]>
Commit | Line | Data |
---|---|---|
1 | import base64 | |
2 | import binascii | |
3 | import calendar | |
4 | import codecs | |
5 | import collections | |
6 | import collections.abc | |
7 | import contextlib | |
8 | import datetime as dt | |
9 | import email.header | |
10 | import email.utils | |
11 | import errno | |
12 | import hashlib | |
13 | import hmac | |
14 | import html.entities | |
15 | import html.parser | |
16 | import inspect | |
17 | import io | |
18 | import itertools | |
19 | import json | |
20 | import locale | |
21 | import math | |
22 | import mimetypes | |
23 | import netrc | |
24 | import operator | |
25 | import os | |
26 | import platform | |
27 | import random | |
28 | import re | |
29 | import shlex | |
30 | import socket | |
31 | import ssl | |
32 | import struct | |
33 | import subprocess | |
34 | import sys | |
35 | import tempfile | |
36 | import time | |
37 | import traceback | |
38 | import types | |
39 | import unicodedata | |
40 | import urllib.error | |
41 | import urllib.parse | |
42 | import urllib.request | |
43 | import xml.etree.ElementTree | |
44 | ||
45 | from . import traversal | |
46 | ||
47 | from ..compat import functools # isort: split | |
48 | from ..compat import ( | |
49 | compat_etree_fromstring, | |
50 | compat_expanduser, | |
51 | compat_HTMLParseError, | |
52 | compat_os_name, | |
53 | ) | |
54 | from ..dependencies import xattr | |
55 | ||
56 | __name__ = __name__.rsplit('.', 1)[0] # Pretend to be the parent module | |
57 | ||
58 | # This is not clearly defined otherwise | |
59 | compiled_regex_type = type(re.compile('')) | |
60 | ||
61 | ||
62 | class NO_DEFAULT: | |
63 | pass | |
64 | ||
65 | ||
66 | def IDENTITY(x): | |
67 | return x | |
68 | ||
69 | ||
70 | ENGLISH_MONTH_NAMES = [ | |
71 | 'January', 'February', 'March', 'April', 'May', 'June', | |
72 | 'July', 'August', 'September', 'October', 'November', 'December'] | |
73 | ||
74 | MONTH_NAMES = { | |
75 | 'en': ENGLISH_MONTH_NAMES, | |
76 | 'fr': [ | |
77 | 'janvier', 'février', 'mars', 'avril', 'mai', 'juin', | |
78 | 'juillet', 'août', 'septembre', 'octobre', 'novembre', 'décembre'], | |
79 | # these follow the genitive grammatical case (dopełniacz) | |
80 | # some websites might be using nominative, which will require another month list | |
81 | # https://en.wikibooks.org/wiki/Polish/Noun_cases | |
82 | 'pl': ['stycznia', 'lutego', 'marca', 'kwietnia', 'maja', 'czerwca', | |
83 | 'lipca', 'sierpnia', 'września', 'października', 'listopada', 'grudnia'], | |
84 | } | |
85 | ||
86 | # From https://github.com/python/cpython/blob/3.11/Lib/email/_parseaddr.py#L36-L42 | |
87 | TIMEZONE_NAMES = { | |
88 | 'UT': 0, 'UTC': 0, 'GMT': 0, 'Z': 0, | |
89 | 'AST': -4, 'ADT': -3, # Atlantic (used in Canada) | |
90 | 'EST': -5, 'EDT': -4, # Eastern | |
91 | 'CST': -6, 'CDT': -5, # Central | |
92 | 'MST': -7, 'MDT': -6, # Mountain | |
93 | 'PST': -8, 'PDT': -7 # Pacific | |
94 | } | |
95 | ||
96 | # needed for sanitizing filenames in restricted mode | |
97 | ACCENT_CHARS = dict(zip('ÂÃÄÀÁÅÆÇÈÉÊËÌÍÎÏÐÑÒÓÔÕÖŐØŒÙÚÛÜŰÝÞßàáâãäåæçèéêëìíîïðñòóôõöőøœùúûüűýþÿ', | |
98 | itertools.chain('AAAAAA', ['AE'], 'CEEEEIIIIDNOOOOOOO', ['OE'], 'UUUUUY', ['TH', 'ss'], | |
99 | 'aaaaaa', ['ae'], 'ceeeeiiiionooooooo', ['oe'], 'uuuuuy', ['th'], 'y'))) | |
100 | ||
101 | DATE_FORMATS = ( | |
102 | '%d %B %Y', | |
103 | '%d %b %Y', | |
104 | '%B %d %Y', | |
105 | '%B %dst %Y', | |
106 | '%B %dnd %Y', | |
107 | '%B %drd %Y', | |
108 | '%B %dth %Y', | |
109 | '%b %d %Y', | |
110 | '%b %dst %Y', | |
111 | '%b %dnd %Y', | |
112 | '%b %drd %Y', | |
113 | '%b %dth %Y', | |
114 | '%b %dst %Y %I:%M', | |
115 | '%b %dnd %Y %I:%M', | |
116 | '%b %drd %Y %I:%M', | |
117 | '%b %dth %Y %I:%M', | |
118 | '%Y %m %d', | |
119 | '%Y-%m-%d', | |
120 | '%Y.%m.%d.', | |
121 | '%Y/%m/%d', | |
122 | '%Y/%m/%d %H:%M', | |
123 | '%Y/%m/%d %H:%M:%S', | |
124 | '%Y%m%d%H%M', | |
125 | '%Y%m%d%H%M%S', | |
126 | '%Y%m%d', | |
127 | '%Y-%m-%d %H:%M', | |
128 | '%Y-%m-%d %H:%M:%S', | |
129 | '%Y-%m-%d %H:%M:%S.%f', | |
130 | '%Y-%m-%d %H:%M:%S:%f', | |
131 | '%d.%m.%Y %H:%M', | |
132 | '%d.%m.%Y %H.%M', | |
133 | '%Y-%m-%dT%H:%M:%SZ', | |
134 | '%Y-%m-%dT%H:%M:%S.%fZ', | |
135 | '%Y-%m-%dT%H:%M:%S.%f0Z', | |
136 | '%Y-%m-%dT%H:%M:%S', | |
137 | '%Y-%m-%dT%H:%M:%S.%f', | |
138 | '%Y-%m-%dT%H:%M', | |
139 | '%b %d %Y at %H:%M', | |
140 | '%b %d %Y at %H:%M:%S', | |
141 | '%B %d %Y at %H:%M', | |
142 | '%B %d %Y at %H:%M:%S', | |
143 | '%H:%M %d-%b-%Y', | |
144 | ) | |
145 | ||
146 | DATE_FORMATS_DAY_FIRST = list(DATE_FORMATS) | |
147 | DATE_FORMATS_DAY_FIRST.extend([ | |
148 | '%d-%m-%Y', | |
149 | '%d.%m.%Y', | |
150 | '%d.%m.%y', | |
151 | '%d/%m/%Y', | |
152 | '%d/%m/%y', | |
153 | '%d/%m/%Y %H:%M:%S', | |
154 | '%d-%m-%Y %H:%M', | |
155 | '%H:%M %d/%m/%Y', | |
156 | ]) | |
157 | ||
158 | DATE_FORMATS_MONTH_FIRST = list(DATE_FORMATS) | |
159 | DATE_FORMATS_MONTH_FIRST.extend([ | |
160 | '%m-%d-%Y', | |
161 | '%m.%d.%Y', | |
162 | '%m/%d/%Y', | |
163 | '%m/%d/%y', | |
164 | '%m/%d/%Y %H:%M:%S', | |
165 | ]) | |
166 | ||
167 | PACKED_CODES_RE = r"}\('(.+)',(\d+),(\d+),'([^']+)'\.split\('\|'\)" | |
168 | JSON_LD_RE = r'(?is)<script[^>]+type=(["\']?)application/ld\+json\1[^>]*>\s*(?P<json_ld>{.+?}|\[.+?\])\s*</script>' | |
169 | ||
170 | NUMBER_RE = r'\d+(?:\.\d+)?' | |
171 | ||
172 | ||
173 | @functools.cache | |
174 | def preferredencoding(): | |
175 | """Get preferred encoding. | |
176 | ||
177 | Returns the best encoding scheme for the system, based on | |
178 | locale.getpreferredencoding() and some further tweaks. | |
179 | """ | |
180 | try: | |
181 | pref = locale.getpreferredencoding() | |
182 | 'TEST'.encode(pref) | |
183 | except Exception: | |
184 | pref = 'UTF-8' | |
185 | ||
186 | return pref | |
187 | ||
188 | ||
189 | def write_json_file(obj, fn): | |
190 | """ Encode obj as JSON and write it to fn, atomically if possible """ | |
191 | ||
192 | tf = tempfile.NamedTemporaryFile( | |
193 | prefix=f'{os.path.basename(fn)}.', dir=os.path.dirname(fn), | |
194 | suffix='.tmp', delete=False, mode='w', encoding='utf-8') | |
195 | ||
196 | try: | |
197 | with tf: | |
198 | json.dump(obj, tf, ensure_ascii=False) | |
199 | if sys.platform == 'win32': | |
200 | # Need to remove existing file on Windows, else os.rename raises | |
201 | # WindowsError or FileExistsError. | |
202 | with contextlib.suppress(OSError): | |
203 | os.unlink(fn) | |
204 | with contextlib.suppress(OSError): | |
205 | mask = os.umask(0) | |
206 | os.umask(mask) | |
207 | os.chmod(tf.name, 0o666 & ~mask) | |
208 | os.rename(tf.name, fn) | |
209 | except Exception: | |
210 | with contextlib.suppress(OSError): | |
211 | os.remove(tf.name) | |
212 | raise | |
213 | ||
214 | ||
215 | def find_xpath_attr(node, xpath, key, val=None): | |
216 | """ Find the xpath xpath[@key=val] """ | |
217 | assert re.match(r'^[a-zA-Z_-]+$', key) | |
218 | expr = xpath + ('[@%s]' % key if val is None else f"[@{key}='{val}']") | |
219 | return node.find(expr) | |
220 | ||
221 | # On python2.6 the xml.etree.ElementTree.Element methods don't support | |
222 | # the namespace parameter | |
223 | ||
224 | ||
225 | def xpath_with_ns(path, ns_map): | |
226 | components = [c.split(':') for c in path.split('/')] | |
227 | replaced = [] | |
228 | for c in components: | |
229 | if len(c) == 1: | |
230 | replaced.append(c[0]) | |
231 | else: | |
232 | ns, tag = c | |
233 | replaced.append('{%s}%s' % (ns_map[ns], tag)) | |
234 | return '/'.join(replaced) | |
235 | ||
236 | ||
237 | def xpath_element(node, xpath, name=None, fatal=False, default=NO_DEFAULT): | |
238 | def _find_xpath(xpath): | |
239 | return node.find(xpath) | |
240 | ||
241 | if isinstance(xpath, str): | |
242 | n = _find_xpath(xpath) | |
243 | else: | |
244 | for xp in xpath: | |
245 | n = _find_xpath(xp) | |
246 | if n is not None: | |
247 | break | |
248 | ||
249 | if n is None: | |
250 | if default is not NO_DEFAULT: | |
251 | return default | |
252 | elif fatal: | |
253 | name = xpath if name is None else name | |
254 | raise ExtractorError('Could not find XML element %s' % name) | |
255 | else: | |
256 | return None | |
257 | return n | |
258 | ||
259 | ||
260 | def xpath_text(node, xpath, name=None, fatal=False, default=NO_DEFAULT): | |
261 | n = xpath_element(node, xpath, name, fatal=fatal, default=default) | |
262 | if n is None or n == default: | |
263 | return n | |
264 | if n.text is None: | |
265 | if default is not NO_DEFAULT: | |
266 | return default | |
267 | elif fatal: | |
268 | name = xpath if name is None else name | |
269 | raise ExtractorError('Could not find XML element\'s text %s' % name) | |
270 | else: | |
271 | return None | |
272 | return n.text | |
273 | ||
274 | ||
275 | def xpath_attr(node, xpath, key, name=None, fatal=False, default=NO_DEFAULT): | |
276 | n = find_xpath_attr(node, xpath, key) | |
277 | if n is None: | |
278 | if default is not NO_DEFAULT: | |
279 | return default | |
280 | elif fatal: | |
281 | name = f'{xpath}[@{key}]' if name is None else name | |
282 | raise ExtractorError('Could not find XML attribute %s' % name) | |
283 | else: | |
284 | return None | |
285 | return n.attrib[key] | |
286 | ||
287 | ||
288 | def get_element_by_id(id, html, **kwargs): | |
289 | """Return the content of the tag with the specified ID in the passed HTML document""" | |
290 | return get_element_by_attribute('id', id, html, **kwargs) | |
291 | ||
292 | ||
293 | def get_element_html_by_id(id, html, **kwargs): | |
294 | """Return the html of the tag with the specified ID in the passed HTML document""" | |
295 | return get_element_html_by_attribute('id', id, html, **kwargs) | |
296 | ||
297 | ||
298 | def get_element_by_class(class_name, html): | |
299 | """Return the content of the first tag with the specified class in the passed HTML document""" | |
300 | retval = get_elements_by_class(class_name, html) | |
301 | return retval[0] if retval else None | |
302 | ||
303 | ||
304 | def get_element_html_by_class(class_name, html): | |
305 | """Return the html of the first tag with the specified class in the passed HTML document""" | |
306 | retval = get_elements_html_by_class(class_name, html) | |
307 | return retval[0] if retval else None | |
308 | ||
309 | ||
310 | def get_element_by_attribute(attribute, value, html, **kwargs): | |
311 | retval = get_elements_by_attribute(attribute, value, html, **kwargs) | |
312 | return retval[0] if retval else None | |
313 | ||
314 | ||
315 | def get_element_html_by_attribute(attribute, value, html, **kargs): | |
316 | retval = get_elements_html_by_attribute(attribute, value, html, **kargs) | |
317 | return retval[0] if retval else None | |
318 | ||
319 | ||
320 | def get_elements_by_class(class_name, html, **kargs): | |
321 | """Return the content of all tags with the specified class in the passed HTML document as a list""" | |
322 | return get_elements_by_attribute( | |
323 | 'class', r'[^\'"]*(?<=[\'"\s])%s(?=[\'"\s])[^\'"]*' % re.escape(class_name), | |
324 | html, escape_value=False) | |
325 | ||
326 | ||
327 | def get_elements_html_by_class(class_name, html): | |
328 | """Return the html of all tags with the specified class in the passed HTML document as a list""" | |
329 | return get_elements_html_by_attribute( | |
330 | 'class', r'[^\'"]*(?<=[\'"\s])%s(?=[\'"\s])[^\'"]*' % re.escape(class_name), | |
331 | html, escape_value=False) | |
332 | ||
333 | ||
334 | def get_elements_by_attribute(*args, **kwargs): | |
335 | """Return the content of the tag with the specified attribute in the passed HTML document""" | |
336 | return [content for content, _ in get_elements_text_and_html_by_attribute(*args, **kwargs)] | |
337 | ||
338 | ||
339 | def get_elements_html_by_attribute(*args, **kwargs): | |
340 | """Return the html of the tag with the specified attribute in the passed HTML document""" | |
341 | return [whole for _, whole in get_elements_text_and_html_by_attribute(*args, **kwargs)] | |
342 | ||
343 | ||
344 | def get_elements_text_and_html_by_attribute(attribute, value, html, *, tag=r'[\w:.-]+', escape_value=True): | |
345 | """ | |
346 | Return the text (content) and the html (whole) of the tag with the specified | |
347 | attribute in the passed HTML document | |
348 | """ | |
349 | if not value: | |
350 | return | |
351 | ||
352 | quote = '' if re.match(r'''[\s"'`=<>]''', value) else '?' | |
353 | ||
354 | value = re.escape(value) if escape_value else value | |
355 | ||
356 | partial_element_re = rf'''(?x) | |
357 | <(?P<tag>{tag}) | |
358 | (?:\s(?:[^>"']|"[^"]*"|'[^']*')*)? | |
359 | \s{re.escape(attribute)}\s*=\s*(?P<_q>['"]{quote})(?-x:{value})(?P=_q) | |
360 | ''' | |
361 | ||
362 | for m in re.finditer(partial_element_re, html): | |
363 | content, whole = get_element_text_and_html_by_tag(m.group('tag'), html[m.start():]) | |
364 | ||
365 | yield ( | |
366 | unescapeHTML(re.sub(r'^(?P<q>["\'])(?P<content>.*)(?P=q)$', r'\g<content>', content, flags=re.DOTALL)), | |
367 | whole | |
368 | ) | |
369 | ||
370 | ||
371 | class HTMLBreakOnClosingTagParser(html.parser.HTMLParser): | |
372 | """ | |
373 | HTML parser which raises HTMLBreakOnClosingTagException upon reaching the | |
374 | closing tag for the first opening tag it has encountered, and can be used | |
375 | as a context manager | |
376 | """ | |
377 | ||
378 | class HTMLBreakOnClosingTagException(Exception): | |
379 | pass | |
380 | ||
381 | def __init__(self): | |
382 | self.tagstack = collections.deque() | |
383 | html.parser.HTMLParser.__init__(self) | |
384 | ||
385 | def __enter__(self): | |
386 | return self | |
387 | ||
388 | def __exit__(self, *_): | |
389 | self.close() | |
390 | ||
391 | def close(self): | |
392 | # handle_endtag does not return upon raising HTMLBreakOnClosingTagException, | |
393 | # so data remains buffered; we no longer have any interest in it, thus | |
394 | # override this method to discard it | |
395 | pass | |
396 | ||
397 | def handle_starttag(self, tag, _): | |
398 | self.tagstack.append(tag) | |
399 | ||
400 | def handle_endtag(self, tag): | |
401 | if not self.tagstack: | |
402 | raise compat_HTMLParseError('no tags in the stack') | |
403 | while self.tagstack: | |
404 | inner_tag = self.tagstack.pop() | |
405 | if inner_tag == tag: | |
406 | break | |
407 | else: | |
408 | raise compat_HTMLParseError(f'matching opening tag for closing {tag} tag not found') | |
409 | if not self.tagstack: | |
410 | raise self.HTMLBreakOnClosingTagException() | |
411 | ||
412 | ||
413 | # XXX: This should be far less strict | |
414 | def get_element_text_and_html_by_tag(tag, html): | |
415 | """ | |
416 | For the first element with the specified tag in the passed HTML document | |
417 | return its' content (text) and the whole element (html) | |
418 | """ | |
419 | def find_or_raise(haystack, needle, exc): | |
420 | try: | |
421 | return haystack.index(needle) | |
422 | except ValueError: | |
423 | raise exc | |
424 | closing_tag = f'</{tag}>' | |
425 | whole_start = find_or_raise( | |
426 | html, f'<{tag}', compat_HTMLParseError(f'opening {tag} tag not found')) | |
427 | content_start = find_or_raise( | |
428 | html[whole_start:], '>', compat_HTMLParseError(f'malformed opening {tag} tag')) | |
429 | content_start += whole_start + 1 | |
430 | with HTMLBreakOnClosingTagParser() as parser: | |
431 | parser.feed(html[whole_start:content_start]) | |
432 | if not parser.tagstack or parser.tagstack[0] != tag: | |
433 | raise compat_HTMLParseError(f'parser did not match opening {tag} tag') | |
434 | offset = content_start | |
435 | while offset < len(html): | |
436 | next_closing_tag_start = find_or_raise( | |
437 | html[offset:], closing_tag, | |
438 | compat_HTMLParseError(f'closing {tag} tag not found')) | |
439 | next_closing_tag_end = next_closing_tag_start + len(closing_tag) | |
440 | try: | |
441 | parser.feed(html[offset:offset + next_closing_tag_end]) | |
442 | offset += next_closing_tag_end | |
443 | except HTMLBreakOnClosingTagParser.HTMLBreakOnClosingTagException: | |
444 | return html[content_start:offset + next_closing_tag_start], \ | |
445 | html[whole_start:offset + next_closing_tag_end] | |
446 | raise compat_HTMLParseError('unexpected end of html') | |
447 | ||
448 | ||
449 | class HTMLAttributeParser(html.parser.HTMLParser): | |
450 | """Trivial HTML parser to gather the attributes for a single element""" | |
451 | ||
452 | def __init__(self): | |
453 | self.attrs = {} | |
454 | html.parser.HTMLParser.__init__(self) | |
455 | ||
456 | def handle_starttag(self, tag, attrs): | |
457 | self.attrs = dict(attrs) | |
458 | raise compat_HTMLParseError('done') | |
459 | ||
460 | ||
461 | class HTMLListAttrsParser(html.parser.HTMLParser): | |
462 | """HTML parser to gather the attributes for the elements of a list""" | |
463 | ||
464 | def __init__(self): | |
465 | html.parser.HTMLParser.__init__(self) | |
466 | self.items = [] | |
467 | self._level = 0 | |
468 | ||
469 | def handle_starttag(self, tag, attrs): | |
470 | if tag == 'li' and self._level == 0: | |
471 | self.items.append(dict(attrs)) | |
472 | self._level += 1 | |
473 | ||
474 | def handle_endtag(self, tag): | |
475 | self._level -= 1 | |
476 | ||
477 | ||
478 | def extract_attributes(html_element): | |
479 | """Given a string for an HTML element such as | |
480 | <el | |
481 | a="foo" B="bar" c="&98;az" d=boz | |
482 | empty= noval entity="&" | |
483 | sq='"' dq="'" | |
484 | > | |
485 | Decode and return a dictionary of attributes. | |
486 | { | |
487 | 'a': 'foo', 'b': 'bar', c: 'baz', d: 'boz', | |
488 | 'empty': '', 'noval': None, 'entity': '&', | |
489 | 'sq': '"', 'dq': '\'' | |
490 | }. | |
491 | """ | |
492 | parser = HTMLAttributeParser() | |
493 | with contextlib.suppress(compat_HTMLParseError): | |
494 | parser.feed(html_element) | |
495 | parser.close() | |
496 | return parser.attrs | |
497 | ||
498 | ||
499 | def parse_list(webpage): | |
500 | """Given a string for an series of HTML <li> elements, | |
501 | return a dictionary of their attributes""" | |
502 | parser = HTMLListAttrsParser() | |
503 | parser.feed(webpage) | |
504 | parser.close() | |
505 | return parser.items | |
506 | ||
507 | ||
508 | def clean_html(html): | |
509 | """Clean an HTML snippet into a readable string""" | |
510 | ||
511 | if html is None: # Convenience for sanitizing descriptions etc. | |
512 | return html | |
513 | ||
514 | html = re.sub(r'\s+', ' ', html) | |
515 | html = re.sub(r'(?u)\s?<\s?br\s?/?\s?>\s?', '\n', html) | |
516 | html = re.sub(r'(?u)<\s?/\s?p\s?>\s?<\s?p[^>]*>', '\n', html) | |
517 | # Strip html tags | |
518 | html = re.sub('<.*?>', '', html) | |
519 | # Replace html entities | |
520 | html = unescapeHTML(html) | |
521 | return html.strip() | |
522 | ||
523 | ||
524 | class LenientJSONDecoder(json.JSONDecoder): | |
525 | # TODO: Write tests | |
526 | def __init__(self, *args, transform_source=None, ignore_extra=False, close_objects=0, **kwargs): | |
527 | self.transform_source, self.ignore_extra = transform_source, ignore_extra | |
528 | self._close_attempts = 2 * close_objects | |
529 | super().__init__(*args, **kwargs) | |
530 | ||
531 | @staticmethod | |
532 | def _close_object(err): | |
533 | doc = err.doc[:err.pos] | |
534 | # We need to add comma first to get the correct error message | |
535 | if err.msg.startswith('Expecting \',\''): | |
536 | return doc + ',' | |
537 | elif not doc.endswith(','): | |
538 | return | |
539 | ||
540 | if err.msg.startswith('Expecting property name'): | |
541 | return doc[:-1] + '}' | |
542 | elif err.msg.startswith('Expecting value'): | |
543 | return doc[:-1] + ']' | |
544 | ||
545 | def decode(self, s): | |
546 | if self.transform_source: | |
547 | s = self.transform_source(s) | |
548 | for attempt in range(self._close_attempts + 1): | |
549 | try: | |
550 | if self.ignore_extra: | |
551 | return self.raw_decode(s.lstrip())[0] | |
552 | return super().decode(s) | |
553 | except json.JSONDecodeError as e: | |
554 | if e.pos is None: | |
555 | raise | |
556 | elif attempt < self._close_attempts: | |
557 | s = self._close_object(e) | |
558 | if s is not None: | |
559 | continue | |
560 | raise type(e)(f'{e.msg} in {s[e.pos - 10:e.pos + 10]!r}', s, e.pos) | |
561 | assert False, 'Too many attempts to decode JSON' | |
562 | ||
563 | ||
564 | def sanitize_open(filename, open_mode): | |
565 | """Try to open the given filename, and slightly tweak it if this fails. | |
566 | ||
567 | Attempts to open the given filename. If this fails, it tries to change | |
568 | the filename slightly, step by step, until it's either able to open it | |
569 | or it fails and raises a final exception, like the standard open() | |
570 | function. | |
571 | ||
572 | It returns the tuple (stream, definitive_file_name). | |
573 | """ | |
574 | if filename == '-': | |
575 | if sys.platform == 'win32': | |
576 | import msvcrt | |
577 | ||
578 | # stdout may be any IO stream, e.g. when using contextlib.redirect_stdout | |
579 | with contextlib.suppress(io.UnsupportedOperation): | |
580 | msvcrt.setmode(sys.stdout.fileno(), os.O_BINARY) | |
581 | return (sys.stdout.buffer if hasattr(sys.stdout, 'buffer') else sys.stdout, filename) | |
582 | ||
583 | for attempt in range(2): | |
584 | try: | |
585 | try: | |
586 | if sys.platform == 'win32': | |
587 | # FIXME: An exclusive lock also locks the file from being read. | |
588 | # Since windows locks are mandatory, don't lock the file on windows (for now). | |
589 | # Ref: https://github.com/yt-dlp/yt-dlp/issues/3124 | |
590 | raise LockingUnsupportedError() | |
591 | stream = locked_file(filename, open_mode, block=False).__enter__() | |
592 | except OSError: | |
593 | stream = open(filename, open_mode) | |
594 | return stream, filename | |
595 | except OSError as err: | |
596 | if attempt or err.errno in (errno.EACCES,): | |
597 | raise | |
598 | old_filename, filename = filename, sanitize_path(filename) | |
599 | if old_filename == filename: | |
600 | raise | |
601 | ||
602 | ||
603 | def timeconvert(timestr): | |
604 | """Convert RFC 2822 defined time string into system timestamp""" | |
605 | timestamp = None | |
606 | timetuple = email.utils.parsedate_tz(timestr) | |
607 | if timetuple is not None: | |
608 | timestamp = email.utils.mktime_tz(timetuple) | |
609 | return timestamp | |
610 | ||
611 | ||
612 | def sanitize_filename(s, restricted=False, is_id=NO_DEFAULT): | |
613 | """Sanitizes a string so it could be used as part of a filename. | |
614 | @param restricted Use a stricter subset of allowed characters | |
615 | @param is_id Whether this is an ID that should be kept unchanged if possible. | |
616 | If unset, yt-dlp's new sanitization rules are in effect | |
617 | """ | |
618 | if s == '': | |
619 | return '' | |
620 | ||
621 | def replace_insane(char): | |
622 | if restricted and char in ACCENT_CHARS: | |
623 | return ACCENT_CHARS[char] | |
624 | elif not restricted and char == '\n': | |
625 | return '\0 ' | |
626 | elif is_id is NO_DEFAULT and not restricted and char in '"*:<>?|/\\': | |
627 | # Replace with their full-width unicode counterparts | |
628 | return {'/': '\u29F8', '\\': '\u29f9'}.get(char, chr(ord(char) + 0xfee0)) | |
629 | elif char == '?' or ord(char) < 32 or ord(char) == 127: | |
630 | return '' | |
631 | elif char == '"': | |
632 | return '' if restricted else '\'' | |
633 | elif char == ':': | |
634 | return '\0_\0-' if restricted else '\0 \0-' | |
635 | elif char in '\\/|*<>': | |
636 | return '\0_' | |
637 | if restricted and (char in '!&\'()[]{}$;`^,#' or char.isspace() or ord(char) > 127): | |
638 | return '' if unicodedata.category(char)[0] in 'CM' else '\0_' | |
639 | return char | |
640 | ||
641 | # Replace look-alike Unicode glyphs | |
642 | if restricted and (is_id is NO_DEFAULT or not is_id): | |
643 | s = unicodedata.normalize('NFKC', s) | |
644 | s = re.sub(r'[0-9]+(?::[0-9]+)+', lambda m: m.group(0).replace(':', '_'), s) # Handle timestamps | |
645 | result = ''.join(map(replace_insane, s)) | |
646 | if is_id is NO_DEFAULT: | |
647 | result = re.sub(r'(\0.)(?:(?=\1)..)+', r'\1', result) # Remove repeated substitute chars | |
648 | STRIP_RE = r'(?:\0.|[ _-])*' | |
649 | result = re.sub(f'^\0.{STRIP_RE}|{STRIP_RE}\0.$', '', result) # Remove substitute chars from start/end | |
650 | result = result.replace('\0', '') or '_' | |
651 | ||
652 | if not is_id: | |
653 | while '__' in result: | |
654 | result = result.replace('__', '_') | |
655 | result = result.strip('_') | |
656 | # Common case of "Foreign band name - English song title" | |
657 | if restricted and result.startswith('-_'): | |
658 | result = result[2:] | |
659 | if result.startswith('-'): | |
660 | result = '_' + result[len('-'):] | |
661 | result = result.lstrip('.') | |
662 | if not result: | |
663 | result = '_' | |
664 | return result | |
665 | ||
666 | ||
667 | def sanitize_path(s, force=False): | |
668 | """Sanitizes and normalizes path on Windows""" | |
669 | # XXX: this handles drive relative paths (c:sth) incorrectly | |
670 | if sys.platform == 'win32': | |
671 | force = False | |
672 | drive_or_unc, _ = os.path.splitdrive(s) | |
673 | elif force: | |
674 | drive_or_unc = '' | |
675 | else: | |
676 | return s | |
677 | ||
678 | norm_path = os.path.normpath(remove_start(s, drive_or_unc)).split(os.path.sep) | |
679 | if drive_or_unc: | |
680 | norm_path.pop(0) | |
681 | sanitized_path = [ | |
682 | path_part if path_part in ['.', '..'] else re.sub(r'(?:[/<>:"\|\\?\*]|[\s.]$)', '#', path_part) | |
683 | for path_part in norm_path] | |
684 | if drive_or_unc: | |
685 | sanitized_path.insert(0, drive_or_unc + os.path.sep) | |
686 | elif force and s and s[0] == os.path.sep: | |
687 | sanitized_path.insert(0, os.path.sep) | |
688 | # TODO: Fix behavioral differences <3.12 | |
689 | # The workaround using `normpath` only superficially passes tests | |
690 | # Ref: https://github.com/python/cpython/pull/100351 | |
691 | return os.path.normpath(os.path.join(*sanitized_path)) | |
692 | ||
693 | ||
694 | def sanitize_url(url, *, scheme='http'): | |
695 | # Prepend protocol-less URLs with `http:` scheme in order to mitigate | |
696 | # the number of unwanted failures due to missing protocol | |
697 | if url is None: | |
698 | return | |
699 | elif url.startswith('//'): | |
700 | return f'{scheme}:{url}' | |
701 | # Fix some common typos seen so far | |
702 | COMMON_TYPOS = ( | |
703 | # https://github.com/ytdl-org/youtube-dl/issues/15649 | |
704 | (r'^httpss://', r'https://'), | |
705 | # https://bx1.be/lives/direct-tv/ | |
706 | (r'^rmtp([es]?)://', r'rtmp\1://'), | |
707 | ) | |
708 | for mistake, fixup in COMMON_TYPOS: | |
709 | if re.match(mistake, url): | |
710 | return re.sub(mistake, fixup, url) | |
711 | return url | |
712 | ||
713 | ||
714 | def extract_basic_auth(url): | |
715 | parts = urllib.parse.urlsplit(url) | |
716 | if parts.username is None: | |
717 | return url, None | |
718 | url = urllib.parse.urlunsplit(parts._replace(netloc=( | |
719 | parts.hostname if parts.port is None | |
720 | else '%s:%d' % (parts.hostname, parts.port)))) | |
721 | auth_payload = base64.b64encode( | |
722 | ('%s:%s' % (parts.username, parts.password or '')).encode()) | |
723 | return url, f'Basic {auth_payload.decode()}' | |
724 | ||
725 | ||
726 | def expand_path(s): | |
727 | """Expand shell variables and ~""" | |
728 | return os.path.expandvars(compat_expanduser(s)) | |
729 | ||
730 | ||
731 | def orderedSet(iterable, *, lazy=False): | |
732 | """Remove all duplicates from the input iterable""" | |
733 | def _iter(): | |
734 | seen = [] # Do not use set since the items can be unhashable | |
735 | for x in iterable: | |
736 | if x not in seen: | |
737 | seen.append(x) | |
738 | yield x | |
739 | ||
740 | return _iter() if lazy else list(_iter()) | |
741 | ||
742 | ||
743 | def _htmlentity_transform(entity_with_semicolon): | |
744 | """Transforms an HTML entity to a character.""" | |
745 | entity = entity_with_semicolon[:-1] | |
746 | ||
747 | # Known non-numeric HTML entity | |
748 | if entity in html.entities.name2codepoint: | |
749 | return chr(html.entities.name2codepoint[entity]) | |
750 | ||
751 | # TODO: HTML5 allows entities without a semicolon. | |
752 | # E.g. 'Éric' should be decoded as 'Éric'. | |
753 | if entity_with_semicolon in html.entities.html5: | |
754 | return html.entities.html5[entity_with_semicolon] | |
755 | ||
756 | mobj = re.match(r'#(x[0-9a-fA-F]+|[0-9]+)', entity) | |
757 | if mobj is not None: | |
758 | numstr = mobj.group(1) | |
759 | if numstr.startswith('x'): | |
760 | base = 16 | |
761 | numstr = '0%s' % numstr | |
762 | else: | |
763 | base = 10 | |
764 | # See https://github.com/ytdl-org/youtube-dl/issues/7518 | |
765 | with contextlib.suppress(ValueError): | |
766 | return chr(int(numstr, base)) | |
767 | ||
768 | # Unknown entity in name, return its literal representation | |
769 | return '&%s;' % entity | |
770 | ||
771 | ||
772 | def unescapeHTML(s): | |
773 | if s is None: | |
774 | return None | |
775 | assert isinstance(s, str) | |
776 | ||
777 | return re.sub( | |
778 | r'&([^&;]+;)', lambda m: _htmlentity_transform(m.group(1)), s) | |
779 | ||
780 | ||
781 | def escapeHTML(text): | |
782 | return ( | |
783 | text | |
784 | .replace('&', '&') | |
785 | .replace('<', '<') | |
786 | .replace('>', '>') | |
787 | .replace('"', '"') | |
788 | .replace("'", ''') | |
789 | ) | |
790 | ||
791 | ||
792 | class netrc_from_content(netrc.netrc): | |
793 | def __init__(self, content): | |
794 | self.hosts, self.macros = {}, {} | |
795 | with io.StringIO(content) as stream: | |
796 | self._parse('-', stream, False) | |
797 | ||
798 | ||
799 | class Popen(subprocess.Popen): | |
800 | if sys.platform == 'win32': | |
801 | _startupinfo = subprocess.STARTUPINFO() | |
802 | _startupinfo.dwFlags |= subprocess.STARTF_USESHOWWINDOW | |
803 | else: | |
804 | _startupinfo = None | |
805 | ||
806 | @staticmethod | |
807 | def _fix_pyinstaller_ld_path(env): | |
808 | """Restore LD_LIBRARY_PATH when using PyInstaller | |
809 | Ref: https://github.com/pyinstaller/pyinstaller/blob/develop/doc/runtime-information.rst#ld_library_path--libpath-considerations | |
810 | https://github.com/yt-dlp/yt-dlp/issues/4573 | |
811 | """ | |
812 | if not hasattr(sys, '_MEIPASS'): | |
813 | return | |
814 | ||
815 | def _fix(key): | |
816 | orig = env.get(f'{key}_ORIG') | |
817 | if orig is None: | |
818 | env.pop(key, None) | |
819 | else: | |
820 | env[key] = orig | |
821 | ||
822 | _fix('LD_LIBRARY_PATH') # Linux | |
823 | _fix('DYLD_LIBRARY_PATH') # macOS | |
824 | ||
825 | def __init__(self, args, *remaining, env=None, text=False, shell=False, **kwargs): | |
826 | if env is None: | |
827 | env = os.environ.copy() | |
828 | self._fix_pyinstaller_ld_path(env) | |
829 | ||
830 | self.__text_mode = kwargs.get('encoding') or kwargs.get('errors') or text or kwargs.get('universal_newlines') | |
831 | if text is True: | |
832 | kwargs['universal_newlines'] = True # For 3.6 compatibility | |
833 | kwargs.setdefault('encoding', 'utf-8') | |
834 | kwargs.setdefault('errors', 'replace') | |
835 | ||
836 | if shell and compat_os_name == 'nt' and kwargs.get('executable') is None: | |
837 | if not isinstance(args, str): | |
838 | args = shell_quote(args, shell=True) | |
839 | shell = False | |
840 | # Set variable for `cmd.exe` newline escaping (see `utils.shell_quote`) | |
841 | env['='] = '"^\n\n"' | |
842 | args = f'{self.__comspec()} /Q /S /D /V:OFF /E:ON /C "{args}"' | |
843 | ||
844 | super().__init__(args, *remaining, env=env, shell=shell, **kwargs, startupinfo=self._startupinfo) | |
845 | ||
846 | def __comspec(self): | |
847 | comspec = os.environ.get('ComSpec') or os.path.join( | |
848 | os.environ.get('SystemRoot', ''), 'System32', 'cmd.exe') | |
849 | if os.path.isabs(comspec): | |
850 | return comspec | |
851 | raise FileNotFoundError('shell not found: neither %ComSpec% nor %SystemRoot% is set') | |
852 | ||
853 | def communicate_or_kill(self, *args, **kwargs): | |
854 | try: | |
855 | return self.communicate(*args, **kwargs) | |
856 | except BaseException: # Including KeyboardInterrupt | |
857 | self.kill(timeout=None) | |
858 | raise | |
859 | ||
860 | def kill(self, *, timeout=0): | |
861 | super().kill() | |
862 | if timeout != 0: | |
863 | self.wait(timeout=timeout) | |
864 | ||
865 | @classmethod | |
866 | def run(cls, *args, timeout=None, **kwargs): | |
867 | with cls(*args, **kwargs) as proc: | |
868 | default = '' if proc.__text_mode else b'' | |
869 | stdout, stderr = proc.communicate_or_kill(timeout=timeout) | |
870 | return stdout or default, stderr or default, proc.returncode | |
871 | ||
872 | ||
873 | def encodeArgument(s): | |
874 | # Legacy code that uses byte strings | |
875 | # Uncomment the following line after fixing all post processors | |
876 | # assert isinstance(s, str), 'Internal error: %r should be of type %r, is %r' % (s, str, type(s)) | |
877 | return s if isinstance(s, str) else s.decode('ascii') | |
878 | ||
879 | ||
880 | _timetuple = collections.namedtuple('Time', ('hours', 'minutes', 'seconds', 'milliseconds')) | |
881 | ||
882 | ||
883 | def timetuple_from_msec(msec): | |
884 | secs, msec = divmod(msec, 1000) | |
885 | mins, secs = divmod(secs, 60) | |
886 | hrs, mins = divmod(mins, 60) | |
887 | return _timetuple(hrs, mins, secs, msec) | |
888 | ||
889 | ||
890 | def formatSeconds(secs, delim=':', msec=False): | |
891 | time = timetuple_from_msec(secs * 1000) | |
892 | if time.hours: | |
893 | ret = '%d%s%02d%s%02d' % (time.hours, delim, time.minutes, delim, time.seconds) | |
894 | elif time.minutes: | |
895 | ret = '%d%s%02d' % (time.minutes, delim, time.seconds) | |
896 | else: | |
897 | ret = '%d' % time.seconds | |
898 | return '%s.%03d' % (ret, time.milliseconds) if msec else ret | |
899 | ||
900 | ||
901 | def bug_reports_message(before=';'): | |
902 | from ..update import REPOSITORY | |
903 | ||
904 | msg = (f'please report this issue on https://github.com/{REPOSITORY}/issues?q= , ' | |
905 | 'filling out the appropriate issue template. Confirm you are on the latest version using yt-dlp -U') | |
906 | ||
907 | before = before.rstrip() | |
908 | if not before or before.endswith(('.', '!', '?')): | |
909 | msg = msg[0].title() + msg[1:] | |
910 | ||
911 | return (before + ' ' if before else '') + msg | |
912 | ||
913 | ||
914 | class YoutubeDLError(Exception): | |
915 | """Base exception for YoutubeDL errors.""" | |
916 | msg = None | |
917 | ||
918 | def __init__(self, msg=None): | |
919 | if msg is not None: | |
920 | self.msg = msg | |
921 | elif self.msg is None: | |
922 | self.msg = type(self).__name__ | |
923 | super().__init__(self.msg) | |
924 | ||
925 | ||
926 | class ExtractorError(YoutubeDLError): | |
927 | """Error during info extraction.""" | |
928 | ||
929 | def __init__(self, msg, tb=None, expected=False, cause=None, video_id=None, ie=None): | |
930 | """ tb, if given, is the original traceback (so that it can be printed out). | |
931 | If expected is set, this is a normal error message and most likely not a bug in yt-dlp. | |
932 | """ | |
933 | from ..networking.exceptions import network_exceptions | |
934 | if sys.exc_info()[0] in network_exceptions: | |
935 | expected = True | |
936 | ||
937 | self.orig_msg = str(msg) | |
938 | self.traceback = tb | |
939 | self.expected = expected | |
940 | self.cause = cause | |
941 | self.video_id = video_id | |
942 | self.ie = ie | |
943 | self.exc_info = sys.exc_info() # preserve original exception | |
944 | if isinstance(self.exc_info[1], ExtractorError): | |
945 | self.exc_info = self.exc_info[1].exc_info | |
946 | super().__init__(self.__msg) | |
947 | ||
948 | @property | |
949 | def __msg(self): | |
950 | return ''.join(( | |
951 | format_field(self.ie, None, '[%s] '), | |
952 | format_field(self.video_id, None, '%s: '), | |
953 | self.orig_msg, | |
954 | format_field(self.cause, None, ' (caused by %r)'), | |
955 | '' if self.expected else bug_reports_message())) | |
956 | ||
957 | def format_traceback(self): | |
958 | return join_nonempty( | |
959 | self.traceback and ''.join(traceback.format_tb(self.traceback)), | |
960 | self.cause and ''.join(traceback.format_exception(None, self.cause, self.cause.__traceback__)[1:]), | |
961 | delim='\n') or None | |
962 | ||
963 | def __setattr__(self, name, value): | |
964 | super().__setattr__(name, value) | |
965 | if getattr(self, 'msg', None) and name not in ('msg', 'args'): | |
966 | self.msg = self.__msg or type(self).__name__ | |
967 | self.args = (self.msg, ) # Cannot be property | |
968 | ||
969 | ||
970 | class UnsupportedError(ExtractorError): | |
971 | def __init__(self, url): | |
972 | super().__init__( | |
973 | 'Unsupported URL: %s' % url, expected=True) | |
974 | self.url = url | |
975 | ||
976 | ||
977 | class RegexNotFoundError(ExtractorError): | |
978 | """Error when a regex didn't match""" | |
979 | pass | |
980 | ||
981 | ||
982 | class GeoRestrictedError(ExtractorError): | |
983 | """Geographic restriction Error exception. | |
984 | ||
985 | This exception may be thrown when a video is not available from your | |
986 | geographic location due to geographic restrictions imposed by a website. | |
987 | """ | |
988 | ||
989 | def __init__(self, msg, countries=None, **kwargs): | |
990 | kwargs['expected'] = True | |
991 | super().__init__(msg, **kwargs) | |
992 | self.countries = countries | |
993 | ||
994 | ||
995 | class UserNotLive(ExtractorError): | |
996 | """Error when a channel/user is not live""" | |
997 | ||
998 | def __init__(self, msg=None, **kwargs): | |
999 | kwargs['expected'] = True | |
1000 | super().__init__(msg or 'The channel is not currently live', **kwargs) | |
1001 | ||
1002 | ||
1003 | class DownloadError(YoutubeDLError): | |
1004 | """Download Error exception. | |
1005 | ||
1006 | This exception may be thrown by FileDownloader objects if they are not | |
1007 | configured to continue on errors. They will contain the appropriate | |
1008 | error message. | |
1009 | """ | |
1010 | ||
1011 | def __init__(self, msg, exc_info=None): | |
1012 | """ exc_info, if given, is the original exception that caused the trouble (as returned by sys.exc_info()). """ | |
1013 | super().__init__(msg) | |
1014 | self.exc_info = exc_info | |
1015 | ||
1016 | ||
1017 | class EntryNotInPlaylist(YoutubeDLError): | |
1018 | """Entry not in playlist exception. | |
1019 | ||
1020 | This exception will be thrown by YoutubeDL when a requested entry | |
1021 | is not found in the playlist info_dict | |
1022 | """ | |
1023 | msg = 'Entry not found in info' | |
1024 | ||
1025 | ||
1026 | class SameFileError(YoutubeDLError): | |
1027 | """Same File exception. | |
1028 | ||
1029 | This exception will be thrown by FileDownloader objects if they detect | |
1030 | multiple files would have to be downloaded to the same file on disk. | |
1031 | """ | |
1032 | msg = 'Fixed output name but more than one file to download' | |
1033 | ||
1034 | def __init__(self, filename=None): | |
1035 | if filename is not None: | |
1036 | self.msg += f': {filename}' | |
1037 | super().__init__(self.msg) | |
1038 | ||
1039 | ||
1040 | class PostProcessingError(YoutubeDLError): | |
1041 | """Post Processing exception. | |
1042 | ||
1043 | This exception may be raised by PostProcessor's .run() method to | |
1044 | indicate an error in the postprocessing task. | |
1045 | """ | |
1046 | ||
1047 | ||
1048 | class DownloadCancelled(YoutubeDLError): | |
1049 | """ Exception raised when the download queue should be interrupted """ | |
1050 | msg = 'The download was cancelled' | |
1051 | ||
1052 | ||
1053 | class ExistingVideoReached(DownloadCancelled): | |
1054 | """ --break-on-existing triggered """ | |
1055 | msg = 'Encountered a video that is already in the archive, stopping due to --break-on-existing' | |
1056 | ||
1057 | ||
1058 | class RejectedVideoReached(DownloadCancelled): | |
1059 | """ --break-match-filter triggered """ | |
1060 | msg = 'Encountered a video that did not match filter, stopping due to --break-match-filter' | |
1061 | ||
1062 | ||
1063 | class MaxDownloadsReached(DownloadCancelled): | |
1064 | """ --max-downloads limit has been reached. """ | |
1065 | msg = 'Maximum number of downloads reached, stopping due to --max-downloads' | |
1066 | ||
1067 | ||
1068 | class ReExtractInfo(YoutubeDLError): | |
1069 | """ Video info needs to be re-extracted. """ | |
1070 | ||
1071 | def __init__(self, msg, expected=False): | |
1072 | super().__init__(msg) | |
1073 | self.expected = expected | |
1074 | ||
1075 | ||
1076 | class ThrottledDownload(ReExtractInfo): | |
1077 | """ Download speed below --throttled-rate. """ | |
1078 | msg = 'The download speed is below throttle limit' | |
1079 | ||
1080 | def __init__(self): | |
1081 | super().__init__(self.msg, expected=False) | |
1082 | ||
1083 | ||
1084 | class UnavailableVideoError(YoutubeDLError): | |
1085 | """Unavailable Format exception. | |
1086 | ||
1087 | This exception will be thrown when a video is requested | |
1088 | in a format that is not available for that video. | |
1089 | """ | |
1090 | msg = 'Unable to download video' | |
1091 | ||
1092 | def __init__(self, err=None): | |
1093 | if err is not None: | |
1094 | self.msg += f': {err}' | |
1095 | super().__init__(self.msg) | |
1096 | ||
1097 | ||
1098 | class ContentTooShortError(YoutubeDLError): | |
1099 | """Content Too Short exception. | |
1100 | ||
1101 | This exception may be raised by FileDownloader objects when a file they | |
1102 | download is too small for what the server announced first, indicating | |
1103 | the connection was probably interrupted. | |
1104 | """ | |
1105 | ||
1106 | def __init__(self, downloaded, expected): | |
1107 | super().__init__(f'Downloaded {downloaded} bytes, expected {expected} bytes') | |
1108 | # Both in bytes | |
1109 | self.downloaded = downloaded | |
1110 | self.expected = expected | |
1111 | ||
1112 | ||
1113 | class XAttrMetadataError(YoutubeDLError): | |
1114 | def __init__(self, code=None, msg='Unknown error'): | |
1115 | super().__init__(msg) | |
1116 | self.code = code | |
1117 | self.msg = msg | |
1118 | ||
1119 | # Parsing code and msg | |
1120 | if (self.code in (errno.ENOSPC, errno.EDQUOT) | |
1121 | or 'No space left' in self.msg or 'Disk quota exceeded' in self.msg): | |
1122 | self.reason = 'NO_SPACE' | |
1123 | elif self.code == errno.E2BIG or 'Argument list too long' in self.msg: | |
1124 | self.reason = 'VALUE_TOO_LONG' | |
1125 | else: | |
1126 | self.reason = 'NOT_SUPPORTED' | |
1127 | ||
1128 | ||
1129 | class XAttrUnavailableError(YoutubeDLError): | |
1130 | pass | |
1131 | ||
1132 | ||
1133 | def is_path_like(f): | |
1134 | return isinstance(f, (str, bytes, os.PathLike)) | |
1135 | ||
1136 | ||
1137 | def extract_timezone(date_str): | |
1138 | m = re.search( | |
1139 | r'''(?x) | |
1140 | ^.{8,}? # >=8 char non-TZ prefix, if present | |
1141 | (?P<tz>Z| # just the UTC Z, or | |
1142 | (?:(?<=.\b\d{4}|\b\d{2}:\d\d)| # preceded by 4 digits or hh:mm or | |
1143 | (?<!.\b[a-zA-Z]{3}|[a-zA-Z]{4}|..\b\d\d)) # not preceded by 3 alpha word or >= 4 alpha or 2 digits | |
1144 | [ ]? # optional space | |
1145 | (?P<sign>\+|-) # +/- | |
1146 | (?P<hours>[0-9]{2}):?(?P<minutes>[0-9]{2}) # hh[:]mm | |
1147 | $) | |
1148 | ''', date_str) | |
1149 | if not m: | |
1150 | m = re.search(r'\d{1,2}:\d{1,2}(?:\.\d+)?(?P<tz>\s*[A-Z]+)$', date_str) | |
1151 | timezone = TIMEZONE_NAMES.get(m and m.group('tz').strip()) | |
1152 | if timezone is not None: | |
1153 | date_str = date_str[:-len(m.group('tz'))] | |
1154 | timezone = dt.timedelta(hours=timezone or 0) | |
1155 | else: | |
1156 | date_str = date_str[:-len(m.group('tz'))] | |
1157 | if not m.group('sign'): | |
1158 | timezone = dt.timedelta() | |
1159 | else: | |
1160 | sign = 1 if m.group('sign') == '+' else -1 | |
1161 | timezone = dt.timedelta( | |
1162 | hours=sign * int(m.group('hours')), | |
1163 | minutes=sign * int(m.group('minutes'))) | |
1164 | return timezone, date_str | |
1165 | ||
1166 | ||
1167 | def parse_iso8601(date_str, delimiter='T', timezone=None): | |
1168 | """ Return a UNIX timestamp from the given date """ | |
1169 | ||
1170 | if date_str is None: | |
1171 | return None | |
1172 | ||
1173 | date_str = re.sub(r'\.[0-9]+', '', date_str) | |
1174 | ||
1175 | if timezone is None: | |
1176 | timezone, date_str = extract_timezone(date_str) | |
1177 | ||
1178 | with contextlib.suppress(ValueError): | |
1179 | date_format = f'%Y-%m-%d{delimiter}%H:%M:%S' | |
1180 | dt_ = dt.datetime.strptime(date_str, date_format) - timezone | |
1181 | return calendar.timegm(dt_.timetuple()) | |
1182 | ||
1183 | ||
1184 | def date_formats(day_first=True): | |
1185 | return DATE_FORMATS_DAY_FIRST if day_first else DATE_FORMATS_MONTH_FIRST | |
1186 | ||
1187 | ||
1188 | def unified_strdate(date_str, day_first=True): | |
1189 | """Return a string with the date in the format YYYYMMDD""" | |
1190 | ||
1191 | if date_str is None: | |
1192 | return None | |
1193 | upload_date = None | |
1194 | # Replace commas | |
1195 | date_str = date_str.replace(',', ' ') | |
1196 | # Remove AM/PM + timezone | |
1197 | date_str = re.sub(r'(?i)\s*(?:AM|PM)(?:\s+[A-Z]+)?', '', date_str) | |
1198 | _, date_str = extract_timezone(date_str) | |
1199 | ||
1200 | for expression in date_formats(day_first): | |
1201 | with contextlib.suppress(ValueError): | |
1202 | upload_date = dt.datetime.strptime(date_str, expression).strftime('%Y%m%d') | |
1203 | if upload_date is None: | |
1204 | timetuple = email.utils.parsedate_tz(date_str) | |
1205 | if timetuple: | |
1206 | with contextlib.suppress(ValueError): | |
1207 | upload_date = dt.datetime(*timetuple[:6]).strftime('%Y%m%d') | |
1208 | if upload_date is not None: | |
1209 | return str(upload_date) | |
1210 | ||
1211 | ||
1212 | def unified_timestamp(date_str, day_first=True): | |
1213 | if not isinstance(date_str, str): | |
1214 | return None | |
1215 | ||
1216 | date_str = re.sub(r'\s+', ' ', re.sub( | |
1217 | r'(?i)[,|]|(mon|tues?|wed(nes)?|thu(rs)?|fri|sat(ur)?)(day)?', '', date_str)) | |
1218 | ||
1219 | pm_delta = 12 if re.search(r'(?i)PM', date_str) else 0 | |
1220 | timezone, date_str = extract_timezone(date_str) | |
1221 | ||
1222 | # Remove AM/PM + timezone | |
1223 | date_str = re.sub(r'(?i)\s*(?:AM|PM)(?:\s+[A-Z]+)?', '', date_str) | |
1224 | ||
1225 | # Remove unrecognized timezones from ISO 8601 alike timestamps | |
1226 | m = re.search(r'\d{1,2}:\d{1,2}(?:\.\d+)?(?P<tz>\s*[A-Z]+)$', date_str) | |
1227 | if m: | |
1228 | date_str = date_str[:-len(m.group('tz'))] | |
1229 | ||
1230 | # Python only supports microseconds, so remove nanoseconds | |
1231 | m = re.search(r'^([0-9]{4,}-[0-9]{1,2}-[0-9]{1,2}T[0-9]{1,2}:[0-9]{1,2}:[0-9]{1,2}\.[0-9]{6})[0-9]+$', date_str) | |
1232 | if m: | |
1233 | date_str = m.group(1) | |
1234 | ||
1235 | for expression in date_formats(day_first): | |
1236 | with contextlib.suppress(ValueError): | |
1237 | dt_ = dt.datetime.strptime(date_str, expression) - timezone + dt.timedelta(hours=pm_delta) | |
1238 | return calendar.timegm(dt_.timetuple()) | |
1239 | ||
1240 | timetuple = email.utils.parsedate_tz(date_str) | |
1241 | if timetuple: | |
1242 | return calendar.timegm(timetuple) + pm_delta * 3600 - timezone.total_seconds() | |
1243 | ||
1244 | ||
1245 | def determine_ext(url, default_ext='unknown_video'): | |
1246 | if url is None or '.' not in url: | |
1247 | return default_ext | |
1248 | guess = url.partition('?')[0].rpartition('.')[2] | |
1249 | if re.match(r'^[A-Za-z0-9]+$', guess): | |
1250 | return guess | |
1251 | # Try extract ext from URLs like http://example.com/foo/bar.mp4/?download | |
1252 | elif guess.rstrip('/') in KNOWN_EXTENSIONS: | |
1253 | return guess.rstrip('/') | |
1254 | else: | |
1255 | return default_ext | |
1256 | ||
1257 | ||
1258 | def subtitles_filename(filename, sub_lang, sub_format, expected_real_ext=None): | |
1259 | return replace_extension(filename, sub_lang + '.' + sub_format, expected_real_ext) | |
1260 | ||
1261 | ||
1262 | def datetime_from_str(date_str, precision='auto', format='%Y%m%d'): | |
1263 | R""" | |
1264 | Return a datetime object from a string. | |
1265 | Supported format: | |
1266 | (now|today|yesterday|DATE)([+-]\d+(microsecond|second|minute|hour|day|week|month|year)s?)? | |
1267 | ||
1268 | @param format strftime format of DATE | |
1269 | @param precision Round the datetime object: auto|microsecond|second|minute|hour|day | |
1270 | auto: round to the unit provided in date_str (if applicable). | |
1271 | """ | |
1272 | auto_precision = False | |
1273 | if precision == 'auto': | |
1274 | auto_precision = True | |
1275 | precision = 'microsecond' | |
1276 | today = datetime_round(dt.datetime.now(dt.timezone.utc), precision) | |
1277 | if date_str in ('now', 'today'): | |
1278 | return today | |
1279 | if date_str == 'yesterday': | |
1280 | return today - dt.timedelta(days=1) | |
1281 | match = re.match( | |
1282 | r'(?P<start>.+)(?P<sign>[+-])(?P<time>\d+)(?P<unit>microsecond|second|minute|hour|day|week|month|year)s?', | |
1283 | date_str) | |
1284 | if match is not None: | |
1285 | start_time = datetime_from_str(match.group('start'), precision, format) | |
1286 | time = int(match.group('time')) * (-1 if match.group('sign') == '-' else 1) | |
1287 | unit = match.group('unit') | |
1288 | if unit == 'month' or unit == 'year': | |
1289 | new_date = datetime_add_months(start_time, time * 12 if unit == 'year' else time) | |
1290 | unit = 'day' | |
1291 | else: | |
1292 | if unit == 'week': | |
1293 | unit = 'day' | |
1294 | time *= 7 | |
1295 | delta = dt.timedelta(**{unit + 's': time}) | |
1296 | new_date = start_time + delta | |
1297 | if auto_precision: | |
1298 | return datetime_round(new_date, unit) | |
1299 | return new_date | |
1300 | ||
1301 | return datetime_round(dt.datetime.strptime(date_str, format), precision) | |
1302 | ||
1303 | ||
1304 | def date_from_str(date_str, format='%Y%m%d', strict=False): | |
1305 | R""" | |
1306 | Return a date object from a string using datetime_from_str | |
1307 | ||
1308 | @param strict Restrict allowed patterns to "YYYYMMDD" and | |
1309 | (now|today|yesterday)(-\d+(day|week|month|year)s?)? | |
1310 | """ | |
1311 | if strict and not re.fullmatch(r'\d{8}|(now|today|yesterday)(-\d+(day|week|month|year)s?)?', date_str): | |
1312 | raise ValueError(f'Invalid date format "{date_str}"') | |
1313 | return datetime_from_str(date_str, precision='microsecond', format=format).date() | |
1314 | ||
1315 | ||
1316 | def datetime_add_months(dt_, months): | |
1317 | """Increment/Decrement a datetime object by months.""" | |
1318 | month = dt_.month + months - 1 | |
1319 | year = dt_.year + month // 12 | |
1320 | month = month % 12 + 1 | |
1321 | day = min(dt_.day, calendar.monthrange(year, month)[1]) | |
1322 | return dt_.replace(year, month, day) | |
1323 | ||
1324 | ||
1325 | def datetime_round(dt_, precision='day'): | |
1326 | """ | |
1327 | Round a datetime object's time to a specific precision | |
1328 | """ | |
1329 | if precision == 'microsecond': | |
1330 | return dt_ | |
1331 | ||
1332 | unit_seconds = { | |
1333 | 'day': 86400, | |
1334 | 'hour': 3600, | |
1335 | 'minute': 60, | |
1336 | 'second': 1, | |
1337 | } | |
1338 | roundto = lambda x, n: ((x + n / 2) // n) * n | |
1339 | timestamp = roundto(calendar.timegm(dt_.timetuple()), unit_seconds[precision]) | |
1340 | return dt.datetime.fromtimestamp(timestamp, dt.timezone.utc) | |
1341 | ||
1342 | ||
1343 | def hyphenate_date(date_str): | |
1344 | """ | |
1345 | Convert a date in 'YYYYMMDD' format to 'YYYY-MM-DD' format""" | |
1346 | match = re.match(r'^(\d\d\d\d)(\d\d)(\d\d)$', date_str) | |
1347 | if match is not None: | |
1348 | return '-'.join(match.groups()) | |
1349 | else: | |
1350 | return date_str | |
1351 | ||
1352 | ||
1353 | class DateRange: | |
1354 | """Represents a time interval between two dates""" | |
1355 | ||
1356 | def __init__(self, start=None, end=None): | |
1357 | """start and end must be strings in the format accepted by date""" | |
1358 | if start is not None: | |
1359 | self.start = date_from_str(start, strict=True) | |
1360 | else: | |
1361 | self.start = dt.datetime.min.date() | |
1362 | if end is not None: | |
1363 | self.end = date_from_str(end, strict=True) | |
1364 | else: | |
1365 | self.end = dt.datetime.max.date() | |
1366 | if self.start > self.end: | |
1367 | raise ValueError('Date range: "%s" , the start date must be before the end date' % self) | |
1368 | ||
1369 | @classmethod | |
1370 | def day(cls, day): | |
1371 | """Returns a range that only contains the given day""" | |
1372 | return cls(day, day) | |
1373 | ||
1374 | def __contains__(self, date): | |
1375 | """Check if the date is in the range""" | |
1376 | if not isinstance(date, dt.date): | |
1377 | date = date_from_str(date) | |
1378 | return self.start <= date <= self.end | |
1379 | ||
1380 | def __repr__(self): | |
1381 | return f'{__name__}.{type(self).__name__}({self.start.isoformat()!r}, {self.end.isoformat()!r})' | |
1382 | ||
1383 | def __str__(self): | |
1384 | return f'{self.start} to {self.end}' | |
1385 | ||
1386 | def __eq__(self, other): | |
1387 | return (isinstance(other, DateRange) | |
1388 | and self.start == other.start and self.end == other.end) | |
1389 | ||
1390 | ||
1391 | @functools.cache | |
1392 | def system_identifier(): | |
1393 | python_implementation = platform.python_implementation() | |
1394 | if python_implementation == 'PyPy' and hasattr(sys, 'pypy_version_info'): | |
1395 | python_implementation += ' version %d.%d.%d' % sys.pypy_version_info[:3] | |
1396 | libc_ver = [] | |
1397 | with contextlib.suppress(OSError): # We may not have access to the executable | |
1398 | libc_ver = platform.libc_ver() | |
1399 | ||
1400 | return 'Python %s (%s %s %s) - %s (%s%s)' % ( | |
1401 | platform.python_version(), | |
1402 | python_implementation, | |
1403 | platform.machine(), | |
1404 | platform.architecture()[0], | |
1405 | platform.platform(), | |
1406 | ssl.OPENSSL_VERSION, | |
1407 | format_field(join_nonempty(*libc_ver, delim=' '), None, ', %s'), | |
1408 | ) | |
1409 | ||
1410 | ||
1411 | @functools.cache | |
1412 | def get_windows_version(): | |
1413 | ''' Get Windows version. returns () if it's not running on Windows ''' | |
1414 | if compat_os_name == 'nt': | |
1415 | return version_tuple(platform.win32_ver()[1]) | |
1416 | else: | |
1417 | return () | |
1418 | ||
1419 | ||
1420 | def write_string(s, out=None, encoding=None): | |
1421 | assert isinstance(s, str) | |
1422 | out = out or sys.stderr | |
1423 | # `sys.stderr` might be `None` (Ref: https://github.com/pyinstaller/pyinstaller/pull/7217) | |
1424 | if not out: | |
1425 | return | |
1426 | ||
1427 | if compat_os_name == 'nt' and supports_terminal_sequences(out): | |
1428 | s = re.sub(r'([\r\n]+)', r' \1', s) | |
1429 | ||
1430 | enc, buffer = None, out | |
1431 | # `mode` might be `None` (Ref: https://github.com/yt-dlp/yt-dlp/issues/8816) | |
1432 | if 'b' in (getattr(out, 'mode', None) or ''): | |
1433 | enc = encoding or preferredencoding() | |
1434 | elif hasattr(out, 'buffer'): | |
1435 | buffer = out.buffer | |
1436 | enc = encoding or getattr(out, 'encoding', None) or preferredencoding() | |
1437 | ||
1438 | buffer.write(s.encode(enc, 'ignore') if enc else s) | |
1439 | out.flush() | |
1440 | ||
1441 | ||
1442 | # TODO: Use global logger | |
1443 | def deprecation_warning(msg, *, printer=None, stacklevel=0, **kwargs): | |
1444 | from .. import _IN_CLI | |
1445 | if _IN_CLI: | |
1446 | if msg in deprecation_warning._cache: | |
1447 | return | |
1448 | deprecation_warning._cache.add(msg) | |
1449 | if printer: | |
1450 | return printer(f'{msg}{bug_reports_message()}', **kwargs) | |
1451 | return write_string(f'ERROR: {msg}{bug_reports_message()}\n', **kwargs) | |
1452 | else: | |
1453 | import warnings | |
1454 | warnings.warn(DeprecationWarning(msg), stacklevel=stacklevel + 3) | |
1455 | ||
1456 | ||
1457 | deprecation_warning._cache = set() | |
1458 | ||
1459 | ||
1460 | def bytes_to_intlist(bs): | |
1461 | if not bs: | |
1462 | return [] | |
1463 | if isinstance(bs[0], int): # Python 3 | |
1464 | return list(bs) | |
1465 | else: | |
1466 | return [ord(c) for c in bs] | |
1467 | ||
1468 | ||
1469 | def intlist_to_bytes(xs): | |
1470 | if not xs: | |
1471 | return b'' | |
1472 | return struct.pack('%dB' % len(xs), *xs) | |
1473 | ||
1474 | ||
1475 | class LockingUnsupportedError(OSError): | |
1476 | msg = 'File locking is not supported' | |
1477 | ||
1478 | def __init__(self): | |
1479 | super().__init__(self.msg) | |
1480 | ||
1481 | ||
1482 | # Cross-platform file locking | |
1483 | if sys.platform == 'win32': | |
1484 | import ctypes | |
1485 | import ctypes.wintypes | |
1486 | import msvcrt | |
1487 | ||
1488 | class OVERLAPPED(ctypes.Structure): | |
1489 | _fields_ = [ | |
1490 | ('Internal', ctypes.wintypes.LPVOID), | |
1491 | ('InternalHigh', ctypes.wintypes.LPVOID), | |
1492 | ('Offset', ctypes.wintypes.DWORD), | |
1493 | ('OffsetHigh', ctypes.wintypes.DWORD), | |
1494 | ('hEvent', ctypes.wintypes.HANDLE), | |
1495 | ] | |
1496 | ||
1497 | kernel32 = ctypes.WinDLL('kernel32') | |
1498 | LockFileEx = kernel32.LockFileEx | |
1499 | LockFileEx.argtypes = [ | |
1500 | ctypes.wintypes.HANDLE, # hFile | |
1501 | ctypes.wintypes.DWORD, # dwFlags | |
1502 | ctypes.wintypes.DWORD, # dwReserved | |
1503 | ctypes.wintypes.DWORD, # nNumberOfBytesToLockLow | |
1504 | ctypes.wintypes.DWORD, # nNumberOfBytesToLockHigh | |
1505 | ctypes.POINTER(OVERLAPPED) # Overlapped | |
1506 | ] | |
1507 | LockFileEx.restype = ctypes.wintypes.BOOL | |
1508 | UnlockFileEx = kernel32.UnlockFileEx | |
1509 | UnlockFileEx.argtypes = [ | |
1510 | ctypes.wintypes.HANDLE, # hFile | |
1511 | ctypes.wintypes.DWORD, # dwReserved | |
1512 | ctypes.wintypes.DWORD, # nNumberOfBytesToLockLow | |
1513 | ctypes.wintypes.DWORD, # nNumberOfBytesToLockHigh | |
1514 | ctypes.POINTER(OVERLAPPED) # Overlapped | |
1515 | ] | |
1516 | UnlockFileEx.restype = ctypes.wintypes.BOOL | |
1517 | whole_low = 0xffffffff | |
1518 | whole_high = 0x7fffffff | |
1519 | ||
1520 | def _lock_file(f, exclusive, block): | |
1521 | overlapped = OVERLAPPED() | |
1522 | overlapped.Offset = 0 | |
1523 | overlapped.OffsetHigh = 0 | |
1524 | overlapped.hEvent = 0 | |
1525 | f._lock_file_overlapped_p = ctypes.pointer(overlapped) | |
1526 | ||
1527 | if not LockFileEx(msvcrt.get_osfhandle(f.fileno()), | |
1528 | (0x2 if exclusive else 0x0) | (0x0 if block else 0x1), | |
1529 | 0, whole_low, whole_high, f._lock_file_overlapped_p): | |
1530 | # NB: No argument form of "ctypes.FormatError" does not work on PyPy | |
1531 | raise BlockingIOError(f'Locking file failed: {ctypes.FormatError(ctypes.GetLastError())!r}') | |
1532 | ||
1533 | def _unlock_file(f): | |
1534 | assert f._lock_file_overlapped_p | |
1535 | handle = msvcrt.get_osfhandle(f.fileno()) | |
1536 | if not UnlockFileEx(handle, 0, whole_low, whole_high, f._lock_file_overlapped_p): | |
1537 | raise OSError('Unlocking file failed: %r' % ctypes.FormatError()) | |
1538 | ||
1539 | else: | |
1540 | try: | |
1541 | import fcntl | |
1542 | ||
1543 | def _lock_file(f, exclusive, block): | |
1544 | flags = fcntl.LOCK_EX if exclusive else fcntl.LOCK_SH | |
1545 | if not block: | |
1546 | flags |= fcntl.LOCK_NB | |
1547 | try: | |
1548 | fcntl.flock(f, flags) | |
1549 | except BlockingIOError: | |
1550 | raise | |
1551 | except OSError: # AOSP does not have flock() | |
1552 | fcntl.lockf(f, flags) | |
1553 | ||
1554 | def _unlock_file(f): | |
1555 | with contextlib.suppress(OSError): | |
1556 | return fcntl.flock(f, fcntl.LOCK_UN) | |
1557 | with contextlib.suppress(OSError): | |
1558 | return fcntl.lockf(f, fcntl.LOCK_UN) # AOSP does not have flock() | |
1559 | return fcntl.flock(f, fcntl.LOCK_UN | fcntl.LOCK_NB) # virtiofs needs LOCK_NB on unlocking | |
1560 | ||
1561 | except ImportError: | |
1562 | ||
1563 | def _lock_file(f, exclusive, block): | |
1564 | raise LockingUnsupportedError() | |
1565 | ||
1566 | def _unlock_file(f): | |
1567 | raise LockingUnsupportedError() | |
1568 | ||
1569 | ||
1570 | class locked_file: | |
1571 | locked = False | |
1572 | ||
1573 | def __init__(self, filename, mode, block=True, encoding=None): | |
1574 | if mode not in {'r', 'rb', 'a', 'ab', 'w', 'wb'}: | |
1575 | raise NotImplementedError(mode) | |
1576 | self.mode, self.block = mode, block | |
1577 | ||
1578 | writable = any(f in mode for f in 'wax+') | |
1579 | readable = any(f in mode for f in 'r+') | |
1580 | flags = functools.reduce(operator.ior, ( | |
1581 | getattr(os, 'O_CLOEXEC', 0), # UNIX only | |
1582 | getattr(os, 'O_BINARY', 0), # Windows only | |
1583 | getattr(os, 'O_NOINHERIT', 0), # Windows only | |
1584 | os.O_CREAT if writable else 0, # O_TRUNC only after locking | |
1585 | os.O_APPEND if 'a' in mode else 0, | |
1586 | os.O_EXCL if 'x' in mode else 0, | |
1587 | os.O_RDONLY if not writable else os.O_RDWR if readable else os.O_WRONLY, | |
1588 | )) | |
1589 | ||
1590 | self.f = os.fdopen(os.open(filename, flags, 0o666), mode, encoding=encoding) | |
1591 | ||
1592 | def __enter__(self): | |
1593 | exclusive = 'r' not in self.mode | |
1594 | try: | |
1595 | _lock_file(self.f, exclusive, self.block) | |
1596 | self.locked = True | |
1597 | except OSError: | |
1598 | self.f.close() | |
1599 | raise | |
1600 | if 'w' in self.mode: | |
1601 | try: | |
1602 | self.f.truncate() | |
1603 | except OSError as e: | |
1604 | if e.errno not in ( | |
1605 | errno.ESPIPE, # Illegal seek - expected for FIFO | |
1606 | errno.EINVAL, # Invalid argument - expected for /dev/null | |
1607 | ): | |
1608 | raise | |
1609 | return self | |
1610 | ||
1611 | def unlock(self): | |
1612 | if not self.locked: | |
1613 | return | |
1614 | try: | |
1615 | _unlock_file(self.f) | |
1616 | finally: | |
1617 | self.locked = False | |
1618 | ||
1619 | def __exit__(self, *_): | |
1620 | try: | |
1621 | self.unlock() | |
1622 | finally: | |
1623 | self.f.close() | |
1624 | ||
1625 | open = __enter__ | |
1626 | close = __exit__ | |
1627 | ||
1628 | def __getattr__(self, attr): | |
1629 | return getattr(self.f, attr) | |
1630 | ||
1631 | def __iter__(self): | |
1632 | return iter(self.f) | |
1633 | ||
1634 | ||
1635 | @functools.cache | |
1636 | def get_filesystem_encoding(): | |
1637 | encoding = sys.getfilesystemencoding() | |
1638 | return encoding if encoding is not None else 'utf-8' | |
1639 | ||
1640 | ||
1641 | _WINDOWS_QUOTE_TRANS = str.maketrans({'"': R'\"'}) | |
1642 | _CMD_QUOTE_TRANS = str.maketrans({ | |
1643 | # Keep quotes balanced by replacing them with `""` instead of `\\"` | |
1644 | '"': '""', | |
1645 | # These require an env-variable `=` containing `"^\n\n"` (set in `utils.Popen`) | |
1646 | # `=` should be unique since variables containing `=` cannot be set using cmd | |
1647 | '\n': '%=%', | |
1648 | '\r': '%=%', | |
1649 | # Use zero length variable replacement so `%` doesn't get expanded | |
1650 | # `cd` is always set as long as extensions are enabled (`/E:ON` in `utils.Popen`) | |
1651 | '%': '%%cd:~,%', | |
1652 | }) | |
1653 | ||
1654 | ||
1655 | def shell_quote(args, *, shell=False): | |
1656 | args = list(variadic(args)) | |
1657 | ||
1658 | if compat_os_name != 'nt': | |
1659 | return shlex.join(args) | |
1660 | ||
1661 | trans = _CMD_QUOTE_TRANS if shell else _WINDOWS_QUOTE_TRANS | |
1662 | return ' '.join( | |
1663 | s if re.fullmatch(r'[\w#$*\-+./:?@\\]+', s, re.ASCII) | |
1664 | else re.sub(r'(\\+)("|$)', r'\1\1\2', s).translate(trans).join('""') | |
1665 | for s in args) | |
1666 | ||
1667 | ||
1668 | def smuggle_url(url, data): | |
1669 | """ Pass additional data in a URL for internal use. """ | |
1670 | ||
1671 | url, idata = unsmuggle_url(url, {}) | |
1672 | data.update(idata) | |
1673 | sdata = urllib.parse.urlencode( | |
1674 | {'__youtubedl_smuggle': json.dumps(data)}) | |
1675 | return url + '#' + sdata | |
1676 | ||
1677 | ||
1678 | def unsmuggle_url(smug_url, default=None): | |
1679 | if '#__youtubedl_smuggle' not in smug_url: | |
1680 | return smug_url, default | |
1681 | url, _, sdata = smug_url.rpartition('#') | |
1682 | jsond = urllib.parse.parse_qs(sdata)['__youtubedl_smuggle'][0] | |
1683 | data = json.loads(jsond) | |
1684 | return url, data | |
1685 | ||
1686 | ||
1687 | def format_decimal_suffix(num, fmt='%d%s', *, factor=1000): | |
1688 | """ Formats numbers with decimal sufixes like K, M, etc """ | |
1689 | num, factor = float_or_none(num), float(factor) | |
1690 | if num is None or num < 0: | |
1691 | return None | |
1692 | POSSIBLE_SUFFIXES = 'kMGTPEZY' | |
1693 | exponent = 0 if num == 0 else min(int(math.log(num, factor)), len(POSSIBLE_SUFFIXES)) | |
1694 | suffix = ['', *POSSIBLE_SUFFIXES][exponent] | |
1695 | if factor == 1024: | |
1696 | suffix = {'k': 'Ki', '': ''}.get(suffix, f'{suffix}i') | |
1697 | converted = num / (factor ** exponent) | |
1698 | return fmt % (converted, suffix) | |
1699 | ||
1700 | ||
1701 | def format_bytes(bytes): | |
1702 | return format_decimal_suffix(bytes, '%.2f%sB', factor=1024) or 'N/A' | |
1703 | ||
1704 | ||
1705 | def lookup_unit_table(unit_table, s, strict=False): | |
1706 | num_re = NUMBER_RE if strict else NUMBER_RE.replace(R'\.', '[,.]') | |
1707 | units_re = '|'.join(re.escape(u) for u in unit_table) | |
1708 | m = (re.fullmatch if strict else re.match)( | |
1709 | rf'(?P<num>{num_re})\s*(?P<unit>{units_re})\b', s) | |
1710 | if not m: | |
1711 | return None | |
1712 | ||
1713 | num = float(m.group('num').replace(',', '.')) | |
1714 | mult = unit_table[m.group('unit')] | |
1715 | return round(num * mult) | |
1716 | ||
1717 | ||
1718 | def parse_bytes(s): | |
1719 | """Parse a string indicating a byte quantity into an integer""" | |
1720 | return lookup_unit_table( | |
1721 | {u: 1024**i for i, u in enumerate(['', *'KMGTPEZY'])}, | |
1722 | s.upper(), strict=True) | |
1723 | ||
1724 | ||
1725 | def parse_filesize(s): | |
1726 | if s is None: | |
1727 | return None | |
1728 | ||
1729 | # The lower-case forms are of course incorrect and unofficial, | |
1730 | # but we support those too | |
1731 | _UNIT_TABLE = { | |
1732 | 'B': 1, | |
1733 | 'b': 1, | |
1734 | 'bytes': 1, | |
1735 | 'KiB': 1024, | |
1736 | 'KB': 1000, | |
1737 | 'kB': 1024, | |
1738 | 'Kb': 1000, | |
1739 | 'kb': 1000, | |
1740 | 'kilobytes': 1000, | |
1741 | 'kibibytes': 1024, | |
1742 | 'MiB': 1024 ** 2, | |
1743 | 'MB': 1000 ** 2, | |
1744 | 'mB': 1024 ** 2, | |
1745 | 'Mb': 1000 ** 2, | |
1746 | 'mb': 1000 ** 2, | |
1747 | 'megabytes': 1000 ** 2, | |
1748 | 'mebibytes': 1024 ** 2, | |
1749 | 'GiB': 1024 ** 3, | |
1750 | 'GB': 1000 ** 3, | |
1751 | 'gB': 1024 ** 3, | |
1752 | 'Gb': 1000 ** 3, | |
1753 | 'gb': 1000 ** 3, | |
1754 | 'gigabytes': 1000 ** 3, | |
1755 | 'gibibytes': 1024 ** 3, | |
1756 | 'TiB': 1024 ** 4, | |
1757 | 'TB': 1000 ** 4, | |
1758 | 'tB': 1024 ** 4, | |
1759 | 'Tb': 1000 ** 4, | |
1760 | 'tb': 1000 ** 4, | |
1761 | 'terabytes': 1000 ** 4, | |
1762 | 'tebibytes': 1024 ** 4, | |
1763 | 'PiB': 1024 ** 5, | |
1764 | 'PB': 1000 ** 5, | |
1765 | 'pB': 1024 ** 5, | |
1766 | 'Pb': 1000 ** 5, | |
1767 | 'pb': 1000 ** 5, | |
1768 | 'petabytes': 1000 ** 5, | |
1769 | 'pebibytes': 1024 ** 5, | |
1770 | 'EiB': 1024 ** 6, | |
1771 | 'EB': 1000 ** 6, | |
1772 | 'eB': 1024 ** 6, | |
1773 | 'Eb': 1000 ** 6, | |
1774 | 'eb': 1000 ** 6, | |
1775 | 'exabytes': 1000 ** 6, | |
1776 | 'exbibytes': 1024 ** 6, | |
1777 | 'ZiB': 1024 ** 7, | |
1778 | 'ZB': 1000 ** 7, | |
1779 | 'zB': 1024 ** 7, | |
1780 | 'Zb': 1000 ** 7, | |
1781 | 'zb': 1000 ** 7, | |
1782 | 'zettabytes': 1000 ** 7, | |
1783 | 'zebibytes': 1024 ** 7, | |
1784 | 'YiB': 1024 ** 8, | |
1785 | 'YB': 1000 ** 8, | |
1786 | 'yB': 1024 ** 8, | |
1787 | 'Yb': 1000 ** 8, | |
1788 | 'yb': 1000 ** 8, | |
1789 | 'yottabytes': 1000 ** 8, | |
1790 | 'yobibytes': 1024 ** 8, | |
1791 | } | |
1792 | ||
1793 | return lookup_unit_table(_UNIT_TABLE, s) | |
1794 | ||
1795 | ||
1796 | def parse_count(s): | |
1797 | if s is None: | |
1798 | return None | |
1799 | ||
1800 | s = re.sub(r'^[^\d]+\s', '', s).strip() | |
1801 | ||
1802 | if re.match(r'^[\d,.]+$', s): | |
1803 | return str_to_int(s) | |
1804 | ||
1805 | _UNIT_TABLE = { | |
1806 | 'k': 1000, | |
1807 | 'K': 1000, | |
1808 | 'm': 1000 ** 2, | |
1809 | 'M': 1000 ** 2, | |
1810 | 'kk': 1000 ** 2, | |
1811 | 'KK': 1000 ** 2, | |
1812 | 'b': 1000 ** 3, | |
1813 | 'B': 1000 ** 3, | |
1814 | } | |
1815 | ||
1816 | ret = lookup_unit_table(_UNIT_TABLE, s) | |
1817 | if ret is not None: | |
1818 | return ret | |
1819 | ||
1820 | mobj = re.match(r'([\d,.]+)(?:$|\s)', s) | |
1821 | if mobj: | |
1822 | return str_to_int(mobj.group(1)) | |
1823 | ||
1824 | ||
1825 | def parse_resolution(s, *, lenient=False): | |
1826 | if s is None: | |
1827 | return {} | |
1828 | ||
1829 | if lenient: | |
1830 | mobj = re.search(r'(?P<w>\d+)\s*[xX×,]\s*(?P<h>\d+)', s) | |
1831 | else: | |
1832 | mobj = re.search(r'(?<![a-zA-Z0-9])(?P<w>\d+)\s*[xX×,]\s*(?P<h>\d+)(?![a-zA-Z0-9])', s) | |
1833 | if mobj: | |
1834 | return { | |
1835 | 'width': int(mobj.group('w')), | |
1836 | 'height': int(mobj.group('h')), | |
1837 | } | |
1838 | ||
1839 | mobj = re.search(r'(?<![a-zA-Z0-9])(\d+)[pPiI](?![a-zA-Z0-9])', s) | |
1840 | if mobj: | |
1841 | return {'height': int(mobj.group(1))} | |
1842 | ||
1843 | mobj = re.search(r'\b([48])[kK]\b', s) | |
1844 | if mobj: | |
1845 | return {'height': int(mobj.group(1)) * 540} | |
1846 | ||
1847 | return {} | |
1848 | ||
1849 | ||
1850 | def parse_bitrate(s): | |
1851 | if not isinstance(s, str): | |
1852 | return | |
1853 | mobj = re.search(r'\b(\d+)\s*kbps', s) | |
1854 | if mobj: | |
1855 | return int(mobj.group(1)) | |
1856 | ||
1857 | ||
1858 | def month_by_name(name, lang='en'): | |
1859 | """ Return the number of a month by (locale-independently) English name """ | |
1860 | ||
1861 | month_names = MONTH_NAMES.get(lang, MONTH_NAMES['en']) | |
1862 | ||
1863 | try: | |
1864 | return month_names.index(name) + 1 | |
1865 | except ValueError: | |
1866 | return None | |
1867 | ||
1868 | ||
1869 | def month_by_abbreviation(abbrev): | |
1870 | """ Return the number of a month by (locale-independently) English | |
1871 | abbreviations """ | |
1872 | ||
1873 | try: | |
1874 | return [s[:3] for s in ENGLISH_MONTH_NAMES].index(abbrev) + 1 | |
1875 | except ValueError: | |
1876 | return None | |
1877 | ||
1878 | ||
1879 | def fix_xml_ampersands(xml_str): | |
1880 | """Replace all the '&' by '&' in XML""" | |
1881 | return re.sub( | |
1882 | r'&(?!amp;|lt;|gt;|apos;|quot;|#x[0-9a-fA-F]{,4};|#[0-9]{,4};)', | |
1883 | '&', | |
1884 | xml_str) | |
1885 | ||
1886 | ||
1887 | def setproctitle(title): | |
1888 | assert isinstance(title, str) | |
1889 | ||
1890 | # Workaround for https://github.com/yt-dlp/yt-dlp/issues/4541 | |
1891 | try: | |
1892 | import ctypes | |
1893 | except ImportError: | |
1894 | return | |
1895 | ||
1896 | try: | |
1897 | libc = ctypes.cdll.LoadLibrary('libc.so.6') | |
1898 | except OSError: | |
1899 | return | |
1900 | except TypeError: | |
1901 | # LoadLibrary in Windows Python 2.7.13 only expects | |
1902 | # a bytestring, but since unicode_literals turns | |
1903 | # every string into a unicode string, it fails. | |
1904 | return | |
1905 | title_bytes = title.encode() | |
1906 | buf = ctypes.create_string_buffer(len(title_bytes)) | |
1907 | buf.value = title_bytes | |
1908 | try: | |
1909 | # PR_SET_NAME = 15 Ref: /usr/include/linux/prctl.h | |
1910 | libc.prctl(15, buf, 0, 0, 0) | |
1911 | except AttributeError: | |
1912 | return # Strange libc, just skip this | |
1913 | ||
1914 | ||
1915 | def remove_start(s, start): | |
1916 | return s[len(start):] if s is not None and s.startswith(start) else s | |
1917 | ||
1918 | ||
1919 | def remove_end(s, end): | |
1920 | return s[:-len(end)] if s is not None and s.endswith(end) else s | |
1921 | ||
1922 | ||
1923 | def remove_quotes(s): | |
1924 | if s is None or len(s) < 2: | |
1925 | return s | |
1926 | for quote in ('"', "'", ): | |
1927 | if s[0] == quote and s[-1] == quote: | |
1928 | return s[1:-1] | |
1929 | return s | |
1930 | ||
1931 | ||
1932 | def get_domain(url): | |
1933 | """ | |
1934 | This implementation is inconsistent, but is kept for compatibility. | |
1935 | Use this only for "webpage_url_domain" | |
1936 | """ | |
1937 | return remove_start(urllib.parse.urlparse(url).netloc, 'www.') or None | |
1938 | ||
1939 | ||
1940 | def url_basename(url): | |
1941 | path = urllib.parse.urlparse(url).path | |
1942 | return path.strip('/').split('/')[-1] | |
1943 | ||
1944 | ||
1945 | def base_url(url): | |
1946 | return re.match(r'https?://[^?#]+/', url).group() | |
1947 | ||
1948 | ||
1949 | def urljoin(base, path): | |
1950 | if isinstance(path, bytes): | |
1951 | path = path.decode() | |
1952 | if not isinstance(path, str) or not path: | |
1953 | return None | |
1954 | if re.match(r'^(?:[a-zA-Z][a-zA-Z0-9+-.]*:)?//', path): | |
1955 | return path | |
1956 | if isinstance(base, bytes): | |
1957 | base = base.decode() | |
1958 | if not isinstance(base, str) or not re.match( | |
1959 | r'^(?:https?:)?//', base): | |
1960 | return None | |
1961 | return urllib.parse.urljoin(base, path) | |
1962 | ||
1963 | ||
1964 | def int_or_none(v, scale=1, default=None, get_attr=None, invscale=1): | |
1965 | if get_attr and v is not None: | |
1966 | v = getattr(v, get_attr, None) | |
1967 | try: | |
1968 | return int(v) * invscale // scale | |
1969 | except (ValueError, TypeError, OverflowError): | |
1970 | return default | |
1971 | ||
1972 | ||
1973 | def str_or_none(v, default=None): | |
1974 | return default if v is None else str(v) | |
1975 | ||
1976 | ||
1977 | def str_to_int(int_str): | |
1978 | """ A more relaxed version of int_or_none """ | |
1979 | if isinstance(int_str, int): | |
1980 | return int_str | |
1981 | elif isinstance(int_str, str): | |
1982 | int_str = re.sub(r'[,\.\+]', '', int_str) | |
1983 | return int_or_none(int_str) | |
1984 | ||
1985 | ||
1986 | def float_or_none(v, scale=1, invscale=1, default=None): | |
1987 | if v is None: | |
1988 | return default | |
1989 | try: | |
1990 | return float(v) * invscale / scale | |
1991 | except (ValueError, TypeError): | |
1992 | return default | |
1993 | ||
1994 | ||
1995 | def bool_or_none(v, default=None): | |
1996 | return v if isinstance(v, bool) else default | |
1997 | ||
1998 | ||
1999 | def strip_or_none(v, default=None): | |
2000 | return v.strip() if isinstance(v, str) else default | |
2001 | ||
2002 | ||
2003 | def url_or_none(url): | |
2004 | if not url or not isinstance(url, str): | |
2005 | return None | |
2006 | url = url.strip() | |
2007 | return url if re.match(r'^(?:(?:https?|rt(?:m(?:pt?[es]?|fp)|sp[su]?)|mms|ftps?):)?//', url) else None | |
2008 | ||
2009 | ||
2010 | def strftime_or_none(timestamp, date_format='%Y%m%d', default=None): | |
2011 | datetime_object = None | |
2012 | try: | |
2013 | if isinstance(timestamp, (int, float)): # unix timestamp | |
2014 | # Using naive datetime here can break timestamp() in Windows | |
2015 | # Ref: https://github.com/yt-dlp/yt-dlp/issues/5185, https://github.com/python/cpython/issues/94414 | |
2016 | # Also, dt.datetime.fromtimestamp breaks for negative timestamps | |
2017 | # Ref: https://github.com/yt-dlp/yt-dlp/issues/6706#issuecomment-1496842642 | |
2018 | datetime_object = (dt.datetime.fromtimestamp(0, dt.timezone.utc) | |
2019 | + dt.timedelta(seconds=timestamp)) | |
2020 | elif isinstance(timestamp, str): # assume YYYYMMDD | |
2021 | datetime_object = dt.datetime.strptime(timestamp, '%Y%m%d') | |
2022 | date_format = re.sub( # Support %s on windows | |
2023 | r'(?<!%)(%%)*%s', rf'\g<1>{int(datetime_object.timestamp())}', date_format) | |
2024 | return datetime_object.strftime(date_format) | |
2025 | except (ValueError, TypeError, AttributeError): | |
2026 | return default | |
2027 | ||
2028 | ||
2029 | def parse_duration(s): | |
2030 | if not isinstance(s, str): | |
2031 | return None | |
2032 | s = s.strip() | |
2033 | if not s: | |
2034 | return None | |
2035 | ||
2036 | days, hours, mins, secs, ms = [None] * 5 | |
2037 | m = re.match(r'''(?x) | |
2038 | (?P<before_secs> | |
2039 | (?:(?:(?P<days>[0-9]+):)?(?P<hours>[0-9]+):)?(?P<mins>[0-9]+):)? | |
2040 | (?P<secs>(?(before_secs)[0-9]{1,2}|[0-9]+)) | |
2041 | (?P<ms>[.:][0-9]+)?Z?$ | |
2042 | ''', s) | |
2043 | if m: | |
2044 | days, hours, mins, secs, ms = m.group('days', 'hours', 'mins', 'secs', 'ms') | |
2045 | else: | |
2046 | m = re.match( | |
2047 | r'''(?ix)(?:P? | |
2048 | (?: | |
2049 | [0-9]+\s*y(?:ears?)?,?\s* | |
2050 | )? | |
2051 | (?: | |
2052 | [0-9]+\s*m(?:onths?)?,?\s* | |
2053 | )? | |
2054 | (?: | |
2055 | [0-9]+\s*w(?:eeks?)?,?\s* | |
2056 | )? | |
2057 | (?: | |
2058 | (?P<days>[0-9]+)\s*d(?:ays?)?,?\s* | |
2059 | )? | |
2060 | T)? | |
2061 | (?: | |
2062 | (?P<hours>[0-9]+)\s*h(?:(?:ou)?rs?)?,?\s* | |
2063 | )? | |
2064 | (?: | |
2065 | (?P<mins>[0-9]+)\s*m(?:in(?:ute)?s?)?,?\s* | |
2066 | )? | |
2067 | (?: | |
2068 | (?P<secs>[0-9]+)(?P<ms>\.[0-9]+)?\s*s(?:ec(?:ond)?s?)?\s* | |
2069 | )?Z?$''', s) | |
2070 | if m: | |
2071 | days, hours, mins, secs, ms = m.groups() | |
2072 | else: | |
2073 | m = re.match(r'(?i)(?:(?P<hours>[0-9.]+)\s*(?:hours?)|(?P<mins>[0-9.]+)\s*(?:mins?\.?|minutes?)\s*)Z?$', s) | |
2074 | if m: | |
2075 | hours, mins = m.groups() | |
2076 | else: | |
2077 | return None | |
2078 | ||
2079 | if ms: | |
2080 | ms = ms.replace(':', '.') | |
2081 | return sum(float(part or 0) * mult for part, mult in ( | |
2082 | (days, 86400), (hours, 3600), (mins, 60), (secs, 1), (ms, 1))) | |
2083 | ||
2084 | ||
2085 | def prepend_extension(filename, ext, expected_real_ext=None): | |
2086 | name, real_ext = os.path.splitext(filename) | |
2087 | return ( | |
2088 | f'{name}.{ext}{real_ext}' | |
2089 | if not expected_real_ext or real_ext[1:] == expected_real_ext | |
2090 | else f'{filename}.{ext}') | |
2091 | ||
2092 | ||
2093 | def replace_extension(filename, ext, expected_real_ext=None): | |
2094 | name, real_ext = os.path.splitext(filename) | |
2095 | return '{}.{}'.format( | |
2096 | name if not expected_real_ext or real_ext[1:] == expected_real_ext else filename, | |
2097 | ext) | |
2098 | ||
2099 | ||
2100 | def check_executable(exe, args=[]): | |
2101 | """ Checks if the given binary is installed somewhere in PATH, and returns its name. | |
2102 | args can be a list of arguments for a short output (like -version) """ | |
2103 | try: | |
2104 | Popen.run([exe] + args, stdout=subprocess.PIPE, stderr=subprocess.PIPE) | |
2105 | except OSError: | |
2106 | return False | |
2107 | return exe | |
2108 | ||
2109 | ||
2110 | def _get_exe_version_output(exe, args): | |
2111 | try: | |
2112 | # STDIN should be redirected too. On UNIX-like systems, ffmpeg triggers | |
2113 | # SIGTTOU if yt-dlp is run in the background. | |
2114 | # See https://github.com/ytdl-org/youtube-dl/issues/955#issuecomment-209789656 | |
2115 | stdout, _, ret = Popen.run([encodeArgument(exe)] + args, text=True, | |
2116 | stdin=subprocess.PIPE, stdout=subprocess.PIPE, stderr=subprocess.STDOUT) | |
2117 | if ret: | |
2118 | return None | |
2119 | except OSError: | |
2120 | return False | |
2121 | return stdout | |
2122 | ||
2123 | ||
2124 | def detect_exe_version(output, version_re=None, unrecognized='present'): | |
2125 | assert isinstance(output, str) | |
2126 | if version_re is None: | |
2127 | version_re = r'version\s+([-0-9._a-zA-Z]+)' | |
2128 | m = re.search(version_re, output) | |
2129 | if m: | |
2130 | return m.group(1) | |
2131 | else: | |
2132 | return unrecognized | |
2133 | ||
2134 | ||
2135 | def get_exe_version(exe, args=['--version'], | |
2136 | version_re=None, unrecognized=('present', 'broken')): | |
2137 | """ Returns the version of the specified executable, | |
2138 | or False if the executable is not present """ | |
2139 | unrecognized = variadic(unrecognized) | |
2140 | assert len(unrecognized) in (1, 2) | |
2141 | out = _get_exe_version_output(exe, args) | |
2142 | if out is None: | |
2143 | return unrecognized[-1] | |
2144 | return out and detect_exe_version(out, version_re, unrecognized[0]) | |
2145 | ||
2146 | ||
2147 | def frange(start=0, stop=None, step=1): | |
2148 | """Float range""" | |
2149 | if stop is None: | |
2150 | start, stop = 0, start | |
2151 | sign = [-1, 1][step > 0] if step else 0 | |
2152 | while sign * start < sign * stop: | |
2153 | yield start | |
2154 | start += step | |
2155 | ||
2156 | ||
2157 | class LazyList(collections.abc.Sequence): | |
2158 | """Lazy immutable list from an iterable | |
2159 | Note that slices of a LazyList are lists and not LazyList""" | |
2160 | ||
2161 | class IndexError(IndexError): | |
2162 | pass | |
2163 | ||
2164 | def __init__(self, iterable, *, reverse=False, _cache=None): | |
2165 | self._iterable = iter(iterable) | |
2166 | self._cache = [] if _cache is None else _cache | |
2167 | self._reversed = reverse | |
2168 | ||
2169 | def __iter__(self): | |
2170 | if self._reversed: | |
2171 | # We need to consume the entire iterable to iterate in reverse | |
2172 | yield from self.exhaust() | |
2173 | return | |
2174 | yield from self._cache | |
2175 | for item in self._iterable: | |
2176 | self._cache.append(item) | |
2177 | yield item | |
2178 | ||
2179 | def _exhaust(self): | |
2180 | self._cache.extend(self._iterable) | |
2181 | self._iterable = [] # Discard the emptied iterable to make it pickle-able | |
2182 | return self._cache | |
2183 | ||
2184 | def exhaust(self): | |
2185 | """Evaluate the entire iterable""" | |
2186 | return self._exhaust()[::-1 if self._reversed else 1] | |
2187 | ||
2188 | @staticmethod | |
2189 | def _reverse_index(x): | |
2190 | return None if x is None else ~x | |
2191 | ||
2192 | def __getitem__(self, idx): | |
2193 | if isinstance(idx, slice): | |
2194 | if self._reversed: | |
2195 | idx = slice(self._reverse_index(idx.start), self._reverse_index(idx.stop), -(idx.step or 1)) | |
2196 | start, stop, step = idx.start, idx.stop, idx.step or 1 | |
2197 | elif isinstance(idx, int): | |
2198 | if self._reversed: | |
2199 | idx = self._reverse_index(idx) | |
2200 | start, stop, step = idx, idx, 0 | |
2201 | else: | |
2202 | raise TypeError('indices must be integers or slices') | |
2203 | if ((start or 0) < 0 or (stop or 0) < 0 | |
2204 | or (start is None and step < 0) | |
2205 | or (stop is None and step > 0)): | |
2206 | # We need to consume the entire iterable to be able to slice from the end | |
2207 | # Obviously, never use this with infinite iterables | |
2208 | self._exhaust() | |
2209 | try: | |
2210 | return self._cache[idx] | |
2211 | except IndexError as e: | |
2212 | raise self.IndexError(e) from e | |
2213 | n = max(start or 0, stop or 0) - len(self._cache) + 1 | |
2214 | if n > 0: | |
2215 | self._cache.extend(itertools.islice(self._iterable, n)) | |
2216 | try: | |
2217 | return self._cache[idx] | |
2218 | except IndexError as e: | |
2219 | raise self.IndexError(e) from e | |
2220 | ||
2221 | def __bool__(self): | |
2222 | try: | |
2223 | self[-1] if self._reversed else self[0] | |
2224 | except self.IndexError: | |
2225 | return False | |
2226 | return True | |
2227 | ||
2228 | def __len__(self): | |
2229 | self._exhaust() | |
2230 | return len(self._cache) | |
2231 | ||
2232 | def __reversed__(self): | |
2233 | return type(self)(self._iterable, reverse=not self._reversed, _cache=self._cache) | |
2234 | ||
2235 | def __copy__(self): | |
2236 | return type(self)(self._iterable, reverse=self._reversed, _cache=self._cache) | |
2237 | ||
2238 | def __repr__(self): | |
2239 | # repr and str should mimic a list. So we exhaust the iterable | |
2240 | return repr(self.exhaust()) | |
2241 | ||
2242 | def __str__(self): | |
2243 | return repr(self.exhaust()) | |
2244 | ||
2245 | ||
2246 | class PagedList: | |
2247 | ||
2248 | class IndexError(IndexError): | |
2249 | pass | |
2250 | ||
2251 | def __len__(self): | |
2252 | # This is only useful for tests | |
2253 | return len(self.getslice()) | |
2254 | ||
2255 | def __init__(self, pagefunc, pagesize, use_cache=True): | |
2256 | self._pagefunc = pagefunc | |
2257 | self._pagesize = pagesize | |
2258 | self._pagecount = float('inf') | |
2259 | self._use_cache = use_cache | |
2260 | self._cache = {} | |
2261 | ||
2262 | def getpage(self, pagenum): | |
2263 | page_results = self._cache.get(pagenum) | |
2264 | if page_results is None: | |
2265 | page_results = [] if pagenum > self._pagecount else list(self._pagefunc(pagenum)) | |
2266 | if self._use_cache: | |
2267 | self._cache[pagenum] = page_results | |
2268 | return page_results | |
2269 | ||
2270 | def getslice(self, start=0, end=None): | |
2271 | return list(self._getslice(start, end)) | |
2272 | ||
2273 | def _getslice(self, start, end): | |
2274 | raise NotImplementedError('This method must be implemented by subclasses') | |
2275 | ||
2276 | def __getitem__(self, idx): | |
2277 | assert self._use_cache, 'Indexing PagedList requires cache' | |
2278 | if not isinstance(idx, int) or idx < 0: | |
2279 | raise TypeError('indices must be non-negative integers') | |
2280 | entries = self.getslice(idx, idx + 1) | |
2281 | if not entries: | |
2282 | raise self.IndexError() | |
2283 | return entries[0] | |
2284 | ||
2285 | def __bool__(self): | |
2286 | return bool(self.getslice(0, 1)) | |
2287 | ||
2288 | ||
2289 | class OnDemandPagedList(PagedList): | |
2290 | """Download pages until a page with less than maximum results""" | |
2291 | ||
2292 | def _getslice(self, start, end): | |
2293 | for pagenum in itertools.count(start // self._pagesize): | |
2294 | firstid = pagenum * self._pagesize | |
2295 | nextfirstid = pagenum * self._pagesize + self._pagesize | |
2296 | if start >= nextfirstid: | |
2297 | continue | |
2298 | ||
2299 | startv = ( | |
2300 | start % self._pagesize | |
2301 | if firstid <= start < nextfirstid | |
2302 | else 0) | |
2303 | endv = ( | |
2304 | ((end - 1) % self._pagesize) + 1 | |
2305 | if (end is not None and firstid <= end <= nextfirstid) | |
2306 | else None) | |
2307 | ||
2308 | try: | |
2309 | page_results = self.getpage(pagenum) | |
2310 | except Exception: | |
2311 | self._pagecount = pagenum - 1 | |
2312 | raise | |
2313 | if startv != 0 or endv is not None: | |
2314 | page_results = page_results[startv:endv] | |
2315 | yield from page_results | |
2316 | ||
2317 | # A little optimization - if current page is not "full", ie. does | |
2318 | # not contain page_size videos then we can assume that this page | |
2319 | # is the last one - there are no more ids on further pages - | |
2320 | # i.e. no need to query again. | |
2321 | if len(page_results) + startv < self._pagesize: | |
2322 | break | |
2323 | ||
2324 | # If we got the whole page, but the next page is not interesting, | |
2325 | # break out early as well | |
2326 | if end == nextfirstid: | |
2327 | break | |
2328 | ||
2329 | ||
2330 | class InAdvancePagedList(PagedList): | |
2331 | """PagedList with total number of pages known in advance""" | |
2332 | ||
2333 | def __init__(self, pagefunc, pagecount, pagesize): | |
2334 | PagedList.__init__(self, pagefunc, pagesize, True) | |
2335 | self._pagecount = pagecount | |
2336 | ||
2337 | def _getslice(self, start, end): | |
2338 | start_page = start // self._pagesize | |
2339 | end_page = self._pagecount if end is None else min(self._pagecount, end // self._pagesize + 1) | |
2340 | skip_elems = start - start_page * self._pagesize | |
2341 | only_more = None if end is None else end - start | |
2342 | for pagenum in range(start_page, end_page): | |
2343 | page_results = self.getpage(pagenum) | |
2344 | if skip_elems: | |
2345 | page_results = page_results[skip_elems:] | |
2346 | skip_elems = None | |
2347 | if only_more is not None: | |
2348 | if len(page_results) < only_more: | |
2349 | only_more -= len(page_results) | |
2350 | else: | |
2351 | yield from page_results[:only_more] | |
2352 | break | |
2353 | yield from page_results | |
2354 | ||
2355 | ||
2356 | class PlaylistEntries: | |
2357 | MissingEntry = object() | |
2358 | is_exhausted = False | |
2359 | ||
2360 | def __init__(self, ydl, info_dict): | |
2361 | self.ydl = ydl | |
2362 | ||
2363 | # _entries must be assigned now since infodict can change during iteration | |
2364 | entries = info_dict.get('entries') | |
2365 | if entries is None: | |
2366 | raise EntryNotInPlaylist('There are no entries') | |
2367 | elif isinstance(entries, list): | |
2368 | self.is_exhausted = True | |
2369 | ||
2370 | requested_entries = info_dict.get('requested_entries') | |
2371 | self.is_incomplete = requested_entries is not None | |
2372 | if self.is_incomplete: | |
2373 | assert self.is_exhausted | |
2374 | self._entries = [self.MissingEntry] * max(requested_entries or [0]) | |
2375 | for i, entry in zip(requested_entries, entries): | |
2376 | self._entries[i - 1] = entry | |
2377 | elif isinstance(entries, (list, PagedList, LazyList)): | |
2378 | self._entries = entries | |
2379 | else: | |
2380 | self._entries = LazyList(entries) | |
2381 | ||
2382 | PLAYLIST_ITEMS_RE = re.compile(r'''(?x) | |
2383 | (?P<start>[+-]?\d+)? | |
2384 | (?P<range>[:-] | |
2385 | (?P<end>[+-]?\d+|inf(?:inite)?)? | |
2386 | (?::(?P<step>[+-]?\d+))? | |
2387 | )?''') | |
2388 | ||
2389 | @classmethod | |
2390 | def parse_playlist_items(cls, string): | |
2391 | for segment in string.split(','): | |
2392 | if not segment: | |
2393 | raise ValueError('There is two or more consecutive commas') | |
2394 | mobj = cls.PLAYLIST_ITEMS_RE.fullmatch(segment) | |
2395 | if not mobj: | |
2396 | raise ValueError(f'{segment!r} is not a valid specification') | |
2397 | start, end, step, has_range = mobj.group('start', 'end', 'step', 'range') | |
2398 | if int_or_none(step) == 0: | |
2399 | raise ValueError(f'Step in {segment!r} cannot be zero') | |
2400 | yield slice(int_or_none(start), float_or_none(end), int_or_none(step)) if has_range else int(start) | |
2401 | ||
2402 | def get_requested_items(self): | |
2403 | playlist_items = self.ydl.params.get('playlist_items') | |
2404 | playlist_start = self.ydl.params.get('playliststart', 1) | |
2405 | playlist_end = self.ydl.params.get('playlistend') | |
2406 | # For backwards compatibility, interpret -1 as whole list | |
2407 | if playlist_end in (-1, None): | |
2408 | playlist_end = '' | |
2409 | if not playlist_items: | |
2410 | playlist_items = f'{playlist_start}:{playlist_end}' | |
2411 | elif playlist_start != 1 or playlist_end: | |
2412 | self.ydl.report_warning('Ignoring playliststart and playlistend because playlistitems was given', only_once=True) | |
2413 | ||
2414 | for index in self.parse_playlist_items(playlist_items): | |
2415 | for i, entry in self[index]: | |
2416 | yield i, entry | |
2417 | if not entry: | |
2418 | continue | |
2419 | try: | |
2420 | # The item may have just been added to archive. Don't break due to it | |
2421 | if not self.ydl.params.get('lazy_playlist'): | |
2422 | # TODO: Add auto-generated fields | |
2423 | self.ydl._match_entry(entry, incomplete=True, silent=True) | |
2424 | except (ExistingVideoReached, RejectedVideoReached): | |
2425 | return | |
2426 | ||
2427 | def get_full_count(self): | |
2428 | if self.is_exhausted and not self.is_incomplete: | |
2429 | return len(self) | |
2430 | elif isinstance(self._entries, InAdvancePagedList): | |
2431 | if self._entries._pagesize == 1: | |
2432 | return self._entries._pagecount | |
2433 | ||
2434 | @functools.cached_property | |
2435 | def _getter(self): | |
2436 | if isinstance(self._entries, list): | |
2437 | def get_entry(i): | |
2438 | try: | |
2439 | entry = self._entries[i] | |
2440 | except IndexError: | |
2441 | entry = self.MissingEntry | |
2442 | if not self.is_incomplete: | |
2443 | raise self.IndexError() | |
2444 | if entry is self.MissingEntry: | |
2445 | raise EntryNotInPlaylist(f'Entry {i + 1} cannot be found') | |
2446 | return entry | |
2447 | else: | |
2448 | def get_entry(i): | |
2449 | try: | |
2450 | return type(self.ydl)._handle_extraction_exceptions(lambda _, i: self._entries[i])(self.ydl, i) | |
2451 | except (LazyList.IndexError, PagedList.IndexError): | |
2452 | raise self.IndexError() | |
2453 | return get_entry | |
2454 | ||
2455 | def __getitem__(self, idx): | |
2456 | if isinstance(idx, int): | |
2457 | idx = slice(idx, idx) | |
2458 | ||
2459 | # NB: PlaylistEntries[1:10] => (0, 1, ... 9) | |
2460 | step = 1 if idx.step is None else idx.step | |
2461 | if idx.start is None: | |
2462 | start = 0 if step > 0 else len(self) - 1 | |
2463 | else: | |
2464 | start = idx.start - 1 if idx.start >= 0 else len(self) + idx.start | |
2465 | ||
2466 | # NB: Do not call len(self) when idx == [:] | |
2467 | if idx.stop is None: | |
2468 | stop = 0 if step < 0 else float('inf') | |
2469 | else: | |
2470 | stop = idx.stop - 1 if idx.stop >= 0 else len(self) + idx.stop | |
2471 | stop += [-1, 1][step > 0] | |
2472 | ||
2473 | for i in frange(start, stop, step): | |
2474 | if i < 0: | |
2475 | continue | |
2476 | try: | |
2477 | entry = self._getter(i) | |
2478 | except self.IndexError: | |
2479 | self.is_exhausted = True | |
2480 | if step > 0: | |
2481 | break | |
2482 | continue | |
2483 | yield i + 1, entry | |
2484 | ||
2485 | def __len__(self): | |
2486 | return len(tuple(self[:])) | |
2487 | ||
2488 | class IndexError(IndexError): | |
2489 | pass | |
2490 | ||
2491 | ||
2492 | def uppercase_escape(s): | |
2493 | unicode_escape = codecs.getdecoder('unicode_escape') | |
2494 | return re.sub( | |
2495 | r'\\U[0-9a-fA-F]{8}', | |
2496 | lambda m: unicode_escape(m.group(0))[0], | |
2497 | s) | |
2498 | ||
2499 | ||
2500 | def lowercase_escape(s): | |
2501 | unicode_escape = codecs.getdecoder('unicode_escape') | |
2502 | return re.sub( | |
2503 | r'\\u[0-9a-fA-F]{4}', | |
2504 | lambda m: unicode_escape(m.group(0))[0], | |
2505 | s) | |
2506 | ||
2507 | ||
2508 | def parse_qs(url, **kwargs): | |
2509 | return urllib.parse.parse_qs(urllib.parse.urlparse(url).query, **kwargs) | |
2510 | ||
2511 | ||
2512 | def read_batch_urls(batch_fd): | |
2513 | def fixup(url): | |
2514 | if not isinstance(url, str): | |
2515 | url = url.decode('utf-8', 'replace') | |
2516 | BOM_UTF8 = ('\xef\xbb\xbf', '\ufeff') | |
2517 | for bom in BOM_UTF8: | |
2518 | if url.startswith(bom): | |
2519 | url = url[len(bom):] | |
2520 | url = url.lstrip() | |
2521 | if not url or url.startswith(('#', ';', ']')): | |
2522 | return False | |
2523 | # "#" cannot be stripped out since it is part of the URI | |
2524 | # However, it can be safely stripped out if following a whitespace | |
2525 | return re.split(r'\s#', url, 1)[0].rstrip() | |
2526 | ||
2527 | with contextlib.closing(batch_fd) as fd: | |
2528 | return [url for url in map(fixup, fd) if url] | |
2529 | ||
2530 | ||
2531 | def urlencode_postdata(*args, **kargs): | |
2532 | return urllib.parse.urlencode(*args, **kargs).encode('ascii') | |
2533 | ||
2534 | ||
2535 | def update_url(url, *, query_update=None, **kwargs): | |
2536 | """Replace URL components specified by kwargs | |
2537 | @param url str or parse url tuple | |
2538 | @param query_update update query | |
2539 | @returns str | |
2540 | """ | |
2541 | if isinstance(url, str): | |
2542 | if not kwargs and not query_update: | |
2543 | return url | |
2544 | else: | |
2545 | url = urllib.parse.urlparse(url) | |
2546 | if query_update: | |
2547 | assert 'query' not in kwargs, 'query_update and query cannot be specified at the same time' | |
2548 | kwargs['query'] = urllib.parse.urlencode({ | |
2549 | **urllib.parse.parse_qs(url.query), | |
2550 | **query_update | |
2551 | }, True) | |
2552 | return urllib.parse.urlunparse(url._replace(**kwargs)) | |
2553 | ||
2554 | ||
2555 | def update_url_query(url, query): | |
2556 | return update_url(url, query_update=query) | |
2557 | ||
2558 | ||
2559 | def _multipart_encode_impl(data, boundary): | |
2560 | content_type = 'multipart/form-data; boundary=%s' % boundary | |
2561 | ||
2562 | out = b'' | |
2563 | for k, v in data.items(): | |
2564 | out += b'--' + boundary.encode('ascii') + b'\r\n' | |
2565 | if isinstance(k, str): | |
2566 | k = k.encode() | |
2567 | if isinstance(v, str): | |
2568 | v = v.encode() | |
2569 | # RFC 2047 requires non-ASCII field names to be encoded, while RFC 7578 | |
2570 | # suggests sending UTF-8 directly. Firefox sends UTF-8, too | |
2571 | content = b'Content-Disposition: form-data; name="' + k + b'"\r\n\r\n' + v + b'\r\n' | |
2572 | if boundary.encode('ascii') in content: | |
2573 | raise ValueError('Boundary overlaps with data') | |
2574 | out += content | |
2575 | ||
2576 | out += b'--' + boundary.encode('ascii') + b'--\r\n' | |
2577 | ||
2578 | return out, content_type | |
2579 | ||
2580 | ||
2581 | def multipart_encode(data, boundary=None): | |
2582 | ''' | |
2583 | Encode a dict to RFC 7578-compliant form-data | |
2584 | ||
2585 | data: | |
2586 | A dict where keys and values can be either Unicode or bytes-like | |
2587 | objects. | |
2588 | boundary: | |
2589 | If specified a Unicode object, it's used as the boundary. Otherwise | |
2590 | a random boundary is generated. | |
2591 | ||
2592 | Reference: https://tools.ietf.org/html/rfc7578 | |
2593 | ''' | |
2594 | has_specified_boundary = boundary is not None | |
2595 | ||
2596 | while True: | |
2597 | if boundary is None: | |
2598 | boundary = '---------------' + str(random.randrange(0x0fffffff, 0xffffffff)) | |
2599 | ||
2600 | try: | |
2601 | out, content_type = _multipart_encode_impl(data, boundary) | |
2602 | break | |
2603 | except ValueError: | |
2604 | if has_specified_boundary: | |
2605 | raise | |
2606 | boundary = None | |
2607 | ||
2608 | return out, content_type | |
2609 | ||
2610 | ||
2611 | def is_iterable_like(x, allowed_types=collections.abc.Iterable, blocked_types=NO_DEFAULT): | |
2612 | if blocked_types is NO_DEFAULT: | |
2613 | blocked_types = (str, bytes, collections.abc.Mapping) | |
2614 | return isinstance(x, allowed_types) and not isinstance(x, blocked_types) | |
2615 | ||
2616 | ||
2617 | def variadic(x, allowed_types=NO_DEFAULT): | |
2618 | if not isinstance(allowed_types, (tuple, type)): | |
2619 | deprecation_warning('allowed_types should be a tuple or a type') | |
2620 | allowed_types = tuple(allowed_types) | |
2621 | return x if is_iterable_like(x, blocked_types=allowed_types) else (x, ) | |
2622 | ||
2623 | ||
2624 | def try_call(*funcs, expected_type=None, args=[], kwargs={}): | |
2625 | for f in funcs: | |
2626 | try: | |
2627 | val = f(*args, **kwargs) | |
2628 | except (AttributeError, KeyError, TypeError, IndexError, ValueError, ZeroDivisionError): | |
2629 | pass | |
2630 | else: | |
2631 | if expected_type is None or isinstance(val, expected_type): | |
2632 | return val | |
2633 | ||
2634 | ||
2635 | def try_get(src, getter, expected_type=None): | |
2636 | return try_call(*variadic(getter), args=(src,), expected_type=expected_type) | |
2637 | ||
2638 | ||
2639 | def filter_dict(dct, cndn=lambda _, v: v is not None): | |
2640 | return {k: v for k, v in dct.items() if cndn(k, v)} | |
2641 | ||
2642 | ||
2643 | def merge_dicts(*dicts): | |
2644 | merged = {} | |
2645 | for a_dict in dicts: | |
2646 | for k, v in a_dict.items(): | |
2647 | if (v is not None and k not in merged | |
2648 | or isinstance(v, str) and merged[k] == ''): | |
2649 | merged[k] = v | |
2650 | return merged | |
2651 | ||
2652 | ||
2653 | def encode_compat_str(string, encoding=preferredencoding(), errors='strict'): | |
2654 | return string if isinstance(string, str) else str(string, encoding, errors) | |
2655 | ||
2656 | ||
2657 | US_RATINGS = { | |
2658 | 'G': 0, | |
2659 | 'PG': 10, | |
2660 | 'PG-13': 13, | |
2661 | 'R': 16, | |
2662 | 'NC': 18, | |
2663 | } | |
2664 | ||
2665 | ||
2666 | TV_PARENTAL_GUIDELINES = { | |
2667 | 'TV-Y': 0, | |
2668 | 'TV-Y7': 7, | |
2669 | 'TV-G': 0, | |
2670 | 'TV-PG': 0, | |
2671 | 'TV-14': 14, | |
2672 | 'TV-MA': 17, | |
2673 | } | |
2674 | ||
2675 | ||
2676 | def parse_age_limit(s): | |
2677 | # isinstance(False, int) is True. So type() must be used instead | |
2678 | if type(s) is int: # noqa: E721 | |
2679 | return s if 0 <= s <= 21 else None | |
2680 | elif not isinstance(s, str): | |
2681 | return None | |
2682 | m = re.match(r'^(?P<age>\d{1,2})\+?$', s) | |
2683 | if m: | |
2684 | return int(m.group('age')) | |
2685 | s = s.upper() | |
2686 | if s in US_RATINGS: | |
2687 | return US_RATINGS[s] | |
2688 | m = re.match(r'^TV[_-]?(%s)$' % '|'.join(k[3:] for k in TV_PARENTAL_GUIDELINES), s) | |
2689 | if m: | |
2690 | return TV_PARENTAL_GUIDELINES['TV-' + m.group(1)] | |
2691 | return None | |
2692 | ||
2693 | ||
2694 | def strip_jsonp(code): | |
2695 | return re.sub( | |
2696 | r'''(?sx)^ | |
2697 | (?:window\.)?(?P<func_name>[a-zA-Z0-9_.$]*) | |
2698 | (?:\s*&&\s*(?P=func_name))? | |
2699 | \s*\(\s*(?P<callback_data>.*)\);? | |
2700 | \s*?(?://[^\n]*)*$''', | |
2701 | r'\g<callback_data>', code) | |
2702 | ||
2703 | ||
2704 | def js_to_json(code, vars={}, *, strict=False): | |
2705 | # vars is a dict of var, val pairs to substitute | |
2706 | STRING_QUOTES = '\'"`' | |
2707 | STRING_RE = '|'.join(rf'{q}(?:\\.|[^\\{q}])*{q}' for q in STRING_QUOTES) | |
2708 | COMMENT_RE = r'/\*(?:(?!\*/).)*?\*/|//[^\n]*\n' | |
2709 | SKIP_RE = fr'\s*(?:{COMMENT_RE})?\s*' | |
2710 | INTEGER_TABLE = ( | |
2711 | (fr'(?s)^(0[xX][0-9a-fA-F]+){SKIP_RE}:?$', 16), | |
2712 | (fr'(?s)^(0+[0-7]+){SKIP_RE}:?$', 8), | |
2713 | ) | |
2714 | ||
2715 | def process_escape(match): | |
2716 | JSON_PASSTHROUGH_ESCAPES = R'"\bfnrtu' | |
2717 | escape = match.group(1) or match.group(2) | |
2718 | ||
2719 | return (Rf'\{escape}' if escape in JSON_PASSTHROUGH_ESCAPES | |
2720 | else R'\u00' if escape == 'x' | |
2721 | else '' if escape == '\n' | |
2722 | else escape) | |
2723 | ||
2724 | def template_substitute(match): | |
2725 | evaluated = js_to_json(match.group(1), vars, strict=strict) | |
2726 | if evaluated[0] == '"': | |
2727 | return json.loads(evaluated) | |
2728 | return evaluated | |
2729 | ||
2730 | def fix_kv(m): | |
2731 | v = m.group(0) | |
2732 | if v in ('true', 'false', 'null'): | |
2733 | return v | |
2734 | elif v in ('undefined', 'void 0'): | |
2735 | return 'null' | |
2736 | elif v.startswith('/*') or v.startswith('//') or v.startswith('!') or v == ',': | |
2737 | return '' | |
2738 | ||
2739 | if v[0] in STRING_QUOTES: | |
2740 | v = re.sub(r'(?s)\${([^}]+)}', template_substitute, v[1:-1]) if v[0] == '`' else v[1:-1] | |
2741 | escaped = re.sub(r'(?s)(")|\\(.)', process_escape, v) | |
2742 | return f'"{escaped}"' | |
2743 | ||
2744 | for regex, base in INTEGER_TABLE: | |
2745 | im = re.match(regex, v) | |
2746 | if im: | |
2747 | i = int(im.group(1), base) | |
2748 | return f'"{i}":' if v.endswith(':') else str(i) | |
2749 | ||
2750 | if v in vars: | |
2751 | try: | |
2752 | if not strict: | |
2753 | json.loads(vars[v]) | |
2754 | except json.JSONDecodeError: | |
2755 | return json.dumps(vars[v]) | |
2756 | else: | |
2757 | return vars[v] | |
2758 | ||
2759 | if not strict: | |
2760 | return f'"{v}"' | |
2761 | ||
2762 | raise ValueError(f'Unknown value: {v}') | |
2763 | ||
2764 | def create_map(mobj): | |
2765 | return json.dumps(dict(json.loads(js_to_json(mobj.group(1) or '[]', vars=vars)))) | |
2766 | ||
2767 | code = re.sub(r'(?:new\s+)?Array\((.*?)\)', r'[\g<1>]', code) | |
2768 | code = re.sub(r'new Map\((\[.*?\])?\)', create_map, code) | |
2769 | if not strict: | |
2770 | code = re.sub(rf'new Date\(({STRING_RE})\)', r'\g<1>', code) | |
2771 | code = re.sub(r'new \w+\((.*?)\)', lambda m: json.dumps(m.group(0)), code) | |
2772 | code = re.sub(r'parseInt\([^\d]+(\d+)[^\d]+\)', r'\1', code) | |
2773 | code = re.sub(r'\(function\([^)]*\)\s*\{[^}]*\}\s*\)\s*\(\s*(["\'][^)]*["\'])\s*\)', r'\1', code) | |
2774 | ||
2775 | return re.sub(rf'''(?sx) | |
2776 | {STRING_RE}| | |
2777 | {COMMENT_RE}|,(?={SKIP_RE}[\]}}])| | |
2778 | void\s0|(?:(?<![0-9])[eE]|[a-df-zA-DF-Z_$])[.a-zA-Z_$0-9]*| | |
2779 | \b(?:0[xX][0-9a-fA-F]+|0+[0-7]+)(?:{SKIP_RE}:)?| | |
2780 | [0-9]+(?={SKIP_RE}:)| | |
2781 | !+ | |
2782 | ''', fix_kv, code) | |
2783 | ||
2784 | ||
2785 | def qualities(quality_ids): | |
2786 | """ Get a numeric quality value out of a list of possible values """ | |
2787 | def q(qid): | |
2788 | try: | |
2789 | return quality_ids.index(qid) | |
2790 | except ValueError: | |
2791 | return -1 | |
2792 | return q | |
2793 | ||
2794 | ||
2795 | POSTPROCESS_WHEN = ('pre_process', 'after_filter', 'video', 'before_dl', 'post_process', 'after_move', 'after_video', 'playlist') | |
2796 | ||
2797 | ||
2798 | DEFAULT_OUTTMPL = { | |
2799 | 'default': '%(title)s [%(id)s].%(ext)s', | |
2800 | 'chapter': '%(title)s - %(section_number)03d %(section_title)s [%(id)s].%(ext)s', | |
2801 | } | |
2802 | OUTTMPL_TYPES = { | |
2803 | 'chapter': None, | |
2804 | 'subtitle': None, | |
2805 | 'thumbnail': None, | |
2806 | 'description': 'description', | |
2807 | 'annotation': 'annotations.xml', | |
2808 | 'infojson': 'info.json', | |
2809 | 'link': None, | |
2810 | 'pl_video': None, | |
2811 | 'pl_thumbnail': None, | |
2812 | 'pl_description': 'description', | |
2813 | 'pl_infojson': 'info.json', | |
2814 | } | |
2815 | ||
2816 | # As of [1] format syntax is: | |
2817 | # %[mapping_key][conversion_flags][minimum_width][.precision][length_modifier]type | |
2818 | # 1. https://docs.python.org/2/library/stdtypes.html#string-formatting | |
2819 | STR_FORMAT_RE_TMPL = r'''(?x) | |
2820 | (?<!%)(?P<prefix>(?:%%)*) | |
2821 | % | |
2822 | (?P<has_key>\((?P<key>{0})\))? | |
2823 | (?P<format> | |
2824 | (?P<conversion>[#0\-+ ]+)? | |
2825 | (?P<min_width>\d+)? | |
2826 | (?P<precision>\.\d+)? | |
2827 | (?P<len_mod>[hlL])? # unused in python | |
2828 | {1} # conversion type | |
2829 | ) | |
2830 | ''' | |
2831 | ||
2832 | ||
2833 | STR_FORMAT_TYPES = 'diouxXeEfFgGcrsa' | |
2834 | ||
2835 | ||
2836 | def limit_length(s, length): | |
2837 | """ Add ellipses to overly long strings """ | |
2838 | if s is None: | |
2839 | return None | |
2840 | ELLIPSES = '...' | |
2841 | if len(s) > length: | |
2842 | return s[:length - len(ELLIPSES)] + ELLIPSES | |
2843 | return s | |
2844 | ||
2845 | ||
2846 | def version_tuple(v): | |
2847 | return tuple(int(e) for e in re.split(r'[-.]', v)) | |
2848 | ||
2849 | ||
2850 | def is_outdated_version(version, limit, assume_new=True): | |
2851 | if not version: | |
2852 | return not assume_new | |
2853 | try: | |
2854 | return version_tuple(version) < version_tuple(limit) | |
2855 | except ValueError: | |
2856 | return not assume_new | |
2857 | ||
2858 | ||
2859 | def ytdl_is_updateable(): | |
2860 | """ Returns if yt-dlp can be updated with -U """ | |
2861 | ||
2862 | from ..update import is_non_updateable | |
2863 | ||
2864 | return not is_non_updateable() | |
2865 | ||
2866 | ||
2867 | def args_to_str(args): | |
2868 | # Get a short string representation for a subprocess command | |
2869 | return shell_quote(args) | |
2870 | ||
2871 | ||
2872 | def error_to_str(err): | |
2873 | return f'{type(err).__name__}: {err}' | |
2874 | ||
2875 | ||
2876 | def mimetype2ext(mt, default=NO_DEFAULT): | |
2877 | if not isinstance(mt, str): | |
2878 | if default is not NO_DEFAULT: | |
2879 | return default | |
2880 | return None | |
2881 | ||
2882 | MAP = { | |
2883 | # video | |
2884 | '3gpp': '3gp', | |
2885 | 'mp2t': 'ts', | |
2886 | 'mp4': 'mp4', | |
2887 | 'mpeg': 'mpeg', | |
2888 | 'mpegurl': 'm3u8', | |
2889 | 'quicktime': 'mov', | |
2890 | 'webm': 'webm', | |
2891 | 'vp9': 'vp9', | |
2892 | 'video/ogg': 'ogv', | |
2893 | 'x-flv': 'flv', | |
2894 | 'x-m4v': 'm4v', | |
2895 | 'x-matroska': 'mkv', | |
2896 | 'x-mng': 'mng', | |
2897 | 'x-mp4-fragmented': 'mp4', | |
2898 | 'x-ms-asf': 'asf', | |
2899 | 'x-ms-wmv': 'wmv', | |
2900 | 'x-msvideo': 'avi', | |
2901 | ||
2902 | # application (streaming playlists) | |
2903 | 'dash+xml': 'mpd', | |
2904 | 'f4m+xml': 'f4m', | |
2905 | 'hds+xml': 'f4m', | |
2906 | 'vnd.apple.mpegurl': 'm3u8', | |
2907 | 'vnd.ms-sstr+xml': 'ism', | |
2908 | 'x-mpegurl': 'm3u8', | |
2909 | ||
2910 | # audio | |
2911 | 'audio/mp4': 'm4a', | |
2912 | # Per RFC 3003, audio/mpeg can be .mp1, .mp2 or .mp3. | |
2913 | # Using .mp3 as it's the most popular one | |
2914 | 'audio/mpeg': 'mp3', | |
2915 | 'audio/webm': 'webm', | |
2916 | 'audio/x-matroska': 'mka', | |
2917 | 'audio/x-mpegurl': 'm3u', | |
2918 | 'midi': 'mid', | |
2919 | 'ogg': 'ogg', | |
2920 | 'wav': 'wav', | |
2921 | 'wave': 'wav', | |
2922 | 'x-aac': 'aac', | |
2923 | 'x-flac': 'flac', | |
2924 | 'x-m4a': 'm4a', | |
2925 | 'x-realaudio': 'ra', | |
2926 | 'x-wav': 'wav', | |
2927 | ||
2928 | # image | |
2929 | 'avif': 'avif', | |
2930 | 'bmp': 'bmp', | |
2931 | 'gif': 'gif', | |
2932 | 'jpeg': 'jpg', | |
2933 | 'png': 'png', | |
2934 | 'svg+xml': 'svg', | |
2935 | 'tiff': 'tif', | |
2936 | 'vnd.wap.wbmp': 'wbmp', | |
2937 | 'webp': 'webp', | |
2938 | 'x-icon': 'ico', | |
2939 | 'x-jng': 'jng', | |
2940 | 'x-ms-bmp': 'bmp', | |
2941 | ||
2942 | # caption | |
2943 | 'filmstrip+json': 'fs', | |
2944 | 'smptett+xml': 'tt', | |
2945 | 'ttaf+xml': 'dfxp', | |
2946 | 'ttml+xml': 'ttml', | |
2947 | 'x-ms-sami': 'sami', | |
2948 | ||
2949 | # misc | |
2950 | 'gzip': 'gz', | |
2951 | 'json': 'json', | |
2952 | 'xml': 'xml', | |
2953 | 'zip': 'zip', | |
2954 | } | |
2955 | ||
2956 | mimetype = mt.partition(';')[0].strip().lower() | |
2957 | _, _, subtype = mimetype.rpartition('/') | |
2958 | ||
2959 | ext = traversal.traverse_obj(MAP, mimetype, subtype, subtype.rsplit('+')[-1]) | |
2960 | if ext: | |
2961 | return ext | |
2962 | elif default is not NO_DEFAULT: | |
2963 | return default | |
2964 | return subtype.replace('+', '.') | |
2965 | ||
2966 | ||
2967 | def ext2mimetype(ext_or_url): | |
2968 | if not ext_or_url: | |
2969 | return None | |
2970 | if '.' not in ext_or_url: | |
2971 | ext_or_url = f'file.{ext_or_url}' | |
2972 | return mimetypes.guess_type(ext_or_url)[0] | |
2973 | ||
2974 | ||
2975 | def parse_codecs(codecs_str): | |
2976 | # http://tools.ietf.org/html/rfc6381 | |
2977 | if not codecs_str: | |
2978 | return {} | |
2979 | split_codecs = list(filter(None, map( | |
2980 | str.strip, codecs_str.strip().strip(',').split(',')))) | |
2981 | vcodec, acodec, scodec, hdr = None, None, None, None | |
2982 | for full_codec in split_codecs: | |
2983 | parts = re.sub(r'0+(?=\d)', '', full_codec).split('.') | |
2984 | if parts[0] in ('avc1', 'avc2', 'avc3', 'avc4', 'vp9', 'vp8', 'hev1', 'hev2', | |
2985 | 'h263', 'h264', 'mp4v', 'hvc1', 'av1', 'theora', 'dvh1', 'dvhe'): | |
2986 | if vcodec: | |
2987 | continue | |
2988 | vcodec = full_codec | |
2989 | if parts[0] in ('dvh1', 'dvhe'): | |
2990 | hdr = 'DV' | |
2991 | elif parts[0] == 'av1' and traversal.traverse_obj(parts, 3) == '10': | |
2992 | hdr = 'HDR10' | |
2993 | elif parts[:2] == ['vp9', '2']: | |
2994 | hdr = 'HDR10' | |
2995 | elif parts[0] in ('flac', 'mp4a', 'opus', 'vorbis', 'mp3', 'aac', 'ac-4', | |
2996 | 'ac-3', 'ec-3', 'eac3', 'dtsc', 'dtse', 'dtsh', 'dtsl'): | |
2997 | acodec = acodec or full_codec | |
2998 | elif parts[0] in ('stpp', 'wvtt'): | |
2999 | scodec = scodec or full_codec | |
3000 | else: | |
3001 | write_string(f'WARNING: Unknown codec {full_codec}\n') | |
3002 | if vcodec or acodec or scodec: | |
3003 | return { | |
3004 | 'vcodec': vcodec or 'none', | |
3005 | 'acodec': acodec or 'none', | |
3006 | 'dynamic_range': hdr, | |
3007 | **({'scodec': scodec} if scodec is not None else {}), | |
3008 | } | |
3009 | elif len(split_codecs) == 2: | |
3010 | return { | |
3011 | 'vcodec': split_codecs[0], | |
3012 | 'acodec': split_codecs[1], | |
3013 | } | |
3014 | return {} | |
3015 | ||
3016 | ||
3017 | def get_compatible_ext(*, vcodecs, acodecs, vexts, aexts, preferences=None): | |
3018 | assert len(vcodecs) == len(vexts) and len(acodecs) == len(aexts) | |
3019 | ||
3020 | allow_mkv = not preferences or 'mkv' in preferences | |
3021 | ||
3022 | if allow_mkv and max(len(acodecs), len(vcodecs)) > 1: | |
3023 | return 'mkv' # TODO: any other format allows this? | |
3024 | ||
3025 | # TODO: All codecs supported by parse_codecs isn't handled here | |
3026 | COMPATIBLE_CODECS = { | |
3027 | 'mp4': { | |
3028 | 'av1', 'hevc', 'avc1', 'mp4a', 'ac-4', # fourcc (m3u8, mpd) | |
3029 | 'h264', 'aacl', 'ec-3', # Set in ISM | |
3030 | }, | |
3031 | 'webm': { | |
3032 | 'av1', 'vp9', 'vp8', 'opus', 'vrbs', | |
3033 | 'vp9x', 'vp8x', # in the webm spec | |
3034 | }, | |
3035 | } | |
3036 | ||
3037 | sanitize_codec = functools.partial( | |
3038 | try_get, getter=lambda x: x[0].split('.')[0].replace('0', '').lower()) | |
3039 | vcodec, acodec = sanitize_codec(vcodecs), sanitize_codec(acodecs) | |
3040 | ||
3041 | for ext in preferences or COMPATIBLE_CODECS.keys(): | |
3042 | codec_set = COMPATIBLE_CODECS.get(ext, set()) | |
3043 | if ext == 'mkv' or codec_set.issuperset((vcodec, acodec)): | |
3044 | return ext | |
3045 | ||
3046 | COMPATIBLE_EXTS = ( | |
3047 | {'mp3', 'mp4', 'm4a', 'm4p', 'm4b', 'm4r', 'm4v', 'ismv', 'isma', 'mov'}, | |
3048 | {'webm', 'weba'}, | |
3049 | ) | |
3050 | for ext in preferences or vexts: | |
3051 | current_exts = {ext, *vexts, *aexts} | |
3052 | if ext == 'mkv' or current_exts == {ext} or any( | |
3053 | ext_sets.issuperset(current_exts) for ext_sets in COMPATIBLE_EXTS): | |
3054 | return ext | |
3055 | return 'mkv' if allow_mkv else preferences[-1] | |
3056 | ||
3057 | ||
3058 | def urlhandle_detect_ext(url_handle, default=NO_DEFAULT): | |
3059 | getheader = url_handle.headers.get | |
3060 | ||
3061 | cd = getheader('Content-Disposition') | |
3062 | if cd: | |
3063 | m = re.match(r'attachment;\s*filename="(?P<filename>[^"]+)"', cd) | |
3064 | if m: | |
3065 | e = determine_ext(m.group('filename'), default_ext=None) | |
3066 | if e: | |
3067 | return e | |
3068 | ||
3069 | meta_ext = getheader('x-amz-meta-name') | |
3070 | if meta_ext: | |
3071 | e = meta_ext.rpartition('.')[2] | |
3072 | if e: | |
3073 | return e | |
3074 | ||
3075 | return mimetype2ext(getheader('Content-Type'), default=default) | |
3076 | ||
3077 | ||
3078 | def encode_data_uri(data, mime_type): | |
3079 | return 'data:%s;base64,%s' % (mime_type, base64.b64encode(data).decode('ascii')) | |
3080 | ||
3081 | ||
3082 | def age_restricted(content_limit, age_limit): | |
3083 | """ Returns True iff the content should be blocked """ | |
3084 | ||
3085 | if age_limit is None: # No limit set | |
3086 | return False | |
3087 | if content_limit is None: | |
3088 | return False # Content available for everyone | |
3089 | return age_limit < content_limit | |
3090 | ||
3091 | ||
3092 | # List of known byte-order-marks (BOM) | |
3093 | BOMS = [ | |
3094 | (b'\xef\xbb\xbf', 'utf-8'), | |
3095 | (b'\x00\x00\xfe\xff', 'utf-32-be'), | |
3096 | (b'\xff\xfe\x00\x00', 'utf-32-le'), | |
3097 | (b'\xff\xfe', 'utf-16-le'), | |
3098 | (b'\xfe\xff', 'utf-16-be'), | |
3099 | ] | |
3100 | ||
3101 | ||
3102 | def is_html(first_bytes): | |
3103 | """ Detect whether a file contains HTML by examining its first bytes. """ | |
3104 | ||
3105 | encoding = 'utf-8' | |
3106 | for bom, enc in BOMS: | |
3107 | while first_bytes.startswith(bom): | |
3108 | encoding, first_bytes = enc, first_bytes[len(bom):] | |
3109 | ||
3110 | return re.match(r'^\s*<', first_bytes.decode(encoding, 'replace')) | |
3111 | ||
3112 | ||
3113 | def determine_protocol(info_dict): | |
3114 | protocol = info_dict.get('protocol') | |
3115 | if protocol is not None: | |
3116 | return protocol | |
3117 | ||
3118 | url = sanitize_url(info_dict['url']) | |
3119 | if url.startswith('rtmp'): | |
3120 | return 'rtmp' | |
3121 | elif url.startswith('mms'): | |
3122 | return 'mms' | |
3123 | elif url.startswith('rtsp'): | |
3124 | return 'rtsp' | |
3125 | ||
3126 | ext = determine_ext(url) | |
3127 | if ext == 'm3u8': | |
3128 | return 'm3u8' if info_dict.get('is_live') else 'm3u8_native' | |
3129 | elif ext == 'f4m': | |
3130 | return 'f4m' | |
3131 | ||
3132 | return urllib.parse.urlparse(url).scheme | |
3133 | ||
3134 | ||
3135 | def render_table(header_row, data, delim=False, extra_gap=0, hide_empty=False): | |
3136 | """ Render a list of rows, each as a list of values. | |
3137 | Text after a \t will be right aligned """ | |
3138 | def width(string): | |
3139 | return len(remove_terminal_sequences(string).replace('\t', '')) | |
3140 | ||
3141 | def get_max_lens(table): | |
3142 | return [max(width(str(v)) for v in col) for col in zip(*table)] | |
3143 | ||
3144 | def filter_using_list(row, filterArray): | |
3145 | return [col for take, col in itertools.zip_longest(filterArray, row, fillvalue=True) if take] | |
3146 | ||
3147 | max_lens = get_max_lens(data) if hide_empty else [] | |
3148 | header_row = filter_using_list(header_row, max_lens) | |
3149 | data = [filter_using_list(row, max_lens) for row in data] | |
3150 | ||
3151 | table = [header_row] + data | |
3152 | max_lens = get_max_lens(table) | |
3153 | extra_gap += 1 | |
3154 | if delim: | |
3155 | table = [header_row, [delim * (ml + extra_gap) for ml in max_lens]] + data | |
3156 | table[1][-1] = table[1][-1][:-extra_gap * len(delim)] # Remove extra_gap from end of delimiter | |
3157 | for row in table: | |
3158 | for pos, text in enumerate(map(str, row)): | |
3159 | if '\t' in text: | |
3160 | row[pos] = text.replace('\t', ' ' * (max_lens[pos] - width(text))) + ' ' * extra_gap | |
3161 | else: | |
3162 | row[pos] = text + ' ' * (max_lens[pos] - width(text) + extra_gap) | |
3163 | ret = '\n'.join(''.join(row).rstrip() for row in table) | |
3164 | return ret | |
3165 | ||
3166 | ||
3167 | def _match_one(filter_part, dct, incomplete): | |
3168 | # TODO: Generalize code with YoutubeDL._build_format_filter | |
3169 | STRING_OPERATORS = { | |
3170 | '*=': operator.contains, | |
3171 | '^=': lambda attr, value: attr.startswith(value), | |
3172 | '$=': lambda attr, value: attr.endswith(value), | |
3173 | '~=': lambda attr, value: re.search(value, attr), | |
3174 | } | |
3175 | COMPARISON_OPERATORS = { | |
3176 | **STRING_OPERATORS, | |
3177 | '<=': operator.le, # "<=" must be defined above "<" | |
3178 | '<': operator.lt, | |
3179 | '>=': operator.ge, | |
3180 | '>': operator.gt, | |
3181 | '=': operator.eq, | |
3182 | } | |
3183 | ||
3184 | if isinstance(incomplete, bool): | |
3185 | is_incomplete = lambda _: incomplete | |
3186 | else: | |
3187 | is_incomplete = lambda k: k in incomplete | |
3188 | ||
3189 | operator_rex = re.compile(r'''(?x) | |
3190 | (?P<key>[a-z_]+) | |
3191 | \s*(?P<negation>!\s*)?(?P<op>%s)(?P<none_inclusive>\s*\?)?\s* | |
3192 | (?: | |
3193 | (?P<quote>["\'])(?P<quotedstrval>.+?)(?P=quote)| | |
3194 | (?P<strval>.+?) | |
3195 | ) | |
3196 | ''' % '|'.join(map(re.escape, COMPARISON_OPERATORS.keys()))) | |
3197 | m = operator_rex.fullmatch(filter_part.strip()) | |
3198 | if m: | |
3199 | m = m.groupdict() | |
3200 | unnegated_op = COMPARISON_OPERATORS[m['op']] | |
3201 | if m['negation']: | |
3202 | op = lambda attr, value: not unnegated_op(attr, value) | |
3203 | else: | |
3204 | op = unnegated_op | |
3205 | comparison_value = m['quotedstrval'] or m['strval'] or m['intval'] | |
3206 | if m['quote']: | |
3207 | comparison_value = comparison_value.replace(r'\%s' % m['quote'], m['quote']) | |
3208 | actual_value = dct.get(m['key']) | |
3209 | numeric_comparison = None | |
3210 | if isinstance(actual_value, (int, float)): | |
3211 | # If the original field is a string and matching comparisonvalue is | |
3212 | # a number we should respect the origin of the original field | |
3213 | # and process comparison value as a string (see | |
3214 | # https://github.com/ytdl-org/youtube-dl/issues/11082) | |
3215 | try: | |
3216 | numeric_comparison = int(comparison_value) | |
3217 | except ValueError: | |
3218 | numeric_comparison = parse_filesize(comparison_value) | |
3219 | if numeric_comparison is None: | |
3220 | numeric_comparison = parse_filesize(f'{comparison_value}B') | |
3221 | if numeric_comparison is None: | |
3222 | numeric_comparison = parse_duration(comparison_value) | |
3223 | if numeric_comparison is not None and m['op'] in STRING_OPERATORS: | |
3224 | raise ValueError('Operator %s only supports string values!' % m['op']) | |
3225 | if actual_value is None: | |
3226 | return is_incomplete(m['key']) or m['none_inclusive'] | |
3227 | return op(actual_value, comparison_value if numeric_comparison is None else numeric_comparison) | |
3228 | ||
3229 | UNARY_OPERATORS = { | |
3230 | '': lambda v: (v is True) if isinstance(v, bool) else (v is not None), | |
3231 | '!': lambda v: (v is False) if isinstance(v, bool) else (v is None), | |
3232 | } | |
3233 | operator_rex = re.compile(r'''(?x) | |
3234 | (?P<op>%s)\s*(?P<key>[a-z_]+) | |
3235 | ''' % '|'.join(map(re.escape, UNARY_OPERATORS.keys()))) | |
3236 | m = operator_rex.fullmatch(filter_part.strip()) | |
3237 | if m: | |
3238 | op = UNARY_OPERATORS[m.group('op')] | |
3239 | actual_value = dct.get(m.group('key')) | |
3240 | if is_incomplete(m.group('key')) and actual_value is None: | |
3241 | return True | |
3242 | return op(actual_value) | |
3243 | ||
3244 | raise ValueError('Invalid filter part %r' % filter_part) | |
3245 | ||
3246 | ||
3247 | def match_str(filter_str, dct, incomplete=False): | |
3248 | """ Filter a dictionary with a simple string syntax. | |
3249 | @returns Whether the filter passes | |
3250 | @param incomplete Set of keys that is expected to be missing from dct. | |
3251 | Can be True/False to indicate all/none of the keys may be missing. | |
3252 | All conditions on incomplete keys pass if the key is missing | |
3253 | """ | |
3254 | return all( | |
3255 | _match_one(filter_part.replace(r'\&', '&'), dct, incomplete) | |
3256 | for filter_part in re.split(r'(?<!\\)&', filter_str)) | |
3257 | ||
3258 | ||
3259 | def match_filter_func(filters, breaking_filters=None): | |
3260 | if not filters and not breaking_filters: | |
3261 | return None | |
3262 | repr_ = f'{match_filter_func.__module__}.{match_filter_func.__qualname__}({filters}, {breaking_filters})' | |
3263 | ||
3264 | breaking_filters = match_filter_func(breaking_filters) or (lambda _, __: None) | |
3265 | filters = set(variadic(filters or [])) | |
3266 | ||
3267 | interactive = '-' in filters | |
3268 | if interactive: | |
3269 | filters.remove('-') | |
3270 | ||
3271 | @function_with_repr.set_repr(repr_) | |
3272 | def _match_func(info_dict, incomplete=False): | |
3273 | ret = breaking_filters(info_dict, incomplete) | |
3274 | if ret is not None: | |
3275 | raise RejectedVideoReached(ret) | |
3276 | ||
3277 | if not filters or any(match_str(f, info_dict, incomplete) for f in filters): | |
3278 | return NO_DEFAULT if interactive and not incomplete else None | |
3279 | else: | |
3280 | video_title = info_dict.get('title') or info_dict.get('id') or 'entry' | |
3281 | filter_str = ') | ('.join(map(str.strip, filters)) | |
3282 | return f'{video_title} does not pass filter ({filter_str}), skipping ..' | |
3283 | return _match_func | |
3284 | ||
3285 | ||
3286 | class download_range_func: | |
3287 | def __init__(self, chapters, ranges, from_info=False): | |
3288 | self.chapters, self.ranges, self.from_info = chapters, ranges, from_info | |
3289 | ||
3290 | def __call__(self, info_dict, ydl): | |
3291 | ||
3292 | warning = ('There are no chapters matching the regex' if info_dict.get('chapters') | |
3293 | else 'Cannot match chapters since chapter information is unavailable') | |
3294 | for regex in self.chapters or []: | |
3295 | for i, chapter in enumerate(info_dict.get('chapters') or []): | |
3296 | if re.search(regex, chapter['title']): | |
3297 | warning = None | |
3298 | yield {**chapter, 'index': i} | |
3299 | if self.chapters and warning: | |
3300 | ydl.to_screen(f'[info] {info_dict["id"]}: {warning}') | |
3301 | ||
3302 | for start, end in self.ranges or []: | |
3303 | yield { | |
3304 | 'start_time': self._handle_negative_timestamp(start, info_dict), | |
3305 | 'end_time': self._handle_negative_timestamp(end, info_dict), | |
3306 | } | |
3307 | ||
3308 | if self.from_info and (info_dict.get('start_time') or info_dict.get('end_time')): | |
3309 | yield { | |
3310 | 'start_time': info_dict.get('start_time') or 0, | |
3311 | 'end_time': info_dict.get('end_time') or float('inf'), | |
3312 | } | |
3313 | elif not self.ranges and not self.chapters: | |
3314 | yield {} | |
3315 | ||
3316 | @staticmethod | |
3317 | def _handle_negative_timestamp(time, info): | |
3318 | return max(info['duration'] + time, 0) if info.get('duration') and time < 0 else time | |
3319 | ||
3320 | def __eq__(self, other): | |
3321 | return (isinstance(other, download_range_func) | |
3322 | and self.chapters == other.chapters and self.ranges == other.ranges) | |
3323 | ||
3324 | def __repr__(self): | |
3325 | return f'{__name__}.{type(self).__name__}({self.chapters}, {self.ranges})' | |
3326 | ||
3327 | ||
3328 | def parse_dfxp_time_expr(time_expr): | |
3329 | if not time_expr: | |
3330 | return | |
3331 | ||
3332 | mobj = re.match(rf'^(?P<time_offset>{NUMBER_RE})s?$', time_expr) | |
3333 | if mobj: | |
3334 | return float(mobj.group('time_offset')) | |
3335 | ||
3336 | mobj = re.match(r'^(\d+):(\d\d):(\d\d(?:(?:\.|:)\d+)?)$', time_expr) | |
3337 | if mobj: | |
3338 | return 3600 * int(mobj.group(1)) + 60 * int(mobj.group(2)) + float(mobj.group(3).replace(':', '.')) | |
3339 | ||
3340 | ||
3341 | def srt_subtitles_timecode(seconds): | |
3342 | return '%02d:%02d:%02d,%03d' % timetuple_from_msec(seconds * 1000) | |
3343 | ||
3344 | ||
3345 | def ass_subtitles_timecode(seconds): | |
3346 | time = timetuple_from_msec(seconds * 1000) | |
3347 | return '%01d:%02d:%02d.%02d' % (*time[:-1], time.milliseconds / 10) | |
3348 | ||
3349 | ||
3350 | def dfxp2srt(dfxp_data): | |
3351 | ''' | |
3352 | @param dfxp_data A bytes-like object containing DFXP data | |
3353 | @returns A unicode object containing converted SRT data | |
3354 | ''' | |
3355 | LEGACY_NAMESPACES = ( | |
3356 | (b'http://www.w3.org/ns/ttml', [ | |
3357 | b'http://www.w3.org/2004/11/ttaf1', | |
3358 | b'http://www.w3.org/2006/04/ttaf1', | |
3359 | b'http://www.w3.org/2006/10/ttaf1', | |
3360 | ]), | |
3361 | (b'http://www.w3.org/ns/ttml#styling', [ | |
3362 | b'http://www.w3.org/ns/ttml#style', | |
3363 | ]), | |
3364 | ) | |
3365 | ||
3366 | SUPPORTED_STYLING = [ | |
3367 | 'color', | |
3368 | 'fontFamily', | |
3369 | 'fontSize', | |
3370 | 'fontStyle', | |
3371 | 'fontWeight', | |
3372 | 'textDecoration' | |
3373 | ] | |
3374 | ||
3375 | _x = functools.partial(xpath_with_ns, ns_map={ | |
3376 | 'xml': 'http://www.w3.org/XML/1998/namespace', | |
3377 | 'ttml': 'http://www.w3.org/ns/ttml', | |
3378 | 'tts': 'http://www.w3.org/ns/ttml#styling', | |
3379 | }) | |
3380 | ||
3381 | styles = {} | |
3382 | default_style = {} | |
3383 | ||
3384 | class TTMLPElementParser: | |
3385 | _out = '' | |
3386 | _unclosed_elements = [] | |
3387 | _applied_styles = [] | |
3388 | ||
3389 | def start(self, tag, attrib): | |
3390 | if tag in (_x('ttml:br'), 'br'): | |
3391 | self._out += '\n' | |
3392 | else: | |
3393 | unclosed_elements = [] | |
3394 | style = {} | |
3395 | element_style_id = attrib.get('style') | |
3396 | if default_style: | |
3397 | style.update(default_style) | |
3398 | if element_style_id: | |
3399 | style.update(styles.get(element_style_id, {})) | |
3400 | for prop in SUPPORTED_STYLING: | |
3401 | prop_val = attrib.get(_x('tts:' + prop)) | |
3402 | if prop_val: | |
3403 | style[prop] = prop_val | |
3404 | if style: | |
3405 | font = '' | |
3406 | for k, v in sorted(style.items()): | |
3407 | if self._applied_styles and self._applied_styles[-1].get(k) == v: | |
3408 | continue | |
3409 | if k == 'color': | |
3410 | font += ' color="%s"' % v | |
3411 | elif k == 'fontSize': | |
3412 | font += ' size="%s"' % v | |
3413 | elif k == 'fontFamily': | |
3414 | font += ' face="%s"' % v | |
3415 | elif k == 'fontWeight' and v == 'bold': | |
3416 | self._out += '<b>' | |
3417 | unclosed_elements.append('b') | |
3418 | elif k == 'fontStyle' and v == 'italic': | |
3419 | self._out += '<i>' | |
3420 | unclosed_elements.append('i') | |
3421 | elif k == 'textDecoration' and v == 'underline': | |
3422 | self._out += '<u>' | |
3423 | unclosed_elements.append('u') | |
3424 | if font: | |
3425 | self._out += '<font' + font + '>' | |
3426 | unclosed_elements.append('font') | |
3427 | applied_style = {} | |
3428 | if self._applied_styles: | |
3429 | applied_style.update(self._applied_styles[-1]) | |
3430 | applied_style.update(style) | |
3431 | self._applied_styles.append(applied_style) | |
3432 | self._unclosed_elements.append(unclosed_elements) | |
3433 | ||
3434 | def end(self, tag): | |
3435 | if tag not in (_x('ttml:br'), 'br'): | |
3436 | unclosed_elements = self._unclosed_elements.pop() | |
3437 | for element in reversed(unclosed_elements): | |
3438 | self._out += '</%s>' % element | |
3439 | if unclosed_elements and self._applied_styles: | |
3440 | self._applied_styles.pop() | |
3441 | ||
3442 | def data(self, data): | |
3443 | self._out += data | |
3444 | ||
3445 | def close(self): | |
3446 | return self._out.strip() | |
3447 | ||
3448 | # Fix UTF-8 encoded file wrongly marked as UTF-16. See https://github.com/yt-dlp/yt-dlp/issues/6543#issuecomment-1477169870 | |
3449 | # This will not trigger false positives since only UTF-8 text is being replaced | |
3450 | dfxp_data = dfxp_data.replace(b'encoding=\'UTF-16\'', b'encoding=\'UTF-8\'') | |
3451 | ||
3452 | def parse_node(node): | |
3453 | target = TTMLPElementParser() | |
3454 | parser = xml.etree.ElementTree.XMLParser(target=target) | |
3455 | parser.feed(xml.etree.ElementTree.tostring(node)) | |
3456 | return parser.close() | |
3457 | ||
3458 | for k, v in LEGACY_NAMESPACES: | |
3459 | for ns in v: | |
3460 | dfxp_data = dfxp_data.replace(ns, k) | |
3461 | ||
3462 | dfxp = compat_etree_fromstring(dfxp_data) | |
3463 | out = [] | |
3464 | paras = dfxp.findall(_x('.//ttml:p')) or dfxp.findall('.//p') | |
3465 | ||
3466 | if not paras: | |
3467 | raise ValueError('Invalid dfxp/TTML subtitle') | |
3468 | ||
3469 | repeat = False | |
3470 | while True: | |
3471 | for style in dfxp.findall(_x('.//ttml:style')): | |
3472 | style_id = style.get('id') or style.get(_x('xml:id')) | |
3473 | if not style_id: | |
3474 | continue | |
3475 | parent_style_id = style.get('style') | |
3476 | if parent_style_id: | |
3477 | if parent_style_id not in styles: | |
3478 | repeat = True | |
3479 | continue | |
3480 | styles[style_id] = styles[parent_style_id].copy() | |
3481 | for prop in SUPPORTED_STYLING: | |
3482 | prop_val = style.get(_x('tts:' + prop)) | |
3483 | if prop_val: | |
3484 | styles.setdefault(style_id, {})[prop] = prop_val | |
3485 | if repeat: | |
3486 | repeat = False | |
3487 | else: | |
3488 | break | |
3489 | ||
3490 | for p in ('body', 'div'): | |
3491 | ele = xpath_element(dfxp, [_x('.//ttml:' + p), './/' + p]) | |
3492 | if ele is None: | |
3493 | continue | |
3494 | style = styles.get(ele.get('style')) | |
3495 | if not style: | |
3496 | continue | |
3497 | default_style.update(style) | |
3498 | ||
3499 | for para, index in zip(paras, itertools.count(1)): | |
3500 | begin_time = parse_dfxp_time_expr(para.attrib.get('begin')) | |
3501 | end_time = parse_dfxp_time_expr(para.attrib.get('end')) | |
3502 | dur = parse_dfxp_time_expr(para.attrib.get('dur')) | |
3503 | if begin_time is None: | |
3504 | continue | |
3505 | if not end_time: | |
3506 | if not dur: | |
3507 | continue | |
3508 | end_time = begin_time + dur | |
3509 | out.append('%d\n%s --> %s\n%s\n\n' % ( | |
3510 | index, | |
3511 | srt_subtitles_timecode(begin_time), | |
3512 | srt_subtitles_timecode(end_time), | |
3513 | parse_node(para))) | |
3514 | ||
3515 | return ''.join(out) | |
3516 | ||
3517 | ||
3518 | def cli_option(params, command_option, param, separator=None): | |
3519 | param = params.get(param) | |
3520 | return ([] if param is None | |
3521 | else [command_option, str(param)] if separator is None | |
3522 | else [f'{command_option}{separator}{param}']) | |
3523 | ||
3524 | ||
3525 | def cli_bool_option(params, command_option, param, true_value='true', false_value='false', separator=None): | |
3526 | param = params.get(param) | |
3527 | assert param in (True, False, None) | |
3528 | return cli_option({True: true_value, False: false_value}, command_option, param, separator) | |
3529 | ||
3530 | ||
3531 | def cli_valueless_option(params, command_option, param, expected_value=True): | |
3532 | return [command_option] if params.get(param) == expected_value else [] | |
3533 | ||
3534 | ||
3535 | def cli_configuration_args(argdict, keys, default=[], use_compat=True): | |
3536 | if isinstance(argdict, (list, tuple)): # for backward compatibility | |
3537 | if use_compat: | |
3538 | return argdict | |
3539 | else: | |
3540 | argdict = None | |
3541 | if argdict is None: | |
3542 | return default | |
3543 | assert isinstance(argdict, dict) | |
3544 | ||
3545 | assert isinstance(keys, (list, tuple)) | |
3546 | for key_list in keys: | |
3547 | arg_list = list(filter( | |
3548 | lambda x: x is not None, | |
3549 | [argdict.get(key.lower()) for key in variadic(key_list)])) | |
3550 | if arg_list: | |
3551 | return [arg for args in arg_list for arg in args] | |
3552 | return default | |
3553 | ||
3554 | ||
3555 | def _configuration_args(main_key, argdict, exe, keys=None, default=[], use_compat=True): | |
3556 | main_key, exe = main_key.lower(), exe.lower() | |
3557 | root_key = exe if main_key == exe else f'{main_key}+{exe}' | |
3558 | keys = [f'{root_key}{k}' for k in (keys or [''])] | |
3559 | if root_key in keys: | |
3560 | if main_key != exe: | |
3561 | keys.append((main_key, exe)) | |
3562 | keys.append('default') | |
3563 | else: | |
3564 | use_compat = False | |
3565 | return cli_configuration_args(argdict, keys, default, use_compat) | |
3566 | ||
3567 | ||
3568 | class ISO639Utils: | |
3569 | # See http://www.loc.gov/standards/iso639-2/ISO-639-2_utf-8.txt | |
3570 | _lang_map = { | |
3571 | 'aa': 'aar', | |
3572 | 'ab': 'abk', | |
3573 | 'ae': 'ave', | |
3574 | 'af': 'afr', | |
3575 | 'ak': 'aka', | |
3576 | 'am': 'amh', | |
3577 | 'an': 'arg', | |
3578 | 'ar': 'ara', | |
3579 | 'as': 'asm', | |
3580 | 'av': 'ava', | |
3581 | 'ay': 'aym', | |
3582 | 'az': 'aze', | |
3583 | 'ba': 'bak', | |
3584 | 'be': 'bel', | |
3585 | 'bg': 'bul', | |
3586 | 'bh': 'bih', | |
3587 | 'bi': 'bis', | |
3588 | 'bm': 'bam', | |
3589 | 'bn': 'ben', | |
3590 | 'bo': 'bod', | |
3591 | 'br': 'bre', | |
3592 | 'bs': 'bos', | |
3593 | 'ca': 'cat', | |
3594 | 'ce': 'che', | |
3595 | 'ch': 'cha', | |
3596 | 'co': 'cos', | |
3597 | 'cr': 'cre', | |
3598 | 'cs': 'ces', | |
3599 | 'cu': 'chu', | |
3600 | 'cv': 'chv', | |
3601 | 'cy': 'cym', | |
3602 | 'da': 'dan', | |
3603 | 'de': 'deu', | |
3604 | 'dv': 'div', | |
3605 | 'dz': 'dzo', | |
3606 | 'ee': 'ewe', | |
3607 | 'el': 'ell', | |
3608 | 'en': 'eng', | |
3609 | 'eo': 'epo', | |
3610 | 'es': 'spa', | |
3611 | 'et': 'est', | |
3612 | 'eu': 'eus', | |
3613 | 'fa': 'fas', | |
3614 | 'ff': 'ful', | |
3615 | 'fi': 'fin', | |
3616 | 'fj': 'fij', | |
3617 | 'fo': 'fao', | |
3618 | 'fr': 'fra', | |
3619 | 'fy': 'fry', | |
3620 | 'ga': 'gle', | |
3621 | 'gd': 'gla', | |
3622 | 'gl': 'glg', | |
3623 | 'gn': 'grn', | |
3624 | 'gu': 'guj', | |
3625 | 'gv': 'glv', | |
3626 | 'ha': 'hau', | |
3627 | 'he': 'heb', | |
3628 | 'iw': 'heb', # Replaced by he in 1989 revision | |
3629 | 'hi': 'hin', | |
3630 | 'ho': 'hmo', | |
3631 | 'hr': 'hrv', | |
3632 | 'ht': 'hat', | |
3633 | 'hu': 'hun', | |
3634 | 'hy': 'hye', | |
3635 | 'hz': 'her', | |
3636 | 'ia': 'ina', | |
3637 | 'id': 'ind', | |
3638 | 'in': 'ind', # Replaced by id in 1989 revision | |
3639 | 'ie': 'ile', | |
3640 | 'ig': 'ibo', | |
3641 | 'ii': 'iii', | |
3642 | 'ik': 'ipk', | |
3643 | 'io': 'ido', | |
3644 | 'is': 'isl', | |
3645 | 'it': 'ita', | |
3646 | 'iu': 'iku', | |
3647 | 'ja': 'jpn', | |
3648 | 'jv': 'jav', | |
3649 | 'ka': 'kat', | |
3650 | 'kg': 'kon', | |
3651 | 'ki': 'kik', | |
3652 | 'kj': 'kua', | |
3653 | 'kk': 'kaz', | |
3654 | 'kl': 'kal', | |
3655 | 'km': 'khm', | |
3656 | 'kn': 'kan', | |
3657 | 'ko': 'kor', | |
3658 | 'kr': 'kau', | |
3659 | 'ks': 'kas', | |
3660 | 'ku': 'kur', | |
3661 | 'kv': 'kom', | |
3662 | 'kw': 'cor', | |
3663 | 'ky': 'kir', | |
3664 | 'la': 'lat', | |
3665 | 'lb': 'ltz', | |
3666 | 'lg': 'lug', | |
3667 | 'li': 'lim', | |
3668 | 'ln': 'lin', | |
3669 | 'lo': 'lao', | |
3670 | 'lt': 'lit', | |
3671 | 'lu': 'lub', | |
3672 | 'lv': 'lav', | |
3673 | 'mg': 'mlg', | |
3674 | 'mh': 'mah', | |
3675 | 'mi': 'mri', | |
3676 | 'mk': 'mkd', | |
3677 | 'ml': 'mal', | |
3678 | 'mn': 'mon', | |
3679 | 'mr': 'mar', | |
3680 | 'ms': 'msa', | |
3681 | 'mt': 'mlt', | |
3682 | 'my': 'mya', | |
3683 | 'na': 'nau', | |
3684 | 'nb': 'nob', | |
3685 | 'nd': 'nde', | |
3686 | 'ne': 'nep', | |
3687 | 'ng': 'ndo', | |
3688 | 'nl': 'nld', | |
3689 | 'nn': 'nno', | |
3690 | 'no': 'nor', | |
3691 | 'nr': 'nbl', | |
3692 | 'nv': 'nav', | |
3693 | 'ny': 'nya', | |
3694 | 'oc': 'oci', | |
3695 | 'oj': 'oji', | |
3696 | 'om': 'orm', | |
3697 | 'or': 'ori', | |
3698 | 'os': 'oss', | |
3699 | 'pa': 'pan', | |
3700 | 'pe': 'per', | |
3701 | 'pi': 'pli', | |
3702 | 'pl': 'pol', | |
3703 | 'ps': 'pus', | |
3704 | 'pt': 'por', | |
3705 | 'qu': 'que', | |
3706 | 'rm': 'roh', | |
3707 | 'rn': 'run', | |
3708 | 'ro': 'ron', | |
3709 | 'ru': 'rus', | |
3710 | 'rw': 'kin', | |
3711 | 'sa': 'san', | |
3712 | 'sc': 'srd', | |
3713 | 'sd': 'snd', | |
3714 | 'se': 'sme', | |
3715 | 'sg': 'sag', | |
3716 | 'si': 'sin', | |
3717 | 'sk': 'slk', | |
3718 | 'sl': 'slv', | |
3719 | 'sm': 'smo', | |
3720 | 'sn': 'sna', | |
3721 | 'so': 'som', | |
3722 | 'sq': 'sqi', | |
3723 | 'sr': 'srp', | |
3724 | 'ss': 'ssw', | |
3725 | 'st': 'sot', | |
3726 | 'su': 'sun', | |
3727 | 'sv': 'swe', | |
3728 | 'sw': 'swa', | |
3729 | 'ta': 'tam', | |
3730 | 'te': 'tel', | |
3731 | 'tg': 'tgk', | |
3732 | 'th': 'tha', | |
3733 | 'ti': 'tir', | |
3734 | 'tk': 'tuk', | |
3735 | 'tl': 'tgl', | |
3736 | 'tn': 'tsn', | |
3737 | 'to': 'ton', | |
3738 | 'tr': 'tur', | |
3739 | 'ts': 'tso', | |
3740 | 'tt': 'tat', | |
3741 | 'tw': 'twi', | |
3742 | 'ty': 'tah', | |
3743 | 'ug': 'uig', | |
3744 | 'uk': 'ukr', | |
3745 | 'ur': 'urd', | |
3746 | 'uz': 'uzb', | |
3747 | 've': 'ven', | |
3748 | 'vi': 'vie', | |
3749 | 'vo': 'vol', | |
3750 | 'wa': 'wln', | |
3751 | 'wo': 'wol', | |
3752 | 'xh': 'xho', | |
3753 | 'yi': 'yid', | |
3754 | 'ji': 'yid', # Replaced by yi in 1989 revision | |
3755 | 'yo': 'yor', | |
3756 | 'za': 'zha', | |
3757 | 'zh': 'zho', | |
3758 | 'zu': 'zul', | |
3759 | } | |
3760 | ||
3761 | @classmethod | |
3762 | def short2long(cls, code): | |
3763 | """Convert language code from ISO 639-1 to ISO 639-2/T""" | |
3764 | return cls._lang_map.get(code[:2]) | |
3765 | ||
3766 | @classmethod | |
3767 | def long2short(cls, code): | |
3768 | """Convert language code from ISO 639-2/T to ISO 639-1""" | |
3769 | for short_name, long_name in cls._lang_map.items(): | |
3770 | if long_name == code: | |
3771 | return short_name | |
3772 | ||
3773 | ||
3774 | class ISO3166Utils: | |
3775 | # From http://data.okfn.org/data/core/country-list | |
3776 | _country_map = { | |
3777 | 'AF': 'Afghanistan', | |
3778 | 'AX': 'Åland Islands', | |
3779 | 'AL': 'Albania', | |
3780 | 'DZ': 'Algeria', | |
3781 | 'AS': 'American Samoa', | |
3782 | 'AD': 'Andorra', | |
3783 | 'AO': 'Angola', | |
3784 | 'AI': 'Anguilla', | |
3785 | 'AQ': 'Antarctica', | |
3786 | 'AG': 'Antigua and Barbuda', | |
3787 | 'AR': 'Argentina', | |
3788 | 'AM': 'Armenia', | |
3789 | 'AW': 'Aruba', | |
3790 | 'AU': 'Australia', | |
3791 | 'AT': 'Austria', | |
3792 | 'AZ': 'Azerbaijan', | |
3793 | 'BS': 'Bahamas', | |
3794 | 'BH': 'Bahrain', | |
3795 | 'BD': 'Bangladesh', | |
3796 | 'BB': 'Barbados', | |
3797 | 'BY': 'Belarus', | |
3798 | 'BE': 'Belgium', | |
3799 | 'BZ': 'Belize', | |
3800 | 'BJ': 'Benin', | |
3801 | 'BM': 'Bermuda', | |
3802 | 'BT': 'Bhutan', | |
3803 | 'BO': 'Bolivia, Plurinational State of', | |
3804 | 'BQ': 'Bonaire, Sint Eustatius and Saba', | |
3805 | 'BA': 'Bosnia and Herzegovina', | |
3806 | 'BW': 'Botswana', | |
3807 | 'BV': 'Bouvet Island', | |
3808 | 'BR': 'Brazil', | |
3809 | 'IO': 'British Indian Ocean Territory', | |
3810 | 'BN': 'Brunei Darussalam', | |
3811 | 'BG': 'Bulgaria', | |
3812 | 'BF': 'Burkina Faso', | |
3813 | 'BI': 'Burundi', | |
3814 | 'KH': 'Cambodia', | |
3815 | 'CM': 'Cameroon', | |
3816 | 'CA': 'Canada', | |
3817 | 'CV': 'Cape Verde', | |
3818 | 'KY': 'Cayman Islands', | |
3819 | 'CF': 'Central African Republic', | |
3820 | 'TD': 'Chad', | |
3821 | 'CL': 'Chile', | |
3822 | 'CN': 'China', | |
3823 | 'CX': 'Christmas Island', | |
3824 | 'CC': 'Cocos (Keeling) Islands', | |
3825 | 'CO': 'Colombia', | |
3826 | 'KM': 'Comoros', | |
3827 | 'CG': 'Congo', | |
3828 | 'CD': 'Congo, the Democratic Republic of the', | |
3829 | 'CK': 'Cook Islands', | |
3830 | 'CR': 'Costa Rica', | |
3831 | 'CI': 'Côte d\'Ivoire', | |
3832 | 'HR': 'Croatia', | |
3833 | 'CU': 'Cuba', | |
3834 | 'CW': 'Curaçao', | |
3835 | 'CY': 'Cyprus', | |
3836 | 'CZ': 'Czech Republic', | |
3837 | 'DK': 'Denmark', | |
3838 | 'DJ': 'Djibouti', | |
3839 | 'DM': 'Dominica', | |
3840 | 'DO': 'Dominican Republic', | |
3841 | 'EC': 'Ecuador', | |
3842 | 'EG': 'Egypt', | |
3843 | 'SV': 'El Salvador', | |
3844 | 'GQ': 'Equatorial Guinea', | |
3845 | 'ER': 'Eritrea', | |
3846 | 'EE': 'Estonia', | |
3847 | 'ET': 'Ethiopia', | |
3848 | 'FK': 'Falkland Islands (Malvinas)', | |
3849 | 'FO': 'Faroe Islands', | |
3850 | 'FJ': 'Fiji', | |
3851 | 'FI': 'Finland', | |
3852 | 'FR': 'France', | |
3853 | 'GF': 'French Guiana', | |
3854 | 'PF': 'French Polynesia', | |
3855 | 'TF': 'French Southern Territories', | |
3856 | 'GA': 'Gabon', | |
3857 | 'GM': 'Gambia', | |
3858 | 'GE': 'Georgia', | |
3859 | 'DE': 'Germany', | |
3860 | 'GH': 'Ghana', | |
3861 | 'GI': 'Gibraltar', | |
3862 | 'GR': 'Greece', | |
3863 | 'GL': 'Greenland', | |
3864 | 'GD': 'Grenada', | |
3865 | 'GP': 'Guadeloupe', | |
3866 | 'GU': 'Guam', | |
3867 | 'GT': 'Guatemala', | |
3868 | 'GG': 'Guernsey', | |
3869 | 'GN': 'Guinea', | |
3870 | 'GW': 'Guinea-Bissau', | |
3871 | 'GY': 'Guyana', | |
3872 | 'HT': 'Haiti', | |
3873 | 'HM': 'Heard Island and McDonald Islands', | |
3874 | 'VA': 'Holy See (Vatican City State)', | |
3875 | 'HN': 'Honduras', | |
3876 | 'HK': 'Hong Kong', | |
3877 | 'HU': 'Hungary', | |
3878 | 'IS': 'Iceland', | |
3879 | 'IN': 'India', | |
3880 | 'ID': 'Indonesia', | |
3881 | 'IR': 'Iran, Islamic Republic of', | |
3882 | 'IQ': 'Iraq', | |
3883 | 'IE': 'Ireland', | |
3884 | 'IM': 'Isle of Man', | |
3885 | 'IL': 'Israel', | |
3886 | 'IT': 'Italy', | |
3887 | 'JM': 'Jamaica', | |
3888 | 'JP': 'Japan', | |
3889 | 'JE': 'Jersey', | |
3890 | 'JO': 'Jordan', | |
3891 | 'KZ': 'Kazakhstan', | |
3892 | 'KE': 'Kenya', | |
3893 | 'KI': 'Kiribati', | |
3894 | 'KP': 'Korea, Democratic People\'s Republic of', | |
3895 | 'KR': 'Korea, Republic of', | |
3896 | 'KW': 'Kuwait', | |
3897 | 'KG': 'Kyrgyzstan', | |
3898 | 'LA': 'Lao People\'s Democratic Republic', | |
3899 | 'LV': 'Latvia', | |
3900 | 'LB': 'Lebanon', | |
3901 | 'LS': 'Lesotho', | |
3902 | 'LR': 'Liberia', | |
3903 | 'LY': 'Libya', | |
3904 | 'LI': 'Liechtenstein', | |
3905 | 'LT': 'Lithuania', | |
3906 | 'LU': 'Luxembourg', | |
3907 | 'MO': 'Macao', | |
3908 | 'MK': 'Macedonia, the Former Yugoslav Republic of', | |
3909 | 'MG': 'Madagascar', | |
3910 | 'MW': 'Malawi', | |
3911 | 'MY': 'Malaysia', | |
3912 | 'MV': 'Maldives', | |
3913 | 'ML': 'Mali', | |
3914 | 'MT': 'Malta', | |
3915 | 'MH': 'Marshall Islands', | |
3916 | 'MQ': 'Martinique', | |
3917 | 'MR': 'Mauritania', | |
3918 | 'MU': 'Mauritius', | |
3919 | 'YT': 'Mayotte', | |
3920 | 'MX': 'Mexico', | |
3921 | 'FM': 'Micronesia, Federated States of', | |
3922 | 'MD': 'Moldova, Republic of', | |
3923 | 'MC': 'Monaco', | |
3924 | 'MN': 'Mongolia', | |
3925 | 'ME': 'Montenegro', | |
3926 | 'MS': 'Montserrat', | |
3927 | 'MA': 'Morocco', | |
3928 | 'MZ': 'Mozambique', | |
3929 | 'MM': 'Myanmar', | |
3930 | 'NA': 'Namibia', | |
3931 | 'NR': 'Nauru', | |
3932 | 'NP': 'Nepal', | |
3933 | 'NL': 'Netherlands', | |
3934 | 'NC': 'New Caledonia', | |
3935 | 'NZ': 'New Zealand', | |
3936 | 'NI': 'Nicaragua', | |
3937 | 'NE': 'Niger', | |
3938 | 'NG': 'Nigeria', | |
3939 | 'NU': 'Niue', | |
3940 | 'NF': 'Norfolk Island', | |
3941 | 'MP': 'Northern Mariana Islands', | |
3942 | 'NO': 'Norway', | |
3943 | 'OM': 'Oman', | |
3944 | 'PK': 'Pakistan', | |
3945 | 'PW': 'Palau', | |
3946 | 'PS': 'Palestine, State of', | |
3947 | 'PA': 'Panama', | |
3948 | 'PG': 'Papua New Guinea', | |
3949 | 'PY': 'Paraguay', | |
3950 | 'PE': 'Peru', | |
3951 | 'PH': 'Philippines', | |
3952 | 'PN': 'Pitcairn', | |
3953 | 'PL': 'Poland', | |
3954 | 'PT': 'Portugal', | |
3955 | 'PR': 'Puerto Rico', | |
3956 | 'QA': 'Qatar', | |
3957 | 'RE': 'Réunion', | |
3958 | 'RO': 'Romania', | |
3959 | 'RU': 'Russian Federation', | |
3960 | 'RW': 'Rwanda', | |
3961 | 'BL': 'Saint Barthélemy', | |
3962 | 'SH': 'Saint Helena, Ascension and Tristan da Cunha', | |
3963 | 'KN': 'Saint Kitts and Nevis', | |
3964 | 'LC': 'Saint Lucia', | |
3965 | 'MF': 'Saint Martin (French part)', | |
3966 | 'PM': 'Saint Pierre and Miquelon', | |
3967 | 'VC': 'Saint Vincent and the Grenadines', | |
3968 | 'WS': 'Samoa', | |
3969 | 'SM': 'San Marino', | |
3970 | 'ST': 'Sao Tome and Principe', | |
3971 | 'SA': 'Saudi Arabia', | |
3972 | 'SN': 'Senegal', | |
3973 | 'RS': 'Serbia', | |
3974 | 'SC': 'Seychelles', | |
3975 | 'SL': 'Sierra Leone', | |
3976 | 'SG': 'Singapore', | |
3977 | 'SX': 'Sint Maarten (Dutch part)', | |
3978 | 'SK': 'Slovakia', | |
3979 | 'SI': 'Slovenia', | |
3980 | 'SB': 'Solomon Islands', | |
3981 | 'SO': 'Somalia', | |
3982 | 'ZA': 'South Africa', | |
3983 | 'GS': 'South Georgia and the South Sandwich Islands', | |
3984 | 'SS': 'South Sudan', | |
3985 | 'ES': 'Spain', | |
3986 | 'LK': 'Sri Lanka', | |
3987 | 'SD': 'Sudan', | |
3988 | 'SR': 'Suriname', | |
3989 | 'SJ': 'Svalbard and Jan Mayen', | |
3990 | 'SZ': 'Swaziland', | |
3991 | 'SE': 'Sweden', | |
3992 | 'CH': 'Switzerland', | |
3993 | 'SY': 'Syrian Arab Republic', | |
3994 | 'TW': 'Taiwan, Province of China', | |
3995 | 'TJ': 'Tajikistan', | |
3996 | 'TZ': 'Tanzania, United Republic of', | |
3997 | 'TH': 'Thailand', | |
3998 | 'TL': 'Timor-Leste', | |
3999 | 'TG': 'Togo', | |
4000 | 'TK': 'Tokelau', | |
4001 | 'TO': 'Tonga', | |
4002 | 'TT': 'Trinidad and Tobago', | |
4003 | 'TN': 'Tunisia', | |
4004 | 'TR': 'Turkey', | |
4005 | 'TM': 'Turkmenistan', | |
4006 | 'TC': 'Turks and Caicos Islands', | |
4007 | 'TV': 'Tuvalu', | |
4008 | 'UG': 'Uganda', | |
4009 | 'UA': 'Ukraine', | |
4010 | 'AE': 'United Arab Emirates', | |
4011 | 'GB': 'United Kingdom', | |
4012 | 'US': 'United States', | |
4013 | 'UM': 'United States Minor Outlying Islands', | |
4014 | 'UY': 'Uruguay', | |
4015 | 'UZ': 'Uzbekistan', | |
4016 | 'VU': 'Vanuatu', | |
4017 | 'VE': 'Venezuela, Bolivarian Republic of', | |
4018 | 'VN': 'Viet Nam', | |
4019 | 'VG': 'Virgin Islands, British', | |
4020 | 'VI': 'Virgin Islands, U.S.', | |
4021 | 'WF': 'Wallis and Futuna', | |
4022 | 'EH': 'Western Sahara', | |
4023 | 'YE': 'Yemen', | |
4024 | 'ZM': 'Zambia', | |
4025 | 'ZW': 'Zimbabwe', | |
4026 | # Not ISO 3166 codes, but used for IP blocks | |
4027 | 'AP': 'Asia/Pacific Region', | |
4028 | 'EU': 'Europe', | |
4029 | } | |
4030 | ||
4031 | @classmethod | |
4032 | def short2full(cls, code): | |
4033 | """Convert an ISO 3166-2 country code to the corresponding full name""" | |
4034 | return cls._country_map.get(code.upper()) | |
4035 | ||
4036 | ||
4037 | class GeoUtils: | |
4038 | # Major IPv4 address blocks per country | |
4039 | _country_ip_map = { | |
4040 | 'AD': '46.172.224.0/19', | |
4041 | 'AE': '94.200.0.0/13', | |
4042 | 'AF': '149.54.0.0/17', | |
4043 | 'AG': '209.59.64.0/18', | |
4044 | 'AI': '204.14.248.0/21', | |
4045 | 'AL': '46.99.0.0/16', | |
4046 | 'AM': '46.70.0.0/15', | |
4047 | 'AO': '105.168.0.0/13', | |
4048 | 'AP': '182.50.184.0/21', | |
4049 | 'AQ': '23.154.160.0/24', | |
4050 | 'AR': '181.0.0.0/12', | |
4051 | 'AS': '202.70.112.0/20', | |
4052 | 'AT': '77.116.0.0/14', | |
4053 | 'AU': '1.128.0.0/11', | |
4054 | 'AW': '181.41.0.0/18', | |
4055 | 'AX': '185.217.4.0/22', | |
4056 | 'AZ': '5.197.0.0/16', | |
4057 | 'BA': '31.176.128.0/17', | |
4058 | 'BB': '65.48.128.0/17', | |
4059 | 'BD': '114.130.0.0/16', | |
4060 | 'BE': '57.0.0.0/8', | |
4061 | 'BF': '102.178.0.0/15', | |
4062 | 'BG': '95.42.0.0/15', | |
4063 | 'BH': '37.131.0.0/17', | |
4064 | 'BI': '154.117.192.0/18', | |
4065 | 'BJ': '137.255.0.0/16', | |
4066 | 'BL': '185.212.72.0/23', | |
4067 | 'BM': '196.12.64.0/18', | |
4068 | 'BN': '156.31.0.0/16', | |
4069 | 'BO': '161.56.0.0/16', | |
4070 | 'BQ': '161.0.80.0/20', | |
4071 | 'BR': '191.128.0.0/12', | |
4072 | 'BS': '24.51.64.0/18', | |
4073 | 'BT': '119.2.96.0/19', | |
4074 | 'BW': '168.167.0.0/16', | |
4075 | 'BY': '178.120.0.0/13', | |
4076 | 'BZ': '179.42.192.0/18', | |
4077 | 'CA': '99.224.0.0/11', | |
4078 | 'CD': '41.243.0.0/16', | |
4079 | 'CF': '197.242.176.0/21', | |
4080 | 'CG': '160.113.0.0/16', | |
4081 | 'CH': '85.0.0.0/13', | |
4082 | 'CI': '102.136.0.0/14', | |
4083 | 'CK': '202.65.32.0/19', | |
4084 | 'CL': '152.172.0.0/14', | |
4085 | 'CM': '102.244.0.0/14', | |
4086 | 'CN': '36.128.0.0/10', | |
4087 | 'CO': '181.240.0.0/12', | |
4088 | 'CR': '201.192.0.0/12', | |
4089 | 'CU': '152.206.0.0/15', | |
4090 | 'CV': '165.90.96.0/19', | |
4091 | 'CW': '190.88.128.0/17', | |
4092 | 'CY': '31.153.0.0/16', | |
4093 | 'CZ': '88.100.0.0/14', | |
4094 | 'DE': '53.0.0.0/8', | |
4095 | 'DJ': '197.241.0.0/17', | |
4096 | 'DK': '87.48.0.0/12', | |
4097 | 'DM': '192.243.48.0/20', | |
4098 | 'DO': '152.166.0.0/15', | |
4099 | 'DZ': '41.96.0.0/12', | |
4100 | 'EC': '186.68.0.0/15', | |
4101 | 'EE': '90.190.0.0/15', | |
4102 | 'EG': '156.160.0.0/11', | |
4103 | 'ER': '196.200.96.0/20', | |
4104 | 'ES': '88.0.0.0/11', | |
4105 | 'ET': '196.188.0.0/14', | |
4106 | 'EU': '2.16.0.0/13', | |
4107 | 'FI': '91.152.0.0/13', | |
4108 | 'FJ': '144.120.0.0/16', | |
4109 | 'FK': '80.73.208.0/21', | |
4110 | 'FM': '119.252.112.0/20', | |
4111 | 'FO': '88.85.32.0/19', | |
4112 | 'FR': '90.0.0.0/9', | |
4113 | 'GA': '41.158.0.0/15', | |
4114 | 'GB': '25.0.0.0/8', | |
4115 | 'GD': '74.122.88.0/21', | |
4116 | 'GE': '31.146.0.0/16', | |
4117 | 'GF': '161.22.64.0/18', | |
4118 | 'GG': '62.68.160.0/19', | |
4119 | 'GH': '154.160.0.0/12', | |
4120 | 'GI': '95.164.0.0/16', | |
4121 | 'GL': '88.83.0.0/19', | |
4122 | 'GM': '160.182.0.0/15', | |
4123 | 'GN': '197.149.192.0/18', | |
4124 | 'GP': '104.250.0.0/19', | |
4125 | 'GQ': '105.235.224.0/20', | |
4126 | 'GR': '94.64.0.0/13', | |
4127 | 'GT': '168.234.0.0/16', | |
4128 | 'GU': '168.123.0.0/16', | |
4129 | 'GW': '197.214.80.0/20', | |
4130 | 'GY': '181.41.64.0/18', | |
4131 | 'HK': '113.252.0.0/14', | |
4132 | 'HN': '181.210.0.0/16', | |
4133 | 'HR': '93.136.0.0/13', | |
4134 | 'HT': '148.102.128.0/17', | |
4135 | 'HU': '84.0.0.0/14', | |
4136 | 'ID': '39.192.0.0/10', | |
4137 | 'IE': '87.32.0.0/12', | |
4138 | 'IL': '79.176.0.0/13', | |
4139 | 'IM': '5.62.80.0/20', | |
4140 | 'IN': '117.192.0.0/10', | |
4141 | 'IO': '203.83.48.0/21', | |
4142 | 'IQ': '37.236.0.0/14', | |
4143 | 'IR': '2.176.0.0/12', | |
4144 | 'IS': '82.221.0.0/16', | |
4145 | 'IT': '79.0.0.0/10', | |
4146 | 'JE': '87.244.64.0/18', | |
4147 | 'JM': '72.27.0.0/17', | |
4148 | 'JO': '176.29.0.0/16', | |
4149 | 'JP': '133.0.0.0/8', | |
4150 | 'KE': '105.48.0.0/12', | |
4151 | 'KG': '158.181.128.0/17', | |
4152 | 'KH': '36.37.128.0/17', | |
4153 | 'KI': '103.25.140.0/22', | |
4154 | 'KM': '197.255.224.0/20', | |
4155 | 'KN': '198.167.192.0/19', | |
4156 | 'KP': '175.45.176.0/22', | |
4157 | 'KR': '175.192.0.0/10', | |
4158 | 'KW': '37.36.0.0/14', | |
4159 | 'KY': '64.96.0.0/15', | |
4160 | 'KZ': '2.72.0.0/13', | |
4161 | 'LA': '115.84.64.0/18', | |
4162 | 'LB': '178.135.0.0/16', | |
4163 | 'LC': '24.92.144.0/20', | |
4164 | 'LI': '82.117.0.0/19', | |
4165 | 'LK': '112.134.0.0/15', | |
4166 | 'LR': '102.183.0.0/16', | |
4167 | 'LS': '129.232.0.0/17', | |
4168 | 'LT': '78.56.0.0/13', | |
4169 | 'LU': '188.42.0.0/16', | |
4170 | 'LV': '46.109.0.0/16', | |
4171 | 'LY': '41.252.0.0/14', | |
4172 | 'MA': '105.128.0.0/11', | |
4173 | 'MC': '88.209.64.0/18', | |
4174 | 'MD': '37.246.0.0/16', | |
4175 | 'ME': '178.175.0.0/17', | |
4176 | 'MF': '74.112.232.0/21', | |
4177 | 'MG': '154.126.0.0/17', | |
4178 | 'MH': '117.103.88.0/21', | |
4179 | 'MK': '77.28.0.0/15', | |
4180 | 'ML': '154.118.128.0/18', | |
4181 | 'MM': '37.111.0.0/17', | |
4182 | 'MN': '49.0.128.0/17', | |
4183 | 'MO': '60.246.0.0/16', | |
4184 | 'MP': '202.88.64.0/20', | |
4185 | 'MQ': '109.203.224.0/19', | |
4186 | 'MR': '41.188.64.0/18', | |
4187 | 'MS': '208.90.112.0/22', | |
4188 | 'MT': '46.11.0.0/16', | |
4189 | 'MU': '105.16.0.0/12', | |
4190 | 'MV': '27.114.128.0/18', | |
4191 | 'MW': '102.70.0.0/15', | |
4192 | 'MX': '187.192.0.0/11', | |
4193 | 'MY': '175.136.0.0/13', | |
4194 | 'MZ': '197.218.0.0/15', | |
4195 | 'NA': '41.182.0.0/16', | |
4196 | 'NC': '101.101.0.0/18', | |
4197 | 'NE': '197.214.0.0/18', | |
4198 | 'NF': '203.17.240.0/22', | |
4199 | 'NG': '105.112.0.0/12', | |
4200 | 'NI': '186.76.0.0/15', | |
4201 | 'NL': '145.96.0.0/11', | |
4202 | 'NO': '84.208.0.0/13', | |
4203 | 'NP': '36.252.0.0/15', | |
4204 | 'NR': '203.98.224.0/19', | |
4205 | 'NU': '49.156.48.0/22', | |
4206 | 'NZ': '49.224.0.0/14', | |
4207 | 'OM': '5.36.0.0/15', | |
4208 | 'PA': '186.72.0.0/15', | |
4209 | 'PE': '186.160.0.0/14', | |
4210 | 'PF': '123.50.64.0/18', | |
4211 | 'PG': '124.240.192.0/19', | |
4212 | 'PH': '49.144.0.0/13', | |
4213 | 'PK': '39.32.0.0/11', | |
4214 | 'PL': '83.0.0.0/11', | |
4215 | 'PM': '70.36.0.0/20', | |
4216 | 'PR': '66.50.0.0/16', | |
4217 | 'PS': '188.161.0.0/16', | |
4218 | 'PT': '85.240.0.0/13', | |
4219 | 'PW': '202.124.224.0/20', | |
4220 | 'PY': '181.120.0.0/14', | |
4221 | 'QA': '37.210.0.0/15', | |
4222 | 'RE': '102.35.0.0/16', | |
4223 | 'RO': '79.112.0.0/13', | |
4224 | 'RS': '93.86.0.0/15', | |
4225 | 'RU': '5.136.0.0/13', | |
4226 | 'RW': '41.186.0.0/16', | |
4227 | 'SA': '188.48.0.0/13', | |
4228 | 'SB': '202.1.160.0/19', | |
4229 | 'SC': '154.192.0.0/11', | |
4230 | 'SD': '102.120.0.0/13', | |
4231 | 'SE': '78.64.0.0/12', | |
4232 | 'SG': '8.128.0.0/10', | |
4233 | 'SI': '188.196.0.0/14', | |
4234 | 'SK': '78.98.0.0/15', | |
4235 | 'SL': '102.143.0.0/17', | |
4236 | 'SM': '89.186.32.0/19', | |
4237 | 'SN': '41.82.0.0/15', | |
4238 | 'SO': '154.115.192.0/18', | |
4239 | 'SR': '186.179.128.0/17', | |
4240 | 'SS': '105.235.208.0/21', | |
4241 | 'ST': '197.159.160.0/19', | |
4242 | 'SV': '168.243.0.0/16', | |
4243 | 'SX': '190.102.0.0/20', | |
4244 | 'SY': '5.0.0.0/16', | |
4245 | 'SZ': '41.84.224.0/19', | |
4246 | 'TC': '65.255.48.0/20', | |
4247 | 'TD': '154.68.128.0/19', | |
4248 | 'TG': '196.168.0.0/14', | |
4249 | 'TH': '171.96.0.0/13', | |
4250 | 'TJ': '85.9.128.0/18', | |
4251 | 'TK': '27.96.24.0/21', | |
4252 | 'TL': '180.189.160.0/20', | |
4253 | 'TM': '95.85.96.0/19', | |
4254 | 'TN': '197.0.0.0/11', | |
4255 | 'TO': '175.176.144.0/21', | |
4256 | 'TR': '78.160.0.0/11', | |
4257 | 'TT': '186.44.0.0/15', | |
4258 | 'TV': '202.2.96.0/19', | |
4259 | 'TW': '120.96.0.0/11', | |
4260 | 'TZ': '156.156.0.0/14', | |
4261 | 'UA': '37.52.0.0/14', | |
4262 | 'UG': '102.80.0.0/13', | |
4263 | 'US': '6.0.0.0/8', | |
4264 | 'UY': '167.56.0.0/13', | |
4265 | 'UZ': '84.54.64.0/18', | |
4266 | 'VA': '212.77.0.0/19', | |
4267 | 'VC': '207.191.240.0/21', | |
4268 | 'VE': '186.88.0.0/13', | |
4269 | 'VG': '66.81.192.0/20', | |
4270 | 'VI': '146.226.0.0/16', | |
4271 | 'VN': '14.160.0.0/11', | |
4272 | 'VU': '202.80.32.0/20', | |
4273 | 'WF': '117.20.32.0/21', | |
4274 | 'WS': '202.4.32.0/19', | |
4275 | 'YE': '134.35.0.0/16', | |
4276 | 'YT': '41.242.116.0/22', | |
4277 | 'ZA': '41.0.0.0/11', | |
4278 | 'ZM': '102.144.0.0/13', | |
4279 | 'ZW': '102.177.192.0/18', | |
4280 | } | |
4281 | ||
4282 | @classmethod | |
4283 | def random_ipv4(cls, code_or_block): | |
4284 | if len(code_or_block) == 2: | |
4285 | block = cls._country_ip_map.get(code_or_block.upper()) | |
4286 | if not block: | |
4287 | return None | |
4288 | else: | |
4289 | block = code_or_block | |
4290 | addr, preflen = block.split('/') | |
4291 | addr_min = struct.unpack('!L', socket.inet_aton(addr))[0] | |
4292 | addr_max = addr_min | (0xffffffff >> int(preflen)) | |
4293 | return str(socket.inet_ntoa( | |
4294 | struct.pack('!L', random.randint(addr_min, addr_max)))) | |
4295 | ||
4296 | ||
4297 | # Both long_to_bytes and bytes_to_long are adapted from PyCrypto, which is | |
4298 | # released into Public Domain | |
4299 | # https://github.com/dlitz/pycrypto/blob/master/lib/Crypto/Util/number.py#L387 | |
4300 | ||
4301 | def long_to_bytes(n, blocksize=0): | |
4302 | """long_to_bytes(n:long, blocksize:int) : string | |
4303 | Convert a long integer to a byte string. | |
4304 | ||
4305 | If optional blocksize is given and greater than zero, pad the front of the | |
4306 | byte string with binary zeros so that the length is a multiple of | |
4307 | blocksize. | |
4308 | """ | |
4309 | # after much testing, this algorithm was deemed to be the fastest | |
4310 | s = b'' | |
4311 | n = int(n) | |
4312 | while n > 0: | |
4313 | s = struct.pack('>I', n & 0xffffffff) + s | |
4314 | n = n >> 32 | |
4315 | # strip off leading zeros | |
4316 | for i in range(len(s)): | |
4317 | if s[i] != b'\000'[0]: | |
4318 | break | |
4319 | else: | |
4320 | # only happens when n == 0 | |
4321 | s = b'\000' | |
4322 | i = 0 | |
4323 | s = s[i:] | |
4324 | # add back some pad bytes. this could be done more efficiently w.r.t. the | |
4325 | # de-padding being done above, but sigh... | |
4326 | if blocksize > 0 and len(s) % blocksize: | |
4327 | s = (blocksize - len(s) % blocksize) * b'\000' + s | |
4328 | return s | |
4329 | ||
4330 | ||
4331 | def bytes_to_long(s): | |
4332 | """bytes_to_long(string) : long | |
4333 | Convert a byte string to a long integer. | |
4334 | ||
4335 | This is (essentially) the inverse of long_to_bytes(). | |
4336 | """ | |
4337 | acc = 0 | |
4338 | length = len(s) | |
4339 | if length % 4: | |
4340 | extra = (4 - length % 4) | |
4341 | s = b'\000' * extra + s | |
4342 | length = length + extra | |
4343 | for i in range(0, length, 4): | |
4344 | acc = (acc << 32) + struct.unpack('>I', s[i:i + 4])[0] | |
4345 | return acc | |
4346 | ||
4347 | ||
4348 | def ohdave_rsa_encrypt(data, exponent, modulus): | |
4349 | ''' | |
4350 | Implement OHDave's RSA algorithm. See http://www.ohdave.com/rsa/ | |
4351 | ||
4352 | Input: | |
4353 | data: data to encrypt, bytes-like object | |
4354 | exponent, modulus: parameter e and N of RSA algorithm, both integer | |
4355 | Output: hex string of encrypted data | |
4356 | ||
4357 | Limitation: supports one block encryption only | |
4358 | ''' | |
4359 | ||
4360 | payload = int(binascii.hexlify(data[::-1]), 16) | |
4361 | encrypted = pow(payload, exponent, modulus) | |
4362 | return '%x' % encrypted | |
4363 | ||
4364 | ||
4365 | def pkcs1pad(data, length): | |
4366 | """ | |
4367 | Padding input data with PKCS#1 scheme | |
4368 | ||
4369 | @param {int[]} data input data | |
4370 | @param {int} length target length | |
4371 | @returns {int[]} padded data | |
4372 | """ | |
4373 | if len(data) > length - 11: | |
4374 | raise ValueError('Input data too long for PKCS#1 padding') | |
4375 | ||
4376 | pseudo_random = [random.randint(0, 254) for _ in range(length - len(data) - 3)] | |
4377 | return [0, 2] + pseudo_random + [0] + data | |
4378 | ||
4379 | ||
4380 | def _base_n_table(n, table): | |
4381 | if not table and not n: | |
4382 | raise ValueError('Either table or n must be specified') | |
4383 | table = (table or '0123456789abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ')[:n] | |
4384 | ||
4385 | if n and n != len(table): | |
4386 | raise ValueError(f'base {n} exceeds table length {len(table)}') | |
4387 | return table | |
4388 | ||
4389 | ||
4390 | def encode_base_n(num, n=None, table=None): | |
4391 | """Convert given int to a base-n string""" | |
4392 | table = _base_n_table(n, table) | |
4393 | if not num: | |
4394 | return table[0] | |
4395 | ||
4396 | result, base = '', len(table) | |
4397 | while num: | |
4398 | result = table[num % base] + result | |
4399 | num = num // base | |
4400 | return result | |
4401 | ||
4402 | ||
4403 | def decode_base_n(string, n=None, table=None): | |
4404 | """Convert given base-n string to int""" | |
4405 | table = {char: index for index, char in enumerate(_base_n_table(n, table))} | |
4406 | result, base = 0, len(table) | |
4407 | for char in string: | |
4408 | result = result * base + table[char] | |
4409 | return result | |
4410 | ||
4411 | ||
4412 | def decode_packed_codes(code): | |
4413 | mobj = re.search(PACKED_CODES_RE, code) | |
4414 | obfuscated_code, base, count, symbols = mobj.groups() | |
4415 | base = int(base) | |
4416 | count = int(count) | |
4417 | symbols = symbols.split('|') | |
4418 | symbol_table = {} | |
4419 | ||
4420 | while count: | |
4421 | count -= 1 | |
4422 | base_n_count = encode_base_n(count, base) | |
4423 | symbol_table[base_n_count] = symbols[count] or base_n_count | |
4424 | ||
4425 | return re.sub( | |
4426 | r'\b(\w+)\b', lambda mobj: symbol_table[mobj.group(0)], | |
4427 | obfuscated_code) | |
4428 | ||
4429 | ||
4430 | def caesar(s, alphabet, shift): | |
4431 | if shift == 0: | |
4432 | return s | |
4433 | l = len(alphabet) | |
4434 | return ''.join( | |
4435 | alphabet[(alphabet.index(c) + shift) % l] if c in alphabet else c | |
4436 | for c in s) | |
4437 | ||
4438 | ||
4439 | def rot47(s): | |
4440 | return caesar(s, r'''!"#$%&'()*+,-./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{|}~''', 47) | |
4441 | ||
4442 | ||
4443 | def parse_m3u8_attributes(attrib): | |
4444 | info = {} | |
4445 | for (key, val) in re.findall(r'(?P<key>[A-Z0-9-]+)=(?P<val>"[^"]+"|[^",]+)(?:,|$)', attrib): | |
4446 | if val.startswith('"'): | |
4447 | val = val[1:-1] | |
4448 | info[key] = val | |
4449 | return info | |
4450 | ||
4451 | ||
4452 | def urshift(val, n): | |
4453 | return val >> n if val >= 0 else (val + 0x100000000) >> n | |
4454 | ||
4455 | ||
4456 | def write_xattr(path, key, value): | |
4457 | # Windows: Write xattrs to NTFS Alternate Data Streams: | |
4458 | # http://en.wikipedia.org/wiki/NTFS#Alternate_data_streams_.28ADS.29 | |
4459 | if compat_os_name == 'nt': | |
4460 | assert ':' not in key | |
4461 | assert os.path.exists(path) | |
4462 | ||
4463 | try: | |
4464 | with open(f'{path}:{key}', 'wb') as f: | |
4465 | f.write(value) | |
4466 | except OSError as e: | |
4467 | raise XAttrMetadataError(e.errno, e.strerror) | |
4468 | return | |
4469 | ||
4470 | # UNIX Method 1. Use os.setxattr/xattrs/pyxattrs modules | |
4471 | ||
4472 | setxattr = None | |
4473 | if callable(getattr(os, 'setxattr', None)): | |
4474 | setxattr = os.setxattr | |
4475 | elif getattr(xattr, '_yt_dlp__identifier', None) == 'pyxattr': | |
4476 | # Unicode arguments are not supported in pyxattr until version 0.5.0 | |
4477 | # See https://github.com/ytdl-org/youtube-dl/issues/5498 | |
4478 | if version_tuple(xattr.__version__) >= (0, 5, 0): | |
4479 | setxattr = xattr.set | |
4480 | elif xattr: | |
4481 | setxattr = xattr.setxattr | |
4482 | ||
4483 | if setxattr: | |
4484 | try: | |
4485 | setxattr(path, key, value) | |
4486 | except OSError as e: | |
4487 | raise XAttrMetadataError(e.errno, e.strerror) | |
4488 | return | |
4489 | ||
4490 | # UNIX Method 2. Use setfattr/xattr executables | |
4491 | exe = ('setfattr' if check_executable('setfattr', ['--version']) | |
4492 | else 'xattr' if check_executable('xattr', ['-h']) else None) | |
4493 | if not exe: | |
4494 | raise XAttrUnavailableError( | |
4495 | 'Couldn\'t find a tool to set the xattrs. Install either the "xattr" or "pyxattr" Python modules or the ' | |
4496 | + ('"xattr" binary' if sys.platform != 'linux' else 'GNU "attr" package (which contains the "setfattr" tool)')) | |
4497 | ||
4498 | value = value.decode() | |
4499 | try: | |
4500 | _, stderr, returncode = Popen.run( | |
4501 | [exe, '-w', key, value, path] if exe == 'xattr' else [exe, '-n', key, '-v', value, path], | |
4502 | text=True, stdout=subprocess.PIPE, stderr=subprocess.PIPE, stdin=subprocess.PIPE) | |
4503 | except OSError as e: | |
4504 | raise XAttrMetadataError(e.errno, e.strerror) | |
4505 | if returncode: | |
4506 | raise XAttrMetadataError(returncode, stderr) | |
4507 | ||
4508 | ||
4509 | def random_birthday(year_field, month_field, day_field): | |
4510 | start_date = dt.date(1950, 1, 1) | |
4511 | end_date = dt.date(1995, 12, 31) | |
4512 | offset = random.randint(0, (end_date - start_date).days) | |
4513 | random_date = start_date + dt.timedelta(offset) | |
4514 | return { | |
4515 | year_field: str(random_date.year), | |
4516 | month_field: str(random_date.month), | |
4517 | day_field: str(random_date.day), | |
4518 | } | |
4519 | ||
4520 | ||
4521 | def find_available_port(interface=''): | |
4522 | try: | |
4523 | with socket.socket() as sock: | |
4524 | sock.bind((interface, 0)) | |
4525 | return sock.getsockname()[1] | |
4526 | except OSError: | |
4527 | return None | |
4528 | ||
4529 | ||
4530 | # Templates for internet shortcut files, which are plain text files. | |
4531 | DOT_URL_LINK_TEMPLATE = '''\ | |
4532 | [InternetShortcut] | |
4533 | URL=%(url)s | |
4534 | ''' | |
4535 | ||
4536 | DOT_WEBLOC_LINK_TEMPLATE = '''\ | |
4537 | <?xml version="1.0" encoding="UTF-8"?> | |
4538 | <!DOCTYPE plist PUBLIC "-//Apple//DTD PLIST 1.0//EN" "http://www.apple.com/DTDs/PropertyList-1.0.dtd"> | |
4539 | <plist version="1.0"> | |
4540 | <dict> | |
4541 | \t<key>URL</key> | |
4542 | \t<string>%(url)s</string> | |
4543 | </dict> | |
4544 | </plist> | |
4545 | ''' | |
4546 | ||
4547 | DOT_DESKTOP_LINK_TEMPLATE = '''\ | |
4548 | [Desktop Entry] | |
4549 | Encoding=UTF-8 | |
4550 | Name=%(filename)s | |
4551 | Type=Link | |
4552 | URL=%(url)s | |
4553 | Icon=text-html | |
4554 | ''' | |
4555 | ||
4556 | LINK_TEMPLATES = { | |
4557 | 'url': DOT_URL_LINK_TEMPLATE, | |
4558 | 'desktop': DOT_DESKTOP_LINK_TEMPLATE, | |
4559 | 'webloc': DOT_WEBLOC_LINK_TEMPLATE, | |
4560 | } | |
4561 | ||
4562 | ||
4563 | def iri_to_uri(iri): | |
4564 | """ | |
4565 | Converts an IRI (Internationalized Resource Identifier, allowing Unicode characters) to a URI (Uniform Resource Identifier, ASCII-only). | |
4566 | ||
4567 | 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. | |
4568 | """ | |
4569 | ||
4570 | iri_parts = urllib.parse.urlparse(iri) | |
4571 | ||
4572 | if '[' in iri_parts.netloc: | |
4573 | raise ValueError('IPv6 URIs are not, yet, supported.') | |
4574 | # Querying `.netloc`, when there's only one bracket, also raises a ValueError. | |
4575 | ||
4576 | # 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. | |
4577 | ||
4578 | net_location = '' | |
4579 | if iri_parts.username: | |
4580 | net_location += urllib.parse.quote(iri_parts.username, safe=r"!$%&'()*+,~") | |
4581 | if iri_parts.password is not None: | |
4582 | net_location += ':' + urllib.parse.quote(iri_parts.password, safe=r"!$%&'()*+,~") | |
4583 | net_location += '@' | |
4584 | ||
4585 | net_location += iri_parts.hostname.encode('idna').decode() # Punycode for Unicode hostnames. | |
4586 | # The 'idna' encoding produces ASCII text. | |
4587 | if iri_parts.port is not None and iri_parts.port != 80: | |
4588 | net_location += ':' + str(iri_parts.port) | |
4589 | ||
4590 | return urllib.parse.urlunparse( | |
4591 | (iri_parts.scheme, | |
4592 | net_location, | |
4593 | ||
4594 | urllib.parse.quote_plus(iri_parts.path, safe=r"!$%&'()*+,/:;=@|~"), | |
4595 | ||
4596 | # Unsure about the `safe` argument, since this is a legacy way of handling parameters. | |
4597 | urllib.parse.quote_plus(iri_parts.params, safe=r"!$%&'()*+,/:;=@|~"), | |
4598 | ||
4599 | # Not totally sure about the `safe` argument, since the source does not explicitly mention the query URI component. | |
4600 | urllib.parse.quote_plus(iri_parts.query, safe=r"!$%&'()*+,/:;=?@{|}~"), | |
4601 | ||
4602 | urllib.parse.quote_plus(iri_parts.fragment, safe=r"!#$%&'()*+,/:;=?@{|}~"))) | |
4603 | ||
4604 | # Source for `safe` arguments: https://url.spec.whatwg.org/#percent-encoded-bytes. | |
4605 | ||
4606 | ||
4607 | def to_high_limit_path(path): | |
4608 | if sys.platform in ['win32', 'cygwin']: | |
4609 | # Work around MAX_PATH limitation on Windows. The maximum allowed length for the individual path segments may still be quite limited. | |
4610 | return '\\\\?\\' + os.path.abspath(path) | |
4611 | ||
4612 | return path | |
4613 | ||
4614 | ||
4615 | def format_field(obj, field=None, template='%s', ignore=NO_DEFAULT, default='', func=IDENTITY): | |
4616 | val = traversal.traverse_obj(obj, *variadic(field)) | |
4617 | if not val if ignore is NO_DEFAULT else val in variadic(ignore): | |
4618 | return default | |
4619 | return template % func(val) | |
4620 | ||
4621 | ||
4622 | def clean_podcast_url(url): | |
4623 | url = re.sub(r'''(?x) | |
4624 | (?: | |
4625 | (?: | |
4626 | chtbl\.com/track| | |
4627 | media\.blubrry\.com| # https://create.blubrry.com/resources/podcast-media-download-statistics/getting-started/ | |
4628 | play\.podtrac\.com| | |
4629 | chrt\.fm/track| | |
4630 | mgln\.ai/e | |
4631 | )(?:/[^/.]+)?| | |
4632 | (?:dts|www)\.podtrac\.com/(?:pts/)?redirect\.[0-9a-z]{3,4}| # http://analytics.podtrac.com/how-to-measure | |
4633 | flex\.acast\.com| | |
4634 | pd(?: | |
4635 | cn\.co| # https://podcorn.com/analytics-prefix/ | |
4636 | st\.fm # https://podsights.com/docs/ | |
4637 | )/e| | |
4638 | [0-9]\.gum\.fm| | |
4639 | pscrb\.fm/rss/p | |
4640 | )/''', '', url) | |
4641 | return re.sub(r'^\w+://(\w+://)', r'\1', url) | |
4642 | ||
4643 | ||
4644 | _HEX_TABLE = '0123456789abcdef' | |
4645 | ||
4646 | ||
4647 | def random_uuidv4(): | |
4648 | return re.sub(r'[xy]', lambda x: _HEX_TABLE[random.randint(0, 15)], 'xxxxxxxx-xxxx-4xxx-yxxx-xxxxxxxxxxxx') | |
4649 | ||
4650 | ||
4651 | def make_dir(path, to_screen=None): | |
4652 | try: | |
4653 | dn = os.path.dirname(path) | |
4654 | if dn: | |
4655 | os.makedirs(dn, exist_ok=True) | |
4656 | return True | |
4657 | except OSError as err: | |
4658 | if callable(to_screen) is not None: | |
4659 | to_screen(f'unable to create directory {err}') | |
4660 | return False | |
4661 | ||
4662 | ||
4663 | def get_executable_path(): | |
4664 | from ..update import _get_variant_and_executable_path | |
4665 | ||
4666 | return os.path.dirname(os.path.abspath(_get_variant_and_executable_path()[1])) | |
4667 | ||
4668 | ||
4669 | def get_user_config_dirs(package_name): | |
4670 | # .config (e.g. ~/.config/package_name) | |
4671 | xdg_config_home = os.getenv('XDG_CONFIG_HOME') or compat_expanduser('~/.config') | |
4672 | yield os.path.join(xdg_config_home, package_name) | |
4673 | ||
4674 | # appdata (%APPDATA%/package_name) | |
4675 | appdata_dir = os.getenv('appdata') | |
4676 | if appdata_dir: | |
4677 | yield os.path.join(appdata_dir, package_name) | |
4678 | ||
4679 | # home (~/.package_name) | |
4680 | yield os.path.join(compat_expanduser('~'), f'.{package_name}') | |
4681 | ||
4682 | ||
4683 | def get_system_config_dirs(package_name): | |
4684 | # /etc/package_name | |
4685 | yield os.path.join('/etc', package_name) | |
4686 | ||
4687 | ||
4688 | def time_seconds(**kwargs): | |
4689 | """ | |
4690 | Returns TZ-aware time in seconds since the epoch (1970-01-01T00:00:00Z) | |
4691 | """ | |
4692 | return time.time() + dt.timedelta(**kwargs).total_seconds() | |
4693 | ||
4694 | ||
4695 | # create a JSON Web Signature (jws) with HS256 algorithm | |
4696 | # the resulting format is in JWS Compact Serialization | |
4697 | # implemented following JWT https://www.rfc-editor.org/rfc/rfc7519.html | |
4698 | # implemented following JWS https://www.rfc-editor.org/rfc/rfc7515.html | |
4699 | def jwt_encode_hs256(payload_data, key, headers={}): | |
4700 | header_data = { | |
4701 | 'alg': 'HS256', | |
4702 | 'typ': 'JWT', | |
4703 | } | |
4704 | if headers: | |
4705 | header_data.update(headers) | |
4706 | header_b64 = base64.b64encode(json.dumps(header_data).encode()) | |
4707 | payload_b64 = base64.b64encode(json.dumps(payload_data).encode()) | |
4708 | h = hmac.new(key.encode(), header_b64 + b'.' + payload_b64, hashlib.sha256) | |
4709 | signature_b64 = base64.b64encode(h.digest()) | |
4710 | token = header_b64 + b'.' + payload_b64 + b'.' + signature_b64 | |
4711 | return token | |
4712 | ||
4713 | ||
4714 | # can be extended in future to verify the signature and parse header and return the algorithm used if it's not HS256 | |
4715 | def jwt_decode_hs256(jwt): | |
4716 | header_b64, payload_b64, signature_b64 = jwt.split('.') | |
4717 | # add trailing ='s that may have been stripped, superfluous ='s are ignored | |
4718 | payload_data = json.loads(base64.urlsafe_b64decode(f'{payload_b64}===')) | |
4719 | return payload_data | |
4720 | ||
4721 | ||
4722 | WINDOWS_VT_MODE = False if compat_os_name == 'nt' else None | |
4723 | ||
4724 | ||
4725 | @functools.cache | |
4726 | def supports_terminal_sequences(stream): | |
4727 | if compat_os_name == 'nt': | |
4728 | if not WINDOWS_VT_MODE: | |
4729 | return False | |
4730 | elif not os.getenv('TERM'): | |
4731 | return False | |
4732 | try: | |
4733 | return stream.isatty() | |
4734 | except BaseException: | |
4735 | return False | |
4736 | ||
4737 | ||
4738 | def windows_enable_vt_mode(): | |
4739 | """Ref: https://bugs.python.org/issue30075 """ | |
4740 | if get_windows_version() < (10, 0, 10586): | |
4741 | return | |
4742 | ||
4743 | import ctypes | |
4744 | import ctypes.wintypes | |
4745 | import msvcrt | |
4746 | ||
4747 | ENABLE_VIRTUAL_TERMINAL_PROCESSING = 0x0004 | |
4748 | ||
4749 | dll = ctypes.WinDLL('kernel32', use_last_error=False) | |
4750 | handle = os.open('CONOUT$', os.O_RDWR) | |
4751 | try: | |
4752 | h_out = ctypes.wintypes.HANDLE(msvcrt.get_osfhandle(handle)) | |
4753 | dw_original_mode = ctypes.wintypes.DWORD() | |
4754 | success = dll.GetConsoleMode(h_out, ctypes.byref(dw_original_mode)) | |
4755 | if not success: | |
4756 | raise Exception('GetConsoleMode failed') | |
4757 | ||
4758 | success = dll.SetConsoleMode(h_out, ctypes.wintypes.DWORD( | |
4759 | dw_original_mode.value | ENABLE_VIRTUAL_TERMINAL_PROCESSING)) | |
4760 | if not success: | |
4761 | raise Exception('SetConsoleMode failed') | |
4762 | finally: | |
4763 | os.close(handle) | |
4764 | ||
4765 | global WINDOWS_VT_MODE | |
4766 | WINDOWS_VT_MODE = True | |
4767 | supports_terminal_sequences.cache_clear() | |
4768 | ||
4769 | ||
4770 | _terminal_sequences_re = re.compile('\033\\[[^m]+m') | |
4771 | ||
4772 | ||
4773 | def remove_terminal_sequences(string): | |
4774 | return _terminal_sequences_re.sub('', string) | |
4775 | ||
4776 | ||
4777 | def number_of_digits(number): | |
4778 | return len('%d' % number) | |
4779 | ||
4780 | ||
4781 | def join_nonempty(*values, delim='-', from_dict=None): | |
4782 | if from_dict is not None: | |
4783 | values = (traversal.traverse_obj(from_dict, variadic(v)) for v in values) | |
4784 | return delim.join(map(str, filter(None, values))) | |
4785 | ||
4786 | ||
4787 | def scale_thumbnails_to_max_format_width(formats, thumbnails, url_width_re): | |
4788 | """ | |
4789 | Find the largest format dimensions in terms of video width and, for each thumbnail: | |
4790 | * Modify the URL: Match the width with the provided regex and replace with the former width | |
4791 | * Update dimensions | |
4792 | ||
4793 | This function is useful with video services that scale the provided thumbnails on demand | |
4794 | """ | |
4795 | _keys = ('width', 'height') | |
4796 | max_dimensions = max( | |
4797 | (tuple(format.get(k) or 0 for k in _keys) for format in formats), | |
4798 | default=(0, 0)) | |
4799 | if not max_dimensions[0]: | |
4800 | return thumbnails | |
4801 | return [ | |
4802 | merge_dicts( | |
4803 | {'url': re.sub(url_width_re, str(max_dimensions[0]), thumbnail['url'])}, | |
4804 | dict(zip(_keys, max_dimensions)), thumbnail) | |
4805 | for thumbnail in thumbnails | |
4806 | ] | |
4807 | ||
4808 | ||
4809 | def parse_http_range(range): | |
4810 | """ Parse value of "Range" or "Content-Range" HTTP header into tuple. """ | |
4811 | if not range: | |
4812 | return None, None, None | |
4813 | crg = re.search(r'bytes[ =](\d+)-(\d+)?(?:/(\d+))?', range) | |
4814 | if not crg: | |
4815 | return None, None, None | |
4816 | return int(crg.group(1)), int_or_none(crg.group(2)), int_or_none(crg.group(3)) | |
4817 | ||
4818 | ||
4819 | def read_stdin(what): | |
4820 | if what: | |
4821 | eof = 'Ctrl+Z' if compat_os_name == 'nt' else 'Ctrl+D' | |
4822 | write_string(f'Reading {what} from STDIN - EOF ({eof}) to end:\n') | |
4823 | return sys.stdin | |
4824 | ||
4825 | ||
4826 | def determine_file_encoding(data): | |
4827 | """ | |
4828 | Detect the text encoding used | |
4829 | @returns (encoding, bytes to skip) | |
4830 | """ | |
4831 | ||
4832 | # BOM marks are given priority over declarations | |
4833 | for bom, enc in BOMS: | |
4834 | if data.startswith(bom): | |
4835 | return enc, len(bom) | |
4836 | ||
4837 | # Strip off all null bytes to match even when UTF-16 or UTF-32 is used. | |
4838 | # We ignore the endianness to get a good enough match | |
4839 | data = data.replace(b'\0', b'') | |
4840 | mobj = re.match(rb'(?m)^#\s*coding\s*:\s*(\S+)\s*$', data) | |
4841 | return mobj.group(1).decode() if mobj else None, 0 | |
4842 | ||
4843 | ||
4844 | class Config: | |
4845 | own_args = None | |
4846 | parsed_args = None | |
4847 | filename = None | |
4848 | __initialized = False | |
4849 | ||
4850 | def __init__(self, parser, label=None): | |
4851 | self.parser, self.label = parser, label | |
4852 | self._loaded_paths, self.configs = set(), [] | |
4853 | ||
4854 | def init(self, args=None, filename=None): | |
4855 | assert not self.__initialized | |
4856 | self.own_args, self.filename = args, filename | |
4857 | return self.load_configs() | |
4858 | ||
4859 | def load_configs(self): | |
4860 | directory = '' | |
4861 | if self.filename: | |
4862 | location = os.path.realpath(self.filename) | |
4863 | directory = os.path.dirname(location) | |
4864 | if location in self._loaded_paths: | |
4865 | return False | |
4866 | self._loaded_paths.add(location) | |
4867 | ||
4868 | self.__initialized = True | |
4869 | opts, _ = self.parser.parse_known_args(self.own_args) | |
4870 | self.parsed_args = self.own_args | |
4871 | for location in opts.config_locations or []: | |
4872 | if location == '-': | |
4873 | if location in self._loaded_paths: | |
4874 | continue | |
4875 | self._loaded_paths.add(location) | |
4876 | self.append_config(shlex.split(read_stdin('options'), comments=True), label='stdin') | |
4877 | continue | |
4878 | location = os.path.join(directory, expand_path(location)) | |
4879 | if os.path.isdir(location): | |
4880 | location = os.path.join(location, 'yt-dlp.conf') | |
4881 | if not os.path.exists(location): | |
4882 | self.parser.error(f'config location {location} does not exist') | |
4883 | self.append_config(self.read_file(location), location) | |
4884 | return True | |
4885 | ||
4886 | def __str__(self): | |
4887 | label = join_nonempty( | |
4888 | self.label, 'config', f'"{self.filename}"' if self.filename else '', | |
4889 | delim=' ') | |
4890 | return join_nonempty( | |
4891 | self.own_args is not None and f'{label[0].upper()}{label[1:]}: {self.hide_login_info(self.own_args)}', | |
4892 | *(f'\n{c}'.replace('\n', '\n| ')[1:] for c in self.configs), | |
4893 | delim='\n') | |
4894 | ||
4895 | @staticmethod | |
4896 | def read_file(filename, default=[]): | |
4897 | try: | |
4898 | optionf = open(filename, 'rb') | |
4899 | except OSError: | |
4900 | return default # silently skip if file is not present | |
4901 | try: | |
4902 | enc, skip = determine_file_encoding(optionf.read(512)) | |
4903 | optionf.seek(skip, io.SEEK_SET) | |
4904 | except OSError: | |
4905 | enc = None # silently skip read errors | |
4906 | try: | |
4907 | # FIXME: https://github.com/ytdl-org/youtube-dl/commit/dfe5fa49aed02cf36ba9f743b11b0903554b5e56 | |
4908 | contents = optionf.read().decode(enc or preferredencoding()) | |
4909 | res = shlex.split(contents, comments=True) | |
4910 | except Exception as err: | |
4911 | raise ValueError(f'Unable to parse "{filename}": {err}') | |
4912 | finally: | |
4913 | optionf.close() | |
4914 | return res | |
4915 | ||
4916 | @staticmethod | |
4917 | def hide_login_info(opts): | |
4918 | PRIVATE_OPTS = {'-p', '--password', '-u', '--username', '--video-password', '--ap-password', '--ap-username'} | |
4919 | eqre = re.compile('^(?P<key>' + ('|'.join(re.escape(po) for po in PRIVATE_OPTS)) + ')=.+$') | |
4920 | ||
4921 | def _scrub_eq(o): | |
4922 | m = eqre.match(o) | |
4923 | if m: | |
4924 | return m.group('key') + '=PRIVATE' | |
4925 | else: | |
4926 | return o | |
4927 | ||
4928 | opts = list(map(_scrub_eq, opts)) | |
4929 | for idx, opt in enumerate(opts): | |
4930 | if opt in PRIVATE_OPTS and idx + 1 < len(opts): | |
4931 | opts[idx + 1] = 'PRIVATE' | |
4932 | return opts | |
4933 | ||
4934 | def append_config(self, *args, label=None): | |
4935 | config = type(self)(self.parser, label) | |
4936 | config._loaded_paths = self._loaded_paths | |
4937 | if config.init(*args): | |
4938 | self.configs.append(config) | |
4939 | ||
4940 | @property | |
4941 | def all_args(self): | |
4942 | for config in reversed(self.configs): | |
4943 | yield from config.all_args | |
4944 | yield from self.parsed_args or [] | |
4945 | ||
4946 | def parse_known_args(self, **kwargs): | |
4947 | return self.parser.parse_known_args(self.all_args, **kwargs) | |
4948 | ||
4949 | def parse_args(self): | |
4950 | return self.parser.parse_args(self.all_args) | |
4951 | ||
4952 | ||
4953 | def merge_headers(*dicts): | |
4954 | """Merge dicts of http headers case insensitively, prioritizing the latter ones""" | |
4955 | return {k.title(): v for k, v in itertools.chain.from_iterable(map(dict.items, dicts))} | |
4956 | ||
4957 | ||
4958 | def cached_method(f): | |
4959 | """Cache a method""" | |
4960 | signature = inspect.signature(f) | |
4961 | ||
4962 | @functools.wraps(f) | |
4963 | def wrapper(self, *args, **kwargs): | |
4964 | bound_args = signature.bind(self, *args, **kwargs) | |
4965 | bound_args.apply_defaults() | |
4966 | key = tuple(bound_args.arguments.values())[1:] | |
4967 | ||
4968 | cache = vars(self).setdefault('_cached_method__cache', {}).setdefault(f.__name__, {}) | |
4969 | if key not in cache: | |
4970 | cache[key] = f(self, *args, **kwargs) | |
4971 | return cache[key] | |
4972 | return wrapper | |
4973 | ||
4974 | ||
4975 | class classproperty: | |
4976 | """property access for class methods with optional caching""" | |
4977 | def __new__(cls, func=None, *args, **kwargs): | |
4978 | if not func: | |
4979 | return functools.partial(cls, *args, **kwargs) | |
4980 | return super().__new__(cls) | |
4981 | ||
4982 | def __init__(self, func, *, cache=False): | |
4983 | functools.update_wrapper(self, func) | |
4984 | self.func = func | |
4985 | self._cache = {} if cache else None | |
4986 | ||
4987 | def __get__(self, _, cls): | |
4988 | if self._cache is None: | |
4989 | return self.func(cls) | |
4990 | elif cls not in self._cache: | |
4991 | self._cache[cls] = self.func(cls) | |
4992 | return self._cache[cls] | |
4993 | ||
4994 | ||
4995 | class function_with_repr: | |
4996 | def __init__(self, func, repr_=None): | |
4997 | functools.update_wrapper(self, func) | |
4998 | self.func, self.__repr = func, repr_ | |
4999 | ||
5000 | def __call__(self, *args, **kwargs): | |
5001 | return self.func(*args, **kwargs) | |
5002 | ||
5003 | @classmethod | |
5004 | def set_repr(cls, repr_): | |
5005 | return functools.partial(cls, repr_=repr_) | |
5006 | ||
5007 | def __repr__(self): | |
5008 | if self.__repr: | |
5009 | return self.__repr | |
5010 | return f'{self.func.__module__}.{self.func.__qualname__}' | |
5011 | ||
5012 | ||
5013 | class Namespace(types.SimpleNamespace): | |
5014 | """Immutable namespace""" | |
5015 | ||
5016 | def __iter__(self): | |
5017 | return iter(self.__dict__.values()) | |
5018 | ||
5019 | @property | |
5020 | def items_(self): | |
5021 | return self.__dict__.items() | |
5022 | ||
5023 | ||
5024 | MEDIA_EXTENSIONS = Namespace( | |
5025 | common_video=('avi', 'flv', 'mkv', 'mov', 'mp4', 'webm'), | |
5026 | video=('3g2', '3gp', 'f4v', 'mk3d', 'divx', 'mpg', 'ogv', 'm4v', 'wmv'), | |
5027 | common_audio=('aiff', 'alac', 'flac', 'm4a', 'mka', 'mp3', 'ogg', 'opus', 'wav'), | |
5028 | audio=('aac', 'ape', 'asf', 'f4a', 'f4b', 'm4b', 'm4p', 'm4r', 'oga', 'ogx', 'spx', 'vorbis', 'wma', 'weba'), | |
5029 | thumbnails=('jpg', 'png', 'webp'), | |
5030 | storyboards=('mhtml', ), | |
5031 | subtitles=('srt', 'vtt', 'ass', 'lrc'), | |
5032 | manifests=('f4f', 'f4m', 'm3u8', 'smil', 'mpd'), | |
5033 | ) | |
5034 | MEDIA_EXTENSIONS.video += MEDIA_EXTENSIONS.common_video | |
5035 | MEDIA_EXTENSIONS.audio += MEDIA_EXTENSIONS.common_audio | |
5036 | ||
5037 | KNOWN_EXTENSIONS = (*MEDIA_EXTENSIONS.video, *MEDIA_EXTENSIONS.audio, *MEDIA_EXTENSIONS.manifests) | |
5038 | ||
5039 | ||
5040 | class RetryManager: | |
5041 | """Usage: | |
5042 | for retry in RetryManager(...): | |
5043 | try: | |
5044 | ... | |
5045 | except SomeException as err: | |
5046 | retry.error = err | |
5047 | continue | |
5048 | """ | |
5049 | attempt, _error = 0, None | |
5050 | ||
5051 | def __init__(self, _retries, _error_callback, **kwargs): | |
5052 | self.retries = _retries or 0 | |
5053 | self.error_callback = functools.partial(_error_callback, **kwargs) | |
5054 | ||
5055 | def _should_retry(self): | |
5056 | return self._error is not NO_DEFAULT and self.attempt <= self.retries | |
5057 | ||
5058 | @property | |
5059 | def error(self): | |
5060 | if self._error is NO_DEFAULT: | |
5061 | return None | |
5062 | return self._error | |
5063 | ||
5064 | @error.setter | |
5065 | def error(self, value): | |
5066 | self._error = value | |
5067 | ||
5068 | def __iter__(self): | |
5069 | while self._should_retry(): | |
5070 | self.error = NO_DEFAULT | |
5071 | self.attempt += 1 | |
5072 | yield self | |
5073 | if self.error: | |
5074 | self.error_callback(self.error, self.attempt, self.retries) | |
5075 | ||
5076 | @staticmethod | |
5077 | def report_retry(e, count, retries, *, sleep_func, info, warn, error=None, suffix=None): | |
5078 | """Utility function for reporting retries""" | |
5079 | if count > retries: | |
5080 | if error: | |
5081 | return error(f'{e}. Giving up after {count - 1} retries') if count > 1 else error(str(e)) | |
5082 | raise e | |
5083 | ||
5084 | if not count: | |
5085 | return warn(e) | |
5086 | elif isinstance(e, ExtractorError): | |
5087 | e = remove_end(str_or_none(e.cause) or e.orig_msg, '.') | |
5088 | warn(f'{e}. Retrying{format_field(suffix, None, " %s")} ({count}/{retries})...') | |
5089 | ||
5090 | delay = float_or_none(sleep_func(n=count - 1)) if callable(sleep_func) else sleep_func | |
5091 | if delay: | |
5092 | info(f'Sleeping {delay:.2f} seconds ...') | |
5093 | time.sleep(delay) | |
5094 | ||
5095 | ||
5096 | def make_archive_id(ie, video_id): | |
5097 | ie_key = ie if isinstance(ie, str) else ie.ie_key() | |
5098 | return f'{ie_key.lower()} {video_id}' | |
5099 | ||
5100 | ||
5101 | def truncate_string(s, left, right=0): | |
5102 | assert left > 3 and right >= 0 | |
5103 | if s is None or len(s) <= left + right: | |
5104 | return s | |
5105 | return f'{s[:left - 3]}...{s[-right:] if right else ""}' | |
5106 | ||
5107 | ||
5108 | def orderedSet_from_options(options, alias_dict, *, use_regex=False, start=None): | |
5109 | assert 'all' in alias_dict, '"all" alias is required' | |
5110 | requested = list(start or []) | |
5111 | for val in options: | |
5112 | discard = val.startswith('-') | |
5113 | if discard: | |
5114 | val = val[1:] | |
5115 | ||
5116 | if val in alias_dict: | |
5117 | val = alias_dict[val] if not discard else [ | |
5118 | i[1:] if i.startswith('-') else f'-{i}' for i in alias_dict[val]] | |
5119 | # NB: Do not allow regex in aliases for performance | |
5120 | requested = orderedSet_from_options(val, alias_dict, start=requested) | |
5121 | continue | |
5122 | ||
5123 | current = (filter(re.compile(val, re.I).fullmatch, alias_dict['all']) if use_regex | |
5124 | else [val] if val in alias_dict['all'] else None) | |
5125 | if current is None: | |
5126 | raise ValueError(val) | |
5127 | ||
5128 | if discard: | |
5129 | for item in current: | |
5130 | while item in requested: | |
5131 | requested.remove(item) | |
5132 | else: | |
5133 | requested.extend(current) | |
5134 | ||
5135 | return orderedSet(requested) | |
5136 | ||
5137 | ||
5138 | # TODO: Rewrite | |
5139 | class FormatSorter: | |
5140 | regex = r' *((?P<reverse>\+)?(?P<field>[a-zA-Z0-9_]+)((?P<separator>[~:])(?P<limit>.*?))?)? *$' | |
5141 | ||
5142 | default = ('hidden', 'aud_or_vid', 'hasvid', 'ie_pref', 'lang', 'quality', | |
5143 | 'res', 'fps', 'hdr:12', 'vcodec:vp9.2', 'channels', 'acodec', | |
5144 | 'size', 'br', 'asr', 'proto', 'ext', 'hasaud', 'source', 'id') # These must not be aliases | |
5145 | ytdl_default = ('hasaud', 'lang', 'quality', 'tbr', 'filesize', 'vbr', | |
5146 | 'height', 'width', 'proto', 'vext', 'abr', 'aext', | |
5147 | 'fps', 'fs_approx', 'source', 'id') | |
5148 | ||
5149 | settings = { | |
5150 | 'vcodec': {'type': 'ordered', 'regex': True, | |
5151 | 'order': ['av0?1', 'vp0?9.2', 'vp0?9', '[hx]265|he?vc?', '[hx]264|avc', 'vp0?8', 'mp4v|h263', 'theora', '', None, 'none']}, | |
5152 | 'acodec': {'type': 'ordered', 'regex': True, | |
5153 | 'order': ['[af]lac', 'wav|aiff', 'opus', 'vorbis|ogg', 'aac', 'mp?4a?', 'mp3', 'ac-?4', 'e-?a?c-?3', 'ac-?3', 'dts', '', None, 'none']}, | |
5154 | 'hdr': {'type': 'ordered', 'regex': True, 'field': 'dynamic_range', | |
5155 | 'order': ['dv', '(hdr)?12', r'(hdr)?10\+', '(hdr)?10', 'hlg', '', 'sdr', None]}, | |
5156 | 'proto': {'type': 'ordered', 'regex': True, 'field': 'protocol', | |
5157 | 'order': ['(ht|f)tps', '(ht|f)tp$', 'm3u8.*', '.*dash', 'websocket_frag', 'rtmpe?', '', 'mms|rtsp', 'ws|websocket', 'f4']}, | |
5158 | 'vext': {'type': 'ordered', 'field': 'video_ext', | |
5159 | 'order': ('mp4', 'mov', 'webm', 'flv', '', 'none'), | |
5160 | 'order_free': ('webm', 'mp4', 'mov', 'flv', '', 'none')}, | |
5161 | 'aext': {'type': 'ordered', 'regex': True, 'field': 'audio_ext', | |
5162 | 'order': ('m4a', 'aac', 'mp3', 'ogg', 'opus', 'web[am]', '', 'none'), | |
5163 | 'order_free': ('ogg', 'opus', 'web[am]', 'mp3', 'm4a', 'aac', '', 'none')}, | |
5164 | 'hidden': {'visible': False, 'forced': True, 'type': 'extractor', 'max': -1000}, | |
5165 | 'aud_or_vid': {'visible': False, 'forced': True, 'type': 'multiple', | |
5166 | 'field': ('vcodec', 'acodec'), | |
5167 | 'function': lambda it: int(any(v != 'none' for v in it))}, | |
5168 | 'ie_pref': {'priority': True, 'type': 'extractor'}, | |
5169 | 'hasvid': {'priority': True, 'field': 'vcodec', 'type': 'boolean', 'not_in_list': ('none',)}, | |
5170 | 'hasaud': {'field': 'acodec', 'type': 'boolean', 'not_in_list': ('none',)}, | |
5171 | 'lang': {'convert': 'float', 'field': 'language_preference', 'default': -1}, | |
5172 | 'quality': {'convert': 'float', 'default': -1}, | |
5173 | 'filesize': {'convert': 'bytes'}, | |
5174 | 'fs_approx': {'convert': 'bytes', 'field': 'filesize_approx'}, | |
5175 | 'id': {'convert': 'string', 'field': 'format_id'}, | |
5176 | 'height': {'convert': 'float_none'}, | |
5177 | 'width': {'convert': 'float_none'}, | |
5178 | 'fps': {'convert': 'float_none'}, | |
5179 | 'channels': {'convert': 'float_none', 'field': 'audio_channels'}, | |
5180 | 'tbr': {'convert': 'float_none'}, | |
5181 | 'vbr': {'convert': 'float_none'}, | |
5182 | 'abr': {'convert': 'float_none'}, | |
5183 | 'asr': {'convert': 'float_none'}, | |
5184 | 'source': {'convert': 'float', 'field': 'source_preference', 'default': -1}, | |
5185 | ||
5186 | 'codec': {'type': 'combined', 'field': ('vcodec', 'acodec')}, | |
5187 | 'br': {'type': 'multiple', 'field': ('tbr', 'vbr', 'abr'), 'convert': 'float_none', | |
5188 | 'function': lambda it: next(filter(None, it), None)}, | |
5189 | 'size': {'type': 'multiple', 'field': ('filesize', 'fs_approx'), 'convert': 'bytes', | |
5190 | 'function': lambda it: next(filter(None, it), None)}, | |
5191 | 'ext': {'type': 'combined', 'field': ('vext', 'aext')}, | |
5192 | 'res': {'type': 'multiple', 'field': ('height', 'width'), | |
5193 | 'function': lambda it: (lambda l: min(l) if l else 0)(tuple(filter(None, it)))}, | |
5194 | ||
5195 | # Actual field names | |
5196 | 'format_id': {'type': 'alias', 'field': 'id'}, | |
5197 | 'preference': {'type': 'alias', 'field': 'ie_pref'}, | |
5198 | 'language_preference': {'type': 'alias', 'field': 'lang'}, | |
5199 | 'source_preference': {'type': 'alias', 'field': 'source'}, | |
5200 | 'protocol': {'type': 'alias', 'field': 'proto'}, | |
5201 | 'filesize_approx': {'type': 'alias', 'field': 'fs_approx'}, | |
5202 | 'audio_channels': {'type': 'alias', 'field': 'channels'}, | |
5203 | ||
5204 | # Deprecated | |
5205 | 'dimension': {'type': 'alias', 'field': 'res', 'deprecated': True}, | |
5206 | 'resolution': {'type': 'alias', 'field': 'res', 'deprecated': True}, | |
5207 | 'extension': {'type': 'alias', 'field': 'ext', 'deprecated': True}, | |
5208 | 'bitrate': {'type': 'alias', 'field': 'br', 'deprecated': True}, | |
5209 | 'total_bitrate': {'type': 'alias', 'field': 'tbr', 'deprecated': True}, | |
5210 | 'video_bitrate': {'type': 'alias', 'field': 'vbr', 'deprecated': True}, | |
5211 | 'audio_bitrate': {'type': 'alias', 'field': 'abr', 'deprecated': True}, | |
5212 | 'framerate': {'type': 'alias', 'field': 'fps', 'deprecated': True}, | |
5213 | 'filesize_estimate': {'type': 'alias', 'field': 'size', 'deprecated': True}, | |
5214 | 'samplerate': {'type': 'alias', 'field': 'asr', 'deprecated': True}, | |
5215 | 'video_ext': {'type': 'alias', 'field': 'vext', 'deprecated': True}, | |
5216 | 'audio_ext': {'type': 'alias', 'field': 'aext', 'deprecated': True}, | |
5217 | 'video_codec': {'type': 'alias', 'field': 'vcodec', 'deprecated': True}, | |
5218 | 'audio_codec': {'type': 'alias', 'field': 'acodec', 'deprecated': True}, | |
5219 | 'video': {'type': 'alias', 'field': 'hasvid', 'deprecated': True}, | |
5220 | 'has_video': {'type': 'alias', 'field': 'hasvid', 'deprecated': True}, | |
5221 | 'audio': {'type': 'alias', 'field': 'hasaud', 'deprecated': True}, | |
5222 | 'has_audio': {'type': 'alias', 'field': 'hasaud', 'deprecated': True}, | |
5223 | 'extractor': {'type': 'alias', 'field': 'ie_pref', 'deprecated': True}, | |
5224 | 'extractor_preference': {'type': 'alias', 'field': 'ie_pref', 'deprecated': True}, | |
5225 | } | |
5226 | ||
5227 | def __init__(self, ydl, field_preference): | |
5228 | self.ydl = ydl | |
5229 | self._order = [] | |
5230 | self.evaluate_params(self.ydl.params, field_preference) | |
5231 | if ydl.params.get('verbose'): | |
5232 | self.print_verbose_info(self.ydl.write_debug) | |
5233 | ||
5234 | def _get_field_setting(self, field, key): | |
5235 | if field not in self.settings: | |
5236 | if key in ('forced', 'priority'): | |
5237 | return False | |
5238 | self.ydl.deprecated_feature(f'Using arbitrary fields ({field}) for format sorting is ' | |
5239 | 'deprecated and may be removed in a future version') | |
5240 | self.settings[field] = {} | |
5241 | propObj = self.settings[field] | |
5242 | if key not in propObj: | |
5243 | type = propObj.get('type') | |
5244 | if key == 'field': | |
5245 | default = 'preference' if type == 'extractor' else (field,) if type in ('combined', 'multiple') else field | |
5246 | elif key == 'convert': | |
5247 | default = 'order' if type == 'ordered' else 'float_string' if field else 'ignore' | |
5248 | else: | |
5249 | default = {'type': 'field', 'visible': True, 'order': [], 'not_in_list': (None,)}.get(key, None) | |
5250 | propObj[key] = default | |
5251 | return propObj[key] | |
5252 | ||
5253 | def _resolve_field_value(self, field, value, convertNone=False): | |
5254 | if value is None: | |
5255 | if not convertNone: | |
5256 | return None | |
5257 | else: | |
5258 | value = value.lower() | |
5259 | conversion = self._get_field_setting(field, 'convert') | |
5260 | if conversion == 'ignore': | |
5261 | return None | |
5262 | if conversion == 'string': | |
5263 | return value | |
5264 | elif conversion == 'float_none': | |
5265 | return float_or_none(value) | |
5266 | elif conversion == 'bytes': | |
5267 | return parse_bytes(value) | |
5268 | elif conversion == 'order': | |
5269 | order_list = (self._use_free_order and self._get_field_setting(field, 'order_free')) or self._get_field_setting(field, 'order') | |
5270 | use_regex = self._get_field_setting(field, 'regex') | |
5271 | list_length = len(order_list) | |
5272 | empty_pos = order_list.index('') if '' in order_list else list_length + 1 | |
5273 | if use_regex and value is not None: | |
5274 | for i, regex in enumerate(order_list): | |
5275 | if regex and re.match(regex, value): | |
5276 | return list_length - i | |
5277 | return list_length - empty_pos # not in list | |
5278 | else: # not regex or value = None | |
5279 | return list_length - (order_list.index(value) if value in order_list else empty_pos) | |
5280 | else: | |
5281 | if value.isnumeric(): | |
5282 | return float(value) | |
5283 | else: | |
5284 | self.settings[field]['convert'] = 'string' | |
5285 | return value | |
5286 | ||
5287 | def evaluate_params(self, params, sort_extractor): | |
5288 | self._use_free_order = params.get('prefer_free_formats', False) | |
5289 | self._sort_user = params.get('format_sort', []) | |
5290 | self._sort_extractor = sort_extractor | |
5291 | ||
5292 | def add_item(field, reverse, closest, limit_text): | |
5293 | field = field.lower() | |
5294 | if field in self._order: | |
5295 | return | |
5296 | self._order.append(field) | |
5297 | limit = self._resolve_field_value(field, limit_text) | |
5298 | data = { | |
5299 | 'reverse': reverse, | |
5300 | 'closest': False if limit is None else closest, | |
5301 | 'limit_text': limit_text, | |
5302 | 'limit': limit} | |
5303 | if field in self.settings: | |
5304 | self.settings[field].update(data) | |
5305 | else: | |
5306 | self.settings[field] = data | |
5307 | ||
5308 | sort_list = ( | |
5309 | tuple(field for field in self.default if self._get_field_setting(field, 'forced')) | |
5310 | + (tuple() if params.get('format_sort_force', False) | |
5311 | else tuple(field for field in self.default if self._get_field_setting(field, 'priority'))) | |
5312 | + tuple(self._sort_user) + tuple(sort_extractor) + self.default) | |
5313 | ||
5314 | for item in sort_list: | |
5315 | match = re.match(self.regex, item) | |
5316 | if match is None: | |
5317 | raise ExtractorError('Invalid format sort string "%s" given by extractor' % item) | |
5318 | field = match.group('field') | |
5319 | if field is None: | |
5320 | continue | |
5321 | if self._get_field_setting(field, 'type') == 'alias': | |
5322 | alias, field = field, self._get_field_setting(field, 'field') | |
5323 | if self._get_field_setting(alias, 'deprecated'): | |
5324 | self.ydl.deprecated_feature(f'Format sorting alias {alias} is deprecated and may ' | |
5325 | f'be removed in a future version. Please use {field} instead') | |
5326 | reverse = match.group('reverse') is not None | |
5327 | closest = match.group('separator') == '~' | |
5328 | limit_text = match.group('limit') | |
5329 | ||
5330 | has_limit = limit_text is not None | |
5331 | has_multiple_fields = self._get_field_setting(field, 'type') == 'combined' | |
5332 | has_multiple_limits = has_limit and has_multiple_fields and not self._get_field_setting(field, 'same_limit') | |
5333 | ||
5334 | fields = self._get_field_setting(field, 'field') if has_multiple_fields else (field,) | |
5335 | limits = limit_text.split(':') if has_multiple_limits else (limit_text,) if has_limit else tuple() | |
5336 | limit_count = len(limits) | |
5337 | for (i, f) in enumerate(fields): | |
5338 | add_item(f, reverse, closest, | |
5339 | limits[i] if i < limit_count | |
5340 | else limits[0] if has_limit and not has_multiple_limits | |
5341 | else None) | |
5342 | ||
5343 | def print_verbose_info(self, write_debug): | |
5344 | if self._sort_user: | |
5345 | write_debug('Sort order given by user: %s' % ', '.join(self._sort_user)) | |
5346 | if self._sort_extractor: | |
5347 | write_debug('Sort order given by extractor: %s' % ', '.join(self._sort_extractor)) | |
5348 | write_debug('Formats sorted by: %s' % ', '.join(['%s%s%s' % ( | |
5349 | '+' if self._get_field_setting(field, 'reverse') else '', field, | |
5350 | '%s%s(%s)' % ('~' if self._get_field_setting(field, 'closest') else ':', | |
5351 | self._get_field_setting(field, 'limit_text'), | |
5352 | self._get_field_setting(field, 'limit')) | |
5353 | if self._get_field_setting(field, 'limit_text') is not None else '') | |
5354 | for field in self._order if self._get_field_setting(field, 'visible')])) | |
5355 | ||
5356 | def _calculate_field_preference_from_value(self, format, field, type, value): | |
5357 | reverse = self._get_field_setting(field, 'reverse') | |
5358 | closest = self._get_field_setting(field, 'closest') | |
5359 | limit = self._get_field_setting(field, 'limit') | |
5360 | ||
5361 | if type == 'extractor': | |
5362 | maximum = self._get_field_setting(field, 'max') | |
5363 | if value is None or (maximum is not None and value >= maximum): | |
5364 | value = -1 | |
5365 | elif type == 'boolean': | |
5366 | in_list = self._get_field_setting(field, 'in_list') | |
5367 | not_in_list = self._get_field_setting(field, 'not_in_list') | |
5368 | value = 0 if ((in_list is None or value in in_list) and (not_in_list is None or value not in not_in_list)) else -1 | |
5369 | elif type == 'ordered': | |
5370 | value = self._resolve_field_value(field, value, True) | |
5371 | ||
5372 | # try to convert to number | |
5373 | val_num = float_or_none(value, default=self._get_field_setting(field, 'default')) | |
5374 | is_num = self._get_field_setting(field, 'convert') != 'string' and val_num is not None | |
5375 | if is_num: | |
5376 | value = val_num | |
5377 | ||
5378 | return ((-10, 0) if value is None | |
5379 | else (1, value, 0) if not is_num # if a field has mixed strings and numbers, strings are sorted higher | |
5380 | else (0, -abs(value - limit), value - limit if reverse else limit - value) if closest | |
5381 | else (0, value, 0) if not reverse and (limit is None or value <= limit) | |
5382 | else (0, -value, 0) if limit is None or (reverse and value == limit) or value > limit | |
5383 | else (-1, value, 0)) | |
5384 | ||
5385 | def _calculate_field_preference(self, format, field): | |
5386 | type = self._get_field_setting(field, 'type') # extractor, boolean, ordered, field, multiple | |
5387 | get_value = lambda f: format.get(self._get_field_setting(f, 'field')) | |
5388 | if type == 'multiple': | |
5389 | type = 'field' # Only 'field' is allowed in multiple for now | |
5390 | actual_fields = self._get_field_setting(field, 'field') | |
5391 | ||
5392 | value = self._get_field_setting(field, 'function')(get_value(f) for f in actual_fields) | |
5393 | else: | |
5394 | value = get_value(field) | |
5395 | return self._calculate_field_preference_from_value(format, field, type, value) | |
5396 | ||
5397 | def calculate_preference(self, format): | |
5398 | # Determine missing protocol | |
5399 | if not format.get('protocol'): | |
5400 | format['protocol'] = determine_protocol(format) | |
5401 | ||
5402 | # Determine missing ext | |
5403 | if not format.get('ext') and 'url' in format: | |
5404 | format['ext'] = determine_ext(format['url']) | |
5405 | if format.get('vcodec') == 'none': | |
5406 | format['audio_ext'] = format['ext'] if format.get('acodec') != 'none' else 'none' | |
5407 | format['video_ext'] = 'none' | |
5408 | else: | |
5409 | format['video_ext'] = format['ext'] | |
5410 | format['audio_ext'] = 'none' | |
5411 | # if format.get('preference') is None and format.get('ext') in ('f4f', 'f4m'): # Not supported? | |
5412 | # format['preference'] = -1000 | |
5413 | ||
5414 | if format.get('preference') is None and format.get('ext') == 'flv' and re.match('[hx]265|he?vc?', format.get('vcodec') or ''): | |
5415 | # HEVC-over-FLV is out-of-spec by FLV's original spec | |
5416 | # ref. https://trac.ffmpeg.org/ticket/6389 | |
5417 | # ref. https://github.com/yt-dlp/yt-dlp/pull/5821 | |
5418 | format['preference'] = -100 | |
5419 | ||
5420 | # Determine missing bitrates | |
5421 | if format.get('vcodec') == 'none': | |
5422 | format['vbr'] = 0 | |
5423 | if format.get('acodec') == 'none': | |
5424 | format['abr'] = 0 | |
5425 | if not format.get('vbr') and format.get('vcodec') != 'none': | |
5426 | format['vbr'] = try_call(lambda: format['tbr'] - format['abr']) or None | |
5427 | if not format.get('abr') and format.get('acodec') != 'none': | |
5428 | format['abr'] = try_call(lambda: format['tbr'] - format['vbr']) or None | |
5429 | if not format.get('tbr'): | |
5430 | format['tbr'] = try_call(lambda: format['vbr'] + format['abr']) or None | |
5431 | ||
5432 | return tuple(self._calculate_field_preference(format, field) for field in self._order) | |
5433 | ||
5434 | ||
5435 | def filesize_from_tbr(tbr, duration): | |
5436 | """ | |
5437 | @param tbr: Total bitrate in kbps (1000 bits/sec) | |
5438 | @param duration: Duration in seconds | |
5439 | @returns Filesize in bytes | |
5440 | """ | |
5441 | if tbr is None or duration is None: | |
5442 | return None | |
5443 | return int(duration * tbr * (1000 / 8)) | |
5444 | ||
5445 | ||
5446 | # XXX: Temporary | |
5447 | class _YDLLogger: | |
5448 | def __init__(self, ydl=None): | |
5449 | self._ydl = ydl | |
5450 | ||
5451 | def debug(self, message): | |
5452 | if self._ydl: | |
5453 | self._ydl.write_debug(message) | |
5454 | ||
5455 | def info(self, message): | |
5456 | if self._ydl: | |
5457 | self._ydl.to_screen(message) | |
5458 | ||
5459 | def warning(self, message, *, once=False): | |
5460 | if self._ydl: | |
5461 | self._ydl.report_warning(message, once) | |
5462 | ||
5463 | def error(self, message, *, is_error=True): | |
5464 | if self._ydl: | |
5465 | self._ydl.report_error(message, is_error=is_error) | |
5466 | ||
5467 | def stdout(self, message): | |
5468 | if self._ydl: | |
5469 | self._ydl.to_stdout(message) | |
5470 | ||
5471 | def stderr(self, message): | |
5472 | if self._ydl: | |
5473 | self._ydl.to_stderr(message) |