]>
Commit | Line | Data |
---|---|---|
d77c3dfd | 1 | #!/usr/bin/env python |
dcdb292f | 2 | # coding: utf-8 |
d77c3dfd | 3 | |
ecc0c5ee PH |
4 | from __future__ import unicode_literals |
5 | ||
1e399778 | 6 | import base64 |
5bc880b9 | 7 | import binascii |
912b38b4 | 8 | import calendar |
676eb3f2 | 9 | import codecs |
62e609ab | 10 | import contextlib |
e3946f98 | 11 | import ctypes |
c496ca96 PH |
12 | import datetime |
13 | import email.utils | |
0c265486 | 14 | import email.header |
f45c185f | 15 | import errno |
be4a824d | 16 | import functools |
d77c3dfd | 17 | import gzip |
03f9daab | 18 | import io |
79a2e94e | 19 | import itertools |
f4bfd65f | 20 | import json |
d77c3dfd | 21 | import locale |
02dbf93f | 22 | import math |
347de493 | 23 | import operator |
d77c3dfd | 24 | import os |
c496ca96 | 25 | import platform |
773f291d | 26 | import random |
d77c3dfd | 27 | import re |
c496ca96 | 28 | import socket |
79a2e94e | 29 | import ssl |
1c088fa8 | 30 | import subprocess |
d77c3dfd | 31 | import sys |
181c8655 | 32 | import tempfile |
01951dda | 33 | import traceback |
bcf89ce6 | 34 | import xml.etree.ElementTree |
d77c3dfd | 35 | import zlib |
d77c3dfd | 36 | |
8c25f81b | 37 | from .compat import ( |
b4a3d461 | 38 | compat_HTMLParseError, |
8bb56eee | 39 | compat_HTMLParser, |
8f9312c3 | 40 | compat_basestring, |
8c25f81b | 41 | compat_chr, |
d7cd9a9e | 42 | compat_ctypes_WINFUNCTYPE, |
36e6f62c | 43 | compat_etree_fromstring, |
51098426 | 44 | compat_expanduser, |
8c25f81b | 45 | compat_html_entities, |
55b2f099 | 46 | compat_html_entities_html5, |
be4a824d | 47 | compat_http_client, |
c86b6142 | 48 | compat_kwargs, |
efa97bdc | 49 | compat_os_name, |
8c25f81b | 50 | compat_parse_qs, |
702ccf2d | 51 | compat_shlex_quote, |
be4a824d | 52 | compat_socket_create_connection, |
8c25f81b | 53 | compat_str, |
edaa23f8 | 54 | compat_struct_pack, |
d3f8e038 | 55 | compat_struct_unpack, |
8c25f81b PH |
56 | compat_urllib_error, |
57 | compat_urllib_parse, | |
15707c7e | 58 | compat_urllib_parse_urlencode, |
8c25f81b | 59 | compat_urllib_parse_urlparse, |
7581bfc9 | 60 | compat_urllib_parse_unquote_plus, |
8c25f81b PH |
61 | compat_urllib_request, |
62 | compat_urlparse, | |
810c10ba | 63 | compat_xpath, |
8c25f81b | 64 | ) |
4644ac55 | 65 | |
71aff188 YCH |
66 | from .socks import ( |
67 | ProxyType, | |
68 | sockssocket, | |
69 | ) | |
70 | ||
4644ac55 | 71 | |
51fb4995 YCH |
72 | def register_socks_protocols(): |
73 | # "Register" SOCKS protocols | |
d5ae6bb5 YCH |
74 | # In Python < 2.6.5, urlsplit() suffers from bug https://bugs.python.org/issue7904 |
75 | # URLs with protocols not in urlparse.uses_netloc are not handled correctly | |
51fb4995 YCH |
76 | for scheme in ('socks', 'socks4', 'socks4a', 'socks5'): |
77 | if scheme not in compat_urlparse.uses_netloc: | |
78 | compat_urlparse.uses_netloc.append(scheme) | |
79 | ||
80 | ||
468e2e92 FV |
81 | # This is not clearly defined otherwise |
82 | compiled_regex_type = type(re.compile('')) | |
83 | ||
3e669f36 | 84 | std_headers = { |
b12cf31b | 85 | 'User-Agent': 'Mozilla/5.0 (X11; Linux x86_64; rv:59.0) Gecko/20100101 Firefox/59.0 (Chrome)', |
59ae15a5 PH |
86 | 'Accept-Charset': 'ISO-8859-1,utf-8;q=0.7,*;q=0.7', |
87 | 'Accept': 'text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8', | |
88 | 'Accept-Encoding': 'gzip, deflate', | |
89 | 'Accept-Language': 'en-us,en;q=0.5', | |
3e669f36 | 90 | } |
f427df17 | 91 | |
5f6a1245 | 92 | |
fb37eb25 S |
93 | USER_AGENTS = { |
94 | 'Safari': 'Mozilla/5.0 (X11; Linux x86_64; rv:10.0) AppleWebKit/533.20.25 (KHTML, like Gecko) Version/5.0.4 Safari/533.20.27', | |
95 | } | |
96 | ||
97 | ||
bf42a990 S |
98 | NO_DEFAULT = object() |
99 | ||
7105440c YCH |
100 | ENGLISH_MONTH_NAMES = [ |
101 | 'January', 'February', 'March', 'April', 'May', 'June', | |
102 | 'July', 'August', 'September', 'October', 'November', 'December'] | |
103 | ||
f6717dec S |
104 | MONTH_NAMES = { |
105 | 'en': ENGLISH_MONTH_NAMES, | |
106 | 'fr': [ | |
3e4185c3 S |
107 | 'janvier', 'février', 'mars', 'avril', 'mai', 'juin', |
108 | 'juillet', 'août', 'septembre', 'octobre', 'novembre', 'décembre'], | |
f6717dec | 109 | } |
a942d6cb | 110 | |
a7aaa398 S |
111 | KNOWN_EXTENSIONS = ( |
112 | 'mp4', 'm4a', 'm4p', 'm4b', 'm4r', 'm4v', 'aac', | |
113 | 'flv', 'f4v', 'f4a', 'f4b', | |
114 | 'webm', 'ogg', 'ogv', 'oga', 'ogx', 'spx', 'opus', | |
115 | 'mkv', 'mka', 'mk3d', | |
116 | 'avi', 'divx', | |
117 | 'mov', | |
118 | 'asf', 'wmv', 'wma', | |
119 | '3gp', '3g2', | |
120 | 'mp3', | |
121 | 'flac', | |
122 | 'ape', | |
123 | 'wav', | |
124 | 'f4f', 'f4m', 'm3u8', 'smil') | |
125 | ||
c587cbb7 | 126 | # needed for sanitizing filenames in restricted mode |
c8827027 | 127 | ACCENT_CHARS = dict(zip('ÂÃÄÀÁÅÆÇÈÉÊËÌÍÎÏÐÑÒÓÔÕÖŐØŒÙÚÛÜŰÝÞßàáâãäåæçèéêëìíîïðñòóôõöőøœùúûüűýþÿ', |
128 | itertools.chain('AAAAAA', ['AE'], 'CEEEEIIIIDNOOOOOOO', ['OE'], 'UUUUUYP', ['ss'], | |
129 | 'aaaaaa', ['ae'], 'ceeeeiiiionooooooo', ['oe'], 'uuuuuypy'))) | |
c587cbb7 | 130 | |
46f59e89 S |
131 | DATE_FORMATS = ( |
132 | '%d %B %Y', | |
133 | '%d %b %Y', | |
134 | '%B %d %Y', | |
cb655f34 S |
135 | '%B %dst %Y', |
136 | '%B %dnd %Y', | |
137 | '%B %dth %Y', | |
46f59e89 | 138 | '%b %d %Y', |
cb655f34 S |
139 | '%b %dst %Y', |
140 | '%b %dnd %Y', | |
141 | '%b %dth %Y', | |
46f59e89 S |
142 | '%b %dst %Y %I:%M', |
143 | '%b %dnd %Y %I:%M', | |
144 | '%b %dth %Y %I:%M', | |
145 | '%Y %m %d', | |
146 | '%Y-%m-%d', | |
147 | '%Y/%m/%d', | |
81c13222 | 148 | '%Y/%m/%d %H:%M', |
46f59e89 | 149 | '%Y/%m/%d %H:%M:%S', |
0c1c6f4b | 150 | '%Y-%m-%d %H:%M', |
46f59e89 S |
151 | '%Y-%m-%d %H:%M:%S', |
152 | '%Y-%m-%d %H:%M:%S.%f', | |
153 | '%d.%m.%Y %H:%M', | |
154 | '%d.%m.%Y %H.%M', | |
155 | '%Y-%m-%dT%H:%M:%SZ', | |
156 | '%Y-%m-%dT%H:%M:%S.%fZ', | |
157 | '%Y-%m-%dT%H:%M:%S.%f0Z', | |
158 | '%Y-%m-%dT%H:%M:%S', | |
159 | '%Y-%m-%dT%H:%M:%S.%f', | |
160 | '%Y-%m-%dT%H:%M', | |
c6eed6b8 S |
161 | '%b %d %Y at %H:%M', |
162 | '%b %d %Y at %H:%M:%S', | |
b555ae9b S |
163 | '%B %d %Y at %H:%M', |
164 | '%B %d %Y at %H:%M:%S', | |
46f59e89 S |
165 | ) |
166 | ||
167 | DATE_FORMATS_DAY_FIRST = list(DATE_FORMATS) | |
168 | DATE_FORMATS_DAY_FIRST.extend([ | |
169 | '%d-%m-%Y', | |
170 | '%d.%m.%Y', | |
171 | '%d.%m.%y', | |
172 | '%d/%m/%Y', | |
173 | '%d/%m/%y', | |
174 | '%d/%m/%Y %H:%M:%S', | |
175 | ]) | |
176 | ||
177 | DATE_FORMATS_MONTH_FIRST = list(DATE_FORMATS) | |
178 | DATE_FORMATS_MONTH_FIRST.extend([ | |
179 | '%m-%d-%Y', | |
180 | '%m.%d.%Y', | |
181 | '%m/%d/%Y', | |
182 | '%m/%d/%y', | |
183 | '%m/%d/%Y %H:%M:%S', | |
184 | ]) | |
185 | ||
06b3fe29 S |
186 | PACKED_CODES_RE = r"}\('(.+)',(\d+),(\d+),'([^']+)'\.split\('\|'\)" |
187 | ||
7105440c | 188 | |
d77c3dfd | 189 | def preferredencoding(): |
59ae15a5 | 190 | """Get preferred encoding. |
d77c3dfd | 191 | |
59ae15a5 PH |
192 | Returns the best encoding scheme for the system, based on |
193 | locale.getpreferredencoding() and some further tweaks. | |
194 | """ | |
195 | try: | |
196 | pref = locale.getpreferredencoding() | |
28e614de | 197 | 'TEST'.encode(pref) |
70a1165b | 198 | except Exception: |
59ae15a5 | 199 | pref = 'UTF-8' |
bae611f2 | 200 | |
59ae15a5 | 201 | return pref |
d77c3dfd | 202 | |
f4bfd65f | 203 | |
181c8655 | 204 | def write_json_file(obj, fn): |
1394646a | 205 | """ Encode obj as JSON and write it to fn, atomically if possible """ |
181c8655 | 206 | |
92120217 | 207 | fn = encodeFilename(fn) |
61ee5aeb | 208 | if sys.version_info < (3, 0) and sys.platform != 'win32': |
ec5f6016 JMF |
209 | encoding = get_filesystem_encoding() |
210 | # os.path.basename returns a bytes object, but NamedTemporaryFile | |
211 | # will fail if the filename contains non ascii characters unless we | |
212 | # use a unicode object | |
213 | path_basename = lambda f: os.path.basename(fn).decode(encoding) | |
214 | # the same for os.path.dirname | |
215 | path_dirname = lambda f: os.path.dirname(fn).decode(encoding) | |
216 | else: | |
217 | path_basename = os.path.basename | |
218 | path_dirname = os.path.dirname | |
219 | ||
73159f99 S |
220 | args = { |
221 | 'suffix': '.tmp', | |
ec5f6016 JMF |
222 | 'prefix': path_basename(fn) + '.', |
223 | 'dir': path_dirname(fn), | |
73159f99 S |
224 | 'delete': False, |
225 | } | |
226 | ||
181c8655 PH |
227 | # In Python 2.x, json.dump expects a bytestream. |
228 | # In Python 3.x, it writes to a character stream | |
229 | if sys.version_info < (3, 0): | |
73159f99 | 230 | args['mode'] = 'wb' |
181c8655 | 231 | else: |
73159f99 S |
232 | args.update({ |
233 | 'mode': 'w', | |
234 | 'encoding': 'utf-8', | |
235 | }) | |
236 | ||
c86b6142 | 237 | tf = tempfile.NamedTemporaryFile(**compat_kwargs(args)) |
181c8655 PH |
238 | |
239 | try: | |
240 | with tf: | |
241 | json.dump(obj, tf) | |
1394646a IK |
242 | if sys.platform == 'win32': |
243 | # Need to remove existing file on Windows, else os.rename raises | |
244 | # WindowsError or FileExistsError. | |
245 | try: | |
246 | os.unlink(fn) | |
247 | except OSError: | |
248 | pass | |
181c8655 | 249 | os.rename(tf.name, fn) |
70a1165b | 250 | except Exception: |
181c8655 PH |
251 | try: |
252 | os.remove(tf.name) | |
253 | except OSError: | |
254 | pass | |
255 | raise | |
256 | ||
257 | ||
258 | if sys.version_info >= (2, 7): | |
ee114368 | 259 | def find_xpath_attr(node, xpath, key, val=None): |
59ae56fa | 260 | """ Find the xpath xpath[@key=val] """ |
5d2354f1 | 261 | assert re.match(r'^[a-zA-Z_-]+$', key) |
ee114368 | 262 | expr = xpath + ('[@%s]' % key if val is None else "[@%s='%s']" % (key, val)) |
59ae56fa PH |
263 | return node.find(expr) |
264 | else: | |
ee114368 | 265 | def find_xpath_attr(node, xpath, key, val=None): |
810c10ba | 266 | for f in node.findall(compat_xpath(xpath)): |
ee114368 S |
267 | if key not in f.attrib: |
268 | continue | |
269 | if val is None or f.attrib.get(key) == val: | |
59ae56fa PH |
270 | return f |
271 | return None | |
272 | ||
d7e66d39 JMF |
273 | # On python2.6 the xml.etree.ElementTree.Element methods don't support |
274 | # the namespace parameter | |
5f6a1245 JW |
275 | |
276 | ||
d7e66d39 JMF |
277 | def xpath_with_ns(path, ns_map): |
278 | components = [c.split(':') for c in path.split('/')] | |
279 | replaced = [] | |
280 | for c in components: | |
281 | if len(c) == 1: | |
282 | replaced.append(c[0]) | |
283 | else: | |
284 | ns, tag = c | |
285 | replaced.append('{%s}%s' % (ns_map[ns], tag)) | |
286 | return '/'.join(replaced) | |
287 | ||
d77c3dfd | 288 | |
a41fb80c | 289 | def xpath_element(node, xpath, name=None, fatal=False, default=NO_DEFAULT): |
578c0745 | 290 | def _find_xpath(xpath): |
810c10ba | 291 | return node.find(compat_xpath(xpath)) |
578c0745 S |
292 | |
293 | if isinstance(xpath, (str, compat_str)): | |
294 | n = _find_xpath(xpath) | |
295 | else: | |
296 | for xp in xpath: | |
297 | n = _find_xpath(xp) | |
298 | if n is not None: | |
299 | break | |
d74bebd5 | 300 | |
8e636da4 | 301 | if n is None: |
bf42a990 S |
302 | if default is not NO_DEFAULT: |
303 | return default | |
304 | elif fatal: | |
bf0ff932 PH |
305 | name = xpath if name is None else name |
306 | raise ExtractorError('Could not find XML element %s' % name) | |
307 | else: | |
308 | return None | |
a41fb80c S |
309 | return n |
310 | ||
311 | ||
312 | def xpath_text(node, xpath, name=None, fatal=False, default=NO_DEFAULT): | |
8e636da4 S |
313 | n = xpath_element(node, xpath, name, fatal=fatal, default=default) |
314 | if n is None or n == default: | |
315 | return n | |
316 | if n.text is None: | |
317 | if default is not NO_DEFAULT: | |
318 | return default | |
319 | elif fatal: | |
320 | name = xpath if name is None else name | |
321 | raise ExtractorError('Could not find XML element\'s text %s' % name) | |
322 | else: | |
323 | return None | |
324 | return n.text | |
a41fb80c S |
325 | |
326 | ||
327 | def xpath_attr(node, xpath, key, name=None, fatal=False, default=NO_DEFAULT): | |
328 | n = find_xpath_attr(node, xpath, key) | |
329 | if n is None: | |
330 | if default is not NO_DEFAULT: | |
331 | return default | |
332 | elif fatal: | |
333 | name = '%s[@%s]' % (xpath, key) if name is None else name | |
334 | raise ExtractorError('Could not find XML attribute %s' % name) | |
335 | else: | |
336 | return None | |
337 | return n.attrib[key] | |
bf0ff932 PH |
338 | |
339 | ||
9e6dd238 | 340 | def get_element_by_id(id, html): |
43e8fafd | 341 | """Return the content of the tag with the specified ID in the passed HTML document""" |
611c1dd9 | 342 | return get_element_by_attribute('id', id, html) |
43e8fafd | 343 | |
12ea2f30 | 344 | |
84c237fb | 345 | def get_element_by_class(class_name, html): |
2af12ad9 TC |
346 | """Return the content of the first tag with the specified class in the passed HTML document""" |
347 | retval = get_elements_by_class(class_name, html) | |
348 | return retval[0] if retval else None | |
349 | ||
350 | ||
351 | def get_element_by_attribute(attribute, value, html, escape_value=True): | |
352 | retval = get_elements_by_attribute(attribute, value, html, escape_value) | |
353 | return retval[0] if retval else None | |
354 | ||
355 | ||
356 | def get_elements_by_class(class_name, html): | |
357 | """Return the content of all tags with the specified class in the passed HTML document as a list""" | |
358 | return get_elements_by_attribute( | |
84c237fb YCH |
359 | 'class', r'[^\'"]*\b%s\b[^\'"]*' % re.escape(class_name), |
360 | html, escape_value=False) | |
361 | ||
362 | ||
2af12ad9 | 363 | def get_elements_by_attribute(attribute, value, html, escape_value=True): |
43e8fafd | 364 | """Return the content of the tag with the specified attribute in the passed HTML document""" |
9e6dd238 | 365 | |
84c237fb YCH |
366 | value = re.escape(value) if escape_value else value |
367 | ||
2af12ad9 TC |
368 | retlist = [] |
369 | for m in re.finditer(r'''(?xs) | |
38285056 | 370 | <([a-zA-Z0-9:._-]+) |
609ff8ca | 371 | (?:\s+[a-zA-Z0-9:._-]+(?:=[a-zA-Z0-9:._-]*|="[^"]*"|='[^']*'|))*? |
38285056 | 372 | \s+%s=['"]?%s['"]? |
609ff8ca | 373 | (?:\s+[a-zA-Z0-9:._-]+(?:=[a-zA-Z0-9:._-]*|="[^"]*"|='[^']*'|))*? |
38285056 PH |
374 | \s*> |
375 | (?P<content>.*?) | |
376 | </\1> | |
2af12ad9 TC |
377 | ''' % (re.escape(attribute), value), html): |
378 | res = m.group('content') | |
38285056 | 379 | |
2af12ad9 TC |
380 | if res.startswith('"') or res.startswith("'"): |
381 | res = res[1:-1] | |
38285056 | 382 | |
2af12ad9 | 383 | retlist.append(unescapeHTML(res)) |
a921f407 | 384 | |
2af12ad9 | 385 | return retlist |
a921f407 | 386 | |
c5229f39 | 387 | |
8bb56eee BF |
388 | class HTMLAttributeParser(compat_HTMLParser): |
389 | """Trivial HTML parser to gather the attributes for a single element""" | |
390 | def __init__(self): | |
c5229f39 | 391 | self.attrs = {} |
8bb56eee BF |
392 | compat_HTMLParser.__init__(self) |
393 | ||
394 | def handle_starttag(self, tag, attrs): | |
395 | self.attrs = dict(attrs) | |
396 | ||
c5229f39 | 397 | |
8bb56eee BF |
398 | def extract_attributes(html_element): |
399 | """Given a string for an HTML element such as | |
400 | <el | |
401 | a="foo" B="bar" c="&98;az" d=boz | |
402 | empty= noval entity="&" | |
403 | sq='"' dq="'" | |
404 | > | |
405 | Decode and return a dictionary of attributes. | |
406 | { | |
407 | 'a': 'foo', 'b': 'bar', c: 'baz', d: 'boz', | |
408 | 'empty': '', 'noval': None, 'entity': '&', | |
409 | 'sq': '"', 'dq': '\'' | |
410 | }. | |
411 | NB HTMLParser is stricter in Python 2.6 & 3.2 than in later versions, | |
412 | but the cases in the unit test will work for all of 2.6, 2.7, 3.2-3.5. | |
413 | """ | |
414 | parser = HTMLAttributeParser() | |
b4a3d461 S |
415 | try: |
416 | parser.feed(html_element) | |
417 | parser.close() | |
418 | # Older Python may throw HTMLParseError in case of malformed HTML | |
419 | except compat_HTMLParseError: | |
420 | pass | |
8bb56eee | 421 | return parser.attrs |
9e6dd238 | 422 | |
c5229f39 | 423 | |
9e6dd238 | 424 | def clean_html(html): |
59ae15a5 | 425 | """Clean an HTML snippet into a readable string""" |
dd622d7c PH |
426 | |
427 | if html is None: # Convenience for sanitizing descriptions etc. | |
428 | return html | |
429 | ||
59ae15a5 PH |
430 | # Newline vs <br /> |
431 | html = html.replace('\n', ' ') | |
edd9221c TF |
432 | html = re.sub(r'(?u)\s*<\s*br\s*/?\s*>\s*', '\n', html) |
433 | html = re.sub(r'(?u)<\s*/\s*p\s*>\s*<\s*p[^>]*>', '\n', html) | |
59ae15a5 PH |
434 | # Strip html tags |
435 | html = re.sub('<.*?>', '', html) | |
436 | # Replace html entities | |
437 | html = unescapeHTML(html) | |
7decf895 | 438 | return html.strip() |
9e6dd238 FV |
439 | |
440 | ||
d77c3dfd | 441 | def sanitize_open(filename, open_mode): |
59ae15a5 PH |
442 | """Try to open the given filename, and slightly tweak it if this fails. |
443 | ||
444 | Attempts to open the given filename. If this fails, it tries to change | |
445 | the filename slightly, step by step, until it's either able to open it | |
446 | or it fails and raises a final exception, like the standard open() | |
447 | function. | |
448 | ||
449 | It returns the tuple (stream, definitive_file_name). | |
450 | """ | |
451 | try: | |
28e614de | 452 | if filename == '-': |
59ae15a5 PH |
453 | if sys.platform == 'win32': |
454 | import msvcrt | |
455 | msvcrt.setmode(sys.stdout.fileno(), os.O_BINARY) | |
898280a0 | 456 | return (sys.stdout.buffer if hasattr(sys.stdout, 'buffer') else sys.stdout, filename) |
59ae15a5 PH |
457 | stream = open(encodeFilename(filename), open_mode) |
458 | return (stream, filename) | |
459 | except (IOError, OSError) as err: | |
f45c185f PH |
460 | if err.errno in (errno.EACCES,): |
461 | raise | |
59ae15a5 | 462 | |
f45c185f | 463 | # In case of error, try to remove win32 forbidden chars |
d55de57b | 464 | alt_filename = sanitize_path(filename) |
f45c185f PH |
465 | if alt_filename == filename: |
466 | raise | |
467 | else: | |
468 | # An exception here should be caught in the caller | |
d55de57b | 469 | stream = open(encodeFilename(alt_filename), open_mode) |
f45c185f | 470 | return (stream, alt_filename) |
d77c3dfd FV |
471 | |
472 | ||
473 | def timeconvert(timestr): | |
59ae15a5 PH |
474 | """Convert RFC 2822 defined time string into system timestamp""" |
475 | timestamp = None | |
476 | timetuple = email.utils.parsedate_tz(timestr) | |
477 | if timetuple is not None: | |
478 | timestamp = email.utils.mktime_tz(timetuple) | |
479 | return timestamp | |
1c469a94 | 480 | |
5f6a1245 | 481 | |
796173d0 | 482 | def sanitize_filename(s, restricted=False, is_id=False): |
59ae15a5 PH |
483 | """Sanitizes a string so it could be used as part of a filename. |
484 | If restricted is set, use a stricter subset of allowed characters. | |
158af524 S |
485 | Set is_id if this is not an arbitrary string, but an ID that should be kept |
486 | if possible. | |
59ae15a5 PH |
487 | """ |
488 | def replace_insane(char): | |
c587cbb7 AT |
489 | if restricted and char in ACCENT_CHARS: |
490 | return ACCENT_CHARS[char] | |
59ae15a5 PH |
491 | if char == '?' or ord(char) < 32 or ord(char) == 127: |
492 | return '' | |
493 | elif char == '"': | |
494 | return '' if restricted else '\'' | |
495 | elif char == ':': | |
496 | return '_-' if restricted else ' -' | |
497 | elif char in '\\/|*<>': | |
498 | return '_' | |
627dcfff | 499 | if restricted and (char in '!&\'()[]{}$;`^,#' or char.isspace()): |
59ae15a5 PH |
500 | return '_' |
501 | if restricted and ord(char) > 127: | |
502 | return '_' | |
503 | return char | |
504 | ||
2aeb06d6 PH |
505 | # Handle timestamps |
506 | s = re.sub(r'[0-9]+(?::[0-9]+)+', lambda m: m.group(0).replace(':', '_'), s) | |
28e614de | 507 | result = ''.join(map(replace_insane, s)) |
796173d0 PH |
508 | if not is_id: |
509 | while '__' in result: | |
510 | result = result.replace('__', '_') | |
511 | result = result.strip('_') | |
512 | # Common case of "Foreign band name - English song title" | |
513 | if restricted and result.startswith('-_'): | |
514 | result = result[2:] | |
5a42414b PH |
515 | if result.startswith('-'): |
516 | result = '_' + result[len('-'):] | |
a7440261 | 517 | result = result.lstrip('.') |
796173d0 PH |
518 | if not result: |
519 | result = '_' | |
59ae15a5 | 520 | return result |
d77c3dfd | 521 | |
5f6a1245 | 522 | |
a2aaf4db S |
523 | def sanitize_path(s): |
524 | """Sanitizes and normalizes path on Windows""" | |
525 | if sys.platform != 'win32': | |
526 | return s | |
be531ef1 S |
527 | drive_or_unc, _ = os.path.splitdrive(s) |
528 | if sys.version_info < (2, 7) and not drive_or_unc: | |
529 | drive_or_unc, _ = os.path.splitunc(s) | |
530 | norm_path = os.path.normpath(remove_start(s, drive_or_unc)).split(os.path.sep) | |
531 | if drive_or_unc: | |
a2aaf4db S |
532 | norm_path.pop(0) |
533 | sanitized_path = [ | |
ec85ded8 | 534 | path_part if path_part in ['.', '..'] else re.sub(r'(?:[/<>:"\|\\?\*]|[\s.]$)', '#', path_part) |
a2aaf4db | 535 | for path_part in norm_path] |
be531ef1 S |
536 | if drive_or_unc: |
537 | sanitized_path.insert(0, drive_or_unc + os.path.sep) | |
a2aaf4db S |
538 | return os.path.join(*sanitized_path) |
539 | ||
540 | ||
17bcc626 | 541 | def sanitize_url(url): |
befa4708 S |
542 | # Prepend protocol-less URLs with `http:` scheme in order to mitigate |
543 | # the number of unwanted failures due to missing protocol | |
544 | if url.startswith('//'): | |
545 | return 'http:%s' % url | |
546 | # Fix some common typos seen so far | |
547 | COMMON_TYPOS = ( | |
548 | # https://github.com/rg3/youtube-dl/issues/15649 | |
549 | (r'^httpss://', r'https://'), | |
550 | # https://bx1.be/lives/direct-tv/ | |
551 | (r'^rmtp([es]?)://', r'rtmp\1://'), | |
552 | ) | |
553 | for mistake, fixup in COMMON_TYPOS: | |
554 | if re.match(mistake, url): | |
555 | return re.sub(mistake, fixup, url) | |
556 | return url | |
17bcc626 S |
557 | |
558 | ||
67dda517 | 559 | def sanitized_Request(url, *args, **kwargs): |
17bcc626 | 560 | return compat_urllib_request.Request(sanitize_url(url), *args, **kwargs) |
67dda517 S |
561 | |
562 | ||
51098426 S |
563 | def expand_path(s): |
564 | """Expand shell variables and ~""" | |
565 | return os.path.expandvars(compat_expanduser(s)) | |
566 | ||
567 | ||
d77c3dfd | 568 | def orderedSet(iterable): |
59ae15a5 PH |
569 | """ Remove all duplicates from the input iterable """ |
570 | res = [] | |
571 | for el in iterable: | |
572 | if el not in res: | |
573 | res.append(el) | |
574 | return res | |
d77c3dfd | 575 | |
912b38b4 | 576 | |
55b2f099 | 577 | def _htmlentity_transform(entity_with_semicolon): |
4e408e47 | 578 | """Transforms an HTML entity to a character.""" |
55b2f099 YCH |
579 | entity = entity_with_semicolon[:-1] |
580 | ||
4e408e47 PH |
581 | # Known non-numeric HTML entity |
582 | if entity in compat_html_entities.name2codepoint: | |
583 | return compat_chr(compat_html_entities.name2codepoint[entity]) | |
584 | ||
55b2f099 YCH |
585 | # TODO: HTML5 allows entities without a semicolon. For example, |
586 | # 'Éric' should be decoded as 'Éric'. | |
587 | if entity_with_semicolon in compat_html_entities_html5: | |
588 | return compat_html_entities_html5[entity_with_semicolon] | |
589 | ||
91757b0f | 590 | mobj = re.match(r'#(x[0-9a-fA-F]+|[0-9]+)', entity) |
4e408e47 PH |
591 | if mobj is not None: |
592 | numstr = mobj.group(1) | |
28e614de | 593 | if numstr.startswith('x'): |
4e408e47 | 594 | base = 16 |
28e614de | 595 | numstr = '0%s' % numstr |
4e408e47 PH |
596 | else: |
597 | base = 10 | |
7aefc49c S |
598 | # See https://github.com/rg3/youtube-dl/issues/7518 |
599 | try: | |
600 | return compat_chr(int(numstr, base)) | |
601 | except ValueError: | |
602 | pass | |
4e408e47 PH |
603 | |
604 | # Unknown entity in name, return its literal representation | |
7a3f0c00 | 605 | return '&%s;' % entity |
4e408e47 PH |
606 | |
607 | ||
d77c3dfd | 608 | def unescapeHTML(s): |
912b38b4 PH |
609 | if s is None: |
610 | return None | |
611 | assert type(s) == compat_str | |
d77c3dfd | 612 | |
4e408e47 | 613 | return re.sub( |
95f3f7c2 | 614 | r'&([^&;]+;)', lambda m: _htmlentity_transform(m.group(1)), s) |
d77c3dfd | 615 | |
8bf48f23 | 616 | |
aa49acd1 S |
617 | def get_subprocess_encoding(): |
618 | if sys.platform == 'win32' and sys.getwindowsversion()[0] >= 5: | |
619 | # For subprocess calls, encode with locale encoding | |
620 | # Refer to http://stackoverflow.com/a/9951851/35070 | |
621 | encoding = preferredencoding() | |
622 | else: | |
623 | encoding = sys.getfilesystemencoding() | |
624 | if encoding is None: | |
625 | encoding = 'utf-8' | |
626 | return encoding | |
627 | ||
628 | ||
8bf48f23 | 629 | def encodeFilename(s, for_subprocess=False): |
59ae15a5 PH |
630 | """ |
631 | @param s The name of the file | |
632 | """ | |
d77c3dfd | 633 | |
8bf48f23 | 634 | assert type(s) == compat_str |
d77c3dfd | 635 | |
59ae15a5 PH |
636 | # Python 3 has a Unicode API |
637 | if sys.version_info >= (3, 0): | |
638 | return s | |
0f00efed | 639 | |
aa49acd1 S |
640 | # Pass '' directly to use Unicode APIs on Windows 2000 and up |
641 | # (Detecting Windows NT 4 is tricky because 'major >= 4' would | |
642 | # match Windows 9x series as well. Besides, NT 4 is obsolete.) | |
643 | if not for_subprocess and sys.platform == 'win32' and sys.getwindowsversion()[0] >= 5: | |
644 | return s | |
645 | ||
8ee239e9 YCH |
646 | # Jython assumes filenames are Unicode strings though reported as Python 2.x compatible |
647 | if sys.platform.startswith('java'): | |
648 | return s | |
649 | ||
aa49acd1 S |
650 | return s.encode(get_subprocess_encoding(), 'ignore') |
651 | ||
652 | ||
653 | def decodeFilename(b, for_subprocess=False): | |
654 | ||
655 | if sys.version_info >= (3, 0): | |
656 | return b | |
657 | ||
658 | if not isinstance(b, bytes): | |
659 | return b | |
660 | ||
661 | return b.decode(get_subprocess_encoding(), 'ignore') | |
8bf48f23 | 662 | |
f07b74fc PH |
663 | |
664 | def encodeArgument(s): | |
665 | if not isinstance(s, compat_str): | |
666 | # Legacy code that uses byte strings | |
667 | # Uncomment the following line after fixing all post processors | |
7af808a5 | 668 | # assert False, 'Internal error: %r should be of type %r, is %r' % (s, compat_str, type(s)) |
f07b74fc PH |
669 | s = s.decode('ascii') |
670 | return encodeFilename(s, True) | |
671 | ||
672 | ||
aa49acd1 S |
673 | def decodeArgument(b): |
674 | return decodeFilename(b, True) | |
675 | ||
676 | ||
8271226a PH |
677 | def decodeOption(optval): |
678 | if optval is None: | |
679 | return optval | |
680 | if isinstance(optval, bytes): | |
681 | optval = optval.decode(preferredencoding()) | |
682 | ||
683 | assert isinstance(optval, compat_str) | |
684 | return optval | |
1c256f70 | 685 | |
5f6a1245 | 686 | |
4539dd30 PH |
687 | def formatSeconds(secs): |
688 | if secs > 3600: | |
689 | return '%d:%02d:%02d' % (secs // 3600, (secs % 3600) // 60, secs % 60) | |
690 | elif secs > 60: | |
691 | return '%d:%02d' % (secs // 60, secs % 60) | |
692 | else: | |
693 | return '%d' % secs | |
694 | ||
a0ddb8a2 | 695 | |
be4a824d PH |
696 | def make_HTTPS_handler(params, **kwargs): |
697 | opts_no_check_certificate = params.get('nocheckcertificate', False) | |
0db261ba | 698 | if hasattr(ssl, 'create_default_context'): # Python >= 3.4 or 2.7.9 |
be5f2c19 | 699 | context = ssl.create_default_context(ssl.Purpose.SERVER_AUTH) |
0db261ba | 700 | if opts_no_check_certificate: |
be5f2c19 | 701 | context.check_hostname = False |
0db261ba | 702 | context.verify_mode = ssl.CERT_NONE |
a2366922 | 703 | try: |
be4a824d | 704 | return YoutubeDLHTTPSHandler(params, context=context, **kwargs) |
a2366922 PH |
705 | except TypeError: |
706 | # Python 2.7.8 | |
707 | # (create_default_context present but HTTPSHandler has no context=) | |
708 | pass | |
709 | ||
710 | if sys.version_info < (3, 2): | |
d7932313 | 711 | return YoutubeDLHTTPSHandler(params, **kwargs) |
aa37e3d4 | 712 | else: # Python < 3.4 |
d7932313 | 713 | context = ssl.SSLContext(ssl.PROTOCOL_TLSv1) |
ea6d901e | 714 | context.verify_mode = (ssl.CERT_NONE |
dca08720 | 715 | if opts_no_check_certificate |
ea6d901e | 716 | else ssl.CERT_REQUIRED) |
303b479e | 717 | context.set_default_verify_paths() |
be4a824d | 718 | return YoutubeDLHTTPSHandler(params, context=context, **kwargs) |
ea6d901e | 719 | |
732ea2f0 | 720 | |
08f2a92c JMF |
721 | def bug_reports_message(): |
722 | if ytdl_is_updateable(): | |
723 | update_cmd = 'type youtube-dl -U to update' | |
724 | else: | |
725 | update_cmd = 'see https://yt-dl.org/update on how to update' | |
726 | msg = '; please report this issue on https://yt-dl.org/bug .' | |
727 | msg += ' Make sure you are using the latest version; %s.' % update_cmd | |
728 | msg += ' Be sure to call youtube-dl with the --verbose flag and include its complete output.' | |
729 | return msg | |
730 | ||
731 | ||
bf5b9d85 PM |
732 | class YoutubeDLError(Exception): |
733 | """Base exception for YoutubeDL errors.""" | |
734 | pass | |
735 | ||
736 | ||
737 | class ExtractorError(YoutubeDLError): | |
1c256f70 | 738 | """Error during info extraction.""" |
5f6a1245 | 739 | |
d11271dd | 740 | def __init__(self, msg, tb=None, expected=False, cause=None, video_id=None): |
9a82b238 PH |
741 | """ tb, if given, is the original traceback (so that it can be printed out). |
742 | If expected is set, this is a normal error message and most likely not a bug in youtube-dl. | |
743 | """ | |
744 | ||
745 | if sys.exc_info()[0] in (compat_urllib_error.URLError, socket.timeout, UnavailableVideoError): | |
746 | expected = True | |
d11271dd PH |
747 | if video_id is not None: |
748 | msg = video_id + ': ' + msg | |
410f3e73 | 749 | if cause: |
28e614de | 750 | msg += ' (caused by %r)' % cause |
9a82b238 | 751 | if not expected: |
08f2a92c | 752 | msg += bug_reports_message() |
1c256f70 | 753 | super(ExtractorError, self).__init__(msg) |
d5979c5d | 754 | |
1c256f70 | 755 | self.traceback = tb |
8cc83b8d | 756 | self.exc_info = sys.exc_info() # preserve original exception |
2eabb802 | 757 | self.cause = cause |
d11271dd | 758 | self.video_id = video_id |
1c256f70 | 759 | |
01951dda PH |
760 | def format_traceback(self): |
761 | if self.traceback is None: | |
762 | return None | |
28e614de | 763 | return ''.join(traceback.format_tb(self.traceback)) |
01951dda | 764 | |
1c256f70 | 765 | |
416c7fcb PH |
766 | class UnsupportedError(ExtractorError): |
767 | def __init__(self, url): | |
768 | super(UnsupportedError, self).__init__( | |
769 | 'Unsupported URL: %s' % url, expected=True) | |
770 | self.url = url | |
771 | ||
772 | ||
55b3e45b JMF |
773 | class RegexNotFoundError(ExtractorError): |
774 | """Error when a regex didn't match""" | |
775 | pass | |
776 | ||
777 | ||
773f291d S |
778 | class GeoRestrictedError(ExtractorError): |
779 | """Geographic restriction Error exception. | |
780 | ||
781 | This exception may be thrown when a video is not available from your | |
782 | geographic location due to geographic restrictions imposed by a website. | |
783 | """ | |
784 | def __init__(self, msg, countries=None): | |
785 | super(GeoRestrictedError, self).__init__(msg, expected=True) | |
786 | self.msg = msg | |
787 | self.countries = countries | |
788 | ||
789 | ||
bf5b9d85 | 790 | class DownloadError(YoutubeDLError): |
59ae15a5 | 791 | """Download Error exception. |
d77c3dfd | 792 | |
59ae15a5 PH |
793 | This exception may be thrown by FileDownloader objects if they are not |
794 | configured to continue on errors. They will contain the appropriate | |
795 | error message. | |
796 | """ | |
5f6a1245 | 797 | |
8cc83b8d FV |
798 | def __init__(self, msg, exc_info=None): |
799 | """ exc_info, if given, is the original exception that caused the trouble (as returned by sys.exc_info()). """ | |
800 | super(DownloadError, self).__init__(msg) | |
801 | self.exc_info = exc_info | |
d77c3dfd FV |
802 | |
803 | ||
bf5b9d85 | 804 | class SameFileError(YoutubeDLError): |
59ae15a5 | 805 | """Same File exception. |
d77c3dfd | 806 | |
59ae15a5 PH |
807 | This exception will be thrown by FileDownloader objects if they detect |
808 | multiple files would have to be downloaded to the same file on disk. | |
809 | """ | |
810 | pass | |
d77c3dfd FV |
811 | |
812 | ||
bf5b9d85 | 813 | class PostProcessingError(YoutubeDLError): |
59ae15a5 | 814 | """Post Processing exception. |
d77c3dfd | 815 | |
59ae15a5 PH |
816 | This exception may be raised by PostProcessor's .run() method to |
817 | indicate an error in the postprocessing task. | |
818 | """ | |
5f6a1245 | 819 | |
7851b379 | 820 | def __init__(self, msg): |
bf5b9d85 | 821 | super(PostProcessingError, self).__init__(msg) |
7851b379 | 822 | self.msg = msg |
d77c3dfd | 823 | |
5f6a1245 | 824 | |
bf5b9d85 | 825 | class MaxDownloadsReached(YoutubeDLError): |
59ae15a5 PH |
826 | """ --max-downloads limit has been reached. """ |
827 | pass | |
d77c3dfd FV |
828 | |
829 | ||
bf5b9d85 | 830 | class UnavailableVideoError(YoutubeDLError): |
59ae15a5 | 831 | """Unavailable Format exception. |
d77c3dfd | 832 | |
59ae15a5 PH |
833 | This exception will be thrown when a video is requested |
834 | in a format that is not available for that video. | |
835 | """ | |
836 | pass | |
d77c3dfd FV |
837 | |
838 | ||
bf5b9d85 | 839 | class ContentTooShortError(YoutubeDLError): |
59ae15a5 | 840 | """Content Too Short exception. |
d77c3dfd | 841 | |
59ae15a5 PH |
842 | This exception may be raised by FileDownloader objects when a file they |
843 | download is too small for what the server announced first, indicating | |
844 | the connection was probably interrupted. | |
845 | """ | |
d77c3dfd | 846 | |
59ae15a5 | 847 | def __init__(self, downloaded, expected): |
bf5b9d85 PM |
848 | super(ContentTooShortError, self).__init__( |
849 | 'Downloaded {0} bytes, expected {1} bytes'.format(downloaded, expected) | |
850 | ) | |
2c7ed247 | 851 | # Both in bytes |
59ae15a5 PH |
852 | self.downloaded = downloaded |
853 | self.expected = expected | |
d77c3dfd | 854 | |
5f6a1245 | 855 | |
bf5b9d85 | 856 | class XAttrMetadataError(YoutubeDLError): |
efa97bdc YCH |
857 | def __init__(self, code=None, msg='Unknown error'): |
858 | super(XAttrMetadataError, self).__init__(msg) | |
859 | self.code = code | |
bd264412 | 860 | self.msg = msg |
efa97bdc YCH |
861 | |
862 | # Parsing code and msg | |
863 | if (self.code in (errno.ENOSPC, errno.EDQUOT) or | |
864 | 'No space left' in self.msg or 'Disk quota excedded' in self.msg): | |
865 | self.reason = 'NO_SPACE' | |
866 | elif self.code == errno.E2BIG or 'Argument list too long' in self.msg: | |
867 | self.reason = 'VALUE_TOO_LONG' | |
868 | else: | |
869 | self.reason = 'NOT_SUPPORTED' | |
870 | ||
871 | ||
bf5b9d85 | 872 | class XAttrUnavailableError(YoutubeDLError): |
efa97bdc YCH |
873 | pass |
874 | ||
875 | ||
c5a59d93 | 876 | def _create_http_connection(ydl_handler, http_class, is_https, *args, **kwargs): |
e5e78797 S |
877 | # Working around python 2 bug (see http://bugs.python.org/issue17849) by limiting |
878 | # expected HTTP responses to meet HTTP/1.0 or later (see also | |
879 | # https://github.com/rg3/youtube-dl/issues/6727) | |
880 | if sys.version_info < (3, 0): | |
65220c3b S |
881 | kwargs['strict'] = True |
882 | hc = http_class(*args, **compat_kwargs(kwargs)) | |
be4a824d PH |
883 | source_address = ydl_handler._params.get('source_address') |
884 | if source_address is not None: | |
885 | sa = (source_address, 0) | |
886 | if hasattr(hc, 'source_address'): # Python 2.7+ | |
887 | hc.source_address = sa | |
888 | else: # Python 2.6 | |
889 | def _hc_connect(self, *args, **kwargs): | |
890 | sock = compat_socket_create_connection( | |
891 | (self.host, self.port), self.timeout, sa) | |
892 | if is_https: | |
d7932313 PH |
893 | self.sock = ssl.wrap_socket( |
894 | sock, self.key_file, self.cert_file, | |
895 | ssl_version=ssl.PROTOCOL_TLSv1) | |
be4a824d PH |
896 | else: |
897 | self.sock = sock | |
898 | hc.connect = functools.partial(_hc_connect, hc) | |
899 | ||
900 | return hc | |
901 | ||
902 | ||
87f0e62d | 903 | def handle_youtubedl_headers(headers): |
992fc9d6 YCH |
904 | filtered_headers = headers |
905 | ||
906 | if 'Youtubedl-no-compression' in filtered_headers: | |
907 | filtered_headers = dict((k, v) for k, v in filtered_headers.items() if k.lower() != 'accept-encoding') | |
87f0e62d | 908 | del filtered_headers['Youtubedl-no-compression'] |
87f0e62d | 909 | |
992fc9d6 | 910 | return filtered_headers |
87f0e62d YCH |
911 | |
912 | ||
acebc9cd | 913 | class YoutubeDLHandler(compat_urllib_request.HTTPHandler): |
59ae15a5 PH |
914 | """Handler for HTTP requests and responses. |
915 | ||
916 | This class, when installed with an OpenerDirector, automatically adds | |
917 | the standard headers to every HTTP request and handles gzipped and | |
918 | deflated responses from web servers. If compression is to be avoided in | |
919 | a particular request, the original request in the program code only has | |
0424ec30 | 920 | to include the HTTP header "Youtubedl-no-compression", which will be |
59ae15a5 PH |
921 | removed before making the real request. |
922 | ||
923 | Part of this code was copied from: | |
924 | ||
925 | http://techknack.net/python-urllib2-handlers/ | |
926 | ||
927 | Andrew Rowls, the author of that code, agreed to release it to the | |
928 | public domain. | |
929 | """ | |
930 | ||
be4a824d PH |
931 | def __init__(self, params, *args, **kwargs): |
932 | compat_urllib_request.HTTPHandler.__init__(self, *args, **kwargs) | |
933 | self._params = params | |
934 | ||
935 | def http_open(self, req): | |
71aff188 YCH |
936 | conn_class = compat_http_client.HTTPConnection |
937 | ||
938 | socks_proxy = req.headers.get('Ytdl-socks-proxy') | |
939 | if socks_proxy: | |
940 | conn_class = make_socks_conn_class(conn_class, socks_proxy) | |
941 | del req.headers['Ytdl-socks-proxy'] | |
942 | ||
be4a824d | 943 | return self.do_open(functools.partial( |
71aff188 | 944 | _create_http_connection, self, conn_class, False), |
be4a824d PH |
945 | req) |
946 | ||
59ae15a5 PH |
947 | @staticmethod |
948 | def deflate(data): | |
949 | try: | |
950 | return zlib.decompress(data, -zlib.MAX_WBITS) | |
951 | except zlib.error: | |
952 | return zlib.decompress(data) | |
953 | ||
acebc9cd | 954 | def http_request(self, req): |
51f267d9 S |
955 | # According to RFC 3986, URLs can not contain non-ASCII characters, however this is not |
956 | # always respected by websites, some tend to give out URLs with non percent-encoded | |
957 | # non-ASCII characters (see telemb.py, ard.py [#3412]) | |
958 | # urllib chokes on URLs with non-ASCII characters (see http://bugs.python.org/issue3991) | |
959 | # To work around aforementioned issue we will replace request's original URL with | |
960 | # percent-encoded one | |
961 | # Since redirects are also affected (e.g. http://www.southpark.de/alle-episoden/s18e09) | |
962 | # the code of this workaround has been moved here from YoutubeDL.urlopen() | |
963 | url = req.get_full_url() | |
964 | url_escaped = escape_url(url) | |
965 | ||
966 | # Substitute URL if any change after escaping | |
967 | if url != url_escaped: | |
15d260eb | 968 | req = update_Request(req, url=url_escaped) |
51f267d9 | 969 | |
33ac271b | 970 | for h, v in std_headers.items(): |
3d5f7a39 JK |
971 | # Capitalize is needed because of Python bug 2275: http://bugs.python.org/issue2275 |
972 | # The dict keys are capitalized because of this bug by urllib | |
973 | if h.capitalize() not in req.headers: | |
33ac271b | 974 | req.add_header(h, v) |
87f0e62d YCH |
975 | |
976 | req.headers = handle_youtubedl_headers(req.headers) | |
989b4b2b PH |
977 | |
978 | if sys.version_info < (2, 7) and '#' in req.get_full_url(): | |
979 | # Python 2.6 is brain-dead when it comes to fragments | |
980 | req._Request__original = req._Request__original.partition('#')[0] | |
981 | req._Request__r_type = req._Request__r_type.partition('#')[0] | |
982 | ||
59ae15a5 PH |
983 | return req |
984 | ||
acebc9cd | 985 | def http_response(self, req, resp): |
59ae15a5 PH |
986 | old_resp = resp |
987 | # gzip | |
988 | if resp.headers.get('Content-encoding', '') == 'gzip': | |
aa3e9507 PH |
989 | content = resp.read() |
990 | gz = gzip.GzipFile(fileobj=io.BytesIO(content), mode='rb') | |
991 | try: | |
992 | uncompressed = io.BytesIO(gz.read()) | |
993 | except IOError as original_ioerror: | |
994 | # There may be junk add the end of the file | |
995 | # See http://stackoverflow.com/q/4928560/35070 for details | |
996 | for i in range(1, 1024): | |
997 | try: | |
998 | gz = gzip.GzipFile(fileobj=io.BytesIO(content[:-i]), mode='rb') | |
999 | uncompressed = io.BytesIO(gz.read()) | |
1000 | except IOError: | |
1001 | continue | |
1002 | break | |
1003 | else: | |
1004 | raise original_ioerror | |
b407d853 | 1005 | resp = compat_urllib_request.addinfourl(uncompressed, old_resp.headers, old_resp.url, old_resp.code) |
59ae15a5 | 1006 | resp.msg = old_resp.msg |
c047270c | 1007 | del resp.headers['Content-encoding'] |
59ae15a5 PH |
1008 | # deflate |
1009 | if resp.headers.get('Content-encoding', '') == 'deflate': | |
1010 | gz = io.BytesIO(self.deflate(resp.read())) | |
b407d853 | 1011 | resp = compat_urllib_request.addinfourl(gz, old_resp.headers, old_resp.url, old_resp.code) |
59ae15a5 | 1012 | resp.msg = old_resp.msg |
c047270c | 1013 | del resp.headers['Content-encoding'] |
ad729172 S |
1014 | # Percent-encode redirect URL of Location HTTP header to satisfy RFC 3986 (see |
1015 | # https://github.com/rg3/youtube-dl/issues/6457). | |
5a4d9ddb S |
1016 | if 300 <= resp.code < 400: |
1017 | location = resp.headers.get('Location') | |
1018 | if location: | |
1019 | # As of RFC 2616 default charset is iso-8859-1 that is respected by python 3 | |
1020 | if sys.version_info >= (3, 0): | |
1021 | location = location.encode('iso-8859-1').decode('utf-8') | |
0ea59007 YCH |
1022 | else: |
1023 | location = location.decode('utf-8') | |
5a4d9ddb S |
1024 | location_escaped = escape_url(location) |
1025 | if location != location_escaped: | |
1026 | del resp.headers['Location'] | |
9a4aec8b YCH |
1027 | if sys.version_info < (3, 0): |
1028 | location_escaped = location_escaped.encode('utf-8') | |
5a4d9ddb | 1029 | resp.headers['Location'] = location_escaped |
59ae15a5 | 1030 | return resp |
0f8d03f8 | 1031 | |
acebc9cd PH |
1032 | https_request = http_request |
1033 | https_response = http_response | |
bf50b038 | 1034 | |
5de90176 | 1035 | |
71aff188 YCH |
1036 | def make_socks_conn_class(base_class, socks_proxy): |
1037 | assert issubclass(base_class, ( | |
1038 | compat_http_client.HTTPConnection, compat_http_client.HTTPSConnection)) | |
1039 | ||
1040 | url_components = compat_urlparse.urlparse(socks_proxy) | |
1041 | if url_components.scheme.lower() == 'socks5': | |
1042 | socks_type = ProxyType.SOCKS5 | |
1043 | elif url_components.scheme.lower() in ('socks', 'socks4'): | |
1044 | socks_type = ProxyType.SOCKS4 | |
51fb4995 YCH |
1045 | elif url_components.scheme.lower() == 'socks4a': |
1046 | socks_type = ProxyType.SOCKS4A | |
71aff188 | 1047 | |
cdd94c2e YCH |
1048 | def unquote_if_non_empty(s): |
1049 | if not s: | |
1050 | return s | |
1051 | return compat_urllib_parse_unquote_plus(s) | |
1052 | ||
71aff188 YCH |
1053 | proxy_args = ( |
1054 | socks_type, | |
1055 | url_components.hostname, url_components.port or 1080, | |
1056 | True, # Remote DNS | |
cdd94c2e YCH |
1057 | unquote_if_non_empty(url_components.username), |
1058 | unquote_if_non_empty(url_components.password), | |
71aff188 YCH |
1059 | ) |
1060 | ||
1061 | class SocksConnection(base_class): | |
1062 | def connect(self): | |
1063 | self.sock = sockssocket() | |
1064 | self.sock.setproxy(*proxy_args) | |
1065 | if type(self.timeout) in (int, float): | |
1066 | self.sock.settimeout(self.timeout) | |
1067 | self.sock.connect((self.host, self.port)) | |
1068 | ||
1069 | if isinstance(self, compat_http_client.HTTPSConnection): | |
1070 | if hasattr(self, '_context'): # Python > 2.6 | |
1071 | self.sock = self._context.wrap_socket( | |
1072 | self.sock, server_hostname=self.host) | |
1073 | else: | |
1074 | self.sock = ssl.wrap_socket(self.sock) | |
1075 | ||
1076 | return SocksConnection | |
1077 | ||
1078 | ||
be4a824d PH |
1079 | class YoutubeDLHTTPSHandler(compat_urllib_request.HTTPSHandler): |
1080 | def __init__(self, params, https_conn_class=None, *args, **kwargs): | |
1081 | compat_urllib_request.HTTPSHandler.__init__(self, *args, **kwargs) | |
1082 | self._https_conn_class = https_conn_class or compat_http_client.HTTPSConnection | |
1083 | self._params = params | |
1084 | ||
1085 | def https_open(self, req): | |
4f264c02 | 1086 | kwargs = {} |
71aff188 YCH |
1087 | conn_class = self._https_conn_class |
1088 | ||
4f264c02 JMF |
1089 | if hasattr(self, '_context'): # python > 2.6 |
1090 | kwargs['context'] = self._context | |
1091 | if hasattr(self, '_check_hostname'): # python 3.x | |
1092 | kwargs['check_hostname'] = self._check_hostname | |
71aff188 YCH |
1093 | |
1094 | socks_proxy = req.headers.get('Ytdl-socks-proxy') | |
1095 | if socks_proxy: | |
1096 | conn_class = make_socks_conn_class(conn_class, socks_proxy) | |
1097 | del req.headers['Ytdl-socks-proxy'] | |
1098 | ||
be4a824d | 1099 | return self.do_open(functools.partial( |
71aff188 | 1100 | _create_http_connection, self, conn_class, True), |
4f264c02 | 1101 | req, **kwargs) |
be4a824d PH |
1102 | |
1103 | ||
a6420bf5 S |
1104 | class YoutubeDLCookieProcessor(compat_urllib_request.HTTPCookieProcessor): |
1105 | def __init__(self, cookiejar=None): | |
1106 | compat_urllib_request.HTTPCookieProcessor.__init__(self, cookiejar) | |
1107 | ||
1108 | def http_response(self, request, response): | |
1109 | # Python 2 will choke on next HTTP request in row if there are non-ASCII | |
1110 | # characters in Set-Cookie HTTP header of last response (see | |
1111 | # https://github.com/rg3/youtube-dl/issues/6769). | |
1112 | # In order to at least prevent crashing we will percent encode Set-Cookie | |
1113 | # header before HTTPCookieProcessor starts processing it. | |
e28034c5 S |
1114 | # if sys.version_info < (3, 0) and response.headers: |
1115 | # for set_cookie_header in ('Set-Cookie', 'Set-Cookie2'): | |
1116 | # set_cookie = response.headers.get(set_cookie_header) | |
1117 | # if set_cookie: | |
1118 | # set_cookie_escaped = compat_urllib_parse.quote(set_cookie, b"%/;:@&=+$,!~*'()?#[] ") | |
1119 | # if set_cookie != set_cookie_escaped: | |
1120 | # del response.headers[set_cookie_header] | |
1121 | # response.headers[set_cookie_header] = set_cookie_escaped | |
a6420bf5 S |
1122 | return compat_urllib_request.HTTPCookieProcessor.http_response(self, request, response) |
1123 | ||
1124 | https_request = compat_urllib_request.HTTPCookieProcessor.http_request | |
1125 | https_response = http_response | |
1126 | ||
1127 | ||
46f59e89 S |
1128 | def extract_timezone(date_str): |
1129 | m = re.search( | |
1130 | r'^.{8,}?(?P<tz>Z$| ?(?P<sign>\+|-)(?P<hours>[0-9]{2}):?(?P<minutes>[0-9]{2})$)', | |
1131 | date_str) | |
1132 | if not m: | |
1133 | timezone = datetime.timedelta() | |
1134 | else: | |
1135 | date_str = date_str[:-len(m.group('tz'))] | |
1136 | if not m.group('sign'): | |
1137 | timezone = datetime.timedelta() | |
1138 | else: | |
1139 | sign = 1 if m.group('sign') == '+' else -1 | |
1140 | timezone = datetime.timedelta( | |
1141 | hours=sign * int(m.group('hours')), | |
1142 | minutes=sign * int(m.group('minutes'))) | |
1143 | return timezone, date_str | |
1144 | ||
1145 | ||
08b38d54 | 1146 | def parse_iso8601(date_str, delimiter='T', timezone=None): |
912b38b4 PH |
1147 | """ Return a UNIX timestamp from the given date """ |
1148 | ||
1149 | if date_str is None: | |
1150 | return None | |
1151 | ||
52c3a6e4 S |
1152 | date_str = re.sub(r'\.[0-9]+', '', date_str) |
1153 | ||
08b38d54 | 1154 | if timezone is None: |
46f59e89 S |
1155 | timezone, date_str = extract_timezone(date_str) |
1156 | ||
52c3a6e4 S |
1157 | try: |
1158 | date_format = '%Y-%m-%d{0}%H:%M:%S'.format(delimiter) | |
1159 | dt = datetime.datetime.strptime(date_str, date_format) - timezone | |
1160 | return calendar.timegm(dt.timetuple()) | |
1161 | except ValueError: | |
1162 | pass | |
912b38b4 PH |
1163 | |
1164 | ||
46f59e89 S |
1165 | def date_formats(day_first=True): |
1166 | return DATE_FORMATS_DAY_FIRST if day_first else DATE_FORMATS_MONTH_FIRST | |
1167 | ||
1168 | ||
42bdd9d0 | 1169 | def unified_strdate(date_str, day_first=True): |
bf50b038 | 1170 | """Return a string with the date in the format YYYYMMDD""" |
64e7ad60 PH |
1171 | |
1172 | if date_str is None: | |
1173 | return None | |
bf50b038 | 1174 | upload_date = None |
5f6a1245 | 1175 | # Replace commas |
026fcc04 | 1176 | date_str = date_str.replace(',', ' ') |
42bdd9d0 | 1177 | # Remove AM/PM + timezone |
9bb8e0a3 | 1178 | date_str = re.sub(r'(?i)\s*(?:AM|PM)(?:\s+[A-Z]+)?', '', date_str) |
46f59e89 | 1179 | _, date_str = extract_timezone(date_str) |
42bdd9d0 | 1180 | |
46f59e89 | 1181 | for expression in date_formats(day_first): |
bf50b038 JMF |
1182 | try: |
1183 | upload_date = datetime.datetime.strptime(date_str, expression).strftime('%Y%m%d') | |
5de90176 | 1184 | except ValueError: |
bf50b038 | 1185 | pass |
42393ce2 PH |
1186 | if upload_date is None: |
1187 | timetuple = email.utils.parsedate_tz(date_str) | |
1188 | if timetuple: | |
c6b9cf05 S |
1189 | try: |
1190 | upload_date = datetime.datetime(*timetuple[:6]).strftime('%Y%m%d') | |
1191 | except ValueError: | |
1192 | pass | |
6a750402 JMF |
1193 | if upload_date is not None: |
1194 | return compat_str(upload_date) | |
bf50b038 | 1195 | |
5f6a1245 | 1196 | |
46f59e89 S |
1197 | def unified_timestamp(date_str, day_first=True): |
1198 | if date_str is None: | |
1199 | return None | |
1200 | ||
2ae2ffda | 1201 | date_str = re.sub(r'[,|]', '', date_str) |
46f59e89 | 1202 | |
7dc2a74e | 1203 | pm_delta = 12 if re.search(r'(?i)PM', date_str) else 0 |
46f59e89 S |
1204 | timezone, date_str = extract_timezone(date_str) |
1205 | ||
1206 | # Remove AM/PM + timezone | |
1207 | date_str = re.sub(r'(?i)\s*(?:AM|PM)(?:\s+[A-Z]+)?', '', date_str) | |
1208 | ||
deef3195 S |
1209 | # Remove unrecognized timezones from ISO 8601 alike timestamps |
1210 | m = re.search(r'\d{1,2}:\d{1,2}(?:\.\d+)?(?P<tz>\s*[A-Z]+)$', date_str) | |
1211 | if m: | |
1212 | date_str = date_str[:-len(m.group('tz'))] | |
1213 | ||
f226880c PH |
1214 | # Python only supports microseconds, so remove nanoseconds |
1215 | m = re.search(r'^([0-9]{4,}-[0-9]{1,2}-[0-9]{1,2}T[0-9]{1,2}:[0-9]{1,2}:[0-9]{1,2}\.[0-9]{6})[0-9]+$', date_str) | |
1216 | if m: | |
1217 | date_str = m.group(1) | |
1218 | ||
46f59e89 S |
1219 | for expression in date_formats(day_first): |
1220 | try: | |
7dc2a74e | 1221 | dt = datetime.datetime.strptime(date_str, expression) - timezone + datetime.timedelta(hours=pm_delta) |
46f59e89 S |
1222 | return calendar.timegm(dt.timetuple()) |
1223 | except ValueError: | |
1224 | pass | |
1225 | timetuple = email.utils.parsedate_tz(date_str) | |
1226 | if timetuple: | |
7dc2a74e | 1227 | return calendar.timegm(timetuple) + pm_delta * 3600 |
46f59e89 S |
1228 | |
1229 | ||
28e614de | 1230 | def determine_ext(url, default_ext='unknown_video'): |
f4776371 S |
1231 | if url is None: |
1232 | return default_ext | |
9cb9a5df | 1233 | guess = url.partition('?')[0].rpartition('.')[2] |
73e79f2a PH |
1234 | if re.match(r'^[A-Za-z0-9]+$', guess): |
1235 | return guess | |
a7aaa398 S |
1236 | # Try extract ext from URLs like http://example.com/foo/bar.mp4/?download |
1237 | elif guess.rstrip('/') in KNOWN_EXTENSIONS: | |
9cb9a5df | 1238 | return guess.rstrip('/') |
73e79f2a | 1239 | else: |
cbdbb766 | 1240 | return default_ext |
73e79f2a | 1241 | |
5f6a1245 | 1242 | |
d4051a8e | 1243 | def subtitles_filename(filename, sub_lang, sub_format): |
28e614de | 1244 | return filename.rsplit('.', 1)[0] + '.' + sub_lang + '.' + sub_format |
d4051a8e | 1245 | |
5f6a1245 | 1246 | |
bd558525 | 1247 | def date_from_str(date_str): |
37254abc JMF |
1248 | """ |
1249 | Return a datetime object from a string in the format YYYYMMDD or | |
1250 | (now|today)[+-][0-9](day|week|month|year)(s)?""" | |
1251 | today = datetime.date.today() | |
f8795e10 | 1252 | if date_str in ('now', 'today'): |
37254abc | 1253 | return today |
f8795e10 PH |
1254 | if date_str == 'yesterday': |
1255 | return today - datetime.timedelta(days=1) | |
ec85ded8 | 1256 | match = re.match(r'(now|today)(?P<sign>[+-])(?P<time>\d+)(?P<unit>day|week|month|year)(s)?', date_str) |
37254abc JMF |
1257 | if match is not None: |
1258 | sign = match.group('sign') | |
1259 | time = int(match.group('time')) | |
1260 | if sign == '-': | |
1261 | time = -time | |
1262 | unit = match.group('unit') | |
dfb1b146 | 1263 | # A bad approximation? |
37254abc JMF |
1264 | if unit == 'month': |
1265 | unit = 'day' | |
1266 | time *= 30 | |
1267 | elif unit == 'year': | |
1268 | unit = 'day' | |
1269 | time *= 365 | |
1270 | unit += 's' | |
1271 | delta = datetime.timedelta(**{unit: time}) | |
1272 | return today + delta | |
611c1dd9 | 1273 | return datetime.datetime.strptime(date_str, '%Y%m%d').date() |
5f6a1245 JW |
1274 | |
1275 | ||
e63fc1be | 1276 | def hyphenate_date(date_str): |
1277 | """ | |
1278 | Convert a date in 'YYYYMMDD' format to 'YYYY-MM-DD' format""" | |
1279 | match = re.match(r'^(\d\d\d\d)(\d\d)(\d\d)$', date_str) | |
1280 | if match is not None: | |
1281 | return '-'.join(match.groups()) | |
1282 | else: | |
1283 | return date_str | |
1284 | ||
5f6a1245 | 1285 | |
bd558525 JMF |
1286 | class DateRange(object): |
1287 | """Represents a time interval between two dates""" | |
5f6a1245 | 1288 | |
bd558525 JMF |
1289 | def __init__(self, start=None, end=None): |
1290 | """start and end must be strings in the format accepted by date""" | |
1291 | if start is not None: | |
1292 | self.start = date_from_str(start) | |
1293 | else: | |
1294 | self.start = datetime.datetime.min.date() | |
1295 | if end is not None: | |
1296 | self.end = date_from_str(end) | |
1297 | else: | |
1298 | self.end = datetime.datetime.max.date() | |
37254abc | 1299 | if self.start > self.end: |
bd558525 | 1300 | raise ValueError('Date range: "%s" , the start date must be before the end date' % self) |
5f6a1245 | 1301 | |
bd558525 JMF |
1302 | @classmethod |
1303 | def day(cls, day): | |
1304 | """Returns a range that only contains the given day""" | |
5f6a1245 JW |
1305 | return cls(day, day) |
1306 | ||
bd558525 JMF |
1307 | def __contains__(self, date): |
1308 | """Check if the date is in the range""" | |
37254abc JMF |
1309 | if not isinstance(date, datetime.date): |
1310 | date = date_from_str(date) | |
1311 | return self.start <= date <= self.end | |
5f6a1245 | 1312 | |
bd558525 | 1313 | def __str__(self): |
5f6a1245 | 1314 | return '%s - %s' % (self.start.isoformat(), self.end.isoformat()) |
c496ca96 PH |
1315 | |
1316 | ||
1317 | def platform_name(): | |
1318 | """ Returns the platform name as a compat_str """ | |
1319 | res = platform.platform() | |
1320 | if isinstance(res, bytes): | |
1321 | res = res.decode(preferredencoding()) | |
1322 | ||
1323 | assert isinstance(res, compat_str) | |
1324 | return res | |
c257baff PH |
1325 | |
1326 | ||
b58ddb32 PH |
1327 | def _windows_write_string(s, out): |
1328 | """ Returns True if the string was written using special methods, | |
1329 | False if it has yet to be written out.""" | |
1330 | # Adapted from http://stackoverflow.com/a/3259271/35070 | |
1331 | ||
1332 | import ctypes | |
1333 | import ctypes.wintypes | |
1334 | ||
1335 | WIN_OUTPUT_IDS = { | |
1336 | 1: -11, | |
1337 | 2: -12, | |
1338 | } | |
1339 | ||
a383a98a PH |
1340 | try: |
1341 | fileno = out.fileno() | |
1342 | except AttributeError: | |
1343 | # If the output stream doesn't have a fileno, it's virtual | |
1344 | return False | |
aa42e873 PH |
1345 | except io.UnsupportedOperation: |
1346 | # Some strange Windows pseudo files? | |
1347 | return False | |
b58ddb32 PH |
1348 | if fileno not in WIN_OUTPUT_IDS: |
1349 | return False | |
1350 | ||
d7cd9a9e | 1351 | GetStdHandle = compat_ctypes_WINFUNCTYPE( |
b58ddb32 | 1352 | ctypes.wintypes.HANDLE, ctypes.wintypes.DWORD)( |
d7cd9a9e | 1353 | ('GetStdHandle', ctypes.windll.kernel32)) |
b58ddb32 PH |
1354 | h = GetStdHandle(WIN_OUTPUT_IDS[fileno]) |
1355 | ||
d7cd9a9e | 1356 | WriteConsoleW = compat_ctypes_WINFUNCTYPE( |
b58ddb32 PH |
1357 | ctypes.wintypes.BOOL, ctypes.wintypes.HANDLE, ctypes.wintypes.LPWSTR, |
1358 | ctypes.wintypes.DWORD, ctypes.POINTER(ctypes.wintypes.DWORD), | |
d7cd9a9e | 1359 | ctypes.wintypes.LPVOID)(('WriteConsoleW', ctypes.windll.kernel32)) |
b58ddb32 PH |
1360 | written = ctypes.wintypes.DWORD(0) |
1361 | ||
d7cd9a9e | 1362 | GetFileType = compat_ctypes_WINFUNCTYPE(ctypes.wintypes.DWORD, ctypes.wintypes.DWORD)(('GetFileType', ctypes.windll.kernel32)) |
b58ddb32 PH |
1363 | FILE_TYPE_CHAR = 0x0002 |
1364 | FILE_TYPE_REMOTE = 0x8000 | |
d7cd9a9e | 1365 | GetConsoleMode = compat_ctypes_WINFUNCTYPE( |
b58ddb32 PH |
1366 | ctypes.wintypes.BOOL, ctypes.wintypes.HANDLE, |
1367 | ctypes.POINTER(ctypes.wintypes.DWORD))( | |
d7cd9a9e | 1368 | ('GetConsoleMode', ctypes.windll.kernel32)) |
b58ddb32 PH |
1369 | INVALID_HANDLE_VALUE = ctypes.wintypes.DWORD(-1).value |
1370 | ||
1371 | def not_a_console(handle): | |
1372 | if handle == INVALID_HANDLE_VALUE or handle is None: | |
1373 | return True | |
8fb3ac36 PH |
1374 | return ((GetFileType(handle) & ~FILE_TYPE_REMOTE) != FILE_TYPE_CHAR or |
1375 | GetConsoleMode(handle, ctypes.byref(ctypes.wintypes.DWORD())) == 0) | |
b58ddb32 PH |
1376 | |
1377 | if not_a_console(h): | |
1378 | return False | |
1379 | ||
d1b9c912 PH |
1380 | def next_nonbmp_pos(s): |
1381 | try: | |
1382 | return next(i for i, c in enumerate(s) if ord(c) > 0xffff) | |
1383 | except StopIteration: | |
1384 | return len(s) | |
1385 | ||
1386 | while s: | |
1387 | count = min(next_nonbmp_pos(s), 1024) | |
1388 | ||
b58ddb32 | 1389 | ret = WriteConsoleW( |
d1b9c912 | 1390 | h, s, count if count else 2, ctypes.byref(written), None) |
b58ddb32 PH |
1391 | if ret == 0: |
1392 | raise OSError('Failed to write string') | |
d1b9c912 PH |
1393 | if not count: # We just wrote a non-BMP character |
1394 | assert written.value == 2 | |
1395 | s = s[1:] | |
1396 | else: | |
1397 | assert written.value > 0 | |
1398 | s = s[written.value:] | |
b58ddb32 PH |
1399 | return True |
1400 | ||
1401 | ||
734f90bb | 1402 | def write_string(s, out=None, encoding=None): |
7459e3a2 PH |
1403 | if out is None: |
1404 | out = sys.stderr | |
8bf48f23 | 1405 | assert type(s) == compat_str |
7459e3a2 | 1406 | |
b58ddb32 PH |
1407 | if sys.platform == 'win32' and encoding is None and hasattr(out, 'fileno'): |
1408 | if _windows_write_string(s, out): | |
1409 | return | |
1410 | ||
7459e3a2 PH |
1411 | if ('b' in getattr(out, 'mode', '') or |
1412 | sys.version_info[0] < 3): # Python 2 lies about mode of sys.stderr | |
104aa738 PH |
1413 | byt = s.encode(encoding or preferredencoding(), 'ignore') |
1414 | out.write(byt) | |
1415 | elif hasattr(out, 'buffer'): | |
1416 | enc = encoding or getattr(out, 'encoding', None) or preferredencoding() | |
1417 | byt = s.encode(enc, 'ignore') | |
1418 | out.buffer.write(byt) | |
1419 | else: | |
8bf48f23 | 1420 | out.write(s) |
7459e3a2 PH |
1421 | out.flush() |
1422 | ||
1423 | ||
48ea9cea PH |
1424 | def bytes_to_intlist(bs): |
1425 | if not bs: | |
1426 | return [] | |
1427 | if isinstance(bs[0], int): # Python 3 | |
1428 | return list(bs) | |
1429 | else: | |
1430 | return [ord(c) for c in bs] | |
1431 | ||
c257baff | 1432 | |
cba892fa | 1433 | def intlist_to_bytes(xs): |
1434 | if not xs: | |
1435 | return b'' | |
edaa23f8 | 1436 | return compat_struct_pack('%dB' % len(xs), *xs) |
c38b1e77 PH |
1437 | |
1438 | ||
c1c9a79c PH |
1439 | # Cross-platform file locking |
1440 | if sys.platform == 'win32': | |
1441 | import ctypes.wintypes | |
1442 | import msvcrt | |
1443 | ||
1444 | class OVERLAPPED(ctypes.Structure): | |
1445 | _fields_ = [ | |
1446 | ('Internal', ctypes.wintypes.LPVOID), | |
1447 | ('InternalHigh', ctypes.wintypes.LPVOID), | |
1448 | ('Offset', ctypes.wintypes.DWORD), | |
1449 | ('OffsetHigh', ctypes.wintypes.DWORD), | |
1450 | ('hEvent', ctypes.wintypes.HANDLE), | |
1451 | ] | |
1452 | ||
1453 | kernel32 = ctypes.windll.kernel32 | |
1454 | LockFileEx = kernel32.LockFileEx | |
1455 | LockFileEx.argtypes = [ | |
1456 | ctypes.wintypes.HANDLE, # hFile | |
1457 | ctypes.wintypes.DWORD, # dwFlags | |
1458 | ctypes.wintypes.DWORD, # dwReserved | |
1459 | ctypes.wintypes.DWORD, # nNumberOfBytesToLockLow | |
1460 | ctypes.wintypes.DWORD, # nNumberOfBytesToLockHigh | |
1461 | ctypes.POINTER(OVERLAPPED) # Overlapped | |
1462 | ] | |
1463 | LockFileEx.restype = ctypes.wintypes.BOOL | |
1464 | UnlockFileEx = kernel32.UnlockFileEx | |
1465 | UnlockFileEx.argtypes = [ | |
1466 | ctypes.wintypes.HANDLE, # hFile | |
1467 | ctypes.wintypes.DWORD, # dwReserved | |
1468 | ctypes.wintypes.DWORD, # nNumberOfBytesToLockLow | |
1469 | ctypes.wintypes.DWORD, # nNumberOfBytesToLockHigh | |
1470 | ctypes.POINTER(OVERLAPPED) # Overlapped | |
1471 | ] | |
1472 | UnlockFileEx.restype = ctypes.wintypes.BOOL | |
1473 | whole_low = 0xffffffff | |
1474 | whole_high = 0x7fffffff | |
1475 | ||
1476 | def _lock_file(f, exclusive): | |
1477 | overlapped = OVERLAPPED() | |
1478 | overlapped.Offset = 0 | |
1479 | overlapped.OffsetHigh = 0 | |
1480 | overlapped.hEvent = 0 | |
1481 | f._lock_file_overlapped_p = ctypes.pointer(overlapped) | |
1482 | handle = msvcrt.get_osfhandle(f.fileno()) | |
1483 | if not LockFileEx(handle, 0x2 if exclusive else 0x0, 0, | |
1484 | whole_low, whole_high, f._lock_file_overlapped_p): | |
1485 | raise OSError('Locking file failed: %r' % ctypes.FormatError()) | |
1486 | ||
1487 | def _unlock_file(f): | |
1488 | assert f._lock_file_overlapped_p | |
1489 | handle = msvcrt.get_osfhandle(f.fileno()) | |
1490 | if not UnlockFileEx(handle, 0, | |
1491 | whole_low, whole_high, f._lock_file_overlapped_p): | |
1492 | raise OSError('Unlocking file failed: %r' % ctypes.FormatError()) | |
1493 | ||
1494 | else: | |
399a76e6 YCH |
1495 | # Some platforms, such as Jython, is missing fcntl |
1496 | try: | |
1497 | import fcntl | |
c1c9a79c | 1498 | |
399a76e6 YCH |
1499 | def _lock_file(f, exclusive): |
1500 | fcntl.flock(f, fcntl.LOCK_EX if exclusive else fcntl.LOCK_SH) | |
c1c9a79c | 1501 | |
399a76e6 YCH |
1502 | def _unlock_file(f): |
1503 | fcntl.flock(f, fcntl.LOCK_UN) | |
1504 | except ImportError: | |
1505 | UNSUPPORTED_MSG = 'file locking is not supported on this platform' | |
1506 | ||
1507 | def _lock_file(f, exclusive): | |
1508 | raise IOError(UNSUPPORTED_MSG) | |
1509 | ||
1510 | def _unlock_file(f): | |
1511 | raise IOError(UNSUPPORTED_MSG) | |
c1c9a79c PH |
1512 | |
1513 | ||
1514 | class locked_file(object): | |
1515 | def __init__(self, filename, mode, encoding=None): | |
1516 | assert mode in ['r', 'a', 'w'] | |
1517 | self.f = io.open(filename, mode, encoding=encoding) | |
1518 | self.mode = mode | |
1519 | ||
1520 | def __enter__(self): | |
1521 | exclusive = self.mode != 'r' | |
1522 | try: | |
1523 | _lock_file(self.f, exclusive) | |
1524 | except IOError: | |
1525 | self.f.close() | |
1526 | raise | |
1527 | return self | |
1528 | ||
1529 | def __exit__(self, etype, value, traceback): | |
1530 | try: | |
1531 | _unlock_file(self.f) | |
1532 | finally: | |
1533 | self.f.close() | |
1534 | ||
1535 | def __iter__(self): | |
1536 | return iter(self.f) | |
1537 | ||
1538 | def write(self, *args): | |
1539 | return self.f.write(*args) | |
1540 | ||
1541 | def read(self, *args): | |
1542 | return self.f.read(*args) | |
4eb7f1d1 JMF |
1543 | |
1544 | ||
4644ac55 S |
1545 | def get_filesystem_encoding(): |
1546 | encoding = sys.getfilesystemencoding() | |
1547 | return encoding if encoding is not None else 'utf-8' | |
1548 | ||
1549 | ||
4eb7f1d1 | 1550 | def shell_quote(args): |
a6a173c2 | 1551 | quoted_args = [] |
4644ac55 | 1552 | encoding = get_filesystem_encoding() |
a6a173c2 JMF |
1553 | for a in args: |
1554 | if isinstance(a, bytes): | |
1555 | # We may get a filename encoded with 'encodeFilename' | |
1556 | a = a.decode(encoding) | |
aefce8e6 | 1557 | quoted_args.append(compat_shlex_quote(a)) |
28e614de | 1558 | return ' '.join(quoted_args) |
9d4660ca PH |
1559 | |
1560 | ||
1561 | def smuggle_url(url, data): | |
1562 | """ Pass additional data in a URL for internal use. """ | |
1563 | ||
81953d1a RA |
1564 | url, idata = unsmuggle_url(url, {}) |
1565 | data.update(idata) | |
15707c7e | 1566 | sdata = compat_urllib_parse_urlencode( |
28e614de PH |
1567 | {'__youtubedl_smuggle': json.dumps(data)}) |
1568 | return url + '#' + sdata | |
9d4660ca PH |
1569 | |
1570 | ||
79f82953 | 1571 | def unsmuggle_url(smug_url, default=None): |
83e865a3 | 1572 | if '#__youtubedl_smuggle' not in smug_url: |
79f82953 | 1573 | return smug_url, default |
28e614de PH |
1574 | url, _, sdata = smug_url.rpartition('#') |
1575 | jsond = compat_parse_qs(sdata)['__youtubedl_smuggle'][0] | |
9d4660ca PH |
1576 | data = json.loads(jsond) |
1577 | return url, data | |
02dbf93f PH |
1578 | |
1579 | ||
02dbf93f PH |
1580 | def format_bytes(bytes): |
1581 | if bytes is None: | |
28e614de | 1582 | return 'N/A' |
02dbf93f PH |
1583 | if type(bytes) is str: |
1584 | bytes = float(bytes) | |
1585 | if bytes == 0.0: | |
1586 | exponent = 0 | |
1587 | else: | |
1588 | exponent = int(math.log(bytes, 1024.0)) | |
28e614de | 1589 | suffix = ['B', 'KiB', 'MiB', 'GiB', 'TiB', 'PiB', 'EiB', 'ZiB', 'YiB'][exponent] |
02dbf93f | 1590 | converted = float(bytes) / float(1024 ** exponent) |
28e614de | 1591 | return '%.2f%s' % (converted, suffix) |
f53c966a | 1592 | |
1c088fa8 | 1593 | |
fb47597b S |
1594 | def lookup_unit_table(unit_table, s): |
1595 | units_re = '|'.join(re.escape(u) for u in unit_table) | |
1596 | m = re.match( | |
782b1b5b | 1597 | r'(?P<num>[0-9]+(?:[,.][0-9]*)?)\s*(?P<unit>%s)\b' % units_re, s) |
fb47597b S |
1598 | if not m: |
1599 | return None | |
1600 | num_str = m.group('num').replace(',', '.') | |
1601 | mult = unit_table[m.group('unit')] | |
1602 | return int(float(num_str) * mult) | |
1603 | ||
1604 | ||
be64b5b0 PH |
1605 | def parse_filesize(s): |
1606 | if s is None: | |
1607 | return None | |
1608 | ||
dfb1b146 | 1609 | # The lower-case forms are of course incorrect and unofficial, |
be64b5b0 PH |
1610 | # but we support those too |
1611 | _UNIT_TABLE = { | |
1612 | 'B': 1, | |
1613 | 'b': 1, | |
70852b47 | 1614 | 'bytes': 1, |
be64b5b0 PH |
1615 | 'KiB': 1024, |
1616 | 'KB': 1000, | |
1617 | 'kB': 1024, | |
1618 | 'Kb': 1000, | |
13585d76 | 1619 | 'kb': 1000, |
70852b47 YCH |
1620 | 'kilobytes': 1000, |
1621 | 'kibibytes': 1024, | |
be64b5b0 PH |
1622 | 'MiB': 1024 ** 2, |
1623 | 'MB': 1000 ** 2, | |
1624 | 'mB': 1024 ** 2, | |
1625 | 'Mb': 1000 ** 2, | |
13585d76 | 1626 | 'mb': 1000 ** 2, |
70852b47 YCH |
1627 | 'megabytes': 1000 ** 2, |
1628 | 'mebibytes': 1024 ** 2, | |
be64b5b0 PH |
1629 | 'GiB': 1024 ** 3, |
1630 | 'GB': 1000 ** 3, | |
1631 | 'gB': 1024 ** 3, | |
1632 | 'Gb': 1000 ** 3, | |
13585d76 | 1633 | 'gb': 1000 ** 3, |
70852b47 YCH |
1634 | 'gigabytes': 1000 ** 3, |
1635 | 'gibibytes': 1024 ** 3, | |
be64b5b0 PH |
1636 | 'TiB': 1024 ** 4, |
1637 | 'TB': 1000 ** 4, | |
1638 | 'tB': 1024 ** 4, | |
1639 | 'Tb': 1000 ** 4, | |
13585d76 | 1640 | 'tb': 1000 ** 4, |
70852b47 YCH |
1641 | 'terabytes': 1000 ** 4, |
1642 | 'tebibytes': 1024 ** 4, | |
be64b5b0 PH |
1643 | 'PiB': 1024 ** 5, |
1644 | 'PB': 1000 ** 5, | |
1645 | 'pB': 1024 ** 5, | |
1646 | 'Pb': 1000 ** 5, | |
13585d76 | 1647 | 'pb': 1000 ** 5, |
70852b47 YCH |
1648 | 'petabytes': 1000 ** 5, |
1649 | 'pebibytes': 1024 ** 5, | |
be64b5b0 PH |
1650 | 'EiB': 1024 ** 6, |
1651 | 'EB': 1000 ** 6, | |
1652 | 'eB': 1024 ** 6, | |
1653 | 'Eb': 1000 ** 6, | |
13585d76 | 1654 | 'eb': 1000 ** 6, |
70852b47 YCH |
1655 | 'exabytes': 1000 ** 6, |
1656 | 'exbibytes': 1024 ** 6, | |
be64b5b0 PH |
1657 | 'ZiB': 1024 ** 7, |
1658 | 'ZB': 1000 ** 7, | |
1659 | 'zB': 1024 ** 7, | |
1660 | 'Zb': 1000 ** 7, | |
13585d76 | 1661 | 'zb': 1000 ** 7, |
70852b47 YCH |
1662 | 'zettabytes': 1000 ** 7, |
1663 | 'zebibytes': 1024 ** 7, | |
be64b5b0 PH |
1664 | 'YiB': 1024 ** 8, |
1665 | 'YB': 1000 ** 8, | |
1666 | 'yB': 1024 ** 8, | |
1667 | 'Yb': 1000 ** 8, | |
13585d76 | 1668 | 'yb': 1000 ** 8, |
70852b47 YCH |
1669 | 'yottabytes': 1000 ** 8, |
1670 | 'yobibytes': 1024 ** 8, | |
be64b5b0 PH |
1671 | } |
1672 | ||
fb47597b S |
1673 | return lookup_unit_table(_UNIT_TABLE, s) |
1674 | ||
1675 | ||
1676 | def parse_count(s): | |
1677 | if s is None: | |
be64b5b0 PH |
1678 | return None |
1679 | ||
fb47597b S |
1680 | s = s.strip() |
1681 | ||
1682 | if re.match(r'^[\d,.]+$', s): | |
1683 | return str_to_int(s) | |
1684 | ||
1685 | _UNIT_TABLE = { | |
1686 | 'k': 1000, | |
1687 | 'K': 1000, | |
1688 | 'm': 1000 ** 2, | |
1689 | 'M': 1000 ** 2, | |
1690 | 'kk': 1000 ** 2, | |
1691 | 'KK': 1000 ** 2, | |
1692 | } | |
be64b5b0 | 1693 | |
fb47597b | 1694 | return lookup_unit_table(_UNIT_TABLE, s) |
be64b5b0 | 1695 | |
2f7ae819 | 1696 | |
b871d7e9 S |
1697 | def parse_resolution(s): |
1698 | if s is None: | |
1699 | return {} | |
1700 | ||
1701 | mobj = re.search(r'\b(?P<w>\d+)\s*[xX×]\s*(?P<h>\d+)\b', s) | |
1702 | if mobj: | |
1703 | return { | |
1704 | 'width': int(mobj.group('w')), | |
1705 | 'height': int(mobj.group('h')), | |
1706 | } | |
1707 | ||
1708 | mobj = re.search(r'\b(\d+)[pPiI]\b', s) | |
1709 | if mobj: | |
1710 | return {'height': int(mobj.group(1))} | |
1711 | ||
1712 | mobj = re.search(r'\b([48])[kK]\b', s) | |
1713 | if mobj: | |
1714 | return {'height': int(mobj.group(1)) * 540} | |
1715 | ||
1716 | return {} | |
1717 | ||
1718 | ||
a942d6cb | 1719 | def month_by_name(name, lang='en'): |
caefb1de PH |
1720 | """ Return the number of a month by (locale-independently) English name """ |
1721 | ||
f6717dec | 1722 | month_names = MONTH_NAMES.get(lang, MONTH_NAMES['en']) |
a942d6cb | 1723 | |
caefb1de | 1724 | try: |
f6717dec | 1725 | return month_names.index(name) + 1 |
7105440c YCH |
1726 | except ValueError: |
1727 | return None | |
1728 | ||
1729 | ||
1730 | def month_by_abbreviation(abbrev): | |
1731 | """ Return the number of a month by (locale-independently) English | |
1732 | abbreviations """ | |
1733 | ||
1734 | try: | |
1735 | return [s[:3] for s in ENGLISH_MONTH_NAMES].index(abbrev) + 1 | |
caefb1de PH |
1736 | except ValueError: |
1737 | return None | |
18258362 JMF |
1738 | |
1739 | ||
5aafe895 | 1740 | def fix_xml_ampersands(xml_str): |
18258362 | 1741 | """Replace all the '&' by '&' in XML""" |
5aafe895 PH |
1742 | return re.sub( |
1743 | r'&(?!amp;|lt;|gt;|apos;|quot;|#x[0-9a-fA-F]{,4};|#[0-9]{,4};)', | |
28e614de | 1744 | '&', |
5aafe895 | 1745 | xml_str) |
e3946f98 PH |
1746 | |
1747 | ||
1748 | def setproctitle(title): | |
8bf48f23 | 1749 | assert isinstance(title, compat_str) |
c1c05c67 YCH |
1750 | |
1751 | # ctypes in Jython is not complete | |
1752 | # http://bugs.jython.org/issue2148 | |
1753 | if sys.platform.startswith('java'): | |
1754 | return | |
1755 | ||
e3946f98 | 1756 | try: |
611c1dd9 | 1757 | libc = ctypes.cdll.LoadLibrary('libc.so.6') |
e3946f98 PH |
1758 | except OSError: |
1759 | return | |
2f49bcd6 RC |
1760 | except TypeError: |
1761 | # LoadLibrary in Windows Python 2.7.13 only expects | |
1762 | # a bytestring, but since unicode_literals turns | |
1763 | # every string into a unicode string, it fails. | |
1764 | return | |
6eefe533 PH |
1765 | title_bytes = title.encode('utf-8') |
1766 | buf = ctypes.create_string_buffer(len(title_bytes)) | |
1767 | buf.value = title_bytes | |
e3946f98 | 1768 | try: |
6eefe533 | 1769 | libc.prctl(15, buf, 0, 0, 0) |
e3946f98 PH |
1770 | except AttributeError: |
1771 | return # Strange libc, just skip this | |
d7dda168 PH |
1772 | |
1773 | ||
1774 | def remove_start(s, start): | |
46bc9b7d | 1775 | return s[len(start):] if s is not None and s.startswith(start) else s |
29eb5174 PH |
1776 | |
1777 | ||
2b9faf55 | 1778 | def remove_end(s, end): |
46bc9b7d | 1779 | return s[:-len(end)] if s is not None and s.endswith(end) else s |
2b9faf55 PH |
1780 | |
1781 | ||
31b2051e S |
1782 | def remove_quotes(s): |
1783 | if s is None or len(s) < 2: | |
1784 | return s | |
1785 | for quote in ('"', "'", ): | |
1786 | if s[0] == quote and s[-1] == quote: | |
1787 | return s[1:-1] | |
1788 | return s | |
1789 | ||
1790 | ||
29eb5174 | 1791 | def url_basename(url): |
9b8aaeed | 1792 | path = compat_urlparse.urlparse(url).path |
28e614de | 1793 | return path.strip('/').split('/')[-1] |
aa94a6d3 PH |
1794 | |
1795 | ||
02dc0a36 S |
1796 | def base_url(url): |
1797 | return re.match(r'https?://[^?#&]+/', url).group() | |
1798 | ||
1799 | ||
e34c3361 | 1800 | def urljoin(base, path): |
4b5de77b S |
1801 | if isinstance(path, bytes): |
1802 | path = path.decode('utf-8') | |
e34c3361 S |
1803 | if not isinstance(path, compat_str) or not path: |
1804 | return None | |
b0c65c67 | 1805 | if re.match(r'^(?:https?:)?//', path): |
e34c3361 | 1806 | return path |
4b5de77b S |
1807 | if isinstance(base, bytes): |
1808 | base = base.decode('utf-8') | |
1809 | if not isinstance(base, compat_str) or not re.match( | |
1810 | r'^(?:https?:)?//', base): | |
e34c3361 S |
1811 | return None |
1812 | return compat_urlparse.urljoin(base, path) | |
1813 | ||
1814 | ||
aa94a6d3 PH |
1815 | class HEADRequest(compat_urllib_request.Request): |
1816 | def get_method(self): | |
611c1dd9 | 1817 | return 'HEAD' |
7217e148 PH |
1818 | |
1819 | ||
95cf60e8 S |
1820 | class PUTRequest(compat_urllib_request.Request): |
1821 | def get_method(self): | |
1822 | return 'PUT' | |
1823 | ||
1824 | ||
9732d77e | 1825 | def int_or_none(v, scale=1, default=None, get_attr=None, invscale=1): |
28746fbd PH |
1826 | if get_attr: |
1827 | if v is not None: | |
1828 | v = getattr(v, get_attr, None) | |
9572013d PH |
1829 | if v == '': |
1830 | v = None | |
1812afb7 S |
1831 | if v is None: |
1832 | return default | |
1833 | try: | |
1834 | return int(v) * invscale // scale | |
1835 | except ValueError: | |
af98f8ff | 1836 | return default |
9732d77e | 1837 | |
9572013d | 1838 | |
40a90862 JMF |
1839 | def str_or_none(v, default=None): |
1840 | return default if v is None else compat_str(v) | |
1841 | ||
9732d77e PH |
1842 | |
1843 | def str_to_int(int_str): | |
48d4681e | 1844 | """ A more relaxed version of int_or_none """ |
9732d77e PH |
1845 | if int_str is None: |
1846 | return None | |
28e614de | 1847 | int_str = re.sub(r'[,\.\+]', '', int_str) |
9732d77e | 1848 | return int(int_str) |
608d11f5 PH |
1849 | |
1850 | ||
9732d77e | 1851 | def float_or_none(v, scale=1, invscale=1, default=None): |
caf80631 S |
1852 | if v is None: |
1853 | return default | |
1854 | try: | |
1855 | return float(v) * invscale / scale | |
1856 | except ValueError: | |
1857 | return default | |
43f775e4 PH |
1858 | |
1859 | ||
c7e327c4 S |
1860 | def bool_or_none(v, default=None): |
1861 | return v if isinstance(v, bool) else default | |
1862 | ||
1863 | ||
b72b4431 S |
1864 | def strip_or_none(v): |
1865 | return None if v is None else v.strip() | |
1866 | ||
1867 | ||
608d11f5 | 1868 | def parse_duration(s): |
8f9312c3 | 1869 | if not isinstance(s, compat_basestring): |
608d11f5 PH |
1870 | return None |
1871 | ||
ca7b3246 S |
1872 | s = s.strip() |
1873 | ||
acaff495 | 1874 | days, hours, mins, secs, ms = [None] * 5 |
15846398 | 1875 | m = re.match(r'(?:(?:(?:(?P<days>[0-9]+):)?(?P<hours>[0-9]+):)?(?P<mins>[0-9]+):)?(?P<secs>[0-9]+)(?P<ms>\.[0-9]+)?Z?$', s) |
acaff495 | 1876 | if m: |
1877 | days, hours, mins, secs, ms = m.groups() | |
1878 | else: | |
1879 | m = re.match( | |
056653bb S |
1880 | r'''(?ix)(?:P? |
1881 | (?: | |
1882 | [0-9]+\s*y(?:ears?)?\s* | |
1883 | )? | |
1884 | (?: | |
1885 | [0-9]+\s*m(?:onths?)?\s* | |
1886 | )? | |
1887 | (?: | |
1888 | [0-9]+\s*w(?:eeks?)?\s* | |
1889 | )? | |
8f4b58d7 | 1890 | (?: |
acaff495 | 1891 | (?P<days>[0-9]+)\s*d(?:ays?)?\s* |
8f4b58d7 | 1892 | )? |
056653bb | 1893 | T)? |
acaff495 | 1894 | (?: |
1895 | (?P<hours>[0-9]+)\s*h(?:ours?)?\s* | |
1896 | )? | |
1897 | (?: | |
1898 | (?P<mins>[0-9]+)\s*m(?:in(?:ute)?s?)?\s* | |
1899 | )? | |
1900 | (?: | |
1901 | (?P<secs>[0-9]+)(?P<ms>\.[0-9]+)?\s*s(?:ec(?:ond)?s?)?\s* | |
15846398 | 1902 | )?Z?$''', s) |
acaff495 | 1903 | if m: |
1904 | days, hours, mins, secs, ms = m.groups() | |
1905 | else: | |
15846398 | 1906 | m = re.match(r'(?i)(?:(?P<hours>[0-9.]+)\s*(?:hours?)|(?P<mins>[0-9.]+)\s*(?:mins?\.?|minutes?)\s*)Z?$', s) |
acaff495 | 1907 | if m: |
1908 | hours, mins = m.groups() | |
1909 | else: | |
1910 | return None | |
1911 | ||
1912 | duration = 0 | |
1913 | if secs: | |
1914 | duration += float(secs) | |
1915 | if mins: | |
1916 | duration += float(mins) * 60 | |
1917 | if hours: | |
1918 | duration += float(hours) * 60 * 60 | |
1919 | if days: | |
1920 | duration += float(days) * 24 * 60 * 60 | |
1921 | if ms: | |
1922 | duration += float(ms) | |
1923 | return duration | |
91d7d0b3 JMF |
1924 | |
1925 | ||
e65e4c88 | 1926 | def prepend_extension(filename, ext, expected_real_ext=None): |
5f6a1245 | 1927 | name, real_ext = os.path.splitext(filename) |
e65e4c88 S |
1928 | return ( |
1929 | '{0}.{1}{2}'.format(name, ext, real_ext) | |
1930 | if not expected_real_ext or real_ext[1:] == expected_real_ext | |
1931 | else '{0}.{1}'.format(filename, ext)) | |
d70ad093 PH |
1932 | |
1933 | ||
b3ed15b7 S |
1934 | def replace_extension(filename, ext, expected_real_ext=None): |
1935 | name, real_ext = os.path.splitext(filename) | |
1936 | return '{0}.{1}'.format( | |
1937 | name if not expected_real_ext or real_ext[1:] == expected_real_ext else filename, | |
1938 | ext) | |
1939 | ||
1940 | ||
d70ad093 PH |
1941 | def check_executable(exe, args=[]): |
1942 | """ Checks if the given binary is installed somewhere in PATH, and returns its name. | |
1943 | args can be a list of arguments for a short output (like -version) """ | |
1944 | try: | |
1945 | subprocess.Popen([exe] + args, stdout=subprocess.PIPE, stderr=subprocess.PIPE).communicate() | |
1946 | except OSError: | |
1947 | return False | |
1948 | return exe | |
b7ab0590 PH |
1949 | |
1950 | ||
95807118 | 1951 | def get_exe_version(exe, args=['--version'], |
cae97f65 | 1952 | version_re=None, unrecognized='present'): |
95807118 PH |
1953 | """ Returns the version of the specified executable, |
1954 | or False if the executable is not present """ | |
1955 | try: | |
b64d04c1 YCH |
1956 | # STDIN should be redirected too. On UNIX-like systems, ffmpeg triggers |
1957 | # SIGTTOU if youtube-dl is run in the background. | |
1958 | # See https://github.com/rg3/youtube-dl/issues/955#issuecomment-209789656 | |
cae97f65 | 1959 | out, _ = subprocess.Popen( |
54116803 | 1960 | [encodeArgument(exe)] + args, |
00ca7552 | 1961 | stdin=subprocess.PIPE, |
95807118 PH |
1962 | stdout=subprocess.PIPE, stderr=subprocess.STDOUT).communicate() |
1963 | except OSError: | |
1964 | return False | |
cae97f65 PH |
1965 | if isinstance(out, bytes): # Python 2.x |
1966 | out = out.decode('ascii', 'ignore') | |
1967 | return detect_exe_version(out, version_re, unrecognized) | |
1968 | ||
1969 | ||
1970 | def detect_exe_version(output, version_re=None, unrecognized='present'): | |
1971 | assert isinstance(output, compat_str) | |
1972 | if version_re is None: | |
1973 | version_re = r'version\s+([-0-9._a-zA-Z]+)' | |
1974 | m = re.search(version_re, output) | |
95807118 PH |
1975 | if m: |
1976 | return m.group(1) | |
1977 | else: | |
1978 | return unrecognized | |
1979 | ||
1980 | ||
b7ab0590 | 1981 | class PagedList(object): |
dd26ced1 PH |
1982 | def __len__(self): |
1983 | # This is only useful for tests | |
1984 | return len(self.getslice()) | |
1985 | ||
9c44d242 PH |
1986 | |
1987 | class OnDemandPagedList(PagedList): | |
6be08ce6 | 1988 | def __init__(self, pagefunc, pagesize, use_cache=True): |
9c44d242 PH |
1989 | self._pagefunc = pagefunc |
1990 | self._pagesize = pagesize | |
b95dc034 YCH |
1991 | self._use_cache = use_cache |
1992 | if use_cache: | |
1993 | self._cache = {} | |
9c44d242 | 1994 | |
b7ab0590 PH |
1995 | def getslice(self, start=0, end=None): |
1996 | res = [] | |
1997 | for pagenum in itertools.count(start // self._pagesize): | |
1998 | firstid = pagenum * self._pagesize | |
1999 | nextfirstid = pagenum * self._pagesize + self._pagesize | |
2000 | if start >= nextfirstid: | |
2001 | continue | |
2002 | ||
b95dc034 YCH |
2003 | page_results = None |
2004 | if self._use_cache: | |
2005 | page_results = self._cache.get(pagenum) | |
2006 | if page_results is None: | |
2007 | page_results = list(self._pagefunc(pagenum)) | |
2008 | if self._use_cache: | |
2009 | self._cache[pagenum] = page_results | |
b7ab0590 PH |
2010 | |
2011 | startv = ( | |
2012 | start % self._pagesize | |
2013 | if firstid <= start < nextfirstid | |
2014 | else 0) | |
2015 | ||
2016 | endv = ( | |
2017 | ((end - 1) % self._pagesize) + 1 | |
2018 | if (end is not None and firstid <= end <= nextfirstid) | |
2019 | else None) | |
2020 | ||
2021 | if startv != 0 or endv is not None: | |
2022 | page_results = page_results[startv:endv] | |
2023 | res.extend(page_results) | |
2024 | ||
2025 | # A little optimization - if current page is not "full", ie. does | |
2026 | # not contain page_size videos then we can assume that this page | |
2027 | # is the last one - there are no more ids on further pages - | |
2028 | # i.e. no need to query again. | |
2029 | if len(page_results) + startv < self._pagesize: | |
2030 | break | |
2031 | ||
2032 | # If we got the whole page, but the next page is not interesting, | |
2033 | # break out early as well | |
2034 | if end == nextfirstid: | |
2035 | break | |
2036 | return res | |
81c2f20b PH |
2037 | |
2038 | ||
9c44d242 PH |
2039 | class InAdvancePagedList(PagedList): |
2040 | def __init__(self, pagefunc, pagecount, pagesize): | |
2041 | self._pagefunc = pagefunc | |
2042 | self._pagecount = pagecount | |
2043 | self._pagesize = pagesize | |
2044 | ||
2045 | def getslice(self, start=0, end=None): | |
2046 | res = [] | |
2047 | start_page = start // self._pagesize | |
2048 | end_page = ( | |
2049 | self._pagecount if end is None else (end // self._pagesize + 1)) | |
2050 | skip_elems = start - start_page * self._pagesize | |
2051 | only_more = None if end is None else end - start | |
2052 | for pagenum in range(start_page, end_page): | |
2053 | page = list(self._pagefunc(pagenum)) | |
2054 | if skip_elems: | |
2055 | page = page[skip_elems:] | |
2056 | skip_elems = None | |
2057 | if only_more is not None: | |
2058 | if len(page) < only_more: | |
2059 | only_more -= len(page) | |
2060 | else: | |
2061 | page = page[:only_more] | |
2062 | res.extend(page) | |
2063 | break | |
2064 | res.extend(page) | |
2065 | return res | |
2066 | ||
2067 | ||
81c2f20b | 2068 | def uppercase_escape(s): |
676eb3f2 | 2069 | unicode_escape = codecs.getdecoder('unicode_escape') |
81c2f20b | 2070 | return re.sub( |
a612753d | 2071 | r'\\U[0-9a-fA-F]{8}', |
676eb3f2 PH |
2072 | lambda m: unicode_escape(m.group(0))[0], |
2073 | s) | |
0fe2ff78 YCH |
2074 | |
2075 | ||
2076 | def lowercase_escape(s): | |
2077 | unicode_escape = codecs.getdecoder('unicode_escape') | |
2078 | return re.sub( | |
2079 | r'\\u[0-9a-fA-F]{4}', | |
2080 | lambda m: unicode_escape(m.group(0))[0], | |
2081 | s) | |
b53466e1 | 2082 | |
d05cfe06 S |
2083 | |
2084 | def escape_rfc3986(s): | |
2085 | """Escape non-ASCII characters as suggested by RFC 3986""" | |
8f9312c3 | 2086 | if sys.version_info < (3, 0) and isinstance(s, compat_str): |
d05cfe06 | 2087 | s = s.encode('utf-8') |
ecc0c5ee | 2088 | return compat_urllib_parse.quote(s, b"%/;:@&=+$,!~*'()?#[]") |
d05cfe06 S |
2089 | |
2090 | ||
2091 | def escape_url(url): | |
2092 | """Escape URL as suggested by RFC 3986""" | |
2093 | url_parsed = compat_urllib_parse_urlparse(url) | |
2094 | return url_parsed._replace( | |
efbed08d | 2095 | netloc=url_parsed.netloc.encode('idna').decode('ascii'), |
d05cfe06 S |
2096 | path=escape_rfc3986(url_parsed.path), |
2097 | params=escape_rfc3986(url_parsed.params), | |
2098 | query=escape_rfc3986(url_parsed.query), | |
2099 | fragment=escape_rfc3986(url_parsed.fragment) | |
2100 | ).geturl() | |
2101 | ||
62e609ab PH |
2102 | |
2103 | def read_batch_urls(batch_fd): | |
2104 | def fixup(url): | |
2105 | if not isinstance(url, compat_str): | |
2106 | url = url.decode('utf-8', 'replace') | |
28e614de | 2107 | BOM_UTF8 = '\xef\xbb\xbf' |
62e609ab PH |
2108 | if url.startswith(BOM_UTF8): |
2109 | url = url[len(BOM_UTF8):] | |
2110 | url = url.strip() | |
2111 | if url.startswith(('#', ';', ']')): | |
2112 | return False | |
2113 | return url | |
2114 | ||
2115 | with contextlib.closing(batch_fd) as fd: | |
2116 | return [url for url in map(fixup, fd) if url] | |
b74fa8cd JMF |
2117 | |
2118 | ||
2119 | def urlencode_postdata(*args, **kargs): | |
15707c7e | 2120 | return compat_urllib_parse_urlencode(*args, **kargs).encode('ascii') |
bcf89ce6 PH |
2121 | |
2122 | ||
38f9ef31 | 2123 | def update_url_query(url, query): |
cacd9966 YCH |
2124 | if not query: |
2125 | return url | |
38f9ef31 | 2126 | parsed_url = compat_urlparse.urlparse(url) |
2127 | qs = compat_parse_qs(parsed_url.query) | |
2128 | qs.update(query) | |
2129 | return compat_urlparse.urlunparse(parsed_url._replace( | |
15707c7e | 2130 | query=compat_urllib_parse_urlencode(qs, True))) |
16392824 | 2131 | |
8e60dc75 | 2132 | |
ed0291d1 S |
2133 | def update_Request(req, url=None, data=None, headers={}, query={}): |
2134 | req_headers = req.headers.copy() | |
2135 | req_headers.update(headers) | |
2136 | req_data = data or req.data | |
2137 | req_url = update_url_query(url or req.get_full_url(), query) | |
95cf60e8 S |
2138 | req_get_method = req.get_method() |
2139 | if req_get_method == 'HEAD': | |
2140 | req_type = HEADRequest | |
2141 | elif req_get_method == 'PUT': | |
2142 | req_type = PUTRequest | |
2143 | else: | |
2144 | req_type = compat_urllib_request.Request | |
ed0291d1 S |
2145 | new_req = req_type( |
2146 | req_url, data=req_data, headers=req_headers, | |
2147 | origin_req_host=req.origin_req_host, unverifiable=req.unverifiable) | |
2148 | if hasattr(req, 'timeout'): | |
2149 | new_req.timeout = req.timeout | |
2150 | return new_req | |
2151 | ||
2152 | ||
10c87c15 | 2153 | def _multipart_encode_impl(data, boundary): |
0c265486 YCH |
2154 | content_type = 'multipart/form-data; boundary=%s' % boundary |
2155 | ||
2156 | out = b'' | |
2157 | for k, v in data.items(): | |
2158 | out += b'--' + boundary.encode('ascii') + b'\r\n' | |
2159 | if isinstance(k, compat_str): | |
2160 | k = k.encode('utf-8') | |
2161 | if isinstance(v, compat_str): | |
2162 | v = v.encode('utf-8') | |
2163 | # RFC 2047 requires non-ASCII field names to be encoded, while RFC 7578 | |
2164 | # suggests sending UTF-8 directly. Firefox sends UTF-8, too | |
b2ad479d | 2165 | content = b'Content-Disposition: form-data; name="' + k + b'"\r\n\r\n' + v + b'\r\n' |
0c265486 YCH |
2166 | if boundary.encode('ascii') in content: |
2167 | raise ValueError('Boundary overlaps with data') | |
2168 | out += content | |
2169 | ||
2170 | out += b'--' + boundary.encode('ascii') + b'--\r\n' | |
2171 | ||
2172 | return out, content_type | |
2173 | ||
2174 | ||
2175 | def multipart_encode(data, boundary=None): | |
2176 | ''' | |
2177 | Encode a dict to RFC 7578-compliant form-data | |
2178 | ||
2179 | data: | |
2180 | A dict where keys and values can be either Unicode or bytes-like | |
2181 | objects. | |
2182 | boundary: | |
2183 | If specified a Unicode object, it's used as the boundary. Otherwise | |
2184 | a random boundary is generated. | |
2185 | ||
2186 | Reference: https://tools.ietf.org/html/rfc7578 | |
2187 | ''' | |
2188 | has_specified_boundary = boundary is not None | |
2189 | ||
2190 | while True: | |
2191 | if boundary is None: | |
2192 | boundary = '---------------' + str(random.randrange(0x0fffffff, 0xffffffff)) | |
2193 | ||
2194 | try: | |
10c87c15 | 2195 | out, content_type = _multipart_encode_impl(data, boundary) |
0c265486 YCH |
2196 | break |
2197 | except ValueError: | |
2198 | if has_specified_boundary: | |
2199 | raise | |
2200 | boundary = None | |
2201 | ||
2202 | return out, content_type | |
2203 | ||
2204 | ||
86296ad2 | 2205 | def dict_get(d, key_or_keys, default=None, skip_false_values=True): |
cbecc9b9 S |
2206 | if isinstance(key_or_keys, (list, tuple)): |
2207 | for key in key_or_keys: | |
86296ad2 S |
2208 | if key not in d or d[key] is None or skip_false_values and not d[key]: |
2209 | continue | |
2210 | return d[key] | |
cbecc9b9 S |
2211 | return default |
2212 | return d.get(key_or_keys, default) | |
2213 | ||
2214 | ||
329ca3be | 2215 | def try_get(src, getter, expected_type=None): |
a32a9a7e S |
2216 | if not isinstance(getter, (list, tuple)): |
2217 | getter = [getter] | |
2218 | for get in getter: | |
2219 | try: | |
2220 | v = get(src) | |
2221 | except (AttributeError, KeyError, TypeError, IndexError): | |
2222 | pass | |
2223 | else: | |
2224 | if expected_type is None or isinstance(v, expected_type): | |
2225 | return v | |
329ca3be S |
2226 | |
2227 | ||
8e60dc75 S |
2228 | def encode_compat_str(string, encoding=preferredencoding(), errors='strict'): |
2229 | return string if isinstance(string, compat_str) else compat_str(string, encoding, errors) | |
2230 | ||
16392824 | 2231 | |
a1a530b0 PH |
2232 | US_RATINGS = { |
2233 | 'G': 0, | |
2234 | 'PG': 10, | |
2235 | 'PG-13': 13, | |
2236 | 'R': 16, | |
2237 | 'NC': 18, | |
2238 | } | |
fac55558 PH |
2239 | |
2240 | ||
a8795327 S |
2241 | TV_PARENTAL_GUIDELINES = { |
2242 | 'TV-Y': 0, | |
2243 | 'TV-Y7': 7, | |
2244 | 'TV-G': 0, | |
2245 | 'TV-PG': 0, | |
2246 | 'TV-14': 14, | |
2247 | 'TV-MA': 17, | |
2248 | } | |
2249 | ||
2250 | ||
146c80e2 | 2251 | def parse_age_limit(s): |
a8795327 S |
2252 | if type(s) == int: |
2253 | return s if 0 <= s <= 21 else None | |
2254 | if not isinstance(s, compat_basestring): | |
d838b1bd | 2255 | return None |
146c80e2 | 2256 | m = re.match(r'^(?P<age>\d{1,2})\+?$', s) |
a8795327 S |
2257 | if m: |
2258 | return int(m.group('age')) | |
2259 | if s in US_RATINGS: | |
2260 | return US_RATINGS[s] | |
2261 | return TV_PARENTAL_GUIDELINES.get(s) | |
146c80e2 S |
2262 | |
2263 | ||
fac55558 | 2264 | def strip_jsonp(code): |
609a61e3 | 2265 | return re.sub( |
5552c9eb YCH |
2266 | r'''(?sx)^ |
2267 | (?:window\.)?(?P<func_name>[a-zA-Z0-9_.$]+) | |
2268 | (?:\s*&&\s*(?P=func_name))? | |
2269 | \s*\(\s*(?P<callback_data>.*)\);? | |
2270 | \s*?(?://[^\n]*)*$''', | |
2271 | r'\g<callback_data>', code) | |
478c2c61 PH |
2272 | |
2273 | ||
e05f6939 | 2274 | def js_to_json(code): |
4195096e S |
2275 | COMMENT_RE = r'/\*(?:(?!\*/).)*?\*/|//[^\n]*' |
2276 | SKIP_RE = r'\s*(?:{comment})?\s*'.format(comment=COMMENT_RE) | |
2277 | INTEGER_TABLE = ( | |
2278 | (r'(?s)^(0[xX][0-9a-fA-F]+){skip}:?$'.format(skip=SKIP_RE), 16), | |
2279 | (r'(?s)^(0+[0-7]+){skip}:?$'.format(skip=SKIP_RE), 8), | |
2280 | ) | |
2281 | ||
e05f6939 | 2282 | def fix_kv(m): |
e7b6d122 PH |
2283 | v = m.group(0) |
2284 | if v in ('true', 'false', 'null'): | |
2285 | return v | |
b3ee552e | 2286 | elif v.startswith('/*') or v.startswith('//') or v == ',': |
bd1e4844 | 2287 | return "" |
2288 | ||
2289 | if v[0] in ("'", '"'): | |
2290 | v = re.sub(r'(?s)\\.|"', lambda m: { | |
e7b6d122 | 2291 | '"': '\\"', |
bd1e4844 | 2292 | "\\'": "'", |
2293 | '\\\n': '', | |
2294 | '\\x': '\\u00', | |
2295 | }.get(m.group(0), m.group(0)), v[1:-1]) | |
2296 | ||
89ac4a19 S |
2297 | for regex, base in INTEGER_TABLE: |
2298 | im = re.match(regex, v) | |
2299 | if im: | |
e4659b45 | 2300 | i = int(im.group(1), base) |
89ac4a19 S |
2301 | return '"%d":' % i if v.endswith(':') else '%d' % i |
2302 | ||
e7b6d122 | 2303 | return '"%s"' % v |
e05f6939 | 2304 | |
bd1e4844 | 2305 | return re.sub(r'''(?sx) |
2306 | "(?:[^"\\]*(?:\\\\|\\['"nurtbfx/\n]))*[^"\\]*"| | |
2307 | '(?:[^'\\]*(?:\\\\|\\['"nurtbfx/\n]))*[^'\\]*'| | |
4195096e | 2308 | {comment}|,(?={skip}[\]}}])| |
c384d537 | 2309 | (?:(?<![0-9])[eE]|[a-df-zA-DF-Z_])[.a-zA-Z_0-9]*| |
4195096e S |
2310 | \b(?:0[xX][0-9a-fA-F]+|0+[0-7]+)(?:{skip}:)?| |
2311 | [0-9]+(?={skip}:) | |
2312 | '''.format(comment=COMMENT_RE, skip=SKIP_RE), fix_kv, code) | |
e05f6939 PH |
2313 | |
2314 | ||
478c2c61 PH |
2315 | def qualities(quality_ids): |
2316 | """ Get a numeric quality value out of a list of possible values """ | |
2317 | def q(qid): | |
2318 | try: | |
2319 | return quality_ids.index(qid) | |
2320 | except ValueError: | |
2321 | return -1 | |
2322 | return q | |
2323 | ||
acd69589 PH |
2324 | |
2325 | DEFAULT_OUTTMPL = '%(title)s-%(id)s.%(ext)s' | |
0a871f68 | 2326 | |
a020a0dc PH |
2327 | |
2328 | def limit_length(s, length): | |
2329 | """ Add ellipses to overly long strings """ | |
2330 | if s is None: | |
2331 | return None | |
2332 | ELLIPSES = '...' | |
2333 | if len(s) > length: | |
2334 | return s[:length - len(ELLIPSES)] + ELLIPSES | |
2335 | return s | |
48844745 PH |
2336 | |
2337 | ||
2338 | def version_tuple(v): | |
5f9b8394 | 2339 | return tuple(int(e) for e in re.split(r'[-.]', v)) |
48844745 PH |
2340 | |
2341 | ||
2342 | def is_outdated_version(version, limit, assume_new=True): | |
2343 | if not version: | |
2344 | return not assume_new | |
2345 | try: | |
2346 | return version_tuple(version) < version_tuple(limit) | |
2347 | except ValueError: | |
2348 | return not assume_new | |
732ea2f0 PH |
2349 | |
2350 | ||
2351 | def ytdl_is_updateable(): | |
2352 | """ Returns if youtube-dl can be updated with -U """ | |
2353 | from zipimport import zipimporter | |
2354 | ||
2355 | return isinstance(globals().get('__loader__'), zipimporter) or hasattr(sys, 'frozen') | |
7d4111ed PH |
2356 | |
2357 | ||
2358 | def args_to_str(args): | |
2359 | # Get a short string representation for a subprocess command | |
702ccf2d | 2360 | return ' '.join(compat_shlex_quote(a) for a in args) |
2ccd1b10 PH |
2361 | |
2362 | ||
9b9c5355 | 2363 | def error_to_compat_str(err): |
fdae2358 S |
2364 | err_str = str(err) |
2365 | # On python 2 error byte string must be decoded with proper | |
2366 | # encoding rather than ascii | |
2367 | if sys.version_info[0] < 3: | |
2368 | err_str = err_str.decode(preferredencoding()) | |
2369 | return err_str | |
2370 | ||
2371 | ||
c460bdd5 | 2372 | def mimetype2ext(mt): |
eb9ee194 S |
2373 | if mt is None: |
2374 | return None | |
2375 | ||
765ac263 JMF |
2376 | ext = { |
2377 | 'audio/mp4': 'm4a', | |
6c33d24b YCH |
2378 | # Per RFC 3003, audio/mpeg can be .mp1, .mp2 or .mp3. Here use .mp3 as |
2379 | # it's the most popular one | |
2380 | 'audio/mpeg': 'mp3', | |
765ac263 JMF |
2381 | }.get(mt) |
2382 | if ext is not None: | |
2383 | return ext | |
2384 | ||
c460bdd5 | 2385 | _, _, res = mt.rpartition('/') |
6562d34a | 2386 | res = res.split(';')[0].strip().lower() |
c460bdd5 PH |
2387 | |
2388 | return { | |
f6861ec9 | 2389 | '3gpp': '3gp', |
cafcf657 | 2390 | 'smptett+xml': 'tt', |
cafcf657 | 2391 | 'ttaf+xml': 'dfxp', |
a0d8d704 | 2392 | 'ttml+xml': 'ttml', |
f6861ec9 | 2393 | 'x-flv': 'flv', |
a0d8d704 | 2394 | 'x-mp4-fragmented': 'mp4', |
d4f05d47 | 2395 | 'x-ms-sami': 'sami', |
a0d8d704 | 2396 | 'x-ms-wmv': 'wmv', |
b4173f15 RA |
2397 | 'mpegurl': 'm3u8', |
2398 | 'x-mpegurl': 'm3u8', | |
2399 | 'vnd.apple.mpegurl': 'm3u8', | |
2400 | 'dash+xml': 'mpd', | |
b4173f15 | 2401 | 'f4m+xml': 'f4m', |
f164b971 | 2402 | 'hds+xml': 'f4m', |
e910fe2f | 2403 | 'vnd.ms-sstr+xml': 'ism', |
c2b2c7e1 | 2404 | 'quicktime': 'mov', |
98ce1a3f | 2405 | 'mp2t': 'ts', |
c460bdd5 PH |
2406 | }.get(res, res) |
2407 | ||
2408 | ||
4f3c5e06 | 2409 | def parse_codecs(codecs_str): |
2410 | # http://tools.ietf.org/html/rfc6381 | |
2411 | if not codecs_str: | |
2412 | return {} | |
2413 | splited_codecs = list(filter(None, map( | |
2414 | lambda str: str.strip(), codecs_str.strip().strip(',').split(',')))) | |
2415 | vcodec, acodec = None, None | |
2416 | for full_codec in splited_codecs: | |
2417 | codec = full_codec.split('.')[0] | |
ffe6979e | 2418 | if codec in ('avc1', 'avc2', 'avc3', 'avc4', 'vp9', 'vp8', 'hev1', 'hev2', 'h263', 'h264', 'mp4v', 'hvc1'): |
4f3c5e06 | 2419 | if not vcodec: |
2420 | vcodec = full_codec | |
60f5c9fb | 2421 | elif codec in ('mp4a', 'opus', 'vorbis', 'mp3', 'aac', 'ac-3', 'ec-3', 'eac3', 'dtsc', 'dtse', 'dtsh', 'dtsl'): |
4f3c5e06 | 2422 | if not acodec: |
2423 | acodec = full_codec | |
2424 | else: | |
60f5c9fb | 2425 | write_string('WARNING: Unknown codec %s\n' % full_codec, sys.stderr) |
4f3c5e06 | 2426 | if not vcodec and not acodec: |
2427 | if len(splited_codecs) == 2: | |
2428 | return { | |
2429 | 'vcodec': vcodec, | |
2430 | 'acodec': acodec, | |
2431 | } | |
2432 | elif len(splited_codecs) == 1: | |
2433 | return { | |
2434 | 'vcodec': 'none', | |
2435 | 'acodec': vcodec, | |
2436 | } | |
2437 | else: | |
2438 | return { | |
2439 | 'vcodec': vcodec or 'none', | |
2440 | 'acodec': acodec or 'none', | |
2441 | } | |
2442 | return {} | |
2443 | ||
2444 | ||
2ccd1b10 | 2445 | def urlhandle_detect_ext(url_handle): |
79298173 | 2446 | getheader = url_handle.headers.get |
2ccd1b10 | 2447 | |
b55ee18f PH |
2448 | cd = getheader('Content-Disposition') |
2449 | if cd: | |
2450 | m = re.match(r'attachment;\s*filename="(?P<filename>[^"]+)"', cd) | |
2451 | if m: | |
2452 | e = determine_ext(m.group('filename'), default_ext=None) | |
2453 | if e: | |
2454 | return e | |
2455 | ||
c460bdd5 | 2456 | return mimetype2ext(getheader('Content-Type')) |
05900629 PH |
2457 | |
2458 | ||
1e399778 YCH |
2459 | def encode_data_uri(data, mime_type): |
2460 | return 'data:%s;base64,%s' % (mime_type, base64.b64encode(data).decode('ascii')) | |
2461 | ||
2462 | ||
05900629 | 2463 | def age_restricted(content_limit, age_limit): |
6ec6cb4e | 2464 | """ Returns True iff the content should be blocked """ |
05900629 PH |
2465 | |
2466 | if age_limit is None: # No limit set | |
2467 | return False | |
2468 | if content_limit is None: | |
2469 | return False # Content available for everyone | |
2470 | return age_limit < content_limit | |
61ca9a80 PH |
2471 | |
2472 | ||
2473 | def is_html(first_bytes): | |
2474 | """ Detect whether a file contains HTML by examining its first bytes. """ | |
2475 | ||
2476 | BOMS = [ | |
2477 | (b'\xef\xbb\xbf', 'utf-8'), | |
2478 | (b'\x00\x00\xfe\xff', 'utf-32-be'), | |
2479 | (b'\xff\xfe\x00\x00', 'utf-32-le'), | |
2480 | (b'\xff\xfe', 'utf-16-le'), | |
2481 | (b'\xfe\xff', 'utf-16-be'), | |
2482 | ] | |
2483 | for bom, enc in BOMS: | |
2484 | if first_bytes.startswith(bom): | |
2485 | s = first_bytes[len(bom):].decode(enc, 'replace') | |
2486 | break | |
2487 | else: | |
2488 | s = first_bytes.decode('utf-8', 'replace') | |
2489 | ||
2490 | return re.match(r'^\s*<', s) | |
a055469f PH |
2491 | |
2492 | ||
2493 | def determine_protocol(info_dict): | |
2494 | protocol = info_dict.get('protocol') | |
2495 | if protocol is not None: | |
2496 | return protocol | |
2497 | ||
2498 | url = info_dict['url'] | |
2499 | if url.startswith('rtmp'): | |
2500 | return 'rtmp' | |
2501 | elif url.startswith('mms'): | |
2502 | return 'mms' | |
2503 | elif url.startswith('rtsp'): | |
2504 | return 'rtsp' | |
2505 | ||
2506 | ext = determine_ext(url) | |
2507 | if ext == 'm3u8': | |
2508 | return 'm3u8' | |
2509 | elif ext == 'f4m': | |
2510 | return 'f4m' | |
2511 | ||
2512 | return compat_urllib_parse_urlparse(url).scheme | |
cfb56d1a PH |
2513 | |
2514 | ||
2515 | def render_table(header_row, data): | |
2516 | """ Render a list of rows, each as a list of values """ | |
2517 | table = [header_row] + data | |
2518 | max_lens = [max(len(compat_str(v)) for v in col) for col in zip(*table)] | |
2519 | format_str = ' '.join('%-' + compat_str(ml + 1) + 's' for ml in max_lens[:-1]) + '%s' | |
2520 | return '\n'.join(format_str % tuple(row) for row in table) | |
347de493 PH |
2521 | |
2522 | ||
2523 | def _match_one(filter_part, dct): | |
2524 | COMPARISON_OPERATORS = { | |
2525 | '<': operator.lt, | |
2526 | '<=': operator.le, | |
2527 | '>': operator.gt, | |
2528 | '>=': operator.ge, | |
2529 | '=': operator.eq, | |
2530 | '!=': operator.ne, | |
2531 | } | |
2532 | operator_rex = re.compile(r'''(?x)\s* | |
2533 | (?P<key>[a-z_]+) | |
2534 | \s*(?P<op>%s)(?P<none_inclusive>\s*\?)?\s* | |
2535 | (?: | |
2536 | (?P<intval>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?)| | |
db13c16e | 2537 | (?P<quote>["\'])(?P<quotedstrval>(?:\\.|(?!(?P=quote)|\\).)+?)(?P=quote)| |
347de493 PH |
2538 | (?P<strval>(?![0-9.])[a-z0-9A-Z]*) |
2539 | ) | |
2540 | \s*$ | |
2541 | ''' % '|'.join(map(re.escape, COMPARISON_OPERATORS.keys()))) | |
2542 | m = operator_rex.search(filter_part) | |
2543 | if m: | |
2544 | op = COMPARISON_OPERATORS[m.group('op')] | |
e5a088dc | 2545 | actual_value = dct.get(m.group('key')) |
db13c16e S |
2546 | if (m.group('quotedstrval') is not None or |
2547 | m.group('strval') is not None or | |
e5a088dc S |
2548 | # If the original field is a string and matching comparisonvalue is |
2549 | # a number we should respect the origin of the original field | |
2550 | # and process comparison value as a string (see | |
2551 | # https://github.com/rg3/youtube-dl/issues/11082). | |
2552 | actual_value is not None and m.group('intval') is not None and | |
2553 | isinstance(actual_value, compat_str)): | |
347de493 PH |
2554 | if m.group('op') not in ('=', '!='): |
2555 | raise ValueError( | |
2556 | 'Operator %s does not support string values!' % m.group('op')) | |
db13c16e S |
2557 | comparison_value = m.group('quotedstrval') or m.group('strval') or m.group('intval') |
2558 | quote = m.group('quote') | |
2559 | if quote is not None: | |
2560 | comparison_value = comparison_value.replace(r'\%s' % quote, quote) | |
347de493 PH |
2561 | else: |
2562 | try: | |
2563 | comparison_value = int(m.group('intval')) | |
2564 | except ValueError: | |
2565 | comparison_value = parse_filesize(m.group('intval')) | |
2566 | if comparison_value is None: | |
2567 | comparison_value = parse_filesize(m.group('intval') + 'B') | |
2568 | if comparison_value is None: | |
2569 | raise ValueError( | |
2570 | 'Invalid integer value %r in filter part %r' % ( | |
2571 | m.group('intval'), filter_part)) | |
347de493 PH |
2572 | if actual_value is None: |
2573 | return m.group('none_inclusive') | |
2574 | return op(actual_value, comparison_value) | |
2575 | ||
2576 | UNARY_OPERATORS = { | |
1cc47c66 S |
2577 | '': lambda v: (v is True) if isinstance(v, bool) else (v is not None), |
2578 | '!': lambda v: (v is False) if isinstance(v, bool) else (v is None), | |
347de493 PH |
2579 | } |
2580 | operator_rex = re.compile(r'''(?x)\s* | |
2581 | (?P<op>%s)\s*(?P<key>[a-z_]+) | |
2582 | \s*$ | |
2583 | ''' % '|'.join(map(re.escape, UNARY_OPERATORS.keys()))) | |
2584 | m = operator_rex.search(filter_part) | |
2585 | if m: | |
2586 | op = UNARY_OPERATORS[m.group('op')] | |
2587 | actual_value = dct.get(m.group('key')) | |
2588 | return op(actual_value) | |
2589 | ||
2590 | raise ValueError('Invalid filter part %r' % filter_part) | |
2591 | ||
2592 | ||
2593 | def match_str(filter_str, dct): | |
2594 | """ Filter a dictionary with a simple string syntax. Returns True (=passes filter) or false """ | |
2595 | ||
2596 | return all( | |
2597 | _match_one(filter_part, dct) for filter_part in filter_str.split('&')) | |
2598 | ||
2599 | ||
2600 | def match_filter_func(filter_str): | |
2601 | def _match_func(info_dict): | |
2602 | if match_str(filter_str, info_dict): | |
2603 | return None | |
2604 | else: | |
2605 | video_title = info_dict.get('title', info_dict.get('id', 'video')) | |
2606 | return '%s does not pass filter %s, skipping ..' % (video_title, filter_str) | |
2607 | return _match_func | |
91410c9b PH |
2608 | |
2609 | ||
bf6427d2 YCH |
2610 | def parse_dfxp_time_expr(time_expr): |
2611 | if not time_expr: | |
d631d5f9 | 2612 | return |
bf6427d2 YCH |
2613 | |
2614 | mobj = re.match(r'^(?P<time_offset>\d+(?:\.\d+)?)s?$', time_expr) | |
2615 | if mobj: | |
2616 | return float(mobj.group('time_offset')) | |
2617 | ||
db2fe38b | 2618 | mobj = re.match(r'^(\d+):(\d\d):(\d\d(?:(?:\.|:)\d+)?)$', time_expr) |
bf6427d2 | 2619 | if mobj: |
db2fe38b | 2620 | return 3600 * int(mobj.group(1)) + 60 * int(mobj.group(2)) + float(mobj.group(3).replace(':', '.')) |
bf6427d2 YCH |
2621 | |
2622 | ||
c1c924ab YCH |
2623 | def srt_subtitles_timecode(seconds): |
2624 | return '%02d:%02d:%02d,%03d' % (seconds / 3600, (seconds % 3600) / 60, seconds % 60, (seconds % 1) * 1000) | |
bf6427d2 YCH |
2625 | |
2626 | ||
2627 | def dfxp2srt(dfxp_data): | |
3869028f YCH |
2628 | ''' |
2629 | @param dfxp_data A bytes-like object containing DFXP data | |
2630 | @returns A unicode object containing converted SRT data | |
2631 | ''' | |
5b995f71 | 2632 | LEGACY_NAMESPACES = ( |
3869028f YCH |
2633 | (b'http://www.w3.org/ns/ttml', [ |
2634 | b'http://www.w3.org/2004/11/ttaf1', | |
2635 | b'http://www.w3.org/2006/04/ttaf1', | |
2636 | b'http://www.w3.org/2006/10/ttaf1', | |
5b995f71 | 2637 | ]), |
3869028f YCH |
2638 | (b'http://www.w3.org/ns/ttml#styling', [ |
2639 | b'http://www.w3.org/ns/ttml#style', | |
5b995f71 RA |
2640 | ]), |
2641 | ) | |
2642 | ||
2643 | SUPPORTED_STYLING = [ | |
2644 | 'color', | |
2645 | 'fontFamily', | |
2646 | 'fontSize', | |
2647 | 'fontStyle', | |
2648 | 'fontWeight', | |
2649 | 'textDecoration' | |
2650 | ] | |
2651 | ||
4e335771 YCH |
2652 | _x = functools.partial(xpath_with_ns, ns_map={ |
2653 | 'ttml': 'http://www.w3.org/ns/ttml', | |
5b995f71 | 2654 | 'tts': 'http://www.w3.org/ns/ttml#styling', |
4e335771 | 2655 | }) |
bf6427d2 | 2656 | |
5b995f71 RA |
2657 | styles = {} |
2658 | default_style = {} | |
2659 | ||
87de7069 | 2660 | class TTMLPElementParser(object): |
5b995f71 RA |
2661 | _out = '' |
2662 | _unclosed_elements = [] | |
2663 | _applied_styles = [] | |
bf6427d2 | 2664 | |
2b14cb56 | 2665 | def start(self, tag, attrib): |
5b995f71 RA |
2666 | if tag in (_x('ttml:br'), 'br'): |
2667 | self._out += '\n' | |
2668 | else: | |
2669 | unclosed_elements = [] | |
2670 | style = {} | |
2671 | element_style_id = attrib.get('style') | |
2672 | if default_style: | |
2673 | style.update(default_style) | |
2674 | if element_style_id: | |
2675 | style.update(styles.get(element_style_id, {})) | |
2676 | for prop in SUPPORTED_STYLING: | |
2677 | prop_val = attrib.get(_x('tts:' + prop)) | |
2678 | if prop_val: | |
2679 | style[prop] = prop_val | |
2680 | if style: | |
2681 | font = '' | |
2682 | for k, v in sorted(style.items()): | |
2683 | if self._applied_styles and self._applied_styles[-1].get(k) == v: | |
2684 | continue | |
2685 | if k == 'color': | |
2686 | font += ' color="%s"' % v | |
2687 | elif k == 'fontSize': | |
2688 | font += ' size="%s"' % v | |
2689 | elif k == 'fontFamily': | |
2690 | font += ' face="%s"' % v | |
2691 | elif k == 'fontWeight' and v == 'bold': | |
2692 | self._out += '<b>' | |
2693 | unclosed_elements.append('b') | |
2694 | elif k == 'fontStyle' and v == 'italic': | |
2695 | self._out += '<i>' | |
2696 | unclosed_elements.append('i') | |
2697 | elif k == 'textDecoration' and v == 'underline': | |
2698 | self._out += '<u>' | |
2699 | unclosed_elements.append('u') | |
2700 | if font: | |
2701 | self._out += '<font' + font + '>' | |
2702 | unclosed_elements.append('font') | |
2703 | applied_style = {} | |
2704 | if self._applied_styles: | |
2705 | applied_style.update(self._applied_styles[-1]) | |
2706 | applied_style.update(style) | |
2707 | self._applied_styles.append(applied_style) | |
2708 | self._unclosed_elements.append(unclosed_elements) | |
bf6427d2 | 2709 | |
2b14cb56 | 2710 | def end(self, tag): |
5b995f71 RA |
2711 | if tag not in (_x('ttml:br'), 'br'): |
2712 | unclosed_elements = self._unclosed_elements.pop() | |
2713 | for element in reversed(unclosed_elements): | |
2714 | self._out += '</%s>' % element | |
2715 | if unclosed_elements and self._applied_styles: | |
2716 | self._applied_styles.pop() | |
bf6427d2 | 2717 | |
2b14cb56 | 2718 | def data(self, data): |
5b995f71 | 2719 | self._out += data |
2b14cb56 | 2720 | |
2721 | def close(self): | |
5b995f71 | 2722 | return self._out.strip() |
2b14cb56 | 2723 | |
2724 | def parse_node(node): | |
2725 | target = TTMLPElementParser() | |
2726 | parser = xml.etree.ElementTree.XMLParser(target=target) | |
2727 | parser.feed(xml.etree.ElementTree.tostring(node)) | |
2728 | return parser.close() | |
bf6427d2 | 2729 | |
5b995f71 RA |
2730 | for k, v in LEGACY_NAMESPACES: |
2731 | for ns in v: | |
2732 | dfxp_data = dfxp_data.replace(ns, k) | |
2733 | ||
3869028f | 2734 | dfxp = compat_etree_fromstring(dfxp_data) |
bf6427d2 | 2735 | out = [] |
5b995f71 | 2736 | paras = dfxp.findall(_x('.//ttml:p')) or dfxp.findall('.//p') |
1b0427e6 YCH |
2737 | |
2738 | if not paras: | |
2739 | raise ValueError('Invalid dfxp/TTML subtitle') | |
bf6427d2 | 2740 | |
5b995f71 RA |
2741 | repeat = False |
2742 | while True: | |
2743 | for style in dfxp.findall(_x('.//ttml:style')): | |
2744 | style_id = style.get('id') | |
2745 | parent_style_id = style.get('style') | |
2746 | if parent_style_id: | |
2747 | if parent_style_id not in styles: | |
2748 | repeat = True | |
2749 | continue | |
2750 | styles[style_id] = styles[parent_style_id].copy() | |
2751 | for prop in SUPPORTED_STYLING: | |
2752 | prop_val = style.get(_x('tts:' + prop)) | |
2753 | if prop_val: | |
2754 | styles.setdefault(style_id, {})[prop] = prop_val | |
2755 | if repeat: | |
2756 | repeat = False | |
2757 | else: | |
2758 | break | |
2759 | ||
2760 | for p in ('body', 'div'): | |
2761 | ele = xpath_element(dfxp, [_x('.//ttml:' + p), './/' + p]) | |
2762 | if ele is None: | |
2763 | continue | |
2764 | style = styles.get(ele.get('style')) | |
2765 | if not style: | |
2766 | continue | |
2767 | default_style.update(style) | |
2768 | ||
bf6427d2 | 2769 | for para, index in zip(paras, itertools.count(1)): |
d631d5f9 | 2770 | begin_time = parse_dfxp_time_expr(para.attrib.get('begin')) |
7dff0363 | 2771 | end_time = parse_dfxp_time_expr(para.attrib.get('end')) |
d631d5f9 YCH |
2772 | dur = parse_dfxp_time_expr(para.attrib.get('dur')) |
2773 | if begin_time is None: | |
2774 | continue | |
7dff0363 | 2775 | if not end_time: |
d631d5f9 YCH |
2776 | if not dur: |
2777 | continue | |
2778 | end_time = begin_time + dur | |
bf6427d2 YCH |
2779 | out.append('%d\n%s --> %s\n%s\n\n' % ( |
2780 | index, | |
c1c924ab YCH |
2781 | srt_subtitles_timecode(begin_time), |
2782 | srt_subtitles_timecode(end_time), | |
bf6427d2 YCH |
2783 | parse_node(para))) |
2784 | ||
2785 | return ''.join(out) | |
2786 | ||
2787 | ||
66e289ba S |
2788 | def cli_option(params, command_option, param): |
2789 | param = params.get(param) | |
98e698f1 RA |
2790 | if param: |
2791 | param = compat_str(param) | |
66e289ba S |
2792 | return [command_option, param] if param is not None else [] |
2793 | ||
2794 | ||
2795 | def cli_bool_option(params, command_option, param, true_value='true', false_value='false', separator=None): | |
2796 | param = params.get(param) | |
5b232f46 S |
2797 | if param is None: |
2798 | return [] | |
66e289ba S |
2799 | assert isinstance(param, bool) |
2800 | if separator: | |
2801 | return [command_option + separator + (true_value if param else false_value)] | |
2802 | return [command_option, true_value if param else false_value] | |
2803 | ||
2804 | ||
2805 | def cli_valueless_option(params, command_option, param, expected_value=True): | |
2806 | param = params.get(param) | |
2807 | return [command_option] if param == expected_value else [] | |
2808 | ||
2809 | ||
2810 | def cli_configuration_args(params, param, default=[]): | |
2811 | ex_args = params.get(param) | |
2812 | if ex_args is None: | |
2813 | return default | |
2814 | assert isinstance(ex_args, list) | |
2815 | return ex_args | |
2816 | ||
2817 | ||
39672624 YCH |
2818 | class ISO639Utils(object): |
2819 | # See http://www.loc.gov/standards/iso639-2/ISO-639-2_utf-8.txt | |
2820 | _lang_map = { | |
2821 | 'aa': 'aar', | |
2822 | 'ab': 'abk', | |
2823 | 'ae': 'ave', | |
2824 | 'af': 'afr', | |
2825 | 'ak': 'aka', | |
2826 | 'am': 'amh', | |
2827 | 'an': 'arg', | |
2828 | 'ar': 'ara', | |
2829 | 'as': 'asm', | |
2830 | 'av': 'ava', | |
2831 | 'ay': 'aym', | |
2832 | 'az': 'aze', | |
2833 | 'ba': 'bak', | |
2834 | 'be': 'bel', | |
2835 | 'bg': 'bul', | |
2836 | 'bh': 'bih', | |
2837 | 'bi': 'bis', | |
2838 | 'bm': 'bam', | |
2839 | 'bn': 'ben', | |
2840 | 'bo': 'bod', | |
2841 | 'br': 'bre', | |
2842 | 'bs': 'bos', | |
2843 | 'ca': 'cat', | |
2844 | 'ce': 'che', | |
2845 | 'ch': 'cha', | |
2846 | 'co': 'cos', | |
2847 | 'cr': 'cre', | |
2848 | 'cs': 'ces', | |
2849 | 'cu': 'chu', | |
2850 | 'cv': 'chv', | |
2851 | 'cy': 'cym', | |
2852 | 'da': 'dan', | |
2853 | 'de': 'deu', | |
2854 | 'dv': 'div', | |
2855 | 'dz': 'dzo', | |
2856 | 'ee': 'ewe', | |
2857 | 'el': 'ell', | |
2858 | 'en': 'eng', | |
2859 | 'eo': 'epo', | |
2860 | 'es': 'spa', | |
2861 | 'et': 'est', | |
2862 | 'eu': 'eus', | |
2863 | 'fa': 'fas', | |
2864 | 'ff': 'ful', | |
2865 | 'fi': 'fin', | |
2866 | 'fj': 'fij', | |
2867 | 'fo': 'fao', | |
2868 | 'fr': 'fra', | |
2869 | 'fy': 'fry', | |
2870 | 'ga': 'gle', | |
2871 | 'gd': 'gla', | |
2872 | 'gl': 'glg', | |
2873 | 'gn': 'grn', | |
2874 | 'gu': 'guj', | |
2875 | 'gv': 'glv', | |
2876 | 'ha': 'hau', | |
2877 | 'he': 'heb', | |
2878 | 'hi': 'hin', | |
2879 | 'ho': 'hmo', | |
2880 | 'hr': 'hrv', | |
2881 | 'ht': 'hat', | |
2882 | 'hu': 'hun', | |
2883 | 'hy': 'hye', | |
2884 | 'hz': 'her', | |
2885 | 'ia': 'ina', | |
2886 | 'id': 'ind', | |
2887 | 'ie': 'ile', | |
2888 | 'ig': 'ibo', | |
2889 | 'ii': 'iii', | |
2890 | 'ik': 'ipk', | |
2891 | 'io': 'ido', | |
2892 | 'is': 'isl', | |
2893 | 'it': 'ita', | |
2894 | 'iu': 'iku', | |
2895 | 'ja': 'jpn', | |
2896 | 'jv': 'jav', | |
2897 | 'ka': 'kat', | |
2898 | 'kg': 'kon', | |
2899 | 'ki': 'kik', | |
2900 | 'kj': 'kua', | |
2901 | 'kk': 'kaz', | |
2902 | 'kl': 'kal', | |
2903 | 'km': 'khm', | |
2904 | 'kn': 'kan', | |
2905 | 'ko': 'kor', | |
2906 | 'kr': 'kau', | |
2907 | 'ks': 'kas', | |
2908 | 'ku': 'kur', | |
2909 | 'kv': 'kom', | |
2910 | 'kw': 'cor', | |
2911 | 'ky': 'kir', | |
2912 | 'la': 'lat', | |
2913 | 'lb': 'ltz', | |
2914 | 'lg': 'lug', | |
2915 | 'li': 'lim', | |
2916 | 'ln': 'lin', | |
2917 | 'lo': 'lao', | |
2918 | 'lt': 'lit', | |
2919 | 'lu': 'lub', | |
2920 | 'lv': 'lav', | |
2921 | 'mg': 'mlg', | |
2922 | 'mh': 'mah', | |
2923 | 'mi': 'mri', | |
2924 | 'mk': 'mkd', | |
2925 | 'ml': 'mal', | |
2926 | 'mn': 'mon', | |
2927 | 'mr': 'mar', | |
2928 | 'ms': 'msa', | |
2929 | 'mt': 'mlt', | |
2930 | 'my': 'mya', | |
2931 | 'na': 'nau', | |
2932 | 'nb': 'nob', | |
2933 | 'nd': 'nde', | |
2934 | 'ne': 'nep', | |
2935 | 'ng': 'ndo', | |
2936 | 'nl': 'nld', | |
2937 | 'nn': 'nno', | |
2938 | 'no': 'nor', | |
2939 | 'nr': 'nbl', | |
2940 | 'nv': 'nav', | |
2941 | 'ny': 'nya', | |
2942 | 'oc': 'oci', | |
2943 | 'oj': 'oji', | |
2944 | 'om': 'orm', | |
2945 | 'or': 'ori', | |
2946 | 'os': 'oss', | |
2947 | 'pa': 'pan', | |
2948 | 'pi': 'pli', | |
2949 | 'pl': 'pol', | |
2950 | 'ps': 'pus', | |
2951 | 'pt': 'por', | |
2952 | 'qu': 'que', | |
2953 | 'rm': 'roh', | |
2954 | 'rn': 'run', | |
2955 | 'ro': 'ron', | |
2956 | 'ru': 'rus', | |
2957 | 'rw': 'kin', | |
2958 | 'sa': 'san', | |
2959 | 'sc': 'srd', | |
2960 | 'sd': 'snd', | |
2961 | 'se': 'sme', | |
2962 | 'sg': 'sag', | |
2963 | 'si': 'sin', | |
2964 | 'sk': 'slk', | |
2965 | 'sl': 'slv', | |
2966 | 'sm': 'smo', | |
2967 | 'sn': 'sna', | |
2968 | 'so': 'som', | |
2969 | 'sq': 'sqi', | |
2970 | 'sr': 'srp', | |
2971 | 'ss': 'ssw', | |
2972 | 'st': 'sot', | |
2973 | 'su': 'sun', | |
2974 | 'sv': 'swe', | |
2975 | 'sw': 'swa', | |
2976 | 'ta': 'tam', | |
2977 | 'te': 'tel', | |
2978 | 'tg': 'tgk', | |
2979 | 'th': 'tha', | |
2980 | 'ti': 'tir', | |
2981 | 'tk': 'tuk', | |
2982 | 'tl': 'tgl', | |
2983 | 'tn': 'tsn', | |
2984 | 'to': 'ton', | |
2985 | 'tr': 'tur', | |
2986 | 'ts': 'tso', | |
2987 | 'tt': 'tat', | |
2988 | 'tw': 'twi', | |
2989 | 'ty': 'tah', | |
2990 | 'ug': 'uig', | |
2991 | 'uk': 'ukr', | |
2992 | 'ur': 'urd', | |
2993 | 'uz': 'uzb', | |
2994 | 've': 'ven', | |
2995 | 'vi': 'vie', | |
2996 | 'vo': 'vol', | |
2997 | 'wa': 'wln', | |
2998 | 'wo': 'wol', | |
2999 | 'xh': 'xho', | |
3000 | 'yi': 'yid', | |
3001 | 'yo': 'yor', | |
3002 | 'za': 'zha', | |
3003 | 'zh': 'zho', | |
3004 | 'zu': 'zul', | |
3005 | } | |
3006 | ||
3007 | @classmethod | |
3008 | def short2long(cls, code): | |
3009 | """Convert language code from ISO 639-1 to ISO 639-2/T""" | |
3010 | return cls._lang_map.get(code[:2]) | |
3011 | ||
3012 | @classmethod | |
3013 | def long2short(cls, code): | |
3014 | """Convert language code from ISO 639-2/T to ISO 639-1""" | |
3015 | for short_name, long_name in cls._lang_map.items(): | |
3016 | if long_name == code: | |
3017 | return short_name | |
3018 | ||
3019 | ||
4eb10f66 YCH |
3020 | class ISO3166Utils(object): |
3021 | # From http://data.okfn.org/data/core/country-list | |
3022 | _country_map = { | |
3023 | 'AF': 'Afghanistan', | |
3024 | 'AX': 'Åland Islands', | |
3025 | 'AL': 'Albania', | |
3026 | 'DZ': 'Algeria', | |
3027 | 'AS': 'American Samoa', | |
3028 | 'AD': 'Andorra', | |
3029 | 'AO': 'Angola', | |
3030 | 'AI': 'Anguilla', | |
3031 | 'AQ': 'Antarctica', | |
3032 | 'AG': 'Antigua and Barbuda', | |
3033 | 'AR': 'Argentina', | |
3034 | 'AM': 'Armenia', | |
3035 | 'AW': 'Aruba', | |
3036 | 'AU': 'Australia', | |
3037 | 'AT': 'Austria', | |
3038 | 'AZ': 'Azerbaijan', | |
3039 | 'BS': 'Bahamas', | |
3040 | 'BH': 'Bahrain', | |
3041 | 'BD': 'Bangladesh', | |
3042 | 'BB': 'Barbados', | |
3043 | 'BY': 'Belarus', | |
3044 | 'BE': 'Belgium', | |
3045 | 'BZ': 'Belize', | |
3046 | 'BJ': 'Benin', | |
3047 | 'BM': 'Bermuda', | |
3048 | 'BT': 'Bhutan', | |
3049 | 'BO': 'Bolivia, Plurinational State of', | |
3050 | 'BQ': 'Bonaire, Sint Eustatius and Saba', | |
3051 | 'BA': 'Bosnia and Herzegovina', | |
3052 | 'BW': 'Botswana', | |
3053 | 'BV': 'Bouvet Island', | |
3054 | 'BR': 'Brazil', | |
3055 | 'IO': 'British Indian Ocean Territory', | |
3056 | 'BN': 'Brunei Darussalam', | |
3057 | 'BG': 'Bulgaria', | |
3058 | 'BF': 'Burkina Faso', | |
3059 | 'BI': 'Burundi', | |
3060 | 'KH': 'Cambodia', | |
3061 | 'CM': 'Cameroon', | |
3062 | 'CA': 'Canada', | |
3063 | 'CV': 'Cape Verde', | |
3064 | 'KY': 'Cayman Islands', | |
3065 | 'CF': 'Central African Republic', | |
3066 | 'TD': 'Chad', | |
3067 | 'CL': 'Chile', | |
3068 | 'CN': 'China', | |
3069 | 'CX': 'Christmas Island', | |
3070 | 'CC': 'Cocos (Keeling) Islands', | |
3071 | 'CO': 'Colombia', | |
3072 | 'KM': 'Comoros', | |
3073 | 'CG': 'Congo', | |
3074 | 'CD': 'Congo, the Democratic Republic of the', | |
3075 | 'CK': 'Cook Islands', | |
3076 | 'CR': 'Costa Rica', | |
3077 | 'CI': 'Côte d\'Ivoire', | |
3078 | 'HR': 'Croatia', | |
3079 | 'CU': 'Cuba', | |
3080 | 'CW': 'Curaçao', | |
3081 | 'CY': 'Cyprus', | |
3082 | 'CZ': 'Czech Republic', | |
3083 | 'DK': 'Denmark', | |
3084 | 'DJ': 'Djibouti', | |
3085 | 'DM': 'Dominica', | |
3086 | 'DO': 'Dominican Republic', | |
3087 | 'EC': 'Ecuador', | |
3088 | 'EG': 'Egypt', | |
3089 | 'SV': 'El Salvador', | |
3090 | 'GQ': 'Equatorial Guinea', | |
3091 | 'ER': 'Eritrea', | |
3092 | 'EE': 'Estonia', | |
3093 | 'ET': 'Ethiopia', | |
3094 | 'FK': 'Falkland Islands (Malvinas)', | |
3095 | 'FO': 'Faroe Islands', | |
3096 | 'FJ': 'Fiji', | |
3097 | 'FI': 'Finland', | |
3098 | 'FR': 'France', | |
3099 | 'GF': 'French Guiana', | |
3100 | 'PF': 'French Polynesia', | |
3101 | 'TF': 'French Southern Territories', | |
3102 | 'GA': 'Gabon', | |
3103 | 'GM': 'Gambia', | |
3104 | 'GE': 'Georgia', | |
3105 | 'DE': 'Germany', | |
3106 | 'GH': 'Ghana', | |
3107 | 'GI': 'Gibraltar', | |
3108 | 'GR': 'Greece', | |
3109 | 'GL': 'Greenland', | |
3110 | 'GD': 'Grenada', | |
3111 | 'GP': 'Guadeloupe', | |
3112 | 'GU': 'Guam', | |
3113 | 'GT': 'Guatemala', | |
3114 | 'GG': 'Guernsey', | |
3115 | 'GN': 'Guinea', | |
3116 | 'GW': 'Guinea-Bissau', | |
3117 | 'GY': 'Guyana', | |
3118 | 'HT': 'Haiti', | |
3119 | 'HM': 'Heard Island and McDonald Islands', | |
3120 | 'VA': 'Holy See (Vatican City State)', | |
3121 | 'HN': 'Honduras', | |
3122 | 'HK': 'Hong Kong', | |
3123 | 'HU': 'Hungary', | |
3124 | 'IS': 'Iceland', | |
3125 | 'IN': 'India', | |
3126 | 'ID': 'Indonesia', | |
3127 | 'IR': 'Iran, Islamic Republic of', | |
3128 | 'IQ': 'Iraq', | |
3129 | 'IE': 'Ireland', | |
3130 | 'IM': 'Isle of Man', | |
3131 | 'IL': 'Israel', | |
3132 | 'IT': 'Italy', | |
3133 | 'JM': 'Jamaica', | |
3134 | 'JP': 'Japan', | |
3135 | 'JE': 'Jersey', | |
3136 | 'JO': 'Jordan', | |
3137 | 'KZ': 'Kazakhstan', | |
3138 | 'KE': 'Kenya', | |
3139 | 'KI': 'Kiribati', | |
3140 | 'KP': 'Korea, Democratic People\'s Republic of', | |
3141 | 'KR': 'Korea, Republic of', | |
3142 | 'KW': 'Kuwait', | |
3143 | 'KG': 'Kyrgyzstan', | |
3144 | 'LA': 'Lao People\'s Democratic Republic', | |
3145 | 'LV': 'Latvia', | |
3146 | 'LB': 'Lebanon', | |
3147 | 'LS': 'Lesotho', | |
3148 | 'LR': 'Liberia', | |
3149 | 'LY': 'Libya', | |
3150 | 'LI': 'Liechtenstein', | |
3151 | 'LT': 'Lithuania', | |
3152 | 'LU': 'Luxembourg', | |
3153 | 'MO': 'Macao', | |
3154 | 'MK': 'Macedonia, the Former Yugoslav Republic of', | |
3155 | 'MG': 'Madagascar', | |
3156 | 'MW': 'Malawi', | |
3157 | 'MY': 'Malaysia', | |
3158 | 'MV': 'Maldives', | |
3159 | 'ML': 'Mali', | |
3160 | 'MT': 'Malta', | |
3161 | 'MH': 'Marshall Islands', | |
3162 | 'MQ': 'Martinique', | |
3163 | 'MR': 'Mauritania', | |
3164 | 'MU': 'Mauritius', | |
3165 | 'YT': 'Mayotte', | |
3166 | 'MX': 'Mexico', | |
3167 | 'FM': 'Micronesia, Federated States of', | |
3168 | 'MD': 'Moldova, Republic of', | |
3169 | 'MC': 'Monaco', | |
3170 | 'MN': 'Mongolia', | |
3171 | 'ME': 'Montenegro', | |
3172 | 'MS': 'Montserrat', | |
3173 | 'MA': 'Morocco', | |
3174 | 'MZ': 'Mozambique', | |
3175 | 'MM': 'Myanmar', | |
3176 | 'NA': 'Namibia', | |
3177 | 'NR': 'Nauru', | |
3178 | 'NP': 'Nepal', | |
3179 | 'NL': 'Netherlands', | |
3180 | 'NC': 'New Caledonia', | |
3181 | 'NZ': 'New Zealand', | |
3182 | 'NI': 'Nicaragua', | |
3183 | 'NE': 'Niger', | |
3184 | 'NG': 'Nigeria', | |
3185 | 'NU': 'Niue', | |
3186 | 'NF': 'Norfolk Island', | |
3187 | 'MP': 'Northern Mariana Islands', | |
3188 | 'NO': 'Norway', | |
3189 | 'OM': 'Oman', | |
3190 | 'PK': 'Pakistan', | |
3191 | 'PW': 'Palau', | |
3192 | 'PS': 'Palestine, State of', | |
3193 | 'PA': 'Panama', | |
3194 | 'PG': 'Papua New Guinea', | |
3195 | 'PY': 'Paraguay', | |
3196 | 'PE': 'Peru', | |
3197 | 'PH': 'Philippines', | |
3198 | 'PN': 'Pitcairn', | |
3199 | 'PL': 'Poland', | |
3200 | 'PT': 'Portugal', | |
3201 | 'PR': 'Puerto Rico', | |
3202 | 'QA': 'Qatar', | |
3203 | 'RE': 'Réunion', | |
3204 | 'RO': 'Romania', | |
3205 | 'RU': 'Russian Federation', | |
3206 | 'RW': 'Rwanda', | |
3207 | 'BL': 'Saint Barthélemy', | |
3208 | 'SH': 'Saint Helena, Ascension and Tristan da Cunha', | |
3209 | 'KN': 'Saint Kitts and Nevis', | |
3210 | 'LC': 'Saint Lucia', | |
3211 | 'MF': 'Saint Martin (French part)', | |
3212 | 'PM': 'Saint Pierre and Miquelon', | |
3213 | 'VC': 'Saint Vincent and the Grenadines', | |
3214 | 'WS': 'Samoa', | |
3215 | 'SM': 'San Marino', | |
3216 | 'ST': 'Sao Tome and Principe', | |
3217 | 'SA': 'Saudi Arabia', | |
3218 | 'SN': 'Senegal', | |
3219 | 'RS': 'Serbia', | |
3220 | 'SC': 'Seychelles', | |
3221 | 'SL': 'Sierra Leone', | |
3222 | 'SG': 'Singapore', | |
3223 | 'SX': 'Sint Maarten (Dutch part)', | |
3224 | 'SK': 'Slovakia', | |
3225 | 'SI': 'Slovenia', | |
3226 | 'SB': 'Solomon Islands', | |
3227 | 'SO': 'Somalia', | |
3228 | 'ZA': 'South Africa', | |
3229 | 'GS': 'South Georgia and the South Sandwich Islands', | |
3230 | 'SS': 'South Sudan', | |
3231 | 'ES': 'Spain', | |
3232 | 'LK': 'Sri Lanka', | |
3233 | 'SD': 'Sudan', | |
3234 | 'SR': 'Suriname', | |
3235 | 'SJ': 'Svalbard and Jan Mayen', | |
3236 | 'SZ': 'Swaziland', | |
3237 | 'SE': 'Sweden', | |
3238 | 'CH': 'Switzerland', | |
3239 | 'SY': 'Syrian Arab Republic', | |
3240 | 'TW': 'Taiwan, Province of China', | |
3241 | 'TJ': 'Tajikistan', | |
3242 | 'TZ': 'Tanzania, United Republic of', | |
3243 | 'TH': 'Thailand', | |
3244 | 'TL': 'Timor-Leste', | |
3245 | 'TG': 'Togo', | |
3246 | 'TK': 'Tokelau', | |
3247 | 'TO': 'Tonga', | |
3248 | 'TT': 'Trinidad and Tobago', | |
3249 | 'TN': 'Tunisia', | |
3250 | 'TR': 'Turkey', | |
3251 | 'TM': 'Turkmenistan', | |
3252 | 'TC': 'Turks and Caicos Islands', | |
3253 | 'TV': 'Tuvalu', | |
3254 | 'UG': 'Uganda', | |
3255 | 'UA': 'Ukraine', | |
3256 | 'AE': 'United Arab Emirates', | |
3257 | 'GB': 'United Kingdom', | |
3258 | 'US': 'United States', | |
3259 | 'UM': 'United States Minor Outlying Islands', | |
3260 | 'UY': 'Uruguay', | |
3261 | 'UZ': 'Uzbekistan', | |
3262 | 'VU': 'Vanuatu', | |
3263 | 'VE': 'Venezuela, Bolivarian Republic of', | |
3264 | 'VN': 'Viet Nam', | |
3265 | 'VG': 'Virgin Islands, British', | |
3266 | 'VI': 'Virgin Islands, U.S.', | |
3267 | 'WF': 'Wallis and Futuna', | |
3268 | 'EH': 'Western Sahara', | |
3269 | 'YE': 'Yemen', | |
3270 | 'ZM': 'Zambia', | |
3271 | 'ZW': 'Zimbabwe', | |
3272 | } | |
3273 | ||
3274 | @classmethod | |
3275 | def short2full(cls, code): | |
3276 | """Convert an ISO 3166-2 country code to the corresponding full name""" | |
3277 | return cls._country_map.get(code.upper()) | |
3278 | ||
3279 | ||
773f291d S |
3280 | class GeoUtils(object): |
3281 | # Major IPv4 address blocks per country | |
3282 | _country_ip_map = { | |
3283 | 'AD': '85.94.160.0/19', | |
3284 | 'AE': '94.200.0.0/13', | |
3285 | 'AF': '149.54.0.0/17', | |
3286 | 'AG': '209.59.64.0/18', | |
3287 | 'AI': '204.14.248.0/21', | |
3288 | 'AL': '46.99.0.0/16', | |
3289 | 'AM': '46.70.0.0/15', | |
3290 | 'AO': '105.168.0.0/13', | |
3291 | 'AP': '159.117.192.0/21', | |
3292 | 'AR': '181.0.0.0/12', | |
3293 | 'AS': '202.70.112.0/20', | |
3294 | 'AT': '84.112.0.0/13', | |
3295 | 'AU': '1.128.0.0/11', | |
3296 | 'AW': '181.41.0.0/18', | |
3297 | 'AZ': '5.191.0.0/16', | |
3298 | 'BA': '31.176.128.0/17', | |
3299 | 'BB': '65.48.128.0/17', | |
3300 | 'BD': '114.130.0.0/16', | |
3301 | 'BE': '57.0.0.0/8', | |
3302 | 'BF': '129.45.128.0/17', | |
3303 | 'BG': '95.42.0.0/15', | |
3304 | 'BH': '37.131.0.0/17', | |
3305 | 'BI': '154.117.192.0/18', | |
3306 | 'BJ': '137.255.0.0/16', | |
3307 | 'BL': '192.131.134.0/24', | |
3308 | 'BM': '196.12.64.0/18', | |
3309 | 'BN': '156.31.0.0/16', | |
3310 | 'BO': '161.56.0.0/16', | |
3311 | 'BQ': '161.0.80.0/20', | |
3312 | 'BR': '152.240.0.0/12', | |
3313 | 'BS': '24.51.64.0/18', | |
3314 | 'BT': '119.2.96.0/19', | |
3315 | 'BW': '168.167.0.0/16', | |
3316 | 'BY': '178.120.0.0/13', | |
3317 | 'BZ': '179.42.192.0/18', | |
3318 | 'CA': '99.224.0.0/11', | |
3319 | 'CD': '41.243.0.0/16', | |
3320 | 'CF': '196.32.200.0/21', | |
3321 | 'CG': '197.214.128.0/17', | |
3322 | 'CH': '85.0.0.0/13', | |
3323 | 'CI': '154.232.0.0/14', | |
3324 | 'CK': '202.65.32.0/19', | |
3325 | 'CL': '152.172.0.0/14', | |
3326 | 'CM': '165.210.0.0/15', | |
3327 | 'CN': '36.128.0.0/10', | |
3328 | 'CO': '181.240.0.0/12', | |
3329 | 'CR': '201.192.0.0/12', | |
3330 | 'CU': '152.206.0.0/15', | |
3331 | 'CV': '165.90.96.0/19', | |
3332 | 'CW': '190.88.128.0/17', | |
3333 | 'CY': '46.198.0.0/15', | |
3334 | 'CZ': '88.100.0.0/14', | |
3335 | 'DE': '53.0.0.0/8', | |
3336 | 'DJ': '197.241.0.0/17', | |
3337 | 'DK': '87.48.0.0/12', | |
3338 | 'DM': '192.243.48.0/20', | |
3339 | 'DO': '152.166.0.0/15', | |
3340 | 'DZ': '41.96.0.0/12', | |
3341 | 'EC': '186.68.0.0/15', | |
3342 | 'EE': '90.190.0.0/15', | |
3343 | 'EG': '156.160.0.0/11', | |
3344 | 'ER': '196.200.96.0/20', | |
3345 | 'ES': '88.0.0.0/11', | |
3346 | 'ET': '196.188.0.0/14', | |
3347 | 'EU': '2.16.0.0/13', | |
3348 | 'FI': '91.152.0.0/13', | |
3349 | 'FJ': '144.120.0.0/16', | |
3350 | 'FM': '119.252.112.0/20', | |
3351 | 'FO': '88.85.32.0/19', | |
3352 | 'FR': '90.0.0.0/9', | |
3353 | 'GA': '41.158.0.0/15', | |
3354 | 'GB': '25.0.0.0/8', | |
3355 | 'GD': '74.122.88.0/21', | |
3356 | 'GE': '31.146.0.0/16', | |
3357 | 'GF': '161.22.64.0/18', | |
3358 | 'GG': '62.68.160.0/19', | |
3359 | 'GH': '45.208.0.0/14', | |
3360 | 'GI': '85.115.128.0/19', | |
3361 | 'GL': '88.83.0.0/19', | |
3362 | 'GM': '160.182.0.0/15', | |
3363 | 'GN': '197.149.192.0/18', | |
3364 | 'GP': '104.250.0.0/19', | |
3365 | 'GQ': '105.235.224.0/20', | |
3366 | 'GR': '94.64.0.0/13', | |
3367 | 'GT': '168.234.0.0/16', | |
3368 | 'GU': '168.123.0.0/16', | |
3369 | 'GW': '197.214.80.0/20', | |
3370 | 'GY': '181.41.64.0/18', | |
3371 | 'HK': '113.252.0.0/14', | |
3372 | 'HN': '181.210.0.0/16', | |
3373 | 'HR': '93.136.0.0/13', | |
3374 | 'HT': '148.102.128.0/17', | |
3375 | 'HU': '84.0.0.0/14', | |
3376 | 'ID': '39.192.0.0/10', | |
3377 | 'IE': '87.32.0.0/12', | |
3378 | 'IL': '79.176.0.0/13', | |
3379 | 'IM': '5.62.80.0/20', | |
3380 | 'IN': '117.192.0.0/10', | |
3381 | 'IO': '203.83.48.0/21', | |
3382 | 'IQ': '37.236.0.0/14', | |
3383 | 'IR': '2.176.0.0/12', | |
3384 | 'IS': '82.221.0.0/16', | |
3385 | 'IT': '79.0.0.0/10', | |
3386 | 'JE': '87.244.64.0/18', | |
3387 | 'JM': '72.27.0.0/17', | |
3388 | 'JO': '176.29.0.0/16', | |
3389 | 'JP': '126.0.0.0/8', | |
3390 | 'KE': '105.48.0.0/12', | |
3391 | 'KG': '158.181.128.0/17', | |
3392 | 'KH': '36.37.128.0/17', | |
3393 | 'KI': '103.25.140.0/22', | |
3394 | 'KM': '197.255.224.0/20', | |
3395 | 'KN': '198.32.32.0/19', | |
3396 | 'KP': '175.45.176.0/22', | |
3397 | 'KR': '175.192.0.0/10', | |
3398 | 'KW': '37.36.0.0/14', | |
3399 | 'KY': '64.96.0.0/15', | |
3400 | 'KZ': '2.72.0.0/13', | |
3401 | 'LA': '115.84.64.0/18', | |
3402 | 'LB': '178.135.0.0/16', | |
3403 | 'LC': '192.147.231.0/24', | |
3404 | 'LI': '82.117.0.0/19', | |
3405 | 'LK': '112.134.0.0/15', | |
3406 | 'LR': '41.86.0.0/19', | |
3407 | 'LS': '129.232.0.0/17', | |
3408 | 'LT': '78.56.0.0/13', | |
3409 | 'LU': '188.42.0.0/16', | |
3410 | 'LV': '46.109.0.0/16', | |
3411 | 'LY': '41.252.0.0/14', | |
3412 | 'MA': '105.128.0.0/11', | |
3413 | 'MC': '88.209.64.0/18', | |
3414 | 'MD': '37.246.0.0/16', | |
3415 | 'ME': '178.175.0.0/17', | |
3416 | 'MF': '74.112.232.0/21', | |
3417 | 'MG': '154.126.0.0/17', | |
3418 | 'MH': '117.103.88.0/21', | |
3419 | 'MK': '77.28.0.0/15', | |
3420 | 'ML': '154.118.128.0/18', | |
3421 | 'MM': '37.111.0.0/17', | |
3422 | 'MN': '49.0.128.0/17', | |
3423 | 'MO': '60.246.0.0/16', | |
3424 | 'MP': '202.88.64.0/20', | |
3425 | 'MQ': '109.203.224.0/19', | |
3426 | 'MR': '41.188.64.0/18', | |
3427 | 'MS': '208.90.112.0/22', | |
3428 | 'MT': '46.11.0.0/16', | |
3429 | 'MU': '105.16.0.0/12', | |
3430 | 'MV': '27.114.128.0/18', | |
3431 | 'MW': '105.234.0.0/16', | |
3432 | 'MX': '187.192.0.0/11', | |
3433 | 'MY': '175.136.0.0/13', | |
3434 | 'MZ': '197.218.0.0/15', | |
3435 | 'NA': '41.182.0.0/16', | |
3436 | 'NC': '101.101.0.0/18', | |
3437 | 'NE': '197.214.0.0/18', | |
3438 | 'NF': '203.17.240.0/22', | |
3439 | 'NG': '105.112.0.0/12', | |
3440 | 'NI': '186.76.0.0/15', | |
3441 | 'NL': '145.96.0.0/11', | |
3442 | 'NO': '84.208.0.0/13', | |
3443 | 'NP': '36.252.0.0/15', | |
3444 | 'NR': '203.98.224.0/19', | |
3445 | 'NU': '49.156.48.0/22', | |
3446 | 'NZ': '49.224.0.0/14', | |
3447 | 'OM': '5.36.0.0/15', | |
3448 | 'PA': '186.72.0.0/15', | |
3449 | 'PE': '186.160.0.0/14', | |
3450 | 'PF': '123.50.64.0/18', | |
3451 | 'PG': '124.240.192.0/19', | |
3452 | 'PH': '49.144.0.0/13', | |
3453 | 'PK': '39.32.0.0/11', | |
3454 | 'PL': '83.0.0.0/11', | |
3455 | 'PM': '70.36.0.0/20', | |
3456 | 'PR': '66.50.0.0/16', | |
3457 | 'PS': '188.161.0.0/16', | |
3458 | 'PT': '85.240.0.0/13', | |
3459 | 'PW': '202.124.224.0/20', | |
3460 | 'PY': '181.120.0.0/14', | |
3461 | 'QA': '37.210.0.0/15', | |
3462 | 'RE': '139.26.0.0/16', | |
3463 | 'RO': '79.112.0.0/13', | |
3464 | 'RS': '178.220.0.0/14', | |
3465 | 'RU': '5.136.0.0/13', | |
3466 | 'RW': '105.178.0.0/15', | |
3467 | 'SA': '188.48.0.0/13', | |
3468 | 'SB': '202.1.160.0/19', | |
3469 | 'SC': '154.192.0.0/11', | |
3470 | 'SD': '154.96.0.0/13', | |
3471 | 'SE': '78.64.0.0/12', | |
3472 | 'SG': '152.56.0.0/14', | |
3473 | 'SI': '188.196.0.0/14', | |
3474 | 'SK': '78.98.0.0/15', | |
3475 | 'SL': '197.215.0.0/17', | |
3476 | 'SM': '89.186.32.0/19', | |
3477 | 'SN': '41.82.0.0/15', | |
3478 | 'SO': '197.220.64.0/19', | |
3479 | 'SR': '186.179.128.0/17', | |
3480 | 'SS': '105.235.208.0/21', | |
3481 | 'ST': '197.159.160.0/19', | |
3482 | 'SV': '168.243.0.0/16', | |
3483 | 'SX': '190.102.0.0/20', | |
3484 | 'SY': '5.0.0.0/16', | |
3485 | 'SZ': '41.84.224.0/19', | |
3486 | 'TC': '65.255.48.0/20', | |
3487 | 'TD': '154.68.128.0/19', | |
3488 | 'TG': '196.168.0.0/14', | |
3489 | 'TH': '171.96.0.0/13', | |
3490 | 'TJ': '85.9.128.0/18', | |
3491 | 'TK': '27.96.24.0/21', | |
3492 | 'TL': '180.189.160.0/20', | |
3493 | 'TM': '95.85.96.0/19', | |
3494 | 'TN': '197.0.0.0/11', | |
3495 | 'TO': '175.176.144.0/21', | |
3496 | 'TR': '78.160.0.0/11', | |
3497 | 'TT': '186.44.0.0/15', | |
3498 | 'TV': '202.2.96.0/19', | |
3499 | 'TW': '120.96.0.0/11', | |
3500 | 'TZ': '156.156.0.0/14', | |
3501 | 'UA': '93.72.0.0/13', | |
3502 | 'UG': '154.224.0.0/13', | |
3503 | 'US': '3.0.0.0/8', | |
3504 | 'UY': '167.56.0.0/13', | |
3505 | 'UZ': '82.215.64.0/18', | |
3506 | 'VA': '212.77.0.0/19', | |
3507 | 'VC': '24.92.144.0/20', | |
3508 | 'VE': '186.88.0.0/13', | |
3509 | 'VG': '172.103.64.0/18', | |
3510 | 'VI': '146.226.0.0/16', | |
3511 | 'VN': '14.160.0.0/11', | |
3512 | 'VU': '202.80.32.0/20', | |
3513 | 'WF': '117.20.32.0/21', | |
3514 | 'WS': '202.4.32.0/19', | |
3515 | 'YE': '134.35.0.0/16', | |
3516 | 'YT': '41.242.116.0/22', | |
3517 | 'ZA': '41.0.0.0/11', | |
3518 | 'ZM': '165.56.0.0/13', | |
3519 | 'ZW': '41.85.192.0/19', | |
3520 | } | |
3521 | ||
3522 | @classmethod | |
3523 | def random_ipv4(cls, code): | |
3524 | block = cls._country_ip_map.get(code.upper()) | |
3525 | if not block: | |
3526 | return None | |
3527 | addr, preflen = block.split('/') | |
3528 | addr_min = compat_struct_unpack('!L', socket.inet_aton(addr))[0] | |
3529 | addr_max = addr_min | (0xffffffff >> int(preflen)) | |
18a0defa | 3530 | return compat_str(socket.inet_ntoa( |
4248dad9 | 3531 | compat_struct_pack('!L', random.randint(addr_min, addr_max)))) |
773f291d S |
3532 | |
3533 | ||
91410c9b | 3534 | class PerRequestProxyHandler(compat_urllib_request.ProxyHandler): |
2461f79d PH |
3535 | def __init__(self, proxies=None): |
3536 | # Set default handlers | |
3537 | for type in ('http', 'https'): | |
3538 | setattr(self, '%s_open' % type, | |
3539 | lambda r, proxy='__noproxy__', type=type, meth=self.proxy_open: | |
3540 | meth(r, proxy, type)) | |
3541 | return compat_urllib_request.ProxyHandler.__init__(self, proxies) | |
3542 | ||
91410c9b | 3543 | def proxy_open(self, req, proxy, type): |
2461f79d | 3544 | req_proxy = req.headers.get('Ytdl-request-proxy') |
91410c9b PH |
3545 | if req_proxy is not None: |
3546 | proxy = req_proxy | |
2461f79d PH |
3547 | del req.headers['Ytdl-request-proxy'] |
3548 | ||
3549 | if proxy == '__noproxy__': | |
3550 | return None # No Proxy | |
51fb4995 | 3551 | if compat_urlparse.urlparse(proxy).scheme.lower() in ('socks', 'socks4', 'socks4a', 'socks5'): |
71aff188 YCH |
3552 | req.add_header('Ytdl-socks-proxy', proxy) |
3553 | # youtube-dl's http/https handlers do wrapping the socket with socks | |
3554 | return None | |
91410c9b PH |
3555 | return compat_urllib_request.ProxyHandler.proxy_open( |
3556 | self, req, proxy, type) | |
5bc880b9 YCH |
3557 | |
3558 | ||
0a5445dd YCH |
3559 | # Both long_to_bytes and bytes_to_long are adapted from PyCrypto, which is |
3560 | # released into Public Domain | |
3561 | # https://github.com/dlitz/pycrypto/blob/master/lib/Crypto/Util/number.py#L387 | |
3562 | ||
3563 | def long_to_bytes(n, blocksize=0): | |
3564 | """long_to_bytes(n:long, blocksize:int) : string | |
3565 | Convert a long integer to a byte string. | |
3566 | ||
3567 | If optional blocksize is given and greater than zero, pad the front of the | |
3568 | byte string with binary zeros so that the length is a multiple of | |
3569 | blocksize. | |
3570 | """ | |
3571 | # after much testing, this algorithm was deemed to be the fastest | |
3572 | s = b'' | |
3573 | n = int(n) | |
3574 | while n > 0: | |
3575 | s = compat_struct_pack('>I', n & 0xffffffff) + s | |
3576 | n = n >> 32 | |
3577 | # strip off leading zeros | |
3578 | for i in range(len(s)): | |
3579 | if s[i] != b'\000'[0]: | |
3580 | break | |
3581 | else: | |
3582 | # only happens when n == 0 | |
3583 | s = b'\000' | |
3584 | i = 0 | |
3585 | s = s[i:] | |
3586 | # add back some pad bytes. this could be done more efficiently w.r.t. the | |
3587 | # de-padding being done above, but sigh... | |
3588 | if blocksize > 0 and len(s) % blocksize: | |
3589 | s = (blocksize - len(s) % blocksize) * b'\000' + s | |
3590 | return s | |
3591 | ||
3592 | ||
3593 | def bytes_to_long(s): | |
3594 | """bytes_to_long(string) : long | |
3595 | Convert a byte string to a long integer. | |
3596 | ||
3597 | This is (essentially) the inverse of long_to_bytes(). | |
3598 | """ | |
3599 | acc = 0 | |
3600 | length = len(s) | |
3601 | if length % 4: | |
3602 | extra = (4 - length % 4) | |
3603 | s = b'\000' * extra + s | |
3604 | length = length + extra | |
3605 | for i in range(0, length, 4): | |
3606 | acc = (acc << 32) + compat_struct_unpack('>I', s[i:i + 4])[0] | |
3607 | return acc | |
3608 | ||
3609 | ||
5bc880b9 YCH |
3610 | def ohdave_rsa_encrypt(data, exponent, modulus): |
3611 | ''' | |
3612 | Implement OHDave's RSA algorithm. See http://www.ohdave.com/rsa/ | |
3613 | ||
3614 | Input: | |
3615 | data: data to encrypt, bytes-like object | |
3616 | exponent, modulus: parameter e and N of RSA algorithm, both integer | |
3617 | Output: hex string of encrypted data | |
3618 | ||
3619 | Limitation: supports one block encryption only | |
3620 | ''' | |
3621 | ||
3622 | payload = int(binascii.hexlify(data[::-1]), 16) | |
3623 | encrypted = pow(payload, exponent, modulus) | |
3624 | return '%x' % encrypted | |
81bdc8fd YCH |
3625 | |
3626 | ||
f48409c7 YCH |
3627 | def pkcs1pad(data, length): |
3628 | """ | |
3629 | Padding input data with PKCS#1 scheme | |
3630 | ||
3631 | @param {int[]} data input data | |
3632 | @param {int} length target length | |
3633 | @returns {int[]} padded data | |
3634 | """ | |
3635 | if len(data) > length - 11: | |
3636 | raise ValueError('Input data too long for PKCS#1 padding') | |
3637 | ||
3638 | pseudo_random = [random.randint(0, 254) for _ in range(length - len(data) - 3)] | |
3639 | return [0, 2] + pseudo_random + [0] + data | |
3640 | ||
3641 | ||
5eb6bdce | 3642 | def encode_base_n(num, n, table=None): |
59f898b7 | 3643 | FULL_TABLE = '0123456789abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ' |
59f898b7 YCH |
3644 | if not table: |
3645 | table = FULL_TABLE[:n] | |
3646 | ||
5eb6bdce YCH |
3647 | if n > len(table): |
3648 | raise ValueError('base %d exceeds table length %d' % (n, len(table))) | |
3649 | ||
3650 | if num == 0: | |
3651 | return table[0] | |
3652 | ||
81bdc8fd YCH |
3653 | ret = '' |
3654 | while num: | |
3655 | ret = table[num % n] + ret | |
3656 | num = num // n | |
3657 | return ret | |
f52354a8 YCH |
3658 | |
3659 | ||
3660 | def decode_packed_codes(code): | |
06b3fe29 | 3661 | mobj = re.search(PACKED_CODES_RE, code) |
f52354a8 YCH |
3662 | obfucasted_code, base, count, symbols = mobj.groups() |
3663 | base = int(base) | |
3664 | count = int(count) | |
3665 | symbols = symbols.split('|') | |
3666 | symbol_table = {} | |
3667 | ||
3668 | while count: | |
3669 | count -= 1 | |
5eb6bdce | 3670 | base_n_count = encode_base_n(count, base) |
f52354a8 YCH |
3671 | symbol_table[base_n_count] = symbols[count] or base_n_count |
3672 | ||
3673 | return re.sub( | |
3674 | r'\b(\w+)\b', lambda mobj: symbol_table[mobj.group(0)], | |
3675 | obfucasted_code) | |
e154c651 | 3676 | |
3677 | ||
3678 | def parse_m3u8_attributes(attrib): | |
3679 | info = {} | |
3680 | for (key, val) in re.findall(r'(?P<key>[A-Z0-9-]+)=(?P<val>"[^"]+"|[^",]+)(?:,|$)', attrib): | |
3681 | if val.startswith('"'): | |
3682 | val = val[1:-1] | |
3683 | info[key] = val | |
3684 | return info | |
1143535d YCH |
3685 | |
3686 | ||
3687 | def urshift(val, n): | |
3688 | return val >> n if val >= 0 else (val + 0x100000000) >> n | |
d3f8e038 YCH |
3689 | |
3690 | ||
3691 | # Based on png2str() written by @gdkchan and improved by @yokrysty | |
3692 | # Originally posted at https://github.com/rg3/youtube-dl/issues/9706 | |
3693 | def decode_png(png_data): | |
3694 | # Reference: https://www.w3.org/TR/PNG/ | |
3695 | header = png_data[8:] | |
3696 | ||
3697 | if png_data[:8] != b'\x89PNG\x0d\x0a\x1a\x0a' or header[4:8] != b'IHDR': | |
3698 | raise IOError('Not a valid PNG file.') | |
3699 | ||
3700 | int_map = {1: '>B', 2: '>H', 4: '>I'} | |
3701 | unpack_integer = lambda x: compat_struct_unpack(int_map[len(x)], x)[0] | |
3702 | ||
3703 | chunks = [] | |
3704 | ||
3705 | while header: | |
3706 | length = unpack_integer(header[:4]) | |
3707 | header = header[4:] | |
3708 | ||
3709 | chunk_type = header[:4] | |
3710 | header = header[4:] | |
3711 | ||
3712 | chunk_data = header[:length] | |
3713 | header = header[length:] | |
3714 | ||
3715 | header = header[4:] # Skip CRC | |
3716 | ||
3717 | chunks.append({ | |
3718 | 'type': chunk_type, | |
3719 | 'length': length, | |
3720 | 'data': chunk_data | |
3721 | }) | |
3722 | ||
3723 | ihdr = chunks[0]['data'] | |
3724 | ||
3725 | width = unpack_integer(ihdr[:4]) | |
3726 | height = unpack_integer(ihdr[4:8]) | |
3727 | ||
3728 | idat = b'' | |
3729 | ||
3730 | for chunk in chunks: | |
3731 | if chunk['type'] == b'IDAT': | |
3732 | idat += chunk['data'] | |
3733 | ||
3734 | if not idat: | |
3735 | raise IOError('Unable to read PNG data.') | |
3736 | ||
3737 | decompressed_data = bytearray(zlib.decompress(idat)) | |
3738 | ||
3739 | stride = width * 3 | |
3740 | pixels = [] | |
3741 | ||
3742 | def _get_pixel(idx): | |
3743 | x = idx % stride | |
3744 | y = idx // stride | |
3745 | return pixels[y][x] | |
3746 | ||
3747 | for y in range(height): | |
3748 | basePos = y * (1 + stride) | |
3749 | filter_type = decompressed_data[basePos] | |
3750 | ||
3751 | current_row = [] | |
3752 | ||
3753 | pixels.append(current_row) | |
3754 | ||
3755 | for x in range(stride): | |
3756 | color = decompressed_data[1 + basePos + x] | |
3757 | basex = y * stride + x | |
3758 | left = 0 | |
3759 | up = 0 | |
3760 | ||
3761 | if x > 2: | |
3762 | left = _get_pixel(basex - 3) | |
3763 | if y > 0: | |
3764 | up = _get_pixel(basex - stride) | |
3765 | ||
3766 | if filter_type == 1: # Sub | |
3767 | color = (color + left) & 0xff | |
3768 | elif filter_type == 2: # Up | |
3769 | color = (color + up) & 0xff | |
3770 | elif filter_type == 3: # Average | |
3771 | color = (color + ((left + up) >> 1)) & 0xff | |
3772 | elif filter_type == 4: # Paeth | |
3773 | a = left | |
3774 | b = up | |
3775 | c = 0 | |
3776 | ||
3777 | if x > 2 and y > 0: | |
3778 | c = _get_pixel(basex - stride - 3) | |
3779 | ||
3780 | p = a + b - c | |
3781 | ||
3782 | pa = abs(p - a) | |
3783 | pb = abs(p - b) | |
3784 | pc = abs(p - c) | |
3785 | ||
3786 | if pa <= pb and pa <= pc: | |
3787 | color = (color + a) & 0xff | |
3788 | elif pb <= pc: | |
3789 | color = (color + b) & 0xff | |
3790 | else: | |
3791 | color = (color + c) & 0xff | |
3792 | ||
3793 | current_row.append(color) | |
3794 | ||
3795 | return width, height, pixels | |
efa97bdc YCH |
3796 | |
3797 | ||
3798 | def write_xattr(path, key, value): | |
3799 | # This mess below finds the best xattr tool for the job | |
3800 | try: | |
3801 | # try the pyxattr module... | |
3802 | import xattr | |
3803 | ||
53a7e3d2 YCH |
3804 | if hasattr(xattr, 'set'): # pyxattr |
3805 | # Unicode arguments are not supported in python-pyxattr until | |
3806 | # version 0.5.0 | |
3807 | # See https://github.com/rg3/youtube-dl/issues/5498 | |
3808 | pyxattr_required_version = '0.5.0' | |
3809 | if version_tuple(xattr.__version__) < version_tuple(pyxattr_required_version): | |
3810 | # TODO: fallback to CLI tools | |
3811 | raise XAttrUnavailableError( | |
3812 | 'python-pyxattr is detected but is too old. ' | |
3813 | 'youtube-dl requires %s or above while your version is %s. ' | |
3814 | 'Falling back to other xattr implementations' % ( | |
3815 | pyxattr_required_version, xattr.__version__)) | |
3816 | ||
3817 | setxattr = xattr.set | |
3818 | else: # xattr | |
3819 | setxattr = xattr.setxattr | |
efa97bdc YCH |
3820 | |
3821 | try: | |
53a7e3d2 | 3822 | setxattr(path, key, value) |
efa97bdc YCH |
3823 | except EnvironmentError as e: |
3824 | raise XAttrMetadataError(e.errno, e.strerror) | |
3825 | ||
3826 | except ImportError: | |
3827 | if compat_os_name == 'nt': | |
3828 | # Write xattrs to NTFS Alternate Data Streams: | |
3829 | # http://en.wikipedia.org/wiki/NTFS#Alternate_data_streams_.28ADS.29 | |
3830 | assert ':' not in key | |
3831 | assert os.path.exists(path) | |
3832 | ||
3833 | ads_fn = path + ':' + key | |
3834 | try: | |
3835 | with open(ads_fn, 'wb') as f: | |
3836 | f.write(value) | |
3837 | except EnvironmentError as e: | |
3838 | raise XAttrMetadataError(e.errno, e.strerror) | |
3839 | else: | |
3840 | user_has_setfattr = check_executable('setfattr', ['--version']) | |
3841 | user_has_xattr = check_executable('xattr', ['-h']) | |
3842 | ||
3843 | if user_has_setfattr or user_has_xattr: | |
3844 | ||
3845 | value = value.decode('utf-8') | |
3846 | if user_has_setfattr: | |
3847 | executable = 'setfattr' | |
3848 | opts = ['-n', key, '-v', value] | |
3849 | elif user_has_xattr: | |
3850 | executable = 'xattr' | |
3851 | opts = ['-w', key, value] | |
3852 | ||
3853 | cmd = ([encodeFilename(executable, True)] + | |
3854 | [encodeArgument(o) for o in opts] + | |
3855 | [encodeFilename(path, True)]) | |
3856 | ||
3857 | try: | |
3858 | p = subprocess.Popen( | |
3859 | cmd, stdout=subprocess.PIPE, stderr=subprocess.PIPE, stdin=subprocess.PIPE) | |
3860 | except EnvironmentError as e: | |
3861 | raise XAttrMetadataError(e.errno, e.strerror) | |
3862 | stdout, stderr = p.communicate() | |
3863 | stderr = stderr.decode('utf-8', 'replace') | |
3864 | if p.returncode != 0: | |
3865 | raise XAttrMetadataError(p.returncode, stderr) | |
3866 | ||
3867 | else: | |
3868 | # On Unix, and can't find pyxattr, setfattr, or xattr. | |
3869 | if sys.platform.startswith('linux'): | |
3870 | raise XAttrUnavailableError( | |
3871 | "Couldn't find a tool to set the xattrs. " | |
3872 | "Install either the python 'pyxattr' or 'xattr' " | |
3873 | "modules, or the GNU 'attr' package " | |
3874 | "(which contains the 'setfattr' tool).") | |
3875 | else: | |
3876 | raise XAttrUnavailableError( | |
3877 | "Couldn't find a tool to set the xattrs. " | |
3878 | "Install either the python 'xattr' module, " | |
3879 | "or the 'xattr' binary.") | |
0c265486 YCH |
3880 | |
3881 | ||
3882 | def random_birthday(year_field, month_field, day_field): | |
3883 | return { | |
3884 | year_field: str(random.randint(1950, 1995)), | |
3885 | month_field: str(random.randint(1, 12)), | |
3886 | day_field: str(random.randint(1, 31)), | |
3887 | } |