Mercurial > hg
annotate mercurial/encoding.py @ 28825:87c6ad2251d8
date: reallow negative timestamp, fix for Windows buggy gmtime() (issue2513)
DVCS are very useful to store various texts (as legislation) written before
Unix epoch. Fri, 13 Dec 1901 is a nice gain over Thu, 01 Jan 1970.
Revert dd24f3e7ca9e and e1002cf9fe54, fix c208dcd0f709. Add tests.
author | Florent Gallaire <fgallaire@gmail.com> |
---|---|
date | Fri, 08 Apr 2016 14:11:03 +0200 |
parents | 3c6e94d0811c |
children | 0f6d6fdd3c2a |
rev | line source |
---|---|
8226
8b2cd04a6e97
put license and copyright info into comment blocks
Martin Geisler <mg@lazybytes.net>
parents:
8225
diff
changeset
|
1 # encoding.py - character transcoding support for Mercurial |
8b2cd04a6e97
put license and copyright info into comment blocks
Martin Geisler <mg@lazybytes.net>
parents:
8225
diff
changeset
|
2 # |
8b2cd04a6e97
put license and copyright info into comment blocks
Martin Geisler <mg@lazybytes.net>
parents:
8225
diff
changeset
|
3 # Copyright 2005-2009 Matt Mackall <mpm@selenic.com> and others |
8b2cd04a6e97
put license and copyright info into comment blocks
Martin Geisler <mg@lazybytes.net>
parents:
8225
diff
changeset
|
4 # |
8b2cd04a6e97
put license and copyright info into comment blocks
Martin Geisler <mg@lazybytes.net>
parents:
8225
diff
changeset
|
5 # This software may be used and distributed according to the terms of the |
10263 | 6 # GNU General Public License version 2 or any later version. |
7948
de377b1a9a84
move encoding bits from util to encoding
Matt Mackall <mpm@selenic.com>
parents:
diff
changeset
|
7 |
27355
b479fc425a81
encoding: use absolute_import
Gregory Szorc <gregory.szorc@gmail.com>
parents:
26963
diff
changeset
|
8 from __future__ import absolute_import |
b479fc425a81
encoding: use absolute_import
Gregory Szorc <gregory.szorc@gmail.com>
parents:
26963
diff
changeset
|
9 |
28068
9ece901f7a19
encoding: add option to escape non-ascii characters in JSON
Yuya Nishihara <yuya@tcha.org>
parents:
28067
diff
changeset
|
10 import array |
27355
b479fc425a81
encoding: use absolute_import
Gregory Szorc <gregory.szorc@gmail.com>
parents:
26963
diff
changeset
|
11 import locale |
b479fc425a81
encoding: use absolute_import
Gregory Szorc <gregory.szorc@gmail.com>
parents:
26963
diff
changeset
|
12 import os |
28507
9bcbd9412225
encoding: make HFS+ ignore code Python 3 compatible
Gregory Szorc <gregory.szorc@gmail.com>
parents:
28069
diff
changeset
|
13 import sys |
27355
b479fc425a81
encoding: use absolute_import
Gregory Szorc <gregory.szorc@gmail.com>
parents:
26963
diff
changeset
|
14 import unicodedata |
b479fc425a81
encoding: use absolute_import
Gregory Szorc <gregory.szorc@gmail.com>
parents:
26963
diff
changeset
|
15 |
b479fc425a81
encoding: use absolute_import
Gregory Szorc <gregory.szorc@gmail.com>
parents:
26963
diff
changeset
|
16 from . import ( |
b479fc425a81
encoding: use absolute_import
Gregory Szorc <gregory.szorc@gmail.com>
parents:
26963
diff
changeset
|
17 error, |
b479fc425a81
encoding: use absolute_import
Gregory Szorc <gregory.szorc@gmail.com>
parents:
26963
diff
changeset
|
18 ) |
7948
de377b1a9a84
move encoding bits from util to encoding
Matt Mackall <mpm@selenic.com>
parents:
diff
changeset
|
19 |
28507
9bcbd9412225
encoding: make HFS+ ignore code Python 3 compatible
Gregory Szorc <gregory.szorc@gmail.com>
parents:
28069
diff
changeset
|
20 if sys.version_info[0] >= 3: |
9bcbd9412225
encoding: make HFS+ ignore code Python 3 compatible
Gregory Szorc <gregory.szorc@gmail.com>
parents:
28069
diff
changeset
|
21 unichr = chr |
9bcbd9412225
encoding: make HFS+ ignore code Python 3 compatible
Gregory Szorc <gregory.szorc@gmail.com>
parents:
28069
diff
changeset
|
22 |
23596
885bd7c5c7e3
encoding: add hfsignoreclean to clean out HFS-ignored characters
Augie Fackler <raf@durin42.com>
parents:
22973
diff
changeset
|
23 # These unicode characters are ignored by HFS+ (Apple Technote 1150, |
885bd7c5c7e3
encoding: add hfsignoreclean to clean out HFS-ignored characters
Augie Fackler <raf@durin42.com>
parents:
22973
diff
changeset
|
24 # "Unicode Subtleties"), so we need to ignore them in some places for |
885bd7c5c7e3
encoding: add hfsignoreclean to clean out HFS-ignored characters
Augie Fackler <raf@durin42.com>
parents:
22973
diff
changeset
|
25 # sanity. |
885bd7c5c7e3
encoding: add hfsignoreclean to clean out HFS-ignored characters
Augie Fackler <raf@durin42.com>
parents:
22973
diff
changeset
|
26 _ignore = [unichr(int(x, 16)).encode("utf-8") for x in |
885bd7c5c7e3
encoding: add hfsignoreclean to clean out HFS-ignored characters
Augie Fackler <raf@durin42.com>
parents:
22973
diff
changeset
|
27 "200c 200d 200e 200f 202a 202b 202c 202d 202e " |
885bd7c5c7e3
encoding: add hfsignoreclean to clean out HFS-ignored characters
Augie Fackler <raf@durin42.com>
parents:
22973
diff
changeset
|
28 "206a 206b 206c 206d 206e 206f feff".split()] |
885bd7c5c7e3
encoding: add hfsignoreclean to clean out HFS-ignored characters
Augie Fackler <raf@durin42.com>
parents:
22973
diff
changeset
|
29 # verify the next function will work |
28507
9bcbd9412225
encoding: make HFS+ ignore code Python 3 compatible
Gregory Szorc <gregory.szorc@gmail.com>
parents:
28069
diff
changeset
|
30 if sys.version_info[0] >= 3: |
9bcbd9412225
encoding: make HFS+ ignore code Python 3 compatible
Gregory Szorc <gregory.szorc@gmail.com>
parents:
28069
diff
changeset
|
31 assert set(i[0] for i in _ignore) == set([ord(b'\xe2'), ord(b'\xef')]) |
9bcbd9412225
encoding: make HFS+ ignore code Python 3 compatible
Gregory Szorc <gregory.szorc@gmail.com>
parents:
28069
diff
changeset
|
32 else: |
9bcbd9412225
encoding: make HFS+ ignore code Python 3 compatible
Gregory Szorc <gregory.szorc@gmail.com>
parents:
28069
diff
changeset
|
33 assert set(i[0] for i in _ignore) == set(["\xe2", "\xef"]) |
23596
885bd7c5c7e3
encoding: add hfsignoreclean to clean out HFS-ignored characters
Augie Fackler <raf@durin42.com>
parents:
22973
diff
changeset
|
34 |
885bd7c5c7e3
encoding: add hfsignoreclean to clean out HFS-ignored characters
Augie Fackler <raf@durin42.com>
parents:
22973
diff
changeset
|
35 def hfsignoreclean(s): |
885bd7c5c7e3
encoding: add hfsignoreclean to clean out HFS-ignored characters
Augie Fackler <raf@durin42.com>
parents:
22973
diff
changeset
|
36 """Remove codepoints ignored by HFS+ from s. |
885bd7c5c7e3
encoding: add hfsignoreclean to clean out HFS-ignored characters
Augie Fackler <raf@durin42.com>
parents:
22973
diff
changeset
|
37 |
885bd7c5c7e3
encoding: add hfsignoreclean to clean out HFS-ignored characters
Augie Fackler <raf@durin42.com>
parents:
22973
diff
changeset
|
38 >>> hfsignoreclean(u'.h\u200cg'.encode('utf-8')) |
885bd7c5c7e3
encoding: add hfsignoreclean to clean out HFS-ignored characters
Augie Fackler <raf@durin42.com>
parents:
22973
diff
changeset
|
39 '.hg' |
885bd7c5c7e3
encoding: add hfsignoreclean to clean out HFS-ignored characters
Augie Fackler <raf@durin42.com>
parents:
22973
diff
changeset
|
40 >>> hfsignoreclean(u'.h\ufeffg'.encode('utf-8')) |
885bd7c5c7e3
encoding: add hfsignoreclean to clean out HFS-ignored characters
Augie Fackler <raf@durin42.com>
parents:
22973
diff
changeset
|
41 '.hg' |
885bd7c5c7e3
encoding: add hfsignoreclean to clean out HFS-ignored characters
Augie Fackler <raf@durin42.com>
parents:
22973
diff
changeset
|
42 """ |
885bd7c5c7e3
encoding: add hfsignoreclean to clean out HFS-ignored characters
Augie Fackler <raf@durin42.com>
parents:
22973
diff
changeset
|
43 if "\xe2" in s or "\xef" in s: |
885bd7c5c7e3
encoding: add hfsignoreclean to clean out HFS-ignored characters
Augie Fackler <raf@durin42.com>
parents:
22973
diff
changeset
|
44 for c in _ignore: |
885bd7c5c7e3
encoding: add hfsignoreclean to clean out HFS-ignored characters
Augie Fackler <raf@durin42.com>
parents:
22973
diff
changeset
|
45 s = s.replace(c, '') |
885bd7c5c7e3
encoding: add hfsignoreclean to clean out HFS-ignored characters
Augie Fackler <raf@durin42.com>
parents:
22973
diff
changeset
|
46 return s |
885bd7c5c7e3
encoding: add hfsignoreclean to clean out HFS-ignored characters
Augie Fackler <raf@durin42.com>
parents:
22973
diff
changeset
|
47 |
11892
2be70ca17311
encoding: improve handling of buggy getpreferredencoding() on Mac OS X
Dan Villiom Podlaski Christiansen <danchr@gmail.com>
parents:
11297
diff
changeset
|
48 def _getpreferredencoding(): |
2be70ca17311
encoding: improve handling of buggy getpreferredencoding() on Mac OS X
Dan Villiom Podlaski Christiansen <danchr@gmail.com>
parents:
11297
diff
changeset
|
49 ''' |
2be70ca17311
encoding: improve handling of buggy getpreferredencoding() on Mac OS X
Dan Villiom Podlaski Christiansen <danchr@gmail.com>
parents:
11297
diff
changeset
|
50 On darwin, getpreferredencoding ignores the locale environment and |
2be70ca17311
encoding: improve handling of buggy getpreferredencoding() on Mac OS X
Dan Villiom Podlaski Christiansen <danchr@gmail.com>
parents:
11297
diff
changeset
|
51 always returns mac-roman. http://bugs.python.org/issue6202 fixes this |
2be70ca17311
encoding: improve handling of buggy getpreferredencoding() on Mac OS X
Dan Villiom Podlaski Christiansen <danchr@gmail.com>
parents:
11297
diff
changeset
|
52 for Python 2.7 and up. This is the same corrected code for earlier |
2be70ca17311
encoding: improve handling of buggy getpreferredencoding() on Mac OS X
Dan Villiom Podlaski Christiansen <danchr@gmail.com>
parents:
11297
diff
changeset
|
53 Python versions. |
2be70ca17311
encoding: improve handling of buggy getpreferredencoding() on Mac OS X
Dan Villiom Podlaski Christiansen <danchr@gmail.com>
parents:
11297
diff
changeset
|
54 |
12770
614f0d8724ab
check-code: find trailing whitespace
Martin Geisler <mg@lazybytes.net>
parents:
12062
diff
changeset
|
55 However, we can't use a version check for this method, as some distributions |
11892
2be70ca17311
encoding: improve handling of buggy getpreferredencoding() on Mac OS X
Dan Villiom Podlaski Christiansen <danchr@gmail.com>
parents:
11297
diff
changeset
|
56 patch Python to fix this. Instead, we use it as a 'fixer' for the mac-roman |
2be70ca17311
encoding: improve handling of buggy getpreferredencoding() on Mac OS X
Dan Villiom Podlaski Christiansen <danchr@gmail.com>
parents:
11297
diff
changeset
|
57 encoding, as it is unlikely that this encoding is the actually expected. |
2be70ca17311
encoding: improve handling of buggy getpreferredencoding() on Mac OS X
Dan Villiom Podlaski Christiansen <danchr@gmail.com>
parents:
11297
diff
changeset
|
58 ''' |
2be70ca17311
encoding: improve handling of buggy getpreferredencoding() on Mac OS X
Dan Villiom Podlaski Christiansen <danchr@gmail.com>
parents:
11297
diff
changeset
|
59 try: |
2be70ca17311
encoding: improve handling of buggy getpreferredencoding() on Mac OS X
Dan Villiom Podlaski Christiansen <danchr@gmail.com>
parents:
11297
diff
changeset
|
60 locale.CODESET |
2be70ca17311
encoding: improve handling of buggy getpreferredencoding() on Mac OS X
Dan Villiom Podlaski Christiansen <danchr@gmail.com>
parents:
11297
diff
changeset
|
61 except AttributeError: |
2be70ca17311
encoding: improve handling of buggy getpreferredencoding() on Mac OS X
Dan Villiom Podlaski Christiansen <danchr@gmail.com>
parents:
11297
diff
changeset
|
62 # Fall back to parsing environment variables :-( |
2be70ca17311
encoding: improve handling of buggy getpreferredencoding() on Mac OS X
Dan Villiom Podlaski Christiansen <danchr@gmail.com>
parents:
11297
diff
changeset
|
63 return locale.getdefaultlocale()[1] |
2be70ca17311
encoding: improve handling of buggy getpreferredencoding() on Mac OS X
Dan Villiom Podlaski Christiansen <danchr@gmail.com>
parents:
11297
diff
changeset
|
64 |
2be70ca17311
encoding: improve handling of buggy getpreferredencoding() on Mac OS X
Dan Villiom Podlaski Christiansen <danchr@gmail.com>
parents:
11297
diff
changeset
|
65 oldloc = locale.setlocale(locale.LC_CTYPE) |
2be70ca17311
encoding: improve handling of buggy getpreferredencoding() on Mac OS X
Dan Villiom Podlaski Christiansen <danchr@gmail.com>
parents:
11297
diff
changeset
|
66 locale.setlocale(locale.LC_CTYPE, "") |
2be70ca17311
encoding: improve handling of buggy getpreferredencoding() on Mac OS X
Dan Villiom Podlaski Christiansen <danchr@gmail.com>
parents:
11297
diff
changeset
|
67 result = locale.nl_langinfo(locale.CODESET) |
2be70ca17311
encoding: improve handling of buggy getpreferredencoding() on Mac OS X
Dan Villiom Podlaski Christiansen <danchr@gmail.com>
parents:
11297
diff
changeset
|
68 locale.setlocale(locale.LC_CTYPE, oldloc) |
2be70ca17311
encoding: improve handling of buggy getpreferredencoding() on Mac OS X
Dan Villiom Podlaski Christiansen <danchr@gmail.com>
parents:
11297
diff
changeset
|
69 |
2be70ca17311
encoding: improve handling of buggy getpreferredencoding() on Mac OS X
Dan Villiom Podlaski Christiansen <danchr@gmail.com>
parents:
11297
diff
changeset
|
70 return result |
2be70ca17311
encoding: improve handling of buggy getpreferredencoding() on Mac OS X
Dan Villiom Podlaski Christiansen <danchr@gmail.com>
parents:
11297
diff
changeset
|
71 |
2be70ca17311
encoding: improve handling of buggy getpreferredencoding() on Mac OS X
Dan Villiom Podlaski Christiansen <danchr@gmail.com>
parents:
11297
diff
changeset
|
72 _encodingfixers = { |
2be70ca17311
encoding: improve handling of buggy getpreferredencoding() on Mac OS X
Dan Villiom Podlaski Christiansen <danchr@gmail.com>
parents:
11297
diff
changeset
|
73 '646': lambda: 'ascii', |
2be70ca17311
encoding: improve handling of buggy getpreferredencoding() on Mac OS X
Dan Villiom Podlaski Christiansen <danchr@gmail.com>
parents:
11297
diff
changeset
|
74 'ANSI_X3.4-1968': lambda: 'ascii', |
2be70ca17311
encoding: improve handling of buggy getpreferredencoding() on Mac OS X
Dan Villiom Podlaski Christiansen <danchr@gmail.com>
parents:
11297
diff
changeset
|
75 'mac-roman': _getpreferredencoding |
2be70ca17311
encoding: improve handling of buggy getpreferredencoding() on Mac OS X
Dan Villiom Podlaski Christiansen <danchr@gmail.com>
parents:
11297
diff
changeset
|
76 } |
7948
de377b1a9a84
move encoding bits from util to encoding
Matt Mackall <mpm@selenic.com>
parents:
diff
changeset
|
77 |
de377b1a9a84
move encoding bits from util to encoding
Matt Mackall <mpm@selenic.com>
parents:
diff
changeset
|
78 try: |
de377b1a9a84
move encoding bits from util to encoding
Matt Mackall <mpm@selenic.com>
parents:
diff
changeset
|
79 encoding = os.environ.get("HGENCODING") |
de377b1a9a84
move encoding bits from util to encoding
Matt Mackall <mpm@selenic.com>
parents:
diff
changeset
|
80 if not encoding: |
de377b1a9a84
move encoding bits from util to encoding
Matt Mackall <mpm@selenic.com>
parents:
diff
changeset
|
81 encoding = locale.getpreferredencoding() or 'ascii' |
11892
2be70ca17311
encoding: improve handling of buggy getpreferredencoding() on Mac OS X
Dan Villiom Podlaski Christiansen <danchr@gmail.com>
parents:
11297
diff
changeset
|
82 encoding = _encodingfixers.get(encoding, lambda: encoding)() |
7948
de377b1a9a84
move encoding bits from util to encoding
Matt Mackall <mpm@selenic.com>
parents:
diff
changeset
|
83 except locale.Error: |
de377b1a9a84
move encoding bits from util to encoding
Matt Mackall <mpm@selenic.com>
parents:
diff
changeset
|
84 encoding = 'ascii' |
de377b1a9a84
move encoding bits from util to encoding
Matt Mackall <mpm@selenic.com>
parents:
diff
changeset
|
85 encodingmode = os.environ.get("HGENCODINGMODE", "strict") |
de377b1a9a84
move encoding bits from util to encoding
Matt Mackall <mpm@selenic.com>
parents:
diff
changeset
|
86 fallbackencoding = 'ISO-8859-1' |
de377b1a9a84
move encoding bits from util to encoding
Matt Mackall <mpm@selenic.com>
parents:
diff
changeset
|
87 |
13046
7cc4263e07a9
encoding: add localstr class to track UTF-8 version of transcoded strings
Matt Mackall <mpm@selenic.com>
parents:
12866
diff
changeset
|
88 class localstr(str): |
7cc4263e07a9
encoding: add localstr class to track UTF-8 version of transcoded strings
Matt Mackall <mpm@selenic.com>
parents:
12866
diff
changeset
|
89 '''This class allows strings that are unmodified to be |
7cc4263e07a9
encoding: add localstr class to track UTF-8 version of transcoded strings
Matt Mackall <mpm@selenic.com>
parents:
12866
diff
changeset
|
90 round-tripped to the local encoding and back''' |
7cc4263e07a9
encoding: add localstr class to track UTF-8 version of transcoded strings
Matt Mackall <mpm@selenic.com>
parents:
12866
diff
changeset
|
91 def __new__(cls, u, l): |
7cc4263e07a9
encoding: add localstr class to track UTF-8 version of transcoded strings
Matt Mackall <mpm@selenic.com>
parents:
12866
diff
changeset
|
92 s = str.__new__(cls, l) |
7cc4263e07a9
encoding: add localstr class to track UTF-8 version of transcoded strings
Matt Mackall <mpm@selenic.com>
parents:
12866
diff
changeset
|
93 s._utf8 = u |
7cc4263e07a9
encoding: add localstr class to track UTF-8 version of transcoded strings
Matt Mackall <mpm@selenic.com>
parents:
12866
diff
changeset
|
94 return s |
7cc4263e07a9
encoding: add localstr class to track UTF-8 version of transcoded strings
Matt Mackall <mpm@selenic.com>
parents:
12866
diff
changeset
|
95 def __hash__(self): |
7cc4263e07a9
encoding: add localstr class to track UTF-8 version of transcoded strings
Matt Mackall <mpm@selenic.com>
parents:
12866
diff
changeset
|
96 return hash(self._utf8) # avoid collisions in local string space |
7cc4263e07a9
encoding: add localstr class to track UTF-8 version of transcoded strings
Matt Mackall <mpm@selenic.com>
parents:
12866
diff
changeset
|
97 |
7948
de377b1a9a84
move encoding bits from util to encoding
Matt Mackall <mpm@selenic.com>
parents:
diff
changeset
|
98 def tolocal(s): |
de377b1a9a84
move encoding bits from util to encoding
Matt Mackall <mpm@selenic.com>
parents:
diff
changeset
|
99 """ |
de377b1a9a84
move encoding bits from util to encoding
Matt Mackall <mpm@selenic.com>
parents:
diff
changeset
|
100 Convert a string from internal UTF-8 to local encoding |
de377b1a9a84
move encoding bits from util to encoding
Matt Mackall <mpm@selenic.com>
parents:
diff
changeset
|
101 |
de377b1a9a84
move encoding bits from util to encoding
Matt Mackall <mpm@selenic.com>
parents:
diff
changeset
|
102 All internal strings should be UTF-8 but some repos before the |
de377b1a9a84
move encoding bits from util to encoding
Matt Mackall <mpm@selenic.com>
parents:
diff
changeset
|
103 implementation of locale support may contain latin1 or possibly |
de377b1a9a84
move encoding bits from util to encoding
Matt Mackall <mpm@selenic.com>
parents:
diff
changeset
|
104 other character sets. We attempt to decode everything strictly |
de377b1a9a84
move encoding bits from util to encoding
Matt Mackall <mpm@selenic.com>
parents:
diff
changeset
|
105 using UTF-8, then Latin-1, and failing that, we use UTF-8 and |
de377b1a9a84
move encoding bits from util to encoding
Matt Mackall <mpm@selenic.com>
parents:
diff
changeset
|
106 replace unknown characters. |
13046
7cc4263e07a9
encoding: add localstr class to track UTF-8 version of transcoded strings
Matt Mackall <mpm@selenic.com>
parents:
12866
diff
changeset
|
107 |
7cc4263e07a9
encoding: add localstr class to track UTF-8 version of transcoded strings
Matt Mackall <mpm@selenic.com>
parents:
12866
diff
changeset
|
108 The localstr class is used to cache the known UTF-8 encoding of |
7cc4263e07a9
encoding: add localstr class to track UTF-8 version of transcoded strings
Matt Mackall <mpm@selenic.com>
parents:
12866
diff
changeset
|
109 strings next to their local representation to allow lossless |
7cc4263e07a9
encoding: add localstr class to track UTF-8 version of transcoded strings
Matt Mackall <mpm@selenic.com>
parents:
12866
diff
changeset
|
110 round-trip conversion back to UTF-8. |
7cc4263e07a9
encoding: add localstr class to track UTF-8 version of transcoded strings
Matt Mackall <mpm@selenic.com>
parents:
12866
diff
changeset
|
111 |
7cc4263e07a9
encoding: add localstr class to track UTF-8 version of transcoded strings
Matt Mackall <mpm@selenic.com>
parents:
12866
diff
changeset
|
112 >>> u = 'foo: \\xc3\\xa4' # utf-8 |
7cc4263e07a9
encoding: add localstr class to track UTF-8 version of transcoded strings
Matt Mackall <mpm@selenic.com>
parents:
12866
diff
changeset
|
113 >>> l = tolocal(u) |
7cc4263e07a9
encoding: add localstr class to track UTF-8 version of transcoded strings
Matt Mackall <mpm@selenic.com>
parents:
12866
diff
changeset
|
114 >>> l |
7cc4263e07a9
encoding: add localstr class to track UTF-8 version of transcoded strings
Matt Mackall <mpm@selenic.com>
parents:
12866
diff
changeset
|
115 'foo: ?' |
7cc4263e07a9
encoding: add localstr class to track UTF-8 version of transcoded strings
Matt Mackall <mpm@selenic.com>
parents:
12866
diff
changeset
|
116 >>> fromlocal(l) |
7cc4263e07a9
encoding: add localstr class to track UTF-8 version of transcoded strings
Matt Mackall <mpm@selenic.com>
parents:
12866
diff
changeset
|
117 'foo: \\xc3\\xa4' |
7cc4263e07a9
encoding: add localstr class to track UTF-8 version of transcoded strings
Matt Mackall <mpm@selenic.com>
parents:
12866
diff
changeset
|
118 >>> u2 = 'foo: \\xc3\\xa1' |
7cc4263e07a9
encoding: add localstr class to track UTF-8 version of transcoded strings
Matt Mackall <mpm@selenic.com>
parents:
12866
diff
changeset
|
119 >>> d = { l: 1, tolocal(u2): 2 } |
18378
404feac78b8a
tests: stabilize doctest output
Mads Kiilerich <mads@kiilerich.com>
parents:
17424
diff
changeset
|
120 >>> len(d) # no collision |
404feac78b8a
tests: stabilize doctest output
Mads Kiilerich <mads@kiilerich.com>
parents:
17424
diff
changeset
|
121 2 |
13046
7cc4263e07a9
encoding: add localstr class to track UTF-8 version of transcoded strings
Matt Mackall <mpm@selenic.com>
parents:
12866
diff
changeset
|
122 >>> 'foo: ?' in d |
7cc4263e07a9
encoding: add localstr class to track UTF-8 version of transcoded strings
Matt Mackall <mpm@selenic.com>
parents:
12866
diff
changeset
|
123 False |
7cc4263e07a9
encoding: add localstr class to track UTF-8 version of transcoded strings
Matt Mackall <mpm@selenic.com>
parents:
12866
diff
changeset
|
124 >>> l1 = 'foo: \\xe4' # historical latin1 fallback |
7cc4263e07a9
encoding: add localstr class to track UTF-8 version of transcoded strings
Matt Mackall <mpm@selenic.com>
parents:
12866
diff
changeset
|
125 >>> l = tolocal(l1) |
7cc4263e07a9
encoding: add localstr class to track UTF-8 version of transcoded strings
Matt Mackall <mpm@selenic.com>
parents:
12866
diff
changeset
|
126 >>> l |
7cc4263e07a9
encoding: add localstr class to track UTF-8 version of transcoded strings
Matt Mackall <mpm@selenic.com>
parents:
12866
diff
changeset
|
127 'foo: ?' |
7cc4263e07a9
encoding: add localstr class to track UTF-8 version of transcoded strings
Matt Mackall <mpm@selenic.com>
parents:
12866
diff
changeset
|
128 >>> fromlocal(l) # magically in utf-8 |
7cc4263e07a9
encoding: add localstr class to track UTF-8 version of transcoded strings
Matt Mackall <mpm@selenic.com>
parents:
12866
diff
changeset
|
129 'foo: \\xc3\\xa4' |
7948
de377b1a9a84
move encoding bits from util to encoding
Matt Mackall <mpm@selenic.com>
parents:
diff
changeset
|
130 """ |
13046
7cc4263e07a9
encoding: add localstr class to track UTF-8 version of transcoded strings
Matt Mackall <mpm@selenic.com>
parents:
12866
diff
changeset
|
131 |
16274
5d75eb8568d1
encoding: tune fast-path of tolocal a bit
Matt Mackall <mpm@selenic.com>
parents:
16133
diff
changeset
|
132 try: |
7948
de377b1a9a84
move encoding bits from util to encoding
Matt Mackall <mpm@selenic.com>
parents:
diff
changeset
|
133 try: |
16274
5d75eb8568d1
encoding: tune fast-path of tolocal a bit
Matt Mackall <mpm@selenic.com>
parents:
16133
diff
changeset
|
134 # make sure string is actually stored in UTF-8 |
5d75eb8568d1
encoding: tune fast-path of tolocal a bit
Matt Mackall <mpm@selenic.com>
parents:
16133
diff
changeset
|
135 u = s.decode('UTF-8') |
5d75eb8568d1
encoding: tune fast-path of tolocal a bit
Matt Mackall <mpm@selenic.com>
parents:
16133
diff
changeset
|
136 if encoding == 'UTF-8': |
5d75eb8568d1
encoding: tune fast-path of tolocal a bit
Matt Mackall <mpm@selenic.com>
parents:
16133
diff
changeset
|
137 # fast path |
5d75eb8568d1
encoding: tune fast-path of tolocal a bit
Matt Mackall <mpm@selenic.com>
parents:
16133
diff
changeset
|
138 return s |
13940
b7b26e54e37a
encoding: avoid localstr when a string can be encoded losslessly (issue2763)
Matt Mackall <mpm@selenic.com>
parents:
13051
diff
changeset
|
139 r = u.encode(encoding, "replace") |
b7b26e54e37a
encoding: avoid localstr when a string can be encoded losslessly (issue2763)
Matt Mackall <mpm@selenic.com>
parents:
13051
diff
changeset
|
140 if u == r.decode(encoding): |
b7b26e54e37a
encoding: avoid localstr when a string can be encoded losslessly (issue2763)
Matt Mackall <mpm@selenic.com>
parents:
13051
diff
changeset
|
141 # r is a safe, non-lossy encoding of s |
b7b26e54e37a
encoding: avoid localstr when a string can be encoded losslessly (issue2763)
Matt Mackall <mpm@selenic.com>
parents:
13051
diff
changeset
|
142 return r |
16274
5d75eb8568d1
encoding: tune fast-path of tolocal a bit
Matt Mackall <mpm@selenic.com>
parents:
16133
diff
changeset
|
143 return localstr(s, r) |
5d75eb8568d1
encoding: tune fast-path of tolocal a bit
Matt Mackall <mpm@selenic.com>
parents:
16133
diff
changeset
|
144 except UnicodeDecodeError: |
5d75eb8568d1
encoding: tune fast-path of tolocal a bit
Matt Mackall <mpm@selenic.com>
parents:
16133
diff
changeset
|
145 # we should only get here if we're looking at an ancient changeset |
5d75eb8568d1
encoding: tune fast-path of tolocal a bit
Matt Mackall <mpm@selenic.com>
parents:
16133
diff
changeset
|
146 try: |
5d75eb8568d1
encoding: tune fast-path of tolocal a bit
Matt Mackall <mpm@selenic.com>
parents:
16133
diff
changeset
|
147 u = s.decode(fallbackencoding) |
5d75eb8568d1
encoding: tune fast-path of tolocal a bit
Matt Mackall <mpm@selenic.com>
parents:
16133
diff
changeset
|
148 r = u.encode(encoding, "replace") |
5d75eb8568d1
encoding: tune fast-path of tolocal a bit
Matt Mackall <mpm@selenic.com>
parents:
16133
diff
changeset
|
149 if u == r.decode(encoding): |
5d75eb8568d1
encoding: tune fast-path of tolocal a bit
Matt Mackall <mpm@selenic.com>
parents:
16133
diff
changeset
|
150 # r is a safe, non-lossy encoding of s |
5d75eb8568d1
encoding: tune fast-path of tolocal a bit
Matt Mackall <mpm@selenic.com>
parents:
16133
diff
changeset
|
151 return r |
13940
b7b26e54e37a
encoding: avoid localstr when a string can be encoded losslessly (issue2763)
Matt Mackall <mpm@selenic.com>
parents:
13051
diff
changeset
|
152 return localstr(u.encode('UTF-8'), r) |
16274
5d75eb8568d1
encoding: tune fast-path of tolocal a bit
Matt Mackall <mpm@selenic.com>
parents:
16133
diff
changeset
|
153 except UnicodeDecodeError: |
5d75eb8568d1
encoding: tune fast-path of tolocal a bit
Matt Mackall <mpm@selenic.com>
parents:
16133
diff
changeset
|
154 u = s.decode("utf-8", "replace") # last ditch |
5d75eb8568d1
encoding: tune fast-path of tolocal a bit
Matt Mackall <mpm@selenic.com>
parents:
16133
diff
changeset
|
155 return u.encode(encoding, "replace") # can't round-trip |
25660
328739ea70c3
global: mass rewrite to use modern exception syntax
Gregory Szorc <gregory.szorc@gmail.com>
parents:
24608
diff
changeset
|
156 except LookupError as k: |
16274
5d75eb8568d1
encoding: tune fast-path of tolocal a bit
Matt Mackall <mpm@selenic.com>
parents:
16133
diff
changeset
|
157 raise error.Abort(k, hint="please check your locale settings") |
7948
de377b1a9a84
move encoding bits from util to encoding
Matt Mackall <mpm@selenic.com>
parents:
diff
changeset
|
158 |
de377b1a9a84
move encoding bits from util to encoding
Matt Mackall <mpm@selenic.com>
parents:
diff
changeset
|
159 def fromlocal(s): |
de377b1a9a84
move encoding bits from util to encoding
Matt Mackall <mpm@selenic.com>
parents:
diff
changeset
|
160 """ |
de377b1a9a84
move encoding bits from util to encoding
Matt Mackall <mpm@selenic.com>
parents:
diff
changeset
|
161 Convert a string from the local character encoding to UTF-8 |
de377b1a9a84
move encoding bits from util to encoding
Matt Mackall <mpm@selenic.com>
parents:
diff
changeset
|
162 |
de377b1a9a84
move encoding bits from util to encoding
Matt Mackall <mpm@selenic.com>
parents:
diff
changeset
|
163 We attempt to decode strings using the encoding mode set by |
de377b1a9a84
move encoding bits from util to encoding
Matt Mackall <mpm@selenic.com>
parents:
diff
changeset
|
164 HGENCODINGMODE, which defaults to 'strict'. In this mode, unknown |
de377b1a9a84
move encoding bits from util to encoding
Matt Mackall <mpm@selenic.com>
parents:
diff
changeset
|
165 characters will cause an error message. Other modes include |
de377b1a9a84
move encoding bits from util to encoding
Matt Mackall <mpm@selenic.com>
parents:
diff
changeset
|
166 'replace', which replaces unknown characters with a special |
de377b1a9a84
move encoding bits from util to encoding
Matt Mackall <mpm@selenic.com>
parents:
diff
changeset
|
167 Unicode character, and 'ignore', which drops the character. |
de377b1a9a84
move encoding bits from util to encoding
Matt Mackall <mpm@selenic.com>
parents:
diff
changeset
|
168 """ |
13046
7cc4263e07a9
encoding: add localstr class to track UTF-8 version of transcoded strings
Matt Mackall <mpm@selenic.com>
parents:
12866
diff
changeset
|
169 |
7cc4263e07a9
encoding: add localstr class to track UTF-8 version of transcoded strings
Matt Mackall <mpm@selenic.com>
parents:
12866
diff
changeset
|
170 # can we do a lossless round-trip? |
7cc4263e07a9
encoding: add localstr class to track UTF-8 version of transcoded strings
Matt Mackall <mpm@selenic.com>
parents:
12866
diff
changeset
|
171 if isinstance(s, localstr): |
7cc4263e07a9
encoding: add localstr class to track UTF-8 version of transcoded strings
Matt Mackall <mpm@selenic.com>
parents:
12866
diff
changeset
|
172 return s._utf8 |
7cc4263e07a9
encoding: add localstr class to track UTF-8 version of transcoded strings
Matt Mackall <mpm@selenic.com>
parents:
12866
diff
changeset
|
173 |
7948
de377b1a9a84
move encoding bits from util to encoding
Matt Mackall <mpm@selenic.com>
parents:
diff
changeset
|
174 try: |
de377b1a9a84
move encoding bits from util to encoding
Matt Mackall <mpm@selenic.com>
parents:
diff
changeset
|
175 return s.decode(encoding, encodingmode).encode("utf-8") |
25660
328739ea70c3
global: mass rewrite to use modern exception syntax
Gregory Szorc <gregory.szorc@gmail.com>
parents:
24608
diff
changeset
|
176 except UnicodeDecodeError as inst: |
10282
08a0f04b56bd
many, many trivial check-code fixups
Matt Mackall <mpm@selenic.com>
parents:
10263
diff
changeset
|
177 sub = s[max(0, inst.start - 10):inst.start + 10] |
7948
de377b1a9a84
move encoding bits from util to encoding
Matt Mackall <mpm@selenic.com>
parents:
diff
changeset
|
178 raise error.Abort("decoding near '%s': %s!" % (sub, inst)) |
25660
328739ea70c3
global: mass rewrite to use modern exception syntax
Gregory Szorc <gregory.szorc@gmail.com>
parents:
24608
diff
changeset
|
179 except LookupError as k: |
15769
afdf4f5bac61
encoding: use hint markup for "please check your locale settings"
Mads Kiilerich <mads@kiilerich.com>
parents:
15672
diff
changeset
|
180 raise error.Abort(k, hint="please check your locale settings") |
7948
de377b1a9a84
move encoding bits from util to encoding
Matt Mackall <mpm@selenic.com>
parents:
diff
changeset
|
181 |
12866
eddc20306ab6
encoding: default ambiguous character to narrow
Matt Mackall <mpm@selenic.com>
parents:
12770
diff
changeset
|
182 # How to treat ambiguous-width characters. Set to 'wide' to treat as wide. |
15066
24efa83d81cb
i18n: calculate terminal columns by width information of each characters
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
14951
diff
changeset
|
183 wide = (os.environ.get("HGENCODINGAMBIGUOUS", "narrow") == "wide" |
24efa83d81cb
i18n: calculate terminal columns by width information of each characters
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
14951
diff
changeset
|
184 and "WFA" or "WF") |
12866
eddc20306ab6
encoding: default ambiguous character to narrow
Matt Mackall <mpm@selenic.com>
parents:
12770
diff
changeset
|
185 |
7948
de377b1a9a84
move encoding bits from util to encoding
Matt Mackall <mpm@selenic.com>
parents:
diff
changeset
|
186 def colwidth(s): |
15142
176882876780
encoding: colwidth input is in the local encoding
Matt Mackall <mpm@selenic.com>
parents:
15066
diff
changeset
|
187 "Find the column width of a string for display in the local encoding" |
15066
24efa83d81cb
i18n: calculate terminal columns by width information of each characters
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
14951
diff
changeset
|
188 return ucolwidth(s.decode(encoding, 'replace')) |
24efa83d81cb
i18n: calculate terminal columns by width information of each characters
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
14951
diff
changeset
|
189 |
24efa83d81cb
i18n: calculate terminal columns by width information of each characters
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
14951
diff
changeset
|
190 def ucolwidth(d): |
24efa83d81cb
i18n: calculate terminal columns by width information of each characters
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
14951
diff
changeset
|
191 "Find the column width of a Unicode string for display" |
14951
61807854004e
encoding: use getattr isntead of hasattr
Augie Fackler <durin42@gmail.com>
parents:
14069
diff
changeset
|
192 eaw = getattr(unicodedata, 'east_asian_width', None) |
61807854004e
encoding: use getattr isntead of hasattr
Augie Fackler <durin42@gmail.com>
parents:
14069
diff
changeset
|
193 if eaw is not None: |
61807854004e
encoding: use getattr isntead of hasattr
Augie Fackler <durin42@gmail.com>
parents:
14069
diff
changeset
|
194 return sum([eaw(c) in wide and 2 or 1 for c in d]) |
7948
de377b1a9a84
move encoding bits from util to encoding
Matt Mackall <mpm@selenic.com>
parents:
diff
changeset
|
195 return len(d) |
de377b1a9a84
move encoding bits from util to encoding
Matt Mackall <mpm@selenic.com>
parents:
diff
changeset
|
196 |
15143
16c129b0f465
encoding: add getcols to extract substrings based on column width
Matt Mackall <mpm@selenic.com>
parents:
15142
diff
changeset
|
197 def getcols(s, start, c): |
16c129b0f465
encoding: add getcols to extract substrings based on column width
Matt Mackall <mpm@selenic.com>
parents:
15142
diff
changeset
|
198 '''Use colwidth to find a c-column substring of s starting at byte |
16c129b0f465
encoding: add getcols to extract substrings based on column width
Matt Mackall <mpm@selenic.com>
parents:
15142
diff
changeset
|
199 index start''' |
16c129b0f465
encoding: add getcols to extract substrings based on column width
Matt Mackall <mpm@selenic.com>
parents:
15142
diff
changeset
|
200 for x in xrange(start + c, len(s)): |
16c129b0f465
encoding: add getcols to extract substrings based on column width
Matt Mackall <mpm@selenic.com>
parents:
15142
diff
changeset
|
201 t = s[start:x] |
16c129b0f465
encoding: add getcols to extract substrings based on column width
Matt Mackall <mpm@selenic.com>
parents:
15142
diff
changeset
|
202 if colwidth(t) == c: |
16c129b0f465
encoding: add getcols to extract substrings based on column width
Matt Mackall <mpm@selenic.com>
parents:
15142
diff
changeset
|
203 return t |
16c129b0f465
encoding: add getcols to extract substrings based on column width
Matt Mackall <mpm@selenic.com>
parents:
15142
diff
changeset
|
204 |
21861
b515c3a63e96
encoding: add 'leftside' argument into 'trim' to switch trimming side
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
21856
diff
changeset
|
205 def trim(s, width, ellipsis='', leftside=False): |
21856
d24969ee272f
encoding: add 'trim' to trim multi-byte characters at most specified columns
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
18378
diff
changeset
|
206 """Trim string 's' to at most 'width' columns (including 'ellipsis'). |
d24969ee272f
encoding: add 'trim' to trim multi-byte characters at most specified columns
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
18378
diff
changeset
|
207 |
21861
b515c3a63e96
encoding: add 'leftside' argument into 'trim' to switch trimming side
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
21856
diff
changeset
|
208 If 'leftside' is True, left side of string 's' is trimmed. |
b515c3a63e96
encoding: add 'leftside' argument into 'trim' to switch trimming side
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
21856
diff
changeset
|
209 'ellipsis' is always placed at trimmed side. |
b515c3a63e96
encoding: add 'leftside' argument into 'trim' to switch trimming side
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
21856
diff
changeset
|
210 |
21856
d24969ee272f
encoding: add 'trim' to trim multi-byte characters at most specified columns
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
18378
diff
changeset
|
211 >>> ellipsis = '+++' |
27355
b479fc425a81
encoding: use absolute_import
Gregory Szorc <gregory.szorc@gmail.com>
parents:
26963
diff
changeset
|
212 >>> from . import encoding |
21856
d24969ee272f
encoding: add 'trim' to trim multi-byte characters at most specified columns
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
18378
diff
changeset
|
213 >>> encoding.encoding = 'utf-8' |
d24969ee272f
encoding: add 'trim' to trim multi-byte characters at most specified columns
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
18378
diff
changeset
|
214 >>> t= '1234567890' |
d24969ee272f
encoding: add 'trim' to trim multi-byte characters at most specified columns
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
18378
diff
changeset
|
215 >>> print trim(t, 12, ellipsis=ellipsis) |
d24969ee272f
encoding: add 'trim' to trim multi-byte characters at most specified columns
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
18378
diff
changeset
|
216 1234567890 |
d24969ee272f
encoding: add 'trim' to trim multi-byte characters at most specified columns
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
18378
diff
changeset
|
217 >>> print trim(t, 10, ellipsis=ellipsis) |
d24969ee272f
encoding: add 'trim' to trim multi-byte characters at most specified columns
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
18378
diff
changeset
|
218 1234567890 |
d24969ee272f
encoding: add 'trim' to trim multi-byte characters at most specified columns
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
18378
diff
changeset
|
219 >>> print trim(t, 8, ellipsis=ellipsis) |
d24969ee272f
encoding: add 'trim' to trim multi-byte characters at most specified columns
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
18378
diff
changeset
|
220 12345+++ |
21861
b515c3a63e96
encoding: add 'leftside' argument into 'trim' to switch trimming side
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
21856
diff
changeset
|
221 >>> print trim(t, 8, ellipsis=ellipsis, leftside=True) |
b515c3a63e96
encoding: add 'leftside' argument into 'trim' to switch trimming side
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
21856
diff
changeset
|
222 +++67890 |
21856
d24969ee272f
encoding: add 'trim' to trim multi-byte characters at most specified columns
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
18378
diff
changeset
|
223 >>> print trim(t, 8) |
d24969ee272f
encoding: add 'trim' to trim multi-byte characters at most specified columns
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
18378
diff
changeset
|
224 12345678 |
21861
b515c3a63e96
encoding: add 'leftside' argument into 'trim' to switch trimming side
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
21856
diff
changeset
|
225 >>> print trim(t, 8, leftside=True) |
b515c3a63e96
encoding: add 'leftside' argument into 'trim' to switch trimming side
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
21856
diff
changeset
|
226 34567890 |
21856
d24969ee272f
encoding: add 'trim' to trim multi-byte characters at most specified columns
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
18378
diff
changeset
|
227 >>> print trim(t, 3, ellipsis=ellipsis) |
d24969ee272f
encoding: add 'trim' to trim multi-byte characters at most specified columns
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
18378
diff
changeset
|
228 +++ |
d24969ee272f
encoding: add 'trim' to trim multi-byte characters at most specified columns
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
18378
diff
changeset
|
229 >>> print trim(t, 1, ellipsis=ellipsis) |
d24969ee272f
encoding: add 'trim' to trim multi-byte characters at most specified columns
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
18378
diff
changeset
|
230 + |
d24969ee272f
encoding: add 'trim' to trim multi-byte characters at most specified columns
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
18378
diff
changeset
|
231 >>> u = u'\u3042\u3044\u3046\u3048\u304a' # 2 x 5 = 10 columns |
d24969ee272f
encoding: add 'trim' to trim multi-byte characters at most specified columns
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
18378
diff
changeset
|
232 >>> t = u.encode(encoding.encoding) |
d24969ee272f
encoding: add 'trim' to trim multi-byte characters at most specified columns
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
18378
diff
changeset
|
233 >>> print trim(t, 12, ellipsis=ellipsis) |
d24969ee272f
encoding: add 'trim' to trim multi-byte characters at most specified columns
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
18378
diff
changeset
|
234 \xe3\x81\x82\xe3\x81\x84\xe3\x81\x86\xe3\x81\x88\xe3\x81\x8a |
d24969ee272f
encoding: add 'trim' to trim multi-byte characters at most specified columns
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
18378
diff
changeset
|
235 >>> print trim(t, 10, ellipsis=ellipsis) |
d24969ee272f
encoding: add 'trim' to trim multi-byte characters at most specified columns
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
18378
diff
changeset
|
236 \xe3\x81\x82\xe3\x81\x84\xe3\x81\x86\xe3\x81\x88\xe3\x81\x8a |
d24969ee272f
encoding: add 'trim' to trim multi-byte characters at most specified columns
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
18378
diff
changeset
|
237 >>> print trim(t, 8, ellipsis=ellipsis) |
d24969ee272f
encoding: add 'trim' to trim multi-byte characters at most specified columns
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
18378
diff
changeset
|
238 \xe3\x81\x82\xe3\x81\x84+++ |
21861
b515c3a63e96
encoding: add 'leftside' argument into 'trim' to switch trimming side
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
21856
diff
changeset
|
239 >>> print trim(t, 8, ellipsis=ellipsis, leftside=True) |
b515c3a63e96
encoding: add 'leftside' argument into 'trim' to switch trimming side
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
21856
diff
changeset
|
240 +++\xe3\x81\x88\xe3\x81\x8a |
21856
d24969ee272f
encoding: add 'trim' to trim multi-byte characters at most specified columns
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
18378
diff
changeset
|
241 >>> print trim(t, 5) |
d24969ee272f
encoding: add 'trim' to trim multi-byte characters at most specified columns
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
18378
diff
changeset
|
242 \xe3\x81\x82\xe3\x81\x84 |
21861
b515c3a63e96
encoding: add 'leftside' argument into 'trim' to switch trimming side
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
21856
diff
changeset
|
243 >>> print trim(t, 5, leftside=True) |
b515c3a63e96
encoding: add 'leftside' argument into 'trim' to switch trimming side
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
21856
diff
changeset
|
244 \xe3\x81\x88\xe3\x81\x8a |
21856
d24969ee272f
encoding: add 'trim' to trim multi-byte characters at most specified columns
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
18378
diff
changeset
|
245 >>> print trim(t, 4, ellipsis=ellipsis) |
d24969ee272f
encoding: add 'trim' to trim multi-byte characters at most specified columns
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
18378
diff
changeset
|
246 +++ |
21861
b515c3a63e96
encoding: add 'leftside' argument into 'trim' to switch trimming side
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
21856
diff
changeset
|
247 >>> print trim(t, 4, ellipsis=ellipsis, leftside=True) |
b515c3a63e96
encoding: add 'leftside' argument into 'trim' to switch trimming side
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
21856
diff
changeset
|
248 +++ |
21856
d24969ee272f
encoding: add 'trim' to trim multi-byte characters at most specified columns
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
18378
diff
changeset
|
249 >>> t = '\x11\x22\x33\x44\x55\x66\x77\x88\x99\xaa' # invalid byte sequence |
d24969ee272f
encoding: add 'trim' to trim multi-byte characters at most specified columns
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
18378
diff
changeset
|
250 >>> print trim(t, 12, ellipsis=ellipsis) |
d24969ee272f
encoding: add 'trim' to trim multi-byte characters at most specified columns
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
18378
diff
changeset
|
251 \x11\x22\x33\x44\x55\x66\x77\x88\x99\xaa |
d24969ee272f
encoding: add 'trim' to trim multi-byte characters at most specified columns
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
18378
diff
changeset
|
252 >>> print trim(t, 10, ellipsis=ellipsis) |
d24969ee272f
encoding: add 'trim' to trim multi-byte characters at most specified columns
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
18378
diff
changeset
|
253 \x11\x22\x33\x44\x55\x66\x77\x88\x99\xaa |
d24969ee272f
encoding: add 'trim' to trim multi-byte characters at most specified columns
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
18378
diff
changeset
|
254 >>> print trim(t, 8, ellipsis=ellipsis) |
d24969ee272f
encoding: add 'trim' to trim multi-byte characters at most specified columns
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
18378
diff
changeset
|
255 \x11\x22\x33\x44\x55+++ |
21861
b515c3a63e96
encoding: add 'leftside' argument into 'trim' to switch trimming side
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
21856
diff
changeset
|
256 >>> print trim(t, 8, ellipsis=ellipsis, leftside=True) |
b515c3a63e96
encoding: add 'leftside' argument into 'trim' to switch trimming side
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
21856
diff
changeset
|
257 +++\x66\x77\x88\x99\xaa |
21856
d24969ee272f
encoding: add 'trim' to trim multi-byte characters at most specified columns
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
18378
diff
changeset
|
258 >>> print trim(t, 8) |
d24969ee272f
encoding: add 'trim' to trim multi-byte characters at most specified columns
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
18378
diff
changeset
|
259 \x11\x22\x33\x44\x55\x66\x77\x88 |
21861
b515c3a63e96
encoding: add 'leftside' argument into 'trim' to switch trimming side
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
21856
diff
changeset
|
260 >>> print trim(t, 8, leftside=True) |
b515c3a63e96
encoding: add 'leftside' argument into 'trim' to switch trimming side
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
21856
diff
changeset
|
261 \x33\x44\x55\x66\x77\x88\x99\xaa |
21856
d24969ee272f
encoding: add 'trim' to trim multi-byte characters at most specified columns
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
18378
diff
changeset
|
262 >>> print trim(t, 3, ellipsis=ellipsis) |
d24969ee272f
encoding: add 'trim' to trim multi-byte characters at most specified columns
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
18378
diff
changeset
|
263 +++ |
d24969ee272f
encoding: add 'trim' to trim multi-byte characters at most specified columns
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
18378
diff
changeset
|
264 >>> print trim(t, 1, ellipsis=ellipsis) |
d24969ee272f
encoding: add 'trim' to trim multi-byte characters at most specified columns
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
18378
diff
changeset
|
265 + |
d24969ee272f
encoding: add 'trim' to trim multi-byte characters at most specified columns
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
18378
diff
changeset
|
266 """ |
d24969ee272f
encoding: add 'trim' to trim multi-byte characters at most specified columns
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
18378
diff
changeset
|
267 try: |
d24969ee272f
encoding: add 'trim' to trim multi-byte characters at most specified columns
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
18378
diff
changeset
|
268 u = s.decode(encoding) |
d24969ee272f
encoding: add 'trim' to trim multi-byte characters at most specified columns
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
18378
diff
changeset
|
269 except UnicodeDecodeError: |
d24969ee272f
encoding: add 'trim' to trim multi-byte characters at most specified columns
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
18378
diff
changeset
|
270 if len(s) <= width: # trimming is not needed |
d24969ee272f
encoding: add 'trim' to trim multi-byte characters at most specified columns
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
18378
diff
changeset
|
271 return s |
d24969ee272f
encoding: add 'trim' to trim multi-byte characters at most specified columns
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
18378
diff
changeset
|
272 width -= len(ellipsis) |
d24969ee272f
encoding: add 'trim' to trim multi-byte characters at most specified columns
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
18378
diff
changeset
|
273 if width <= 0: # no enough room even for ellipsis |
d24969ee272f
encoding: add 'trim' to trim multi-byte characters at most specified columns
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
18378
diff
changeset
|
274 return ellipsis[:width + len(ellipsis)] |
21861
b515c3a63e96
encoding: add 'leftside' argument into 'trim' to switch trimming side
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
21856
diff
changeset
|
275 if leftside: |
b515c3a63e96
encoding: add 'leftside' argument into 'trim' to switch trimming side
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
21856
diff
changeset
|
276 return ellipsis + s[-width:] |
21856
d24969ee272f
encoding: add 'trim' to trim multi-byte characters at most specified columns
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
18378
diff
changeset
|
277 return s[:width] + ellipsis |
d24969ee272f
encoding: add 'trim' to trim multi-byte characters at most specified columns
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
18378
diff
changeset
|
278 |
d24969ee272f
encoding: add 'trim' to trim multi-byte characters at most specified columns
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
18378
diff
changeset
|
279 if ucolwidth(u) <= width: # trimming is not needed |
d24969ee272f
encoding: add 'trim' to trim multi-byte characters at most specified columns
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
18378
diff
changeset
|
280 return s |
d24969ee272f
encoding: add 'trim' to trim multi-byte characters at most specified columns
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
18378
diff
changeset
|
281 |
d24969ee272f
encoding: add 'trim' to trim multi-byte characters at most specified columns
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
18378
diff
changeset
|
282 width -= len(ellipsis) |
d24969ee272f
encoding: add 'trim' to trim multi-byte characters at most specified columns
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
18378
diff
changeset
|
283 if width <= 0: # no enough room even for ellipsis |
d24969ee272f
encoding: add 'trim' to trim multi-byte characters at most specified columns
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
18378
diff
changeset
|
284 return ellipsis[:width + len(ellipsis)] |
d24969ee272f
encoding: add 'trim' to trim multi-byte characters at most specified columns
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
18378
diff
changeset
|
285 |
21861
b515c3a63e96
encoding: add 'leftside' argument into 'trim' to switch trimming side
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
21856
diff
changeset
|
286 if leftside: |
b515c3a63e96
encoding: add 'leftside' argument into 'trim' to switch trimming side
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
21856
diff
changeset
|
287 uslice = lambda i: u[i:] |
b515c3a63e96
encoding: add 'leftside' argument into 'trim' to switch trimming side
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
21856
diff
changeset
|
288 concat = lambda s: ellipsis + s |
b515c3a63e96
encoding: add 'leftside' argument into 'trim' to switch trimming side
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
21856
diff
changeset
|
289 else: |
b515c3a63e96
encoding: add 'leftside' argument into 'trim' to switch trimming side
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
21856
diff
changeset
|
290 uslice = lambda i: u[:-i] |
b515c3a63e96
encoding: add 'leftside' argument into 'trim' to switch trimming side
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
21856
diff
changeset
|
291 concat = lambda s: s + ellipsis |
21856
d24969ee272f
encoding: add 'trim' to trim multi-byte characters at most specified columns
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
18378
diff
changeset
|
292 for i in xrange(1, len(u)): |
d24969ee272f
encoding: add 'trim' to trim multi-byte characters at most specified columns
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
18378
diff
changeset
|
293 usub = uslice(i) |
d24969ee272f
encoding: add 'trim' to trim multi-byte characters at most specified columns
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
18378
diff
changeset
|
294 if ucolwidth(usub) <= width: |
d24969ee272f
encoding: add 'trim' to trim multi-byte characters at most specified columns
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
18378
diff
changeset
|
295 return concat(usub.encode(encoding)) |
d24969ee272f
encoding: add 'trim' to trim multi-byte characters at most specified columns
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
18378
diff
changeset
|
296 return ellipsis # no enough room for multi-column characters |
d24969ee272f
encoding: add 'trim' to trim multi-byte characters at most specified columns
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
18378
diff
changeset
|
297 |
22973
bcff9ecdaae0
encoding: avoid cyclic dependency around "parsers" in pure Python build
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
22779
diff
changeset
|
298 def _asciilower(s): |
22778
80f2b63dd83a
parsers: add a function to efficiently lowercase ASCII strings
Siddharth Agarwal <sid0@fb.com>
parents:
22426
diff
changeset
|
299 '''convert a string to lowercase if ASCII |
80f2b63dd83a
parsers: add a function to efficiently lowercase ASCII strings
Siddharth Agarwal <sid0@fb.com>
parents:
22426
diff
changeset
|
300 |
80f2b63dd83a
parsers: add a function to efficiently lowercase ASCII strings
Siddharth Agarwal <sid0@fb.com>
parents:
22426
diff
changeset
|
301 Raises UnicodeDecodeError if non-ASCII characters are found.''' |
80f2b63dd83a
parsers: add a function to efficiently lowercase ASCII strings
Siddharth Agarwal <sid0@fb.com>
parents:
22426
diff
changeset
|
302 s.decode('ascii') |
80f2b63dd83a
parsers: add a function to efficiently lowercase ASCII strings
Siddharth Agarwal <sid0@fb.com>
parents:
22426
diff
changeset
|
303 return s.lower() |
80f2b63dd83a
parsers: add a function to efficiently lowercase ASCII strings
Siddharth Agarwal <sid0@fb.com>
parents:
22426
diff
changeset
|
304 |
22973
bcff9ecdaae0
encoding: avoid cyclic dependency around "parsers" in pure Python build
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
22779
diff
changeset
|
305 def asciilower(s): |
bcff9ecdaae0
encoding: avoid cyclic dependency around "parsers" in pure Python build
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
22779
diff
changeset
|
306 # delay importing avoids cyclic dependency around "parsers" in |
bcff9ecdaae0
encoding: avoid cyclic dependency around "parsers" in pure Python build
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
22779
diff
changeset
|
307 # pure Python build (util => i18n => encoding => parsers => util) |
27355
b479fc425a81
encoding: use absolute_import
Gregory Szorc <gregory.szorc@gmail.com>
parents:
26963
diff
changeset
|
308 from . import parsers |
22973
bcff9ecdaae0
encoding: avoid cyclic dependency around "parsers" in pure Python build
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
22779
diff
changeset
|
309 impl = getattr(parsers, 'asciilower', _asciilower) |
bcff9ecdaae0
encoding: avoid cyclic dependency around "parsers" in pure Python build
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
22779
diff
changeset
|
310 global asciilower |
bcff9ecdaae0
encoding: avoid cyclic dependency around "parsers" in pure Python build
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
22779
diff
changeset
|
311 asciilower = impl |
bcff9ecdaae0
encoding: avoid cyclic dependency around "parsers" in pure Python build
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
22779
diff
changeset
|
312 return impl(s) |
22778
80f2b63dd83a
parsers: add a function to efficiently lowercase ASCII strings
Siddharth Agarwal <sid0@fb.com>
parents:
22426
diff
changeset
|
313 |
24578
ac08de78de7f
encoding: use parsers.asciiupper when available
Siddharth Agarwal <sid0@fb.com>
parents:
23596
diff
changeset
|
314 def _asciiupper(s): |
ac08de78de7f
encoding: use parsers.asciiupper when available
Siddharth Agarwal <sid0@fb.com>
parents:
23596
diff
changeset
|
315 '''convert a string to uppercase if ASCII |
ac08de78de7f
encoding: use parsers.asciiupper when available
Siddharth Agarwal <sid0@fb.com>
parents:
23596
diff
changeset
|
316 |
ac08de78de7f
encoding: use parsers.asciiupper when available
Siddharth Agarwal <sid0@fb.com>
parents:
23596
diff
changeset
|
317 Raises UnicodeDecodeError if non-ASCII characters are found.''' |
ac08de78de7f
encoding: use parsers.asciiupper when available
Siddharth Agarwal <sid0@fb.com>
parents:
23596
diff
changeset
|
318 s.decode('ascii') |
ac08de78de7f
encoding: use parsers.asciiupper when available
Siddharth Agarwal <sid0@fb.com>
parents:
23596
diff
changeset
|
319 return s.upper() |
ac08de78de7f
encoding: use parsers.asciiupper when available
Siddharth Agarwal <sid0@fb.com>
parents:
23596
diff
changeset
|
320 |
ac08de78de7f
encoding: use parsers.asciiupper when available
Siddharth Agarwal <sid0@fb.com>
parents:
23596
diff
changeset
|
321 def asciiupper(s): |
ac08de78de7f
encoding: use parsers.asciiupper when available
Siddharth Agarwal <sid0@fb.com>
parents:
23596
diff
changeset
|
322 # delay importing avoids cyclic dependency around "parsers" in |
ac08de78de7f
encoding: use parsers.asciiupper when available
Siddharth Agarwal <sid0@fb.com>
parents:
23596
diff
changeset
|
323 # pure Python build (util => i18n => encoding => parsers => util) |
27355
b479fc425a81
encoding: use absolute_import
Gregory Szorc <gregory.szorc@gmail.com>
parents:
26963
diff
changeset
|
324 from . import parsers |
24578
ac08de78de7f
encoding: use parsers.asciiupper when available
Siddharth Agarwal <sid0@fb.com>
parents:
23596
diff
changeset
|
325 impl = getattr(parsers, 'asciiupper', _asciiupper) |
ac08de78de7f
encoding: use parsers.asciiupper when available
Siddharth Agarwal <sid0@fb.com>
parents:
23596
diff
changeset
|
326 global asciiupper |
ac08de78de7f
encoding: use parsers.asciiupper when available
Siddharth Agarwal <sid0@fb.com>
parents:
23596
diff
changeset
|
327 asciiupper = impl |
ac08de78de7f
encoding: use parsers.asciiupper when available
Siddharth Agarwal <sid0@fb.com>
parents:
23596
diff
changeset
|
328 return impl(s) |
ac08de78de7f
encoding: use parsers.asciiupper when available
Siddharth Agarwal <sid0@fb.com>
parents:
23596
diff
changeset
|
329 |
14069
e38846a79a23
encoding: add an encoding-aware lower function
Matt Mackall <mpm@selenic.com>
parents:
13940
diff
changeset
|
330 def lower(s): |
e38846a79a23
encoding: add an encoding-aware lower function
Matt Mackall <mpm@selenic.com>
parents:
13940
diff
changeset
|
331 "best-effort encoding-aware case-folding of local string s" |
e38846a79a23
encoding: add an encoding-aware lower function
Matt Mackall <mpm@selenic.com>
parents:
13940
diff
changeset
|
332 try: |
22779
d9585dda63c3
encoding.lower: use fast ASCII lower
Siddharth Agarwal <sid0@fb.com>
parents:
22778
diff
changeset
|
333 return asciilower(s) |
17235
3745ae495ce5
encoding: use s.decode to trigger UnicodeDecodeError
Martin Geisler <mg@aragost.com>
parents:
16493
diff
changeset
|
334 except UnicodeDecodeError: |
16387
c481761033bd
encoding: add fast-path for ASCII lowercase
Matt Mackall <mpm@selenic.com>
parents:
16274
diff
changeset
|
335 pass |
c481761033bd
encoding: add fast-path for ASCII lowercase
Matt Mackall <mpm@selenic.com>
parents:
16274
diff
changeset
|
336 try: |
14069
e38846a79a23
encoding: add an encoding-aware lower function
Matt Mackall <mpm@selenic.com>
parents:
13940
diff
changeset
|
337 if isinstance(s, localstr): |
e38846a79a23
encoding: add an encoding-aware lower function
Matt Mackall <mpm@selenic.com>
parents:
13940
diff
changeset
|
338 u = s._utf8.decode("utf-8") |
e38846a79a23
encoding: add an encoding-aware lower function
Matt Mackall <mpm@selenic.com>
parents:
13940
diff
changeset
|
339 else: |
e38846a79a23
encoding: add an encoding-aware lower function
Matt Mackall <mpm@selenic.com>
parents:
13940
diff
changeset
|
340 u = s.decode(encoding, encodingmode) |
e38846a79a23
encoding: add an encoding-aware lower function
Matt Mackall <mpm@selenic.com>
parents:
13940
diff
changeset
|
341 |
e38846a79a23
encoding: add an encoding-aware lower function
Matt Mackall <mpm@selenic.com>
parents:
13940
diff
changeset
|
342 lu = u.lower() |
e38846a79a23
encoding: add an encoding-aware lower function
Matt Mackall <mpm@selenic.com>
parents:
13940
diff
changeset
|
343 if u == lu: |
e38846a79a23
encoding: add an encoding-aware lower function
Matt Mackall <mpm@selenic.com>
parents:
13940
diff
changeset
|
344 return s # preserve localstring |
e38846a79a23
encoding: add an encoding-aware lower function
Matt Mackall <mpm@selenic.com>
parents:
13940
diff
changeset
|
345 return lu.encode(encoding) |
e38846a79a23
encoding: add an encoding-aware lower function
Matt Mackall <mpm@selenic.com>
parents:
13940
diff
changeset
|
346 except UnicodeError: |
e38846a79a23
encoding: add an encoding-aware lower function
Matt Mackall <mpm@selenic.com>
parents:
13940
diff
changeset
|
347 return s.lower() # we don't know how to fold this except in ASCII |
25660
328739ea70c3
global: mass rewrite to use modern exception syntax
Gregory Szorc <gregory.szorc@gmail.com>
parents:
24608
diff
changeset
|
348 except LookupError as k: |
15672
2ebe3d0ce91d
i18n: use encoding.lower/upper for encoding aware case folding
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
15143
diff
changeset
|
349 raise error.Abort(k, hint="please check your locale settings") |
2ebe3d0ce91d
i18n: use encoding.lower/upper for encoding aware case folding
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
15143
diff
changeset
|
350 |
2ebe3d0ce91d
i18n: use encoding.lower/upper for encoding aware case folding
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
15143
diff
changeset
|
351 def upper(s): |
2ebe3d0ce91d
i18n: use encoding.lower/upper for encoding aware case folding
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
15143
diff
changeset
|
352 "best-effort encoding-aware case-folding of local string s" |
2ebe3d0ce91d
i18n: use encoding.lower/upper for encoding aware case folding
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
15143
diff
changeset
|
353 try: |
24578
ac08de78de7f
encoding: use parsers.asciiupper when available
Siddharth Agarwal <sid0@fb.com>
parents:
23596
diff
changeset
|
354 return asciiupper(s) |
17236
9fb8312dbdbd
encoding: add fast-path for ASCII uppercase.
Martin Geisler <mg@aragost.com>
parents:
17235
diff
changeset
|
355 except UnicodeDecodeError: |
24597
b4258d5a1600
encoding.upper: factor out fallback code
Siddharth Agarwal <sid0@fb.com>
parents:
24593
diff
changeset
|
356 return upperfallback(s) |
b4258d5a1600
encoding.upper: factor out fallback code
Siddharth Agarwal <sid0@fb.com>
parents:
24593
diff
changeset
|
357 |
b4258d5a1600
encoding.upper: factor out fallback code
Siddharth Agarwal <sid0@fb.com>
parents:
24593
diff
changeset
|
358 def upperfallback(s): |
17236
9fb8312dbdbd
encoding: add fast-path for ASCII uppercase.
Martin Geisler <mg@aragost.com>
parents:
17235
diff
changeset
|
359 try: |
15672
2ebe3d0ce91d
i18n: use encoding.lower/upper for encoding aware case folding
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
15143
diff
changeset
|
360 if isinstance(s, localstr): |
2ebe3d0ce91d
i18n: use encoding.lower/upper for encoding aware case folding
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
15143
diff
changeset
|
361 u = s._utf8.decode("utf-8") |
2ebe3d0ce91d
i18n: use encoding.lower/upper for encoding aware case folding
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
15143
diff
changeset
|
362 else: |
2ebe3d0ce91d
i18n: use encoding.lower/upper for encoding aware case folding
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
15143
diff
changeset
|
363 u = s.decode(encoding, encodingmode) |
2ebe3d0ce91d
i18n: use encoding.lower/upper for encoding aware case folding
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
15143
diff
changeset
|
364 |
2ebe3d0ce91d
i18n: use encoding.lower/upper for encoding aware case folding
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
15143
diff
changeset
|
365 uu = u.upper() |
2ebe3d0ce91d
i18n: use encoding.lower/upper for encoding aware case folding
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
15143
diff
changeset
|
366 if u == uu: |
2ebe3d0ce91d
i18n: use encoding.lower/upper for encoding aware case folding
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
15143
diff
changeset
|
367 return s # preserve localstring |
2ebe3d0ce91d
i18n: use encoding.lower/upper for encoding aware case folding
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
15143
diff
changeset
|
368 return uu.encode(encoding) |
2ebe3d0ce91d
i18n: use encoding.lower/upper for encoding aware case folding
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
15143
diff
changeset
|
369 except UnicodeError: |
2ebe3d0ce91d
i18n: use encoding.lower/upper for encoding aware case folding
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
15143
diff
changeset
|
370 return s.upper() # we don't know how to fold this except in ASCII |
25660
328739ea70c3
global: mass rewrite to use modern exception syntax
Gregory Szorc <gregory.szorc@gmail.com>
parents:
24608
diff
changeset
|
371 except LookupError as k: |
15672
2ebe3d0ce91d
i18n: use encoding.lower/upper for encoding aware case folding
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents:
15143
diff
changeset
|
372 raise error.Abort(k, hint="please check your locale settings") |
16133
84c58da3a1f8
encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents:
15769
diff
changeset
|
373 |
24593
f473a1fe5c7c
encoding: define an enum that specifies what normcase does to ASCII strings
Siddharth Agarwal <sid0@fb.com>
parents:
24578
diff
changeset
|
374 class normcasespecs(object): |
f473a1fe5c7c
encoding: define an enum that specifies what normcase does to ASCII strings
Siddharth Agarwal <sid0@fb.com>
parents:
24578
diff
changeset
|
375 '''what a platform's normcase does to ASCII strings |
f473a1fe5c7c
encoding: define an enum that specifies what normcase does to ASCII strings
Siddharth Agarwal <sid0@fb.com>
parents:
24578
diff
changeset
|
376 |
f473a1fe5c7c
encoding: define an enum that specifies what normcase does to ASCII strings
Siddharth Agarwal <sid0@fb.com>
parents:
24578
diff
changeset
|
377 This is specified per platform, and should be consistent with what normcase |
f473a1fe5c7c
encoding: define an enum that specifies what normcase does to ASCII strings
Siddharth Agarwal <sid0@fb.com>
parents:
24578
diff
changeset
|
378 on that platform actually does. |
f473a1fe5c7c
encoding: define an enum that specifies what normcase does to ASCII strings
Siddharth Agarwal <sid0@fb.com>
parents:
24578
diff
changeset
|
379 |
f473a1fe5c7c
encoding: define an enum that specifies what normcase does to ASCII strings
Siddharth Agarwal <sid0@fb.com>
parents:
24578
diff
changeset
|
380 lower: normcase lowercases ASCII strings |
f473a1fe5c7c
encoding: define an enum that specifies what normcase does to ASCII strings
Siddharth Agarwal <sid0@fb.com>
parents:
24578
diff
changeset
|
381 upper: normcase uppercases ASCII strings |
24608
1c533e23ce95
util.h: define an enum for normcase specs
Siddharth Agarwal <sid0@fb.com>
parents:
24597
diff
changeset
|
382 other: the fallback function should always be called |
1c533e23ce95
util.h: define an enum for normcase specs
Siddharth Agarwal <sid0@fb.com>
parents:
24597
diff
changeset
|
383 |
1c533e23ce95
util.h: define an enum for normcase specs
Siddharth Agarwal <sid0@fb.com>
parents:
24597
diff
changeset
|
384 This should be kept in sync with normcase_spec in util.h.''' |
24593
f473a1fe5c7c
encoding: define an enum that specifies what normcase does to ASCII strings
Siddharth Agarwal <sid0@fb.com>
parents:
24578
diff
changeset
|
385 lower = -1 |
f473a1fe5c7c
encoding: define an enum that specifies what normcase does to ASCII strings
Siddharth Agarwal <sid0@fb.com>
parents:
24578
diff
changeset
|
386 upper = 1 |
f473a1fe5c7c
encoding: define an enum that specifies what normcase does to ASCII strings
Siddharth Agarwal <sid0@fb.com>
parents:
24578
diff
changeset
|
387 other = 0 |
f473a1fe5c7c
encoding: define an enum that specifies what normcase does to ASCII strings
Siddharth Agarwal <sid0@fb.com>
parents:
24578
diff
changeset
|
388 |
28066
d1cc07123243
encoding: change jsonmap to a list indexed by code point
Yuya Nishihara <yuya@tcha.org>
parents:
27881
diff
changeset
|
389 _jsonmap = [] |
28508
3c6e94d0811c
encoding: use range() instead of xrange()
Gregory Szorc <gregory.szorc@gmail.com>
parents:
28507
diff
changeset
|
390 _jsonmap.extend("\\u%04x" % x for x in range(32)) |
3c6e94d0811c
encoding: use range() instead of xrange()
Gregory Szorc <gregory.szorc@gmail.com>
parents:
28507
diff
changeset
|
391 _jsonmap.extend(chr(x) for x in range(32, 127)) |
28068
9ece901f7a19
encoding: add option to escape non-ascii characters in JSON
Yuya Nishihara <yuya@tcha.org>
parents:
28067
diff
changeset
|
392 _jsonmap.append('\\u007f') |
28067
69a02b1e947c
encoding: initialize jsonmap when module is loaded
Yuya Nishihara <yuya@tcha.org>
parents:
28066
diff
changeset
|
393 _jsonmap[0x09] = '\\t' |
69a02b1e947c
encoding: initialize jsonmap when module is loaded
Yuya Nishihara <yuya@tcha.org>
parents:
28066
diff
changeset
|
394 _jsonmap[0x0a] = '\\n' |
69a02b1e947c
encoding: initialize jsonmap when module is loaded
Yuya Nishihara <yuya@tcha.org>
parents:
28066
diff
changeset
|
395 _jsonmap[0x22] = '\\"' |
69a02b1e947c
encoding: initialize jsonmap when module is loaded
Yuya Nishihara <yuya@tcha.org>
parents:
28066
diff
changeset
|
396 _jsonmap[0x5c] = '\\\\' |
69a02b1e947c
encoding: initialize jsonmap when module is loaded
Yuya Nishihara <yuya@tcha.org>
parents:
28066
diff
changeset
|
397 _jsonmap[0x08] = '\\b' |
69a02b1e947c
encoding: initialize jsonmap when module is loaded
Yuya Nishihara <yuya@tcha.org>
parents:
28066
diff
changeset
|
398 _jsonmap[0x0c] = '\\f' |
69a02b1e947c
encoding: initialize jsonmap when module is loaded
Yuya Nishihara <yuya@tcha.org>
parents:
28066
diff
changeset
|
399 _jsonmap[0x0d] = '\\r' |
28068
9ece901f7a19
encoding: add option to escape non-ascii characters in JSON
Yuya Nishihara <yuya@tcha.org>
parents:
28067
diff
changeset
|
400 _paranoidjsonmap = _jsonmap[:] |
28069
b2d24c2898f9
encoding: backport paranoid escaping from templatefilters.jsonescape()
Yuya Nishihara <yuya@tcha.org>
parents:
28068
diff
changeset
|
401 _paranoidjsonmap[0x3c] = '\\u003c' # '<' (e.g. escape "</script>") |
b2d24c2898f9
encoding: backport paranoid escaping from templatefilters.jsonescape()
Yuya Nishihara <yuya@tcha.org>
parents:
28068
diff
changeset
|
402 _paranoidjsonmap[0x3e] = '\\u003e' # '>' |
28508
3c6e94d0811c
encoding: use range() instead of xrange()
Gregory Szorc <gregory.szorc@gmail.com>
parents:
28507
diff
changeset
|
403 _jsonmap.extend(chr(x) for x in range(128, 256)) |
22426
f6b533e64ed6
encoding: add json escaping filter
Matt Mackall <mpm@selenic.com>
parents:
22425
diff
changeset
|
404 |
28068
9ece901f7a19
encoding: add option to escape non-ascii characters in JSON
Yuya Nishihara <yuya@tcha.org>
parents:
28067
diff
changeset
|
405 def jsonescape(s, paranoid=False): |
22426
f6b533e64ed6
encoding: add json escaping filter
Matt Mackall <mpm@selenic.com>
parents:
22425
diff
changeset
|
406 '''returns a string suitable for JSON |
f6b533e64ed6
encoding: add json escaping filter
Matt Mackall <mpm@selenic.com>
parents:
22425
diff
changeset
|
407 |
f6b533e64ed6
encoding: add json escaping filter
Matt Mackall <mpm@selenic.com>
parents:
22425
diff
changeset
|
408 JSON is problematic for us because it doesn't support non-Unicode |
f6b533e64ed6
encoding: add json escaping filter
Matt Mackall <mpm@selenic.com>
parents:
22425
diff
changeset
|
409 bytes. To deal with this, we take the following approach: |
f6b533e64ed6
encoding: add json escaping filter
Matt Mackall <mpm@selenic.com>
parents:
22425
diff
changeset
|
410 |
f6b533e64ed6
encoding: add json escaping filter
Matt Mackall <mpm@selenic.com>
parents:
22425
diff
changeset
|
411 - localstr objects are converted back to UTF-8 |
f6b533e64ed6
encoding: add json escaping filter
Matt Mackall <mpm@selenic.com>
parents:
22425
diff
changeset
|
412 - valid UTF-8/ASCII strings are passed as-is |
f6b533e64ed6
encoding: add json escaping filter
Matt Mackall <mpm@selenic.com>
parents:
22425
diff
changeset
|
413 - other strings are converted to UTF-8b surrogate encoding |
f6b533e64ed6
encoding: add json escaping filter
Matt Mackall <mpm@selenic.com>
parents:
22425
diff
changeset
|
414 - apply JSON-specified string escaping |
f6b533e64ed6
encoding: add json escaping filter
Matt Mackall <mpm@selenic.com>
parents:
22425
diff
changeset
|
415 |
f6b533e64ed6
encoding: add json escaping filter
Matt Mackall <mpm@selenic.com>
parents:
22425
diff
changeset
|
416 (escapes are doubled in these tests) |
f6b533e64ed6
encoding: add json escaping filter
Matt Mackall <mpm@selenic.com>
parents:
22425
diff
changeset
|
417 |
f6b533e64ed6
encoding: add json escaping filter
Matt Mackall <mpm@selenic.com>
parents:
22425
diff
changeset
|
418 >>> jsonescape('this is a test') |
f6b533e64ed6
encoding: add json escaping filter
Matt Mackall <mpm@selenic.com>
parents:
22425
diff
changeset
|
419 'this is a test' |
27881
ffa599f3f503
encoding: escape U+007F (DEL) character in JSON
Yuya Nishihara <yuya@tcha.org>
parents:
27699
diff
changeset
|
420 >>> jsonescape('escape characters: \\0 \\x0b \\x7f') |
ffa599f3f503
encoding: escape U+007F (DEL) character in JSON
Yuya Nishihara <yuya@tcha.org>
parents:
27699
diff
changeset
|
421 'escape characters: \\\\u0000 \\\\u000b \\\\u007f' |
ffa599f3f503
encoding: escape U+007F (DEL) character in JSON
Yuya Nishihara <yuya@tcha.org>
parents:
27699
diff
changeset
|
422 >>> jsonescape('escape characters: \\t \\n \\r \\" \\\\') |
ffa599f3f503
encoding: escape U+007F (DEL) character in JSON
Yuya Nishihara <yuya@tcha.org>
parents:
27699
diff
changeset
|
423 'escape characters: \\\\t \\\\n \\\\r \\\\" \\\\\\\\' |
22426
f6b533e64ed6
encoding: add json escaping filter
Matt Mackall <mpm@selenic.com>
parents:
22425
diff
changeset
|
424 >>> jsonescape('a weird byte: \\xdd') |
f6b533e64ed6
encoding: add json escaping filter
Matt Mackall <mpm@selenic.com>
parents:
22425
diff
changeset
|
425 'a weird byte: \\xed\\xb3\\x9d' |
f6b533e64ed6
encoding: add json escaping filter
Matt Mackall <mpm@selenic.com>
parents:
22425
diff
changeset
|
426 >>> jsonescape('utf-8: caf\\xc3\\xa9') |
f6b533e64ed6
encoding: add json escaping filter
Matt Mackall <mpm@selenic.com>
parents:
22425
diff
changeset
|
427 'utf-8: caf\\xc3\\xa9' |
f6b533e64ed6
encoding: add json escaping filter
Matt Mackall <mpm@selenic.com>
parents:
22425
diff
changeset
|
428 >>> jsonescape('') |
f6b533e64ed6
encoding: add json escaping filter
Matt Mackall <mpm@selenic.com>
parents:
22425
diff
changeset
|
429 '' |
28068
9ece901f7a19
encoding: add option to escape non-ascii characters in JSON
Yuya Nishihara <yuya@tcha.org>
parents:
28067
diff
changeset
|
430 |
28069
b2d24c2898f9
encoding: backport paranoid escaping from templatefilters.jsonescape()
Yuya Nishihara <yuya@tcha.org>
parents:
28068
diff
changeset
|
431 If paranoid, non-ascii and common troublesome characters are also escaped. |
b2d24c2898f9
encoding: backport paranoid escaping from templatefilters.jsonescape()
Yuya Nishihara <yuya@tcha.org>
parents:
28068
diff
changeset
|
432 This is suitable for web output. |
28068
9ece901f7a19
encoding: add option to escape non-ascii characters in JSON
Yuya Nishihara <yuya@tcha.org>
parents:
28067
diff
changeset
|
433 |
9ece901f7a19
encoding: add option to escape non-ascii characters in JSON
Yuya Nishihara <yuya@tcha.org>
parents:
28067
diff
changeset
|
434 >>> jsonescape('escape boundary: \\x7e \\x7f \\xc2\\x80', paranoid=True) |
9ece901f7a19
encoding: add option to escape non-ascii characters in JSON
Yuya Nishihara <yuya@tcha.org>
parents:
28067
diff
changeset
|
435 'escape boundary: ~ \\\\u007f \\\\u0080' |
9ece901f7a19
encoding: add option to escape non-ascii characters in JSON
Yuya Nishihara <yuya@tcha.org>
parents:
28067
diff
changeset
|
436 >>> jsonescape('a weird byte: \\xdd', paranoid=True) |
9ece901f7a19
encoding: add option to escape non-ascii characters in JSON
Yuya Nishihara <yuya@tcha.org>
parents:
28067
diff
changeset
|
437 'a weird byte: \\\\udcdd' |
9ece901f7a19
encoding: add option to escape non-ascii characters in JSON
Yuya Nishihara <yuya@tcha.org>
parents:
28067
diff
changeset
|
438 >>> jsonescape('utf-8: caf\\xc3\\xa9', paranoid=True) |
9ece901f7a19
encoding: add option to escape non-ascii characters in JSON
Yuya Nishihara <yuya@tcha.org>
parents:
28067
diff
changeset
|
439 'utf-8: caf\\\\u00e9' |
9ece901f7a19
encoding: add option to escape non-ascii characters in JSON
Yuya Nishihara <yuya@tcha.org>
parents:
28067
diff
changeset
|
440 >>> jsonescape('non-BMP: \\xf0\\x9d\\x84\\x9e', paranoid=True) |
9ece901f7a19
encoding: add option to escape non-ascii characters in JSON
Yuya Nishihara <yuya@tcha.org>
parents:
28067
diff
changeset
|
441 'non-BMP: \\\\ud834\\\\udd1e' |
28069
b2d24c2898f9
encoding: backport paranoid escaping from templatefilters.jsonescape()
Yuya Nishihara <yuya@tcha.org>
parents:
28068
diff
changeset
|
442 >>> jsonescape('<foo@example.org>', paranoid=True) |
b2d24c2898f9
encoding: backport paranoid escaping from templatefilters.jsonescape()
Yuya Nishihara <yuya@tcha.org>
parents:
28068
diff
changeset
|
443 '\\\\u003cfoo@example.org\\\\u003e' |
22426
f6b533e64ed6
encoding: add json escaping filter
Matt Mackall <mpm@selenic.com>
parents:
22425
diff
changeset
|
444 ''' |
f6b533e64ed6
encoding: add json escaping filter
Matt Mackall <mpm@selenic.com>
parents:
22425
diff
changeset
|
445 |
28068
9ece901f7a19
encoding: add option to escape non-ascii characters in JSON
Yuya Nishihara <yuya@tcha.org>
parents:
28067
diff
changeset
|
446 if paranoid: |
9ece901f7a19
encoding: add option to escape non-ascii characters in JSON
Yuya Nishihara <yuya@tcha.org>
parents:
28067
diff
changeset
|
447 jm = _paranoidjsonmap |
9ece901f7a19
encoding: add option to escape non-ascii characters in JSON
Yuya Nishihara <yuya@tcha.org>
parents:
28067
diff
changeset
|
448 else: |
9ece901f7a19
encoding: add option to escape non-ascii characters in JSON
Yuya Nishihara <yuya@tcha.org>
parents:
28067
diff
changeset
|
449 jm = _jsonmap |
9ece901f7a19
encoding: add option to escape non-ascii characters in JSON
Yuya Nishihara <yuya@tcha.org>
parents:
28067
diff
changeset
|
450 |
9ece901f7a19
encoding: add option to escape non-ascii characters in JSON
Yuya Nishihara <yuya@tcha.org>
parents:
28067
diff
changeset
|
451 u8chars = toutf8b(s) |
9ece901f7a19
encoding: add option to escape non-ascii characters in JSON
Yuya Nishihara <yuya@tcha.org>
parents:
28067
diff
changeset
|
452 try: |
9ece901f7a19
encoding: add option to escape non-ascii characters in JSON
Yuya Nishihara <yuya@tcha.org>
parents:
28067
diff
changeset
|
453 return ''.join(jm[x] for x in bytearray(u8chars)) # fast path |
9ece901f7a19
encoding: add option to escape non-ascii characters in JSON
Yuya Nishihara <yuya@tcha.org>
parents:
28067
diff
changeset
|
454 except IndexError: |
9ece901f7a19
encoding: add option to escape non-ascii characters in JSON
Yuya Nishihara <yuya@tcha.org>
parents:
28067
diff
changeset
|
455 pass |
9ece901f7a19
encoding: add option to escape non-ascii characters in JSON
Yuya Nishihara <yuya@tcha.org>
parents:
28067
diff
changeset
|
456 # non-BMP char is represented as UTF-16 surrogate pair |
9ece901f7a19
encoding: add option to escape non-ascii characters in JSON
Yuya Nishihara <yuya@tcha.org>
parents:
28067
diff
changeset
|
457 u16codes = array.array('H', u8chars.decode('utf-8').encode('utf-16')) |
9ece901f7a19
encoding: add option to escape non-ascii characters in JSON
Yuya Nishihara <yuya@tcha.org>
parents:
28067
diff
changeset
|
458 u16codes.pop(0) # drop BOM |
9ece901f7a19
encoding: add option to escape non-ascii characters in JSON
Yuya Nishihara <yuya@tcha.org>
parents:
28067
diff
changeset
|
459 return ''.join(jm[x] if x < 128 else '\\u%04x' % x for x in u16codes) |
22426
f6b533e64ed6
encoding: add json escaping filter
Matt Mackall <mpm@selenic.com>
parents:
22425
diff
changeset
|
460 |
26875
cf47bdb2183c
encoding: add getutf8char helper
Matt Mackall <mpm@selenic.com>
parents:
25660
diff
changeset
|
461 _utf8len = [0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 2, 2, 3, 4] |
cf47bdb2183c
encoding: add getutf8char helper
Matt Mackall <mpm@selenic.com>
parents:
25660
diff
changeset
|
462 |
cf47bdb2183c
encoding: add getutf8char helper
Matt Mackall <mpm@selenic.com>
parents:
25660
diff
changeset
|
463 def getutf8char(s, pos): |
cf47bdb2183c
encoding: add getutf8char helper
Matt Mackall <mpm@selenic.com>
parents:
25660
diff
changeset
|
464 '''get the next full utf-8 character in the given string, starting at pos |
cf47bdb2183c
encoding: add getutf8char helper
Matt Mackall <mpm@selenic.com>
parents:
25660
diff
changeset
|
465 |
cf47bdb2183c
encoding: add getutf8char helper
Matt Mackall <mpm@selenic.com>
parents:
25660
diff
changeset
|
466 Raises a UnicodeError if the given location does not start a valid |
cf47bdb2183c
encoding: add getutf8char helper
Matt Mackall <mpm@selenic.com>
parents:
25660
diff
changeset
|
467 utf-8 character. |
cf47bdb2183c
encoding: add getutf8char helper
Matt Mackall <mpm@selenic.com>
parents:
25660
diff
changeset
|
468 ''' |
cf47bdb2183c
encoding: add getutf8char helper
Matt Mackall <mpm@selenic.com>
parents:
25660
diff
changeset
|
469 |
cf47bdb2183c
encoding: add getutf8char helper
Matt Mackall <mpm@selenic.com>
parents:
25660
diff
changeset
|
470 # find how many bytes to attempt decoding from first nibble |
cf47bdb2183c
encoding: add getutf8char helper
Matt Mackall <mpm@selenic.com>
parents:
25660
diff
changeset
|
471 l = _utf8len[ord(s[pos]) >> 4] |
cf47bdb2183c
encoding: add getutf8char helper
Matt Mackall <mpm@selenic.com>
parents:
25660
diff
changeset
|
472 if not l: # ascii |
cf47bdb2183c
encoding: add getutf8char helper
Matt Mackall <mpm@selenic.com>
parents:
25660
diff
changeset
|
473 return s[pos] |
cf47bdb2183c
encoding: add getutf8char helper
Matt Mackall <mpm@selenic.com>
parents:
25660
diff
changeset
|
474 |
cf47bdb2183c
encoding: add getutf8char helper
Matt Mackall <mpm@selenic.com>
parents:
25660
diff
changeset
|
475 c = s[pos:pos + l] |
cf47bdb2183c
encoding: add getutf8char helper
Matt Mackall <mpm@selenic.com>
parents:
25660
diff
changeset
|
476 # validate with attempted decode |
cf47bdb2183c
encoding: add getutf8char helper
Matt Mackall <mpm@selenic.com>
parents:
25660
diff
changeset
|
477 c.decode("utf-8") |
cf47bdb2183c
encoding: add getutf8char helper
Matt Mackall <mpm@selenic.com>
parents:
25660
diff
changeset
|
478 return c |
cf47bdb2183c
encoding: add getutf8char helper
Matt Mackall <mpm@selenic.com>
parents:
25660
diff
changeset
|
479 |
16133
84c58da3a1f8
encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents:
15769
diff
changeset
|
480 def toutf8b(s): |
84c58da3a1f8
encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents:
15769
diff
changeset
|
481 '''convert a local, possibly-binary string into UTF-8b |
84c58da3a1f8
encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents:
15769
diff
changeset
|
482 |
84c58da3a1f8
encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents:
15769
diff
changeset
|
483 This is intended as a generic method to preserve data when working |
84c58da3a1f8
encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents:
15769
diff
changeset
|
484 with schemes like JSON and XML that have no provision for |
84c58da3a1f8
encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents:
15769
diff
changeset
|
485 arbitrary byte strings. As Mercurial often doesn't know |
84c58da3a1f8
encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents:
15769
diff
changeset
|
486 what encoding data is in, we use so-called UTF-8b. |
84c58da3a1f8
encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents:
15769
diff
changeset
|
487 |
84c58da3a1f8
encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents:
15769
diff
changeset
|
488 If a string is already valid UTF-8 (or ASCII), it passes unmodified. |
84c58da3a1f8
encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents:
15769
diff
changeset
|
489 Otherwise, unsupported bytes are mapped to UTF-16 surrogate range, |
84c58da3a1f8
encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents:
15769
diff
changeset
|
490 uDC00-uDCFF. |
84c58da3a1f8
encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents:
15769
diff
changeset
|
491 |
84c58da3a1f8
encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents:
15769
diff
changeset
|
492 Principles of operation: |
84c58da3a1f8
encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents:
15769
diff
changeset
|
493 |
17424
e7cfe3587ea4
fix trivial spelling errors
Mads Kiilerich <mads@kiilerich.com>
parents:
17236
diff
changeset
|
494 - ASCII and UTF-8 data successfully round-trips and is understood |
16133
84c58da3a1f8
encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents:
15769
diff
changeset
|
495 by Unicode-oriented clients |
84c58da3a1f8
encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents:
15769
diff
changeset
|
496 - filenames and file contents in arbitrary other encodings can have |
84c58da3a1f8
encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents:
15769
diff
changeset
|
497 be round-tripped or recovered by clueful clients |
84c58da3a1f8
encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents:
15769
diff
changeset
|
498 - local strings that have a cached known UTF-8 encoding (aka |
84c58da3a1f8
encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents:
15769
diff
changeset
|
499 localstr) get sent as UTF-8 so Unicode-oriented clients get the |
84c58da3a1f8
encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents:
15769
diff
changeset
|
500 Unicode data they want |
84c58da3a1f8
encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents:
15769
diff
changeset
|
501 - because we must preserve UTF-8 bytestring in places such as |
84c58da3a1f8
encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents:
15769
diff
changeset
|
502 filenames, metadata can't be roundtripped without help |
84c58da3a1f8
encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents:
15769
diff
changeset
|
503 |
84c58da3a1f8
encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents:
15769
diff
changeset
|
504 (Note: "UTF-8b" often refers to decoding a mix of valid UTF-8 and |
84c58da3a1f8
encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents:
15769
diff
changeset
|
505 arbitrary bytes into an internal Unicode format that can be |
84c58da3a1f8
encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents:
15769
diff
changeset
|
506 re-encoded back into the original. Here we are exposing the |
84c58da3a1f8
encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents:
15769
diff
changeset
|
507 internal surrogate encoding as a UTF-8 string.) |
84c58da3a1f8
encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents:
15769
diff
changeset
|
508 ''' |
84c58da3a1f8
encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents:
15769
diff
changeset
|
509 |
26879
a24b98f4e03c
encoding: re-escape U+DCxx characters in toutf8b input (issue4927)
Matt Mackall <mpm@selenic.com>
parents:
26878
diff
changeset
|
510 if "\xed" not in s: |
a24b98f4e03c
encoding: re-escape U+DCxx characters in toutf8b input (issue4927)
Matt Mackall <mpm@selenic.com>
parents:
26878
diff
changeset
|
511 if isinstance(s, localstr): |
a24b98f4e03c
encoding: re-escape U+DCxx characters in toutf8b input (issue4927)
Matt Mackall <mpm@selenic.com>
parents:
26878
diff
changeset
|
512 return s._utf8 |
a24b98f4e03c
encoding: re-escape U+DCxx characters in toutf8b input (issue4927)
Matt Mackall <mpm@selenic.com>
parents:
26878
diff
changeset
|
513 try: |
a24b98f4e03c
encoding: re-escape U+DCxx characters in toutf8b input (issue4927)
Matt Mackall <mpm@selenic.com>
parents:
26878
diff
changeset
|
514 s.decode('utf-8') |
a24b98f4e03c
encoding: re-escape U+DCxx characters in toutf8b input (issue4927)
Matt Mackall <mpm@selenic.com>
parents:
26878
diff
changeset
|
515 return s |
a24b98f4e03c
encoding: re-escape U+DCxx characters in toutf8b input (issue4927)
Matt Mackall <mpm@selenic.com>
parents:
26878
diff
changeset
|
516 except UnicodeDecodeError: |
a24b98f4e03c
encoding: re-escape U+DCxx characters in toutf8b input (issue4927)
Matt Mackall <mpm@selenic.com>
parents:
26878
diff
changeset
|
517 pass |
26878
d7e83f106459
encoding: use getutf8char in toutf8b
Matt Mackall <mpm@selenic.com>
parents:
26877
diff
changeset
|
518 |
d7e83f106459
encoding: use getutf8char in toutf8b
Matt Mackall <mpm@selenic.com>
parents:
26877
diff
changeset
|
519 r = "" |
d7e83f106459
encoding: use getutf8char in toutf8b
Matt Mackall <mpm@selenic.com>
parents:
26877
diff
changeset
|
520 pos = 0 |
d7e83f106459
encoding: use getutf8char in toutf8b
Matt Mackall <mpm@selenic.com>
parents:
26877
diff
changeset
|
521 l = len(s) |
d7e83f106459
encoding: use getutf8char in toutf8b
Matt Mackall <mpm@selenic.com>
parents:
26877
diff
changeset
|
522 while pos < l: |
d7e83f106459
encoding: use getutf8char in toutf8b
Matt Mackall <mpm@selenic.com>
parents:
26877
diff
changeset
|
523 try: |
d7e83f106459
encoding: use getutf8char in toutf8b
Matt Mackall <mpm@selenic.com>
parents:
26877
diff
changeset
|
524 c = getutf8char(s, pos) |
26879
a24b98f4e03c
encoding: re-escape U+DCxx characters in toutf8b input (issue4927)
Matt Mackall <mpm@selenic.com>
parents:
26878
diff
changeset
|
525 if "\xed\xb0\x80" <= c <= "\xed\xb3\xbf": |
a24b98f4e03c
encoding: re-escape U+DCxx characters in toutf8b input (issue4927)
Matt Mackall <mpm@selenic.com>
parents:
26878
diff
changeset
|
526 # have to re-escape existing U+DCxx characters |
a24b98f4e03c
encoding: re-escape U+DCxx characters in toutf8b input (issue4927)
Matt Mackall <mpm@selenic.com>
parents:
26878
diff
changeset
|
527 c = unichr(0xdc00 + ord(s[pos])).encode('utf-8') |
a24b98f4e03c
encoding: re-escape U+DCxx characters in toutf8b input (issue4927)
Matt Mackall <mpm@selenic.com>
parents:
26878
diff
changeset
|
528 pos += 1 |
a24b98f4e03c
encoding: re-escape U+DCxx characters in toutf8b input (issue4927)
Matt Mackall <mpm@selenic.com>
parents:
26878
diff
changeset
|
529 else: |
a24b98f4e03c
encoding: re-escape U+DCxx characters in toutf8b input (issue4927)
Matt Mackall <mpm@selenic.com>
parents:
26878
diff
changeset
|
530 pos += len(c) |
26878
d7e83f106459
encoding: use getutf8char in toutf8b
Matt Mackall <mpm@selenic.com>
parents:
26877
diff
changeset
|
531 except UnicodeDecodeError: |
d7e83f106459
encoding: use getutf8char in toutf8b
Matt Mackall <mpm@selenic.com>
parents:
26877
diff
changeset
|
532 c = unichr(0xdc00 + ord(s[pos])).encode('utf-8') |
d7e83f106459
encoding: use getutf8char in toutf8b
Matt Mackall <mpm@selenic.com>
parents:
26877
diff
changeset
|
533 pos += 1 |
d7e83f106459
encoding: use getutf8char in toutf8b
Matt Mackall <mpm@selenic.com>
parents:
26877
diff
changeset
|
534 r += c |
d7e83f106459
encoding: use getutf8char in toutf8b
Matt Mackall <mpm@selenic.com>
parents:
26877
diff
changeset
|
535 return r |
16133
84c58da3a1f8
encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents:
15769
diff
changeset
|
536 |
84c58da3a1f8
encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents:
15769
diff
changeset
|
537 def fromutf8b(s): |
84c58da3a1f8
encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents:
15769
diff
changeset
|
538 '''Given a UTF-8b string, return a local, possibly-binary string. |
84c58da3a1f8
encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents:
15769
diff
changeset
|
539 |
84c58da3a1f8
encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents:
15769
diff
changeset
|
540 return the original binary string. This |
84c58da3a1f8
encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents:
15769
diff
changeset
|
541 is a round-trip process for strings like filenames, but metadata |
84c58da3a1f8
encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents:
15769
diff
changeset
|
542 that's was passed through tolocal will remain in UTF-8. |
84c58da3a1f8
encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents:
15769
diff
changeset
|
543 |
26963
de5ae97ce9f4
encoding: extend test cases for utf8b
Matt Mackall <mpm@selenic.com>
parents:
26879
diff
changeset
|
544 >>> roundtrip = lambda x: fromutf8b(toutf8b(x)) == x |
16133
84c58da3a1f8
encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents:
15769
diff
changeset
|
545 >>> m = "\\xc3\\xa9\\x99abcd" |
26963
de5ae97ce9f4
encoding: extend test cases for utf8b
Matt Mackall <mpm@selenic.com>
parents:
26879
diff
changeset
|
546 >>> toutf8b(m) |
16133
84c58da3a1f8
encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents:
15769
diff
changeset
|
547 '\\xc3\\xa9\\xed\\xb2\\x99abcd' |
26963
de5ae97ce9f4
encoding: extend test cases for utf8b
Matt Mackall <mpm@selenic.com>
parents:
26879
diff
changeset
|
548 >>> roundtrip(m) |
de5ae97ce9f4
encoding: extend test cases for utf8b
Matt Mackall <mpm@selenic.com>
parents:
26879
diff
changeset
|
549 True |
de5ae97ce9f4
encoding: extend test cases for utf8b
Matt Mackall <mpm@selenic.com>
parents:
26879
diff
changeset
|
550 >>> roundtrip("\\xc2\\xc2\\x80") |
de5ae97ce9f4
encoding: extend test cases for utf8b
Matt Mackall <mpm@selenic.com>
parents:
26879
diff
changeset
|
551 True |
de5ae97ce9f4
encoding: extend test cases for utf8b
Matt Mackall <mpm@selenic.com>
parents:
26879
diff
changeset
|
552 >>> roundtrip("\\xef\\xbf\\xbd") |
de5ae97ce9f4
encoding: extend test cases for utf8b
Matt Mackall <mpm@selenic.com>
parents:
26879
diff
changeset
|
553 True |
de5ae97ce9f4
encoding: extend test cases for utf8b
Matt Mackall <mpm@selenic.com>
parents:
26879
diff
changeset
|
554 >>> roundtrip("\\xef\\xef\\xbf\\xbd") |
16133
84c58da3a1f8
encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents:
15769
diff
changeset
|
555 True |
27699
c8d3392f76e1
encoding: handle UTF-16 internal limit with fromutf8b (issue5031)
Matt Mackall <mpm@selenic.com>
parents:
27356
diff
changeset
|
556 >>> roundtrip("\\xf1\\x80\\x80\\x80\\x80") |
c8d3392f76e1
encoding: handle UTF-16 internal limit with fromutf8b (issue5031)
Matt Mackall <mpm@selenic.com>
parents:
27356
diff
changeset
|
557 True |
16133
84c58da3a1f8
encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents:
15769
diff
changeset
|
558 ''' |
84c58da3a1f8
encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents:
15769
diff
changeset
|
559 |
84c58da3a1f8
encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents:
15769
diff
changeset
|
560 # fast path - look for uDxxx prefixes in s |
84c58da3a1f8
encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents:
15769
diff
changeset
|
561 if "\xed" not in s: |
84c58da3a1f8
encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents:
15769
diff
changeset
|
562 return s |
84c58da3a1f8
encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents:
15769
diff
changeset
|
563 |
27699
c8d3392f76e1
encoding: handle UTF-16 internal limit with fromutf8b (issue5031)
Matt Mackall <mpm@selenic.com>
parents:
27356
diff
changeset
|
564 # We could do this with the unicode type but some Python builds |
c8d3392f76e1
encoding: handle UTF-16 internal limit with fromutf8b (issue5031)
Matt Mackall <mpm@selenic.com>
parents:
27356
diff
changeset
|
565 # use UTF-16 internally (issue5031) which causes non-BMP code |
c8d3392f76e1
encoding: handle UTF-16 internal limit with fromutf8b (issue5031)
Matt Mackall <mpm@selenic.com>
parents:
27356
diff
changeset
|
566 # points to be escaped. Instead, we use our handy getutf8char |
c8d3392f76e1
encoding: handle UTF-16 internal limit with fromutf8b (issue5031)
Matt Mackall <mpm@selenic.com>
parents:
27356
diff
changeset
|
567 # helper again to walk the string without "decoding" it. |
c8d3392f76e1
encoding: handle UTF-16 internal limit with fromutf8b (issue5031)
Matt Mackall <mpm@selenic.com>
parents:
27356
diff
changeset
|
568 |
16133
84c58da3a1f8
encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents:
15769
diff
changeset
|
569 r = "" |
27699
c8d3392f76e1
encoding: handle UTF-16 internal limit with fromutf8b (issue5031)
Matt Mackall <mpm@selenic.com>
parents:
27356
diff
changeset
|
570 pos = 0 |
c8d3392f76e1
encoding: handle UTF-16 internal limit with fromutf8b (issue5031)
Matt Mackall <mpm@selenic.com>
parents:
27356
diff
changeset
|
571 l = len(s) |
c8d3392f76e1
encoding: handle UTF-16 internal limit with fromutf8b (issue5031)
Matt Mackall <mpm@selenic.com>
parents:
27356
diff
changeset
|
572 while pos < l: |
c8d3392f76e1
encoding: handle UTF-16 internal limit with fromutf8b (issue5031)
Matt Mackall <mpm@selenic.com>
parents:
27356
diff
changeset
|
573 c = getutf8char(s, pos) |
c8d3392f76e1
encoding: handle UTF-16 internal limit with fromutf8b (issue5031)
Matt Mackall <mpm@selenic.com>
parents:
27356
diff
changeset
|
574 pos += len(c) |
c8d3392f76e1
encoding: handle UTF-16 internal limit with fromutf8b (issue5031)
Matt Mackall <mpm@selenic.com>
parents:
27356
diff
changeset
|
575 # unescape U+DCxx characters |
c8d3392f76e1
encoding: handle UTF-16 internal limit with fromutf8b (issue5031)
Matt Mackall <mpm@selenic.com>
parents:
27356
diff
changeset
|
576 if "\xed\xb0\x80" <= c <= "\xed\xb3\xbf": |
c8d3392f76e1
encoding: handle UTF-16 internal limit with fromutf8b (issue5031)
Matt Mackall <mpm@selenic.com>
parents:
27356
diff
changeset
|
577 c = chr(ord(c.decode("utf-8")) & 0xff) |
c8d3392f76e1
encoding: handle UTF-16 internal limit with fromutf8b (issue5031)
Matt Mackall <mpm@selenic.com>
parents:
27356
diff
changeset
|
578 r += c |
16133
84c58da3a1f8
encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents:
15769
diff
changeset
|
579 return r |