for ext in preferences or COMPATIBLE_CODECS.keys(): | |
3508 | codec_set = COMPATIBLE_CODECS.get(ext, set()) | |
3509 | if ext == 'mkv' or codec_set.issuperset((vcodec, acodec)): | |
3510 | return ext | |
3511 | ||
3512 | COMPATIBLE_EXTS = ( | |
3513 | {'mp3', 'mp4', 'm4a', 'm4p', 'm4b', 'm4r', 'm4v', 'ismv', 'isma', 'mov'}, | |
3514 | {'webm'}, | |
3515 | ) | |
3516 | for ext in preferences or vexts: | |
3517 | current_exts = {ext, *vexts, *aexts} | |
3518 | if ext == 'mkv' or current_exts == {ext} or any( | |
3519 | ext_sets.issuperset(current_exts) for ext_sets in COMPATIBLE_EXTS): | |
3520 | return ext | |
3521 | return 'mkv' if allow_mkv else preferences[-1] | |
3522 | ||
3523 | ||
2ccd1b10 | 3524 | def urlhandle_detect_ext(url_handle): |
79298173 | 3525 | getheader = url_handle.headers.get |
2ccd1b10 | 3526 | |
b55ee18f PH |
3527 | cd = getheader('Content-Disposition') |
3528 | if cd: | |
3529 | m = re.match(r'attachment;\s*filename="(?P<filename>[^"]+)"', cd) | |
3530 | if m: | |
3531 | e = determine_ext(m.group('filename'), default_ext=None) | |
3532 | if e: | |
3533 | return e | |
3534 | ||
c460bdd5 | 3535 | return mimetype2ext(getheader('Content-Type')) |
05900629 PH |
3536 | |
3537 | ||
1e399778 YCH |
3538 | def encode_data_uri(data, mime_type): |
3539 | return 'data:%s;base64,%s' % (mime_type, base64.b64encode(data).decode('ascii')) | |
3540 | ||
3541 | ||
05900629 | 3542 | def age_restricted(content_limit, age_limit): |
6ec6cb4e | 3543 | """ Returns True iff the content should be blocked """ |
05900629 PH |
3544 | |
3545 | if age_limit is None: # No limit set | |
3546 | return False | |
3547 | if content_limit is None: | |
3548 | return False # Content available for everyone | |
3549 | return age_limit < content_limit | |
61ca9a80 PH |
3550 | |
3551 | ||
88f60feb | 3552 | # List of known byte-order-marks (BOM) |
a904a7f8 L |
3553 | BOMS = [ |
3554 | (b'\xef\xbb\xbf', 'utf-8'), | |
3555 | (b'\x00\x00\xfe\xff', 'utf-32-be'), | |
3556 | (b'\xff\xfe\x00\x00', 'utf-32-le'), | |
3557 | (b'\xff\xfe', 'utf-16-le'), | |
3558 | (b'\xfe\xff', 'utf-16-be'), | |
3559 | ] | |
a904a7f8 L |
3560 | |
3561 | ||
61ca9a80 PH |
3562 | def is_html(first_bytes): |
3563 | """ Detect whether a file contains HTML by examining its first bytes. """ | |
3564 | ||
80e8493e | 3565 | encoding = 'utf-8' |
61ca9a80 | 3566 | for bom, enc in BOMS: |
80e8493e | 3567 | while first_bytes.startswith(bom): |
3568 | encoding, first_bytes = enc, first_bytes[len(bom):] | |
61ca9a80 | 3569 | |
80e8493e | 3570 | return re.match(r'^\s*<', first_bytes.decode(encoding, 'replace')) |
a055469f PH |
3571 | |
3572 | ||
3573 | def determine_protocol(info_dict): | |
3574 | protocol = info_dict.get('protocol') | |
3575 | if protocol is not None: | |
3576 | return protocol | |
3577 | ||
7de837a5 | 3578 | url = sanitize_url(info_dict['url']) |
a055469f PH |
3579 | if url.startswith('rtmp'): |
3580 | return 'rtmp' | |
3581 | elif url.startswith('mms'): | |
3582 | return 'mms' | |
3583 | elif url.startswith('rtsp'): | |
3584 | return 'rtsp' | |
3585 | ||
3586 | ext = determine_ext(url) | |
3587 | if ext == 'm3u8': | |
3588 | return 'm3u8' | |
3589 | elif ext == 'f4m': | |
3590 | return 'f4m' | |
3591 | ||
14f25df2 | 3592 | return urllib.parse.urlparse(url).scheme |
cfb56d1a PH |
3593 | |
3594 | ||
c5e3f849 | 3595 | def render_table(header_row, data, delim=False, extra_gap=0, hide_empty=False): |
3596 | """ Render a list of rows, each as a list of values. | |
3597 | Text after a \t will be right aligned """ | |
ec11a9f4 | 3598 | def width(string): |
c5e3f849 | 3599 | return len(remove_terminal_sequences(string).replace('\t', '')) |
76d321f6 | 3600 | |
3601 | def get_max_lens(table): | |
ec11a9f4 | 3602 | return [max(width(str(v)) for v in col) for col in zip(*table)] |
76d321f6 | 3603 | |
3604 | def filter_using_list(row, filterArray): | |
d16df59d | 3605 | return [col for take, col in itertools.zip_longest(filterArray, row, fillvalue=True) if take] |
76d321f6 | 3606 | |
d16df59d | 3607 | max_lens = get_max_lens(data) if hide_empty else [] |
3608 | header_row = filter_using_list(header_row, max_lens) | |
3609 | data = [filter_using_list(row, max_lens) for row in data] | |
76d321f6 | 3610 | |
cfb56d1a | 3611 | table = [header_row] + data |
76d321f6 | 3612 | max_lens = get_max_lens(table) |
c5e3f849 | 3613 | extra_gap += 1 |
76d321f6 | 3614 | if delim: |
c5e3f849 | 3615 | table = [header_row, [delim * (ml + extra_gap) for ml in max_lens]] + data |
1ed7953a | 3616 | table[1][-1] = table[1][-1][:-extra_gap * len(delim)] # Remove extra_gap from end of delimiter |
ec11a9f4 | 3617 | for row in table: |
3618 | for pos, text in enumerate(map(str, row)): | |
c5e3f849 | 3619 | if '\t' in text: |
3620 | row[pos] = text.replace('\t', ' ' * (max_lens[pos] - width(text))) + ' ' * extra_gap | |
3621 | else: | |
3622 | row[pos] = text + ' ' * (max_lens[pos] - width(text) + extra_gap) | |
3623 | ret = '\n'.join(''.join(row).rstrip() for row in table) | |
ec11a9f4 | 3624 | return ret |
347de493 PH |
3625 | |
3626 | ||
8f18aca8 | 3627 | def _match_one(filter_part, dct, incomplete): |
77b87f05 | 3628 | # TODO: Generalize code with YoutubeDL._build_format_filter |
a047eeb6 | 3629 | STRING_OPERATORS = { |
3630 | '*=': operator.contains, | |
3631 | '^=': lambda attr, value: attr.startswith(value), | |
3632 | '$=': lambda attr, value: attr.endswith(value), | |
3633 | '~=': lambda attr, value: re.search(value, attr), | |
3634 | } | |
347de493 | 3635 | COMPARISON_OPERATORS = { |
a047eeb6 | 3636 | **STRING_OPERATORS, |
3637 | '<=': operator.le, # "<=" must be defined above "<" | |
347de493 | 3638 | '<': operator.lt, |
347de493 | 3639 | '>=': operator.ge, |
a047eeb6 | 3640 | '>': operator.gt, |
347de493 | 3641 | '=': operator.eq, |
347de493 | 3642 | } |
a047eeb6 | 3643 | |
6db9c4d5 | 3644 | if isinstance(incomplete, bool): |
3645 | is_incomplete = lambda _: incomplete | |
3646 | else: | |
3647 | is_incomplete = lambda k: k in incomplete | |
3648 | ||
64fa820c | 3649 | operator_rex = re.compile(r'''(?x) |
347de493 | 3650 | (?P<key>[a-z_]+) |
77b87f05 | 3651 | \s*(?P<negation>!\s*)?(?P<op>%s)(?P<none_inclusive>\s*\?)?\s* |
347de493 | 3652 | (?: |
a047eeb6 | 3653 | (?P<quote>["\'])(?P<quotedstrval>.+?)(?P=quote)| |
3654 | (?P<strval>.+?) | |
347de493 | 3655 | ) |
347de493 | 3656 | ''' % '|'.join(map(re.escape, COMPARISON_OPERATORS.keys()))) |
64fa820c | 3657 | m = operator_rex.fullmatch(filter_part.strip()) |
347de493 | 3658 | if m: |
18f96d12 | 3659 | m = m.groupdict() |
3660 | unnegated_op = COMPARISON_OPERATORS[m['op']] | |
3661 | if m['negation']: | |
77b87f05 MT |
3662 | op = lambda attr, value: not unnegated_op(attr, value) |
3663 | else: | |
3664 | op = unnegated_op | |
18f96d12 | 3665 | comparison_value = m['quotedstrval'] or m['strval'] or m['intval'] |
3666 | if m['quote']: | |
3667 | comparison_value = comparison_value.replace(r'\%s' % m['quote'], m['quote']) | |
3668 | actual_value = dct.get(m['key']) | |
3669 | numeric_comparison = None | |
f9934b96 | 3670 | if isinstance(actual_value, (int, float)): |
e5a088dc S |
3671 | # If the original field is a string and matching comparisonvalue is |
3672 | # a number we should respect the origin of the original field | |
3673 | # and process comparison value as a string (see | |
18f96d12 | 3674 | # https://github.com/ytdl-org/youtube-dl/issues/11082) |
347de493 | 3675 | try: |
18f96d12 | 3676 | numeric_comparison = int(comparison_value) |
347de493 | 3677 | except ValueError: |
18f96d12 | 3678 | numeric_comparison = parse_filesize(comparison_value) |
3679 | if numeric_comparison is None: | |
3680 | numeric_comparison = parse_filesize(f'{comparison_value}B') | |
3681 | if numeric_comparison is None: | |
3682 | numeric_comparison = parse_duration(comparison_value) | |
3683 | if numeric_comparison is not None and m['op'] in STRING_OPERATORS: | |
3684 | raise ValueError('Operator %s only supports string values!' % m['op']) | |
347de493 | 3685 | if actual_value is None: |
6db9c4d5 | 3686 | return is_incomplete(m['key']) or m['none_inclusive'] |
18f96d12 | 3687 | return op(actual_value, comparison_value if numeric_comparison is None else numeric_comparison) |
347de493 PH |
3688 | |
3689 | UNARY_OPERATORS = { | |
1cc47c66 S |
3690 | '': lambda v: (v is True) if isinstance(v, bool) else (v is not None), |
3691 | '!': lambda v: (v is False) if isinstance(v, bool) else (v is None), | |
347de493 | 3692 | } |
64fa820c | 3693 | operator_rex = re.compile(r'''(?x) |
347de493 | 3694 | (?P<op>%s)\s*(?P<key>[a-z_]+) |
347de493 | 3695 | ''' % '|'.join(map(re.escape, UNARY_OPERATORS.keys()))) |
64fa820c | 3696 | m = operator_rex.fullmatch(filter_part.strip()) |
347de493 PH |
3697 | if m: |
3698 | op = UNARY_OPERATORS[m.group('op')] | |
3699 | actual_value = dct.get(m.group('key')) | |
6db9c4d5 | 3700 | if is_incomplete(m.group('key')) and actual_value is None: |
8f18aca8 | 3701 | return True |
347de493 PH |
3702 | return op(actual_value) |
3703 | ||
3704 | raise ValueError('Invalid filter part %r' % filter_part) | |
3705 | ||
3706 | ||
8f18aca8 | 3707 | def match_str(filter_str, dct, incomplete=False): |
6db9c4d5 | 3708 | """ Filter a dictionary with a simple string syntax. |
3709 | @returns Whether the filter passes | |
3710 | @param incomplete Set of keys that is expected to be missing from dct. | |
3711 | Can be True/False to indicate all/none of the keys may be missing. | |
3712 | All conditions on incomplete keys pass if the key is missing | |
8f18aca8 | 3713 | """ |
347de493 | 3714 | return all( |
8f18aca8 | 3715 | _match_one(filter_part.replace(r'\&', '&'), dct, incomplete) |
a047eeb6 | 3716 | for filter_part in re.split(r'(?<!\\)&', filter_str)) |
347de493 PH |
3717 | |
3718 | ||
b1a7cd05 | 3719 | def match_filter_func(filters): |
3720 | if not filters: | |
d1b5f70b | 3721 | return None |
492272fe | 3722 | filters = set(variadic(filters)) |
d1b5f70b | 3723 | |
492272fe | 3724 | interactive = '-' in filters |
3725 | if interactive: | |
3726 | filters.remove('-') | |
3727 | ||
3728 | def _match_func(info_dict, incomplete=False): | |
3729 | if not filters or any(match_str(f, info_dict, incomplete) for f in filters): | |
3730 | return NO_DEFAULT if interactive and not incomplete else None | |
347de493 | 3731 | else: |
3bec830a | 3732 | video_title = info_dict.get('title') or info_dict.get('id') or 'entry' |
b1a7cd05 | 3733 | filter_str = ') | ('.join(map(str.strip, filters)) |
3734 | return f'{video_title} does not pass filter ({filter_str}), skipping ..' | |
347de493 | 3735 | return _match_func |
91410c9b PH |
3736 | |
3737 | ||
f2df4071 | 3738 | class download_range_func: |
3739 | def __init__(self, chapters, ranges): | |
3740 | self.chapters, self.ranges = chapters, ranges | |
3741 | ||
3742 | def __call__(self, info_dict, ydl): | |
5ec1b6b7 | 3743 | warning = ('There are no chapters matching the regex' if info_dict.get('chapters') |
56ba69e4 | 3744 | else 'Cannot match chapters since chapter information is unavailable') |
f2df4071 | 3745 | for regex in self.chapters or []: |
5ec1b6b7 | 3746 | for i, chapter in enumerate(info_dict.get('chapters') or []): |
3747 | if re.search(regex, chapter['title']): | |
3748 | warning = None | |
3749 | yield {**chapter, 'index': i} | |
f2df4071 | 3750 | if self.chapters and warning: |
5ec1b6b7 | 3751 | ydl.to_screen(f'[info] {info_dict["id"]}: {warning}') |
3752 | ||
f2df4071 | 3753 | yield from ({'start_time': start, 'end_time': end} for start, end in self.ranges or []) |
5ec1b6b7 | 3754 | |
f2df4071 | 3755 | def __eq__(self, other): |
3756 | return (isinstance(other, download_range_func) | |
3757 | and self.chapters == other.chapters and self.ranges == other.ranges) | |
5ec1b6b7 | 3758 | |
3759 | ||
bf6427d2 YCH |
3760 | def parse_dfxp_time_expr(time_expr): |
3761 | if not time_expr: | |
d631d5f9 | 3762 | return |
bf6427d2 | 3763 | |
1d485a1a | 3764 | mobj = re.match(rf'^(?P<time_offset>{NUMBER_RE})s?$', time_expr) |
bf6427d2 YCH |
3765 | if mobj: |
3766 | return float(mobj.group('time_offset')) | |
3767 | ||
db2fe38b | 3768 | mobj = re.match(r'^(\d+):(\d\d):(\d\d(?:(?:\.|:)\d+)?)$', time_expr) |
bf6427d2 | 3769 | if mobj: |
db2fe38b | 3770 | return 3600 * int(mobj.group(1)) + 60 * int(mobj.group(2)) + float(mobj.group(3).replace(':', '.')) |
bf6427d2 YCH |
3771 | |
3772 | ||
c1c924ab | 3773 | def srt_subtitles_timecode(seconds): |
aa7785f8 | 3774 | return '%02d:%02d:%02d,%03d' % timetuple_from_msec(seconds * 1000) |
3775 | ||
3776 | ||
3777 | def ass_subtitles_timecode(seconds): | |
3778 | time = timetuple_from_msec(seconds * 1000) | |
3779 | return '%01d:%02d:%02d.%02d' % (*time[:-1], time.milliseconds / 10) | |
bf6427d2 YCH |
3780 | |
3781 | ||
3782 | def dfxp2srt(dfxp_data): | |
3869028f YCH |
3783 | ''' |
3784 | @param dfxp_data A bytes-like object containing DFXP data | |
3785 | @returns A unicode object containing converted SRT data | |
3786 | ''' | |
5b995f71 | 3787 | LEGACY_NAMESPACES = ( |
3869028f YCH |
3788 | (b'http://www.w3.org/ns/ttml', [ |
3789 | b'http://www.w3.org/2004/11/ttaf1', | |
3790 | b'http://www.w3.org/2006/04/ttaf1', | |
3791 | b'http://www.w3.org/2006/10/ttaf1', | |
5b995f71 | 3792 | ]), |
3869028f YCH |
3793 | (b'http://www.w3.org/ns/ttml#styling', [ |
3794 | b'http://www.w3.org/ns/ttml#style', | |
5b995f71 RA |
3795 | ]), |
3796 | ) | |
3797 | ||
3798 | SUPPORTED_STYLING = [ | |
3799 | 'color', | |
3800 | 'fontFamily', | |
3801 | 'fontSize', | |
3802 | 'fontStyle', | |
3803 | 'fontWeight', | |
3804 | 'textDecoration' | |
3805 | ] | |
3806 | ||
4e335771 | 3807 | _x = functools.partial(xpath_with_ns, ns_map={ |
261f4730 | 3808 | 'xml': 'http://www.w3.org/XML/1998/namespace', |
4e335771 | 3809 | 'ttml': 'http://www.w3.org/ns/ttml', |
5b995f71 | 3810 | 'tts': 'http://www.w3.org/ns/ttml#styling', |
4e335771 | 3811 | }) |
bf6427d2 | 3812 | |
5b995f71 RA |
3813 | styles = {} |
3814 | default_style = {} | |
3815 | ||
86e5f3ed | 3816 | class TTMLPElementParser: |
5b995f71 RA |
3817 | _out = '' |
3818 | _unclosed_elements = [] | |
3819 | _applied_styles = [] | |
bf6427d2 | 3820 | |
2b14cb56 | 3821 | def start(self, tag, attrib): |
5b995f71 RA |
3822 | if tag in (_x('ttml:br'), 'br'): |
3823 | self._out += '\n' | |
3824 | else: | |
3825 | unclosed_elements = [] | |
3826 | style = {} | |
3827 | element_style_id = attrib.get('style') | |
3828 | if default_style: | |
3829 | style.update(default_style) | |
3830 | if element_style_id: | |
3831 | style.update(styles.get(element_style_id, {})) | |
3832 | for prop in SUPPORTED_STYLING: | |
3833 | prop_val = attrib.get(_x('tts:' + prop)) | |
3834 | if prop_val: | |
3835 | style[prop] = prop_val | |
3836 | if style: | |
3837 | font = '' | |
3838 | for k, v in sorted(style.items()): | |
3839 | if self._applied_styles and self._applied_styles[-1].get(k) == v: | |
3840 | continue | |
3841 | if k == 'color': | |
3842 | font += ' color="%s"' % v | |
3843 | elif k == 'fontSize': | |
3844 | font += ' size="%s"' % v | |
3845 | elif k == 'fontFamily': | |
3846 | font += ' face="%s"' % v | |
3847 | elif k == 'fontWeight' and v == 'bold': | |
3848 | self._out += '<b>' | |
3849 | unclosed_elements.append('b') | |
3850 | elif k == 'fontStyle' and v == 'italic': | |
3851 | self._out += '<i>' | |
3852 | unclosed_elements.append('i') | |
3853 | elif k == 'textDecoration' and v == 'underline': | |
3854 | self._out += '<u>' | |
3855 | unclosed_elements.append('u') | |
3856 | if font: | |
3857 | self._out += '<font' + font + '>' | |
3858 | unclosed_elements.append('font') | |
3859 | applied_style = {} | |
3860 | if self._applied_styles: | |
3861 | applied_style.update(self._applied_styles[-1]) | |
3862 | applied_style.update(style) | |
3863 | self._applied_styles.append(applied_style) | |
3864 | self._unclosed_elements.append(unclosed_elements) | |
bf6427d2 | 3865 | |
2b14cb56 | 3866 | def end(self, tag): |
5b995f71 RA |
3867 | if tag not in (_x('ttml:br'), 'br'): |
3868 | unclosed_elements = self._unclosed_elements.pop() | |
3869 | for element in reversed(unclosed_elements): | |
3870 | self._out += '</%s>' % element | |
3871 | if unclosed_elements and self._applied_styles: | |
3872 | self._applied_styles.pop() | |
bf6427d2 | 3873 | |
2b14cb56 | 3874 | def data(self, data): |
5b995f71 | 3875 | self._out += data |
2b14cb56 | 3876 | |
3877 | def close(self): | |
5b995f71 | 3878 | return self._out.strip() |
2b14cb56 | 3879 | |
3880 | def parse_node(node): | |
3881 | target = TTMLPElementParser() | |
3882 | parser = xml.etree.ElementTree.XMLParser(target=target) | |
3883 | parser.feed(xml.etree.ElementTree.tostring(node)) | |
3884 | return parser.close() | |
bf6427d2 | 3885 | |
5b995f71 RA |
3886 | for k, v in LEGACY_NAMESPACES: |
3887 | for ns in v: | |
3888 | dfxp_data = dfxp_data.replace(ns, k) | |
3889 | ||
3869028f | 3890 | dfxp = compat_etree_fromstring(dfxp_data) |
bf6427d2 | 3891 | out = [] |
5b995f71 | 3892 | paras = dfxp.findall(_x('.//ttml:p')) or dfxp.findall('.//p') |
1b0427e6 YCH |
3893 | |
3894 | if not paras: | |
3895 | raise ValueError('Invalid dfxp/TTML subtitle') | |
bf6427d2 | 3896 | |
5b995f71 RA |
3897 | repeat = False |
3898 | while True: | |
3899 | for style in dfxp.findall(_x('.//ttml:style')): | |
261f4730 RA |
3900 | style_id = style.get('id') or style.get(_x('xml:id')) |
3901 | if not style_id: | |
3902 | continue | |
5b995f71 RA |
3903 | parent_style_id = style.get('style') |
3904 | if parent_style_id: | |
3905 | if parent_style_id not in styles: | |
3906 | repeat = True | |
3907 | continue | |
3908 | styles[style_id] = styles[parent_style_id].copy() | |
3909 | for prop in SUPPORTED_STYLING: | |
3910 | prop_val = style.get(_x('tts:' + prop)) | |
3911 | if prop_val: | |
3912 | styles.setdefault(style_id, {})[prop] = prop_val | |
3913 | if repeat: | |
3914 | repeat = False | |
3915 | else: | |
3916 | break | |
3917 | ||
3918 | for p in ('body', 'div'): | |
3919 | ele = xpath_element(dfxp, [_x('.//ttml:' + p), './/' + p]) | |
3920 | if ele is None: | |
3921 | continue | |
3922 | style = styles.get(ele.get('style')) | |
3923 | if not style: | |
3924 | continue | |
3925 | default_style.update(style) | |
3926 | ||
bf6427d2 | 3927 | for para, index in zip(paras, itertools.count(1)): |
d631d5f9 | 3928 | begin_time = parse_dfxp_time_expr(para.attrib.get('begin')) |
7dff0363 | 3929 | end_time = parse_dfxp_time_expr(para.attrib.get('end')) |
d631d5f9 YCH |
3930 | dur = parse_dfxp_time_expr(para.attrib.get('dur')) |
3931 | if begin_time is None: | |
3932 | continue | |
7dff0363 | 3933 | if not end_time: |
d631d5f9 YCH |
3934 | if not dur: |
3935 | continue | |
3936 | end_time = begin_time + dur | |
bf6427d2 YCH |
3937 | out.append('%d\n%s --> %s\n%s\n\n' % ( |
3938 | index, | |
c1c924ab YCH |
3939 | srt_subtitles_timecode(begin_time), |
3940 | srt_subtitles_timecode(end_time), | |
bf6427d2 YCH |
3941 | parse_node(para))) |
3942 | ||
3943 | return ''.join(out) | |
3944 | ||
3945 | ||
c487cf00 | 3946 | def cli_option(params, command_option, param, separator=None): |
66e289ba | 3947 | param = params.get(param) |
c487cf00 | 3948 | return ([] if param is None |
3949 | else [command_option, str(param)] if separator is None | |
3950 | else [f'{command_option}{separator}{param}']) | |
66e289ba S |
3951 | |
3952 | ||
3953 | def cli_bool_option(params, command_option, param, true_value='true', false_value='false', separator=None): | |
3954 | param = params.get(param) | |
c487cf00 | 3955 | assert param in (True, False, None) |
3956 | return cli_option({True: true_value, False: false_value}, command_option, param, separator) | |
66e289ba S |
3957 | |
3958 | ||
3959 | def cli_valueless_option(params, command_option, param, expected_value=True): | |
c487cf00 | 3960 | return [command_option] if params.get(param) == expected_value else [] |
66e289ba S |
3961 | |
3962 | ||
e92caff5 | 3963 | def cli_configuration_args(argdict, keys, default=[], use_compat=True): |
eab9b2bc | 3964 | if isinstance(argdict, (list, tuple)): # for backward compatibility |
e92caff5 | 3965 | if use_compat: |
5b1ecbb3 | 3966 | return argdict |
3967 | else: | |
3968 | argdict = None | |
eab9b2bc | 3969 | if argdict is None: |
5b1ecbb3 | 3970 | return default |
eab9b2bc | 3971 | assert isinstance(argdict, dict) |
3972 | ||
e92caff5 | 3973 | assert isinstance(keys, (list, tuple)) |
3974 | for key_list in keys: | |
e92caff5 | 3975 | arg_list = list(filter( |
3976 | lambda x: x is not None, | |
6606817a | 3977 | [argdict.get(key.lower()) for key in variadic(key_list)])) |
e92caff5 | 3978 | if arg_list: |
3979 | return [arg for args in arg_list for arg in args] | |
3980 | return default | |
66e289ba | 3981 | |
6251555f | 3982 | |
330690a2 | 3983 | def _configuration_args(main_key, argdict, exe, keys=None, default=[], use_compat=True): |
3984 | main_key, exe = main_key.lower(), exe.lower() | |
3985 | root_key = exe if main_key == exe else f'{main_key}+{exe}' | |
3986 | keys = [f'{root_key}{k}' for k in (keys or [''])] | |
3987 | if root_key in keys: | |
3988 | if main_key != exe: | |
3989 | keys.append((main_key, exe)) | |
3990 | keys.append('default') | |
3991 | else: | |
3992 | use_compat = False | |
3993 | return cli_configuration_args(argdict, keys, default, use_compat) | |
3994 | ||
66e289ba | 3995 | |
86e5f3ed | 3996 | class ISO639Utils: |
39672624 YCH |
3997 | # See http://www.loc.gov/standards/iso639-2/ISO-639-2_utf-8.txt |
3998 | _lang_map = { | |
3999 | 'aa': 'aar', | |
4000 | 'ab': 'abk', | |
4001 | 'ae': 'ave', | |
4002 | 'af': 'afr', | |
4003 | 'ak': 'aka', | |
4004 | 'am': 'amh', | |
4005 | 'an': 'arg', | |
4006 | 'ar': 'ara', | |
4007 | 'as': 'asm', | |
4008 | 'av': 'ava', | |
4009 | 'ay': 'aym', | |
4010 | 'az': 'aze', | |
4011 | 'ba': 'bak', | |
4012 | 'be': 'bel', | |
4013 | 'bg': 'bul', | |
4014 | 'bh': 'bih', | |
4015 | 'bi': 'bis', | |
4016 | 'bm': 'bam', | |
4017 | 'bn': 'ben', | |
4018 | 'bo': 'bod', | |
4019 | 'br': 'bre', | |
4020 | 'bs': 'bos', | |
4021 | 'ca': 'cat', | |
4022 | 'ce': 'che', | |
4023 | 'ch': 'cha', | |
4024 | 'co': 'cos', | |
4025 | 'cr': 'cre', | |
4026 | 'cs': 'ces', | |
4027 | 'cu': 'chu', | |
4028 | 'cv': 'chv', | |
4029 | 'cy': 'cym', | |
4030 | 'da': 'dan', | |
4031 | 'de': 'deu', | |
4032 | 'dv': 'div', | |
4033 | 'dz': 'dzo', | |
4034 | 'ee': 'ewe', | |
4035 | 'el': 'ell', | |
4036 | 'en': 'eng', | |
4037 | 'eo': 'epo', | |
4038 | 'es': 'spa', | |
4039 | 'et': 'est', | |
4040 | 'eu': 'eus', | |
4041 | 'fa': 'fas', | |
4042 | 'ff': 'ful', | |
4043 | 'fi': 'fin', | |
4044 | 'fj': 'fij', | |
4045 | 'fo': 'fao', | |
4046 | 'fr': 'fra', | |
4047 | 'fy': 'fry', | |
4048 | 'ga': 'gle', | |
4049 | 'gd': 'gla', | |
4050 | 'gl': 'glg', | |
4051 | 'gn': 'grn', | |
4052 | 'gu': 'guj', | |
4053 | 'gv': 'glv', | |
4054 | 'ha': 'hau', | |
4055 | 'he': 'heb', | |
b7acc835 | 4056 | 'iw': 'heb', # Replaced by he in 1989 revision |
39672624 YCH |
4057 | 'hi': 'hin', |
4058 | 'ho': 'hmo', | |
4059 | 'hr': 'hrv', | |
4060 | 'ht': 'hat', | |
4061 | 'hu': 'hun', | |
4062 | 'hy': 'hye', | |
4063 | 'hz': 'her', | |
4064 | 'ia': 'ina', | |
4065 | 'id': 'ind', | |
b7acc835 | 4066 | 'in': 'ind', # Replaced by id in 1989 revision |
39672624 YCH |
4067 | 'ie': 'ile', |
4068 | 'ig': 'ibo', | |
4069 | 'ii': 'iii', | |
4070 | 'ik': 'ipk', | |
4071 | 'io': 'ido', | |
4072 | 'is': 'isl', | |
4073 | 'it': 'ita', | |
4074 | 'iu': 'iku', | |
4075 | 'ja': 'jpn', | |
4076 | 'jv': 'jav', | |
4077 | 'ka': 'kat', | |
4078 | 'kg': 'kon', | |
4079 | 'ki': 'kik', | |
4080 | 'kj': 'kua', | |
4081 | 'kk': 'kaz', | |
4082 | 'kl': 'kal', | |
4083 | 'km': 'khm', | |
4084 | 'kn': 'kan', | |
4085 | 'ko': 'kor', | |
4086 | 'kr': 'kau', | |
4087 | 'ks': 'kas', | |
4088 | 'ku': 'kur', | |
4089 | 'kv': 'kom', | |
4090 | 'kw': 'cor', | |
4091 | 'ky': 'kir', | |
4092 | 'la': 'lat', | |
4093 | 'lb': 'ltz', | |
4094 | 'lg': 'lug', | |
4095 | 'li': 'lim', | |
4096 | 'ln': 'lin', | |
4097 | 'lo': 'lao', | |
4098 | 'lt': 'lit', | |
4099 | 'lu': 'lub', | |
4100 | 'lv': 'lav', | |
4101 | 'mg': 'mlg', | |
4102 | 'mh': 'mah', | |
4103 | 'mi': 'mri', | |
4104 | 'mk': 'mkd', | |
4105 | 'ml': 'mal', | |
4106 | 'mn': 'mon', | |
4107 | 'mr': 'mar', | |
4108 | 'ms': 'msa', | |
4109 | 'mt': 'mlt', | |
4110 | 'my': 'mya', | |
4111 | 'na': 'nau', | |
4112 | 'nb': 'nob', | |
4113 | 'nd': 'nde', | |
4114 | 'ne': 'nep', | |
4115 | 'ng': 'ndo', | |
4116 | 'nl': 'nld', | |
4117 | 'nn': 'nno', | |
4118 | 'no': 'nor', | |
4119 | 'nr': 'nbl', | |
4120 | 'nv': 'nav', | |
4121 | 'ny': 'nya', | |
4122 | 'oc': 'oci', | |
4123 | 'oj': 'oji', | |
4124 | 'om': 'orm', | |
4125 | 'or': 'ori', | |
4126 | 'os': 'oss', | |
4127 | 'pa': 'pan', | |
4128 | 'pi': 'pli', | |
4129 | 'pl': 'pol', | |
4130 | 'ps': 'pus', | |
4131 | 'pt': 'por', | |
4132 | 'qu': 'que', | |
4133 | 'rm': 'roh', | |
4134 | 'rn': 'run', | |
4135 | 'ro': 'ron', | |
4136 | 'ru': 'rus', | |
4137 | 'rw': 'kin', | |
4138 | 'sa': 'san', | |
4139 | 'sc': 'srd', | |
4140 | 'sd': 'snd', | |
4141 | 'se': 'sme', | |
4142 | 'sg': 'sag', | |
4143 | 'si': 'sin', | |
4144 | 'sk': 'slk', | |
4145 | 'sl': 'slv', | |
4146 | 'sm': 'smo', | |
4147 | 'sn': 'sna', | |
4148 | 'so': 'som', | |
4149 | 'sq': 'sqi', | |
4150 | 'sr': 'srp', | |
4151 | 'ss': 'ssw', | |
4152 | 'st': 'sot', | |
4153 | 'su': 'sun', | |
4154 | 'sv': 'swe', | |
4155 | 'sw': 'swa', | |
4156 | 'ta': 'tam', | |
4157 | 'te': 'tel', | |
4158 | 'tg': 'tgk', | |
4159 | 'th': 'tha', | |
4160 | 'ti': 'tir', | |
4161 | 'tk': 'tuk', | |
4162 | 'tl': 'tgl', | |
4163 | 'tn': 'tsn', | |
4164 | 'to': 'ton', | |
4165 | 'tr': 'tur', | |
4166 | 'ts': 'tso', | |
4167 | 'tt': 'tat', | |
4168 | 'tw': 'twi', | |
4169 | 'ty': 'tah', | |
4170 | 'ug': 'uig', | |
4171 | 'uk': 'ukr', | |
4172 | 'ur': 'urd', | |
4173 | 'uz': 'uzb', | |
4174 | 've': 'ven', | |
4175 | 'vi': 'vie', | |
4176 | 'vo': 'vol', | |
4177 | 'wa': 'wln', | |
4178 | 'wo': 'wol', | |
4179 | 'xh': 'xho', | |
4180 | 'yi': 'yid', | |
e9a50fba | 4181 | 'ji': 'yid', # Replaced by yi in 1989 revision |
39672624 YCH |
4182 | 'yo': 'yor', |
4183 | 'za': 'zha', | |
4184 | 'zh': 'zho', | |
4185 | 'zu': 'zul', | |
4186 | } | |
4187 | ||
4188 | @classmethod | |
4189 | def short2long(cls, code): | |
4190 | """Convert language code from ISO 639-1 to ISO 639-2/T""" | |
4191 | return cls._lang_map.get(code[:2]) | |
4192 | ||
4193 | @classmethod | |
4194 | def long2short(cls, code): | |
4195 | """Convert language code from ISO 639-2/T to ISO 639-1""" | |
4196 | for short_name, long_name in cls._lang_map.items(): | |
4197 | if long_name == code: | |
4198 | return short_name | |
4199 | ||
4200 | ||
86e5f3ed | 4201 | class ISO3166Utils: |
4eb10f66 YCH |
4202 | # From http://data.okfn.org/data/core/country-list |
4203 | _country_map = { | |
4204 | 'AF': 'Afghanistan', | |
4205 | 'AX': 'Åland Islands', | |
4206 | 'AL': 'Albania', | |
4207 | 'DZ': 'Algeria', | |
4208 | 'AS': 'American Samoa', | |
4209 | 'AD': 'Andorra', | |
4210 | 'AO': 'Angola', | |
4211 | 'AI': 'Anguilla', | |
4212 | 'AQ': 'Antarctica', | |
4213 | 'AG': 'Antigua and Barbuda', | |
4214 | 'AR': 'Argentina', | |
4215 | 'AM': 'Armenia', | |
4216 | 'AW': 'Aruba', | |
4217 | 'AU': 'Australia', | |
4218 | 'AT': 'Austria', | |
4219 | 'AZ': 'Azerbaijan', | |
4220 | 'BS': 'Bahamas', | |
4221 | 'BH': 'Bahrain', | |
4222 | 'BD': 'Bangladesh', | |
4223 | 'BB': 'Barbados', | |
4224 | 'BY': 'Belarus', | |
4225 | 'BE': 'Belgium', | |
4226 | 'BZ': 'Belize', | |
4227 | 'BJ': 'Benin', | |
4228 | 'BM': 'Bermuda', | |
4229 | 'BT': 'Bhutan', | |
4230 | 'BO': 'Bolivia, Plurinational State of', | |
4231 | 'BQ': 'Bonaire, Sint Eustatius and Saba', | |
4232 | 'BA': 'Bosnia and Herzegovina', | |
4233 | 'BW': 'Botswana', | |
4234 | 'BV': 'Bouvet Island', | |
4235 | 'BR': 'Brazil', | |
4236 | 'IO': 'British Indian Ocean Territory', | |
4237 | 'BN': 'Brunei Darussalam', | |
4238 | 'BG': 'Bulgaria', | |
4239 | 'BF': 'Burkina Faso', | |
4240 | 'BI': 'Burundi', | |
4241 | 'KH': 'Cambodia', | |
4242 | 'CM': 'Cameroon', | |
4243 | 'CA': 'Canada', | |
4244 | 'CV': 'Cape Verde', | |
4245 | 'KY': 'Cayman Islands', | |
4246 | 'CF': 'Central African Republic', | |
4247 | 'TD': 'Chad', | |
4248 | 'CL': 'Chile', | |
4249 | 'CN': 'China', | |
4250 | 'CX': 'Christmas Island', | |
4251 | 'CC': 'Cocos (Keeling) Islands', | |
4252 | 'CO': 'Colombia', | |
4253 | 'KM': 'Comoros', | |
4254 | 'CG': 'Congo', | |
4255 | 'CD': 'Congo, the Democratic Republic of the', | |
4256 | 'CK': 'Cook Islands', | |
4257 | 'CR': 'Costa Rica', | |
4258 | 'CI': 'Côte d\'Ivoire', | |
4259 | 'HR': 'Croatia', | |
4260 | 'CU': 'Cuba', | |
4261 | 'CW': 'Curaçao', | |
4262 | 'CY': 'Cyprus', | |
4263 | 'CZ': 'Czech Republic', | |
4264 | 'DK': 'Denmark', | |
4265 | 'DJ': 'Djibouti', | |
4266 | 'DM': 'Dominica', | |
4267 | 'DO': 'Dominican Republic', | |
4268 | 'EC': 'Ecuador', | |
4269 | 'EG': 'Egypt', | |
4270 | 'SV': 'El Salvador', | |
4271 | 'GQ': 'Equatorial Guinea', | |
4272 | 'ER': 'Eritrea', | |
4273 | 'EE': 'Estonia', | |
4274 | 'ET': 'Ethiopia', | |
4275 | 'FK': 'Falkland Islands (Malvinas)', | |
4276 | 'FO': 'Faroe Islands', | |
4277 | 'FJ': 'Fiji', | |
4278 | 'FI': 'Finland', | |
4279 | 'FR': 'France', | |
4280 | 'GF': 'French Guiana', | |
4281 | 'PF': 'French Polynesia', | |
4282 | 'TF': 'French Southern Territories', | |
4283 | 'GA': 'Gabon', | |
4284 | 'GM': 'Gambia', | |
4285 | 'GE': 'Georgia', | |
4286 | 'DE': 'Germany', | |
4287 | 'GH': 'Ghana', | |
4288 | 'GI': 'Gibraltar', | |
4289 | 'GR': 'Greece', | |
4290 | 'GL': 'Greenland', | |
4291 | 'GD': 'Grenada', | |
4292 | 'GP': 'Guadeloupe', | |
4293 | 'GU': 'Guam', | |
4294 | 'GT': 'Guatemala', | |
4295 | 'GG': 'Guernsey', | |
4296 | 'GN': 'Guinea', | |
4297 | 'GW': 'Guinea-Bissau', | |
4298 | 'GY': 'Guyana', | |
4299 | 'HT': 'Haiti', | |
4300 | 'HM': 'Heard Island and McDonald Islands', | |
4301 | 'VA': 'Holy See (Vatican City State)', | |
4302 | 'HN': 'Honduras', | |
4303 | 'HK': 'Hong Kong', | |
4304 | 'HU': 'Hungary', | |
4305 | 'IS': 'Iceland', | |
4306 | 'IN': 'India', | |
4307 | 'ID': 'Indonesia', | |
4308 | 'IR': 'Iran, Islamic Republic of', | |
4309 | 'IQ': 'Iraq', | |
4310 | 'IE': 'Ireland', | |
4311 | 'IM': 'Isle of Man', | |
4312 | 'IL': 'Israel', | |
4313 | 'IT': 'Italy', | |
4314 | 'JM': 'Jamaica', | |
4315 | 'JP': 'Japan', | |
4316 | 'JE': 'Jersey', | |
4317 | 'JO': 'Jordan', | |
4318 | 'KZ': 'Kazakhstan', | |
4319 | 'KE': 'Kenya', | |
4320 | 'KI': 'Kiribati', | |
4321 | 'KP': 'Korea, Democratic People\'s Republic of', | |
4322 | 'KR': 'Korea, Republic of', | |
4323 | 'KW': 'Kuwait', | |
4324 | 'KG': 'Kyrgyzstan', | |
4325 | 'LA': 'Lao People\'s Democratic Republic', | |
4326 | 'LV': 'Latvia', | |
4327 | 'LB': 'Lebanon', | |
4328 | 'LS': 'Lesotho', | |
4329 | 'LR': 'Liberia', | |
4330 | 'LY': 'Libya', | |
4331 | 'LI': 'Liechtenstein', | |
4332 | 'LT': 'Lithuania', | |
4333 | 'LU': 'Luxembourg', | |
4334 | 'MO': 'Macao', | |
4335 | 'MK': 'Macedonia, the Former Yugoslav Republic of', | |
4336 | 'MG': 'Madagascar', | |
4337 | 'MW': 'Malawi', | |
4338 | 'MY': 'Malaysia', | |
4339 | 'MV': 'Maldives', | |
4340 | 'ML': 'Mali', | |
4341 | 'MT': 'Malta', | |
4342 | 'MH': 'Marshall Islands', | |
4343 | 'MQ': 'Martinique', | |
4344 | 'MR': 'Mauritania', | |
4345 | 'MU': 'Mauritius', | |
4346 | 'YT': 'Mayotte', | |
4347 | 'MX': 'Mexico', | |
4348 | 'FM': 'Micronesia, Federated States of', | |
4349 | 'MD': 'Moldova, Republic of', | |
4350 | 'MC': 'Monaco', | |
4351 | 'MN': 'Mongolia', | |
4352 | 'ME': 'Montenegro', | |
4353 | 'MS': 'Montserrat', | |
4354 | 'MA': 'Morocco', | |
4355 | 'MZ': 'Mozambique', | |
4356 | 'MM': 'Myanmar', | |
4357 | 'NA': 'Namibia', | |
4358 | 'NR': 'Nauru', | |
4359 | 'NP': 'Nepal', | |
4360 | 'NL': 'Netherlands', | |
4361 | 'NC': 'New Caledonia', | |
4362 | 'NZ': 'New Zealand', | |
4363 | 'NI': 'Nicaragua', | |
4364 | 'NE': 'Niger', | |
4365 | 'NG': 'Nigeria', | |
4366 | 'NU': 'Niue', | |
4367 | 'NF': 'Norfolk Island', | |
4368 | 'MP': 'Northern Mariana Islands', | |
4369 | 'NO': 'Norway', | |
4370 | 'OM': 'Oman', | |
4371 | 'PK': 'Pakistan', | |
4372 | 'PW': 'Palau', | |
4373 | 'PS': 'Palestine, State of', | |
4374 | 'PA': 'Panama', | |
4375 | 'PG': 'Papua New Guinea', | |
4376 | 'PY': 'Paraguay', | |
4377 | 'PE': 'Peru', | |
4378 | 'PH': 'Philippines', | |
4379 | 'PN': 'Pitcairn', | |
4380 | 'PL': 'Poland', | |
4381 | 'PT': 'Portugal', | |
4382 | 'PR': 'Puerto Rico', | |
4383 | 'QA': 'Qatar', | |
4384 | 'RE': 'Réunion', | |
4385 | 'RO': 'Romania', | |
4386 | 'RU': 'Russian Federation', | |
4387 | 'RW': 'Rwanda', | |
4388 | 'BL': 'Saint Barthélemy', | |
4389 | 'SH': 'Saint Helena, Ascension and Tristan da Cunha', | |
4390 | 'KN': 'Saint Kitts and Nevis', | |
4391 | 'LC': 'Saint Lucia', | |
4392 | 'MF': 'Saint Martin (French part)', | |
4393 | 'PM': 'Saint Pierre and Miquelon', | |
4394 | 'VC': 'Saint Vincent and the Grenadines', | |
4395 | 'WS': 'Samoa', | |
4396 | 'SM': 'San Marino', | |
4397 | 'ST': 'Sao Tome and Principe', | |
4398 | 'SA': 'Saudi Arabia', | |
4399 | 'SN': 'Senegal', | |
4400 | 'RS': 'Serbia', | |
4401 | 'SC': 'Seychelles', | |
4402 | 'SL': 'Sierra Leone', | |
4403 | 'SG': 'Singapore', | |
4404 | 'SX': 'Sint Maarten (Dutch part)', | |
4405 | 'SK': 'Slovakia', | |
4406 | 'SI': 'Slovenia', | |
4407 | 'SB': 'Solomon Islands', | |
4408 | 'SO': 'Somalia', | |
4409 | 'ZA': 'South Africa', | |
4410 | 'GS': 'South Georgia and the South Sandwich Islands', | |
4411 | 'SS': 'South Sudan', | |
4412 | 'ES': 'Spain', | |
4413 | 'LK': 'Sri Lanka', | |
4414 | 'SD': 'Sudan', | |
4415 | 'SR': 'Suriname', | |
4416 | 'SJ': 'Svalbard and Jan Mayen', | |
4417 | 'SZ': 'Swaziland', | |
4418 | 'SE': 'Sweden', | |
4419 | 'CH': 'Switzerland', | |
4420 | 'SY': 'Syrian Arab Republic', | |
4421 | 'TW': 'Taiwan, Province of China', | |
4422 | 'TJ': 'Tajikistan', | |
4423 | 'TZ': 'Tanzania, United Republic of', | |
4424 | 'TH': 'Thailand', | |
4425 | 'TL': 'Timor-Leste', | |
4426 | 'TG': 'Togo', | |
4427 | 'TK': 'Tokelau', | |
4428 | 'TO': 'Tonga', | |
4429 | 'TT': 'Trinidad and Tobago', | |
4430 | 'TN': 'Tunisia', | |
4431 | 'TR': 'Turkey', | |
4432 | 'TM': 'Turkmenistan', | |
4433 | 'TC': 'Turks and Caicos Islands', | |
4434 | 'TV': 'Tuvalu', | |
4435 | 'UG': 'Uganda', | |
4436 | 'UA': 'Ukraine', | |
4437 | 'AE': 'United Arab Emirates', | |
4438 | 'GB': 'United Kingdom', | |
4439 | 'US': 'United States', | |
4440 | 'UM': 'United States Minor Outlying Islands', | |
4441 | 'UY': 'Uruguay', | |
4442 | 'UZ': 'Uzbekistan', | |
4443 | 'VU': 'Vanuatu', | |
4444 | 'VE': 'Venezuela, Bolivarian Republic of', | |
4445 | 'VN': 'Viet Nam', | |
4446 | 'VG': 'Virgin Islands, British', | |
4447 | 'VI': 'Virgin Islands, U.S.', | |
4448 | 'WF': 'Wallis and Futuna', | |
4449 | 'EH': 'Western Sahara', | |
4450 | 'YE': 'Yemen', | |
4451 | 'ZM': 'Zambia', | |
4452 | 'ZW': 'Zimbabwe', | |
2f97cc61 | 4453 | # Not ISO 3166 codes, but used for IP blocks |
4454 | 'AP': 'Asia/Pacific Region', | |
4455 | 'EU': 'Europe', | |
4eb10f66 YCH |
4456 | } |
4457 | ||
4458 | @classmethod | |
4459 | def short2full(cls, code): | |
4460 | """Convert an ISO 3166-2 country code to the corresponding full name""" | |
4461 | return cls._country_map.get(code.upper()) | |
4462 | ||
4463 | ||
86e5f3ed | 4464 | class GeoUtils: |
773f291d S |
4465 | # Major IPv4 address blocks per country |
4466 | _country_ip_map = { | |
53896ca5 | 4467 | 'AD': '46.172.224.0/19', |
773f291d S |
4468 | 'AE': '94.200.0.0/13', |
4469 | 'AF': '149.54.0.0/17', | |
4470 | 'AG': '209.59.64.0/18', | |
4471 | 'AI': '204.14.248.0/21', | |
4472 | 'AL': '46.99.0.0/16', | |
4473 | 'AM': '46.70.0.0/15', | |
4474 | 'AO': '105.168.0.0/13', | |
53896ca5 S |
4475 | 'AP': '182.50.184.0/21', |
4476 | 'AQ': '23.154.160.0/24', | |
773f291d S |
4477 | 'AR': '181.0.0.0/12', |
4478 | 'AS': '202.70.112.0/20', | |
53896ca5 | 4479 | 'AT': '77.116.0.0/14', |
773f291d S |
4480 | 'AU': '1.128.0.0/11', |
4481 | 'AW': '181.41.0.0/18', | |
53896ca5 S |
4482 | 'AX': '185.217.4.0/22', |
4483 | 'AZ': '5.197.0.0/16', | |
773f291d S |
4484 | 'BA': '31.176.128.0/17', |
4485 | 'BB': '65.48.128.0/17', | |
4486 | 'BD': '114.130.0.0/16', | |
4487 | 'BE': '57.0.0.0/8', | |
53896ca5 | 4488 | 'BF': '102.178.0.0/15', |
773f291d S |
4489 | 'BG': '95.42.0.0/15', |
4490 | 'BH': '37.131.0.0/17', | |
4491 | 'BI': '154.117.192.0/18', | |
4492 | 'BJ': '137.255.0.0/16', | |
53896ca5 | 4493 | 'BL': '185.212.72.0/23', |
773f291d S |
4494 | 'BM': '196.12.64.0/18', |
4495 | 'BN': '156.31.0.0/16', | |
4496 | 'BO': '161.56.0.0/16', | |
4497 | 'BQ': '161.0.80.0/20', | |
53896ca5 | 4498 | 'BR': '191.128.0.0/12', |
773f291d S |
4499 | 'BS': '24.51.64.0/18', |
4500 | 'BT': '119.2.96.0/19', | |
4501 | 'BW': '168.167.0.0/16', | |
4502 | 'BY': '178.120.0.0/13', | |
4503 | 'BZ': '179.42.192.0/18', | |
4504 | 'CA': '99.224.0.0/11', | |
4505 | 'CD': '41.243.0.0/16', | |
53896ca5 S |
4506 | 'CF': '197.242.176.0/21', |
4507 | 'CG': '160.113.0.0/16', | |
773f291d | 4508 | 'CH': '85.0.0.0/13', |
53896ca5 | 4509 | 'CI': '102.136.0.0/14', |
773f291d S |
4510 | 'CK': '202.65.32.0/19', |
4511 | 'CL': '152.172.0.0/14', | |
53896ca5 | 4512 | 'CM': '102.244.0.0/14', |
773f291d S |
4513 | 'CN': '36.128.0.0/10', |
4514 | 'CO': '181.240.0.0/12', | |
4515 | 'CR': '201.192.0.0/12', | |
4516 | 'CU': '152.206.0.0/15', | |
4517 | 'CV': '165.90.96.0/19', | |
4518 | 'CW': '190.88.128.0/17', | |
53896ca5 | 4519 | 'CY': '31.153.0.0/16', |
773f291d S |
4520 | 'CZ': '88.100.0.0/14', |
4521 | 'DE': '53.0.0.0/8', | |
4522 | 'DJ': '197.241.0.0/17', | |
4523 | 'DK': '87.48.0.0/12', | |
4524 | 'DM': '192.243.48.0/20', | |
4525 | 'DO': '152.166.0.0/15', | |
4526 | 'DZ': '41.96.0.0/12', | |
4527 | 'EC': '186.68.0.0/15', | |
4528 | 'EE': '90.190.0.0/15', | |
4529 | 'EG': '156.160.0.0/11', | |
4530 | 'ER': '196.200.96.0/20', | |
4531 | 'ES': '88.0.0.0/11', | |
4532 | 'ET': '196.188.0.0/14', | |
4533 | 'EU': '2.16.0.0/13', | |
4534 | 'FI': '91.152.0.0/13', | |
4535 | 'FJ': '144.120.0.0/16', | |
53896ca5 | 4536 | 'FK': '80.73.208.0/21', |
773f291d S |
4537 | 'FM': '119.252.112.0/20', |
4538 | 'FO': '88.85.32.0/19', | |
4539 | 'FR': '90.0.0.0/9', | |
4540 | 'GA': '41.158.0.0/15', | |
4541 | 'GB': '25.0.0.0/8', | |
4542 | 'GD': '74.122.88.0/21', | |
4543 | 'GE': '31.146.0.0/16', | |
4544 | 'GF': '161.22.64.0/18', | |
4545 | 'GG': '62.68.160.0/19', | |
53896ca5 S |
4546 | 'GH': '154.160.0.0/12', |
4547 | 'GI': '95.164.0.0/16', | |
773f291d S |
4548 | 'GL': '88.83.0.0/19', |
4549 | 'GM': '160.182.0.0/15', | |
4550 | 'GN': '197.149.192.0/18', | |
4551 | 'GP': '104.250.0.0/19', | |
4552 | 'GQ': '105.235.224.0/20', | |
4553 | 'GR': '94.64.0.0/13', | |
4554 | 'GT': '168.234.0.0/16', | |
4555 | 'GU': '168.123.0.0/16', | |
4556 | 'GW': '197.214.80.0/20', | |
4557 | 'GY': '181.41.64.0/18', | |
4558 | 'HK': '113.252.0.0/14', | |
4559 | 'HN': '181.210.0.0/16', | |
4560 | 'HR': '93.136.0.0/13', | |
4561 | 'HT': '148.102.128.0/17', | |
4562 | 'HU': '84.0.0.0/14', | |
4563 | 'ID': '39.192.0.0/10', | |
4564 | 'IE': '87.32.0.0/12', | |
4565 | 'IL': '79.176.0.0/13', | |
4566 | 'IM': '5.62.80.0/20', | |
4567 | 'IN': '117.192.0.0/10', | |
4568 | 'IO': '203.83.48.0/21', | |
4569 | 'IQ': '37.236.0.0/14', | |
4570 | 'IR': '2.176.0.0/12', | |
4571 | 'IS': '82.221.0.0/16', | |
4572 | 'IT': '79.0.0.0/10', | |
4573 | 'JE': '87.244.64.0/18', | |
4574 | 'JM': '72.27.0.0/17', | |
4575 | 'JO': '176.29.0.0/16', | |
53896ca5 | 4576 | 'JP': '133.0.0.0/8', |
773f291d S |
4577 | 'KE': '105.48.0.0/12', |
4578 | 'KG': '158.181.128.0/17', | |
4579 | 'KH': '36.37.128.0/17', | |
4580 | 'KI': '103.25.140.0/22', | |
4581 | 'KM': '197.255.224.0/20', | |
53896ca5 | 4582 | 'KN': '198.167.192.0/19', |
773f291d S |
4583 | 'KP': '175.45.176.0/22', |
4584 | 'KR': '175.192.0.0/10', | |
4585 | 'KW': '37.36.0.0/14', | |
4586 | 'KY': '64.96.0.0/15', | |
4587 | 'KZ': '2.72.0.0/13', | |
4588 | 'LA': '115.84.64.0/18', | |
4589 | 'LB': '178.135.0.0/16', | |
53896ca5 | 4590 | 'LC': '24.92.144.0/20', |
773f291d S |
4591 | 'LI': '82.117.0.0/19', |
4592 | 'LK': '112.134.0.0/15', | |
53896ca5 | 4593 | 'LR': '102.183.0.0/16', |
773f291d S |
4594 | 'LS': '129.232.0.0/17', |
4595 | 'LT': '78.56.0.0/13', | |
4596 | 'LU': '188.42.0.0/16', | |
4597 | 'LV': '46.109.0.0/16', | |
4598 | 'LY': '41.252.0.0/14', | |
4599 | 'MA': '105.128.0.0/11', | |
4600 | 'MC': '88.209.64.0/18', | |
4601 | 'MD': '37.246.0.0/16', | |
4602 | 'ME': '178.175.0.0/17', | |
4603 | 'MF': '74.112.232.0/21', | |
4604 | 'MG': '154.126.0.0/17', | |
4605 | 'MH': '117.103.88.0/21', | |
4606 | 'MK': '77.28.0.0/15', | |
4607 | 'ML': '154.118.128.0/18', | |
4608 | 'MM': '37.111.0.0/17', | |
4609 | 'MN': '49.0.128.0/17', | |
4610 | 'MO': '60.246.0.0/16', | |
4611 | 'MP': '202.88.64.0/20', | |
4612 | 'MQ': '109.203.224.0/19', | |
4613 | 'MR': '41.188.64.0/18', | |
4614 | 'MS': '208.90.112.0/22', | |
4615 | 'MT': '46.11.0.0/16', | |
4616 | 'MU': '105.16.0.0/12', | |
4617 | 'MV': '27.114.128.0/18', | |
53896ca5 | 4618 | 'MW': '102.70.0.0/15', |
773f291d S |
4619 | 'MX': '187.192.0.0/11', |
4620 | 'MY': '175.136.0.0/13', | |
4621 | 'MZ': '197.218.0.0/15', | |
4622 | 'NA': '41.182.0.0/16', | |
4623 | 'NC': '101.101.0.0/18', | |
4624 | 'NE': '197.214.0.0/18', | |
4625 | 'NF': '203.17.240.0/22', | |
4626 | 'NG': '105.112.0.0/12', | |
4627 | 'NI': '186.76.0.0/15', | |
4628 | 'NL': '145.96.0.0/11', | |
4629 | 'NO': '84.208.0.0/13', | |
4630 | 'NP': '36.252.0.0/15', | |
4631 | 'NR': '203.98.224.0/19', | |
4632 | 'NU': '49.156.48.0/22', | |
4633 | 'NZ': '49.224.0.0/14', | |
4634 | 'OM': '5.36.0.0/15', | |
4635 | 'PA': '186.72.0.0/15', | |
4636 | 'PE': '186.160.0.0/14', | |
4637 | 'PF': '123.50.64.0/18', | |
4638 | 'PG': '124.240.192.0/19', | |
4639 | 'PH': '49.144.0.0/13', | |
4640 | 'PK': '39.32.0.0/11', | |
4641 | 'PL': '83.0.0.0/11', | |
4642 | 'PM': '70.36.0.0/20', | |
4643 | 'PR': '66.50.0.0/16', | |
4644 | 'PS': '188.161.0.0/16', | |
4645 | 'PT': '85.240.0.0/13', | |
4646 | 'PW': '202.124.224.0/20', | |
4647 | 'PY': '181.120.0.0/14', | |
4648 | 'QA': '37.210.0.0/15', | |
53896ca5 | 4649 | 'RE': '102.35.0.0/16', |
773f291d | 4650 | 'RO': '79.112.0.0/13', |
53896ca5 | 4651 | 'RS': '93.86.0.0/15', |
773f291d | 4652 | 'RU': '5.136.0.0/13', |
53896ca5 | 4653 | 'RW': '41.186.0.0/16', |
773f291d S |
4654 | 'SA': '188.48.0.0/13', |
4655 | 'SB': '202.1.160.0/19', | |
4656 | 'SC': '154.192.0.0/11', | |
53896ca5 | 4657 | 'SD': '102.120.0.0/13', |
773f291d | 4658 | 'SE': '78.64.0.0/12', |
53896ca5 | 4659 | 'SG': '8.128.0.0/10', |
773f291d S |
4660 | 'SI': '188.196.0.0/14', |
4661 | 'SK': '78.98.0.0/15', | |
53896ca5 | 4662 | 'SL': '102.143.0.0/17', |
773f291d S |
4663 | 'SM': '89.186.32.0/19', |
4664 | 'SN': '41.82.0.0/15', | |
53896ca5 | 4665 | 'SO': '154.115.192.0/18', |
773f291d S |
4666 | 'SR': '186.179.128.0/17', |
4667 | 'SS': '105.235.208.0/21', | |
4668 | 'ST': '197.159.160.0/19', | |
4669 | 'SV': '168.243.0.0/16', | |
4670 | 'SX': '190.102.0.0/20', | |
4671 | 'SY': '5.0.0.0/16', | |
4672 | 'SZ': '41.84.224.0/19', | |
4673 | 'TC': '65.255.48.0/20', | |
4674 | 'TD': '154.68.128.0/19', | |
4675 | 'TG': '196.168.0.0/14', | |
4676 | 'TH': '171.96.0.0/13', | |
4677 | 'TJ': '85.9.128.0/18', | |
4678 | 'TK': '27.96.24.0/21', | |
4679 | 'TL': '180.189.160.0/20', | |
4680 | 'TM': '95.85.96.0/19', | |
4681 | 'TN': '197.0.0.0/11', | |
4682 | 'TO': '175.176.144.0/21', | |
4683 | 'TR': '78.160.0.0/11', | |
4684 | 'TT': '186.44.0.0/15', | |
4685 | 'TV': '202.2.96.0/19', | |
4686 | 'TW': '120.96.0.0/11', | |
4687 | 'TZ': '156.156.0.0/14', | |
53896ca5 S |
4688 | 'UA': '37.52.0.0/14', |
4689 | 'UG': '102.80.0.0/13', | |
4690 | 'US': '6.0.0.0/8', | |
773f291d | 4691 | 'UY': '167.56.0.0/13', |
53896ca5 | 4692 | 'UZ': '84.54.64.0/18', |
773f291d | 4693 | 'VA': '212.77.0.0/19', |
53896ca5 | 4694 | 'VC': '207.191.240.0/21', |
773f291d | 4695 | 'VE': '186.88.0.0/13', |
53896ca5 | 4696 | 'VG': '66.81.192.0/20', |
773f291d S |
4697 | 'VI': '146.226.0.0/16', |
4698 | 'VN': '14.160.0.0/11', | |
4699 | 'VU': '202.80.32.0/20', | |
4700 | 'WF': '117.20.32.0/21', | |
4701 | 'WS': '202.4.32.0/19', | |
4702 | 'YE': '134.35.0.0/16', | |
4703 | 'YT': '41.242.116.0/22', | |
4704 | 'ZA': '41.0.0.0/11', | |
53896ca5 S |
4705 | 'ZM': '102.144.0.0/13', |
4706 | 'ZW': '102.177.192.0/18', | |
773f291d S |
4707 | } |
4708 | ||
4709 | @classmethod | |
5f95927a S |
4710 | def random_ipv4(cls, code_or_block): |
4711 | if len(code_or_block) == 2: | |
4712 | block = cls._country_ip_map.get(code_or_block.upper()) | |
4713 | if not block: | |
4714 | return None | |
4715 | else: | |
4716 | block = code_or_block | |
773f291d | 4717 | addr, preflen = block.split('/') |
ac668111 | 4718 | addr_min = struct.unpack('!L', socket.inet_aton(addr))[0] |
773f291d | 4719 | addr_max = addr_min | (0xffffffff >> int(preflen)) |
14f25df2 | 4720 | return str(socket.inet_ntoa( |
ac668111 | 4721 | struct.pack('!L', random.randint(addr_min, addr_max)))) |
773f291d S |
4722 | |
4723 | ||
ac668111 | 4724 | class PerRequestProxyHandler(urllib.request.ProxyHandler): |
2461f79d PH |
4725 | def __init__(self, proxies=None): |
4726 | # Set default handlers | |
4727 | for type in ('http', 'https'): | |
4728 | setattr(self, '%s_open' % type, | |
4729 | lambda r, proxy='__noproxy__', type=type, meth=self.proxy_open: | |
4730 | meth(r, proxy, type)) | |
ac668111 | 4731 | urllib.request.ProxyHandler.__init__(self, proxies) |
2461f79d | 4732 | |
91410c9b | 4733 | def proxy_open(self, req, proxy, type): |
2461f79d | 4734 | req_proxy = req.headers.get('Ytdl-request-proxy') |
91410c9b PH |
4735 | if req_proxy is not None: |
4736 | proxy = req_proxy | |
2461f79d PH |
4737 | del req.headers['Ytdl-request-proxy'] |
4738 | ||
4739 | if proxy == '__noproxy__': | |
4740 | return None # No Proxy | |
14f25df2 | 4741 | if urllib.parse.urlparse(proxy).scheme.lower() in ('socks', 'socks4', 'socks4a', 'socks5'): |
71aff188 | 4742 | req.add_header('Ytdl-socks-proxy', proxy) |
7a5c1cfe | 4743 | # yt-dlp's http/https handlers do wrapping the socket with socks |
71aff188 | 4744 | return None |
ac668111 | 4745 | return urllib.request.ProxyHandler.proxy_open( |
91410c9b | 4746 | self, req, proxy, type) |
5bc880b9 YCH |
4747 | |
4748 | ||
0a5445dd YCH |
4749 | # Both long_to_bytes and bytes_to_long are adapted from PyCrypto, which is |
4750 | # released into Public Domain | |
4751 | # https://github.com/dlitz/pycrypto/blob/master/lib/Crypto/Util/number.py#L387 | |
4752 | ||
4753 | def long_to_bytes(n, blocksize=0): | |
4754 | """long_to_bytes(n:long, blocksize:int) : string | |
4755 | Convert a long integer to a byte string. | |
4756 | ||
4757 | If optional blocksize is given and greater than zero, pad the front of the | |
4758 | byte string with binary zeros so that the length is a multiple of | |
4759 | blocksize. | |
4760 | """ | |
4761 | # after much testing, this algorithm was deemed to be the fastest | |
4762 | s = b'' | |
4763 | n = int(n) | |
4764 | while n > 0: | |
ac668111 | 4765 | s = struct.pack('>I', n & 0xffffffff) + s |
0a5445dd YCH |
4766 | n = n >> 32 |
4767 | # strip off leading zeros | |
4768 | for i in range(len(s)): | |
4769 | if s[i] != b'\000'[0]: | |
4770 | break | |
4771 | else: | |
4772 | # only happens when n == 0 | |
4773 | s = b'\000' | |
4774 | i = 0 | |
4775 | s = s[i:] | |
4776 | # add back some pad bytes. this could be done more efficiently w.r.t. the | |
4777 | # de-padding being done above, but sigh... | |
4778 | if blocksize > 0 and len(s) % blocksize: | |
4779 | s = (blocksize - len(s) % blocksize) * b'\000' + s | |
4780 | return s | |
4781 | ||
4782 | ||
4783 | def bytes_to_long(s): | |
4784 | """bytes_to_long(string) : long | |
4785 | Convert a byte string to a long integer. | |
4786 | ||
4787 | This is (essentially) the inverse of long_to_bytes(). | |
4788 | """ | |
4789 | acc = 0 | |
4790 | length = len(s) | |
4791 | if length % 4: | |
4792 | extra = (4 - length % 4) | |
4793 | s = b'\000' * extra + s | |
4794 | length = length + extra | |
4795 | for i in range(0, length, 4): | |
ac668111 | 4796 | acc = (acc << 32) + struct.unpack('>I', s[i:i + 4])[0] |
0a5445dd YCH |
4797 | return acc |
4798 | ||
4799 | ||
5bc880b9 YCH |
4800 | def ohdave_rsa_encrypt(data, exponent, modulus): |
4801 | ''' | |
4802 | Implement OHDave's RSA algorithm. See http://www.ohdave.com/rsa/ | |
4803 | ||
4804 | Input: | |
4805 | data: data to encrypt, bytes-like object | |
4806 | exponent, modulus: parameter e and N of RSA algorithm, both integer | |
4807 | Output: hex string of encrypted data | |
4808 | ||
4809 | Limitation: supports one block encryption only | |
4810 | ''' | |
4811 | ||
4812 | payload = int(binascii.hexlify(data[::-1]), 16) | |
4813 | encrypted = pow(payload, exponent, modulus) | |
4814 | return '%x' % encrypted | |
81bdc8fd YCH |
4815 | |
4816 | ||
f48409c7 YCH |
4817 | def pkcs1pad(data, length): |
4818 | """ | |
4819 | Padding input data with PKCS#1 scheme | |
4820 | ||
4821 | @param {int[]} data input data | |
4822 | @param {int} length target length | |
4823 | @returns {int[]} padded data | |
4824 | """ | |
4825 | if len(data) > length - 11: | |
4826 | raise ValueError('Input data too long for PKCS#1 padding') | |
4827 | ||
4828 | pseudo_random = [random.randint(0, 254) for _ in range(length - len(data) - 3)] | |
4829 | return [0, 2] + pseudo_random + [0] + data | |
4830 | ||
4831 | ||
7b2c3f47 | 4832 | def _base_n_table(n, table): |
4833 | if not table and not n: | |
4834 | raise ValueError('Either table or n must be specified') | |
612f2be5 | 4835 | table = (table or '0123456789abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ')[:n] |
4836 | ||
44f14eb4 | 4837 | if n and n != len(table): |
612f2be5 | 4838 | raise ValueError(f'base {n} exceeds table length {len(table)}') |
4839 | return table | |
59f898b7 | 4840 | |
5eb6bdce | 4841 | |
7b2c3f47 | 4842 | def encode_base_n(num, n=None, table=None): |
4843 | """Convert given int to a base-n string""" | |
612f2be5 | 4844 | table = _base_n_table(n, table) |
7b2c3f47 | 4845 | if not num: |
5eb6bdce YCH |
4846 | return table[0] |
4847 | ||
7b2c3f47 | 4848 | result, base = '', len(table) |
81bdc8fd | 4849 | while num: |
7b2c3f47 | 4850 | result = table[num % base] + result |
612f2be5 | 4851 | num = num // base |
7b2c3f47 | 4852 | return result |
4853 | ||
4854 | ||
4855 | def decode_base_n(string, n=None, table=None): | |
4856 | """Convert given base-n string to int""" | |
4857 | table = {char: index for index, char in enumerate(_base_n_table(n, table))} | |
4858 | result, base = 0, len(table) | |
4859 | for char in string: | |
4860 | result = result * base + table[char] | |
4861 | return result | |
4862 | ||
4863 | ||
4864 | def decode_base(value, digits): | |
4865 | write_string('DeprecationWarning: yt_dlp.utils.decode_base is deprecated ' | |
4866 | 'and may be removed in a future version. Use yt_dlp.decode_base_n instead') | |
4867 | return decode_base_n(value, table=digits) | |
f52354a8 YCH |
4868 | |
4869 | ||
4870 | def decode_packed_codes(code): | |
06b3fe29 | 4871 | mobj = re.search(PACKED_CODES_RE, code) |
a0566bbf | 4872 | obfuscated_code, base, count, symbols = mobj.groups() |
f52354a8 YCH |
4873 | base = int(base) |
4874 | count = int(count) | |
4875 | symbols = symbols.split('|') | |
4876 | symbol_table = {} | |
4877 | ||
4878 | while count: | |
4879 | count -= 1 | |
5eb6bdce | 4880 | base_n_count = encode_base_n(count, base) |
f52354a8 YCH |
4881 | symbol_table[base_n_count] = symbols[count] or base_n_count |
4882 | ||
4883 | return re.sub( | |
4884 | r'\b(\w+)\b', lambda mobj: symbol_table[mobj.group(0)], | |
a0566bbf | 4885 | obfuscated_code) |
e154c651 | 4886 | |
4887 | ||
1ced2221 S |
4888 | def caesar(s, alphabet, shift): |
4889 | if shift == 0: | |
4890 | return s | |
4891 | l = len(alphabet) | |
4892 | return ''.join( | |
4893 | alphabet[(alphabet.index(c) + shift) % l] if c in alphabet else c | |
4894 | for c in s) | |
4895 | ||
4896 | ||
4897 | def rot47(s): | |
4898 | return caesar(s, r'''!"#$%&'()*+,-./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{|}~''', 47) | |
4899 | ||
4900 | ||
e154c651 | 4901 | def parse_m3u8_attributes(attrib): |
4902 | info = {} | |
4903 | for (key, val) in re.findall(r'(?P<key>[A-Z0-9-]+)=(?P<val>"[^"]+"|[^",]+)(?:,|$)', attrib): | |
4904 | if val.startswith('"'): | |
4905 | val = val[1:-1] | |
4906 | info[key] = val | |
4907 | return info | |
1143535d YCH |
4908 | |
4909 | ||
4910 | def urshift(val, n): | |
4911 | return val >> n if val >= 0 else (val + 0x100000000) >> n | |
d3f8e038 YCH |
4912 | |
4913 | ||
4914 | # Based on png2str() written by @gdkchan and improved by @yokrysty | |
067aa17e | 4915 | # Originally posted at https://github.com/ytdl-org/youtube-dl/issues/9706 |
d3f8e038 YCH |
4916 | def decode_png(png_data): |
4917 | # Reference: https://www.w3.org/TR/PNG/ | |
4918 | header = png_data[8:] | |
4919 | ||
4920 | if png_data[:8] != b'\x89PNG\x0d\x0a\x1a\x0a' or header[4:8] != b'IHDR': | |
86e5f3ed | 4921 | raise OSError('Not a valid PNG file.') |
d3f8e038 YCH |
4922 | |
4923 | int_map = {1: '>B', 2: '>H', 4: '>I'} | |
ac668111 | 4924 | unpack_integer = lambda x: struct.unpack(int_map[len(x)], x)[0] |
d3f8e038 YCH |
4925 | |
4926 | chunks = [] | |
4927 | ||
4928 | while header: | |
4929 | length = unpack_integer(header[:4]) | |
4930 | header = header[4:] | |
4931 | ||
4932 | chunk_type = header[:4] | |
4933 | header = header[4:] | |
4934 | ||
4935 | chunk_data = header[:length] | |
4936 | header = header[length:] | |
4937 | ||
4938 | header = header[4:] # Skip CRC | |
4939 | ||
4940 | chunks.append({ | |
4941 | 'type': chunk_type, | |
4942 | 'length': length, | |
4943 | 'data': chunk_data | |
4944 | }) | |
4945 | ||
4946 | ihdr = chunks[0]['data'] | |
4947 | ||
4948 | width = unpack_integer(ihdr[:4]) | |
4949 | height = unpack_integer(ihdr[4:8]) | |
4950 | ||
4951 | idat = b'' | |
4952 | ||
4953 | for chunk in chunks: | |
4954 | if chunk['type'] == b'IDAT': | |
4955 | idat += chunk['data'] | |
4956 | ||
4957 | if not idat: | |
86e5f3ed | 4958 | raise OSError('Unable to read PNG data.') |
d3f8e038 YCH |
4959 | |
4960 | decompressed_data = bytearray(zlib.decompress(idat)) | |
4961 | ||
4962 | stride = width * 3 | |
4963 | pixels = [] | |
4964 | ||
4965 | def _get_pixel(idx): | |
4966 | x = idx % stride | |
4967 | y = idx // stride | |
4968 | return pixels[y][x] | |
4969 | ||
4970 | for y in range(height): | |
4971 | basePos = y * (1 + stride) | |
4972 | filter_type = decompressed_data[basePos] | |
4973 | ||
4974 | current_row = [] | |
4975 | ||
4976 | pixels.append(current_row) | |
4977 | ||
4978 | for x in range(stride): | |
4979 | color = decompressed_data[1 + basePos + x] | |
4980 | basex = y * stride + x | |
4981 | left = 0 | |
4982 | up = 0 | |
4983 | ||
4984 | if x > 2: | |
4985 | left = _get_pixel(basex - 3) | |
4986 | if y > 0: | |
4987 | up = _get_pixel(basex - stride) | |
4988 | ||
4989 | if filter_type == 1: # Sub | |
4990 | color = (color + left) & 0xff | |
4991 | elif filter_type == 2: # Up | |
4992 | color = (color + up) & 0xff | |
4993 | elif filter_type == 3: # Average | |
4994 | color = (color + ((left + up) >> 1)) & 0xff | |
4995 | elif filter_type == 4: # Paeth | |
4996 | a = left | |
4997 | b = up | |
4998 | c = 0 | |
4999 | ||
5000 | if x > 2 and y > 0: | |
5001 | c = _get_pixel(basex - stride - 3) | |
5002 | ||
5003 | p = a + b - c | |
5004 | ||
5005 | pa = abs(p - a) | |
5006 | pb = abs(p - b) | |
5007 | pc = abs(p - c) | |
5008 | ||
5009 | if pa <= pb and pa <= pc: | |
5010 | color = (color + a) & 0xff | |
5011 | elif pb <= pc: | |
5012 | color = (color + b) & 0xff | |
5013 | else: | |
5014 | color = (color + c) & 0xff | |
5015 | ||
5016 | current_row.append(color) | |
5017 | ||
5018 | return width, height, pixels | |
efa97bdc YCH |
5019 | |
5020 | ||
5021 | def write_xattr(path, key, value): | |
6f7563be | 5022 | # Windows: Write xattrs to NTFS Alternate Data Streams: |
5023 | # http://en.wikipedia.org/wiki/NTFS#Alternate_data_streams_.28ADS.29 | |
5024 | if compat_os_name == 'nt': | |
5025 | assert ':' not in key | |
5026 | assert os.path.exists(path) | |
efa97bdc YCH |
5027 | |
5028 | try: | |
6f7563be | 5029 | with open(f'{path}:{key}', 'wb') as f: |
5030 | f.write(value) | |
86e5f3ed | 5031 | except OSError as e: |
efa97bdc | 5032 | raise XAttrMetadataError(e.errno, e.strerror) |
6f7563be | 5033 | return |
efa97bdc | 5034 | |
6f7563be | 5035 | # UNIX Method 1. Use xattrs/pyxattrs modules |
efa97bdc | 5036 | |
6f7563be | 5037 | setxattr = None |
5038 | if getattr(xattr, '_yt_dlp__identifier', None) == 'pyxattr': | |
5039 | # Unicode arguments are not supported in pyxattr until version 0.5.0 | |
5040 | # See https://github.com/ytdl-org/youtube-dl/issues/5498 | |
5041 | if version_tuple(xattr.__version__) >= (0, 5, 0): | |
5042 | setxattr = xattr.set | |
5043 | elif xattr: | |
5044 | setxattr = xattr.setxattr | |
efa97bdc | 5045 | |
6f7563be | 5046 | if setxattr: |
5047 | try: | |
5048 | setxattr(path, key, value) | |
5049 | except OSError as e: | |
5050 | raise XAttrMetadataError(e.errno, e.strerror) | |
5051 | return | |
efa97bdc | 5052 | |
6f7563be | 5053 | # UNIX Method 2. Use setfattr/xattr executables |
5054 | exe = ('setfattr' if check_executable('setfattr', ['--version']) | |
5055 | else 'xattr' if check_executable('xattr', ['-h']) else None) | |
5056 | if not exe: | |
5057 | raise XAttrUnavailableError( | |
5058 | 'Couldn\'t find a tool to set the xattrs. Install either the python "xattr" or "pyxattr" modules or the ' | |
5059 | + ('"xattr" binary' if sys.platform != 'linux' else 'GNU "attr" package (which contains the "setfattr" tool)')) | |
efa97bdc | 5060 | |
0f06bcd7 | 5061 | value = value.decode() |
6f7563be | 5062 | try: |
f0c9fb96 | 5063 | _, stderr, returncode = Popen.run( |
6f7563be | 5064 | [exe, '-w', key, value, path] if exe == 'xattr' else [exe, '-n', key, '-v', value, path], |
e121e3ce | 5065 | text=True, stdout=subprocess.PIPE, stderr=subprocess.PIPE, stdin=subprocess.PIPE) |
6f7563be | 5066 | except OSError as e: |
5067 | raise XAttrMetadataError(e.errno, e.strerror) | |
f0c9fb96 | 5068 | if returncode: |
5069 | raise XAttrMetadataError(returncode, stderr) | |
0c265486 YCH |
5070 | |
5071 | ||
5072 | def random_birthday(year_field, month_field, day_field): | |
aa374bc7 AS |
5073 | start_date = datetime.date(1950, 1, 1) |
5074 | end_date = datetime.date(1995, 12, 31) | |
5075 | offset = random.randint(0, (end_date - start_date).days) | |
5076 | random_date = start_date + datetime.timedelta(offset) | |
0c265486 | 5077 | return { |
aa374bc7 AS |
5078 | year_field: str(random_date.year), |
5079 | month_field: str(random_date.month), | |
5080 | day_field: str(random_date.day), | |
0c265486 | 5081 | } |
732044af | 5082 | |
c76eb41b | 5083 | |
732044af | 5084 | # Templates for internet shortcut files, which are plain text files. |
e5a998f3 | 5085 | DOT_URL_LINK_TEMPLATE = '''\ |
732044af | 5086 | [InternetShortcut] |
5087 | URL=%(url)s | |
e5a998f3 | 5088 | ''' |
732044af | 5089 | |
e5a998f3 | 5090 | DOT_WEBLOC_LINK_TEMPLATE = '''\ |
732044af | 5091 | <?xml version="1.0" encoding="UTF-8"?> |
5092 | <!DOCTYPE plist PUBLIC "-//Apple//DTD PLIST 1.0//EN" "http://www.apple.com/DTDs/PropertyList-1.0.dtd"> | |
5093 | <plist version="1.0"> | |
5094 | <dict> | |
5095 | \t<key>URL</key> | |
5096 | \t<string>%(url)s</string> | |
5097 | </dict> | |
5098 | </plist> | |
e5a998f3 | 5099 | ''' |
732044af | 5100 | |
e5a998f3 | 5101 | DOT_DESKTOP_LINK_TEMPLATE = '''\ |
732044af | 5102 | [Desktop Entry] |
5103 | Encoding=UTF-8 | |
5104 | Name=%(filename)s | |
5105 | Type=Link | |
5106 | URL=%(url)s | |
5107 | Icon=text-html | |
e5a998f3 | 5108 | ''' |
732044af | 5109 | |
08438d2c | 5110 | LINK_TEMPLATES = { |
5111 | 'url': DOT_URL_LINK_TEMPLATE, | |
5112 | 'desktop': DOT_DESKTOP_LINK_TEMPLATE, | |
5113 | 'webloc': DOT_WEBLOC_LINK_TEMPLATE, | |
5114 | } | |
5115 | ||
732044af | 5116 | |
5117 | def iri_to_uri(iri): | |
5118 | """ | |
5119 | Converts an IRI (Internationalized Resource Identifier, allowing Unicode characters) to a URI (Uniform Resource Identifier, ASCII-only). | |
5120 | ||
5121 | 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. | |
5122 | """ | |
5123 | ||
14f25df2 | 5124 | iri_parts = urllib.parse.urlparse(iri) |
732044af | 5125 | |
5126 | if '[' in iri_parts.netloc: | |
5127 | raise ValueError('IPv6 URIs are not, yet, supported.') | |
5128 | # Querying `.netloc`, when there's only one bracket, also raises a ValueError. | |
5129 | ||
5130 | # 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. | |
5131 | ||
5132 | net_location = '' | |
5133 | if iri_parts.username: | |
f9934b96 | 5134 | net_location += urllib.parse.quote(iri_parts.username, safe=r"!$%&'()*+,~") |
732044af | 5135 | if iri_parts.password is not None: |
f9934b96 | 5136 | net_location += ':' + urllib.parse.quote(iri_parts.password, safe=r"!$%&'()*+,~") |
732044af | 5137 | net_location += '@' |
5138 | ||
0f06bcd7 | 5139 | net_location += iri_parts.hostname.encode('idna').decode() # Punycode for Unicode hostnames. |
732044af | 5140 | # The 'idna' encoding produces ASCII text. |
5141 | if iri_parts.port is not None and iri_parts.port != 80: | |
5142 | net_location += ':' + str(iri_parts.port) | |
5143 | ||
f9934b96 | 5144 | return urllib.parse.urlunparse( |
732044af | 5145 | (iri_parts.scheme, |
5146 | net_location, | |
5147 | ||
f9934b96 | 5148 | urllib.parse.quote_plus(iri_parts.path, safe=r"!$%&'()*+,/:;=@|~"), |
732044af | 5149 | |
5150 | # Unsure about the `safe` argument, since this is a legacy way of handling parameters. | |
f9934b96 | 5151 | urllib.parse.quote_plus(iri_parts.params, safe=r"!$%&'()*+,/:;=@|~"), |
732044af | 5152 | |
5153 | # Not totally sure about the `safe` argument, since the source does not explicitly mention the query URI component. | |
f9934b96 | 5154 | urllib.parse.quote_plus(iri_parts.query, safe=r"!$%&'()*+,/:;=?@{|}~"), |
732044af | 5155 | |
f9934b96 | 5156 | urllib.parse.quote_plus(iri_parts.fragment, safe=r"!#$%&'()*+,/:;=?@{|}~"))) |
732044af | 5157 | |
5158 | # Source for `safe` arguments: https://url.spec.whatwg.org/#percent-encoded-bytes. | |
5159 | ||
5160 | ||
5161 | def to_high_limit_path(path): | |
5162 | if sys.platform in ['win32', 'cygwin']: | |
5163 | # Work around MAX_PATH limitation on Windows. The maximum allowed length for the individual path segments may still be quite limited. | |
e5a998f3 | 5164 | return '\\\\?\\' + os.path.abspath(path) |
732044af | 5165 | |
5166 | return path | |
76d321f6 | 5167 | |
c76eb41b | 5168 | |
7b2c3f47 | 5169 | def format_field(obj, field=None, template='%s', ignore=NO_DEFAULT, default='', func=IDENTITY): |
e0ddbd02 | 5170 | val = traverse_obj(obj, *variadic(field)) |
7b2c3f47 | 5171 | if (not val and val != 0) if ignore is NO_DEFAULT else val in variadic(ignore): |
e0ddbd02 | 5172 | return default |
7b2c3f47 | 5173 | return template % func(val) |
00dd0cd5 | 5174 | |
5175 | ||
5176 | def clean_podcast_url(url): | |
5177 | return re.sub(r'''(?x) | |
5178 | (?: | |
5179 | (?: | |
5180 | chtbl\.com/track| | |
5181 | media\.blubrry\.com| # https://create.blubrry.com/resources/podcast-media-download-statistics/getting-started/ | |
5182 | play\.podtrac\.com | |
5183 | )/[^/]+| | |
5184 | (?:dts|www)\.podtrac\.com/(?:pts/)?redirect\.[0-9a-z]{3,4}| # http://analytics.podtrac.com/how-to-measure | |
5185 | flex\.acast\.com| | |
5186 | pd(?: | |
5187 | cn\.co| # https://podcorn.com/analytics-prefix/ | |
5188 | st\.fm # https://podsights.com/docs/ | |
5189 | )/e | |
5190 | )/''', '', url) | |
ffcb8191 THD |
5191 | |
5192 | ||
5193 | _HEX_TABLE = '0123456789abcdef' | |
5194 | ||
5195 | ||
5196 | def random_uuidv4(): | |
5197 | return re.sub(r'[xy]', lambda x: _HEX_TABLE[random.randint(0, 15)], 'xxxxxxxx-xxxx-4xxx-yxxx-xxxxxxxxxxxx') | |
0202b52a | 5198 | |
5199 | ||
5200 | def make_dir(path, to_screen=None): | |
5201 | try: | |
5202 | dn = os.path.dirname(path) | |
5203 | if dn and not os.path.exists(dn): | |
5204 | os.makedirs(dn) | |
5205 | return True | |
86e5f3ed | 5206 | except OSError as err: |
0202b52a | 5207 | if callable(to_screen) is not None: |
5208 | to_screen('unable to create directory ' + error_to_compat_str(err)) | |
5209 | return False | |
f74980cb | 5210 | |
5211 | ||
5212 | def get_executable_path(): | |
b5899f4f | 5213 | from .update import _get_variant_and_executable_path |
c487cf00 | 5214 | |
b5899f4f | 5215 | return os.path.dirname(os.path.abspath(_get_variant_and_executable_path()[1])) |
f74980cb | 5216 | |
5217 | ||
2f567473 | 5218 | def load_plugins(name, suffix, namespace): |
3ae5e797 | 5219 | classes = {} |
19a03940 | 5220 | with contextlib.suppress(FileNotFoundError): |
019a94f7 ÁS |
5221 | plugins_spec = importlib.util.spec_from_file_location( |
5222 | name, os.path.join(get_executable_path(), 'ytdlp_plugins', name, '__init__.py')) | |
5223 | plugins = importlib.util.module_from_spec(plugins_spec) | |
5224 | sys.modules[plugins_spec.name] = plugins | |
5225 | plugins_spec.loader.exec_module(plugins) | |
f74980cb | 5226 | for name in dir(plugins): |
2f567473 | 5227 | if name in namespace: |
5228 | continue | |
5229 | if not name.endswith(suffix): | |
f74980cb | 5230 | continue |
5231 | klass = getattr(plugins, name) | |
3ae5e797 | 5232 | classes[name] = namespace[name] = klass |
f74980cb | 5233 | return classes |
06167fbb | 5234 | |
5235 | ||
325ebc17 | 5236 | def traverse_obj( |
352d63fd | 5237 | obj, *path_list, default=None, expected_type=None, get_all=True, |
325ebc17 | 5238 | casesense=True, is_user_input=False, traverse_string=False): |
324ad820 | 5239 | ''' Traverse nested list/dict/tuple |
8f334380 | 5240 | @param path_list A list of paths which are checked one by one. |
19a03940 | 5241 | Each path is a list of keys where each key is a: |
5242 | - None: Do nothing | |
5243 | - string: A dictionary key | |
5244 | - int: An index into a list | |
5245 | - tuple: A list of keys all of which will be traversed | |
5246 | - Ellipsis: Fetch all values in the object | |
5247 | - Function: Takes the key and value as arguments | |
5248 | and returns whether the key matches or not | |
325ebc17 | 5249 | @param default Default value to return |
352d63fd | 5250 | @param expected_type Only accept final value of this type (Can also be any callable) |
5251 | @param get_all Return all the values obtained from a path or only the first one | |
324ad820 | 5252 | @param casesense Whether to consider dictionary keys as case sensitive |
5253 | @param is_user_input Whether the keys are generated from user input. If True, | |
5254 | strings are converted to int/slice if necessary | |
5255 | @param traverse_string Whether to traverse inside strings. If True, any | |
5256 | non-compatible object will also be converted into a string | |
8f334380 | 5257 | # TODO: Write tests |
324ad820 | 5258 | ''' |
325ebc17 | 5259 | if not casesense: |
dbf5416a | 5260 | _lower = lambda k: (k.lower() if isinstance(k, str) else k) |
8f334380 | 5261 | path_list = (map(_lower, variadic(path)) for path in path_list) |
5262 | ||
5263 | def _traverse_obj(obj, path, _current_depth=0): | |
5264 | nonlocal depth | |
5265 | path = tuple(variadic(path)) | |
5266 | for i, key in enumerate(path): | |
1797b073 | 5267 | if None in (key, obj): |
5268 | return obj | |
8f334380 | 5269 | if isinstance(key, (list, tuple)): |
5270 | obj = [_traverse_obj(obj, sub_key, _current_depth) for sub_key in key] | |
5271 | key = ... | |
5272 | if key is ...: | |
5273 | obj = (obj.values() if isinstance(obj, dict) | |
5274 | else obj if isinstance(obj, (list, tuple, LazyList)) | |
5275 | else str(obj) if traverse_string else []) | |
5276 | _current_depth += 1 | |
5277 | depth = max(depth, _current_depth) | |
5278 | return [_traverse_obj(inner_obj, path[i + 1:], _current_depth) for inner_obj in obj] | |
2614f646 | 5279 | elif callable(key): |
5280 | if isinstance(obj, (list, tuple, LazyList)): | |
5281 | obj = enumerate(obj) | |
5282 | elif isinstance(obj, dict): | |
5283 | obj = obj.items() | |
5284 | else: | |
5285 | if not traverse_string: | |
5286 | return None | |
5287 | obj = str(obj) | |
5288 | _current_depth += 1 | |
5289 | depth = max(depth, _current_depth) | |
e6f868a6 | 5290 | return [_traverse_obj(v, path[i + 1:], _current_depth) for k, v in obj if try_call(key, args=(k, v))] |
575e17a1 | 5291 | elif isinstance(obj, dict) and not (is_user_input and key == ':'): |
325ebc17 | 5292 | obj = (obj.get(key) if casesense or (key in obj) |
5293 | else next((v for k, v in obj.items() if _lower(k) == key), None)) | |
5294 | else: | |
5295 | if is_user_input: | |
5296 | key = (int_or_none(key) if ':' not in key | |
5297 | else slice(*map(int_or_none, key.split(':')))) | |
8f334380 | 5298 | if key == slice(None): |
575e17a1 | 5299 | return _traverse_obj(obj, (..., *path[i + 1:]), _current_depth) |
325ebc17 | 5300 | if not isinstance(key, (int, slice)): |
9fea350f | 5301 | return None |
8f334380 | 5302 | if not isinstance(obj, (list, tuple, LazyList)): |
325ebc17 | 5303 | if not traverse_string: |
5304 | return None | |
5305 | obj = str(obj) | |
5306 | try: | |
5307 | obj = obj[key] | |
5308 | except IndexError: | |
324ad820 | 5309 | return None |
325ebc17 | 5310 | return obj |
5311 | ||
352d63fd | 5312 | if isinstance(expected_type, type): |
5313 | type_test = lambda val: val if isinstance(val, expected_type) else None | |
352d63fd | 5314 | else: |
7b2c3f47 | 5315 | type_test = expected_type or IDENTITY |
352d63fd | 5316 | |
8f334380 | 5317 | for path in path_list: |
5318 | depth = 0 | |
5319 | val = _traverse_obj(obj, path) | |
325ebc17 | 5320 | if val is not None: |
8f334380 | 5321 | if depth: |
5322 | for _ in range(depth - 1): | |
6586bca9 | 5323 | val = itertools.chain.from_iterable(v for v in val if v is not None) |
352d63fd | 5324 | val = [v for v in map(type_test, val) if v is not None] |
8f334380 | 5325 | if val: |
352d63fd | 5326 | return val if get_all else val[0] |
5327 | else: | |
5328 | val = type_test(val) | |
5329 | if val is not None: | |
8f334380 | 5330 | return val |
325ebc17 | 5331 | return default |
324ad820 | 5332 | |
5333 | ||
5334 | def traverse_dict(dictn, keys, casesense=True): | |
ee8dd27a | 5335 | write_string('DeprecationWarning: yt_dlp.utils.traverse_dict is deprecated ' |
5336 | 'and may be removed in a future version. Use yt_dlp.utils.traverse_obj instead') | |
5337 | return traverse_obj(dictn, keys, casesense=casesense, is_user_input=True, traverse_string=True) | |
6606817a | 5338 | |
5339 | ||
ff91cf74 | 5340 | def get_first(obj, keys, **kwargs): |
5341 | return traverse_obj(obj, (..., *variadic(keys)), **kwargs, get_all=False) | |
5342 | ||
5343 | ||
4b4b7f74 | 5344 | def variadic(x, allowed_types=(str, bytes, dict)): |
cb89cfc1 | 5345 | return x if isinstance(x, collections.abc.Iterable) and not isinstance(x, allowed_types) else (x,) |
bd50a52b THD |
5346 | |
5347 | ||
3e9b66d7 LNO |
5348 | def time_seconds(**kwargs): |
5349 | t = datetime.datetime.now(datetime.timezone(datetime.timedelta(**kwargs))) | |
5350 | return t.timestamp() | |
5351 | ||
5352 | ||
49fa4d9a N |
5353 | # create a JSON Web Signature (jws) with HS256 algorithm |
5354 | # the resulting format is in JWS Compact Serialization | |
5355 | # implemented following JWT https://www.rfc-editor.org/rfc/rfc7519.html | |
5356 | # implemented following JWS https://www.rfc-editor.org/rfc/rfc7515.html | |
5357 | def jwt_encode_hs256(payload_data, key, headers={}): | |
5358 | header_data = { | |
5359 | 'alg': 'HS256', | |
5360 | 'typ': 'JWT', | |
5361 | } | |
5362 | if headers: | |
5363 | header_data.update(headers) | |
0f06bcd7 | 5364 | header_b64 = base64.b64encode(json.dumps(header_data).encode()) |
5365 | payload_b64 = base64.b64encode(json.dumps(payload_data).encode()) | |
5366 | h = hmac.new(key.encode(), header_b64 + b'.' + payload_b64, hashlib.sha256) | |
49fa4d9a N |
5367 | signature_b64 = base64.b64encode(h.digest()) |
5368 | token = header_b64 + b'.' + payload_b64 + b'.' + signature_b64 | |
5369 | return token | |
819e0531 | 5370 | |
5371 | ||
16b0d7e6 | 5372 | # can be extended in future to verify the signature and parse header and return the algorithm used if it's not HS256 |
5373 | def jwt_decode_hs256(jwt): | |
5374 | header_b64, payload_b64, signature_b64 = jwt.split('.') | |
5375 | payload_data = json.loads(base64.urlsafe_b64decode(payload_b64)) | |
5376 | return payload_data | |
5377 | ||
5378 | ||
53973b4d | 5379 | WINDOWS_VT_MODE = False if compat_os_name == 'nt' else None |
5380 | ||
5381 | ||
0b9c08b4 | 5382 | @functools.cache |
819e0531 | 5383 | def supports_terminal_sequences(stream): |
5384 | if compat_os_name == 'nt': | |
8a82af35 | 5385 | if not WINDOWS_VT_MODE: |
819e0531 | 5386 | return False |
5387 | elif not os.getenv('TERM'): | |
5388 | return False | |
5389 | try: | |
5390 | return stream.isatty() | |
5391 | except BaseException: | |
5392 | return False | |
5393 | ||
5394 | ||
53973b4d | 5395 | def windows_enable_vt_mode(): # TODO: Do this the proper way https://bugs.python.org/issue30075 |
8a82af35 | 5396 | if get_windows_version() < (10, 0, 10586): |
53973b4d | 5397 | return |
5398 | global WINDOWS_VT_MODE | |
53973b4d | 5399 | try: |
f0c9fb96 | 5400 | Popen.run('', shell=True) |
53973b4d | 5401 | except Exception: |
5402 | return | |
5403 | ||
5404 | WINDOWS_VT_MODE = True | |
5405 | supports_terminal_sequences.cache_clear() | |
5406 | ||
5407 | ||
ec11a9f4 | 5408 | _terminal_sequences_re = re.compile('\033\\[[^m]+m') |
5409 | ||
5410 | ||
5411 | def remove_terminal_sequences(string): | |
5412 | return _terminal_sequences_re.sub('', string) | |
5413 | ||
5414 | ||
5415 | def number_of_digits(number): | |
5416 | return len('%d' % number) | |
34921b43 | 5417 | |
5418 | ||
5419 | def join_nonempty(*values, delim='-', from_dict=None): | |
5420 | if from_dict is not None: | |
7b2c3f47 | 5421 | values = (traverse_obj(from_dict, variadic(v)) for v in values) |
34921b43 | 5422 | return delim.join(map(str, filter(None, values))) |
06e57990 | 5423 | |
5424 | ||
27231526 ZM |
5425 | def scale_thumbnails_to_max_format_width(formats, thumbnails, url_width_re): |
5426 | """ | |
5427 | Find the largest format dimensions in terms of video width and, for each thumbnail: | |
5428 | * Modify the URL: Match the width with the provided regex and replace with the former width | |
5429 | * Update dimensions | |
5430 | ||
5431 | This function is useful with video services that scale the provided thumbnails on demand | |
5432 | """ | |
5433 | _keys = ('width', 'height') | |
5434 | max_dimensions = max( | |
86e5f3ed | 5435 | (tuple(format.get(k) or 0 for k in _keys) for format in formats), |
27231526 ZM |
5436 | default=(0, 0)) |
5437 | if not max_dimensions[0]: | |
5438 | return thumbnails | |
5439 | return [ | |
5440 | merge_dicts( | |
5441 | {'url': re.sub(url_width_re, str(max_dimensions[0]), thumbnail['url'])}, | |
5442 | dict(zip(_keys, max_dimensions)), thumbnail) | |
5443 | for thumbnail in thumbnails | |
5444 | ] | |
5445 | ||
5446 | ||
93c8410d LNO |
5447 | def parse_http_range(range): |
5448 | """ Parse value of "Range" or "Content-Range" HTTP header into tuple. """ | |
5449 | if not range: | |
5450 | return None, None, None | |
5451 | crg = re.search(r'bytes[ =](\d+)-(\d+)?(?:/(\d+))?', range) | |
5452 | if not crg: | |
5453 | return None, None, None | |
5454 | return int(crg.group(1)), int_or_none(crg.group(2)), int_or_none(crg.group(3)) | |
5455 | ||
5456 | ||
6b9e832d | 5457 | def read_stdin(what): |
5458 | eof = 'Ctrl+Z' if compat_os_name == 'nt' else 'Ctrl+D' | |
5459 | write_string(f'Reading {what} from STDIN - EOF ({eof}) to end:\n') | |
5460 | return sys.stdin | |
5461 | ||
5462 | ||
a904a7f8 L |
5463 | def determine_file_encoding(data): |
5464 | """ | |
88f60feb | 5465 | Detect the text encoding used |
a904a7f8 L |
5466 | @returns (encoding, bytes to skip) |
5467 | """ | |
5468 | ||
88f60feb | 5469 | # BOM marks are given priority over declarations |
a904a7f8 | 5470 | for bom, enc in BOMS: |
a904a7f8 L |
5471 | if data.startswith(bom): |
5472 | return enc, len(bom) | |
5473 | ||
88f60feb | 5474 | # Strip off all null bytes to match even when UTF-16 or UTF-32 is used. |
5475 | # We ignore the endianness to get a good enough match | |
a904a7f8 | 5476 | data = data.replace(b'\0', b'') |
88f60feb | 5477 | mobj = re.match(rb'(?m)^#\s*coding\s*:\s*(\S+)\s*$', data) |
5478 | return mobj.group(1).decode() if mobj else None, 0 | |
a904a7f8 L |
5479 | |
5480 | ||
06e57990 | 5481 | class Config: |
5482 | own_args = None | |
9e491463 | 5483 | parsed_args = None |
06e57990 | 5484 | filename = None |
5485 | __initialized = False | |
5486 | ||
5487 | def __init__(self, parser, label=None): | |
9e491463 | 5488 | self.parser, self.label = parser, label |
06e57990 | 5489 | self._loaded_paths, self.configs = set(), [] |
5490 | ||
5491 | def init(self, args=None, filename=None): | |
5492 | assert not self.__initialized | |
284a60c5 | 5493 | self.own_args, self.filename = args, filename |
5494 | return self.load_configs() | |
5495 | ||
5496 | def load_configs(self): | |
65662dff | 5497 | directory = '' |
284a60c5 | 5498 | if self.filename: |
5499 | location = os.path.realpath(self.filename) | |
65662dff | 5500 | directory = os.path.dirname(location) |
06e57990 | 5501 | if location in self._loaded_paths: |
5502 | return False | |
5503 | self._loaded_paths.add(location) | |
5504 | ||
284a60c5 | 5505 | self.__initialized = True |
5506 | opts, _ = self.parser.parse_known_args(self.own_args) | |
5507 | self.parsed_args = self.own_args | |
9e491463 | 5508 | for location in opts.config_locations or []: |
6b9e832d | 5509 | if location == '-': |
5510 | self.append_config(shlex.split(read_stdin('options'), comments=True), label='stdin') | |
5511 | continue | |
65662dff | 5512 | location = os.path.join(directory, expand_path(location)) |
06e57990 | 5513 | if os.path.isdir(location): |
5514 | location = os.path.join(location, 'yt-dlp.conf') | |
5515 | if not os.path.exists(location): | |
9e491463 | 5516 | self.parser.error(f'config location {location} does not exist') |
06e57990 | 5517 | self.append_config(self.read_file(location), location) |
5518 | return True | |
5519 | ||
5520 | def __str__(self): | |
5521 | label = join_nonempty( | |
5522 | self.label, 'config', f'"{self.filename}"' if self.filename else '', | |
5523 | delim=' ') | |
5524 | return join_nonempty( | |
5525 | self.own_args is not None and f'{label[0].upper()}{label[1:]}: {self.hide_login_info(self.own_args)}', | |
5526 | *(f'\n{c}'.replace('\n', '\n| ')[1:] for c in self.configs), | |
5527 | delim='\n') | |
5528 | ||
5529 | @staticmethod | |
5530 | def read_file(filename, default=[]): | |
5531 | try: | |
a904a7f8 | 5532 | optionf = open(filename, 'rb') |
86e5f3ed | 5533 | except OSError: |
06e57990 | 5534 | return default # silently skip if file is not present |
a904a7f8 L |
5535 | try: |
5536 | enc, skip = determine_file_encoding(optionf.read(512)) | |
5537 | optionf.seek(skip, io.SEEK_SET) | |
5538 | except OSError: | |
5539 | enc = None # silently skip read errors | |
06e57990 | 5540 | try: |
5541 | # FIXME: https://github.com/ytdl-org/youtube-dl/commit/dfe5fa49aed02cf36ba9f743b11b0903554b5e56 | |
a904a7f8 | 5542 | contents = optionf.read().decode(enc or preferredencoding()) |
f9934b96 | 5543 | res = shlex.split(contents, comments=True) |
44a6fcff | 5544 | except Exception as err: |
5545 | raise ValueError(f'Unable to parse "{filename}": {err}') | |
06e57990 | 5546 | finally: |
5547 | optionf.close() | |
5548 | return res | |
5549 | ||
5550 | @staticmethod | |
5551 | def hide_login_info(opts): | |
86e5f3ed | 5552 | PRIVATE_OPTS = {'-p', '--password', '-u', '--username', '--video-password', '--ap-password', '--ap-username'} |
06e57990 | 5553 | eqre = re.compile('^(?P<key>' + ('|'.join(re.escape(po) for po in PRIVATE_OPTS)) + ')=.+$') |
5554 | ||
5555 | def _scrub_eq(o): | |
5556 | m = eqre.match(o) | |
5557 | if m: | |
5558 | return m.group('key') + '=PRIVATE' | |
5559 | else: | |
5560 | return o | |
5561 | ||
5562 | opts = list(map(_scrub_eq, opts)) | |
5563 | for idx, opt in enumerate(opts): | |
5564 | if opt in PRIVATE_OPTS and idx + 1 < len(opts): | |
5565 | opts[idx + 1] = 'PRIVATE' | |
5566 | return opts | |
5567 | ||
5568 | def append_config(self, *args, label=None): | |
9e491463 | 5569 | config = type(self)(self.parser, label) |
06e57990 | 5570 | config._loaded_paths = self._loaded_paths |
5571 | if config.init(*args): | |
5572 | self.configs.append(config) | |
5573 | ||
5574 | @property | |
5575 | def all_args(self): | |
5576 | for config in reversed(self.configs): | |
5577 | yield from config.all_args | |
9e491463 | 5578 | yield from self.parsed_args or [] |
5579 | ||
5580 | def parse_known_args(self, **kwargs): | |
5581 | return self.parser.parse_known_args(self.all_args, **kwargs) | |
06e57990 | 5582 | |
5583 | def parse_args(self): | |
9e491463 | 5584 | return self.parser.parse_args(self.all_args) |
da42679b LNO |
5585 | |
5586 | ||
5587 | class WebSocketsWrapper(): | |
5588 | """Wraps websockets module to use in non-async scopes""" | |
abfecb7b | 5589 | pool = None |
da42679b | 5590 | |
3cea3edd | 5591 | def __init__(self, url, headers=None, connect=True): |
059bc4db | 5592 | self.loop = asyncio.new_event_loop() |
9cd08050 | 5593 | # XXX: "loop" is deprecated |
5594 | self.conn = websockets.connect( | |
5595 | url, extra_headers=headers, ping_interval=None, | |
5596 | close_timeout=float('inf'), loop=self.loop, ping_timeout=float('inf')) | |
3cea3edd LNO |
5597 | if connect: |
5598 | self.__enter__() | |
15dfb392 | 5599 | atexit.register(self.__exit__, None, None, None) |
da42679b LNO |
5600 | |
5601 | def __enter__(self): | |
3cea3edd | 5602 | if not self.pool: |
9cd08050 | 5603 | self.pool = self.run_with_loop(self.conn.__aenter__(), self.loop) |
da42679b LNO |
5604 | return self |
5605 | ||
5606 | def send(self, *args): | |
5607 | self.run_with_loop(self.pool.send(*args), self.loop) | |
5608 | ||
5609 | def recv(self, *args): | |
5610 | return self.run_with_loop(self.pool.recv(*args), self.loop) | |
5611 | ||
5612 | def __exit__(self, type, value, traceback): | |
5613 | try: | |
5614 | return self.run_with_loop(self.conn.__aexit__(type, value, traceback), self.loop) | |
5615 | finally: | |
5616 | self.loop.close() | |
15dfb392 | 5617 | self._cancel_all_tasks(self.loop) |
da42679b LNO |
5618 | |
5619 | # taken from https://github.com/python/cpython/blob/3.9/Lib/asyncio/runners.py with modifications | |
5620 | # for contributors: If there's any new library using asyncio needs to be run in non-async, move these function out of this class | |
5621 | @staticmethod | |
5622 | def run_with_loop(main, loop): | |
059bc4db | 5623 | if not asyncio.iscoroutine(main): |
da42679b LNO |
5624 | raise ValueError(f'a coroutine was expected, got {main!r}') |
5625 | ||
5626 | try: | |
5627 | return loop.run_until_complete(main) | |
5628 | finally: | |
5629 | loop.run_until_complete(loop.shutdown_asyncgens()) | |
5630 | if hasattr(loop, 'shutdown_default_executor'): | |
5631 | loop.run_until_complete(loop.shutdown_default_executor()) | |
5632 | ||
5633 | @staticmethod | |
5634 | def _cancel_all_tasks(loop): | |
059bc4db | 5635 | to_cancel = asyncio.all_tasks(loop) |
da42679b LNO |
5636 | |
5637 | if not to_cancel: | |
5638 | return | |
5639 | ||
5640 | for task in to_cancel: | |
5641 | task.cancel() | |
5642 | ||
9cd08050 | 5643 | # XXX: "loop" is removed in python 3.10+ |
da42679b | 5644 | loop.run_until_complete( |
059bc4db | 5645 | asyncio.gather(*to_cancel, loop=loop, return_exceptions=True)) |
da42679b LNO |
5646 | |
5647 | for task in to_cancel: | |
5648 | if task.cancelled(): | |
5649 | continue | |
5650 | if task.exception() is not None: | |
5651 | loop.call_exception_handler({ | |
5652 | 'message': 'unhandled exception during asyncio.run() shutdown', | |
5653 | 'exception': task.exception(), | |
5654 | 'task': task, | |
5655 | }) | |
5656 | ||
5657 | ||
8b7539d2 | 5658 | def merge_headers(*dicts): |
08d30158 | 5659 | """Merge dicts of http headers case insensitively, prioritizing the latter ones""" |
76aa9913 | 5660 | return {k.title(): v for k, v in itertools.chain.from_iterable(map(dict.items, dicts))} |
28787f16 | 5661 | |
5662 | ||
b1f94422 | 5663 | def cached_method(f): |
5664 | """Cache a method""" | |
5665 | signature = inspect.signature(f) | |
5666 | ||
5667 | @functools.wraps(f) | |
5668 | def wrapper(self, *args, **kwargs): | |
5669 | bound_args = signature.bind(self, *args, **kwargs) | |
5670 | bound_args.apply_defaults() | |
5671 | key = tuple(bound_args.arguments.values()) | |
5672 | ||
5673 | if not hasattr(self, '__cached_method__cache'): | |
5674 | self.__cached_method__cache = {} | |
5675 | cache = self.__cached_method__cache.setdefault(f.__name__, {}) | |
5676 | if key not in cache: | |
5677 | cache[key] = f(self, *args, **kwargs) | |
5678 | return cache[key] | |
5679 | return wrapper | |
5680 | ||
5681 | ||
28787f16 | 5682 | class classproperty: |
b1f94422 | 5683 | """property access for class methods""" |
c487cf00 | 5684 | |
5685 | def __init__(self, func): | |
5686 | functools.update_wrapper(self, func) | |
5687 | self.func = func | |
28787f16 | 5688 | |
5689 | def __get__(self, _, cls): | |
c487cf00 | 5690 | return self.func(cls) |
19a03940 | 5691 | |
5692 | ||
64fa820c | 5693 | class Namespace(types.SimpleNamespace): |
591bb9d3 | 5694 | """Immutable namespace""" |
591bb9d3 | 5695 | |
7896214c | 5696 | def __iter__(self): |
64fa820c | 5697 | return iter(self.__dict__.values()) |
7896214c | 5698 | |
64fa820c | 5699 | @property |
5700 | def items_(self): | |
5701 | return self.__dict__.items() | |
9b8ee23b | 5702 | |
5703 | ||
8dc59305 | 5704 | MEDIA_EXTENSIONS = Namespace( |
5705 | common_video=('avi', 'flv', 'mkv', 'mov', 'mp4', 'webm'), | |
5706 | video=('3g2', '3gp', 'f4v', 'mk3d', 'divx', 'mpg', 'ogv', 'm4v', 'wmv'), | |
5707 | common_audio=('aiff', 'alac', 'flac', 'm4a', 'mka', 'mp3', 'ogg', 'opus', 'wav'), | |
5708 | audio=('aac', 'ape', 'asf', 'f4a', 'f4b', 'm4b', 'm4p', 'm4r', 'oga', 'ogx', 'spx', 'vorbis', 'wma'), | |
5709 | thumbnails=('jpg', 'png', 'webp'), | |
5710 | storyboards=('mhtml', ), | |
5711 | subtitles=('srt', 'vtt', 'ass', 'lrc'), | |
5712 | manifests=('f4f', 'f4m', 'm3u8', 'smil', 'mpd'), | |
5713 | ) | |
5714 | MEDIA_EXTENSIONS.video += MEDIA_EXTENSIONS.common_video | |
5715 | MEDIA_EXTENSIONS.audio += MEDIA_EXTENSIONS.common_audio | |
5716 | ||
5717 | KNOWN_EXTENSIONS = (*MEDIA_EXTENSIONS.video, *MEDIA_EXTENSIONS.audio, *MEDIA_EXTENSIONS.manifests) | |
5718 | ||
5719 | ||
be5c1ae8 | 5720 | class RetryManager: |
5721 | """Usage: | |
5722 | for retry in RetryManager(...): | |
5723 | try: | |
5724 | ... | |
5725 | except SomeException as err: | |
5726 | retry.error = err | |
5727 | continue | |
5728 | """ | |
5729 | attempt, _error = 0, None | |
5730 | ||
5731 | def __init__(self, _retries, _error_callback, **kwargs): | |
5732 | self.retries = _retries or 0 | |
5733 | self.error_callback = functools.partial(_error_callback, **kwargs) | |
5734 | ||
5735 | def _should_retry(self): | |
5736 | return self._error is not NO_DEFAULT and self.attempt <= self.retries | |
5737 | ||
5738 | @property | |
5739 | def error(self): | |
5740 | if self._error is NO_DEFAULT: | |
5741 | return None | |
5742 | return self._error | |
5743 | ||
5744 | @error.setter | |
5745 | def error(self, value): | |
5746 | self._error = value | |
5747 | ||
5748 | def __iter__(self): | |
5749 | while self._should_retry(): | |
5750 | self.error = NO_DEFAULT | |
5751 | self.attempt += 1 | |
5752 | yield self | |
5753 | if self.error: | |
5754 | self.error_callback(self.error, self.attempt, self.retries) | |
5755 | ||
5756 | @staticmethod | |
5757 | def report_retry(e, count, retries, *, sleep_func, info, warn, error=None, suffix=None): | |
5758 | """Utility function for reporting retries""" | |
5759 | if count > retries: | |
5760 | if error: | |
5761 | return error(f'{e}. Giving up after {count - 1} retries') if count > 1 else error(str(e)) | |
5762 | raise e | |
5763 | ||
5764 | if not count: | |
5765 | return warn(e) | |
5766 | elif isinstance(e, ExtractorError): | |
3ce29336 | 5767 | e = remove_end(str_or_none(e.cause) or e.orig_msg, '.') |
be5c1ae8 | 5768 | warn(f'{e}. Retrying{format_field(suffix, None, " %s")} ({count}/{retries})...') |
5769 | ||
5770 | delay = float_or_none(sleep_func(n=count - 1)) if callable(sleep_func) else sleep_func | |
5771 | if delay: | |
5772 | info(f'Sleeping {delay:.2f} seconds ...') | |
5773 | time.sleep(delay) | |
5774 | ||
5775 | ||
0647d925 | 5776 | def make_archive_id(ie, video_id): |
5777 | ie_key = ie if isinstance(ie, str) else ie.ie_key() | |
5778 | return f'{ie_key.lower()} {video_id}' | |
5779 | ||
5780 | ||
a1c5bd82 | 5781 | def truncate_string(s, left, right=0): |
5782 | assert left > 3 and right >= 0 | |
5783 | if s is None or len(s) <= left + right: | |
5784 | return s | |
5785 | return f'{s[:left-3]}...{s[-right:]}' | |
5786 | ||
5787 | ||
9b8ee23b | 5788 | # Deprecated |
5789 | has_certifi = bool(certifi) | |
5790 | has_websockets = bool(websockets) |