]>
Commit | Line | Data |
---|---|---|
451948b2 PH |
1 | from __future__ import unicode_literals |
2 | ||
003c69a8 | 3 | import collections |
8c25f81b | 4 | import getpass |
e07e9313 | 5 | import optparse |
8c25f81b | 6 | import os |
7d4111ed | 7 | import re |
003c69a8 | 8 | import shutil |
be4a824d | 9 | import socket |
8c25f81b PH |
10 | import subprocess |
11 | import sys | |
a0e060ac | 12 | import itertools |
8c25f81b PH |
13 | |
14 | ||
15 | try: | |
16 | import urllib.request as compat_urllib_request | |
5f6a1245 | 17 | except ImportError: # Python 2 |
8c25f81b PH |
18 | import urllib2 as compat_urllib_request |
19 | ||
20 | try: | |
21 | import urllib.error as compat_urllib_error | |
5f6a1245 | 22 | except ImportError: # Python 2 |
8c25f81b PH |
23 | import urllib2 as compat_urllib_error |
24 | ||
25 | try: | |
26 | import urllib.parse as compat_urllib_parse | |
5f6a1245 | 27 | except ImportError: # Python 2 |
8c25f81b PH |
28 | import urllib as compat_urllib_parse |
29 | ||
30 | try: | |
31 | from urllib.parse import urlparse as compat_urllib_parse_urlparse | |
5f6a1245 | 32 | except ImportError: # Python 2 |
8c25f81b PH |
33 | from urlparse import urlparse as compat_urllib_parse_urlparse |
34 | ||
35 | try: | |
36 | import urllib.parse as compat_urlparse | |
5f6a1245 | 37 | except ImportError: # Python 2 |
8c25f81b PH |
38 | import urlparse as compat_urlparse |
39 | ||
40 | try: | |
41 | import http.cookiejar as compat_cookiejar | |
5f6a1245 | 42 | except ImportError: # Python 2 |
8c25f81b PH |
43 | import cookielib as compat_cookiejar |
44 | ||
45 | try: | |
46 | import html.entities as compat_html_entities | |
5f6a1245 | 47 | except ImportError: # Python 2 |
8c25f81b PH |
48 | import htmlentitydefs as compat_html_entities |
49 | ||
8c25f81b PH |
50 | try: |
51 | import http.client as compat_http_client | |
5f6a1245 | 52 | except ImportError: # Python 2 |
8c25f81b PH |
53 | import httplib as compat_http_client |
54 | ||
55 | try: | |
56 | from urllib.error import HTTPError as compat_HTTPError | |
57 | except ImportError: # Python 2 | |
58 | from urllib2 import HTTPError as compat_HTTPError | |
59 | ||
60 | try: | |
61 | from urllib.request import urlretrieve as compat_urlretrieve | |
62 | except ImportError: # Python 2 | |
63 | from urllib import urlretrieve as compat_urlretrieve | |
64 | ||
65 | ||
66 | try: | |
67 | from subprocess import DEVNULL | |
68 | compat_subprocess_get_DEVNULL = lambda: DEVNULL | |
69 | except ImportError: | |
70 | compat_subprocess_get_DEVNULL = lambda: open(os.path.devnull, 'w') | |
71 | ||
83fda3c0 PH |
72 | try: |
73 | import http.server as compat_http_server | |
74 | except ImportError: | |
75 | import BaseHTTPServer as compat_http_server | |
76 | ||
8c25f81b | 77 | try: |
55139679 | 78 | from urllib.parse import unquote_to_bytes as compat_urllib_parse_unquote_to_bytes |
8c25f81b | 79 | from urllib.parse import unquote as compat_urllib_parse_unquote |
aa99aa4e | 80 | from urllib.parse import unquote_plus as compat_urllib_parse_unquote_plus |
55139679 | 81 | except ImportError: # Python 2 |
4d08161a | 82 | # HACK: The following are the correct unquote_to_bytes, unquote and unquote_plus |
55139679 S |
83 | # implementations from cpython 3.4.3's stdlib. Python 2's version |
84 | # is apparently broken (see https://github.com/rg3/youtube-dl/pull/6244) | |
85 | ||
c9c854ce | 86 | def compat_urllib_parse_unquote_to_bytes(string): |
87 | """unquote_to_bytes('abc%20def') -> b'abc def'.""" | |
88 | # Note: strings are encoded as UTF-8. This is only an issue if it contains | |
89 | # unescaped non-ASCII characters, which URIs should not. | |
90 | if not string: | |
91 | # Is it a string-like object? | |
92 | string.split | |
93 | return b'' | |
55139679 | 94 | if isinstance(string, unicode): |
c9c854ce | 95 | string = string.encode('utf-8') |
55139679 | 96 | bits = string.split(b'%') |
c9c854ce | 97 | if len(bits) == 1: |
98 | return string | |
99 | res = [bits[0]] | |
100 | append = res.append | |
c9c854ce | 101 | for item in bits[1:]: |
102 | try: | |
55139679 | 103 | append(compat_urllib_parse._hextochr[item[:2]]) |
c9c854ce | 104 | append(item[2:]) |
55139679 | 105 | except KeyError: |
c9c854ce | 106 | append(b'%') |
107 | append(item) | |
108 | return b''.join(res) | |
109 | ||
a0f28f90 | 110 | def compat_urllib_parse_unquote(string, encoding='utf-8', errors='replace'): |
c9c854ce | 111 | """Replace %xx escapes by their single-character equivalent. The optional |
112 | encoding and errors parameters specify how to decode percent-encoded | |
113 | sequences into Unicode characters, as accepted by the bytes.decode() | |
114 | method. | |
115 | By default, percent-encoded sequences are decoded with UTF-8, and invalid | |
116 | sequences are replaced by a placeholder character. | |
117 | ||
118 | unquote('abc%20def') -> 'abc def'. | |
119 | """ | |
c9c854ce | 120 | if '%' not in string: |
121 | string.split | |
122 | return string | |
123 | if encoding is None: | |
124 | encoding = 'utf-8' | |
125 | if errors is None: | |
126 | errors = 'replace' | |
55139679 | 127 | bits = compat_urllib_parse._asciire.split(string) |
c9c854ce | 128 | res = [bits[0]] |
129 | append = res.append | |
130 | for i in range(1, len(bits), 2): | |
55139679 S |
131 | append(compat_urllib_parse_unquote_to_bytes(bits[i]).decode(encoding, errors)) |
132 | append(bits[i + 1]) | |
c9c854ce | 133 | return ''.join(res) |
134 | ||
aa99aa4e S |
135 | def compat_urllib_parse_unquote_plus(string, encoding='utf-8', errors='replace'): |
136 | """Like unquote(), but also replace plus signs by spaces, as required for | |
137 | unquoting HTML form values. | |
138 | ||
139 | unquote_plus('%7e/abc+def') -> '~/abc def' | |
140 | """ | |
141 | string = string.replace('+', ' ') | |
142 | return compat_urllib_parse_unquote(string, encoding, errors) | |
143 | ||
8f9312c3 PH |
144 | try: |
145 | compat_str = unicode # Python 2 | |
146 | except NameError: | |
147 | compat_str = str | |
148 | ||
149 | try: | |
0196149c | 150 | compat_basestring = basestring # Python 2 |
8f9312c3 | 151 | except NameError: |
0196149c | 152 | compat_basestring = str |
8f9312c3 PH |
153 | |
154 | try: | |
155 | compat_chr = unichr # Python 2 | |
156 | except NameError: | |
157 | compat_chr = chr | |
158 | ||
159 | try: | |
160 | from xml.etree.ElementTree import ParseError as compat_xml_parse_error | |
161 | except ImportError: # Python 2.6 | |
162 | from xml.parsers.expat import ExpatError as compat_xml_parse_error | |
163 | ||
8c25f81b PH |
164 | |
165 | try: | |
166 | from urllib.parse import parse_qs as compat_parse_qs | |
5f6a1245 | 167 | except ImportError: # Python 2 |
8c25f81b PH |
168 | # HACK: The following is the correct parse_qs implementation from cpython 3's stdlib. |
169 | # Python 2's version is apparently totally broken | |
170 | ||
171 | def _parse_qsl(qs, keep_blank_values=False, strict_parsing=False, | |
9e1a5b84 | 172 | encoding='utf-8', errors='replace'): |
8f9312c3 | 173 | qs, _coerce_result = qs, compat_str |
8c25f81b PH |
174 | pairs = [s2 for s1 in qs.split('&') for s2 in s1.split(';')] |
175 | r = [] | |
176 | for name_value in pairs: | |
177 | if not name_value and not strict_parsing: | |
178 | continue | |
179 | nv = name_value.split('=', 1) | |
180 | if len(nv) != 2: | |
181 | if strict_parsing: | |
182 | raise ValueError("bad query field: %r" % (name_value,)) | |
183 | # Handle case of a control-name with no equal sign | |
184 | if keep_blank_values: | |
185 | nv.append('') | |
186 | else: | |
187 | continue | |
188 | if len(nv[1]) or keep_blank_values: | |
189 | name = nv[0].replace('+', ' ') | |
190 | name = compat_urllib_parse_unquote( | |
191 | name, encoding=encoding, errors=errors) | |
192 | name = _coerce_result(name) | |
193 | value = nv[1].replace('+', ' ') | |
194 | value = compat_urllib_parse_unquote( | |
195 | value, encoding=encoding, errors=errors) | |
196 | value = _coerce_result(value) | |
197 | r.append((name, value)) | |
198 | return r | |
199 | ||
200 | def compat_parse_qs(qs, keep_blank_values=False, strict_parsing=False, | |
9e1a5b84 | 201 | encoding='utf-8', errors='replace'): |
8c25f81b PH |
202 | parsed_result = {} |
203 | pairs = _parse_qsl(qs, keep_blank_values, strict_parsing, | |
9e1a5b84 | 204 | encoding=encoding, errors=errors) |
8c25f81b PH |
205 | for name, value in pairs: |
206 | if name in parsed_result: | |
207 | parsed_result[name].append(value) | |
208 | else: | |
209 | parsed_result[name] = [value] | |
210 | return parsed_result | |
211 | ||
8c25f81b PH |
212 | try: |
213 | from shlex import quote as shlex_quote | |
214 | except ImportError: # Python < 3.3 | |
215 | def shlex_quote(s): | |
7d4111ed PH |
216 | if re.match(r'^[-_\w./]+$', s): |
217 | return s | |
218 | else: | |
219 | return "'" + s.replace("'", "'\"'\"'") + "'" | |
8c25f81b PH |
220 | |
221 | ||
222 | def compat_ord(c): | |
5f6a1245 JW |
223 | if type(c) is int: |
224 | return c | |
225 | else: | |
226 | return ord(c) | |
8c25f81b PH |
227 | |
228 | ||
229 | if sys.version_info >= (3, 0): | |
230 | compat_getenv = os.getenv | |
231 | compat_expanduser = os.path.expanduser | |
232 | else: | |
233 | # Environment variables should be decoded with filesystem encoding. | |
234 | # Otherwise it will fail if any non-ASCII characters present (see #3854 #3217 #2918) | |
235 | ||
236 | def compat_getenv(key, default=None): | |
237 | from .utils import get_filesystem_encoding | |
238 | env = os.getenv(key, default) | |
239 | if env: | |
240 | env = env.decode(get_filesystem_encoding()) | |
241 | return env | |
242 | ||
243 | # HACK: The default implementations of os.path.expanduser from cpython do not decode | |
244 | # environment variables with filesystem encoding. We will work around this by | |
245 | # providing adjusted implementations. | |
246 | # The following are os.path.expanduser implementations from cpython 2.7.8 stdlib | |
247 | # for different platforms with correct environment variables decoding. | |
248 | ||
249 | if os.name == 'posix': | |
250 | def compat_expanduser(path): | |
251 | """Expand ~ and ~user constructions. If user or $HOME is unknown, | |
252 | do nothing.""" | |
253 | if not path.startswith('~'): | |
254 | return path | |
255 | i = path.find('/', 1) | |
256 | if i < 0: | |
257 | i = len(path) | |
258 | if i == 1: | |
259 | if 'HOME' not in os.environ: | |
260 | import pwd | |
261 | userhome = pwd.getpwuid(os.getuid()).pw_dir | |
262 | else: | |
263 | userhome = compat_getenv('HOME') | |
264 | else: | |
265 | import pwd | |
266 | try: | |
267 | pwent = pwd.getpwnam(path[1:i]) | |
268 | except KeyError: | |
269 | return path | |
270 | userhome = pwent.pw_dir | |
271 | userhome = userhome.rstrip('/') | |
272 | return (userhome + path[i:]) or '/' | |
273 | elif os.name == 'nt' or os.name == 'ce': | |
274 | def compat_expanduser(path): | |
275 | """Expand ~ and ~user constructs. | |
276 | ||
277 | If user or $HOME is unknown, do nothing.""" | |
278 | if path[:1] != '~': | |
279 | return path | |
280 | i, n = 1, len(path) | |
281 | while i < n and path[i] not in '/\\': | |
282 | i = i + 1 | |
283 | ||
284 | if 'HOME' in os.environ: | |
285 | userhome = compat_getenv('HOME') | |
286 | elif 'USERPROFILE' in os.environ: | |
287 | userhome = compat_getenv('USERPROFILE') | |
83e865a3 | 288 | elif 'HOMEPATH' not in os.environ: |
8c25f81b PH |
289 | return path |
290 | else: | |
291 | try: | |
292 | drive = compat_getenv('HOMEDRIVE') | |
293 | except KeyError: | |
294 | drive = '' | |
295 | userhome = os.path.join(drive, compat_getenv('HOMEPATH')) | |
296 | ||
5f6a1245 | 297 | if i != 1: # ~user |
8c25f81b PH |
298 | userhome = os.path.join(os.path.dirname(userhome), path[1:i]) |
299 | ||
300 | return userhome + path[i:] | |
301 | else: | |
302 | compat_expanduser = os.path.expanduser | |
303 | ||
304 | ||
305 | if sys.version_info < (3, 0): | |
306 | def compat_print(s): | |
307 | from .utils import preferredencoding | |
308 | print(s.encode(preferredencoding(), 'xmlcharrefreplace')) | |
309 | else: | |
310 | def compat_print(s): | |
b061ea6e | 311 | assert isinstance(s, compat_str) |
8c25f81b PH |
312 | print(s) |
313 | ||
314 | ||
315 | try: | |
316 | subprocess_check_output = subprocess.check_output | |
317 | except AttributeError: | |
318 | def subprocess_check_output(*args, **kwargs): | |
319 | assert 'input' not in kwargs | |
320 | p = subprocess.Popen(*args, stdout=subprocess.PIPE, **kwargs) | |
321 | output, _ = p.communicate() | |
322 | ret = p.poll() | |
323 | if ret: | |
324 | raise subprocess.CalledProcessError(ret, p.args, output=output) | |
325 | return output | |
326 | ||
327 | if sys.version_info < (3, 0) and sys.platform == 'win32': | |
328 | def compat_getpass(prompt, *args, **kwargs): | |
329 | if isinstance(prompt, compat_str): | |
baa70803 | 330 | from .utils import preferredencoding |
8c25f81b PH |
331 | prompt = prompt.encode(preferredencoding()) |
332 | return getpass.getpass(prompt, *args, **kwargs) | |
333 | else: | |
334 | compat_getpass = getpass.getpass | |
335 | ||
c7b0add8 PH |
336 | # Old 2.6 and 2.7 releases require kwargs to be bytes |
337 | try: | |
c6973bd4 PH |
338 | def _testfunc(x): |
339 | pass | |
340 | _testfunc(**{'x': 0}) | |
c7b0add8 PH |
341 | except TypeError: |
342 | def compat_kwargs(kwargs): | |
343 | return dict((bytes(k), v) for k, v in kwargs.items()) | |
344 | else: | |
345 | compat_kwargs = lambda kwargs: kwargs | |
8c25f81b | 346 | |
e07e9313 | 347 | |
be4a824d PH |
348 | if sys.version_info < (2, 7): |
349 | def compat_socket_create_connection(address, timeout, source_address=None): | |
350 | host, port = address | |
351 | err = None | |
352 | for res in socket.getaddrinfo(host, port, 0, socket.SOCK_STREAM): | |
353 | af, socktype, proto, canonname, sa = res | |
354 | sock = None | |
355 | try: | |
356 | sock = socket.socket(af, socktype, proto) | |
357 | sock.settimeout(timeout) | |
358 | if source_address: | |
359 | sock.bind(source_address) | |
360 | sock.connect(sa) | |
361 | return sock | |
362 | except socket.error as _: | |
363 | err = _ | |
364 | if sock is not None: | |
365 | sock.close() | |
366 | if err is not None: | |
367 | raise err | |
368 | else: | |
8ad6b5ed | 369 | raise socket.error("getaddrinfo returns an empty list") |
be4a824d PH |
370 | else: |
371 | compat_socket_create_connection = socket.create_connection | |
372 | ||
373 | ||
e07e9313 PH |
374 | # Fix https://github.com/rg3/youtube-dl/issues/4223 |
375 | # See http://bugs.python.org/issue9161 for what is broken | |
376 | def workaround_optparse_bug9161(): | |
07e378fa PH |
377 | op = optparse.OptionParser() |
378 | og = optparse.OptionGroup(op, 'foo') | |
e07e9313 | 379 | try: |
07e378fa | 380 | og.add_option('-t') |
b244b5c3 | 381 | except TypeError: |
e07e9313 PH |
382 | real_add_option = optparse.OptionGroup.add_option |
383 | ||
384 | def _compat_add_option(self, *args, **kwargs): | |
385 | enc = lambda v: ( | |
386 | v.encode('ascii', 'replace') if isinstance(v, compat_str) | |
387 | else v) | |
388 | bargs = [enc(a) for a in args] | |
389 | bkwargs = dict( | |
390 | (k, enc(v)) for k, v in kwargs.items()) | |
391 | return real_add_option(self, *bargs, **bkwargs) | |
392 | optparse.OptionGroup.add_option = _compat_add_option | |
393 | ||
003c69a8 JMF |
394 | if hasattr(shutil, 'get_terminal_size'): # Python >= 3.3 |
395 | compat_get_terminal_size = shutil.get_terminal_size | |
396 | else: | |
397 | _terminal_size = collections.namedtuple('terminal_size', ['columns', 'lines']) | |
398 | ||
399 | def compat_get_terminal_size(): | |
400 | columns = compat_getenv('COLUMNS', None) | |
401 | if columns: | |
402 | columns = int(columns) | |
403 | else: | |
404 | columns = None | |
405 | lines = compat_getenv('LINES', None) | |
406 | if lines: | |
407 | lines = int(lines) | |
408 | else: | |
409 | lines = None | |
410 | ||
411 | try: | |
412 | sp = subprocess.Popen( | |
413 | ['stty', 'size'], | |
414 | stdout=subprocess.PIPE, stderr=subprocess.PIPE) | |
415 | out, err = sp.communicate() | |
416 | lines, columns = map(int, out.split()) | |
70a1165b | 417 | except Exception: |
003c69a8 JMF |
418 | pass |
419 | return _terminal_size(columns, lines) | |
420 | ||
a0e060ac YCH |
421 | try: |
422 | itertools.count(start=0, step=1) | |
423 | compat_itertools_count = itertools.count | |
424 | except TypeError: # Python 2.6 | |
425 | def compat_itertools_count(start=0, step=1): | |
426 | n = start | |
427 | while True: | |
428 | yield n | |
429 | n += step | |
e07e9313 | 430 | |
8c25f81b PH |
431 | __all__ = [ |
432 | 'compat_HTTPError', | |
0196149c | 433 | 'compat_basestring', |
8c25f81b PH |
434 | 'compat_chr', |
435 | 'compat_cookiejar', | |
436 | 'compat_expanduser', | |
003c69a8 | 437 | 'compat_get_terminal_size', |
8c25f81b PH |
438 | 'compat_getenv', |
439 | 'compat_getpass', | |
440 | 'compat_html_entities', | |
8c25f81b | 441 | 'compat_http_client', |
83fda3c0 | 442 | 'compat_http_server', |
a0e060ac | 443 | 'compat_itertools_count', |
c7b0add8 | 444 | 'compat_kwargs', |
8c25f81b PH |
445 | 'compat_ord', |
446 | 'compat_parse_qs', | |
447 | 'compat_print', | |
be4a824d | 448 | 'compat_socket_create_connection', |
987493ae | 449 | 'compat_str', |
8c25f81b PH |
450 | 'compat_subprocess_get_DEVNULL', |
451 | 'compat_urllib_error', | |
452 | 'compat_urllib_parse', | |
453 | 'compat_urllib_parse_unquote', | |
aa99aa4e | 454 | 'compat_urllib_parse_unquote_plus', |
9fefc886 | 455 | 'compat_urllib_parse_unquote_to_bytes', |
8c25f81b PH |
456 | 'compat_urllib_parse_urlparse', |
457 | 'compat_urllib_request', | |
458 | 'compat_urlparse', | |
459 | 'compat_urlretrieve', | |
460 | 'compat_xml_parse_error', | |
461 | 'shlex_quote', | |
462 | 'subprocess_check_output', | |
e07e9313 | 463 | 'workaround_optparse_bug9161', |
8c25f81b | 464 | ] |