|
1 """ |
|
2 encoding.py - character transcoding support for Mercurial |
|
3 |
|
4 Copyright 2005-2009 Matt Mackall <mpm@selenic.com> and others |
|
5 |
|
6 This software may be used and distributed according to the terms of |
|
7 the GNU General Public License version 2, incorporated herein by |
|
8 reference. |
|
9 """ |
|
10 |
|
11 import sys, unicodedata, locale, os, error |
|
12 |
|
13 _encodingfixup = {'646': 'ascii', 'ANSI_X3.4-1968': 'ascii'} |
|
14 |
|
15 try: |
|
16 encoding = os.environ.get("HGENCODING") |
|
17 if sys.platform == 'darwin' and not encoding: |
|
18 # On darwin, getpreferredencoding ignores the locale environment and |
|
19 # always returns mac-roman. We override this if the environment is |
|
20 # not C (has been customized by the user). |
|
21 locale.setlocale(locale.LC_CTYPE, '') |
|
22 encoding = locale.getlocale()[1] |
|
23 if not encoding: |
|
24 encoding = locale.getpreferredencoding() or 'ascii' |
|
25 encoding = _encodingfixup.get(encoding, encoding) |
|
26 except locale.Error: |
|
27 encoding = 'ascii' |
|
28 encodingmode = os.environ.get("HGENCODINGMODE", "strict") |
|
29 fallbackencoding = 'ISO-8859-1' |
|
30 |
|
31 def tolocal(s): |
|
32 """ |
|
33 Convert a string from internal UTF-8 to local encoding |
|
34 |
|
35 All internal strings should be UTF-8 but some repos before the |
|
36 implementation of locale support may contain latin1 or possibly |
|
37 other character sets. We attempt to decode everything strictly |
|
38 using UTF-8, then Latin-1, and failing that, we use UTF-8 and |
|
39 replace unknown characters. |
|
40 """ |
|
41 for e in ('UTF-8', fallbackencoding): |
|
42 try: |
|
43 u = s.decode(e) # attempt strict decoding |
|
44 return u.encode(encoding, "replace") |
|
45 except LookupError, k: |
|
46 raise error.Abort("%s, please check your locale settings" % k) |
|
47 except UnicodeDecodeError: |
|
48 pass |
|
49 u = s.decode("utf-8", "replace") # last ditch |
|
50 return u.encode(encoding, "replace") |
|
51 |
|
52 def fromlocal(s): |
|
53 """ |
|
54 Convert a string from the local character encoding to UTF-8 |
|
55 |
|
56 We attempt to decode strings using the encoding mode set by |
|
57 HGENCODINGMODE, which defaults to 'strict'. In this mode, unknown |
|
58 characters will cause an error message. Other modes include |
|
59 'replace', which replaces unknown characters with a special |
|
60 Unicode character, and 'ignore', which drops the character. |
|
61 """ |
|
62 try: |
|
63 return s.decode(encoding, encodingmode).encode("utf-8") |
|
64 except UnicodeDecodeError, inst: |
|
65 sub = s[max(0, inst.start-10):inst.start+10] |
|
66 raise error.Abort("decoding near '%s': %s!" % (sub, inst)) |
|
67 except LookupError, k: |
|
68 raise error.Abort("%s, please check your locale settings" % k) |
|
69 |
|
70 def colwidth(s): |
|
71 "Find the column width of a UTF-8 string for display" |
|
72 d = s.decode(encoding, 'replace') |
|
73 if hasattr(unicodedata, 'east_asian_width'): |
|
74 w = unicodedata.east_asian_width |
|
75 return sum([w(c) in 'WF' and 2 or 1 for c in d]) |
|
76 return len(d) |
|
77 |