Repository navigation
Expand file tree
/
Copy pathcompat.py
More file actions
313 lines (261 loc) · 9.14 KB
/
Copy pathcompat.py
File metadata and controls
313 lines (261 loc) · 9.14 KB
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
# -*- coding: utf-8 -*-
# config.py
# Copyright (C) 2008, 2009 Michael Trier (mtrier@gmail.com) and contributors
#
# This module is part of GitPython and is released under
# the BSD License: http://www.opensource.org/licenses/bsd-license.php
"""utilities to help provide compatibility with python 3"""
# flake8: noqa
import locale
import os
import sys
import codecs
from gitdb.utils.compat import (
xrange,
MAXSIZE, # @UnusedImport
izip, # @UnusedImport
)
from gitdb.utils.encoding import (
string_types, # @UnusedImport
text_type, # @UnusedImport
force_bytes, # @UnusedImport
force_text # @UnusedImport
)
PY3 = sys.version_info[0] >= 3
is_win = (os.name == 'nt')
is_posix = (os.name == 'posix')
is_darwin = (os.name == 'darwin')
defenc = sys.getdefaultencoding()
if PY3:
import io
FileType = io.IOBase
def byte_ord(b):
return b
def bchr(n):
return bytes([n])
def mviter(d):
return d.values()
range = xrange # @ReservedAssignment
unicode = str
binary_type = bytes
else:
FileType = file # @UndefinedVariable on PY3
# usually, this is just ascii, which might not enough for our encoding needs
# Unless it's set specifically, we override it to be utf-8
if defenc == 'ascii':
defenc = 'utf-8'
byte_ord = ord
bchr = chr
unicode = unicode
binary_type = str
range = xrange # @ReservedAssignment
def mviter(d):
return d.itervalues()
def safe_decode(s):
"""Safely decodes a binary string to unicode"""
if isinstance(s, unicode):
return s
elif isinstance(s, bytes):
return s.decode(defenc, 'surrogateescape')
elif s is not None:
raise TypeError('Expected bytes or text, but got %r' % (s,))
def safe_encode(s):
"""Safely decodes a binary string to unicode"""
if isinstance(s, unicode):
return s.encode(defenc)
elif isinstance(s, bytes):
return s
elif s is not None:
raise TypeError('Expected bytes or text, but got %r' % (s,))
def win_encode(s):
"""Encode unicodes for process arguments on Windows."""
if isinstance(s, unicode):
return s.encode(locale.getpreferredencoding(False))
elif isinstance(s, bytes):
return s
elif s is not None:
raise TypeError('Expected bytes or text, but got %r' % (s,))
def with_metaclass(meta, *bases):
"""copied from https://github.com/Byron/bcore/blob/master/src/python/butility/future.py#L15"""
class metaclass(meta):
__call__ = type.__call__
__init__ = type.__init__
def __new__(cls, name, nbases, d):
if nbases is None:
return type.__new__(cls, name, (), d)
# There may be clients who rely on this attribute to be set to a reasonable value, which is why
# we set the __metaclass__ attribute explicitly
if not PY3 and '___metaclass__' not in d:
d['__metaclass__'] = meta
return meta(name, bases, d)
return metaclass(meta.__name__ + 'Helper', None, {})
## From https://docs.python.org/3.3/howto/pyporting.html
class UnicodeMixin(object):
"""Mixin class to handle defining the proper __str__/__unicode__
methods in Python 2 or 3."""
if PY3:
def __str__(self):
return self.__unicode__()
else: # Python 2
def __str__(self):
return self.__unicode__().encode(defenc)
"""
This is Victor Stinner's pure-Python implementation of PEP 383: the "surrogateescape" error
handler of Python 3.
Source: misc/python/surrogateescape.py in https://bitbucket.org/haypo/misc
"""
# This code is released under the Python license and the BSD 2-clause license
FS_ERRORS = 'surrogateescape'
# # -- Python 2/3 compatibility -------------------------------------
# FS_ERRORS = 'my_surrogateescape'
def u(text):
if PY3:
return text
else:
return text.decode('unicode_escape')
def b(data):
if PY3:
return data.encode('latin1')
else:
return data
if PY3:
_unichr = chr
bytes_chr = lambda code: bytes((code,))
else:
_unichr = unichr
bytes_chr = chr
def surrogateescape_handler(exc):
"""
Pure Python implementation of the PEP 383: the "surrogateescape" error
handler of Python 3. Undecodable bytes will be replaced by a Unicode
character U+DCxx on decoding, and these are translated into the
original bytes on encoding.
"""
mystring = exc.object[exc.start:exc.end]
try:
if isinstance(exc, UnicodeDecodeError):
# mystring is a byte-string in this case
decoded = replace_surrogate_decode(mystring)
elif isinstance(exc, UnicodeEncodeError):
# In the case of u'\udcc3'.encode('ascii',
# 'this_surrogateescape_handler'), both Python 2.x and 3.x raise an
# exception anyway after this function is called, even though I think
# it's doing what it should. It seems that the strict encoder is called
# to encode the unicode string that this function returns ...
decoded = replace_surrogate_encode(mystring, exc)
else:
raise exc
except NotASurrogateError:
raise exc
return (decoded, exc.end)
class NotASurrogateError(Exception):
pass
def replace_surrogate_encode(mystring, exc):
"""
Returns a (unicode) string, not the more logical bytes, because the codecs
register_error functionality expects this.
"""
decoded = []
for ch in mystring:
# if PY3:
# code = ch
# else:
code = ord(ch)
# The following magic comes from Py3.3's Python/codecs.c file:
if not 0xD800 <= code <= 0xDCFF:
# Not a surrogate. Fail with the original exception.
raise exc
# mybytes = [0xe0 | (code >> 12),
# 0x80 | ((code >> 6) & 0x3f),
# 0x80 | (code & 0x3f)]
# Is this a good idea?
if 0xDC00 <= code <= 0xDC7F:
decoded.append(_unichr(code - 0xDC00))
elif code <= 0xDCFF:
decoded.append(_unichr(code - 0xDC00))
else:
raise NotASurrogateError
return str().join(decoded)
def replace_surrogate_decode(mybytes):
"""
Returns a (unicode) string
"""
decoded = []
for ch in mybytes:
# We may be parsing newbytes (in which case ch is an int) or a native
# str on Py2
if isinstance(ch, int):
code = ch
else:
code = ord(ch)
if 0x80 <= code <= 0xFF:
decoded.append(_unichr(0xDC00 + code))
elif code <= 0x7F:
decoded.append(_unichr(code))
else:
# # It may be a bad byte
# # Try swallowing it.
# continue
# print("RAISE!")
raise NotASurrogateError
return str().join(decoded)
def encodefilename(fn):
if FS_ENCODING == 'ascii':
# ASCII encoder of Python 2 expects that the error handler returns a
# Unicode string encodable to ASCII, whereas our surrogateescape error
# handler has to return bytes in 0x80-0xFF range.
encoded = []
for index, ch in enumerate(fn):
code = ord(ch)
if code < 128:
ch = bytes_chr(code)
elif 0xDC80 <= code <= 0xDCFF:
ch = bytes_chr(code - 0xDC00)
else:
raise UnicodeEncodeError(FS_ENCODING,
fn, index, index+1,
'ordinal not in range(128)')
encoded.append(ch)
return bytes().join(encoded)
elif FS_ENCODING == 'utf-8':
# UTF-8 encoder of Python 2 encodes surrogates, so U+DC80-U+DCFF
# doesn't go through our error handler
encoded = []
for index, ch in enumerate(fn):
code = ord(ch)
if 0xD800 <= code <= 0xDFFF:
if 0xDC80 <= code <= 0xDCFF:
ch = bytes_chr(code - 0xDC00)
encoded.append(ch)
else:
raise UnicodeEncodeError(
FS_ENCODING,
fn, index, index+1, 'surrogates not allowed')
else:
ch_utf8 = ch.encode('utf-8')
encoded.append(ch_utf8)
return bytes().join(encoded)
else:
return fn.encode(FS_ENCODING, FS_ERRORS)
def decodefilename(fn):
return fn.decode(FS_ENCODING, FS_ERRORS)
FS_ENCODING = 'ascii'; fn = b('[abc\xff]'); encoded = u('[abc\udcff]')
# FS_ENCODING = 'cp932'; fn = b('[abc\x81\x00]'); encoded = u('[abc\udc81\x00]')
# FS_ENCODING = 'UTF-8'; fn = b('[abc\xff]'); encoded = u('[abc\udcff]')
# normalize the filesystem encoding name.
# For example, we expect "utf-8", not "UTF8".
FS_ENCODING = codecs.lookup(FS_ENCODING).name
def register_surrogateescape():
"""
Registers the surrogateescape error handler on Python 2 (only)
"""
if PY3:
return
try:
codecs.lookup_error(FS_ERRORS)
except LookupError:
codecs.register_error(FS_ERRORS, surrogateescape_handler)
try:
b"100644 \x9f\0aaa".decode(defenc, "surrogateescape")
except:
register_surrogateescape()