]>
Commit | Line | Data |
---|---|---|
d77c3dfd FV |
1 | #!/usr/bin/env python |
2 | # -*- coding: utf-8 -*- | |
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 | |
f45c185f | 14 | import errno |
be4a824d | 15 | import functools |
d77c3dfd | 16 | import gzip |
03f9daab | 17 | import io |
79a2e94e | 18 | import itertools |
f4bfd65f | 19 | import json |
d77c3dfd | 20 | import locale |
02dbf93f | 21 | import math |
347de493 | 22 | import operator |
d77c3dfd | 23 | import os |
4eb7f1d1 | 24 | import pipes |
c496ca96 | 25 | import platform |
d77c3dfd | 26 | import re |
c496ca96 | 27 | import socket |
79a2e94e | 28 | import ssl |
1c088fa8 | 29 | import subprocess |
d77c3dfd | 30 | import sys |
181c8655 | 31 | import tempfile |
01951dda | 32 | import traceback |
bcf89ce6 | 33 | import xml.etree.ElementTree |
d77c3dfd | 34 | import zlib |
d77c3dfd | 35 | |
8c25f81b | 36 | from .compat import ( |
8bb56eee | 37 | compat_HTMLParser, |
8f9312c3 | 38 | compat_basestring, |
8c25f81b | 39 | compat_chr, |
36e6f62c | 40 | compat_etree_fromstring, |
8c25f81b | 41 | compat_html_entities, |
55b2f099 | 42 | compat_html_entities_html5, |
be4a824d | 43 | compat_http_client, |
c86b6142 | 44 | compat_kwargs, |
8c25f81b | 45 | compat_parse_qs, |
702ccf2d | 46 | compat_shlex_quote, |
be4a824d | 47 | compat_socket_create_connection, |
8c25f81b | 48 | compat_str, |
edaa23f8 | 49 | compat_struct_pack, |
8c25f81b PH |
50 | compat_urllib_error, |
51 | compat_urllib_parse, | |
15707c7e | 52 | compat_urllib_parse_urlencode, |
8c25f81b | 53 | compat_urllib_parse_urlparse, |
7581bfc9 | 54 | compat_urllib_parse_unquote_plus, |
8c25f81b PH |
55 | compat_urllib_request, |
56 | compat_urlparse, | |
810c10ba | 57 | compat_xpath, |
8c25f81b | 58 | ) |
4644ac55 | 59 | |
71aff188 YCH |
60 | from .socks import ( |
61 | ProxyType, | |
62 | sockssocket, | |
63 | ) | |
64 | ||
4644ac55 | 65 | |
51fb4995 YCH |
66 | def register_socks_protocols(): |
67 | # "Register" SOCKS protocols | |
d5ae6bb5 YCH |
68 | # In Python < 2.6.5, urlsplit() suffers from bug https://bugs.python.org/issue7904 |
69 | # URLs with protocols not in urlparse.uses_netloc are not handled correctly | |
51fb4995 YCH |
70 | for scheme in ('socks', 'socks4', 'socks4a', 'socks5'): |
71 | if scheme not in compat_urlparse.uses_netloc: | |
72 | compat_urlparse.uses_netloc.append(scheme) | |
73 | ||
74 | ||
468e2e92 FV |
75 | # This is not clearly defined otherwise |
76 | compiled_regex_type = type(re.compile('')) | |
77 | ||
3e669f36 | 78 | std_headers = { |
15d10678 | 79 | 'User-Agent': 'Mozilla/5.0 (X11; Linux x86_64; rv:10.0) Gecko/20150101 Firefox/47.0 (Chrome)', |
59ae15a5 PH |
80 | 'Accept-Charset': 'ISO-8859-1,utf-8;q=0.7,*;q=0.7', |
81 | 'Accept': 'text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8', | |
82 | 'Accept-Encoding': 'gzip, deflate', | |
83 | 'Accept-Language': 'en-us,en;q=0.5', | |
3e669f36 | 84 | } |
f427df17 | 85 | |
5f6a1245 | 86 | |
bf42a990 S |
87 | NO_DEFAULT = object() |
88 | ||
7105440c YCH |
89 | ENGLISH_MONTH_NAMES = [ |
90 | 'January', 'February', 'March', 'April', 'May', 'June', | |
91 | 'July', 'August', 'September', 'October', 'November', 'December'] | |
92 | ||
a7aaa398 S |
93 | KNOWN_EXTENSIONS = ( |
94 | 'mp4', 'm4a', 'm4p', 'm4b', 'm4r', 'm4v', 'aac', | |
95 | 'flv', 'f4v', 'f4a', 'f4b', | |
96 | 'webm', 'ogg', 'ogv', 'oga', 'ogx', 'spx', 'opus', | |
97 | 'mkv', 'mka', 'mk3d', | |
98 | 'avi', 'divx', | |
99 | 'mov', | |
100 | 'asf', 'wmv', 'wma', | |
101 | '3gp', '3g2', | |
102 | 'mp3', | |
103 | 'flac', | |
104 | 'ape', | |
105 | 'wav', | |
106 | 'f4f', 'f4m', 'm3u8', 'smil') | |
107 | ||
c587cbb7 | 108 | # needed for sanitizing filenames in restricted mode |
c8827027 | 109 | ACCENT_CHARS = dict(zip('ÂÃÄÀÁÅÆÇÈÉÊËÌÍÎÏÐÑÒÓÔÕÖŐØŒÙÚÛÜŰÝÞßàáâãäåæçèéêëìíîïðñòóôõöőøœùúûüűýþÿ', |
110 | itertools.chain('AAAAAA', ['AE'], 'CEEEEIIIIDNOOOOOOO', ['OE'], 'UUUUUYP', ['ss'], | |
111 | 'aaaaaa', ['ae'], 'ceeeeiiiionooooooo', ['oe'], 'uuuuuypy'))) | |
c587cbb7 | 112 | |
46f59e89 S |
113 | DATE_FORMATS = ( |
114 | '%d %B %Y', | |
115 | '%d %b %Y', | |
116 | '%B %d %Y', | |
117 | '%b %d %Y', | |
118 | '%b %dst %Y %I:%M', | |
119 | '%b %dnd %Y %I:%M', | |
120 | '%b %dth %Y %I:%M', | |
121 | '%Y %m %d', | |
122 | '%Y-%m-%d', | |
123 | '%Y/%m/%d', | |
124 | '%Y/%m/%d %H:%M:%S', | |
125 | '%Y-%m-%d %H:%M:%S', | |
126 | '%Y-%m-%d %H:%M:%S.%f', | |
127 | '%d.%m.%Y %H:%M', | |
128 | '%d.%m.%Y %H.%M', | |
129 | '%Y-%m-%dT%H:%M:%SZ', | |
130 | '%Y-%m-%dT%H:%M:%S.%fZ', | |
131 | '%Y-%m-%dT%H:%M:%S.%f0Z', | |
132 | '%Y-%m-%dT%H:%M:%S', | |
133 | '%Y-%m-%dT%H:%M:%S.%f', | |
134 | '%Y-%m-%dT%H:%M', | |
135 | ) | |
136 | ||
137 | DATE_FORMATS_DAY_FIRST = list(DATE_FORMATS) | |
138 | DATE_FORMATS_DAY_FIRST.extend([ | |
139 | '%d-%m-%Y', | |
140 | '%d.%m.%Y', | |
141 | '%d.%m.%y', | |
142 | '%d/%m/%Y', | |
143 | '%d/%m/%y', | |
144 | '%d/%m/%Y %H:%M:%S', | |
145 | ]) | |
146 | ||
147 | DATE_FORMATS_MONTH_FIRST = list(DATE_FORMATS) | |
148 | DATE_FORMATS_MONTH_FIRST.extend([ | |
149 | '%m-%d-%Y', | |
150 | '%m.%d.%Y', | |
151 | '%m/%d/%Y', | |
152 | '%m/%d/%y', | |
153 | '%m/%d/%Y %H:%M:%S', | |
154 | ]) | |
155 | ||
7105440c | 156 | |
d77c3dfd | 157 | def preferredencoding(): |
59ae15a5 | 158 | """Get preferred encoding. |
d77c3dfd | 159 | |
59ae15a5 PH |
160 | Returns the best encoding scheme for the system, based on |
161 | locale.getpreferredencoding() and some further tweaks. | |
162 | """ | |
163 | try: | |
164 | pref = locale.getpreferredencoding() | |
28e614de | 165 | 'TEST'.encode(pref) |
70a1165b | 166 | except Exception: |
59ae15a5 | 167 | pref = 'UTF-8' |
bae611f2 | 168 | |
59ae15a5 | 169 | return pref |
d77c3dfd | 170 | |
f4bfd65f | 171 | |
181c8655 | 172 | def write_json_file(obj, fn): |
1394646a | 173 | """ Encode obj as JSON and write it to fn, atomically if possible """ |
181c8655 | 174 | |
92120217 | 175 | fn = encodeFilename(fn) |
61ee5aeb | 176 | if sys.version_info < (3, 0) and sys.platform != 'win32': |
ec5f6016 JMF |
177 | encoding = get_filesystem_encoding() |
178 | # os.path.basename returns a bytes object, but NamedTemporaryFile | |
179 | # will fail if the filename contains non ascii characters unless we | |
180 | # use a unicode object | |
181 | path_basename = lambda f: os.path.basename(fn).decode(encoding) | |
182 | # the same for os.path.dirname | |
183 | path_dirname = lambda f: os.path.dirname(fn).decode(encoding) | |
184 | else: | |
185 | path_basename = os.path.basename | |
186 | path_dirname = os.path.dirname | |
187 | ||
73159f99 S |
188 | args = { |
189 | 'suffix': '.tmp', | |
ec5f6016 JMF |
190 | 'prefix': path_basename(fn) + '.', |
191 | 'dir': path_dirname(fn), | |
73159f99 S |
192 | 'delete': False, |
193 | } | |
194 | ||
181c8655 PH |
195 | # In Python 2.x, json.dump expects a bytestream. |
196 | # In Python 3.x, it writes to a character stream | |
197 | if sys.version_info < (3, 0): | |
73159f99 | 198 | args['mode'] = 'wb' |
181c8655 | 199 | else: |
73159f99 S |
200 | args.update({ |
201 | 'mode': 'w', | |
202 | 'encoding': 'utf-8', | |
203 | }) | |
204 | ||
c86b6142 | 205 | tf = tempfile.NamedTemporaryFile(**compat_kwargs(args)) |
181c8655 PH |
206 | |
207 | try: | |
208 | with tf: | |
209 | json.dump(obj, tf) | |
1394646a IK |
210 | if sys.platform == 'win32': |
211 | # Need to remove existing file on Windows, else os.rename raises | |
212 | # WindowsError or FileExistsError. | |
213 | try: | |
214 | os.unlink(fn) | |
215 | except OSError: | |
216 | pass | |
181c8655 | 217 | os.rename(tf.name, fn) |
70a1165b | 218 | except Exception: |
181c8655 PH |
219 | try: |
220 | os.remove(tf.name) | |
221 | except OSError: | |
222 | pass | |
223 | raise | |
224 | ||
225 | ||
226 | if sys.version_info >= (2, 7): | |
ee114368 | 227 | def find_xpath_attr(node, xpath, key, val=None): |
59ae56fa | 228 | """ Find the xpath xpath[@key=val] """ |
5d2354f1 | 229 | assert re.match(r'^[a-zA-Z_-]+$', key) |
ee114368 | 230 | expr = xpath + ('[@%s]' % key if val is None else "[@%s='%s']" % (key, val)) |
59ae56fa PH |
231 | return node.find(expr) |
232 | else: | |
ee114368 | 233 | def find_xpath_attr(node, xpath, key, val=None): |
810c10ba | 234 | for f in node.findall(compat_xpath(xpath)): |
ee114368 S |
235 | if key not in f.attrib: |
236 | continue | |
237 | if val is None or f.attrib.get(key) == val: | |
59ae56fa PH |
238 | return f |
239 | return None | |
240 | ||
d7e66d39 JMF |
241 | # On python2.6 the xml.etree.ElementTree.Element methods don't support |
242 | # the namespace parameter | |
5f6a1245 JW |
243 | |
244 | ||
d7e66d39 JMF |
245 | def xpath_with_ns(path, ns_map): |
246 | components = [c.split(':') for c in path.split('/')] | |
247 | replaced = [] | |
248 | for c in components: | |
249 | if len(c) == 1: | |
250 | replaced.append(c[0]) | |
251 | else: | |
252 | ns, tag = c | |
253 | replaced.append('{%s}%s' % (ns_map[ns], tag)) | |
254 | return '/'.join(replaced) | |
255 | ||
d77c3dfd | 256 | |
a41fb80c | 257 | def xpath_element(node, xpath, name=None, fatal=False, default=NO_DEFAULT): |
578c0745 | 258 | def _find_xpath(xpath): |
810c10ba | 259 | return node.find(compat_xpath(xpath)) |
578c0745 S |
260 | |
261 | if isinstance(xpath, (str, compat_str)): | |
262 | n = _find_xpath(xpath) | |
263 | else: | |
264 | for xp in xpath: | |
265 | n = _find_xpath(xp) | |
266 | if n is not None: | |
267 | break | |
d74bebd5 | 268 | |
8e636da4 | 269 | if n is None: |
bf42a990 S |
270 | if default is not NO_DEFAULT: |
271 | return default | |
272 | elif fatal: | |
bf0ff932 PH |
273 | name = xpath if name is None else name |
274 | raise ExtractorError('Could not find XML element %s' % name) | |
275 | else: | |
276 | return None | |
a41fb80c S |
277 | return n |
278 | ||
279 | ||
280 | def xpath_text(node, xpath, name=None, fatal=False, default=NO_DEFAULT): | |
8e636da4 S |
281 | n = xpath_element(node, xpath, name, fatal=fatal, default=default) |
282 | if n is None or n == default: | |
283 | return n | |
284 | if n.text is None: | |
285 | if default is not NO_DEFAULT: | |
286 | return default | |
287 | elif fatal: | |
288 | name = xpath if name is None else name | |
289 | raise ExtractorError('Could not find XML element\'s text %s' % name) | |
290 | else: | |
291 | return None | |
292 | return n.text | |
a41fb80c S |
293 | |
294 | ||
295 | def xpath_attr(node, xpath, key, name=None, fatal=False, default=NO_DEFAULT): | |
296 | n = find_xpath_attr(node, xpath, key) | |
297 | if n is None: | |
298 | if default is not NO_DEFAULT: | |
299 | return default | |
300 | elif fatal: | |
301 | name = '%s[@%s]' % (xpath, key) if name is None else name | |
302 | raise ExtractorError('Could not find XML attribute %s' % name) | |
303 | else: | |
304 | return None | |
305 | return n.attrib[key] | |
bf0ff932 PH |
306 | |
307 | ||
9e6dd238 | 308 | def get_element_by_id(id, html): |
43e8fafd | 309 | """Return the content of the tag with the specified ID in the passed HTML document""" |
611c1dd9 | 310 | return get_element_by_attribute('id', id, html) |
43e8fafd | 311 | |
12ea2f30 | 312 | |
43e8fafd ND |
313 | def get_element_by_attribute(attribute, value, html): |
314 | """Return the content of the tag with the specified attribute in the passed HTML document""" | |
9e6dd238 | 315 | |
38285056 PH |
316 | m = re.search(r'''(?xs) |
317 | <([a-zA-Z0-9:._-]+) | |
abc97b5e | 318 | (?:\s+[a-zA-Z0-9:._-]+(?:=[a-zA-Z0-9:._-]*|="[^"]*"|='[^']*'))*? |
38285056 | 319 | \s+%s=['"]?%s['"]? |
abc97b5e | 320 | (?:\s+[a-zA-Z0-9:._-]+(?:=[a-zA-Z0-9:._-]*|="[^"]*"|='[^']*'))*? |
38285056 PH |
321 | \s*> |
322 | (?P<content>.*?) | |
323 | </\1> | |
324 | ''' % (re.escape(attribute), re.escape(value)), html) | |
325 | ||
326 | if not m: | |
327 | return None | |
328 | res = m.group('content') | |
329 | ||
330 | if res.startswith('"') or res.startswith("'"): | |
331 | res = res[1:-1] | |
a921f407 | 332 | |
38285056 | 333 | return unescapeHTML(res) |
a921f407 | 334 | |
c5229f39 | 335 | |
8bb56eee BF |
336 | class HTMLAttributeParser(compat_HTMLParser): |
337 | """Trivial HTML parser to gather the attributes for a single element""" | |
338 | def __init__(self): | |
c5229f39 | 339 | self.attrs = {} |
8bb56eee BF |
340 | compat_HTMLParser.__init__(self) |
341 | ||
342 | def handle_starttag(self, tag, attrs): | |
343 | self.attrs = dict(attrs) | |
344 | ||
c5229f39 | 345 | |
8bb56eee BF |
346 | def extract_attributes(html_element): |
347 | """Given a string for an HTML element such as | |
348 | <el | |
349 | a="foo" B="bar" c="&98;az" d=boz | |
350 | empty= noval entity="&" | |
351 | sq='"' dq="'" | |
352 | > | |
353 | Decode and return a dictionary of attributes. | |
354 | { | |
355 | 'a': 'foo', 'b': 'bar', c: 'baz', d: 'boz', | |
356 | 'empty': '', 'noval': None, 'entity': '&', | |
357 | 'sq': '"', 'dq': '\'' | |
358 | }. | |
359 | NB HTMLParser is stricter in Python 2.6 & 3.2 than in later versions, | |
360 | but the cases in the unit test will work for all of 2.6, 2.7, 3.2-3.5. | |
361 | """ | |
362 | parser = HTMLAttributeParser() | |
363 | parser.feed(html_element) | |
364 | parser.close() | |
365 | return parser.attrs | |
9e6dd238 | 366 | |
c5229f39 | 367 | |
9e6dd238 | 368 | def clean_html(html): |
59ae15a5 | 369 | """Clean an HTML snippet into a readable string""" |
dd622d7c PH |
370 | |
371 | if html is None: # Convenience for sanitizing descriptions etc. | |
372 | return html | |
373 | ||
59ae15a5 PH |
374 | # Newline vs <br /> |
375 | html = html.replace('\n', ' ') | |
6b3aef80 FV |
376 | html = re.sub(r'\s*<\s*br\s*/?\s*>\s*', '\n', html) |
377 | html = re.sub(r'<\s*/\s*p\s*>\s*<\s*p[^>]*>', '\n', html) | |
59ae15a5 PH |
378 | # Strip html tags |
379 | html = re.sub('<.*?>', '', html) | |
380 | # Replace html entities | |
381 | html = unescapeHTML(html) | |
7decf895 | 382 | return html.strip() |
9e6dd238 FV |
383 | |
384 | ||
d77c3dfd | 385 | def sanitize_open(filename, open_mode): |
59ae15a5 PH |
386 | """Try to open the given filename, and slightly tweak it if this fails. |
387 | ||
388 | Attempts to open the given filename. If this fails, it tries to change | |
389 | the filename slightly, step by step, until it's either able to open it | |
390 | or it fails and raises a final exception, like the standard open() | |
391 | function. | |
392 | ||
393 | It returns the tuple (stream, definitive_file_name). | |
394 | """ | |
395 | try: | |
28e614de | 396 | if filename == '-': |
59ae15a5 PH |
397 | if sys.platform == 'win32': |
398 | import msvcrt | |
399 | msvcrt.setmode(sys.stdout.fileno(), os.O_BINARY) | |
898280a0 | 400 | return (sys.stdout.buffer if hasattr(sys.stdout, 'buffer') else sys.stdout, filename) |
59ae15a5 PH |
401 | stream = open(encodeFilename(filename), open_mode) |
402 | return (stream, filename) | |
403 | except (IOError, OSError) as err: | |
f45c185f PH |
404 | if err.errno in (errno.EACCES,): |
405 | raise | |
59ae15a5 | 406 | |
f45c185f | 407 | # In case of error, try to remove win32 forbidden chars |
d55de57b | 408 | alt_filename = sanitize_path(filename) |
f45c185f PH |
409 | if alt_filename == filename: |
410 | raise | |
411 | else: | |
412 | # An exception here should be caught in the caller | |
d55de57b | 413 | stream = open(encodeFilename(alt_filename), open_mode) |
f45c185f | 414 | return (stream, alt_filename) |
d77c3dfd FV |
415 | |
416 | ||
417 | def timeconvert(timestr): | |
59ae15a5 PH |
418 | """Convert RFC 2822 defined time string into system timestamp""" |
419 | timestamp = None | |
420 | timetuple = email.utils.parsedate_tz(timestr) | |
421 | if timetuple is not None: | |
422 | timestamp = email.utils.mktime_tz(timetuple) | |
423 | return timestamp | |
1c469a94 | 424 | |
5f6a1245 | 425 | |
796173d0 | 426 | def sanitize_filename(s, restricted=False, is_id=False): |
59ae15a5 PH |
427 | """Sanitizes a string so it could be used as part of a filename. |
428 | If restricted is set, use a stricter subset of allowed characters. | |
796173d0 | 429 | Set is_id if this is not an arbitrary string, but an ID that should be kept if possible |
59ae15a5 PH |
430 | """ |
431 | def replace_insane(char): | |
c587cbb7 AT |
432 | if restricted and char in ACCENT_CHARS: |
433 | return ACCENT_CHARS[char] | |
59ae15a5 PH |
434 | if char == '?' or ord(char) < 32 or ord(char) == 127: |
435 | return '' | |
436 | elif char == '"': | |
437 | return '' if restricted else '\'' | |
438 | elif char == ':': | |
439 | return '_-' if restricted else ' -' | |
440 | elif char in '\\/|*<>': | |
441 | return '_' | |
627dcfff | 442 | if restricted and (char in '!&\'()[]{}$;`^,#' or char.isspace()): |
59ae15a5 PH |
443 | return '_' |
444 | if restricted and ord(char) > 127: | |
445 | return '_' | |
446 | return char | |
447 | ||
2aeb06d6 PH |
448 | # Handle timestamps |
449 | s = re.sub(r'[0-9]+(?::[0-9]+)+', lambda m: m.group(0).replace(':', '_'), s) | |
28e614de | 450 | result = ''.join(map(replace_insane, s)) |
796173d0 PH |
451 | if not is_id: |
452 | while '__' in result: | |
453 | result = result.replace('__', '_') | |
454 | result = result.strip('_') | |
455 | # Common case of "Foreign band name - English song title" | |
456 | if restricted and result.startswith('-_'): | |
457 | result = result[2:] | |
5a42414b PH |
458 | if result.startswith('-'): |
459 | result = '_' + result[len('-'):] | |
a7440261 | 460 | result = result.lstrip('.') |
796173d0 PH |
461 | if not result: |
462 | result = '_' | |
59ae15a5 | 463 | return result |
d77c3dfd | 464 | |
5f6a1245 | 465 | |
a2aaf4db S |
466 | def sanitize_path(s): |
467 | """Sanitizes and normalizes path on Windows""" | |
468 | if sys.platform != 'win32': | |
469 | return s | |
be531ef1 S |
470 | drive_or_unc, _ = os.path.splitdrive(s) |
471 | if sys.version_info < (2, 7) and not drive_or_unc: | |
472 | drive_or_unc, _ = os.path.splitunc(s) | |
473 | norm_path = os.path.normpath(remove_start(s, drive_or_unc)).split(os.path.sep) | |
474 | if drive_or_unc: | |
a2aaf4db S |
475 | norm_path.pop(0) |
476 | sanitized_path = [ | |
c90d16cf | 477 | path_part if path_part in ['.', '..'] else re.sub('(?:[/<>:"\\|\\\\?\\*]|[\s.]$)', '#', path_part) |
a2aaf4db | 478 | for path_part in norm_path] |
be531ef1 S |
479 | if drive_or_unc: |
480 | sanitized_path.insert(0, drive_or_unc + os.path.sep) | |
a2aaf4db S |
481 | return os.path.join(*sanitized_path) |
482 | ||
483 | ||
67dda517 S |
484 | # Prepend protocol-less URLs with `http:` scheme in order to mitigate the number of |
485 | # unwanted failures due to missing protocol | |
17bcc626 S |
486 | def sanitize_url(url): |
487 | return 'http:%s' % url if url.startswith('//') else url | |
488 | ||
489 | ||
67dda517 | 490 | def sanitized_Request(url, *args, **kwargs): |
17bcc626 | 491 | return compat_urllib_request.Request(sanitize_url(url), *args, **kwargs) |
67dda517 S |
492 | |
493 | ||
d77c3dfd | 494 | def orderedSet(iterable): |
59ae15a5 PH |
495 | """ Remove all duplicates from the input iterable """ |
496 | res = [] | |
497 | for el in iterable: | |
498 | if el not in res: | |
499 | res.append(el) | |
500 | return res | |
d77c3dfd | 501 | |
912b38b4 | 502 | |
55b2f099 | 503 | def _htmlentity_transform(entity_with_semicolon): |
4e408e47 | 504 | """Transforms an HTML entity to a character.""" |
55b2f099 YCH |
505 | entity = entity_with_semicolon[:-1] |
506 | ||
4e408e47 PH |
507 | # Known non-numeric HTML entity |
508 | if entity in compat_html_entities.name2codepoint: | |
509 | return compat_chr(compat_html_entities.name2codepoint[entity]) | |
510 | ||
55b2f099 YCH |
511 | # TODO: HTML5 allows entities without a semicolon. For example, |
512 | # 'Éric' should be decoded as 'Éric'. | |
513 | if entity_with_semicolon in compat_html_entities_html5: | |
514 | return compat_html_entities_html5[entity_with_semicolon] | |
515 | ||
91757b0f | 516 | mobj = re.match(r'#(x[0-9a-fA-F]+|[0-9]+)', entity) |
4e408e47 PH |
517 | if mobj is not None: |
518 | numstr = mobj.group(1) | |
28e614de | 519 | if numstr.startswith('x'): |
4e408e47 | 520 | base = 16 |
28e614de | 521 | numstr = '0%s' % numstr |
4e408e47 PH |
522 | else: |
523 | base = 10 | |
7aefc49c S |
524 | # See https://github.com/rg3/youtube-dl/issues/7518 |
525 | try: | |
526 | return compat_chr(int(numstr, base)) | |
527 | except ValueError: | |
528 | pass | |
4e408e47 PH |
529 | |
530 | # Unknown entity in name, return its literal representation | |
7a3f0c00 | 531 | return '&%s;' % entity |
4e408e47 PH |
532 | |
533 | ||
d77c3dfd | 534 | def unescapeHTML(s): |
912b38b4 PH |
535 | if s is None: |
536 | return None | |
537 | assert type(s) == compat_str | |
d77c3dfd | 538 | |
4e408e47 | 539 | return re.sub( |
55b2f099 | 540 | r'&([^;]+;)', lambda m: _htmlentity_transform(m.group(1)), s) |
d77c3dfd | 541 | |
8bf48f23 | 542 | |
aa49acd1 S |
543 | def get_subprocess_encoding(): |
544 | if sys.platform == 'win32' and sys.getwindowsversion()[0] >= 5: | |
545 | # For subprocess calls, encode with locale encoding | |
546 | # Refer to http://stackoverflow.com/a/9951851/35070 | |
547 | encoding = preferredencoding() | |
548 | else: | |
549 | encoding = sys.getfilesystemencoding() | |
550 | if encoding is None: | |
551 | encoding = 'utf-8' | |
552 | return encoding | |
553 | ||
554 | ||
8bf48f23 | 555 | def encodeFilename(s, for_subprocess=False): |
59ae15a5 PH |
556 | """ |
557 | @param s The name of the file | |
558 | """ | |
d77c3dfd | 559 | |
8bf48f23 | 560 | assert type(s) == compat_str |
d77c3dfd | 561 | |
59ae15a5 PH |
562 | # Python 3 has a Unicode API |
563 | if sys.version_info >= (3, 0): | |
564 | return s | |
0f00efed | 565 | |
aa49acd1 S |
566 | # Pass '' directly to use Unicode APIs on Windows 2000 and up |
567 | # (Detecting Windows NT 4 is tricky because 'major >= 4' would | |
568 | # match Windows 9x series as well. Besides, NT 4 is obsolete.) | |
569 | if not for_subprocess and sys.platform == 'win32' and sys.getwindowsversion()[0] >= 5: | |
570 | return s | |
571 | ||
8ee239e9 YCH |
572 | # Jython assumes filenames are Unicode strings though reported as Python 2.x compatible |
573 | if sys.platform.startswith('java'): | |
574 | return s | |
575 | ||
aa49acd1 S |
576 | return s.encode(get_subprocess_encoding(), 'ignore') |
577 | ||
578 | ||
579 | def decodeFilename(b, for_subprocess=False): | |
580 | ||
581 | if sys.version_info >= (3, 0): | |
582 | return b | |
583 | ||
584 | if not isinstance(b, bytes): | |
585 | return b | |
586 | ||
587 | return b.decode(get_subprocess_encoding(), 'ignore') | |
8bf48f23 | 588 | |
f07b74fc PH |
589 | |
590 | def encodeArgument(s): | |
591 | if not isinstance(s, compat_str): | |
592 | # Legacy code that uses byte strings | |
593 | # Uncomment the following line after fixing all post processors | |
7af808a5 | 594 | # assert False, 'Internal error: %r should be of type %r, is %r' % (s, compat_str, type(s)) |
f07b74fc PH |
595 | s = s.decode('ascii') |
596 | return encodeFilename(s, True) | |
597 | ||
598 | ||
aa49acd1 S |
599 | def decodeArgument(b): |
600 | return decodeFilename(b, True) | |
601 | ||
602 | ||
8271226a PH |
603 | def decodeOption(optval): |
604 | if optval is None: | |
605 | return optval | |
606 | if isinstance(optval, bytes): | |
607 | optval = optval.decode(preferredencoding()) | |
608 | ||
609 | assert isinstance(optval, compat_str) | |
610 | return optval | |
1c256f70 | 611 | |
5f6a1245 | 612 | |
4539dd30 PH |
613 | def formatSeconds(secs): |
614 | if secs > 3600: | |
615 | return '%d:%02d:%02d' % (secs // 3600, (secs % 3600) // 60, secs % 60) | |
616 | elif secs > 60: | |
617 | return '%d:%02d' % (secs // 60, secs % 60) | |
618 | else: | |
619 | return '%d' % secs | |
620 | ||
a0ddb8a2 | 621 | |
be4a824d PH |
622 | def make_HTTPS_handler(params, **kwargs): |
623 | opts_no_check_certificate = params.get('nocheckcertificate', False) | |
0db261ba | 624 | if hasattr(ssl, 'create_default_context'): # Python >= 3.4 or 2.7.9 |
be5f2c19 | 625 | context = ssl.create_default_context(ssl.Purpose.SERVER_AUTH) |
0db261ba | 626 | if opts_no_check_certificate: |
be5f2c19 | 627 | context.check_hostname = False |
0db261ba | 628 | context.verify_mode = ssl.CERT_NONE |
a2366922 | 629 | try: |
be4a824d | 630 | return YoutubeDLHTTPSHandler(params, context=context, **kwargs) |
a2366922 PH |
631 | except TypeError: |
632 | # Python 2.7.8 | |
633 | # (create_default_context present but HTTPSHandler has no context=) | |
634 | pass | |
635 | ||
636 | if sys.version_info < (3, 2): | |
d7932313 | 637 | return YoutubeDLHTTPSHandler(params, **kwargs) |
aa37e3d4 | 638 | else: # Python < 3.4 |
d7932313 | 639 | context = ssl.SSLContext(ssl.PROTOCOL_TLSv1) |
ea6d901e | 640 | context.verify_mode = (ssl.CERT_NONE |
dca08720 | 641 | if opts_no_check_certificate |
ea6d901e | 642 | else ssl.CERT_REQUIRED) |
303b479e | 643 | context.set_default_verify_paths() |
be4a824d | 644 | return YoutubeDLHTTPSHandler(params, context=context, **kwargs) |
ea6d901e | 645 | |
732ea2f0 | 646 | |
08f2a92c JMF |
647 | def bug_reports_message(): |
648 | if ytdl_is_updateable(): | |
649 | update_cmd = 'type youtube-dl -U to update' | |
650 | else: | |
651 | update_cmd = 'see https://yt-dl.org/update on how to update' | |
652 | msg = '; please report this issue on https://yt-dl.org/bug .' | |
653 | msg += ' Make sure you are using the latest version; %s.' % update_cmd | |
654 | msg += ' Be sure to call youtube-dl with the --verbose flag and include its complete output.' | |
655 | return msg | |
656 | ||
657 | ||
1c256f70 PH |
658 | class ExtractorError(Exception): |
659 | """Error during info extraction.""" | |
5f6a1245 | 660 | |
d11271dd | 661 | def __init__(self, msg, tb=None, expected=False, cause=None, video_id=None): |
9a82b238 PH |
662 | """ tb, if given, is the original traceback (so that it can be printed out). |
663 | If expected is set, this is a normal error message and most likely not a bug in youtube-dl. | |
664 | """ | |
665 | ||
666 | if sys.exc_info()[0] in (compat_urllib_error.URLError, socket.timeout, UnavailableVideoError): | |
667 | expected = True | |
d11271dd PH |
668 | if video_id is not None: |
669 | msg = video_id + ': ' + msg | |
410f3e73 | 670 | if cause: |
28e614de | 671 | msg += ' (caused by %r)' % cause |
9a82b238 | 672 | if not expected: |
08f2a92c | 673 | msg += bug_reports_message() |
1c256f70 | 674 | super(ExtractorError, self).__init__(msg) |
d5979c5d | 675 | |
1c256f70 | 676 | self.traceback = tb |
8cc83b8d | 677 | self.exc_info = sys.exc_info() # preserve original exception |
2eabb802 | 678 | self.cause = cause |
d11271dd | 679 | self.video_id = video_id |
1c256f70 | 680 | |
01951dda PH |
681 | def format_traceback(self): |
682 | if self.traceback is None: | |
683 | return None | |
28e614de | 684 | return ''.join(traceback.format_tb(self.traceback)) |
01951dda | 685 | |
1c256f70 | 686 | |
416c7fcb PH |
687 | class UnsupportedError(ExtractorError): |
688 | def __init__(self, url): | |
689 | super(UnsupportedError, self).__init__( | |
690 | 'Unsupported URL: %s' % url, expected=True) | |
691 | self.url = url | |
692 | ||
693 | ||
55b3e45b JMF |
694 | class RegexNotFoundError(ExtractorError): |
695 | """Error when a regex didn't match""" | |
696 | pass | |
697 | ||
698 | ||
d77c3dfd | 699 | class DownloadError(Exception): |
59ae15a5 | 700 | """Download Error exception. |
d77c3dfd | 701 | |
59ae15a5 PH |
702 | This exception may be thrown by FileDownloader objects if they are not |
703 | configured to continue on errors. They will contain the appropriate | |
704 | error message. | |
705 | """ | |
5f6a1245 | 706 | |
8cc83b8d FV |
707 | def __init__(self, msg, exc_info=None): |
708 | """ exc_info, if given, is the original exception that caused the trouble (as returned by sys.exc_info()). """ | |
709 | super(DownloadError, self).__init__(msg) | |
710 | self.exc_info = exc_info | |
d77c3dfd FV |
711 | |
712 | ||
713 | class SameFileError(Exception): | |
59ae15a5 | 714 | """Same File exception. |
d77c3dfd | 715 | |
59ae15a5 PH |
716 | This exception will be thrown by FileDownloader objects if they detect |
717 | multiple files would have to be downloaded to the same file on disk. | |
718 | """ | |
719 | pass | |
d77c3dfd FV |
720 | |
721 | ||
722 | class PostProcessingError(Exception): | |
59ae15a5 | 723 | """Post Processing exception. |
d77c3dfd | 724 | |
59ae15a5 PH |
725 | This exception may be raised by PostProcessor's .run() method to |
726 | indicate an error in the postprocessing task. | |
727 | """ | |
5f6a1245 | 728 | |
7851b379 PH |
729 | def __init__(self, msg): |
730 | self.msg = msg | |
d77c3dfd | 731 | |
5f6a1245 | 732 | |
d77c3dfd | 733 | class MaxDownloadsReached(Exception): |
59ae15a5 PH |
734 | """ --max-downloads limit has been reached. """ |
735 | pass | |
d77c3dfd FV |
736 | |
737 | ||
738 | class UnavailableVideoError(Exception): | |
59ae15a5 | 739 | """Unavailable Format exception. |
d77c3dfd | 740 | |
59ae15a5 PH |
741 | This exception will be thrown when a video is requested |
742 | in a format that is not available for that video. | |
743 | """ | |
744 | pass | |
d77c3dfd FV |
745 | |
746 | ||
747 | class ContentTooShortError(Exception): | |
59ae15a5 | 748 | """Content Too Short exception. |
d77c3dfd | 749 | |
59ae15a5 PH |
750 | This exception may be raised by FileDownloader objects when a file they |
751 | download is too small for what the server announced first, indicating | |
752 | the connection was probably interrupted. | |
753 | """ | |
d77c3dfd | 754 | |
59ae15a5 | 755 | def __init__(self, downloaded, expected): |
2c7ed247 | 756 | # Both in bytes |
59ae15a5 PH |
757 | self.downloaded = downloaded |
758 | self.expected = expected | |
d77c3dfd | 759 | |
5f6a1245 | 760 | |
c5a59d93 | 761 | def _create_http_connection(ydl_handler, http_class, is_https, *args, **kwargs): |
e5e78797 S |
762 | # Working around python 2 bug (see http://bugs.python.org/issue17849) by limiting |
763 | # expected HTTP responses to meet HTTP/1.0 or later (see also | |
764 | # https://github.com/rg3/youtube-dl/issues/6727) | |
765 | if sys.version_info < (3, 0): | |
5a1a2e94 | 766 | kwargs[b'strict'] = True |
be4a824d PH |
767 | hc = http_class(*args, **kwargs) |
768 | source_address = ydl_handler._params.get('source_address') | |
769 | if source_address is not None: | |
770 | sa = (source_address, 0) | |
771 | if hasattr(hc, 'source_address'): # Python 2.7+ | |
772 | hc.source_address = sa | |
773 | else: # Python 2.6 | |
774 | def _hc_connect(self, *args, **kwargs): | |
775 | sock = compat_socket_create_connection( | |
776 | (self.host, self.port), self.timeout, sa) | |
777 | if is_https: | |
d7932313 PH |
778 | self.sock = ssl.wrap_socket( |
779 | sock, self.key_file, self.cert_file, | |
780 | ssl_version=ssl.PROTOCOL_TLSv1) | |
be4a824d PH |
781 | else: |
782 | self.sock = sock | |
783 | hc.connect = functools.partial(_hc_connect, hc) | |
784 | ||
785 | return hc | |
786 | ||
787 | ||
87f0e62d | 788 | def handle_youtubedl_headers(headers): |
992fc9d6 YCH |
789 | filtered_headers = headers |
790 | ||
791 | if 'Youtubedl-no-compression' in filtered_headers: | |
792 | filtered_headers = dict((k, v) for k, v in filtered_headers.items() if k.lower() != 'accept-encoding') | |
87f0e62d | 793 | del filtered_headers['Youtubedl-no-compression'] |
87f0e62d | 794 | |
992fc9d6 | 795 | return filtered_headers |
87f0e62d YCH |
796 | |
797 | ||
acebc9cd | 798 | class YoutubeDLHandler(compat_urllib_request.HTTPHandler): |
59ae15a5 PH |
799 | """Handler for HTTP requests and responses. |
800 | ||
801 | This class, when installed with an OpenerDirector, automatically adds | |
802 | the standard headers to every HTTP request and handles gzipped and | |
803 | deflated responses from web servers. If compression is to be avoided in | |
804 | a particular request, the original request in the program code only has | |
0424ec30 | 805 | to include the HTTP header "Youtubedl-no-compression", which will be |
59ae15a5 PH |
806 | removed before making the real request. |
807 | ||
808 | Part of this code was copied from: | |
809 | ||
810 | http://techknack.net/python-urllib2-handlers/ | |
811 | ||
812 | Andrew Rowls, the author of that code, agreed to release it to the | |
813 | public domain. | |
814 | """ | |
815 | ||
be4a824d PH |
816 | def __init__(self, params, *args, **kwargs): |
817 | compat_urllib_request.HTTPHandler.__init__(self, *args, **kwargs) | |
818 | self._params = params | |
819 | ||
820 | def http_open(self, req): | |
71aff188 YCH |
821 | conn_class = compat_http_client.HTTPConnection |
822 | ||
823 | socks_proxy = req.headers.get('Ytdl-socks-proxy') | |
824 | if socks_proxy: | |
825 | conn_class = make_socks_conn_class(conn_class, socks_proxy) | |
826 | del req.headers['Ytdl-socks-proxy'] | |
827 | ||
be4a824d | 828 | return self.do_open(functools.partial( |
71aff188 | 829 | _create_http_connection, self, conn_class, False), |
be4a824d PH |
830 | req) |
831 | ||
59ae15a5 PH |
832 | @staticmethod |
833 | def deflate(data): | |
834 | try: | |
835 | return zlib.decompress(data, -zlib.MAX_WBITS) | |
836 | except zlib.error: | |
837 | return zlib.decompress(data) | |
838 | ||
839 | @staticmethod | |
840 | def addinfourl_wrapper(stream, headers, url, code): | |
841 | if hasattr(compat_urllib_request.addinfourl, 'getcode'): | |
842 | return compat_urllib_request.addinfourl(stream, headers, url, code) | |
843 | ret = compat_urllib_request.addinfourl(stream, headers, url) | |
844 | ret.code = code | |
845 | return ret | |
846 | ||
acebc9cd | 847 | def http_request(self, req): |
51f267d9 S |
848 | # According to RFC 3986, URLs can not contain non-ASCII characters, however this is not |
849 | # always respected by websites, some tend to give out URLs with non percent-encoded | |
850 | # non-ASCII characters (see telemb.py, ard.py [#3412]) | |
851 | # urllib chokes on URLs with non-ASCII characters (see http://bugs.python.org/issue3991) | |
852 | # To work around aforementioned issue we will replace request's original URL with | |
853 | # percent-encoded one | |
854 | # Since redirects are also affected (e.g. http://www.southpark.de/alle-episoden/s18e09) | |
855 | # the code of this workaround has been moved here from YoutubeDL.urlopen() | |
856 | url = req.get_full_url() | |
857 | url_escaped = escape_url(url) | |
858 | ||
859 | # Substitute URL if any change after escaping | |
860 | if url != url_escaped: | |
15d260eb | 861 | req = update_Request(req, url=url_escaped) |
51f267d9 | 862 | |
33ac271b | 863 | for h, v in std_headers.items(): |
3d5f7a39 JK |
864 | # Capitalize is needed because of Python bug 2275: http://bugs.python.org/issue2275 |
865 | # The dict keys are capitalized because of this bug by urllib | |
866 | if h.capitalize() not in req.headers: | |
33ac271b | 867 | req.add_header(h, v) |
87f0e62d YCH |
868 | |
869 | req.headers = handle_youtubedl_headers(req.headers) | |
989b4b2b PH |
870 | |
871 | if sys.version_info < (2, 7) and '#' in req.get_full_url(): | |
872 | # Python 2.6 is brain-dead when it comes to fragments | |
873 | req._Request__original = req._Request__original.partition('#')[0] | |
874 | req._Request__r_type = req._Request__r_type.partition('#')[0] | |
875 | ||
59ae15a5 PH |
876 | return req |
877 | ||
acebc9cd | 878 | def http_response(self, req, resp): |
59ae15a5 PH |
879 | old_resp = resp |
880 | # gzip | |
881 | if resp.headers.get('Content-encoding', '') == 'gzip': | |
aa3e9507 PH |
882 | content = resp.read() |
883 | gz = gzip.GzipFile(fileobj=io.BytesIO(content), mode='rb') | |
884 | try: | |
885 | uncompressed = io.BytesIO(gz.read()) | |
886 | except IOError as original_ioerror: | |
887 | # There may be junk add the end of the file | |
888 | # See http://stackoverflow.com/q/4928560/35070 for details | |
889 | for i in range(1, 1024): | |
890 | try: | |
891 | gz = gzip.GzipFile(fileobj=io.BytesIO(content[:-i]), mode='rb') | |
892 | uncompressed = io.BytesIO(gz.read()) | |
893 | except IOError: | |
894 | continue | |
895 | break | |
896 | else: | |
897 | raise original_ioerror | |
898 | resp = self.addinfourl_wrapper(uncompressed, old_resp.headers, old_resp.url, old_resp.code) | |
59ae15a5 | 899 | resp.msg = old_resp.msg |
c047270c | 900 | del resp.headers['Content-encoding'] |
59ae15a5 PH |
901 | # deflate |
902 | if resp.headers.get('Content-encoding', '') == 'deflate': | |
903 | gz = io.BytesIO(self.deflate(resp.read())) | |
904 | resp = self.addinfourl_wrapper(gz, old_resp.headers, old_resp.url, old_resp.code) | |
905 | resp.msg = old_resp.msg | |
c047270c | 906 | del resp.headers['Content-encoding'] |
ad729172 S |
907 | # Percent-encode redirect URL of Location HTTP header to satisfy RFC 3986 (see |
908 | # https://github.com/rg3/youtube-dl/issues/6457). | |
5a4d9ddb S |
909 | if 300 <= resp.code < 400: |
910 | location = resp.headers.get('Location') | |
911 | if location: | |
912 | # As of RFC 2616 default charset is iso-8859-1 that is respected by python 3 | |
913 | if sys.version_info >= (3, 0): | |
914 | location = location.encode('iso-8859-1').decode('utf-8') | |
0ea59007 YCH |
915 | else: |
916 | location = location.decode('utf-8') | |
5a4d9ddb S |
917 | location_escaped = escape_url(location) |
918 | if location != location_escaped: | |
919 | del resp.headers['Location'] | |
9a4aec8b YCH |
920 | if sys.version_info < (3, 0): |
921 | location_escaped = location_escaped.encode('utf-8') | |
5a4d9ddb | 922 | resp.headers['Location'] = location_escaped |
59ae15a5 | 923 | return resp |
0f8d03f8 | 924 | |
acebc9cd PH |
925 | https_request = http_request |
926 | https_response = http_response | |
bf50b038 | 927 | |
5de90176 | 928 | |
71aff188 YCH |
929 | def make_socks_conn_class(base_class, socks_proxy): |
930 | assert issubclass(base_class, ( | |
931 | compat_http_client.HTTPConnection, compat_http_client.HTTPSConnection)) | |
932 | ||
933 | url_components = compat_urlparse.urlparse(socks_proxy) | |
934 | if url_components.scheme.lower() == 'socks5': | |
935 | socks_type = ProxyType.SOCKS5 | |
936 | elif url_components.scheme.lower() in ('socks', 'socks4'): | |
937 | socks_type = ProxyType.SOCKS4 | |
51fb4995 YCH |
938 | elif url_components.scheme.lower() == 'socks4a': |
939 | socks_type = ProxyType.SOCKS4A | |
71aff188 | 940 | |
cdd94c2e YCH |
941 | def unquote_if_non_empty(s): |
942 | if not s: | |
943 | return s | |
944 | return compat_urllib_parse_unquote_plus(s) | |
945 | ||
71aff188 YCH |
946 | proxy_args = ( |
947 | socks_type, | |
948 | url_components.hostname, url_components.port or 1080, | |
949 | True, # Remote DNS | |
cdd94c2e YCH |
950 | unquote_if_non_empty(url_components.username), |
951 | unquote_if_non_empty(url_components.password), | |
71aff188 YCH |
952 | ) |
953 | ||
954 | class SocksConnection(base_class): | |
955 | def connect(self): | |
956 | self.sock = sockssocket() | |
957 | self.sock.setproxy(*proxy_args) | |
958 | if type(self.timeout) in (int, float): | |
959 | self.sock.settimeout(self.timeout) | |
960 | self.sock.connect((self.host, self.port)) | |
961 | ||
962 | if isinstance(self, compat_http_client.HTTPSConnection): | |
963 | if hasattr(self, '_context'): # Python > 2.6 | |
964 | self.sock = self._context.wrap_socket( | |
965 | self.sock, server_hostname=self.host) | |
966 | else: | |
967 | self.sock = ssl.wrap_socket(self.sock) | |
968 | ||
969 | return SocksConnection | |
970 | ||
971 | ||
be4a824d PH |
972 | class YoutubeDLHTTPSHandler(compat_urllib_request.HTTPSHandler): |
973 | def __init__(self, params, https_conn_class=None, *args, **kwargs): | |
974 | compat_urllib_request.HTTPSHandler.__init__(self, *args, **kwargs) | |
975 | self._https_conn_class = https_conn_class or compat_http_client.HTTPSConnection | |
976 | self._params = params | |
977 | ||
978 | def https_open(self, req): | |
4f264c02 | 979 | kwargs = {} |
71aff188 YCH |
980 | conn_class = self._https_conn_class |
981 | ||
4f264c02 JMF |
982 | if hasattr(self, '_context'): # python > 2.6 |
983 | kwargs['context'] = self._context | |
984 | if hasattr(self, '_check_hostname'): # python 3.x | |
985 | kwargs['check_hostname'] = self._check_hostname | |
71aff188 YCH |
986 | |
987 | socks_proxy = req.headers.get('Ytdl-socks-proxy') | |
988 | if socks_proxy: | |
989 | conn_class = make_socks_conn_class(conn_class, socks_proxy) | |
990 | del req.headers['Ytdl-socks-proxy'] | |
991 | ||
be4a824d | 992 | return self.do_open(functools.partial( |
71aff188 | 993 | _create_http_connection, self, conn_class, True), |
4f264c02 | 994 | req, **kwargs) |
be4a824d PH |
995 | |
996 | ||
a6420bf5 S |
997 | class YoutubeDLCookieProcessor(compat_urllib_request.HTTPCookieProcessor): |
998 | def __init__(self, cookiejar=None): | |
999 | compat_urllib_request.HTTPCookieProcessor.__init__(self, cookiejar) | |
1000 | ||
1001 | def http_response(self, request, response): | |
1002 | # Python 2 will choke on next HTTP request in row if there are non-ASCII | |
1003 | # characters in Set-Cookie HTTP header of last response (see | |
1004 | # https://github.com/rg3/youtube-dl/issues/6769). | |
1005 | # In order to at least prevent crashing we will percent encode Set-Cookie | |
1006 | # header before HTTPCookieProcessor starts processing it. | |
e28034c5 S |
1007 | # if sys.version_info < (3, 0) and response.headers: |
1008 | # for set_cookie_header in ('Set-Cookie', 'Set-Cookie2'): | |
1009 | # set_cookie = response.headers.get(set_cookie_header) | |
1010 | # if set_cookie: | |
1011 | # set_cookie_escaped = compat_urllib_parse.quote(set_cookie, b"%/;:@&=+$,!~*'()?#[] ") | |
1012 | # if set_cookie != set_cookie_escaped: | |
1013 | # del response.headers[set_cookie_header] | |
1014 | # response.headers[set_cookie_header] = set_cookie_escaped | |
a6420bf5 S |
1015 | return compat_urllib_request.HTTPCookieProcessor.http_response(self, request, response) |
1016 | ||
1017 | https_request = compat_urllib_request.HTTPCookieProcessor.http_request | |
1018 | https_response = http_response | |
1019 | ||
1020 | ||
46f59e89 S |
1021 | def extract_timezone(date_str): |
1022 | m = re.search( | |
1023 | r'^.{8,}?(?P<tz>Z$| ?(?P<sign>\+|-)(?P<hours>[0-9]{2}):?(?P<minutes>[0-9]{2})$)', | |
1024 | date_str) | |
1025 | if not m: | |
1026 | timezone = datetime.timedelta() | |
1027 | else: | |
1028 | date_str = date_str[:-len(m.group('tz'))] | |
1029 | if not m.group('sign'): | |
1030 | timezone = datetime.timedelta() | |
1031 | else: | |
1032 | sign = 1 if m.group('sign') == '+' else -1 | |
1033 | timezone = datetime.timedelta( | |
1034 | hours=sign * int(m.group('hours')), | |
1035 | minutes=sign * int(m.group('minutes'))) | |
1036 | return timezone, date_str | |
1037 | ||
1038 | ||
08b38d54 | 1039 | def parse_iso8601(date_str, delimiter='T', timezone=None): |
912b38b4 PH |
1040 | """ Return a UNIX timestamp from the given date """ |
1041 | ||
1042 | if date_str is None: | |
1043 | return None | |
1044 | ||
52c3a6e4 S |
1045 | date_str = re.sub(r'\.[0-9]+', '', date_str) |
1046 | ||
08b38d54 | 1047 | if timezone is None: |
46f59e89 S |
1048 | timezone, date_str = extract_timezone(date_str) |
1049 | ||
52c3a6e4 S |
1050 | try: |
1051 | date_format = '%Y-%m-%d{0}%H:%M:%S'.format(delimiter) | |
1052 | dt = datetime.datetime.strptime(date_str, date_format) - timezone | |
1053 | return calendar.timegm(dt.timetuple()) | |
1054 | except ValueError: | |
1055 | pass | |
912b38b4 PH |
1056 | |
1057 | ||
46f59e89 S |
1058 | def date_formats(day_first=True): |
1059 | return DATE_FORMATS_DAY_FIRST if day_first else DATE_FORMATS_MONTH_FIRST | |
1060 | ||
1061 | ||
42bdd9d0 | 1062 | def unified_strdate(date_str, day_first=True): |
bf50b038 | 1063 | """Return a string with the date in the format YYYYMMDD""" |
64e7ad60 PH |
1064 | |
1065 | if date_str is None: | |
1066 | return None | |
bf50b038 | 1067 | upload_date = None |
5f6a1245 | 1068 | # Replace commas |
026fcc04 | 1069 | date_str = date_str.replace(',', ' ') |
42bdd9d0 | 1070 | # Remove AM/PM + timezone |
9bb8e0a3 | 1071 | date_str = re.sub(r'(?i)\s*(?:AM|PM)(?:\s+[A-Z]+)?', '', date_str) |
46f59e89 | 1072 | _, date_str = extract_timezone(date_str) |
42bdd9d0 | 1073 | |
46f59e89 | 1074 | for expression in date_formats(day_first): |
bf50b038 JMF |
1075 | try: |
1076 | upload_date = datetime.datetime.strptime(date_str, expression).strftime('%Y%m%d') | |
5de90176 | 1077 | except ValueError: |
bf50b038 | 1078 | pass |
42393ce2 PH |
1079 | if upload_date is None: |
1080 | timetuple = email.utils.parsedate_tz(date_str) | |
1081 | if timetuple: | |
c6b9cf05 S |
1082 | try: |
1083 | upload_date = datetime.datetime(*timetuple[:6]).strftime('%Y%m%d') | |
1084 | except ValueError: | |
1085 | pass | |
6a750402 JMF |
1086 | if upload_date is not None: |
1087 | return compat_str(upload_date) | |
bf50b038 | 1088 | |
5f6a1245 | 1089 | |
46f59e89 S |
1090 | def unified_timestamp(date_str, day_first=True): |
1091 | if date_str is None: | |
1092 | return None | |
1093 | ||
1094 | date_str = date_str.replace(',', ' ') | |
1095 | ||
1096 | pm_delta = datetime.timedelta(hours=12 if re.search(r'(?i)PM', date_str) else 0) | |
1097 | timezone, date_str = extract_timezone(date_str) | |
1098 | ||
1099 | # Remove AM/PM + timezone | |
1100 | date_str = re.sub(r'(?i)\s*(?:AM|PM)(?:\s+[A-Z]+)?', '', date_str) | |
1101 | ||
1102 | for expression in date_formats(day_first): | |
1103 | try: | |
1104 | dt = datetime.datetime.strptime(date_str, expression) - timezone + pm_delta | |
1105 | return calendar.timegm(dt.timetuple()) | |
1106 | except ValueError: | |
1107 | pass | |
1108 | timetuple = email.utils.parsedate_tz(date_str) | |
1109 | if timetuple: | |
1110 | return calendar.timegm(timetuple.timetuple()) | |
1111 | ||
1112 | ||
28e614de | 1113 | def determine_ext(url, default_ext='unknown_video'): |
f4776371 S |
1114 | if url is None: |
1115 | return default_ext | |
9cb9a5df | 1116 | guess = url.partition('?')[0].rpartition('.')[2] |
73e79f2a PH |
1117 | if re.match(r'^[A-Za-z0-9]+$', guess): |
1118 | return guess | |
a7aaa398 S |
1119 | # Try extract ext from URLs like http://example.com/foo/bar.mp4/?download |
1120 | elif guess.rstrip('/') in KNOWN_EXTENSIONS: | |
9cb9a5df | 1121 | return guess.rstrip('/') |
73e79f2a | 1122 | else: |
cbdbb766 | 1123 | return default_ext |
73e79f2a | 1124 | |
5f6a1245 | 1125 | |
d4051a8e | 1126 | def subtitles_filename(filename, sub_lang, sub_format): |
28e614de | 1127 | return filename.rsplit('.', 1)[0] + '.' + sub_lang + '.' + sub_format |
d4051a8e | 1128 | |
5f6a1245 | 1129 | |
bd558525 | 1130 | def date_from_str(date_str): |
37254abc JMF |
1131 | """ |
1132 | Return a datetime object from a string in the format YYYYMMDD or | |
1133 | (now|today)[+-][0-9](day|week|month|year)(s)?""" | |
1134 | today = datetime.date.today() | |
f8795e10 | 1135 | if date_str in ('now', 'today'): |
37254abc | 1136 | return today |
f8795e10 PH |
1137 | if date_str == 'yesterday': |
1138 | return today - datetime.timedelta(days=1) | |
37254abc JMF |
1139 | match = re.match('(now|today)(?P<sign>[+-])(?P<time>\d+)(?P<unit>day|week|month|year)(s)?', date_str) |
1140 | if match is not None: | |
1141 | sign = match.group('sign') | |
1142 | time = int(match.group('time')) | |
1143 | if sign == '-': | |
1144 | time = -time | |
1145 | unit = match.group('unit') | |
dfb1b146 | 1146 | # A bad approximation? |
37254abc JMF |
1147 | if unit == 'month': |
1148 | unit = 'day' | |
1149 | time *= 30 | |
1150 | elif unit == 'year': | |
1151 | unit = 'day' | |
1152 | time *= 365 | |
1153 | unit += 's' | |
1154 | delta = datetime.timedelta(**{unit: time}) | |
1155 | return today + delta | |
611c1dd9 | 1156 | return datetime.datetime.strptime(date_str, '%Y%m%d').date() |
5f6a1245 JW |
1157 | |
1158 | ||
e63fc1be | 1159 | def hyphenate_date(date_str): |
1160 | """ | |
1161 | Convert a date in 'YYYYMMDD' format to 'YYYY-MM-DD' format""" | |
1162 | match = re.match(r'^(\d\d\d\d)(\d\d)(\d\d)$', date_str) | |
1163 | if match is not None: | |
1164 | return '-'.join(match.groups()) | |
1165 | else: | |
1166 | return date_str | |
1167 | ||
5f6a1245 | 1168 | |
bd558525 JMF |
1169 | class DateRange(object): |
1170 | """Represents a time interval between two dates""" | |
5f6a1245 | 1171 | |
bd558525 JMF |
1172 | def __init__(self, start=None, end=None): |
1173 | """start and end must be strings in the format accepted by date""" | |
1174 | if start is not None: | |
1175 | self.start = date_from_str(start) | |
1176 | else: | |
1177 | self.start = datetime.datetime.min.date() | |
1178 | if end is not None: | |
1179 | self.end = date_from_str(end) | |
1180 | else: | |
1181 | self.end = datetime.datetime.max.date() | |
37254abc | 1182 | if self.start > self.end: |
bd558525 | 1183 | raise ValueError('Date range: "%s" , the start date must be before the end date' % self) |
5f6a1245 | 1184 | |
bd558525 JMF |
1185 | @classmethod |
1186 | def day(cls, day): | |
1187 | """Returns a range that only contains the given day""" | |
5f6a1245 JW |
1188 | return cls(day, day) |
1189 | ||
bd558525 JMF |
1190 | def __contains__(self, date): |
1191 | """Check if the date is in the range""" | |
37254abc JMF |
1192 | if not isinstance(date, datetime.date): |
1193 | date = date_from_str(date) | |
1194 | return self.start <= date <= self.end | |
5f6a1245 | 1195 | |
bd558525 | 1196 | def __str__(self): |
5f6a1245 | 1197 | return '%s - %s' % (self.start.isoformat(), self.end.isoformat()) |
c496ca96 PH |
1198 | |
1199 | ||
1200 | def platform_name(): | |
1201 | """ Returns the platform name as a compat_str """ | |
1202 | res = platform.platform() | |
1203 | if isinstance(res, bytes): | |
1204 | res = res.decode(preferredencoding()) | |
1205 | ||
1206 | assert isinstance(res, compat_str) | |
1207 | return res | |
c257baff PH |
1208 | |
1209 | ||
b58ddb32 PH |
1210 | def _windows_write_string(s, out): |
1211 | """ Returns True if the string was written using special methods, | |
1212 | False if it has yet to be written out.""" | |
1213 | # Adapted from http://stackoverflow.com/a/3259271/35070 | |
1214 | ||
1215 | import ctypes | |
1216 | import ctypes.wintypes | |
1217 | ||
1218 | WIN_OUTPUT_IDS = { | |
1219 | 1: -11, | |
1220 | 2: -12, | |
1221 | } | |
1222 | ||
a383a98a PH |
1223 | try: |
1224 | fileno = out.fileno() | |
1225 | except AttributeError: | |
1226 | # If the output stream doesn't have a fileno, it's virtual | |
1227 | return False | |
aa42e873 PH |
1228 | except io.UnsupportedOperation: |
1229 | # Some strange Windows pseudo files? | |
1230 | return False | |
b58ddb32 PH |
1231 | if fileno not in WIN_OUTPUT_IDS: |
1232 | return False | |
1233 | ||
e2f89ec7 | 1234 | GetStdHandle = ctypes.WINFUNCTYPE( |
b58ddb32 | 1235 | ctypes.wintypes.HANDLE, ctypes.wintypes.DWORD)( |
611c1dd9 | 1236 | (b'GetStdHandle', ctypes.windll.kernel32)) |
b58ddb32 PH |
1237 | h = GetStdHandle(WIN_OUTPUT_IDS[fileno]) |
1238 | ||
e2f89ec7 | 1239 | WriteConsoleW = ctypes.WINFUNCTYPE( |
b58ddb32 PH |
1240 | ctypes.wintypes.BOOL, ctypes.wintypes.HANDLE, ctypes.wintypes.LPWSTR, |
1241 | ctypes.wintypes.DWORD, ctypes.POINTER(ctypes.wintypes.DWORD), | |
611c1dd9 | 1242 | ctypes.wintypes.LPVOID)((b'WriteConsoleW', ctypes.windll.kernel32)) |
b58ddb32 PH |
1243 | written = ctypes.wintypes.DWORD(0) |
1244 | ||
611c1dd9 | 1245 | GetFileType = ctypes.WINFUNCTYPE(ctypes.wintypes.DWORD, ctypes.wintypes.DWORD)((b'GetFileType', ctypes.windll.kernel32)) |
b58ddb32 PH |
1246 | FILE_TYPE_CHAR = 0x0002 |
1247 | FILE_TYPE_REMOTE = 0x8000 | |
e2f89ec7 | 1248 | GetConsoleMode = ctypes.WINFUNCTYPE( |
b58ddb32 PH |
1249 | ctypes.wintypes.BOOL, ctypes.wintypes.HANDLE, |
1250 | ctypes.POINTER(ctypes.wintypes.DWORD))( | |
611c1dd9 | 1251 | (b'GetConsoleMode', ctypes.windll.kernel32)) |
b58ddb32 PH |
1252 | INVALID_HANDLE_VALUE = ctypes.wintypes.DWORD(-1).value |
1253 | ||
1254 | def not_a_console(handle): | |
1255 | if handle == INVALID_HANDLE_VALUE or handle is None: | |
1256 | return True | |
8fb3ac36 PH |
1257 | return ((GetFileType(handle) & ~FILE_TYPE_REMOTE) != FILE_TYPE_CHAR or |
1258 | GetConsoleMode(handle, ctypes.byref(ctypes.wintypes.DWORD())) == 0) | |
b58ddb32 PH |
1259 | |
1260 | if not_a_console(h): | |
1261 | return False | |
1262 | ||
d1b9c912 PH |
1263 | def next_nonbmp_pos(s): |
1264 | try: | |
1265 | return next(i for i, c in enumerate(s) if ord(c) > 0xffff) | |
1266 | except StopIteration: | |
1267 | return len(s) | |
1268 | ||
1269 | while s: | |
1270 | count = min(next_nonbmp_pos(s), 1024) | |
1271 | ||
b58ddb32 | 1272 | ret = WriteConsoleW( |
d1b9c912 | 1273 | h, s, count if count else 2, ctypes.byref(written), None) |
b58ddb32 PH |
1274 | if ret == 0: |
1275 | raise OSError('Failed to write string') | |
d1b9c912 PH |
1276 | if not count: # We just wrote a non-BMP character |
1277 | assert written.value == 2 | |
1278 | s = s[1:] | |
1279 | else: | |
1280 | assert written.value > 0 | |
1281 | s = s[written.value:] | |
b58ddb32 PH |
1282 | return True |
1283 | ||
1284 | ||
734f90bb | 1285 | def write_string(s, out=None, encoding=None): |
7459e3a2 PH |
1286 | if out is None: |
1287 | out = sys.stderr | |
8bf48f23 | 1288 | assert type(s) == compat_str |
7459e3a2 | 1289 | |
b58ddb32 PH |
1290 | if sys.platform == 'win32' and encoding is None and hasattr(out, 'fileno'): |
1291 | if _windows_write_string(s, out): | |
1292 | return | |
1293 | ||
7459e3a2 PH |
1294 | if ('b' in getattr(out, 'mode', '') or |
1295 | sys.version_info[0] < 3): # Python 2 lies about mode of sys.stderr | |
104aa738 PH |
1296 | byt = s.encode(encoding or preferredencoding(), 'ignore') |
1297 | out.write(byt) | |
1298 | elif hasattr(out, 'buffer'): | |
1299 | enc = encoding or getattr(out, 'encoding', None) or preferredencoding() | |
1300 | byt = s.encode(enc, 'ignore') | |
1301 | out.buffer.write(byt) | |
1302 | else: | |
8bf48f23 | 1303 | out.write(s) |
7459e3a2 PH |
1304 | out.flush() |
1305 | ||
1306 | ||
48ea9cea PH |
1307 | def bytes_to_intlist(bs): |
1308 | if not bs: | |
1309 | return [] | |
1310 | if isinstance(bs[0], int): # Python 3 | |
1311 | return list(bs) | |
1312 | else: | |
1313 | return [ord(c) for c in bs] | |
1314 | ||
c257baff | 1315 | |
cba892fa | 1316 | def intlist_to_bytes(xs): |
1317 | if not xs: | |
1318 | return b'' | |
edaa23f8 | 1319 | return compat_struct_pack('%dB' % len(xs), *xs) |
c38b1e77 PH |
1320 | |
1321 | ||
c1c9a79c PH |
1322 | # Cross-platform file locking |
1323 | if sys.platform == 'win32': | |
1324 | import ctypes.wintypes | |
1325 | import msvcrt | |
1326 | ||
1327 | class OVERLAPPED(ctypes.Structure): | |
1328 | _fields_ = [ | |
1329 | ('Internal', ctypes.wintypes.LPVOID), | |
1330 | ('InternalHigh', ctypes.wintypes.LPVOID), | |
1331 | ('Offset', ctypes.wintypes.DWORD), | |
1332 | ('OffsetHigh', ctypes.wintypes.DWORD), | |
1333 | ('hEvent', ctypes.wintypes.HANDLE), | |
1334 | ] | |
1335 | ||
1336 | kernel32 = ctypes.windll.kernel32 | |
1337 | LockFileEx = kernel32.LockFileEx | |
1338 | LockFileEx.argtypes = [ | |
1339 | ctypes.wintypes.HANDLE, # hFile | |
1340 | ctypes.wintypes.DWORD, # dwFlags | |
1341 | ctypes.wintypes.DWORD, # dwReserved | |
1342 | ctypes.wintypes.DWORD, # nNumberOfBytesToLockLow | |
1343 | ctypes.wintypes.DWORD, # nNumberOfBytesToLockHigh | |
1344 | ctypes.POINTER(OVERLAPPED) # Overlapped | |
1345 | ] | |
1346 | LockFileEx.restype = ctypes.wintypes.BOOL | |
1347 | UnlockFileEx = kernel32.UnlockFileEx | |
1348 | UnlockFileEx.argtypes = [ | |
1349 | ctypes.wintypes.HANDLE, # hFile | |
1350 | ctypes.wintypes.DWORD, # dwReserved | |
1351 | ctypes.wintypes.DWORD, # nNumberOfBytesToLockLow | |
1352 | ctypes.wintypes.DWORD, # nNumberOfBytesToLockHigh | |
1353 | ctypes.POINTER(OVERLAPPED) # Overlapped | |
1354 | ] | |
1355 | UnlockFileEx.restype = ctypes.wintypes.BOOL | |
1356 | whole_low = 0xffffffff | |
1357 | whole_high = 0x7fffffff | |
1358 | ||
1359 | def _lock_file(f, exclusive): | |
1360 | overlapped = OVERLAPPED() | |
1361 | overlapped.Offset = 0 | |
1362 | overlapped.OffsetHigh = 0 | |
1363 | overlapped.hEvent = 0 | |
1364 | f._lock_file_overlapped_p = ctypes.pointer(overlapped) | |
1365 | handle = msvcrt.get_osfhandle(f.fileno()) | |
1366 | if not LockFileEx(handle, 0x2 if exclusive else 0x0, 0, | |
1367 | whole_low, whole_high, f._lock_file_overlapped_p): | |
1368 | raise OSError('Locking file failed: %r' % ctypes.FormatError()) | |
1369 | ||
1370 | def _unlock_file(f): | |
1371 | assert f._lock_file_overlapped_p | |
1372 | handle = msvcrt.get_osfhandle(f.fileno()) | |
1373 | if not UnlockFileEx(handle, 0, | |
1374 | whole_low, whole_high, f._lock_file_overlapped_p): | |
1375 | raise OSError('Unlocking file failed: %r' % ctypes.FormatError()) | |
1376 | ||
1377 | else: | |
399a76e6 YCH |
1378 | # Some platforms, such as Jython, is missing fcntl |
1379 | try: | |
1380 | import fcntl | |
c1c9a79c | 1381 | |
399a76e6 YCH |
1382 | def _lock_file(f, exclusive): |
1383 | fcntl.flock(f, fcntl.LOCK_EX if exclusive else fcntl.LOCK_SH) | |
c1c9a79c | 1384 | |
399a76e6 YCH |
1385 | def _unlock_file(f): |
1386 | fcntl.flock(f, fcntl.LOCK_UN) | |
1387 | except ImportError: | |
1388 | UNSUPPORTED_MSG = 'file locking is not supported on this platform' | |
1389 | ||
1390 | def _lock_file(f, exclusive): | |
1391 | raise IOError(UNSUPPORTED_MSG) | |
1392 | ||
1393 | def _unlock_file(f): | |
1394 | raise IOError(UNSUPPORTED_MSG) | |
c1c9a79c PH |
1395 | |
1396 | ||
1397 | class locked_file(object): | |
1398 | def __init__(self, filename, mode, encoding=None): | |
1399 | assert mode in ['r', 'a', 'w'] | |
1400 | self.f = io.open(filename, mode, encoding=encoding) | |
1401 | self.mode = mode | |
1402 | ||
1403 | def __enter__(self): | |
1404 | exclusive = self.mode != 'r' | |
1405 | try: | |
1406 | _lock_file(self.f, exclusive) | |
1407 | except IOError: | |
1408 | self.f.close() | |
1409 | raise | |
1410 | return self | |
1411 | ||
1412 | def __exit__(self, etype, value, traceback): | |
1413 | try: | |
1414 | _unlock_file(self.f) | |
1415 | finally: | |
1416 | self.f.close() | |
1417 | ||
1418 | def __iter__(self): | |
1419 | return iter(self.f) | |
1420 | ||
1421 | def write(self, *args): | |
1422 | return self.f.write(*args) | |
1423 | ||
1424 | def read(self, *args): | |
1425 | return self.f.read(*args) | |
4eb7f1d1 JMF |
1426 | |
1427 | ||
4644ac55 S |
1428 | def get_filesystem_encoding(): |
1429 | encoding = sys.getfilesystemencoding() | |
1430 | return encoding if encoding is not None else 'utf-8' | |
1431 | ||
1432 | ||
4eb7f1d1 | 1433 | def shell_quote(args): |
a6a173c2 | 1434 | quoted_args = [] |
4644ac55 | 1435 | encoding = get_filesystem_encoding() |
a6a173c2 JMF |
1436 | for a in args: |
1437 | if isinstance(a, bytes): | |
1438 | # We may get a filename encoded with 'encodeFilename' | |
1439 | a = a.decode(encoding) | |
1440 | quoted_args.append(pipes.quote(a)) | |
28e614de | 1441 | return ' '.join(quoted_args) |
9d4660ca PH |
1442 | |
1443 | ||
1444 | def smuggle_url(url, data): | |
1445 | """ Pass additional data in a URL for internal use. """ | |
1446 | ||
15707c7e | 1447 | sdata = compat_urllib_parse_urlencode( |
28e614de PH |
1448 | {'__youtubedl_smuggle': json.dumps(data)}) |
1449 | return url + '#' + sdata | |
9d4660ca PH |
1450 | |
1451 | ||
79f82953 | 1452 | def unsmuggle_url(smug_url, default=None): |
83e865a3 | 1453 | if '#__youtubedl_smuggle' not in smug_url: |
79f82953 | 1454 | return smug_url, default |
28e614de PH |
1455 | url, _, sdata = smug_url.rpartition('#') |
1456 | jsond = compat_parse_qs(sdata)['__youtubedl_smuggle'][0] | |
9d4660ca PH |
1457 | data = json.loads(jsond) |
1458 | return url, data | |
02dbf93f PH |
1459 | |
1460 | ||
02dbf93f PH |
1461 | def format_bytes(bytes): |
1462 | if bytes is None: | |
28e614de | 1463 | return 'N/A' |
02dbf93f PH |
1464 | if type(bytes) is str: |
1465 | bytes = float(bytes) | |
1466 | if bytes == 0.0: | |
1467 | exponent = 0 | |
1468 | else: | |
1469 | exponent = int(math.log(bytes, 1024.0)) | |
28e614de | 1470 | suffix = ['B', 'KiB', 'MiB', 'GiB', 'TiB', 'PiB', 'EiB', 'ZiB', 'YiB'][exponent] |
02dbf93f | 1471 | converted = float(bytes) / float(1024 ** exponent) |
28e614de | 1472 | return '%.2f%s' % (converted, suffix) |
f53c966a | 1473 | |
1c088fa8 | 1474 | |
fb47597b S |
1475 | def lookup_unit_table(unit_table, s): |
1476 | units_re = '|'.join(re.escape(u) for u in unit_table) | |
1477 | m = re.match( | |
782b1b5b | 1478 | r'(?P<num>[0-9]+(?:[,.][0-9]*)?)\s*(?P<unit>%s)\b' % units_re, s) |
fb47597b S |
1479 | if not m: |
1480 | return None | |
1481 | num_str = m.group('num').replace(',', '.') | |
1482 | mult = unit_table[m.group('unit')] | |
1483 | return int(float(num_str) * mult) | |
1484 | ||
1485 | ||
be64b5b0 PH |
1486 | def parse_filesize(s): |
1487 | if s is None: | |
1488 | return None | |
1489 | ||
dfb1b146 | 1490 | # The lower-case forms are of course incorrect and unofficial, |
be64b5b0 PH |
1491 | # but we support those too |
1492 | _UNIT_TABLE = { | |
1493 | 'B': 1, | |
1494 | 'b': 1, | |
1495 | 'KiB': 1024, | |
1496 | 'KB': 1000, | |
1497 | 'kB': 1024, | |
1498 | 'Kb': 1000, | |
1499 | 'MiB': 1024 ** 2, | |
1500 | 'MB': 1000 ** 2, | |
1501 | 'mB': 1024 ** 2, | |
1502 | 'Mb': 1000 ** 2, | |
1503 | 'GiB': 1024 ** 3, | |
1504 | 'GB': 1000 ** 3, | |
1505 | 'gB': 1024 ** 3, | |
1506 | 'Gb': 1000 ** 3, | |
1507 | 'TiB': 1024 ** 4, | |
1508 | 'TB': 1000 ** 4, | |
1509 | 'tB': 1024 ** 4, | |
1510 | 'Tb': 1000 ** 4, | |
1511 | 'PiB': 1024 ** 5, | |
1512 | 'PB': 1000 ** 5, | |
1513 | 'pB': 1024 ** 5, | |
1514 | 'Pb': 1000 ** 5, | |
1515 | 'EiB': 1024 ** 6, | |
1516 | 'EB': 1000 ** 6, | |
1517 | 'eB': 1024 ** 6, | |
1518 | 'Eb': 1000 ** 6, | |
1519 | 'ZiB': 1024 ** 7, | |
1520 | 'ZB': 1000 ** 7, | |
1521 | 'zB': 1024 ** 7, | |
1522 | 'Zb': 1000 ** 7, | |
1523 | 'YiB': 1024 ** 8, | |
1524 | 'YB': 1000 ** 8, | |
1525 | 'yB': 1024 ** 8, | |
1526 | 'Yb': 1000 ** 8, | |
1527 | } | |
1528 | ||
fb47597b S |
1529 | return lookup_unit_table(_UNIT_TABLE, s) |
1530 | ||
1531 | ||
1532 | def parse_count(s): | |
1533 | if s is None: | |
be64b5b0 PH |
1534 | return None |
1535 | ||
fb47597b S |
1536 | s = s.strip() |
1537 | ||
1538 | if re.match(r'^[\d,.]+$', s): | |
1539 | return str_to_int(s) | |
1540 | ||
1541 | _UNIT_TABLE = { | |
1542 | 'k': 1000, | |
1543 | 'K': 1000, | |
1544 | 'm': 1000 ** 2, | |
1545 | 'M': 1000 ** 2, | |
1546 | 'kk': 1000 ** 2, | |
1547 | 'KK': 1000 ** 2, | |
1548 | } | |
be64b5b0 | 1549 | |
fb47597b | 1550 | return lookup_unit_table(_UNIT_TABLE, s) |
be64b5b0 | 1551 | |
2f7ae819 | 1552 | |
caefb1de PH |
1553 | def month_by_name(name): |
1554 | """ Return the number of a month by (locale-independently) English name """ | |
1555 | ||
caefb1de | 1556 | try: |
7105440c YCH |
1557 | return ENGLISH_MONTH_NAMES.index(name) + 1 |
1558 | except ValueError: | |
1559 | return None | |
1560 | ||
1561 | ||
1562 | def month_by_abbreviation(abbrev): | |
1563 | """ Return the number of a month by (locale-independently) English | |
1564 | abbreviations """ | |
1565 | ||
1566 | try: | |
1567 | return [s[:3] for s in ENGLISH_MONTH_NAMES].index(abbrev) + 1 | |
caefb1de PH |
1568 | except ValueError: |
1569 | return None | |
18258362 JMF |
1570 | |
1571 | ||
5aafe895 | 1572 | def fix_xml_ampersands(xml_str): |
18258362 | 1573 | """Replace all the '&' by '&' in XML""" |
5aafe895 PH |
1574 | return re.sub( |
1575 | r'&(?!amp;|lt;|gt;|apos;|quot;|#x[0-9a-fA-F]{,4};|#[0-9]{,4};)', | |
28e614de | 1576 | '&', |
5aafe895 | 1577 | xml_str) |
e3946f98 PH |
1578 | |
1579 | ||
1580 | def setproctitle(title): | |
8bf48f23 | 1581 | assert isinstance(title, compat_str) |
c1c05c67 YCH |
1582 | |
1583 | # ctypes in Jython is not complete | |
1584 | # http://bugs.jython.org/issue2148 | |
1585 | if sys.platform.startswith('java'): | |
1586 | return | |
1587 | ||
e3946f98 | 1588 | try: |
611c1dd9 | 1589 | libc = ctypes.cdll.LoadLibrary('libc.so.6') |
e3946f98 PH |
1590 | except OSError: |
1591 | return | |
6eefe533 PH |
1592 | title_bytes = title.encode('utf-8') |
1593 | buf = ctypes.create_string_buffer(len(title_bytes)) | |
1594 | buf.value = title_bytes | |
e3946f98 | 1595 | try: |
6eefe533 | 1596 | libc.prctl(15, buf, 0, 0, 0) |
e3946f98 PH |
1597 | except AttributeError: |
1598 | return # Strange libc, just skip this | |
d7dda168 PH |
1599 | |
1600 | ||
1601 | def remove_start(s, start): | |
46bc9b7d | 1602 | return s[len(start):] if s is not None and s.startswith(start) else s |
29eb5174 PH |
1603 | |
1604 | ||
2b9faf55 | 1605 | def remove_end(s, end): |
46bc9b7d | 1606 | return s[:-len(end)] if s is not None and s.endswith(end) else s |
2b9faf55 PH |
1607 | |
1608 | ||
31b2051e S |
1609 | def remove_quotes(s): |
1610 | if s is None or len(s) < 2: | |
1611 | return s | |
1612 | for quote in ('"', "'", ): | |
1613 | if s[0] == quote and s[-1] == quote: | |
1614 | return s[1:-1] | |
1615 | return s | |
1616 | ||
1617 | ||
29eb5174 | 1618 | def url_basename(url): |
9b8aaeed | 1619 | path = compat_urlparse.urlparse(url).path |
28e614de | 1620 | return path.strip('/').split('/')[-1] |
aa94a6d3 PH |
1621 | |
1622 | ||
1623 | class HEADRequest(compat_urllib_request.Request): | |
1624 | def get_method(self): | |
611c1dd9 | 1625 | return 'HEAD' |
7217e148 PH |
1626 | |
1627 | ||
9732d77e | 1628 | def int_or_none(v, scale=1, default=None, get_attr=None, invscale=1): |
28746fbd PH |
1629 | if get_attr: |
1630 | if v is not None: | |
1631 | v = getattr(v, get_attr, None) | |
9572013d PH |
1632 | if v == '': |
1633 | v = None | |
1812afb7 S |
1634 | if v is None: |
1635 | return default | |
1636 | try: | |
1637 | return int(v) * invscale // scale | |
1638 | except ValueError: | |
af98f8ff | 1639 | return default |
9732d77e | 1640 | |
9572013d | 1641 | |
40a90862 JMF |
1642 | def str_or_none(v, default=None): |
1643 | return default if v is None else compat_str(v) | |
1644 | ||
9732d77e PH |
1645 | |
1646 | def str_to_int(int_str): | |
48d4681e | 1647 | """ A more relaxed version of int_or_none """ |
9732d77e PH |
1648 | if int_str is None: |
1649 | return None | |
28e614de | 1650 | int_str = re.sub(r'[,\.\+]', '', int_str) |
9732d77e | 1651 | return int(int_str) |
608d11f5 PH |
1652 | |
1653 | ||
9732d77e | 1654 | def float_or_none(v, scale=1, invscale=1, default=None): |
caf80631 S |
1655 | if v is None: |
1656 | return default | |
1657 | try: | |
1658 | return float(v) * invscale / scale | |
1659 | except ValueError: | |
1660 | return default | |
43f775e4 PH |
1661 | |
1662 | ||
608d11f5 | 1663 | def parse_duration(s): |
8f9312c3 | 1664 | if not isinstance(s, compat_basestring): |
608d11f5 PH |
1665 | return None |
1666 | ||
ca7b3246 S |
1667 | s = s.strip() |
1668 | ||
acaff495 | 1669 | days, hours, mins, secs, ms = [None] * 5 |
1670 | m = re.match(r'(?:(?:(?:(?P<days>[0-9]+):)?(?P<hours>[0-9]+):)?(?P<mins>[0-9]+):)?(?P<secs>[0-9]+)(?P<ms>\.[0-9]+)?$', s) | |
1671 | if m: | |
1672 | days, hours, mins, secs, ms = m.groups() | |
1673 | else: | |
1674 | m = re.match( | |
1675 | r'''(?ix)(?:P?T)? | |
8f4b58d7 | 1676 | (?: |
acaff495 | 1677 | (?P<days>[0-9]+)\s*d(?:ays?)?\s* |
8f4b58d7 | 1678 | )? |
acaff495 | 1679 | (?: |
1680 | (?P<hours>[0-9]+)\s*h(?:ours?)?\s* | |
1681 | )? | |
1682 | (?: | |
1683 | (?P<mins>[0-9]+)\s*m(?:in(?:ute)?s?)?\s* | |
1684 | )? | |
1685 | (?: | |
1686 | (?P<secs>[0-9]+)(?P<ms>\.[0-9]+)?\s*s(?:ec(?:ond)?s?)?\s* | |
1687 | )?$''', s) | |
1688 | if m: | |
1689 | days, hours, mins, secs, ms = m.groups() | |
1690 | else: | |
1691 | m = re.match(r'(?i)(?:(?P<hours>[0-9.]+)\s*(?:hours?)|(?P<mins>[0-9.]+)\s*(?:mins?\.?|minutes?)\s*)$', s) | |
1692 | if m: | |
1693 | hours, mins = m.groups() | |
1694 | else: | |
1695 | return None | |
1696 | ||
1697 | duration = 0 | |
1698 | if secs: | |
1699 | duration += float(secs) | |
1700 | if mins: | |
1701 | duration += float(mins) * 60 | |
1702 | if hours: | |
1703 | duration += float(hours) * 60 * 60 | |
1704 | if days: | |
1705 | duration += float(days) * 24 * 60 * 60 | |
1706 | if ms: | |
1707 | duration += float(ms) | |
1708 | return duration | |
91d7d0b3 JMF |
1709 | |
1710 | ||
e65e4c88 | 1711 | def prepend_extension(filename, ext, expected_real_ext=None): |
5f6a1245 | 1712 | name, real_ext = os.path.splitext(filename) |
e65e4c88 S |
1713 | return ( |
1714 | '{0}.{1}{2}'.format(name, ext, real_ext) | |
1715 | if not expected_real_ext or real_ext[1:] == expected_real_ext | |
1716 | else '{0}.{1}'.format(filename, ext)) | |
d70ad093 PH |
1717 | |
1718 | ||
b3ed15b7 S |
1719 | def replace_extension(filename, ext, expected_real_ext=None): |
1720 | name, real_ext = os.path.splitext(filename) | |
1721 | return '{0}.{1}'.format( | |
1722 | name if not expected_real_ext or real_ext[1:] == expected_real_ext else filename, | |
1723 | ext) | |
1724 | ||
1725 | ||
d70ad093 PH |
1726 | def check_executable(exe, args=[]): |
1727 | """ Checks if the given binary is installed somewhere in PATH, and returns its name. | |
1728 | args can be a list of arguments for a short output (like -version) """ | |
1729 | try: | |
1730 | subprocess.Popen([exe] + args, stdout=subprocess.PIPE, stderr=subprocess.PIPE).communicate() | |
1731 | except OSError: | |
1732 | return False | |
1733 | return exe | |
b7ab0590 PH |
1734 | |
1735 | ||
95807118 | 1736 | def get_exe_version(exe, args=['--version'], |
cae97f65 | 1737 | version_re=None, unrecognized='present'): |
95807118 PH |
1738 | """ Returns the version of the specified executable, |
1739 | or False if the executable is not present """ | |
1740 | try: | |
cae97f65 | 1741 | out, _ = subprocess.Popen( |
54116803 | 1742 | [encodeArgument(exe)] + args, |
95807118 PH |
1743 | stdout=subprocess.PIPE, stderr=subprocess.STDOUT).communicate() |
1744 | except OSError: | |
1745 | return False | |
cae97f65 PH |
1746 | if isinstance(out, bytes): # Python 2.x |
1747 | out = out.decode('ascii', 'ignore') | |
1748 | return detect_exe_version(out, version_re, unrecognized) | |
1749 | ||
1750 | ||
1751 | def detect_exe_version(output, version_re=None, unrecognized='present'): | |
1752 | assert isinstance(output, compat_str) | |
1753 | if version_re is None: | |
1754 | version_re = r'version\s+([-0-9._a-zA-Z]+)' | |
1755 | m = re.search(version_re, output) | |
95807118 PH |
1756 | if m: |
1757 | return m.group(1) | |
1758 | else: | |
1759 | return unrecognized | |
1760 | ||
1761 | ||
b7ab0590 | 1762 | class PagedList(object): |
dd26ced1 PH |
1763 | def __len__(self): |
1764 | # This is only useful for tests | |
1765 | return len(self.getslice()) | |
1766 | ||
9c44d242 PH |
1767 | |
1768 | class OnDemandPagedList(PagedList): | |
b95dc034 | 1769 | def __init__(self, pagefunc, pagesize, use_cache=False): |
9c44d242 PH |
1770 | self._pagefunc = pagefunc |
1771 | self._pagesize = pagesize | |
b95dc034 YCH |
1772 | self._use_cache = use_cache |
1773 | if use_cache: | |
1774 | self._cache = {} | |
9c44d242 | 1775 | |
b7ab0590 PH |
1776 | def getslice(self, start=0, end=None): |
1777 | res = [] | |
1778 | for pagenum in itertools.count(start // self._pagesize): | |
1779 | firstid = pagenum * self._pagesize | |
1780 | nextfirstid = pagenum * self._pagesize + self._pagesize | |
1781 | if start >= nextfirstid: | |
1782 | continue | |
1783 | ||
b95dc034 YCH |
1784 | page_results = None |
1785 | if self._use_cache: | |
1786 | page_results = self._cache.get(pagenum) | |
1787 | if page_results is None: | |
1788 | page_results = list(self._pagefunc(pagenum)) | |
1789 | if self._use_cache: | |
1790 | self._cache[pagenum] = page_results | |
b7ab0590 PH |
1791 | |
1792 | startv = ( | |
1793 | start % self._pagesize | |
1794 | if firstid <= start < nextfirstid | |
1795 | else 0) | |
1796 | ||
1797 | endv = ( | |
1798 | ((end - 1) % self._pagesize) + 1 | |
1799 | if (end is not None and firstid <= end <= nextfirstid) | |
1800 | else None) | |
1801 | ||
1802 | if startv != 0 or endv is not None: | |
1803 | page_results = page_results[startv:endv] | |
1804 | res.extend(page_results) | |
1805 | ||
1806 | # A little optimization - if current page is not "full", ie. does | |
1807 | # not contain page_size videos then we can assume that this page | |
1808 | # is the last one - there are no more ids on further pages - | |
1809 | # i.e. no need to query again. | |
1810 | if len(page_results) + startv < self._pagesize: | |
1811 | break | |
1812 | ||
1813 | # If we got the whole page, but the next page is not interesting, | |
1814 | # break out early as well | |
1815 | if end == nextfirstid: | |
1816 | break | |
1817 | return res | |
81c2f20b PH |
1818 | |
1819 | ||
9c44d242 PH |
1820 | class InAdvancePagedList(PagedList): |
1821 | def __init__(self, pagefunc, pagecount, pagesize): | |
1822 | self._pagefunc = pagefunc | |
1823 | self._pagecount = pagecount | |
1824 | self._pagesize = pagesize | |
1825 | ||
1826 | def getslice(self, start=0, end=None): | |
1827 | res = [] | |
1828 | start_page = start // self._pagesize | |
1829 | end_page = ( | |
1830 | self._pagecount if end is None else (end // self._pagesize + 1)) | |
1831 | skip_elems = start - start_page * self._pagesize | |
1832 | only_more = None if end is None else end - start | |
1833 | for pagenum in range(start_page, end_page): | |
1834 | page = list(self._pagefunc(pagenum)) | |
1835 | if skip_elems: | |
1836 | page = page[skip_elems:] | |
1837 | skip_elems = None | |
1838 | if only_more is not None: | |
1839 | if len(page) < only_more: | |
1840 | only_more -= len(page) | |
1841 | else: | |
1842 | page = page[:only_more] | |
1843 | res.extend(page) | |
1844 | break | |
1845 | res.extend(page) | |
1846 | return res | |
1847 | ||
1848 | ||
81c2f20b | 1849 | def uppercase_escape(s): |
676eb3f2 | 1850 | unicode_escape = codecs.getdecoder('unicode_escape') |
81c2f20b | 1851 | return re.sub( |
a612753d | 1852 | r'\\U[0-9a-fA-F]{8}', |
676eb3f2 PH |
1853 | lambda m: unicode_escape(m.group(0))[0], |
1854 | s) | |
0fe2ff78 YCH |
1855 | |
1856 | ||
1857 | def lowercase_escape(s): | |
1858 | unicode_escape = codecs.getdecoder('unicode_escape') | |
1859 | return re.sub( | |
1860 | r'\\u[0-9a-fA-F]{4}', | |
1861 | lambda m: unicode_escape(m.group(0))[0], | |
1862 | s) | |
b53466e1 | 1863 | |
d05cfe06 S |
1864 | |
1865 | def escape_rfc3986(s): | |
1866 | """Escape non-ASCII characters as suggested by RFC 3986""" | |
8f9312c3 | 1867 | if sys.version_info < (3, 0) and isinstance(s, compat_str): |
d05cfe06 | 1868 | s = s.encode('utf-8') |
ecc0c5ee | 1869 | return compat_urllib_parse.quote(s, b"%/;:@&=+$,!~*'()?#[]") |
d05cfe06 S |
1870 | |
1871 | ||
1872 | def escape_url(url): | |
1873 | """Escape URL as suggested by RFC 3986""" | |
1874 | url_parsed = compat_urllib_parse_urlparse(url) | |
1875 | return url_parsed._replace( | |
efbed08d | 1876 | netloc=url_parsed.netloc.encode('idna').decode('ascii'), |
d05cfe06 S |
1877 | path=escape_rfc3986(url_parsed.path), |
1878 | params=escape_rfc3986(url_parsed.params), | |
1879 | query=escape_rfc3986(url_parsed.query), | |
1880 | fragment=escape_rfc3986(url_parsed.fragment) | |
1881 | ).geturl() | |
1882 | ||
62e609ab PH |
1883 | |
1884 | def read_batch_urls(batch_fd): | |
1885 | def fixup(url): | |
1886 | if not isinstance(url, compat_str): | |
1887 | url = url.decode('utf-8', 'replace') | |
28e614de | 1888 | BOM_UTF8 = '\xef\xbb\xbf' |
62e609ab PH |
1889 | if url.startswith(BOM_UTF8): |
1890 | url = url[len(BOM_UTF8):] | |
1891 | url = url.strip() | |
1892 | if url.startswith(('#', ';', ']')): | |
1893 | return False | |
1894 | return url | |
1895 | ||
1896 | with contextlib.closing(batch_fd) as fd: | |
1897 | return [url for url in map(fixup, fd) if url] | |
b74fa8cd JMF |
1898 | |
1899 | ||
1900 | def urlencode_postdata(*args, **kargs): | |
15707c7e | 1901 | return compat_urllib_parse_urlencode(*args, **kargs).encode('ascii') |
bcf89ce6 PH |
1902 | |
1903 | ||
38f9ef31 | 1904 | def update_url_query(url, query): |
cacd9966 YCH |
1905 | if not query: |
1906 | return url | |
38f9ef31 | 1907 | parsed_url = compat_urlparse.urlparse(url) |
1908 | qs = compat_parse_qs(parsed_url.query) | |
1909 | qs.update(query) | |
1910 | return compat_urlparse.urlunparse(parsed_url._replace( | |
15707c7e | 1911 | query=compat_urllib_parse_urlencode(qs, True))) |
16392824 | 1912 | |
8e60dc75 | 1913 | |
ed0291d1 S |
1914 | def update_Request(req, url=None, data=None, headers={}, query={}): |
1915 | req_headers = req.headers.copy() | |
1916 | req_headers.update(headers) | |
1917 | req_data = data or req.data | |
1918 | req_url = update_url_query(url or req.get_full_url(), query) | |
1919 | req_type = HEADRequest if req.get_method() == 'HEAD' else compat_urllib_request.Request | |
1920 | new_req = req_type( | |
1921 | req_url, data=req_data, headers=req_headers, | |
1922 | origin_req_host=req.origin_req_host, unverifiable=req.unverifiable) | |
1923 | if hasattr(req, 'timeout'): | |
1924 | new_req.timeout = req.timeout | |
1925 | return new_req | |
1926 | ||
1927 | ||
86296ad2 | 1928 | def dict_get(d, key_or_keys, default=None, skip_false_values=True): |
cbecc9b9 S |
1929 | if isinstance(key_or_keys, (list, tuple)): |
1930 | for key in key_or_keys: | |
86296ad2 S |
1931 | if key not in d or d[key] is None or skip_false_values and not d[key]: |
1932 | continue | |
1933 | return d[key] | |
cbecc9b9 S |
1934 | return default |
1935 | return d.get(key_or_keys, default) | |
1936 | ||
1937 | ||
329ca3be S |
1938 | def try_get(src, getter, expected_type=None): |
1939 | try: | |
1940 | v = getter(src) | |
1941 | except (AttributeError, KeyError, TypeError, IndexError): | |
1942 | pass | |
1943 | else: | |
1944 | if expected_type is None or isinstance(v, expected_type): | |
1945 | return v | |
1946 | ||
1947 | ||
8e60dc75 S |
1948 | def encode_compat_str(string, encoding=preferredencoding(), errors='strict'): |
1949 | return string if isinstance(string, compat_str) else compat_str(string, encoding, errors) | |
1950 | ||
16392824 | 1951 | |
a1a530b0 PH |
1952 | US_RATINGS = { |
1953 | 'G': 0, | |
1954 | 'PG': 10, | |
1955 | 'PG-13': 13, | |
1956 | 'R': 16, | |
1957 | 'NC': 18, | |
1958 | } | |
fac55558 PH |
1959 | |
1960 | ||
146c80e2 S |
1961 | def parse_age_limit(s): |
1962 | if s is None: | |
d838b1bd | 1963 | return None |
146c80e2 | 1964 | m = re.match(r'^(?P<age>\d{1,2})\+?$', s) |
d800609c | 1965 | return int(m.group('age')) if m else US_RATINGS.get(s) |
146c80e2 S |
1966 | |
1967 | ||
fac55558 | 1968 | def strip_jsonp(code): |
609a61e3 | 1969 | return re.sub( |
5950cb1d | 1970 | r'(?s)^[a-zA-Z0-9_.$]+\s*\(\s*(.*)\);?\s*?(?://[^\n]*)*$', r'\1', code) |
478c2c61 PH |
1971 | |
1972 | ||
e05f6939 PH |
1973 | def js_to_json(code): |
1974 | def fix_kv(m): | |
e7b6d122 PH |
1975 | v = m.group(0) |
1976 | if v in ('true', 'false', 'null'): | |
1977 | return v | |
bd1e4844 | 1978 | elif v.startswith('/*') or v == ',': |
1979 | return "" | |
1980 | ||
1981 | if v[0] in ("'", '"'): | |
1982 | v = re.sub(r'(?s)\\.|"', lambda m: { | |
e7b6d122 | 1983 | '"': '\\"', |
bd1e4844 | 1984 | "\\'": "'", |
1985 | '\\\n': '', | |
1986 | '\\x': '\\u00', | |
1987 | }.get(m.group(0), m.group(0)), v[1:-1]) | |
1988 | ||
89ac4a19 | 1989 | INTEGER_TABLE = ( |
cda6d47a S |
1990 | (r'^0[xX][0-9a-fA-F]+', 16), |
1991 | (r'^0+[0-7]+', 8), | |
89ac4a19 S |
1992 | ) |
1993 | ||
1994 | for regex, base in INTEGER_TABLE: | |
1995 | im = re.match(regex, v) | |
1996 | if im: | |
cda6d47a | 1997 | i = int(im.group(0), base) |
89ac4a19 S |
1998 | return '"%d":' % i if v.endswith(':') else '%d' % i |
1999 | ||
e7b6d122 | 2000 | return '"%s"' % v |
e05f6939 | 2001 | |
bd1e4844 | 2002 | return re.sub(r'''(?sx) |
2003 | "(?:[^"\\]*(?:\\\\|\\['"nurtbfx/\n]))*[^"\\]*"| | |
2004 | '(?:[^'\\]*(?:\\\\|\\['"nurtbfx/\n]))*[^'\\]*'| | |
2005 | /\*.*?\*/|,(?=\s*[\]}])| | |
2006 | [a-zA-Z_][.a-zA-Z_0-9]*| | |
47212f7b | 2007 | \b(?:0[xX][0-9a-fA-F]+|0+[0-7]+)(?:\s*:)?| |
bd1e4844 | 2008 | [0-9]+(?=\s*:) |
e05f6939 | 2009 | ''', fix_kv, code) |
e05f6939 PH |
2010 | |
2011 | ||
478c2c61 PH |
2012 | def qualities(quality_ids): |
2013 | """ Get a numeric quality value out of a list of possible values """ | |
2014 | def q(qid): | |
2015 | try: | |
2016 | return quality_ids.index(qid) | |
2017 | except ValueError: | |
2018 | return -1 | |
2019 | return q | |
2020 | ||
acd69589 PH |
2021 | |
2022 | DEFAULT_OUTTMPL = '%(title)s-%(id)s.%(ext)s' | |
0a871f68 | 2023 | |
a020a0dc PH |
2024 | |
2025 | def limit_length(s, length): | |
2026 | """ Add ellipses to overly long strings """ | |
2027 | if s is None: | |
2028 | return None | |
2029 | ELLIPSES = '...' | |
2030 | if len(s) > length: | |
2031 | return s[:length - len(ELLIPSES)] + ELLIPSES | |
2032 | return s | |
48844745 PH |
2033 | |
2034 | ||
2035 | def version_tuple(v): | |
5f9b8394 | 2036 | return tuple(int(e) for e in re.split(r'[-.]', v)) |
48844745 PH |
2037 | |
2038 | ||
2039 | def is_outdated_version(version, limit, assume_new=True): | |
2040 | if not version: | |
2041 | return not assume_new | |
2042 | try: | |
2043 | return version_tuple(version) < version_tuple(limit) | |
2044 | except ValueError: | |
2045 | return not assume_new | |
732ea2f0 PH |
2046 | |
2047 | ||
2048 | def ytdl_is_updateable(): | |
2049 | """ Returns if youtube-dl can be updated with -U """ | |
2050 | from zipimport import zipimporter | |
2051 | ||
2052 | return isinstance(globals().get('__loader__'), zipimporter) or hasattr(sys, 'frozen') | |
7d4111ed PH |
2053 | |
2054 | ||
2055 | def args_to_str(args): | |
2056 | # Get a short string representation for a subprocess command | |
702ccf2d | 2057 | return ' '.join(compat_shlex_quote(a) for a in args) |
2ccd1b10 PH |
2058 | |
2059 | ||
9b9c5355 | 2060 | def error_to_compat_str(err): |
fdae2358 S |
2061 | err_str = str(err) |
2062 | # On python 2 error byte string must be decoded with proper | |
2063 | # encoding rather than ascii | |
2064 | if sys.version_info[0] < 3: | |
2065 | err_str = err_str.decode(preferredencoding()) | |
2066 | return err_str | |
2067 | ||
2068 | ||
c460bdd5 | 2069 | def mimetype2ext(mt): |
eb9ee194 S |
2070 | if mt is None: |
2071 | return None | |
2072 | ||
765ac263 JMF |
2073 | ext = { |
2074 | 'audio/mp4': 'm4a', | |
6c33d24b YCH |
2075 | # Per RFC 3003, audio/mpeg can be .mp1, .mp2 or .mp3. Here use .mp3 as |
2076 | # it's the most popular one | |
2077 | 'audio/mpeg': 'mp3', | |
765ac263 JMF |
2078 | }.get(mt) |
2079 | if ext is not None: | |
2080 | return ext | |
2081 | ||
c460bdd5 PH |
2082 | _, _, res = mt.rpartition('/') |
2083 | ||
2084 | return { | |
f6861ec9 | 2085 | '3gpp': '3gp', |
cafcf657 | 2086 | 'smptett+xml': 'tt', |
2087 | 'srt': 'srt', | |
2088 | 'ttaf+xml': 'dfxp', | |
a0d8d704 | 2089 | 'ttml+xml': 'ttml', |
cafcf657 | 2090 | 'vtt': 'vtt', |
f6861ec9 | 2091 | 'x-flv': 'flv', |
a0d8d704 YCH |
2092 | 'x-mp4-fragmented': 'mp4', |
2093 | 'x-ms-wmv': 'wmv', | |
c460bdd5 PH |
2094 | }.get(res, res) |
2095 | ||
2096 | ||
2ccd1b10 | 2097 | def urlhandle_detect_ext(url_handle): |
79298173 | 2098 | getheader = url_handle.headers.get |
2ccd1b10 | 2099 | |
b55ee18f PH |
2100 | cd = getheader('Content-Disposition') |
2101 | if cd: | |
2102 | m = re.match(r'attachment;\s*filename="(?P<filename>[^"]+)"', cd) | |
2103 | if m: | |
2104 | e = determine_ext(m.group('filename'), default_ext=None) | |
2105 | if e: | |
2106 | return e | |
2107 | ||
c460bdd5 | 2108 | return mimetype2ext(getheader('Content-Type')) |
05900629 PH |
2109 | |
2110 | ||
1e399778 YCH |
2111 | def encode_data_uri(data, mime_type): |
2112 | return 'data:%s;base64,%s' % (mime_type, base64.b64encode(data).decode('ascii')) | |
2113 | ||
2114 | ||
05900629 | 2115 | def age_restricted(content_limit, age_limit): |
6ec6cb4e | 2116 | """ Returns True iff the content should be blocked """ |
05900629 PH |
2117 | |
2118 | if age_limit is None: # No limit set | |
2119 | return False | |
2120 | if content_limit is None: | |
2121 | return False # Content available for everyone | |
2122 | return age_limit < content_limit | |
61ca9a80 PH |
2123 | |
2124 | ||
2125 | def is_html(first_bytes): | |
2126 | """ Detect whether a file contains HTML by examining its first bytes. """ | |
2127 | ||
2128 | BOMS = [ | |
2129 | (b'\xef\xbb\xbf', 'utf-8'), | |
2130 | (b'\x00\x00\xfe\xff', 'utf-32-be'), | |
2131 | (b'\xff\xfe\x00\x00', 'utf-32-le'), | |
2132 | (b'\xff\xfe', 'utf-16-le'), | |
2133 | (b'\xfe\xff', 'utf-16-be'), | |
2134 | ] | |
2135 | for bom, enc in BOMS: | |
2136 | if first_bytes.startswith(bom): | |
2137 | s = first_bytes[len(bom):].decode(enc, 'replace') | |
2138 | break | |
2139 | else: | |
2140 | s = first_bytes.decode('utf-8', 'replace') | |
2141 | ||
2142 | return re.match(r'^\s*<', s) | |
a055469f PH |
2143 | |
2144 | ||
2145 | def determine_protocol(info_dict): | |
2146 | protocol = info_dict.get('protocol') | |
2147 | if protocol is not None: | |
2148 | return protocol | |
2149 | ||
2150 | url = info_dict['url'] | |
2151 | if url.startswith('rtmp'): | |
2152 | return 'rtmp' | |
2153 | elif url.startswith('mms'): | |
2154 | return 'mms' | |
2155 | elif url.startswith('rtsp'): | |
2156 | return 'rtsp' | |
2157 | ||
2158 | ext = determine_ext(url) | |
2159 | if ext == 'm3u8': | |
2160 | return 'm3u8' | |
2161 | elif ext == 'f4m': | |
2162 | return 'f4m' | |
2163 | ||
2164 | return compat_urllib_parse_urlparse(url).scheme | |
cfb56d1a PH |
2165 | |
2166 | ||
2167 | def render_table(header_row, data): | |
2168 | """ Render a list of rows, each as a list of values """ | |
2169 | table = [header_row] + data | |
2170 | max_lens = [max(len(compat_str(v)) for v in col) for col in zip(*table)] | |
2171 | format_str = ' '.join('%-' + compat_str(ml + 1) + 's' for ml in max_lens[:-1]) + '%s' | |
2172 | return '\n'.join(format_str % tuple(row) for row in table) | |
347de493 PH |
2173 | |
2174 | ||
2175 | def _match_one(filter_part, dct): | |
2176 | COMPARISON_OPERATORS = { | |
2177 | '<': operator.lt, | |
2178 | '<=': operator.le, | |
2179 | '>': operator.gt, | |
2180 | '>=': operator.ge, | |
2181 | '=': operator.eq, | |
2182 | '!=': operator.ne, | |
2183 | } | |
2184 | operator_rex = re.compile(r'''(?x)\s* | |
2185 | (?P<key>[a-z_]+) | |
2186 | \s*(?P<op>%s)(?P<none_inclusive>\s*\?)?\s* | |
2187 | (?: | |
2188 | (?P<intval>[0-9.]+(?:[kKmMgGtTpPeEzZyY]i?[Bb]?)?)| | |
2189 | (?P<strval>(?![0-9.])[a-z0-9A-Z]*) | |
2190 | ) | |
2191 | \s*$ | |
2192 | ''' % '|'.join(map(re.escape, COMPARISON_OPERATORS.keys()))) | |
2193 | m = operator_rex.search(filter_part) | |
2194 | if m: | |
2195 | op = COMPARISON_OPERATORS[m.group('op')] | |
2196 | if m.group('strval') is not None: | |
2197 | if m.group('op') not in ('=', '!='): | |
2198 | raise ValueError( | |
2199 | 'Operator %s does not support string values!' % m.group('op')) | |
2200 | comparison_value = m.group('strval') | |
2201 | else: | |
2202 | try: | |
2203 | comparison_value = int(m.group('intval')) | |
2204 | except ValueError: | |
2205 | comparison_value = parse_filesize(m.group('intval')) | |
2206 | if comparison_value is None: | |
2207 | comparison_value = parse_filesize(m.group('intval') + 'B') | |
2208 | if comparison_value is None: | |
2209 | raise ValueError( | |
2210 | 'Invalid integer value %r in filter part %r' % ( | |
2211 | m.group('intval'), filter_part)) | |
2212 | actual_value = dct.get(m.group('key')) | |
2213 | if actual_value is None: | |
2214 | return m.group('none_inclusive') | |
2215 | return op(actual_value, comparison_value) | |
2216 | ||
2217 | UNARY_OPERATORS = { | |
2218 | '': lambda v: v is not None, | |
2219 | '!': lambda v: v is None, | |
2220 | } | |
2221 | operator_rex = re.compile(r'''(?x)\s* | |
2222 | (?P<op>%s)\s*(?P<key>[a-z_]+) | |
2223 | \s*$ | |
2224 | ''' % '|'.join(map(re.escape, UNARY_OPERATORS.keys()))) | |
2225 | m = operator_rex.search(filter_part) | |
2226 | if m: | |
2227 | op = UNARY_OPERATORS[m.group('op')] | |
2228 | actual_value = dct.get(m.group('key')) | |
2229 | return op(actual_value) | |
2230 | ||
2231 | raise ValueError('Invalid filter part %r' % filter_part) | |
2232 | ||
2233 | ||
2234 | def match_str(filter_str, dct): | |
2235 | """ Filter a dictionary with a simple string syntax. Returns True (=passes filter) or false """ | |
2236 | ||
2237 | return all( | |
2238 | _match_one(filter_part, dct) for filter_part in filter_str.split('&')) | |
2239 | ||
2240 | ||
2241 | def match_filter_func(filter_str): | |
2242 | def _match_func(info_dict): | |
2243 | if match_str(filter_str, info_dict): | |
2244 | return None | |
2245 | else: | |
2246 | video_title = info_dict.get('title', info_dict.get('id', 'video')) | |
2247 | return '%s does not pass filter %s, skipping ..' % (video_title, filter_str) | |
2248 | return _match_func | |
91410c9b PH |
2249 | |
2250 | ||
bf6427d2 YCH |
2251 | def parse_dfxp_time_expr(time_expr): |
2252 | if not time_expr: | |
d631d5f9 | 2253 | return |
bf6427d2 YCH |
2254 | |
2255 | mobj = re.match(r'^(?P<time_offset>\d+(?:\.\d+)?)s?$', time_expr) | |
2256 | if mobj: | |
2257 | return float(mobj.group('time_offset')) | |
2258 | ||
db2fe38b | 2259 | mobj = re.match(r'^(\d+):(\d\d):(\d\d(?:(?:\.|:)\d+)?)$', time_expr) |
bf6427d2 | 2260 | if mobj: |
db2fe38b | 2261 | return 3600 * int(mobj.group(1)) + 60 * int(mobj.group(2)) + float(mobj.group(3).replace(':', '.')) |
bf6427d2 YCH |
2262 | |
2263 | ||
c1c924ab YCH |
2264 | def srt_subtitles_timecode(seconds): |
2265 | return '%02d:%02d:%02d,%03d' % (seconds / 3600, (seconds % 3600) / 60, seconds % 60, (seconds % 1) * 1000) | |
bf6427d2 YCH |
2266 | |
2267 | ||
2268 | def dfxp2srt(dfxp_data): | |
4e335771 YCH |
2269 | _x = functools.partial(xpath_with_ns, ns_map={ |
2270 | 'ttml': 'http://www.w3.org/ns/ttml', | |
2271 | 'ttaf1': 'http://www.w3.org/2006/10/ttaf1', | |
5bf28d78 | 2272 | 'ttaf1_0604': 'http://www.w3.org/2006/04/ttaf1', |
4e335771 | 2273 | }) |
bf6427d2 | 2274 | |
87de7069 | 2275 | class TTMLPElementParser(object): |
2b14cb56 | 2276 | out = '' |
bf6427d2 | 2277 | |
2b14cb56 | 2278 | def start(self, tag, attrib): |
2279 | if tag in (_x('ttml:br'), _x('ttaf1:br'), 'br'): | |
2280 | self.out += '\n' | |
bf6427d2 | 2281 | |
2b14cb56 | 2282 | def end(self, tag): |
2283 | pass | |
bf6427d2 | 2284 | |
2b14cb56 | 2285 | def data(self, data): |
2286 | self.out += data | |
2287 | ||
2288 | def close(self): | |
2289 | return self.out.strip() | |
2290 | ||
2291 | def parse_node(node): | |
2292 | target = TTMLPElementParser() | |
2293 | parser = xml.etree.ElementTree.XMLParser(target=target) | |
2294 | parser.feed(xml.etree.ElementTree.tostring(node)) | |
2295 | return parser.close() | |
bf6427d2 | 2296 | |
36e6f62c | 2297 | dfxp = compat_etree_fromstring(dfxp_data.encode('utf-8')) |
bf6427d2 | 2298 | out = [] |
5bf28d78 | 2299 | paras = dfxp.findall(_x('.//ttml:p')) or dfxp.findall(_x('.//ttaf1:p')) or dfxp.findall(_x('.//ttaf1_0604:p')) or dfxp.findall('.//p') |
1b0427e6 YCH |
2300 | |
2301 | if not paras: | |
2302 | raise ValueError('Invalid dfxp/TTML subtitle') | |
bf6427d2 YCH |
2303 | |
2304 | for para, index in zip(paras, itertools.count(1)): | |
d631d5f9 | 2305 | begin_time = parse_dfxp_time_expr(para.attrib.get('begin')) |
7dff0363 | 2306 | end_time = parse_dfxp_time_expr(para.attrib.get('end')) |
d631d5f9 YCH |
2307 | dur = parse_dfxp_time_expr(para.attrib.get('dur')) |
2308 | if begin_time is None: | |
2309 | continue | |
7dff0363 | 2310 | if not end_time: |
d631d5f9 YCH |
2311 | if not dur: |
2312 | continue | |
2313 | end_time = begin_time + dur | |
bf6427d2 YCH |
2314 | out.append('%d\n%s --> %s\n%s\n\n' % ( |
2315 | index, | |
c1c924ab YCH |
2316 | srt_subtitles_timecode(begin_time), |
2317 | srt_subtitles_timecode(end_time), | |
bf6427d2 YCH |
2318 | parse_node(para))) |
2319 | ||
2320 | return ''.join(out) | |
2321 | ||
2322 | ||
66e289ba S |
2323 | def cli_option(params, command_option, param): |
2324 | param = params.get(param) | |
2325 | return [command_option, param] if param is not None else [] | |
2326 | ||
2327 | ||
2328 | def cli_bool_option(params, command_option, param, true_value='true', false_value='false', separator=None): | |
2329 | param = params.get(param) | |
2330 | assert isinstance(param, bool) | |
2331 | if separator: | |
2332 | return [command_option + separator + (true_value if param else false_value)] | |
2333 | return [command_option, true_value if param else false_value] | |
2334 | ||
2335 | ||
2336 | def cli_valueless_option(params, command_option, param, expected_value=True): | |
2337 | param = params.get(param) | |
2338 | return [command_option] if param == expected_value else [] | |
2339 | ||
2340 | ||
2341 | def cli_configuration_args(params, param, default=[]): | |
2342 | ex_args = params.get(param) | |
2343 | if ex_args is None: | |
2344 | return default | |
2345 | assert isinstance(ex_args, list) | |
2346 | return ex_args | |
2347 | ||
2348 | ||
39672624 YCH |
2349 | class ISO639Utils(object): |
2350 | # See http://www.loc.gov/standards/iso639-2/ISO-639-2_utf-8.txt | |
2351 | _lang_map = { | |
2352 | 'aa': 'aar', | |
2353 | 'ab': 'abk', | |
2354 | 'ae': 'ave', | |
2355 | 'af': 'afr', | |
2356 | 'ak': 'aka', | |
2357 | 'am': 'amh', | |
2358 | 'an': 'arg', | |
2359 | 'ar': 'ara', | |
2360 | 'as': 'asm', | |
2361 | 'av': 'ava', | |
2362 | 'ay': 'aym', | |
2363 | 'az': 'aze', | |
2364 | 'ba': 'bak', | |
2365 | 'be': 'bel', | |
2366 | 'bg': 'bul', | |
2367 | 'bh': 'bih', | |
2368 | 'bi': 'bis', | |
2369 | 'bm': 'bam', | |
2370 | 'bn': 'ben', | |
2371 | 'bo': 'bod', | |
2372 | 'br': 'bre', | |
2373 | 'bs': 'bos', | |
2374 | 'ca': 'cat', | |
2375 | 'ce': 'che', | |
2376 | 'ch': 'cha', | |
2377 | 'co': 'cos', | |
2378 | 'cr': 'cre', | |
2379 | 'cs': 'ces', | |
2380 | 'cu': 'chu', | |
2381 | 'cv': 'chv', | |
2382 | 'cy': 'cym', | |
2383 | 'da': 'dan', | |
2384 | 'de': 'deu', | |
2385 | 'dv': 'div', | |
2386 | 'dz': 'dzo', | |
2387 | 'ee': 'ewe', | |
2388 | 'el': 'ell', | |
2389 | 'en': 'eng', | |
2390 | 'eo': 'epo', | |
2391 | 'es': 'spa', | |
2392 | 'et': 'est', | |
2393 | 'eu': 'eus', | |
2394 | 'fa': 'fas', | |
2395 | 'ff': 'ful', | |
2396 | 'fi': 'fin', | |
2397 | 'fj': 'fij', | |
2398 | 'fo': 'fao', | |
2399 | 'fr': 'fra', | |
2400 | 'fy': 'fry', | |
2401 | 'ga': 'gle', | |
2402 | 'gd': 'gla', | |
2403 | 'gl': 'glg', | |
2404 | 'gn': 'grn', | |
2405 | 'gu': 'guj', | |
2406 | 'gv': 'glv', | |
2407 | 'ha': 'hau', | |
2408 | 'he': 'heb', | |
2409 | 'hi': 'hin', | |
2410 | 'ho': 'hmo', | |
2411 | 'hr': 'hrv', | |
2412 | 'ht': 'hat', | |
2413 | 'hu': 'hun', | |
2414 | 'hy': 'hye', | |
2415 | 'hz': 'her', | |
2416 | 'ia': 'ina', | |
2417 | 'id': 'ind', | |
2418 | 'ie': 'ile', | |
2419 | 'ig': 'ibo', | |
2420 | 'ii': 'iii', | |
2421 | 'ik': 'ipk', | |
2422 | 'io': 'ido', | |
2423 | 'is': 'isl', | |
2424 | 'it': 'ita', | |
2425 | 'iu': 'iku', | |
2426 | 'ja': 'jpn', | |
2427 | 'jv': 'jav', | |
2428 | 'ka': 'kat', | |
2429 | 'kg': 'kon', | |
2430 | 'ki': 'kik', | |
2431 | 'kj': 'kua', | |
2432 | 'kk': 'kaz', | |
2433 | 'kl': 'kal', | |
2434 | 'km': 'khm', | |
2435 | 'kn': 'kan', | |
2436 | 'ko': 'kor', | |
2437 | 'kr': 'kau', | |
2438 | 'ks': 'kas', | |
2439 | 'ku': 'kur', | |
2440 | 'kv': 'kom', | |
2441 | 'kw': 'cor', | |
2442 | 'ky': 'kir', | |
2443 | 'la': 'lat', | |
2444 | 'lb': 'ltz', | |
2445 | 'lg': 'lug', | |
2446 | 'li': 'lim', | |
2447 | 'ln': 'lin', | |
2448 | 'lo': 'lao', | |
2449 | 'lt': 'lit', | |
2450 | 'lu': 'lub', | |
2451 | 'lv': 'lav', | |
2452 | 'mg': 'mlg', | |
2453 | 'mh': 'mah', | |
2454 | 'mi': 'mri', | |
2455 | 'mk': 'mkd', | |
2456 | 'ml': 'mal', | |
2457 | 'mn': 'mon', | |
2458 | 'mr': 'mar', | |
2459 | 'ms': 'msa', | |
2460 | 'mt': 'mlt', | |
2461 | 'my': 'mya', | |
2462 | 'na': 'nau', | |
2463 | 'nb': 'nob', | |
2464 | 'nd': 'nde', | |
2465 | 'ne': 'nep', | |
2466 | 'ng': 'ndo', | |
2467 | 'nl': 'nld', | |
2468 | 'nn': 'nno', | |
2469 | 'no': 'nor', | |
2470 | 'nr': 'nbl', | |
2471 | 'nv': 'nav', | |
2472 | 'ny': 'nya', | |
2473 | 'oc': 'oci', | |
2474 | 'oj': 'oji', | |
2475 | 'om': 'orm', | |
2476 | 'or': 'ori', | |
2477 | 'os': 'oss', | |
2478 | 'pa': 'pan', | |
2479 | 'pi': 'pli', | |
2480 | 'pl': 'pol', | |
2481 | 'ps': 'pus', | |
2482 | 'pt': 'por', | |
2483 | 'qu': 'que', | |
2484 | 'rm': 'roh', | |
2485 | 'rn': 'run', | |
2486 | 'ro': 'ron', | |
2487 | 'ru': 'rus', | |
2488 | 'rw': 'kin', | |
2489 | 'sa': 'san', | |
2490 | 'sc': 'srd', | |
2491 | 'sd': 'snd', | |
2492 | 'se': 'sme', | |
2493 | 'sg': 'sag', | |
2494 | 'si': 'sin', | |
2495 | 'sk': 'slk', | |
2496 | 'sl': 'slv', | |
2497 | 'sm': 'smo', | |
2498 | 'sn': 'sna', | |
2499 | 'so': 'som', | |
2500 | 'sq': 'sqi', | |
2501 | 'sr': 'srp', | |
2502 | 'ss': 'ssw', | |
2503 | 'st': 'sot', | |
2504 | 'su': 'sun', | |
2505 | 'sv': 'swe', | |
2506 | 'sw': 'swa', | |
2507 | 'ta': 'tam', | |
2508 | 'te': 'tel', | |
2509 | 'tg': 'tgk', | |
2510 | 'th': 'tha', | |
2511 | 'ti': 'tir', | |
2512 | 'tk': 'tuk', | |
2513 | 'tl': 'tgl', | |
2514 | 'tn': 'tsn', | |
2515 | 'to': 'ton', | |
2516 | 'tr': 'tur', | |
2517 | 'ts': 'tso', | |
2518 | 'tt': 'tat', | |
2519 | 'tw': 'twi', | |
2520 | 'ty': 'tah', | |
2521 | 'ug': 'uig', | |
2522 | 'uk': 'ukr', | |
2523 | 'ur': 'urd', | |
2524 | 'uz': 'uzb', | |
2525 | 've': 'ven', | |
2526 | 'vi': 'vie', | |
2527 | 'vo': 'vol', | |
2528 | 'wa': 'wln', | |
2529 | 'wo': 'wol', | |
2530 | 'xh': 'xho', | |
2531 | 'yi': 'yid', | |
2532 | 'yo': 'yor', | |
2533 | 'za': 'zha', | |
2534 | 'zh': 'zho', | |
2535 | 'zu': 'zul', | |
2536 | } | |
2537 | ||
2538 | @classmethod | |
2539 | def short2long(cls, code): | |
2540 | """Convert language code from ISO 639-1 to ISO 639-2/T""" | |
2541 | return cls._lang_map.get(code[:2]) | |
2542 | ||
2543 | @classmethod | |
2544 | def long2short(cls, code): | |
2545 | """Convert language code from ISO 639-2/T to ISO 639-1""" | |
2546 | for short_name, long_name in cls._lang_map.items(): | |
2547 | if long_name == code: | |
2548 | return short_name | |
2549 | ||
2550 | ||
4eb10f66 YCH |
2551 | class ISO3166Utils(object): |
2552 | # From http://data.okfn.org/data/core/country-list | |
2553 | _country_map = { | |
2554 | 'AF': 'Afghanistan', | |
2555 | 'AX': 'Åland Islands', | |
2556 | 'AL': 'Albania', | |
2557 | 'DZ': 'Algeria', | |
2558 | 'AS': 'American Samoa', | |
2559 | 'AD': 'Andorra', | |
2560 | 'AO': 'Angola', | |
2561 | 'AI': 'Anguilla', | |
2562 | 'AQ': 'Antarctica', | |
2563 | 'AG': 'Antigua and Barbuda', | |
2564 | 'AR': 'Argentina', | |
2565 | 'AM': 'Armenia', | |
2566 | 'AW': 'Aruba', | |
2567 | 'AU': 'Australia', | |
2568 | 'AT': 'Austria', | |
2569 | 'AZ': 'Azerbaijan', | |
2570 | 'BS': 'Bahamas', | |
2571 | 'BH': 'Bahrain', | |
2572 | 'BD': 'Bangladesh', | |
2573 | 'BB': 'Barbados', | |
2574 | 'BY': 'Belarus', | |
2575 | 'BE': 'Belgium', | |
2576 | 'BZ': 'Belize', | |
2577 | 'BJ': 'Benin', | |
2578 | 'BM': 'Bermuda', | |
2579 | 'BT': 'Bhutan', | |
2580 | 'BO': 'Bolivia, Plurinational State of', | |
2581 | 'BQ': 'Bonaire, Sint Eustatius and Saba', | |
2582 | 'BA': 'Bosnia and Herzegovina', | |
2583 | 'BW': 'Botswana', | |
2584 | 'BV': 'Bouvet Island', | |
2585 | 'BR': 'Brazil', | |
2586 | 'IO': 'British Indian Ocean Territory', | |
2587 | 'BN': 'Brunei Darussalam', | |
2588 | 'BG': 'Bulgaria', | |
2589 | 'BF': 'Burkina Faso', | |
2590 | 'BI': 'Burundi', | |
2591 | 'KH': 'Cambodia', | |
2592 | 'CM': 'Cameroon', | |
2593 | 'CA': 'Canada', | |
2594 | 'CV': 'Cape Verde', | |
2595 | 'KY': 'Cayman Islands', | |
2596 | 'CF': 'Central African Republic', | |
2597 | 'TD': 'Chad', | |
2598 | 'CL': 'Chile', | |
2599 | 'CN': 'China', | |
2600 | 'CX': 'Christmas Island', | |
2601 | 'CC': 'Cocos (Keeling) Islands', | |
2602 | 'CO': 'Colombia', | |
2603 | 'KM': 'Comoros', | |
2604 | 'CG': 'Congo', | |
2605 | 'CD': 'Congo, the Democratic Republic of the', | |
2606 | 'CK': 'Cook Islands', | |
2607 | 'CR': 'Costa Rica', | |
2608 | 'CI': 'Côte d\'Ivoire', | |
2609 | 'HR': 'Croatia', | |
2610 | 'CU': 'Cuba', | |
2611 | 'CW': 'Curaçao', | |
2612 | 'CY': 'Cyprus', | |
2613 | 'CZ': 'Czech Republic', | |
2614 | 'DK': 'Denmark', | |
2615 | 'DJ': 'Djibouti', | |
2616 | 'DM': 'Dominica', | |
2617 | 'DO': 'Dominican Republic', | |
2618 | 'EC': 'Ecuador', | |
2619 | 'EG': 'Egypt', | |
2620 | 'SV': 'El Salvador', | |
2621 | 'GQ': 'Equatorial Guinea', | |
2622 | 'ER': 'Eritrea', | |
2623 | 'EE': 'Estonia', | |
2624 | 'ET': 'Ethiopia', | |
2625 | 'FK': 'Falkland Islands (Malvinas)', | |
2626 | 'FO': 'Faroe Islands', | |
2627 | 'FJ': 'Fiji', | |
2628 | 'FI': 'Finland', | |
2629 | 'FR': 'France', | |
2630 | 'GF': 'French Guiana', | |
2631 | 'PF': 'French Polynesia', | |
2632 | 'TF': 'French Southern Territories', | |
2633 | 'GA': 'Gabon', | |
2634 | 'GM': 'Gambia', | |
2635 | 'GE': 'Georgia', | |
2636 | 'DE': 'Germany', | |
2637 | 'GH': 'Ghana', | |
2638 | 'GI': 'Gibraltar', | |
2639 | 'GR': 'Greece', | |
2640 | 'GL': 'Greenland', | |
2641 | 'GD': 'Grenada', | |
2642 | 'GP': 'Guadeloupe', | |
2643 | 'GU': 'Guam', | |
2644 | 'GT': 'Guatemala', | |
2645 | 'GG': 'Guernsey', | |
2646 | 'GN': 'Guinea', | |
2647 | 'GW': 'Guinea-Bissau', | |
2648 | 'GY': 'Guyana', | |
2649 | 'HT': 'Haiti', | |
2650 | 'HM': 'Heard Island and McDonald Islands', | |
2651 | 'VA': 'Holy See (Vatican City State)', | |
2652 | 'HN': 'Honduras', | |
2653 | 'HK': 'Hong Kong', | |
2654 | 'HU': 'Hungary', | |
2655 | 'IS': 'Iceland', | |
2656 | 'IN': 'India', | |
2657 | 'ID': 'Indonesia', | |
2658 | 'IR': 'Iran, Islamic Republic of', | |
2659 | 'IQ': 'Iraq', | |
2660 | 'IE': 'Ireland', | |
2661 | 'IM': 'Isle of Man', | |
2662 | 'IL': 'Israel', | |
2663 | 'IT': 'Italy', | |
2664 | 'JM': 'Jamaica', | |
2665 | 'JP': 'Japan', | |
2666 | 'JE': 'Jersey', | |
2667 | 'JO': 'Jordan', | |
2668 | 'KZ': 'Kazakhstan', | |
2669 | 'KE': 'Kenya', | |
2670 | 'KI': 'Kiribati', | |
2671 | 'KP': 'Korea, Democratic People\'s Republic of', | |
2672 | 'KR': 'Korea, Republic of', | |
2673 | 'KW': 'Kuwait', | |
2674 | 'KG': 'Kyrgyzstan', | |
2675 | 'LA': 'Lao People\'s Democratic Republic', | |
2676 | 'LV': 'Latvia', | |
2677 | 'LB': 'Lebanon', | |
2678 | 'LS': 'Lesotho', | |
2679 | 'LR': 'Liberia', | |
2680 | 'LY': 'Libya', | |
2681 | 'LI': 'Liechtenstein', | |
2682 | 'LT': 'Lithuania', | |
2683 | 'LU': 'Luxembourg', | |
2684 | 'MO': 'Macao', | |
2685 | 'MK': 'Macedonia, the Former Yugoslav Republic of', | |
2686 | 'MG': 'Madagascar', | |
2687 | 'MW': 'Malawi', | |
2688 | 'MY': 'Malaysia', | |
2689 | 'MV': 'Maldives', | |
2690 | 'ML': 'Mali', | |
2691 | 'MT': 'Malta', | |
2692 | 'MH': 'Marshall Islands', | |
2693 | 'MQ': 'Martinique', | |
2694 | 'MR': 'Mauritania', | |
2695 | 'MU': 'Mauritius', | |
2696 | 'YT': 'Mayotte', | |
2697 | 'MX': 'Mexico', | |
2698 | 'FM': 'Micronesia, Federated States of', | |
2699 | 'MD': 'Moldova, Republic of', | |
2700 | 'MC': 'Monaco', | |
2701 | 'MN': 'Mongolia', | |
2702 | 'ME': 'Montenegro', | |
2703 | 'MS': 'Montserrat', | |
2704 | 'MA': 'Morocco', | |
2705 | 'MZ': 'Mozambique', | |
2706 | 'MM': 'Myanmar', | |
2707 | 'NA': 'Namibia', | |
2708 | 'NR': 'Nauru', | |
2709 | 'NP': 'Nepal', | |
2710 | 'NL': 'Netherlands', | |
2711 | 'NC': 'New Caledonia', | |
2712 | 'NZ': 'New Zealand', | |
2713 | 'NI': 'Nicaragua', | |
2714 | 'NE': 'Niger', | |
2715 | 'NG': 'Nigeria', | |
2716 | 'NU': 'Niue', | |
2717 | 'NF': 'Norfolk Island', | |
2718 | 'MP': 'Northern Mariana Islands', | |
2719 | 'NO': 'Norway', | |
2720 | 'OM': 'Oman', | |
2721 | 'PK': 'Pakistan', | |
2722 | 'PW': 'Palau', | |
2723 | 'PS': 'Palestine, State of', | |
2724 | 'PA': 'Panama', | |
2725 | 'PG': 'Papua New Guinea', | |
2726 | 'PY': 'Paraguay', | |
2727 | 'PE': 'Peru', | |
2728 | 'PH': 'Philippines', | |
2729 | 'PN': 'Pitcairn', | |
2730 | 'PL': 'Poland', | |
2731 | 'PT': 'Portugal', | |
2732 | 'PR': 'Puerto Rico', | |
2733 | 'QA': 'Qatar', | |
2734 | 'RE': 'Réunion', | |
2735 | 'RO': 'Romania', | |
2736 | 'RU': 'Russian Federation', | |
2737 | 'RW': 'Rwanda', | |
2738 | 'BL': 'Saint Barthélemy', | |
2739 | 'SH': 'Saint Helena, Ascension and Tristan da Cunha', | |
2740 | 'KN': 'Saint Kitts and Nevis', | |
2741 | 'LC': 'Saint Lucia', | |
2742 | 'MF': 'Saint Martin (French part)', | |
2743 | 'PM': 'Saint Pierre and Miquelon', | |
2744 | 'VC': 'Saint Vincent and the Grenadines', | |
2745 | 'WS': 'Samoa', | |
2746 | 'SM': 'San Marino', | |
2747 | 'ST': 'Sao Tome and Principe', | |
2748 | 'SA': 'Saudi Arabia', | |
2749 | 'SN': 'Senegal', | |
2750 | 'RS': 'Serbia', | |
2751 | 'SC': 'Seychelles', | |
2752 | 'SL': 'Sierra Leone', | |
2753 | 'SG': 'Singapore', | |
2754 | 'SX': 'Sint Maarten (Dutch part)', | |
2755 | 'SK': 'Slovakia', | |
2756 | 'SI': 'Slovenia', | |
2757 | 'SB': 'Solomon Islands', | |
2758 | 'SO': 'Somalia', | |
2759 | 'ZA': 'South Africa', | |
2760 | 'GS': 'South Georgia and the South Sandwich Islands', | |
2761 | 'SS': 'South Sudan', | |
2762 | 'ES': 'Spain', | |
2763 | 'LK': 'Sri Lanka', | |
2764 | 'SD': 'Sudan', | |
2765 | 'SR': 'Suriname', | |
2766 | 'SJ': 'Svalbard and Jan Mayen', | |
2767 | 'SZ': 'Swaziland', | |
2768 | 'SE': 'Sweden', | |
2769 | 'CH': 'Switzerland', | |
2770 | 'SY': 'Syrian Arab Republic', | |
2771 | 'TW': 'Taiwan, Province of China', | |
2772 | 'TJ': 'Tajikistan', | |
2773 | 'TZ': 'Tanzania, United Republic of', | |
2774 | 'TH': 'Thailand', | |
2775 | 'TL': 'Timor-Leste', | |
2776 | 'TG': 'Togo', | |
2777 | 'TK': 'Tokelau', | |
2778 | 'TO': 'Tonga', | |
2779 | 'TT': 'Trinidad and Tobago', | |
2780 | 'TN': 'Tunisia', | |
2781 | 'TR': 'Turkey', | |
2782 | 'TM': 'Turkmenistan', | |
2783 | 'TC': 'Turks and Caicos Islands', | |
2784 | 'TV': 'Tuvalu', | |
2785 | 'UG': 'Uganda', | |
2786 | 'UA': 'Ukraine', | |
2787 | 'AE': 'United Arab Emirates', | |
2788 | 'GB': 'United Kingdom', | |
2789 | 'US': 'United States', | |
2790 | 'UM': 'United States Minor Outlying Islands', | |
2791 | 'UY': 'Uruguay', | |
2792 | 'UZ': 'Uzbekistan', | |
2793 | 'VU': 'Vanuatu', | |
2794 | 'VE': 'Venezuela, Bolivarian Republic of', | |
2795 | 'VN': 'Viet Nam', | |
2796 | 'VG': 'Virgin Islands, British', | |
2797 | 'VI': 'Virgin Islands, U.S.', | |
2798 | 'WF': 'Wallis and Futuna', | |
2799 | 'EH': 'Western Sahara', | |
2800 | 'YE': 'Yemen', | |
2801 | 'ZM': 'Zambia', | |
2802 | 'ZW': 'Zimbabwe', | |
2803 | } | |
2804 | ||
2805 | @classmethod | |
2806 | def short2full(cls, code): | |
2807 | """Convert an ISO 3166-2 country code to the corresponding full name""" | |
2808 | return cls._country_map.get(code.upper()) | |
2809 | ||
2810 | ||
91410c9b | 2811 | class PerRequestProxyHandler(compat_urllib_request.ProxyHandler): |
2461f79d PH |
2812 | def __init__(self, proxies=None): |
2813 | # Set default handlers | |
2814 | for type in ('http', 'https'): | |
2815 | setattr(self, '%s_open' % type, | |
2816 | lambda r, proxy='__noproxy__', type=type, meth=self.proxy_open: | |
2817 | meth(r, proxy, type)) | |
2818 | return compat_urllib_request.ProxyHandler.__init__(self, proxies) | |
2819 | ||
91410c9b | 2820 | def proxy_open(self, req, proxy, type): |
2461f79d | 2821 | req_proxy = req.headers.get('Ytdl-request-proxy') |
91410c9b PH |
2822 | if req_proxy is not None: |
2823 | proxy = req_proxy | |
2461f79d PH |
2824 | del req.headers['Ytdl-request-proxy'] |
2825 | ||
2826 | if proxy == '__noproxy__': | |
2827 | return None # No Proxy | |
51fb4995 | 2828 | if compat_urlparse.urlparse(proxy).scheme.lower() in ('socks', 'socks4', 'socks4a', 'socks5'): |
71aff188 YCH |
2829 | req.add_header('Ytdl-socks-proxy', proxy) |
2830 | # youtube-dl's http/https handlers do wrapping the socket with socks | |
2831 | return None | |
91410c9b PH |
2832 | return compat_urllib_request.ProxyHandler.proxy_open( |
2833 | self, req, proxy, type) | |
5bc880b9 YCH |
2834 | |
2835 | ||
2836 | def ohdave_rsa_encrypt(data, exponent, modulus): | |
2837 | ''' | |
2838 | Implement OHDave's RSA algorithm. See http://www.ohdave.com/rsa/ | |
2839 | ||
2840 | Input: | |
2841 | data: data to encrypt, bytes-like object | |
2842 | exponent, modulus: parameter e and N of RSA algorithm, both integer | |
2843 | Output: hex string of encrypted data | |
2844 | ||
2845 | Limitation: supports one block encryption only | |
2846 | ''' | |
2847 | ||
2848 | payload = int(binascii.hexlify(data[::-1]), 16) | |
2849 | encrypted = pow(payload, exponent, modulus) | |
2850 | return '%x' % encrypted | |
81bdc8fd YCH |
2851 | |
2852 | ||
5eb6bdce | 2853 | def encode_base_n(num, n, table=None): |
59f898b7 | 2854 | FULL_TABLE = '0123456789abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ' |
59f898b7 YCH |
2855 | if not table: |
2856 | table = FULL_TABLE[:n] | |
2857 | ||
5eb6bdce YCH |
2858 | if n > len(table): |
2859 | raise ValueError('base %d exceeds table length %d' % (n, len(table))) | |
2860 | ||
2861 | if num == 0: | |
2862 | return table[0] | |
2863 | ||
81bdc8fd YCH |
2864 | ret = '' |
2865 | while num: | |
2866 | ret = table[num % n] + ret | |
2867 | num = num // n | |
2868 | return ret | |
f52354a8 YCH |
2869 | |
2870 | ||
2871 | def decode_packed_codes(code): | |
2872 | mobj = re.search( | |
680079be | 2873 | r"}\('(.+)',(\d+),(\d+),'([^']+)'\.split\('\|'\)", |
f52354a8 YCH |
2874 | code) |
2875 | obfucasted_code, base, count, symbols = mobj.groups() | |
2876 | base = int(base) | |
2877 | count = int(count) | |
2878 | symbols = symbols.split('|') | |
2879 | symbol_table = {} | |
2880 | ||
2881 | while count: | |
2882 | count -= 1 | |
5eb6bdce | 2883 | base_n_count = encode_base_n(count, base) |
f52354a8 YCH |
2884 | symbol_table[base_n_count] = symbols[count] or base_n_count |
2885 | ||
2886 | return re.sub( | |
2887 | r'\b(\w+)\b', lambda mobj: symbol_table[mobj.group(0)], | |
2888 | obfucasted_code) | |
e154c651 | 2889 | |
2890 | ||
2891 | def parse_m3u8_attributes(attrib): | |
2892 | info = {} | |
2893 | for (key, val) in re.findall(r'(?P<key>[A-Z0-9-]+)=(?P<val>"[^"]+"|[^",]+)(?:,|$)', attrib): | |
2894 | if val.startswith('"'): | |
2895 | val = val[1:-1] | |
2896 | info[key] = val | |
2897 | return info |