mercurial/encoding.py
author Bryan O'Sullivan <bos@serpentine.com>
Sun, 27 Dec 2015 23:55:54 +0900
changeset 27524 f5b6b4e574c1
parent 27356 c2effd1ecebf
child 27699 c8d3392f76e1
permissions -rw-r--r--
eol: make output stable This eliminates a divergence in behaviour between PyPy and Python.
Ignore whitespace changes - Everywhere: Within whitespace: At end of lines:
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
25e572394f5c Update license to GPLv2+
Matt Mackall <mpm@selenic.com>
parents: 9574
diff changeset
     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
b479fc425a81 encoding: use absolute_import
Gregory Szorc <gregory.szorc@gmail.com>
parents: 26963
diff changeset
    10
import locale
b479fc425a81 encoding: use absolute_import
Gregory Szorc <gregory.szorc@gmail.com>
parents: 26963
diff changeset
    11
import os
b479fc425a81 encoding: use absolute_import
Gregory Szorc <gregory.szorc@gmail.com>
parents: 26963
diff changeset
    12
import unicodedata
b479fc425a81 encoding: use absolute_import
Gregory Szorc <gregory.szorc@gmail.com>
parents: 26963
diff changeset
    13
b479fc425a81 encoding: use absolute_import
Gregory Szorc <gregory.szorc@gmail.com>
parents: 26963
diff changeset
    14
from . import (
b479fc425a81 encoding: use absolute_import
Gregory Szorc <gregory.szorc@gmail.com>
parents: 26963
diff changeset
    15
    error,
b479fc425a81 encoding: use absolute_import
Gregory Szorc <gregory.szorc@gmail.com>
parents: 26963
diff changeset
    16
)
7948
de377b1a9a84 move encoding bits from util to encoding
Matt Mackall <mpm@selenic.com>
parents:
diff changeset
    17
23596
885bd7c5c7e3 encoding: add hfsignoreclean to clean out HFS-ignored characters
Augie Fackler <raf@durin42.com>
parents: 22973
diff changeset
    18
# 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
    19
# "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
    20
# sanity.
885bd7c5c7e3 encoding: add hfsignoreclean to clean out HFS-ignored characters
Augie Fackler <raf@durin42.com>
parents: 22973
diff changeset
    21
_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
    22
           "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
    23
           "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
    24
# verify the next function will work
885bd7c5c7e3 encoding: add hfsignoreclean to clean out HFS-ignored characters
Augie Fackler <raf@durin42.com>
parents: 22973
diff changeset
    25
assert set([i[0] for i in _ignore]) == set(["\xe2", "\xef"])
885bd7c5c7e3 encoding: add hfsignoreclean to clean out HFS-ignored characters
Augie Fackler <raf@durin42.com>
parents: 22973
diff changeset
    26
885bd7c5c7e3 encoding: add hfsignoreclean to clean out HFS-ignored characters
Augie Fackler <raf@durin42.com>
parents: 22973
diff changeset
    27
def hfsignoreclean(s):
885bd7c5c7e3 encoding: add hfsignoreclean to clean out HFS-ignored characters
Augie Fackler <raf@durin42.com>
parents: 22973
diff changeset
    28
    """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
    29
885bd7c5c7e3 encoding: add hfsignoreclean to clean out HFS-ignored characters
Augie Fackler <raf@durin42.com>
parents: 22973
diff changeset
    30
    >>> 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
    31
    '.hg'
885bd7c5c7e3 encoding: add hfsignoreclean to clean out HFS-ignored characters
Augie Fackler <raf@durin42.com>
parents: 22973
diff changeset
    32
    >>> 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
    33
    '.hg'
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
    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
    36
        for c in _ignore:
885bd7c5c7e3 encoding: add hfsignoreclean to clean out HFS-ignored characters
Augie Fackler <raf@durin42.com>
parents: 22973
diff changeset
    37
            s = s.replace(c, '')
885bd7c5c7e3 encoding: add hfsignoreclean to clean out HFS-ignored characters
Augie Fackler <raf@durin42.com>
parents: 22973
diff changeset
    38
    return s
885bd7c5c7e3 encoding: add hfsignoreclean to clean out HFS-ignored characters
Augie Fackler <raf@durin42.com>
parents: 22973
diff changeset
    39
11892
2be70ca17311 encoding: improve handling of buggy getpreferredencoding() on Mac OS X
Dan Villiom Podlaski Christiansen <danchr@gmail.com>
parents: 11297
diff changeset
    40
def _getpreferredencoding():
2be70ca17311 encoding: improve handling of buggy getpreferredencoding() on Mac OS X
Dan Villiom Podlaski Christiansen <danchr@gmail.com>
parents: 11297
diff changeset
    41
    '''
2be70ca17311 encoding: improve handling of buggy getpreferredencoding() on Mac OS X
Dan Villiom Podlaski Christiansen <danchr@gmail.com>
parents: 11297
diff changeset
    42
    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
    43
    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
    44
    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
    45
    Python versions.
2be70ca17311 encoding: improve handling of buggy getpreferredencoding() on Mac OS X
Dan Villiom Podlaski Christiansen <danchr@gmail.com>
parents: 11297
diff changeset
    46
12770
614f0d8724ab check-code: find trailing whitespace
Martin Geisler <mg@lazybytes.net>
parents: 12062
diff changeset
    47
    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
    48
    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
    49
    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
    50
    '''
2be70ca17311 encoding: improve handling of buggy getpreferredencoding() on Mac OS X
Dan Villiom Podlaski Christiansen <danchr@gmail.com>
parents: 11297
diff changeset
    51
    try:
2be70ca17311 encoding: improve handling of buggy getpreferredencoding() on Mac OS X
Dan Villiom Podlaski Christiansen <danchr@gmail.com>
parents: 11297
diff changeset
    52
        locale.CODESET
2be70ca17311 encoding: improve handling of buggy getpreferredencoding() on Mac OS X
Dan Villiom Podlaski Christiansen <danchr@gmail.com>
parents: 11297
diff changeset
    53
    except AttributeError:
2be70ca17311 encoding: improve handling of buggy getpreferredencoding() on Mac OS X
Dan Villiom Podlaski Christiansen <danchr@gmail.com>
parents: 11297
diff changeset
    54
        # 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
    55
        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
    56
2be70ca17311 encoding: improve handling of buggy getpreferredencoding() on Mac OS X
Dan Villiom Podlaski Christiansen <danchr@gmail.com>
parents: 11297
diff changeset
    57
    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
    58
    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
    59
    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
    60
    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
    61
2be70ca17311 encoding: improve handling of buggy getpreferredencoding() on Mac OS X
Dan Villiom Podlaski Christiansen <danchr@gmail.com>
parents: 11297
diff changeset
    62
    return result
2be70ca17311 encoding: improve handling of buggy getpreferredencoding() on Mac OS X
Dan Villiom Podlaski Christiansen <danchr@gmail.com>
parents: 11297
diff changeset
    63
2be70ca17311 encoding: improve handling of buggy getpreferredencoding() on Mac OS X
Dan Villiom Podlaski Christiansen <danchr@gmail.com>
parents: 11297
diff changeset
    64
_encodingfixers = {
2be70ca17311 encoding: improve handling of buggy getpreferredencoding() on Mac OS X
Dan Villiom Podlaski Christiansen <danchr@gmail.com>
parents: 11297
diff changeset
    65
    '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
    66
    '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
    67
    '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
    68
}
7948
de377b1a9a84 move encoding bits from util to encoding
Matt Mackall <mpm@selenic.com>
parents:
diff changeset
    69
de377b1a9a84 move encoding bits from util to encoding
Matt Mackall <mpm@selenic.com>
parents:
diff changeset
    70
try:
de377b1a9a84 move encoding bits from util to encoding
Matt Mackall <mpm@selenic.com>
parents:
diff changeset
    71
    encoding = os.environ.get("HGENCODING")
de377b1a9a84 move encoding bits from util to encoding
Matt Mackall <mpm@selenic.com>
parents:
diff changeset
    72
    if not encoding:
de377b1a9a84 move encoding bits from util to encoding
Matt Mackall <mpm@selenic.com>
parents:
diff changeset
    73
        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
    74
        encoding = _encodingfixers.get(encoding, lambda: encoding)()
7948
de377b1a9a84 move encoding bits from util to encoding
Matt Mackall <mpm@selenic.com>
parents:
diff changeset
    75
except locale.Error:
de377b1a9a84 move encoding bits from util to encoding
Matt Mackall <mpm@selenic.com>
parents:
diff changeset
    76
    encoding = 'ascii'
de377b1a9a84 move encoding bits from util to encoding
Matt Mackall <mpm@selenic.com>
parents:
diff changeset
    77
encodingmode = os.environ.get("HGENCODINGMODE", "strict")
de377b1a9a84 move encoding bits from util to encoding
Matt Mackall <mpm@selenic.com>
parents:
diff changeset
    78
fallbackencoding = 'ISO-8859-1'
de377b1a9a84 move encoding bits from util to encoding
Matt Mackall <mpm@selenic.com>
parents:
diff changeset
    79
13046
7cc4263e07a9 encoding: add localstr class to track UTF-8 version of transcoded strings
Matt Mackall <mpm@selenic.com>
parents: 12866
diff changeset
    80
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
    81
    '''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
    82
    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
    83
    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
    84
        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
    85
        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
    86
        return s
7cc4263e07a9 encoding: add localstr class to track UTF-8 version of transcoded strings
Matt Mackall <mpm@selenic.com>
parents: 12866
diff changeset
    87
    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
    88
        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
    89
7948
de377b1a9a84 move encoding bits from util to encoding
Matt Mackall <mpm@selenic.com>
parents:
diff changeset
    90
def tolocal(s):
de377b1a9a84 move encoding bits from util to encoding
Matt Mackall <mpm@selenic.com>
parents:
diff changeset
    91
    """
de377b1a9a84 move encoding bits from util to encoding
Matt Mackall <mpm@selenic.com>
parents:
diff changeset
    92
    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
    93
de377b1a9a84 move encoding bits from util to encoding
Matt Mackall <mpm@selenic.com>
parents:
diff changeset
    94
    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
    95
    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
    96
    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
    97
    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
    98
    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
    99
7cc4263e07a9 encoding: add localstr class to track UTF-8 version of transcoded strings
Matt Mackall <mpm@selenic.com>
parents: 12866
diff changeset
   100
    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
   101
    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
   102
    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
   103
7cc4263e07a9 encoding: add localstr class to track UTF-8 version of transcoded strings
Matt Mackall <mpm@selenic.com>
parents: 12866
diff changeset
   104
    >>> 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
   105
    >>> 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
   106
    >>> l
7cc4263e07a9 encoding: add localstr class to track UTF-8 version of transcoded strings
Matt Mackall <mpm@selenic.com>
parents: 12866
diff changeset
   107
    'foo: ?'
7cc4263e07a9 encoding: add localstr class to track UTF-8 version of transcoded strings
Matt Mackall <mpm@selenic.com>
parents: 12866
diff changeset
   108
    >>> fromlocal(l)
7cc4263e07a9 encoding: add localstr class to track UTF-8 version of transcoded strings
Matt Mackall <mpm@selenic.com>
parents: 12866
diff changeset
   109
    '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
   110
    >>> 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
   111
    >>> d = { l: 1, tolocal(u2): 2 }
18378
404feac78b8a tests: stabilize doctest output
Mads Kiilerich <mads@kiilerich.com>
parents: 17424
diff changeset
   112
    >>> len(d) # no collision
404feac78b8a tests: stabilize doctest output
Mads Kiilerich <mads@kiilerich.com>
parents: 17424
diff changeset
   113
    2
13046
7cc4263e07a9 encoding: add localstr class to track UTF-8 version of transcoded strings
Matt Mackall <mpm@selenic.com>
parents: 12866
diff changeset
   114
    >>> '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
   115
    False
7cc4263e07a9 encoding: add localstr class to track UTF-8 version of transcoded strings
Matt Mackall <mpm@selenic.com>
parents: 12866
diff changeset
   116
    >>> 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
   117
    >>> 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
   118
    >>> l
7cc4263e07a9 encoding: add localstr class to track UTF-8 version of transcoded strings
Matt Mackall <mpm@selenic.com>
parents: 12866
diff changeset
   119
    'foo: ?'
7cc4263e07a9 encoding: add localstr class to track UTF-8 version of transcoded strings
Matt Mackall <mpm@selenic.com>
parents: 12866
diff changeset
   120
    >>> 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
   121
    'foo: \\xc3\\xa4'
7948
de377b1a9a84 move encoding bits from util to encoding
Matt Mackall <mpm@selenic.com>
parents:
diff changeset
   122
    """
13046
7cc4263e07a9 encoding: add localstr class to track UTF-8 version of transcoded strings
Matt Mackall <mpm@selenic.com>
parents: 12866
diff changeset
   123
16274
5d75eb8568d1 encoding: tune fast-path of tolocal a bit
Matt Mackall <mpm@selenic.com>
parents: 16133
diff changeset
   124
    try:
7948
de377b1a9a84 move encoding bits from util to encoding
Matt Mackall <mpm@selenic.com>
parents:
diff changeset
   125
        try:
16274
5d75eb8568d1 encoding: tune fast-path of tolocal a bit
Matt Mackall <mpm@selenic.com>
parents: 16133
diff changeset
   126
            # 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
   127
            u = s.decode('UTF-8')
5d75eb8568d1 encoding: tune fast-path of tolocal a bit
Matt Mackall <mpm@selenic.com>
parents: 16133
diff changeset
   128
            if encoding == 'UTF-8':
5d75eb8568d1 encoding: tune fast-path of tolocal a bit
Matt Mackall <mpm@selenic.com>
parents: 16133
diff changeset
   129
                # fast path
5d75eb8568d1 encoding: tune fast-path of tolocal a bit
Matt Mackall <mpm@selenic.com>
parents: 16133
diff changeset
   130
                return s
13940
b7b26e54e37a encoding: avoid localstr when a string can be encoded losslessly (issue2763)
Matt Mackall <mpm@selenic.com>
parents: 13051
diff changeset
   131
            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
   132
            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
   133
                # 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
   134
                return r
16274
5d75eb8568d1 encoding: tune fast-path of tolocal a bit
Matt Mackall <mpm@selenic.com>
parents: 16133
diff changeset
   135
            return localstr(s, r)
5d75eb8568d1 encoding: tune fast-path of tolocal a bit
Matt Mackall <mpm@selenic.com>
parents: 16133
diff changeset
   136
        except UnicodeDecodeError:
5d75eb8568d1 encoding: tune fast-path of tolocal a bit
Matt Mackall <mpm@selenic.com>
parents: 16133
diff changeset
   137
            # 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
   138
            try:
5d75eb8568d1 encoding: tune fast-path of tolocal a bit
Matt Mackall <mpm@selenic.com>
parents: 16133
diff changeset
   139
                u = s.decode(fallbackencoding)
5d75eb8568d1 encoding: tune fast-path of tolocal a bit
Matt Mackall <mpm@selenic.com>
parents: 16133
diff changeset
   140
                r = u.encode(encoding, "replace")
5d75eb8568d1 encoding: tune fast-path of tolocal a bit
Matt Mackall <mpm@selenic.com>
parents: 16133
diff changeset
   141
                if u == r.decode(encoding):
5d75eb8568d1 encoding: tune fast-path of tolocal a bit
Matt Mackall <mpm@selenic.com>
parents: 16133
diff changeset
   142
                    # 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
   143
                    return r
13940
b7b26e54e37a encoding: avoid localstr when a string can be encoded losslessly (issue2763)
Matt Mackall <mpm@selenic.com>
parents: 13051
diff changeset
   144
                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
   145
            except UnicodeDecodeError:
5d75eb8568d1 encoding: tune fast-path of tolocal a bit
Matt Mackall <mpm@selenic.com>
parents: 16133
diff changeset
   146
                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
   147
                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
   148
    except LookupError as k:
16274
5d75eb8568d1 encoding: tune fast-path of tolocal a bit
Matt Mackall <mpm@selenic.com>
parents: 16133
diff changeset
   149
        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
   150
de377b1a9a84 move encoding bits from util to encoding
Matt Mackall <mpm@selenic.com>
parents:
diff changeset
   151
def fromlocal(s):
de377b1a9a84 move encoding bits from util to encoding
Matt Mackall <mpm@selenic.com>
parents:
diff changeset
   152
    """
de377b1a9a84 move encoding bits from util to encoding
Matt Mackall <mpm@selenic.com>
parents:
diff changeset
   153
    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
   154
de377b1a9a84 move encoding bits from util to encoding
Matt Mackall <mpm@selenic.com>
parents:
diff changeset
   155
    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
   156
    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
   157
    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
   158
    'replace', which replaces unknown characters with a special
de377b1a9a84 move encoding bits from util to encoding
Matt Mackall <mpm@selenic.com>
parents:
diff changeset
   159
    Unicode character, and 'ignore', which drops the character.
de377b1a9a84 move encoding bits from util to encoding
Matt Mackall <mpm@selenic.com>
parents:
diff changeset
   160
    """
13046
7cc4263e07a9 encoding: add localstr class to track UTF-8 version of transcoded strings
Matt Mackall <mpm@selenic.com>
parents: 12866
diff changeset
   161
7cc4263e07a9 encoding: add localstr class to track UTF-8 version of transcoded strings
Matt Mackall <mpm@selenic.com>
parents: 12866
diff changeset
   162
    # 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
   163
    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
   164
        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
   165
7948
de377b1a9a84 move encoding bits from util to encoding
Matt Mackall <mpm@selenic.com>
parents:
diff changeset
   166
    try:
de377b1a9a84 move encoding bits from util to encoding
Matt Mackall <mpm@selenic.com>
parents:
diff changeset
   167
        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
   168
    except UnicodeDecodeError as inst:
10282
08a0f04b56bd many, many trivial check-code fixups
Matt Mackall <mpm@selenic.com>
parents: 10263
diff changeset
   169
        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
   170
        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
   171
    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
   172
        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
   173
12866
eddc20306ab6 encoding: default ambiguous character to narrow
Matt Mackall <mpm@selenic.com>
parents: 12770
diff changeset
   174
# 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
   175
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
   176
        and "WFA" or "WF")
12866
eddc20306ab6 encoding: default ambiguous character to narrow
Matt Mackall <mpm@selenic.com>
parents: 12770
diff changeset
   177
7948
de377b1a9a84 move encoding bits from util to encoding
Matt Mackall <mpm@selenic.com>
parents:
diff changeset
   178
def colwidth(s):
15142
176882876780 encoding: colwidth input is in the local encoding
Matt Mackall <mpm@selenic.com>
parents: 15066
diff changeset
   179
    "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
   180
    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
   181
24efa83d81cb i18n: calculate terminal columns by width information of each characters
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents: 14951
diff changeset
   182
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
   183
    "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
   184
    eaw = getattr(unicodedata, 'east_asian_width', None)
61807854004e encoding: use getattr isntead of hasattr
Augie Fackler <durin42@gmail.com>
parents: 14069
diff changeset
   185
    if eaw is not None:
61807854004e encoding: use getattr isntead of hasattr
Augie Fackler <durin42@gmail.com>
parents: 14069
diff changeset
   186
        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
   187
    return len(d)
de377b1a9a84 move encoding bits from util to encoding
Matt Mackall <mpm@selenic.com>
parents:
diff changeset
   188
15143
16c129b0f465 encoding: add getcols to extract substrings based on column width
Matt Mackall <mpm@selenic.com>
parents: 15142
diff changeset
   189
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
   190
    '''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
   191
    index start'''
16c129b0f465 encoding: add getcols to extract substrings based on column width
Matt Mackall <mpm@selenic.com>
parents: 15142
diff changeset
   192
    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
   193
        t = s[start:x]
16c129b0f465 encoding: add getcols to extract substrings based on column width
Matt Mackall <mpm@selenic.com>
parents: 15142
diff changeset
   194
        if colwidth(t) == c:
16c129b0f465 encoding: add getcols to extract substrings based on column width
Matt Mackall <mpm@selenic.com>
parents: 15142
diff changeset
   195
            return t
16c129b0f465 encoding: add getcols to extract substrings based on column width
Matt Mackall <mpm@selenic.com>
parents: 15142
diff changeset
   196
21861
b515c3a63e96 encoding: add 'leftside' argument into 'trim' to switch trimming side
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents: 21856
diff changeset
   197
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
   198
    """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
   199
21861
b515c3a63e96 encoding: add 'leftside' argument into 'trim' to switch trimming side
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents: 21856
diff changeset
   200
    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
   201
    '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
   202
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
   203
    >>> ellipsis = '+++'
27355
b479fc425a81 encoding: use absolute_import
Gregory Szorc <gregory.szorc@gmail.com>
parents: 26963
diff changeset
   204
    >>> 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
   205
    >>> 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
   206
    >>> 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
   207
    >>> 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
   208
    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
   209
    >>> 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
   210
    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
   211
    >>> 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
   212
    12345+++
21861
b515c3a63e96 encoding: add 'leftside' argument into 'trim' to switch trimming side
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents: 21856
diff changeset
   213
    >>> 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
   214
    +++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
   215
    >>> 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
   216
    12345678
21861
b515c3a63e96 encoding: add 'leftside' argument into 'trim' to switch trimming side
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents: 21856
diff changeset
   217
    >>> 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
   218
    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
   219
    >>> 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
   220
    +++
d24969ee272f encoding: add 'trim' to trim multi-byte characters at most specified columns
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents: 18378
diff changeset
   221
    >>> 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
   222
    +
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
    >>> 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
   224
    >>> 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
   225
    >>> 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
   226
    \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
   227
    >>> 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
   228
    \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
   229
    >>> 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
   230
    \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
   231
    >>> 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
   232
    +++\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
   233
    >>> 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
   234
    \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
   235
    >>> 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
   236
    \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
   237
    >>> 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
   238
    +++
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, 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
   240
    +++
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
    >>> 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
   242
    >>> 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
   243
    \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
   244
    >>> 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
   245
    \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
   246
    >>> 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
   247
    \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
   248
    >>> 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
   249
    +++\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
   250
    >>> 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
   251
    \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
   252
    >>> 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
   253
    \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
   254
    >>> 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
   255
    +++
d24969ee272f encoding: add 'trim' to trim multi-byte characters at most specified columns
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents: 18378
diff changeset
   256
    >>> 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
   257
    +
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
    """
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
    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
   260
        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
   261
    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
   262
        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
   263
            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
   264
        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
   265
        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
   266
            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
   267
        if leftside:
b515c3a63e96 encoding: add 'leftside' argument into 'trim' to switch trimming side
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents: 21856
diff changeset
   268
            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
   269
        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
   270
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
    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
   272
        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
   273
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
    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
   275
    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
   276
        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
   277
21861
b515c3a63e96 encoding: add 'leftside' argument into 'trim' to switch trimming side
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents: 21856
diff changeset
   278
    if leftside:
b515c3a63e96 encoding: add 'leftside' argument into 'trim' to switch trimming side
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents: 21856
diff changeset
   279
        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
   280
        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
   281
    else:
b515c3a63e96 encoding: add 'leftside' argument into 'trim' to switch trimming side
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents: 21856
diff changeset
   282
        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
   283
        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
   284
    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
   285
        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
   286
        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
   287
            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
   288
    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
   289
22973
bcff9ecdaae0 encoding: avoid cyclic dependency around "parsers" in pure Python build
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents: 22779
diff changeset
   290
def _asciilower(s):
22778
80f2b63dd83a parsers: add a function to efficiently lowercase ASCII strings
Siddharth Agarwal <sid0@fb.com>
parents: 22426
diff changeset
   291
    '''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
   292
80f2b63dd83a parsers: add a function to efficiently lowercase ASCII strings
Siddharth Agarwal <sid0@fb.com>
parents: 22426
diff changeset
   293
    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
   294
    s.decode('ascii')
80f2b63dd83a parsers: add a function to efficiently lowercase ASCII strings
Siddharth Agarwal <sid0@fb.com>
parents: 22426
diff changeset
   295
    return s.lower()
80f2b63dd83a parsers: add a function to efficiently lowercase ASCII strings
Siddharth Agarwal <sid0@fb.com>
parents: 22426
diff changeset
   296
22973
bcff9ecdaae0 encoding: avoid cyclic dependency around "parsers" in pure Python build
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents: 22779
diff changeset
   297
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
   298
    # 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
   299
    # pure Python build (util => i18n => encoding => parsers => util)
27355
b479fc425a81 encoding: use absolute_import
Gregory Szorc <gregory.szorc@gmail.com>
parents: 26963
diff changeset
   300
    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
   301
    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
   302
    global asciilower
bcff9ecdaae0 encoding: avoid cyclic dependency around "parsers" in pure Python build
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents: 22779
diff changeset
   303
    asciilower = impl
bcff9ecdaae0 encoding: avoid cyclic dependency around "parsers" in pure Python build
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents: 22779
diff changeset
   304
    return impl(s)
22778
80f2b63dd83a parsers: add a function to efficiently lowercase ASCII strings
Siddharth Agarwal <sid0@fb.com>
parents: 22426
diff changeset
   305
24578
ac08de78de7f encoding: use parsers.asciiupper when available
Siddharth Agarwal <sid0@fb.com>
parents: 23596
diff changeset
   306
def _asciiupper(s):
ac08de78de7f encoding: use parsers.asciiupper when available
Siddharth Agarwal <sid0@fb.com>
parents: 23596
diff changeset
   307
    '''convert a string to uppercase if ASCII
ac08de78de7f encoding: use parsers.asciiupper when available
Siddharth Agarwal <sid0@fb.com>
parents: 23596
diff changeset
   308
ac08de78de7f encoding: use parsers.asciiupper when available
Siddharth Agarwal <sid0@fb.com>
parents: 23596
diff changeset
   309
    Raises UnicodeDecodeError if non-ASCII characters are found.'''
ac08de78de7f encoding: use parsers.asciiupper when available
Siddharth Agarwal <sid0@fb.com>
parents: 23596
diff changeset
   310
    s.decode('ascii')
ac08de78de7f encoding: use parsers.asciiupper when available
Siddharth Agarwal <sid0@fb.com>
parents: 23596
diff changeset
   311
    return s.upper()
ac08de78de7f encoding: use parsers.asciiupper when available
Siddharth Agarwal <sid0@fb.com>
parents: 23596
diff changeset
   312
ac08de78de7f encoding: use parsers.asciiupper when available
Siddharth Agarwal <sid0@fb.com>
parents: 23596
diff changeset
   313
def asciiupper(s):
ac08de78de7f encoding: use parsers.asciiupper when available
Siddharth Agarwal <sid0@fb.com>
parents: 23596
diff changeset
   314
    # delay importing avoids cyclic dependency around "parsers" in
ac08de78de7f encoding: use parsers.asciiupper when available
Siddharth Agarwal <sid0@fb.com>
parents: 23596
diff changeset
   315
    # pure Python build (util => i18n => encoding => parsers => util)
27355
b479fc425a81 encoding: use absolute_import
Gregory Szorc <gregory.szorc@gmail.com>
parents: 26963
diff changeset
   316
    from . import parsers
24578
ac08de78de7f encoding: use parsers.asciiupper when available
Siddharth Agarwal <sid0@fb.com>
parents: 23596
diff changeset
   317
    impl = getattr(parsers, 'asciiupper', _asciiupper)
ac08de78de7f encoding: use parsers.asciiupper when available
Siddharth Agarwal <sid0@fb.com>
parents: 23596
diff changeset
   318
    global asciiupper
ac08de78de7f encoding: use parsers.asciiupper when available
Siddharth Agarwal <sid0@fb.com>
parents: 23596
diff changeset
   319
    asciiupper = impl
ac08de78de7f encoding: use parsers.asciiupper when available
Siddharth Agarwal <sid0@fb.com>
parents: 23596
diff changeset
   320
    return impl(s)
ac08de78de7f encoding: use parsers.asciiupper when available
Siddharth Agarwal <sid0@fb.com>
parents: 23596
diff changeset
   321
14069
e38846a79a23 encoding: add an encoding-aware lower function
Matt Mackall <mpm@selenic.com>
parents: 13940
diff changeset
   322
def lower(s):
e38846a79a23 encoding: add an encoding-aware lower function
Matt Mackall <mpm@selenic.com>
parents: 13940
diff changeset
   323
    "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
   324
    try:
22779
d9585dda63c3 encoding.lower: use fast ASCII lower
Siddharth Agarwal <sid0@fb.com>
parents: 22778
diff changeset
   325
        return asciilower(s)
17235
3745ae495ce5 encoding: use s.decode to trigger UnicodeDecodeError
Martin Geisler <mg@aragost.com>
parents: 16493
diff changeset
   326
    except UnicodeDecodeError:
16387
c481761033bd encoding: add fast-path for ASCII lowercase
Matt Mackall <mpm@selenic.com>
parents: 16274
diff changeset
   327
        pass
c481761033bd encoding: add fast-path for ASCII lowercase
Matt Mackall <mpm@selenic.com>
parents: 16274
diff changeset
   328
    try:
14069
e38846a79a23 encoding: add an encoding-aware lower function
Matt Mackall <mpm@selenic.com>
parents: 13940
diff changeset
   329
        if isinstance(s, localstr):
e38846a79a23 encoding: add an encoding-aware lower function
Matt Mackall <mpm@selenic.com>
parents: 13940
diff changeset
   330
            u = s._utf8.decode("utf-8")
e38846a79a23 encoding: add an encoding-aware lower function
Matt Mackall <mpm@selenic.com>
parents: 13940
diff changeset
   331
        else:
e38846a79a23 encoding: add an encoding-aware lower function
Matt Mackall <mpm@selenic.com>
parents: 13940
diff changeset
   332
            u = s.decode(encoding, encodingmode)
e38846a79a23 encoding: add an encoding-aware lower function
Matt Mackall <mpm@selenic.com>
parents: 13940
diff changeset
   333
e38846a79a23 encoding: add an encoding-aware lower function
Matt Mackall <mpm@selenic.com>
parents: 13940
diff changeset
   334
        lu = u.lower()
e38846a79a23 encoding: add an encoding-aware lower function
Matt Mackall <mpm@selenic.com>
parents: 13940
diff changeset
   335
        if u == lu:
e38846a79a23 encoding: add an encoding-aware lower function
Matt Mackall <mpm@selenic.com>
parents: 13940
diff changeset
   336
            return s # preserve localstring
e38846a79a23 encoding: add an encoding-aware lower function
Matt Mackall <mpm@selenic.com>
parents: 13940
diff changeset
   337
        return lu.encode(encoding)
e38846a79a23 encoding: add an encoding-aware lower function
Matt Mackall <mpm@selenic.com>
parents: 13940
diff changeset
   338
    except UnicodeError:
e38846a79a23 encoding: add an encoding-aware lower function
Matt Mackall <mpm@selenic.com>
parents: 13940
diff changeset
   339
        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
   340
    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
   341
        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
   342
2ebe3d0ce91d i18n: use encoding.lower/upper for encoding aware case folding
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents: 15143
diff changeset
   343
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
   344
    "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
   345
    try:
24578
ac08de78de7f encoding: use parsers.asciiupper when available
Siddharth Agarwal <sid0@fb.com>
parents: 23596
diff changeset
   346
        return asciiupper(s)
17236
9fb8312dbdbd encoding: add fast-path for ASCII uppercase.
Martin Geisler <mg@aragost.com>
parents: 17235
diff changeset
   347
    except UnicodeDecodeError:
24597
b4258d5a1600 encoding.upper: factor out fallback code
Siddharth Agarwal <sid0@fb.com>
parents: 24593
diff changeset
   348
        return upperfallback(s)
b4258d5a1600 encoding.upper: factor out fallback code
Siddharth Agarwal <sid0@fb.com>
parents: 24593
diff changeset
   349
b4258d5a1600 encoding.upper: factor out fallback code
Siddharth Agarwal <sid0@fb.com>
parents: 24593
diff changeset
   350
def upperfallback(s):
17236
9fb8312dbdbd encoding: add fast-path for ASCII uppercase.
Martin Geisler <mg@aragost.com>
parents: 17235
diff changeset
   351
    try:
15672
2ebe3d0ce91d i18n: use encoding.lower/upper for encoding aware case folding
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents: 15143
diff changeset
   352
        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
   353
            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
   354
        else:
2ebe3d0ce91d i18n: use encoding.lower/upper for encoding aware case folding
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents: 15143
diff changeset
   355
            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
   356
2ebe3d0ce91d i18n: use encoding.lower/upper for encoding aware case folding
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents: 15143
diff changeset
   357
        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
   358
        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
   359
            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
   360
        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
   361
    except UnicodeError:
2ebe3d0ce91d i18n: use encoding.lower/upper for encoding aware case folding
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents: 15143
diff changeset
   362
        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
   363
    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
   364
        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
   365
24593
f473a1fe5c7c encoding: define an enum that specifies what normcase does to ASCII strings
Siddharth Agarwal <sid0@fb.com>
parents: 24578
diff changeset
   366
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
   367
    '''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
   368
f473a1fe5c7c encoding: define an enum that specifies what normcase does to ASCII strings
Siddharth Agarwal <sid0@fb.com>
parents: 24578
diff changeset
   369
    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
   370
    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
   371
f473a1fe5c7c encoding: define an enum that specifies what normcase does to ASCII strings
Siddharth Agarwal <sid0@fb.com>
parents: 24578
diff changeset
   372
    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
   373
    upper: normcase uppercases ASCII strings
24608
1c533e23ce95 util.h: define an enum for normcase specs
Siddharth Agarwal <sid0@fb.com>
parents: 24597
diff changeset
   374
    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
   375
1c533e23ce95 util.h: define an enum for normcase specs
Siddharth Agarwal <sid0@fb.com>
parents: 24597
diff changeset
   376
    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
   377
    lower = -1
f473a1fe5c7c encoding: define an enum that specifies what normcase does to ASCII strings
Siddharth Agarwal <sid0@fb.com>
parents: 24578
diff changeset
   378
    upper = 1
f473a1fe5c7c encoding: define an enum that specifies what normcase does to ASCII strings
Siddharth Agarwal <sid0@fb.com>
parents: 24578
diff changeset
   379
    other = 0
f473a1fe5c7c encoding: define an enum that specifies what normcase does to ASCII strings
Siddharth Agarwal <sid0@fb.com>
parents: 24578
diff changeset
   380
22426
f6b533e64ed6 encoding: add json escaping filter
Matt Mackall <mpm@selenic.com>
parents: 22425
diff changeset
   381
_jsonmap = {}
f6b533e64ed6 encoding: add json escaping filter
Matt Mackall <mpm@selenic.com>
parents: 22425
diff changeset
   382
f6b533e64ed6 encoding: add json escaping filter
Matt Mackall <mpm@selenic.com>
parents: 22425
diff changeset
   383
def jsonescape(s):
f6b533e64ed6 encoding: add json escaping filter
Matt Mackall <mpm@selenic.com>
parents: 22425
diff changeset
   384
    '''returns a string suitable for JSON
f6b533e64ed6 encoding: add json escaping filter
Matt Mackall <mpm@selenic.com>
parents: 22425
diff changeset
   385
f6b533e64ed6 encoding: add json escaping filter
Matt Mackall <mpm@selenic.com>
parents: 22425
diff changeset
   386
    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
   387
    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
   388
f6b533e64ed6 encoding: add json escaping filter
Matt Mackall <mpm@selenic.com>
parents: 22425
diff changeset
   389
    - localstr objects are converted back to UTF-8
f6b533e64ed6 encoding: add json escaping filter
Matt Mackall <mpm@selenic.com>
parents: 22425
diff changeset
   390
    - valid UTF-8/ASCII strings are passed as-is
f6b533e64ed6 encoding: add json escaping filter
Matt Mackall <mpm@selenic.com>
parents: 22425
diff changeset
   391
    - other strings are converted to UTF-8b surrogate encoding
f6b533e64ed6 encoding: add json escaping filter
Matt Mackall <mpm@selenic.com>
parents: 22425
diff changeset
   392
    - apply JSON-specified string escaping
f6b533e64ed6 encoding: add json escaping filter
Matt Mackall <mpm@selenic.com>
parents: 22425
diff changeset
   393
f6b533e64ed6 encoding: add json escaping filter
Matt Mackall <mpm@selenic.com>
parents: 22425
diff changeset
   394
    (escapes are doubled in these tests)
f6b533e64ed6 encoding: add json escaping filter
Matt Mackall <mpm@selenic.com>
parents: 22425
diff changeset
   395
f6b533e64ed6 encoding: add json escaping filter
Matt Mackall <mpm@selenic.com>
parents: 22425
diff changeset
   396
    >>> jsonescape('this is a test')
f6b533e64ed6 encoding: add json escaping filter
Matt Mackall <mpm@selenic.com>
parents: 22425
diff changeset
   397
    'this is a test'
f6b533e64ed6 encoding: add json escaping filter
Matt Mackall <mpm@selenic.com>
parents: 22425
diff changeset
   398
    >>> jsonescape('escape characters: \\0 \\x0b \\t \\n \\r \\" \\\\')
f6b533e64ed6 encoding: add json escaping filter
Matt Mackall <mpm@selenic.com>
parents: 22425
diff changeset
   399
    'escape characters: \\\\u0000 \\\\u000b \\\\t \\\\n \\\\r \\\\" \\\\\\\\'
f6b533e64ed6 encoding: add json escaping filter
Matt Mackall <mpm@selenic.com>
parents: 22425
diff changeset
   400
    >>> jsonescape('a weird byte: \\xdd')
f6b533e64ed6 encoding: add json escaping filter
Matt Mackall <mpm@selenic.com>
parents: 22425
diff changeset
   401
    'a weird byte: \\xed\\xb3\\x9d'
f6b533e64ed6 encoding: add json escaping filter
Matt Mackall <mpm@selenic.com>
parents: 22425
diff changeset
   402
    >>> jsonescape('utf-8: caf\\xc3\\xa9')
f6b533e64ed6 encoding: add json escaping filter
Matt Mackall <mpm@selenic.com>
parents: 22425
diff changeset
   403
    'utf-8: caf\\xc3\\xa9'
f6b533e64ed6 encoding: add json escaping filter
Matt Mackall <mpm@selenic.com>
parents: 22425
diff changeset
   404
    >>> jsonescape('')
f6b533e64ed6 encoding: add json escaping filter
Matt Mackall <mpm@selenic.com>
parents: 22425
diff changeset
   405
    ''
f6b533e64ed6 encoding: add json escaping filter
Matt Mackall <mpm@selenic.com>
parents: 22425
diff changeset
   406
    '''
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
    if not _jsonmap:
f6b533e64ed6 encoding: add json escaping filter
Matt Mackall <mpm@selenic.com>
parents: 22425
diff changeset
   409
        for x in xrange(32):
27356
c2effd1ecebf encoding: use double backslash
Gregory Szorc <gregory.szorc@gmail.com>
parents: 27355
diff changeset
   410
            _jsonmap[chr(x)] = "\\u%04x" % x
22426
f6b533e64ed6 encoding: add json escaping filter
Matt Mackall <mpm@selenic.com>
parents: 22425
diff changeset
   411
        for x in xrange(32, 256):
f6b533e64ed6 encoding: add json escaping filter
Matt Mackall <mpm@selenic.com>
parents: 22425
diff changeset
   412
            c = chr(x)
f6b533e64ed6 encoding: add json escaping filter
Matt Mackall <mpm@selenic.com>
parents: 22425
diff changeset
   413
            _jsonmap[c] = c
f6b533e64ed6 encoding: add json escaping filter
Matt Mackall <mpm@selenic.com>
parents: 22425
diff changeset
   414
        _jsonmap['\t'] = '\\t'
f6b533e64ed6 encoding: add json escaping filter
Matt Mackall <mpm@selenic.com>
parents: 22425
diff changeset
   415
        _jsonmap['\n'] = '\\n'
f6b533e64ed6 encoding: add json escaping filter
Matt Mackall <mpm@selenic.com>
parents: 22425
diff changeset
   416
        _jsonmap['\"'] = '\\"'
f6b533e64ed6 encoding: add json escaping filter
Matt Mackall <mpm@selenic.com>
parents: 22425
diff changeset
   417
        _jsonmap['\\'] = '\\\\'
f6b533e64ed6 encoding: add json escaping filter
Matt Mackall <mpm@selenic.com>
parents: 22425
diff changeset
   418
        _jsonmap['\b'] = '\\b'
f6b533e64ed6 encoding: add json escaping filter
Matt Mackall <mpm@selenic.com>
parents: 22425
diff changeset
   419
        _jsonmap['\f'] = '\\f'
f6b533e64ed6 encoding: add json escaping filter
Matt Mackall <mpm@selenic.com>
parents: 22425
diff changeset
   420
        _jsonmap['\r'] = '\\r'
f6b533e64ed6 encoding: add json escaping filter
Matt Mackall <mpm@selenic.com>
parents: 22425
diff changeset
   421
f6b533e64ed6 encoding: add json escaping filter
Matt Mackall <mpm@selenic.com>
parents: 22425
diff changeset
   422
    return ''.join(_jsonmap[c] for c in toutf8b(s))
f6b533e64ed6 encoding: add json escaping filter
Matt Mackall <mpm@selenic.com>
parents: 22425
diff changeset
   423
26875
cf47bdb2183c encoding: add getutf8char helper
Matt Mackall <mpm@selenic.com>
parents: 25660
diff changeset
   424
_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
   425
cf47bdb2183c encoding: add getutf8char helper
Matt Mackall <mpm@selenic.com>
parents: 25660
diff changeset
   426
def getutf8char(s, pos):
cf47bdb2183c encoding: add getutf8char helper
Matt Mackall <mpm@selenic.com>
parents: 25660
diff changeset
   427
    '''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
   428
cf47bdb2183c encoding: add getutf8char helper
Matt Mackall <mpm@selenic.com>
parents: 25660
diff changeset
   429
    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
   430
    utf-8 character.
cf47bdb2183c encoding: add getutf8char helper
Matt Mackall <mpm@selenic.com>
parents: 25660
diff changeset
   431
    '''
cf47bdb2183c encoding: add getutf8char helper
Matt Mackall <mpm@selenic.com>
parents: 25660
diff changeset
   432
cf47bdb2183c encoding: add getutf8char helper
Matt Mackall <mpm@selenic.com>
parents: 25660
diff changeset
   433
    # find how many bytes to attempt decoding from first nibble
cf47bdb2183c encoding: add getutf8char helper
Matt Mackall <mpm@selenic.com>
parents: 25660
diff changeset
   434
    l = _utf8len[ord(s[pos]) >> 4]
cf47bdb2183c encoding: add getutf8char helper
Matt Mackall <mpm@selenic.com>
parents: 25660
diff changeset
   435
    if not l: # ascii
cf47bdb2183c encoding: add getutf8char helper
Matt Mackall <mpm@selenic.com>
parents: 25660
diff changeset
   436
        return s[pos]
cf47bdb2183c encoding: add getutf8char helper
Matt Mackall <mpm@selenic.com>
parents: 25660
diff changeset
   437
cf47bdb2183c encoding: add getutf8char helper
Matt Mackall <mpm@selenic.com>
parents: 25660
diff changeset
   438
    c = s[pos:pos + l]
cf47bdb2183c encoding: add getutf8char helper
Matt Mackall <mpm@selenic.com>
parents: 25660
diff changeset
   439
    # validate with attempted decode
cf47bdb2183c encoding: add getutf8char helper
Matt Mackall <mpm@selenic.com>
parents: 25660
diff changeset
   440
    c.decode("utf-8")
cf47bdb2183c encoding: add getutf8char helper
Matt Mackall <mpm@selenic.com>
parents: 25660
diff changeset
   441
    return c
cf47bdb2183c encoding: add getutf8char helper
Matt Mackall <mpm@selenic.com>
parents: 25660
diff changeset
   442
16133
84c58da3a1f8 encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents: 15769
diff changeset
   443
def toutf8b(s):
84c58da3a1f8 encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents: 15769
diff changeset
   444
    '''convert a local, possibly-binary string into UTF-8b
84c58da3a1f8 encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents: 15769
diff changeset
   445
84c58da3a1f8 encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents: 15769
diff changeset
   446
    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
   447
    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
   448
    arbitrary byte strings. As Mercurial often doesn't know
84c58da3a1f8 encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents: 15769
diff changeset
   449
    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
   450
84c58da3a1f8 encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents: 15769
diff changeset
   451
    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
   452
    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
   453
    uDC00-uDCFF.
84c58da3a1f8 encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents: 15769
diff changeset
   454
84c58da3a1f8 encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents: 15769
diff changeset
   455
    Principles of operation:
84c58da3a1f8 encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents: 15769
diff changeset
   456
17424
e7cfe3587ea4 fix trivial spelling errors
Mads Kiilerich <mads@kiilerich.com>
parents: 17236
diff changeset
   457
    - 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
   458
      by Unicode-oriented clients
84c58da3a1f8 encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents: 15769
diff changeset
   459
    - 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
   460
      be round-tripped or recovered by clueful clients
84c58da3a1f8 encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents: 15769
diff changeset
   461
    - 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
   462
      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
   463
      Unicode data they want
84c58da3a1f8 encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents: 15769
diff changeset
   464
    - 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
   465
      filenames, metadata can't be roundtripped without help
84c58da3a1f8 encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents: 15769
diff changeset
   466
84c58da3a1f8 encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents: 15769
diff changeset
   467
    (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
   468
    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
   469
    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
   470
    internal surrogate encoding as a UTF-8 string.)
84c58da3a1f8 encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents: 15769
diff changeset
   471
    '''
84c58da3a1f8 encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents: 15769
diff changeset
   472
26879
a24b98f4e03c encoding: re-escape U+DCxx characters in toutf8b input (issue4927)
Matt Mackall <mpm@selenic.com>
parents: 26878
diff changeset
   473
    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
   474
        if isinstance(s, localstr):
a24b98f4e03c encoding: re-escape U+DCxx characters in toutf8b input (issue4927)
Matt Mackall <mpm@selenic.com>
parents: 26878
diff changeset
   475
            return s._utf8
a24b98f4e03c encoding: re-escape U+DCxx characters in toutf8b input (issue4927)
Matt Mackall <mpm@selenic.com>
parents: 26878
diff changeset
   476
        try:
a24b98f4e03c encoding: re-escape U+DCxx characters in toutf8b input (issue4927)
Matt Mackall <mpm@selenic.com>
parents: 26878
diff changeset
   477
            s.decode('utf-8')
a24b98f4e03c encoding: re-escape U+DCxx characters in toutf8b input (issue4927)
Matt Mackall <mpm@selenic.com>
parents: 26878
diff changeset
   478
            return s
a24b98f4e03c encoding: re-escape U+DCxx characters in toutf8b input (issue4927)
Matt Mackall <mpm@selenic.com>
parents: 26878
diff changeset
   479
        except UnicodeDecodeError:
a24b98f4e03c encoding: re-escape U+DCxx characters in toutf8b input (issue4927)
Matt Mackall <mpm@selenic.com>
parents: 26878
diff changeset
   480
            pass
26878
d7e83f106459 encoding: use getutf8char in toutf8b
Matt Mackall <mpm@selenic.com>
parents: 26877
diff changeset
   481
d7e83f106459 encoding: use getutf8char in toutf8b
Matt Mackall <mpm@selenic.com>
parents: 26877
diff changeset
   482
    r = ""
d7e83f106459 encoding: use getutf8char in toutf8b
Matt Mackall <mpm@selenic.com>
parents: 26877
diff changeset
   483
    pos = 0
d7e83f106459 encoding: use getutf8char in toutf8b
Matt Mackall <mpm@selenic.com>
parents: 26877
diff changeset
   484
    l = len(s)
d7e83f106459 encoding: use getutf8char in toutf8b
Matt Mackall <mpm@selenic.com>
parents: 26877
diff changeset
   485
    while pos < l:
d7e83f106459 encoding: use getutf8char in toutf8b
Matt Mackall <mpm@selenic.com>
parents: 26877
diff changeset
   486
        try:
d7e83f106459 encoding: use getutf8char in toutf8b
Matt Mackall <mpm@selenic.com>
parents: 26877
diff changeset
   487
            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
   488
            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
   489
                # 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
   490
                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
   491
                pos += 1
a24b98f4e03c encoding: re-escape U+DCxx characters in toutf8b input (issue4927)
Matt Mackall <mpm@selenic.com>
parents: 26878
diff changeset
   492
            else:
a24b98f4e03c encoding: re-escape U+DCxx characters in toutf8b input (issue4927)
Matt Mackall <mpm@selenic.com>
parents: 26878
diff changeset
   493
                pos += len(c)
26878
d7e83f106459 encoding: use getutf8char in toutf8b
Matt Mackall <mpm@selenic.com>
parents: 26877
diff changeset
   494
        except UnicodeDecodeError:
d7e83f106459 encoding: use getutf8char in toutf8b
Matt Mackall <mpm@selenic.com>
parents: 26877
diff changeset
   495
            c = unichr(0xdc00 + ord(s[pos])).encode('utf-8')
d7e83f106459 encoding: use getutf8char in toutf8b
Matt Mackall <mpm@selenic.com>
parents: 26877
diff changeset
   496
            pos += 1
d7e83f106459 encoding: use getutf8char in toutf8b
Matt Mackall <mpm@selenic.com>
parents: 26877
diff changeset
   497
        r += c
d7e83f106459 encoding: use getutf8char in toutf8b
Matt Mackall <mpm@selenic.com>
parents: 26877
diff changeset
   498
    return r
16133
84c58da3a1f8 encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents: 15769
diff changeset
   499
84c58da3a1f8 encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents: 15769
diff changeset
   500
def fromutf8b(s):
84c58da3a1f8 encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents: 15769
diff changeset
   501
    '''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
   502
84c58da3a1f8 encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents: 15769
diff changeset
   503
    return the original binary string. This
84c58da3a1f8 encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents: 15769
diff changeset
   504
    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
   505
    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
   506
26963
de5ae97ce9f4 encoding: extend test cases for utf8b
Matt Mackall <mpm@selenic.com>
parents: 26879
diff changeset
   507
    >>> roundtrip = lambda x: fromutf8b(toutf8b(x)) == x
16133
84c58da3a1f8 encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents: 15769
diff changeset
   508
    >>> m = "\\xc3\\xa9\\x99abcd"
26963
de5ae97ce9f4 encoding: extend test cases for utf8b
Matt Mackall <mpm@selenic.com>
parents: 26879
diff changeset
   509
    >>> toutf8b(m)
16133
84c58da3a1f8 encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents: 15769
diff changeset
   510
    '\\xc3\\xa9\\xed\\xb2\\x99abcd'
26963
de5ae97ce9f4 encoding: extend test cases for utf8b
Matt Mackall <mpm@selenic.com>
parents: 26879
diff changeset
   511
    >>> roundtrip(m)
de5ae97ce9f4 encoding: extend test cases for utf8b
Matt Mackall <mpm@selenic.com>
parents: 26879
diff changeset
   512
    True
de5ae97ce9f4 encoding: extend test cases for utf8b
Matt Mackall <mpm@selenic.com>
parents: 26879
diff changeset
   513
    >>> roundtrip("\\xc2\\xc2\\x80")
de5ae97ce9f4 encoding: extend test cases for utf8b
Matt Mackall <mpm@selenic.com>
parents: 26879
diff changeset
   514
    True
de5ae97ce9f4 encoding: extend test cases for utf8b
Matt Mackall <mpm@selenic.com>
parents: 26879
diff changeset
   515
    >>> roundtrip("\\xef\\xbf\\xbd")
de5ae97ce9f4 encoding: extend test cases for utf8b
Matt Mackall <mpm@selenic.com>
parents: 26879
diff changeset
   516
    True
de5ae97ce9f4 encoding: extend test cases for utf8b
Matt Mackall <mpm@selenic.com>
parents: 26879
diff changeset
   517
    >>> roundtrip("\\xef\\xef\\xbf\\xbd")
16133
84c58da3a1f8 encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents: 15769
diff changeset
   518
    True
84c58da3a1f8 encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents: 15769
diff changeset
   519
    '''
84c58da3a1f8 encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents: 15769
diff changeset
   520
84c58da3a1f8 encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents: 15769
diff changeset
   521
    # fast path - look for uDxxx prefixes in s
84c58da3a1f8 encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents: 15769
diff changeset
   522
    if "\xed" not in s:
84c58da3a1f8 encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents: 15769
diff changeset
   523
        return s
84c58da3a1f8 encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents: 15769
diff changeset
   524
84c58da3a1f8 encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents: 15769
diff changeset
   525
    u = s.decode("utf-8")
84c58da3a1f8 encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents: 15769
diff changeset
   526
    r = ""
84c58da3a1f8 encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents: 15769
diff changeset
   527
    for c in u:
26877
cb467a9d7593 encoding: handle non-BMP characters in fromutf8b
Matt Mackall <mpm@selenic.com>
parents: 26875
diff changeset
   528
        if ord(c) & 0xffff00 == 0xdc00:
16133
84c58da3a1f8 encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents: 15769
diff changeset
   529
            r += chr(ord(c) & 0xff)
84c58da3a1f8 encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents: 15769
diff changeset
   530
        else:
84c58da3a1f8 encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents: 15769
diff changeset
   531
            r += c.encode("utf-8")
84c58da3a1f8 encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents: 15769
diff changeset
   532
    return r