mercurial/encoding.py
author Yuya Nishihara <yuya@tcha.org>
Sun, 23 Apr 2017 13:15:30 +0900
changeset 37947 3ea3c96ada54
parent 37946 57b0c7221dba
child 38739 7acec9408e1c
permissions -rw-r--r--
encoding: introduce tagging type for non-lossy non-ASCII string This fixes the weird behavior of toutf8b(), which would convert a local string back to UTF-8 *only if* it was lossy in the system encoding. Before b7b26e54e37a "encoding: avoid localstr when a string can be encoded losslessly (issue2763)", all local strings were wrapped by the localstr class. I think this would justify the round-trip behavior of toutf8b(). ASCII strings are special-cased, so the cost of wrapping with safelocalstr is negligible. (with mercurial repo) $ export HGRCPATH=/dev/null HGPLAIN= HGENCODING=latin-1 $ hg log --time --config experimental.evolution=all > /dev/null (original) time: real 11.340 secs (user 11.290+0.000 sys 0.060+0.000) time: real 11.390 secs (user 11.300+0.000 sys 0.080+0.000) time: real 11.430 secs (user 11.360+0.000 sys 0.070+0.000) (this patch) time: real 11.200 secs (user 11.100+0.000 sys 0.100+0.000) time: real 11.370 secs (user 11.300+0.000 sys 0.070+0.000) time: real 11.190 secs (user 11.130+0.000 sys 0.060+0.000)
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
34137
a8994d08e4a2 doctest: use print_function and convert bytes to unicode where needed
Yuya Nishihara <yuya@tcha.org>
parents: 34136
diff changeset
     8
from __future__ import absolute_import, print_function
27355
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,
32372
df448de7cf3b parsers: switch to policy importer
Yuya Nishihara <yuya@tcha.org>
parents: 32299
diff changeset
    16
    policy,
30030
0f6d6fdd3c2a pycompat: provide 'ispy3' constant
Yuya Nishihara <yuya@tcha.org>
parents: 28508
diff changeset
    17
    pycompat,
27355
b479fc425a81 encoding: use absolute_import
Gregory Szorc <gregory.szorc@gmail.com>
parents: 26963
diff changeset
    18
)
7948
de377b1a9a84 move encoding bits from util to encoding
Matt Mackall <mpm@selenic.com>
parents:
diff changeset
    19
33924
b9101467d88b encoding: extract stub for fast JSON escape
Yuya Nishihara <yuya@tcha.org>
parents: 33852
diff changeset
    20
from .pure import (
b9101467d88b encoding: extract stub for fast JSON escape
Yuya Nishihara <yuya@tcha.org>
parents: 33852
diff changeset
    21
    charencode as charencodepure,
b9101467d88b encoding: extract stub for fast JSON escape
Yuya Nishihara <yuya@tcha.org>
parents: 33852
diff changeset
    22
)
b9101467d88b encoding: extract stub for fast JSON escape
Yuya Nishihara <yuya@tcha.org>
parents: 33852
diff changeset
    23
33761
f5fc54e7e467 encoding: drop circular import by proxying through '<policy>.charencode'
Yuya Nishihara <yuya@tcha.org>
parents: 33022
diff changeset
    24
charencode = policy.importmod(r'charencode')
f5fc54e7e467 encoding: drop circular import by proxying through '<policy>.charencode'
Yuya Nishihara <yuya@tcha.org>
parents: 33022
diff changeset
    25
33926
f4433f2713d0 encoding: add function to test if a str consists of ASCII characters
Yuya Nishihara <yuya@tcha.org>
parents: 33925
diff changeset
    26
isasciistr = charencode.isasciistr
33761
f5fc54e7e467 encoding: drop circular import by proxying through '<policy>.charencode'
Yuya Nishihara <yuya@tcha.org>
parents: 33022
diff changeset
    27
asciilower = charencode.asciilower
f5fc54e7e467 encoding: drop circular import by proxying through '<policy>.charencode'
Yuya Nishihara <yuya@tcha.org>
parents: 33022
diff changeset
    28
asciiupper = charencode.asciiupper
33925
2c37f9dabc32 encoding: add fast path of jsonescape() (issue5533)
Yuya Nishihara <yuya@tcha.org>
parents: 33924
diff changeset
    29
_jsonescapeu8fast = charencode.jsonescapeu8fast
33761
f5fc54e7e467 encoding: drop circular import by proxying through '<policy>.charencode'
Yuya Nishihara <yuya@tcha.org>
parents: 33022
diff changeset
    30
30033
02dbfaa6df0b py3: convert encoding name and mode to str
Yuya Nishihara <yuya@tcha.org>
parents: 30030
diff changeset
    31
_sysstr = pycompat.sysstr
02dbfaa6df0b py3: convert encoding name and mode to str
Yuya Nishihara <yuya@tcha.org>
parents: 30030
diff changeset
    32
30030
0f6d6fdd3c2a pycompat: provide 'ispy3' constant
Yuya Nishihara <yuya@tcha.org>
parents: 28508
diff changeset
    33
if pycompat.ispy3:
28507
9bcbd9412225 encoding: make HFS+ ignore code Python 3 compatible
Gregory Szorc <gregory.szorc@gmail.com>
parents: 28069
diff changeset
    34
    unichr = chr
9bcbd9412225 encoding: make HFS+ ignore code Python 3 compatible
Gregory Szorc <gregory.szorc@gmail.com>
parents: 28069
diff changeset
    35
23596
885bd7c5c7e3 encoding: add hfsignoreclean to clean out HFS-ignored characters
Augie Fackler <raf@durin42.com>
parents: 22973
diff changeset
    36
# 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
    37
# "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
    38
# sanity.
885bd7c5c7e3 encoding: add hfsignoreclean to clean out HFS-ignored characters
Augie Fackler <raf@durin42.com>
parents: 22973
diff changeset
    39
_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
    40
           "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
    41
           "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
    42
# verify the next function will work
32299
7040f5131454 encoding: use i.startswith() instead of i[0] to eliminate py2/3 divergence
Yuya Nishihara <yuya@tcha.org>
parents: 32291
diff changeset
    43
assert all(i.startswith(("\xe2", "\xef")) for i in _ignore)
23596
885bd7c5c7e3 encoding: add hfsignoreclean to clean out HFS-ignored characters
Augie Fackler <raf@durin42.com>
parents: 22973
diff changeset
    44
885bd7c5c7e3 encoding: add hfsignoreclean to clean out HFS-ignored characters
Augie Fackler <raf@durin42.com>
parents: 22973
diff changeset
    45
def hfsignoreclean(s):
885bd7c5c7e3 encoding: add hfsignoreclean to clean out HFS-ignored characters
Augie Fackler <raf@durin42.com>
parents: 22973
diff changeset
    46
    """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
    47
885bd7c5c7e3 encoding: add hfsignoreclean to clean out HFS-ignored characters
Augie Fackler <raf@durin42.com>
parents: 22973
diff changeset
    48
    >>> 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
    49
    '.hg'
885bd7c5c7e3 encoding: add hfsignoreclean to clean out HFS-ignored characters
Augie Fackler <raf@durin42.com>
parents: 22973
diff changeset
    50
    >>> 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
    51
    '.hg'
885bd7c5c7e3 encoding: add hfsignoreclean to clean out HFS-ignored characters
Augie Fackler <raf@durin42.com>
parents: 22973
diff changeset
    52
    """
885bd7c5c7e3 encoding: add hfsignoreclean to clean out HFS-ignored characters
Augie Fackler <raf@durin42.com>
parents: 22973
diff changeset
    53
    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
    54
        for c in _ignore:
885bd7c5c7e3 encoding: add hfsignoreclean to clean out HFS-ignored characters
Augie Fackler <raf@durin42.com>
parents: 22973
diff changeset
    55
            s = s.replace(c, '')
885bd7c5c7e3 encoding: add hfsignoreclean to clean out HFS-ignored characters
Augie Fackler <raf@durin42.com>
parents: 22973
diff changeset
    56
    return s
885bd7c5c7e3 encoding: add hfsignoreclean to clean out HFS-ignored characters
Augie Fackler <raf@durin42.com>
parents: 22973
diff changeset
    57
30034
e4a6b439acc5 py3: provide encoding.environ which is a dict of bytes
Yuya Nishihara <yuya@tcha.org>
parents: 30033
diff changeset
    58
# encoding.environ is provided read-only, which may not be used to modify
e4a6b439acc5 py3: provide encoding.environ which is a dict of bytes
Yuya Nishihara <yuya@tcha.org>
parents: 30033
diff changeset
    59
# the process environment
e4a6b439acc5 py3: provide encoding.environ which is a dict of bytes
Yuya Nishihara <yuya@tcha.org>
parents: 30033
diff changeset
    60
_nativeenviron = (not pycompat.ispy3 or os.supports_bytes_environ)
e4a6b439acc5 py3: provide encoding.environ which is a dict of bytes
Yuya Nishihara <yuya@tcha.org>
parents: 30033
diff changeset
    61
if not pycompat.ispy3:
32184
cf424dae5dc7 check-code: ignore re-exports of os.environ in encoding.py
Yuya Nishihara <yuya@tcha.org>
parents: 32156
diff changeset
    62
    environ = os.environ  # re-exports
30034
e4a6b439acc5 py3: provide encoding.environ which is a dict of bytes
Yuya Nishihara <yuya@tcha.org>
parents: 30033
diff changeset
    63
elif _nativeenviron:
32184
cf424dae5dc7 check-code: ignore re-exports of os.environ in encoding.py
Yuya Nishihara <yuya@tcha.org>
parents: 32156
diff changeset
    64
    environ = os.environb  # re-exports
30034
e4a6b439acc5 py3: provide encoding.environ which is a dict of bytes
Yuya Nishihara <yuya@tcha.org>
parents: 30033
diff changeset
    65
else:
e4a6b439acc5 py3: provide encoding.environ which is a dict of bytes
Yuya Nishihara <yuya@tcha.org>
parents: 30033
diff changeset
    66
    # preferred encoding isn't known yet; use utf-8 to avoid unicode error
e4a6b439acc5 py3: provide encoding.environ which is a dict of bytes
Yuya Nishihara <yuya@tcha.org>
parents: 30033
diff changeset
    67
    # and recreate it once encoding is settled
e4a6b439acc5 py3: provide encoding.environ which is a dict of bytes
Yuya Nishihara <yuya@tcha.org>
parents: 30033
diff changeset
    68
    environ = dict((k.encode(u'utf-8'), v.encode(u'utf-8'))
32184
cf424dae5dc7 check-code: ignore re-exports of os.environ in encoding.py
Yuya Nishihara <yuya@tcha.org>
parents: 32156
diff changeset
    69
                   for k, v in os.environ.items())  # re-exports
30034
e4a6b439acc5 py3: provide encoding.environ which is a dict of bytes
Yuya Nishihara <yuya@tcha.org>
parents: 30033
diff changeset
    70
11892
2be70ca17311 encoding: improve handling of buggy getpreferredencoding() on Mac OS X
Dan Villiom Podlaski Christiansen <danchr@gmail.com>
parents: 11297
diff changeset
    71
_encodingfixers = {
2be70ca17311 encoding: improve handling of buggy getpreferredencoding() on Mac OS X
Dan Villiom Podlaski Christiansen <danchr@gmail.com>
parents: 11297
diff changeset
    72
    '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
    73
    '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
    74
}
7948
de377b1a9a84 move encoding bits from util to encoding
Matt Mackall <mpm@selenic.com>
parents:
diff changeset
    75
de377b1a9a84 move encoding bits from util to encoding
Matt Mackall <mpm@selenic.com>
parents:
diff changeset
    76
try:
30034
e4a6b439acc5 py3: provide encoding.environ which is a dict of bytes
Yuya Nishihara <yuya@tcha.org>
parents: 30033
diff changeset
    77
    encoding = environ.get("HGENCODING")
7948
de377b1a9a84 move encoding bits from util to encoding
Matt Mackall <mpm@selenic.com>
parents:
diff changeset
    78
    if not encoding:
30622
ce36fa9b140c py3: make sure encoding.encoding is a bytes variable
Pulkit Goyal <7895pulkit@gmail.com>
parents: 30034
diff changeset
    79
        encoding = locale.getpreferredencoding().encode('ascii') 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
    80
        encoding = _encodingfixers.get(encoding, lambda: encoding)()
7948
de377b1a9a84 move encoding bits from util to encoding
Matt Mackall <mpm@selenic.com>
parents:
diff changeset
    81
except locale.Error:
de377b1a9a84 move encoding bits from util to encoding
Matt Mackall <mpm@selenic.com>
parents:
diff changeset
    82
    encoding = 'ascii'
30034
e4a6b439acc5 py3: provide encoding.environ which is a dict of bytes
Yuya Nishihara <yuya@tcha.org>
parents: 30033
diff changeset
    83
encodingmode = environ.get("HGENCODINGMODE", "strict")
7948
de377b1a9a84 move encoding bits from util to encoding
Matt Mackall <mpm@selenic.com>
parents:
diff changeset
    84
fallbackencoding = 'ISO-8859-1'
de377b1a9a84 move encoding bits from util to encoding
Matt Mackall <mpm@selenic.com>
parents:
diff changeset
    85
33811
dabe1f11ae3a py3: change encoding.localstr to a subclass of bytes, not str
Yuya Nishihara <yuya@tcha.org>
parents: 33761
diff changeset
    86
class localstr(bytes):
13046
7cc4263e07a9 encoding: add localstr class to track UTF-8 version of transcoded strings
Matt Mackall <mpm@selenic.com>
parents: 12866
diff changeset
    87
    '''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
    88
    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
    89
    def __new__(cls, u, l):
33811
dabe1f11ae3a py3: change encoding.localstr to a subclass of bytes, not str
Yuya Nishihara <yuya@tcha.org>
parents: 33761
diff changeset
    90
        s = bytes.__new__(cls, l)
13046
7cc4263e07a9 encoding: add localstr class to track UTF-8 version of transcoded strings
Matt Mackall <mpm@selenic.com>
parents: 12866
diff changeset
    91
        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
    92
        return s
7cc4263e07a9 encoding: add localstr class to track UTF-8 version of transcoded strings
Matt Mackall <mpm@selenic.com>
parents: 12866
diff changeset
    93
    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
    94
        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
    95
37947
3ea3c96ada54 encoding: introduce tagging type for non-lossy non-ASCII string
Yuya Nishihara <yuya@tcha.org>
parents: 37946
diff changeset
    96
class safelocalstr(bytes):
3ea3c96ada54 encoding: introduce tagging type for non-lossy non-ASCII string
Yuya Nishihara <yuya@tcha.org>
parents: 37946
diff changeset
    97
    """Tagged string denoting it was previously an internal UTF-8 string,
3ea3c96ada54 encoding: introduce tagging type for non-lossy non-ASCII string
Yuya Nishihara <yuya@tcha.org>
parents: 37946
diff changeset
    98
    and can be converted back to UTF-8 losslessly
3ea3c96ada54 encoding: introduce tagging type for non-lossy non-ASCII string
Yuya Nishihara <yuya@tcha.org>
parents: 37946
diff changeset
    99
3ea3c96ada54 encoding: introduce tagging type for non-lossy non-ASCII string
Yuya Nishihara <yuya@tcha.org>
parents: 37946
diff changeset
   100
    >>> assert safelocalstr(b'\\xc3') == b'\\xc3'
3ea3c96ada54 encoding: introduce tagging type for non-lossy non-ASCII string
Yuya Nishihara <yuya@tcha.org>
parents: 37946
diff changeset
   101
    >>> assert b'\\xc3' == safelocalstr(b'\\xc3')
3ea3c96ada54 encoding: introduce tagging type for non-lossy non-ASCII string
Yuya Nishihara <yuya@tcha.org>
parents: 37946
diff changeset
   102
    >>> assert b'\\xc3' in {safelocalstr(b'\\xc3'): 0}
3ea3c96ada54 encoding: introduce tagging type for non-lossy non-ASCII string
Yuya Nishihara <yuya@tcha.org>
parents: 37946
diff changeset
   103
    >>> assert safelocalstr(b'\\xc3') in {b'\\xc3': 0}
3ea3c96ada54 encoding: introduce tagging type for non-lossy non-ASCII string
Yuya Nishihara <yuya@tcha.org>
parents: 37946
diff changeset
   104
    """
3ea3c96ada54 encoding: introduce tagging type for non-lossy non-ASCII string
Yuya Nishihara <yuya@tcha.org>
parents: 37946
diff changeset
   105
7948
de377b1a9a84 move encoding bits from util to encoding
Matt Mackall <mpm@selenic.com>
parents:
diff changeset
   106
def tolocal(s):
de377b1a9a84 move encoding bits from util to encoding
Matt Mackall <mpm@selenic.com>
parents:
diff changeset
   107
    """
de377b1a9a84 move encoding bits from util to encoding
Matt Mackall <mpm@selenic.com>
parents:
diff changeset
   108
    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
   109
de377b1a9a84 move encoding bits from util to encoding
Matt Mackall <mpm@selenic.com>
parents:
diff changeset
   110
    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
   111
    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
   112
    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
   113
    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
   114
    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
   115
7cc4263e07a9 encoding: add localstr class to track UTF-8 version of transcoded strings
Matt Mackall <mpm@selenic.com>
parents: 12866
diff changeset
   116
    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
   117
    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
   118
    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
   119
34131
0fa781320203 doctest: bulk-replace string literals with b'' for Python 3
Yuya Nishihara <yuya@tcha.org>
parents: 33928
diff changeset
   120
    >>> u = b'foo: \\xc3\\xa4' # utf-8
13046
7cc4263e07a9 encoding: add localstr class to track UTF-8 version of transcoded strings
Matt Mackall <mpm@selenic.com>
parents: 12866
diff changeset
   121
    >>> 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
   122
    >>> l
7cc4263e07a9 encoding: add localstr class to track UTF-8 version of transcoded strings
Matt Mackall <mpm@selenic.com>
parents: 12866
diff changeset
   123
    'foo: ?'
7cc4263e07a9 encoding: add localstr class to track UTF-8 version of transcoded strings
Matt Mackall <mpm@selenic.com>
parents: 12866
diff changeset
   124
    >>> fromlocal(l)
7cc4263e07a9 encoding: add localstr class to track UTF-8 version of transcoded strings
Matt Mackall <mpm@selenic.com>
parents: 12866
diff changeset
   125
    'foo: \\xc3\\xa4'
34131
0fa781320203 doctest: bulk-replace string literals with b'' for Python 3
Yuya Nishihara <yuya@tcha.org>
parents: 33928
diff changeset
   126
    >>> u2 = b'foo: \\xc3\\xa1'
13046
7cc4263e07a9 encoding: add localstr class to track UTF-8 version of transcoded strings
Matt Mackall <mpm@selenic.com>
parents: 12866
diff changeset
   127
    >>> d = { l: 1, tolocal(u2): 2 }
18378
404feac78b8a tests: stabilize doctest output
Mads Kiilerich <mads@kiilerich.com>
parents: 17424
diff changeset
   128
    >>> len(d) # no collision
404feac78b8a tests: stabilize doctest output
Mads Kiilerich <mads@kiilerich.com>
parents: 17424
diff changeset
   129
    2
34131
0fa781320203 doctest: bulk-replace string literals with b'' for Python 3
Yuya Nishihara <yuya@tcha.org>
parents: 33928
diff changeset
   130
    >>> b'foo: ?' in d
13046
7cc4263e07a9 encoding: add localstr class to track UTF-8 version of transcoded strings
Matt Mackall <mpm@selenic.com>
parents: 12866
diff changeset
   131
    False
34131
0fa781320203 doctest: bulk-replace string literals with b'' for Python 3
Yuya Nishihara <yuya@tcha.org>
parents: 33928
diff changeset
   132
    >>> l1 = b'foo: \\xe4' # historical latin1 fallback
13046
7cc4263e07a9 encoding: add localstr class to track UTF-8 version of transcoded strings
Matt Mackall <mpm@selenic.com>
parents: 12866
diff changeset
   133
    >>> 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
   134
    >>> l
7cc4263e07a9 encoding: add localstr class to track UTF-8 version of transcoded strings
Matt Mackall <mpm@selenic.com>
parents: 12866
diff changeset
   135
    'foo: ?'
7cc4263e07a9 encoding: add localstr class to track UTF-8 version of transcoded strings
Matt Mackall <mpm@selenic.com>
parents: 12866
diff changeset
   136
    >>> 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
   137
    'foo: \\xc3\\xa4'
7948
de377b1a9a84 move encoding bits from util to encoding
Matt Mackall <mpm@selenic.com>
parents:
diff changeset
   138
    """
13046
7cc4263e07a9 encoding: add localstr class to track UTF-8 version of transcoded strings
Matt Mackall <mpm@selenic.com>
parents: 12866
diff changeset
   139
33927
853574db5b12 encoding: add fast path of from/tolocal() for ASCII strings
Yuya Nishihara <yuya@tcha.org>
parents: 33926
diff changeset
   140
    if isasciistr(s):
853574db5b12 encoding: add fast path of from/tolocal() for ASCII strings
Yuya Nishihara <yuya@tcha.org>
parents: 33926
diff changeset
   141
        return s
853574db5b12 encoding: add fast path of from/tolocal() for ASCII strings
Yuya Nishihara <yuya@tcha.org>
parents: 33926
diff changeset
   142
16274
5d75eb8568d1 encoding: tune fast-path of tolocal a bit
Matt Mackall <mpm@selenic.com>
parents: 16133
diff changeset
   143
    try:
7948
de377b1a9a84 move encoding bits from util to encoding
Matt Mackall <mpm@selenic.com>
parents:
diff changeset
   144
        try:
16274
5d75eb8568d1 encoding: tune fast-path of tolocal a bit
Matt Mackall <mpm@selenic.com>
parents: 16133
diff changeset
   145
            # 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
   146
            u = s.decode('UTF-8')
5d75eb8568d1 encoding: tune fast-path of tolocal a bit
Matt Mackall <mpm@selenic.com>
parents: 16133
diff changeset
   147
            if encoding == 'UTF-8':
5d75eb8568d1 encoding: tune fast-path of tolocal a bit
Matt Mackall <mpm@selenic.com>
parents: 16133
diff changeset
   148
                # fast path
5d75eb8568d1 encoding: tune fast-path of tolocal a bit
Matt Mackall <mpm@selenic.com>
parents: 16133
diff changeset
   149
                return s
30033
02dbfaa6df0b py3: convert encoding name and mode to str
Yuya Nishihara <yuya@tcha.org>
parents: 30030
diff changeset
   150
            r = u.encode(_sysstr(encoding), u"replace")
02dbfaa6df0b py3: convert encoding name and mode to str
Yuya Nishihara <yuya@tcha.org>
parents: 30030
diff changeset
   151
            if u == r.decode(_sysstr(encoding)):
13940
b7b26e54e37a encoding: avoid localstr when a string can be encoded losslessly (issue2763)
Matt Mackall <mpm@selenic.com>
parents: 13051
diff changeset
   152
                # r is a safe, non-lossy encoding of s
37947
3ea3c96ada54 encoding: introduce tagging type for non-lossy non-ASCII string
Yuya Nishihara <yuya@tcha.org>
parents: 37946
diff changeset
   153
                return safelocalstr(r)
16274
5d75eb8568d1 encoding: tune fast-path of tolocal a bit
Matt Mackall <mpm@selenic.com>
parents: 16133
diff changeset
   154
            return localstr(s, r)
5d75eb8568d1 encoding: tune fast-path of tolocal a bit
Matt Mackall <mpm@selenic.com>
parents: 16133
diff changeset
   155
        except UnicodeDecodeError:
5d75eb8568d1 encoding: tune fast-path of tolocal a bit
Matt Mackall <mpm@selenic.com>
parents: 16133
diff changeset
   156
            # 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
   157
            try:
30033
02dbfaa6df0b py3: convert encoding name and mode to str
Yuya Nishihara <yuya@tcha.org>
parents: 30030
diff changeset
   158
                u = s.decode(_sysstr(fallbackencoding))
02dbfaa6df0b py3: convert encoding name and mode to str
Yuya Nishihara <yuya@tcha.org>
parents: 30030
diff changeset
   159
                r = u.encode(_sysstr(encoding), u"replace")
02dbfaa6df0b py3: convert encoding name and mode to str
Yuya Nishihara <yuya@tcha.org>
parents: 30030
diff changeset
   160
                if u == r.decode(_sysstr(encoding)):
16274
5d75eb8568d1 encoding: tune fast-path of tolocal a bit
Matt Mackall <mpm@selenic.com>
parents: 16133
diff changeset
   161
                    # r is a safe, non-lossy encoding of s
37947
3ea3c96ada54 encoding: introduce tagging type for non-lossy non-ASCII string
Yuya Nishihara <yuya@tcha.org>
parents: 37946
diff changeset
   162
                    return safelocalstr(r)
13940
b7b26e54e37a encoding: avoid localstr when a string can be encoded losslessly (issue2763)
Matt Mackall <mpm@selenic.com>
parents: 13051
diff changeset
   163
                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
   164
            except UnicodeDecodeError:
5d75eb8568d1 encoding: tune fast-path of tolocal a bit
Matt Mackall <mpm@selenic.com>
parents: 16133
diff changeset
   165
                u = s.decode("utf-8", "replace") # last ditch
30033
02dbfaa6df0b py3: convert encoding name and mode to str
Yuya Nishihara <yuya@tcha.org>
parents: 30030
diff changeset
   166
                # can't round-trip
02dbfaa6df0b py3: convert encoding name and mode to str
Yuya Nishihara <yuya@tcha.org>
parents: 30030
diff changeset
   167
                return u.encode(_sysstr(encoding), u"replace")
25660
328739ea70c3 global: mass rewrite to use modern exception syntax
Gregory Szorc <gregory.szorc@gmail.com>
parents: 24608
diff changeset
   168
    except LookupError as k:
16274
5d75eb8568d1 encoding: tune fast-path of tolocal a bit
Matt Mackall <mpm@selenic.com>
parents: 16133
diff changeset
   169
        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
   170
de377b1a9a84 move encoding bits from util to encoding
Matt Mackall <mpm@selenic.com>
parents:
diff changeset
   171
def fromlocal(s):
de377b1a9a84 move encoding bits from util to encoding
Matt Mackall <mpm@selenic.com>
parents:
diff changeset
   172
    """
de377b1a9a84 move encoding bits from util to encoding
Matt Mackall <mpm@selenic.com>
parents:
diff changeset
   173
    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
   174
de377b1a9a84 move encoding bits from util to encoding
Matt Mackall <mpm@selenic.com>
parents:
diff changeset
   175
    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
   176
    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
   177
    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
   178
    'replace', which replaces unknown characters with a special
de377b1a9a84 move encoding bits from util to encoding
Matt Mackall <mpm@selenic.com>
parents:
diff changeset
   179
    Unicode character, and 'ignore', which drops the character.
de377b1a9a84 move encoding bits from util to encoding
Matt Mackall <mpm@selenic.com>
parents:
diff changeset
   180
    """
13046
7cc4263e07a9 encoding: add localstr class to track UTF-8 version of transcoded strings
Matt Mackall <mpm@selenic.com>
parents: 12866
diff changeset
   181
7cc4263e07a9 encoding: add localstr class to track UTF-8 version of transcoded strings
Matt Mackall <mpm@selenic.com>
parents: 12866
diff changeset
   182
    # 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
   183
    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
   184
        return s._utf8
33927
853574db5b12 encoding: add fast path of from/tolocal() for ASCII strings
Yuya Nishihara <yuya@tcha.org>
parents: 33926
diff changeset
   185
    if isasciistr(s):
853574db5b12 encoding: add fast path of from/tolocal() for ASCII strings
Yuya Nishihara <yuya@tcha.org>
parents: 33926
diff changeset
   186
        return s
13046
7cc4263e07a9 encoding: add localstr class to track UTF-8 version of transcoded strings
Matt Mackall <mpm@selenic.com>
parents: 12866
diff changeset
   187
7948
de377b1a9a84 move encoding bits from util to encoding
Matt Mackall <mpm@selenic.com>
parents:
diff changeset
   188
    try:
30033
02dbfaa6df0b py3: convert encoding name and mode to str
Yuya Nishihara <yuya@tcha.org>
parents: 30030
diff changeset
   189
        u = s.decode(_sysstr(encoding), _sysstr(encodingmode))
02dbfaa6df0b py3: convert encoding name and mode to str
Yuya Nishihara <yuya@tcha.org>
parents: 30030
diff changeset
   190
        return u.encode("utf-8")
25660
328739ea70c3 global: mass rewrite to use modern exception syntax
Gregory Szorc <gregory.szorc@gmail.com>
parents: 24608
diff changeset
   191
    except UnicodeDecodeError as inst:
10282
08a0f04b56bd many, many trivial check-code fixups
Matt Mackall <mpm@selenic.com>
parents: 10263
diff changeset
   192
        sub = s[max(0, inst.start - 10):inst.start + 10]
36549
3696efeab66f py3: don't crash when re-raising encoding error
Yuya Nishihara <yuya@tcha.org>
parents: 34218
diff changeset
   193
        raise error.Abort("decoding near '%s': %s!"
3696efeab66f py3: don't crash when re-raising encoding error
Yuya Nishihara <yuya@tcha.org>
parents: 34218
diff changeset
   194
                          % (sub, pycompat.bytestr(inst)))
25660
328739ea70c3 global: mass rewrite to use modern exception syntax
Gregory Szorc <gregory.szorc@gmail.com>
parents: 24608
diff changeset
   195
    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
   196
        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
   197
31447
067add650129 encoding: factor out unicode variants of from/tolocal()
Yuya Nishihara <yuya@tcha.org>
parents: 30622
diff changeset
   198
def unitolocal(u):
067add650129 encoding: factor out unicode variants of from/tolocal()
Yuya Nishihara <yuya@tcha.org>
parents: 30622
diff changeset
   199
    """Convert a unicode string to a byte string of local encoding"""
067add650129 encoding: factor out unicode variants of from/tolocal()
Yuya Nishihara <yuya@tcha.org>
parents: 30622
diff changeset
   200
    return tolocal(u.encode('utf-8'))
067add650129 encoding: factor out unicode variants of from/tolocal()
Yuya Nishihara <yuya@tcha.org>
parents: 30622
diff changeset
   201
067add650129 encoding: factor out unicode variants of from/tolocal()
Yuya Nishihara <yuya@tcha.org>
parents: 30622
diff changeset
   202
def unifromlocal(s):
067add650129 encoding: factor out unicode variants of from/tolocal()
Yuya Nishihara <yuya@tcha.org>
parents: 30622
diff changeset
   203
    """Convert a byte string of local encoding to a unicode string"""
067add650129 encoding: factor out unicode variants of from/tolocal()
Yuya Nishihara <yuya@tcha.org>
parents: 30622
diff changeset
   204
    return fromlocal(s).decode('utf-8')
067add650129 encoding: factor out unicode variants of from/tolocal()
Yuya Nishihara <yuya@tcha.org>
parents: 30622
diff changeset
   205
33022
ce96efec8112 py3: add utility to forward __str__() to __bytes__()
Yuya Nishihara <yuya@tcha.org>
parents: 32537
diff changeset
   206
def unimethod(bytesfunc):
ce96efec8112 py3: add utility to forward __str__() to __bytes__()
Yuya Nishihara <yuya@tcha.org>
parents: 32537
diff changeset
   207
    """Create a proxy method that forwards __unicode__() and __str__() of
ce96efec8112 py3: add utility to forward __str__() to __bytes__()
Yuya Nishihara <yuya@tcha.org>
parents: 32537
diff changeset
   208
    Python 3 to __bytes__()"""
ce96efec8112 py3: add utility to forward __str__() to __bytes__()
Yuya Nishihara <yuya@tcha.org>
parents: 32537
diff changeset
   209
    def unifunc(obj):
ce96efec8112 py3: add utility to forward __str__() to __bytes__()
Yuya Nishihara <yuya@tcha.org>
parents: 32537
diff changeset
   210
        return unifromlocal(bytesfunc(obj))
ce96efec8112 py3: add utility to forward __str__() to __bytes__()
Yuya Nishihara <yuya@tcha.org>
parents: 32537
diff changeset
   211
    return unifunc
ce96efec8112 py3: add utility to forward __str__() to __bytes__()
Yuya Nishihara <yuya@tcha.org>
parents: 32537
diff changeset
   212
31448
6419cd243017 encoding: add converter between native str and byte string
Yuya Nishihara <yuya@tcha.org>
parents: 31447
diff changeset
   213
# converter functions between native str and byte string. use these if the
6419cd243017 encoding: add converter between native str and byte string
Yuya Nishihara <yuya@tcha.org>
parents: 31447
diff changeset
   214
# character encoding is not aware (e.g. exception message) or is known to
6419cd243017 encoding: add converter between native str and byte string
Yuya Nishihara <yuya@tcha.org>
parents: 31447
diff changeset
   215
# be locale dependent (e.g. date formatting.)
6419cd243017 encoding: add converter between native str and byte string
Yuya Nishihara <yuya@tcha.org>
parents: 31447
diff changeset
   216
if pycompat.ispy3:
6419cd243017 encoding: add converter between native str and byte string
Yuya Nishihara <yuya@tcha.org>
parents: 31447
diff changeset
   217
    strtolocal = unitolocal
6419cd243017 encoding: add converter between native str and byte string
Yuya Nishihara <yuya@tcha.org>
parents: 31447
diff changeset
   218
    strfromlocal = unifromlocal
33022
ce96efec8112 py3: add utility to forward __str__() to __bytes__()
Yuya Nishihara <yuya@tcha.org>
parents: 32537
diff changeset
   219
    strmethod = unimethod
31448
6419cd243017 encoding: add converter between native str and byte string
Yuya Nishihara <yuya@tcha.org>
parents: 31447
diff changeset
   220
else:
31774
7d2cbe11ae48 pycompat: introduce identity function as a compat stub
Yuya Nishihara <yuya@tcha.org>
parents: 31448
diff changeset
   221
    strtolocal = pycompat.identity
7d2cbe11ae48 pycompat: introduce identity function as a compat stub
Yuya Nishihara <yuya@tcha.org>
parents: 31448
diff changeset
   222
    strfromlocal = pycompat.identity
33022
ce96efec8112 py3: add utility to forward __str__() to __bytes__()
Yuya Nishihara <yuya@tcha.org>
parents: 32537
diff changeset
   223
    strmethod = pycompat.identity
31448
6419cd243017 encoding: add converter between native str and byte string
Yuya Nishihara <yuya@tcha.org>
parents: 31447
diff changeset
   224
30034
e4a6b439acc5 py3: provide encoding.environ which is a dict of bytes
Yuya Nishihara <yuya@tcha.org>
parents: 30033
diff changeset
   225
if not _nativeenviron:
e4a6b439acc5 py3: provide encoding.environ which is a dict of bytes
Yuya Nishihara <yuya@tcha.org>
parents: 30033
diff changeset
   226
    # now encoding and helper functions are available, recreate the environ
e4a6b439acc5 py3: provide encoding.environ which is a dict of bytes
Yuya Nishihara <yuya@tcha.org>
parents: 30033
diff changeset
   227
    # dict to be exported to other modules
e4a6b439acc5 py3: provide encoding.environ which is a dict of bytes
Yuya Nishihara <yuya@tcha.org>
parents: 30033
diff changeset
   228
    environ = dict((tolocal(k.encode(u'utf-8')), tolocal(v.encode(u'utf-8')))
32184
cf424dae5dc7 check-code: ignore re-exports of os.environ in encoding.py
Yuya Nishihara <yuya@tcha.org>
parents: 32156
diff changeset
   229
                   for k, v in os.environ.items())  # re-exports
30034
e4a6b439acc5 py3: provide encoding.environ which is a dict of bytes
Yuya Nishihara <yuya@tcha.org>
parents: 30033
diff changeset
   230
12866
eddc20306ab6 encoding: default ambiguous character to narrow
Matt Mackall <mpm@selenic.com>
parents: 12770
diff changeset
   231
# How to treat ambiguous-width characters. Set to 'wide' to treat as wide.
32537
044f3d7eb9ae encoding: make sure "wide" variable never be referenced from other modules
Yuya Nishihara <yuya@tcha.org>
parents: 32529
diff changeset
   232
_wide = _sysstr(environ.get("HGENCODINGAMBIGUOUS", "narrow") == "wide"
044f3d7eb9ae encoding: make sure "wide" variable never be referenced from other modules
Yuya Nishihara <yuya@tcha.org>
parents: 32529
diff changeset
   233
                and "WFA" or "WF")
12866
eddc20306ab6 encoding: default ambiguous character to narrow
Matt Mackall <mpm@selenic.com>
parents: 12770
diff changeset
   234
7948
de377b1a9a84 move encoding bits from util to encoding
Matt Mackall <mpm@selenic.com>
parents:
diff changeset
   235
def colwidth(s):
15142
176882876780 encoding: colwidth input is in the local encoding
Matt Mackall <mpm@selenic.com>
parents: 15066
diff changeset
   236
    "Find the column width of a string for display in the local encoding"
30033
02dbfaa6df0b py3: convert encoding name and mode to str
Yuya Nishihara <yuya@tcha.org>
parents: 30030
diff changeset
   237
    return ucolwidth(s.decode(_sysstr(encoding), u'replace'))
15066
24efa83d81cb i18n: calculate terminal columns by width information of each characters
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents: 14951
diff changeset
   238
24efa83d81cb i18n: calculate terminal columns by width information of each characters
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents: 14951
diff changeset
   239
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
   240
    "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
   241
    eaw = getattr(unicodedata, 'east_asian_width', None)
61807854004e encoding: use getattr isntead of hasattr
Augie Fackler <durin42@gmail.com>
parents: 14069
diff changeset
   242
    if eaw is not None:
32537
044f3d7eb9ae encoding: make sure "wide" variable never be referenced from other modules
Yuya Nishihara <yuya@tcha.org>
parents: 32529
diff changeset
   243
        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
   244
    return len(d)
de377b1a9a84 move encoding bits from util to encoding
Matt Mackall <mpm@selenic.com>
parents:
diff changeset
   245
15143
16c129b0f465 encoding: add getcols to extract substrings based on column width
Matt Mackall <mpm@selenic.com>
parents: 15142
diff changeset
   246
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
   247
    '''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
   248
    index start'''
16c129b0f465 encoding: add getcols to extract substrings based on column width
Matt Mackall <mpm@selenic.com>
parents: 15142
diff changeset
   249
    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
   250
        t = s[start:x]
16c129b0f465 encoding: add getcols to extract substrings based on column width
Matt Mackall <mpm@selenic.com>
parents: 15142
diff changeset
   251
        if colwidth(t) == c:
16c129b0f465 encoding: add getcols to extract substrings based on column width
Matt Mackall <mpm@selenic.com>
parents: 15142
diff changeset
   252
            return t
16c129b0f465 encoding: add getcols to extract substrings based on column width
Matt Mackall <mpm@selenic.com>
parents: 15142
diff changeset
   253
21861
b515c3a63e96 encoding: add 'leftside' argument into 'trim' to switch trimming side
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents: 21856
diff changeset
   254
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
   255
    """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
   256
21861
b515c3a63e96 encoding: add 'leftside' argument into 'trim' to switch trimming side
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents: 21856
diff changeset
   257
    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
   258
    '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
   259
34136
414a3513c2bd doctest: do not embed non-ascii characters in docstring
Yuya Nishihara <yuya@tcha.org>
parents: 34135
diff changeset
   260
    >>> from .node import bin
34137
a8994d08e4a2 doctest: use print_function and convert bytes to unicode where needed
Yuya Nishihara <yuya@tcha.org>
parents: 34136
diff changeset
   261
    >>> def bprint(s):
a8994d08e4a2 doctest: use print_function and convert bytes to unicode where needed
Yuya Nishihara <yuya@tcha.org>
parents: 34136
diff changeset
   262
    ...     print(pycompat.sysstr(s))
34131
0fa781320203 doctest: bulk-replace string literals with b'' for Python 3
Yuya Nishihara <yuya@tcha.org>
parents: 33928
diff changeset
   263
    >>> ellipsis = b'+++'
27355
b479fc425a81 encoding: use absolute_import
Gregory Szorc <gregory.szorc@gmail.com>
parents: 26963
diff changeset
   264
    >>> from . import encoding
34131
0fa781320203 doctest: bulk-replace string literals with b'' for Python 3
Yuya Nishihara <yuya@tcha.org>
parents: 33928
diff changeset
   265
    >>> encoding.encoding = b'utf-8'
0fa781320203 doctest: bulk-replace string literals with b'' for Python 3
Yuya Nishihara <yuya@tcha.org>
parents: 33928
diff changeset
   266
    >>> t = b'1234567890'
34137
a8994d08e4a2 doctest: use print_function and convert bytes to unicode where needed
Yuya Nishihara <yuya@tcha.org>
parents: 34136
diff changeset
   267
    >>> bprint(trim(t, 12, ellipsis=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
   268
    1234567890
34137
a8994d08e4a2 doctest: use print_function and convert bytes to unicode where needed
Yuya Nishihara <yuya@tcha.org>
parents: 34136
diff changeset
   269
    >>> bprint(trim(t, 10, ellipsis=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
   270
    1234567890
34137
a8994d08e4a2 doctest: use print_function and convert bytes to unicode where needed
Yuya Nishihara <yuya@tcha.org>
parents: 34136
diff changeset
   271
    >>> bprint(trim(t, 8, ellipsis=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
   272
    12345+++
34137
a8994d08e4a2 doctest: use print_function and convert bytes to unicode where needed
Yuya Nishihara <yuya@tcha.org>
parents: 34136
diff changeset
   273
    >>> bprint(trim(t, 8, ellipsis=ellipsis, leftside=True))
21861
b515c3a63e96 encoding: add 'leftside' argument into 'trim' to switch trimming side
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents: 21856
diff changeset
   274
    +++67890
34137
a8994d08e4a2 doctest: use print_function and convert bytes to unicode where needed
Yuya Nishihara <yuya@tcha.org>
parents: 34136
diff changeset
   275
    >>> bprint(trim(t, 8))
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
   276
    12345678
34137
a8994d08e4a2 doctest: use print_function and convert bytes to unicode where needed
Yuya Nishihara <yuya@tcha.org>
parents: 34136
diff changeset
   277
    >>> bprint(trim(t, 8, leftside=True))
21861
b515c3a63e96 encoding: add 'leftside' argument into 'trim' to switch trimming side
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents: 21856
diff changeset
   278
    34567890
34137
a8994d08e4a2 doctest: use print_function and convert bytes to unicode where needed
Yuya Nishihara <yuya@tcha.org>
parents: 34136
diff changeset
   279
    >>> bprint(trim(t, 3, ellipsis=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
   280
    +++
34137
a8994d08e4a2 doctest: use print_function and convert bytes to unicode where needed
Yuya Nishihara <yuya@tcha.org>
parents: 34136
diff changeset
   281
    >>> bprint(trim(t, 1, ellipsis=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
   282
    +
d24969ee272f encoding: add 'trim' to trim multi-byte characters at most specified columns
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents: 18378
diff changeset
   283
    >>> u = u'\u3042\u3044\u3046\u3048\u304a' # 2 x 5 = 10 columns
34135
e9e225f16932 doctest: pass encoding name as system string
Yuya Nishihara <yuya@tcha.org>
parents: 34131
diff changeset
   284
    >>> t = u.encode(pycompat.sysstr(encoding.encoding))
34137
a8994d08e4a2 doctest: use print_function and convert bytes to unicode where needed
Yuya Nishihara <yuya@tcha.org>
parents: 34136
diff changeset
   285
    >>> bprint(trim(t, 12, ellipsis=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
   286
    \xe3\x81\x82\xe3\x81\x84\xe3\x81\x86\xe3\x81\x88\xe3\x81\x8a
34137
a8994d08e4a2 doctest: use print_function and convert bytes to unicode where needed
Yuya Nishihara <yuya@tcha.org>
parents: 34136
diff changeset
   287
    >>> bprint(trim(t, 10, ellipsis=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
   288
    \xe3\x81\x82\xe3\x81\x84\xe3\x81\x86\xe3\x81\x88\xe3\x81\x8a
34137
a8994d08e4a2 doctest: use print_function and convert bytes to unicode where needed
Yuya Nishihara <yuya@tcha.org>
parents: 34136
diff changeset
   289
    >>> bprint(trim(t, 8, ellipsis=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
   290
    \xe3\x81\x82\xe3\x81\x84+++
34137
a8994d08e4a2 doctest: use print_function and convert bytes to unicode where needed
Yuya Nishihara <yuya@tcha.org>
parents: 34136
diff changeset
   291
    >>> bprint(trim(t, 8, ellipsis=ellipsis, leftside=True))
21861
b515c3a63e96 encoding: add 'leftside' argument into 'trim' to switch trimming side
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents: 21856
diff changeset
   292
    +++\xe3\x81\x88\xe3\x81\x8a
34137
a8994d08e4a2 doctest: use print_function and convert bytes to unicode where needed
Yuya Nishihara <yuya@tcha.org>
parents: 34136
diff changeset
   293
    >>> bprint(trim(t, 5))
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
   294
    \xe3\x81\x82\xe3\x81\x84
34137
a8994d08e4a2 doctest: use print_function and convert bytes to unicode where needed
Yuya Nishihara <yuya@tcha.org>
parents: 34136
diff changeset
   295
    >>> bprint(trim(t, 5, leftside=True))
21861
b515c3a63e96 encoding: add 'leftside' argument into 'trim' to switch trimming side
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents: 21856
diff changeset
   296
    \xe3\x81\x88\xe3\x81\x8a
34137
a8994d08e4a2 doctest: use print_function and convert bytes to unicode where needed
Yuya Nishihara <yuya@tcha.org>
parents: 34136
diff changeset
   297
    >>> bprint(trim(t, 4, ellipsis=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
   298
    +++
34137
a8994d08e4a2 doctest: use print_function and convert bytes to unicode where needed
Yuya Nishihara <yuya@tcha.org>
parents: 34136
diff changeset
   299
    >>> bprint(trim(t, 4, ellipsis=ellipsis, leftside=True))
21861
b515c3a63e96 encoding: add 'leftside' argument into 'trim' to switch trimming side
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents: 21856
diff changeset
   300
    +++
34136
414a3513c2bd doctest: do not embed non-ascii characters in docstring
Yuya Nishihara <yuya@tcha.org>
parents: 34135
diff changeset
   301
    >>> t = bin(b'112233445566778899aa') # invalid byte sequence
34137
a8994d08e4a2 doctest: use print_function and convert bytes to unicode where needed
Yuya Nishihara <yuya@tcha.org>
parents: 34136
diff changeset
   302
    >>> bprint(trim(t, 12, ellipsis=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
   303
    \x11\x22\x33\x44\x55\x66\x77\x88\x99\xaa
34137
a8994d08e4a2 doctest: use print_function and convert bytes to unicode where needed
Yuya Nishihara <yuya@tcha.org>
parents: 34136
diff changeset
   304
    >>> bprint(trim(t, 10, ellipsis=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
   305
    \x11\x22\x33\x44\x55\x66\x77\x88\x99\xaa
34137
a8994d08e4a2 doctest: use print_function and convert bytes to unicode where needed
Yuya Nishihara <yuya@tcha.org>
parents: 34136
diff changeset
   306
    >>> bprint(trim(t, 8, ellipsis=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
   307
    \x11\x22\x33\x44\x55+++
34137
a8994d08e4a2 doctest: use print_function and convert bytes to unicode where needed
Yuya Nishihara <yuya@tcha.org>
parents: 34136
diff changeset
   308
    >>> bprint(trim(t, 8, ellipsis=ellipsis, leftside=True))
21861
b515c3a63e96 encoding: add 'leftside' argument into 'trim' to switch trimming side
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents: 21856
diff changeset
   309
    +++\x66\x77\x88\x99\xaa
34137
a8994d08e4a2 doctest: use print_function and convert bytes to unicode where needed
Yuya Nishihara <yuya@tcha.org>
parents: 34136
diff changeset
   310
    >>> bprint(trim(t, 8))
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
   311
    \x11\x22\x33\x44\x55\x66\x77\x88
34137
a8994d08e4a2 doctest: use print_function and convert bytes to unicode where needed
Yuya Nishihara <yuya@tcha.org>
parents: 34136
diff changeset
   312
    >>> bprint(trim(t, 8, leftside=True))
21861
b515c3a63e96 encoding: add 'leftside' argument into 'trim' to switch trimming side
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents: 21856
diff changeset
   313
    \x33\x44\x55\x66\x77\x88\x99\xaa
34137
a8994d08e4a2 doctest: use print_function and convert bytes to unicode where needed
Yuya Nishihara <yuya@tcha.org>
parents: 34136
diff changeset
   314
    >>> bprint(trim(t, 3, ellipsis=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
   315
    +++
34137
a8994d08e4a2 doctest: use print_function and convert bytes to unicode where needed
Yuya Nishihara <yuya@tcha.org>
parents: 34136
diff changeset
   316
    >>> bprint(trim(t, 1, ellipsis=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
   317
    +
d24969ee272f encoding: add 'trim' to trim multi-byte characters at most specified columns
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents: 18378
diff changeset
   318
    """
d24969ee272f encoding: add 'trim' to trim multi-byte characters at most specified columns
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents: 18378
diff changeset
   319
    try:
30033
02dbfaa6df0b py3: convert encoding name and mode to str
Yuya Nishihara <yuya@tcha.org>
parents: 30030
diff changeset
   320
        u = s.decode(_sysstr(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
   321
    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
   322
        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
   323
            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
   324
        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
   325
        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
   326
            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
   327
        if leftside:
b515c3a63e96 encoding: add 'leftside' argument into 'trim' to switch trimming side
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents: 21856
diff changeset
   328
            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
   329
        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
   330
d24969ee272f encoding: add 'trim' to trim multi-byte characters at most specified columns
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents: 18378
diff changeset
   331
    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
   332
        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
   333
d24969ee272f encoding: add 'trim' to trim multi-byte characters at most specified columns
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents: 18378
diff changeset
   334
    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
   335
    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
   336
        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
   337
21861
b515c3a63e96 encoding: add 'leftside' argument into 'trim' to switch trimming side
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents: 21856
diff changeset
   338
    if leftside:
b515c3a63e96 encoding: add 'leftside' argument into 'trim' to switch trimming side
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents: 21856
diff changeset
   339
        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
   340
        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
   341
    else:
b515c3a63e96 encoding: add 'leftside' argument into 'trim' to switch trimming side
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents: 21856
diff changeset
   342
        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
   343
        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
   344
    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
   345
        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
   346
        if ucolwidth(usub) <= width:
30033
02dbfaa6df0b py3: convert encoding name and mode to str
Yuya Nishihara <yuya@tcha.org>
parents: 30030
diff changeset
   347
            return concat(usub.encode(_sysstr(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
   348
    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
   349
14069
e38846a79a23 encoding: add an encoding-aware lower function
Matt Mackall <mpm@selenic.com>
parents: 13940
diff changeset
   350
def lower(s):
e38846a79a23 encoding: add an encoding-aware lower function
Matt Mackall <mpm@selenic.com>
parents: 13940
diff changeset
   351
    "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
   352
    try:
22779
d9585dda63c3 encoding.lower: use fast ASCII lower
Siddharth Agarwal <sid0@fb.com>
parents: 22778
diff changeset
   353
        return asciilower(s)
17235
3745ae495ce5 encoding: use s.decode to trigger UnicodeDecodeError
Martin Geisler <mg@aragost.com>
parents: 16493
diff changeset
   354
    except UnicodeDecodeError:
16387
c481761033bd encoding: add fast-path for ASCII lowercase
Matt Mackall <mpm@selenic.com>
parents: 16274
diff changeset
   355
        pass
c481761033bd encoding: add fast-path for ASCII lowercase
Matt Mackall <mpm@selenic.com>
parents: 16274
diff changeset
   356
    try:
14069
e38846a79a23 encoding: add an encoding-aware lower function
Matt Mackall <mpm@selenic.com>
parents: 13940
diff changeset
   357
        if isinstance(s, localstr):
e38846a79a23 encoding: add an encoding-aware lower function
Matt Mackall <mpm@selenic.com>
parents: 13940
diff changeset
   358
            u = s._utf8.decode("utf-8")
e38846a79a23 encoding: add an encoding-aware lower function
Matt Mackall <mpm@selenic.com>
parents: 13940
diff changeset
   359
        else:
30033
02dbfaa6df0b py3: convert encoding name and mode to str
Yuya Nishihara <yuya@tcha.org>
parents: 30030
diff changeset
   360
            u = s.decode(_sysstr(encoding), _sysstr(encodingmode))
14069
e38846a79a23 encoding: add an encoding-aware lower function
Matt Mackall <mpm@selenic.com>
parents: 13940
diff changeset
   361
e38846a79a23 encoding: add an encoding-aware lower function
Matt Mackall <mpm@selenic.com>
parents: 13940
diff changeset
   362
        lu = u.lower()
e38846a79a23 encoding: add an encoding-aware lower function
Matt Mackall <mpm@selenic.com>
parents: 13940
diff changeset
   363
        if u == lu:
e38846a79a23 encoding: add an encoding-aware lower function
Matt Mackall <mpm@selenic.com>
parents: 13940
diff changeset
   364
            return s # preserve localstring
30033
02dbfaa6df0b py3: convert encoding name and mode to str
Yuya Nishihara <yuya@tcha.org>
parents: 30030
diff changeset
   365
        return lu.encode(_sysstr(encoding))
14069
e38846a79a23 encoding: add an encoding-aware lower function
Matt Mackall <mpm@selenic.com>
parents: 13940
diff changeset
   366
    except UnicodeError:
e38846a79a23 encoding: add an encoding-aware lower function
Matt Mackall <mpm@selenic.com>
parents: 13940
diff changeset
   367
        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
   368
    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
   369
        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
   370
2ebe3d0ce91d i18n: use encoding.lower/upper for encoding aware case folding
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents: 15143
diff changeset
   371
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
   372
    "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
   373
    try:
24578
ac08de78de7f encoding: use parsers.asciiupper when available
Siddharth Agarwal <sid0@fb.com>
parents: 23596
diff changeset
   374
        return asciiupper(s)
17236
9fb8312dbdbd encoding: add fast-path for ASCII uppercase.
Martin Geisler <mg@aragost.com>
parents: 17235
diff changeset
   375
    except UnicodeDecodeError:
24597
b4258d5a1600 encoding.upper: factor out fallback code
Siddharth Agarwal <sid0@fb.com>
parents: 24593
diff changeset
   376
        return upperfallback(s)
b4258d5a1600 encoding.upper: factor out fallback code
Siddharth Agarwal <sid0@fb.com>
parents: 24593
diff changeset
   377
b4258d5a1600 encoding.upper: factor out fallback code
Siddharth Agarwal <sid0@fb.com>
parents: 24593
diff changeset
   378
def upperfallback(s):
17236
9fb8312dbdbd encoding: add fast-path for ASCII uppercase.
Martin Geisler <mg@aragost.com>
parents: 17235
diff changeset
   379
    try:
15672
2ebe3d0ce91d i18n: use encoding.lower/upper for encoding aware case folding
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents: 15143
diff changeset
   380
        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
   381
            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
   382
        else:
30033
02dbfaa6df0b py3: convert encoding name and mode to str
Yuya Nishihara <yuya@tcha.org>
parents: 30030
diff changeset
   383
            u = s.decode(_sysstr(encoding), _sysstr(encodingmode))
15672
2ebe3d0ce91d i18n: use encoding.lower/upper for encoding aware case folding
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents: 15143
diff changeset
   384
2ebe3d0ce91d i18n: use encoding.lower/upper for encoding aware case folding
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents: 15143
diff changeset
   385
        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
   386
        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
   387
            return s # preserve localstring
30033
02dbfaa6df0b py3: convert encoding name and mode to str
Yuya Nishihara <yuya@tcha.org>
parents: 30030
diff changeset
   388
        return uu.encode(_sysstr(encoding))
15672
2ebe3d0ce91d i18n: use encoding.lower/upper for encoding aware case folding
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents: 15143
diff changeset
   389
    except UnicodeError:
2ebe3d0ce91d i18n: use encoding.lower/upper for encoding aware case folding
FUJIWARA Katsunori <foozy@lares.dti.ne.jp>
parents: 15143
diff changeset
   390
        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
   391
    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
   392
        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
   393
24593
f473a1fe5c7c encoding: define an enum that specifies what normcase does to ASCII strings
Siddharth Agarwal <sid0@fb.com>
parents: 24578
diff changeset
   394
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
   395
    '''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
   396
f473a1fe5c7c encoding: define an enum that specifies what normcase does to ASCII strings
Siddharth Agarwal <sid0@fb.com>
parents: 24578
diff changeset
   397
    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
   398
    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
   399
f473a1fe5c7c encoding: define an enum that specifies what normcase does to ASCII strings
Siddharth Agarwal <sid0@fb.com>
parents: 24578
diff changeset
   400
    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
   401
    upper: normcase uppercases ASCII strings
24608
1c533e23ce95 util.h: define an enum for normcase specs
Siddharth Agarwal <sid0@fb.com>
parents: 24597
diff changeset
   402
    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
   403
1c533e23ce95 util.h: define an enum for normcase specs
Siddharth Agarwal <sid0@fb.com>
parents: 24597
diff changeset
   404
    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
   405
    lower = -1
f473a1fe5c7c encoding: define an enum that specifies what normcase does to ASCII strings
Siddharth Agarwal <sid0@fb.com>
parents: 24578
diff changeset
   406
    upper = 1
f473a1fe5c7c encoding: define an enum that specifies what normcase does to ASCII strings
Siddharth Agarwal <sid0@fb.com>
parents: 24578
diff changeset
   407
    other = 0
f473a1fe5c7c encoding: define an enum that specifies what normcase does to ASCII strings
Siddharth Agarwal <sid0@fb.com>
parents: 24578
diff changeset
   408
28068
9ece901f7a19 encoding: add option to escape non-ascii characters in JSON
Yuya Nishihara <yuya@tcha.org>
parents: 28067
diff changeset
   409
def jsonescape(s, paranoid=False):
22426
f6b533e64ed6 encoding: add json escaping filter
Matt Mackall <mpm@selenic.com>
parents: 22425
diff changeset
   410
    '''returns a string suitable for JSON
f6b533e64ed6 encoding: add json escaping filter
Matt Mackall <mpm@selenic.com>
parents: 22425
diff changeset
   411
f6b533e64ed6 encoding: add json escaping filter
Matt Mackall <mpm@selenic.com>
parents: 22425
diff changeset
   412
    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
   413
    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
   414
37947
3ea3c96ada54 encoding: introduce tagging type for non-lossy non-ASCII string
Yuya Nishihara <yuya@tcha.org>
parents: 37946
diff changeset
   415
    - localstr/safelocalstr objects are converted back to UTF-8
22426
f6b533e64ed6 encoding: add json escaping filter
Matt Mackall <mpm@selenic.com>
parents: 22425
diff changeset
   416
    - valid UTF-8/ASCII strings are passed as-is
f6b533e64ed6 encoding: add json escaping filter
Matt Mackall <mpm@selenic.com>
parents: 22425
diff changeset
   417
    - other strings are converted to UTF-8b surrogate encoding
f6b533e64ed6 encoding: add json escaping filter
Matt Mackall <mpm@selenic.com>
parents: 22425
diff changeset
   418
    - apply JSON-specified string escaping
f6b533e64ed6 encoding: add json escaping filter
Matt Mackall <mpm@selenic.com>
parents: 22425
diff changeset
   419
f6b533e64ed6 encoding: add json escaping filter
Matt Mackall <mpm@selenic.com>
parents: 22425
diff changeset
   420
    (escapes are doubled in these tests)
f6b533e64ed6 encoding: add json escaping filter
Matt Mackall <mpm@selenic.com>
parents: 22425
diff changeset
   421
34131
0fa781320203 doctest: bulk-replace string literals with b'' for Python 3
Yuya Nishihara <yuya@tcha.org>
parents: 33928
diff changeset
   422
    >>> jsonescape(b'this is a test')
22426
f6b533e64ed6 encoding: add json escaping filter
Matt Mackall <mpm@selenic.com>
parents: 22425
diff changeset
   423
    'this is a test'
34131
0fa781320203 doctest: bulk-replace string literals with b'' for Python 3
Yuya Nishihara <yuya@tcha.org>
parents: 33928
diff changeset
   424
    >>> jsonescape(b'escape characters: \\0 \\x0b \\x7f')
27881
ffa599f3f503 encoding: escape U+007F (DEL) character in JSON
Yuya Nishihara <yuya@tcha.org>
parents: 27699
diff changeset
   425
    'escape characters: \\\\u0000 \\\\u000b \\\\u007f'
34131
0fa781320203 doctest: bulk-replace string literals with b'' for Python 3
Yuya Nishihara <yuya@tcha.org>
parents: 33928
diff changeset
   426
    >>> jsonescape(b'escape characters: \\b \\t \\n \\f \\r \\" \\\\')
33925
2c37f9dabc32 encoding: add fast path of jsonescape() (issue5533)
Yuya Nishihara <yuya@tcha.org>
parents: 33924
diff changeset
   427
    'escape characters: \\\\b \\\\t \\\\n \\\\f \\\\r \\\\" \\\\\\\\'
34131
0fa781320203 doctest: bulk-replace string literals with b'' for Python 3
Yuya Nishihara <yuya@tcha.org>
parents: 33928
diff changeset
   428
    >>> jsonescape(b'a weird byte: \\xdd')
22426
f6b533e64ed6 encoding: add json escaping filter
Matt Mackall <mpm@selenic.com>
parents: 22425
diff changeset
   429
    'a weird byte: \\xed\\xb3\\x9d'
34131
0fa781320203 doctest: bulk-replace string literals with b'' for Python 3
Yuya Nishihara <yuya@tcha.org>
parents: 33928
diff changeset
   430
    >>> jsonescape(b'utf-8: caf\\xc3\\xa9')
22426
f6b533e64ed6 encoding: add json escaping filter
Matt Mackall <mpm@selenic.com>
parents: 22425
diff changeset
   431
    'utf-8: caf\\xc3\\xa9'
34131
0fa781320203 doctest: bulk-replace string literals with b'' for Python 3
Yuya Nishihara <yuya@tcha.org>
parents: 33928
diff changeset
   432
    >>> jsonescape(b'')
22426
f6b533e64ed6 encoding: add json escaping filter
Matt Mackall <mpm@selenic.com>
parents: 22425
diff changeset
   433
    ''
28068
9ece901f7a19 encoding: add option to escape non-ascii characters in JSON
Yuya Nishihara <yuya@tcha.org>
parents: 28067
diff changeset
   434
28069
b2d24c2898f9 encoding: backport paranoid escaping from templatefilters.jsonescape()
Yuya Nishihara <yuya@tcha.org>
parents: 28068
diff changeset
   435
    If paranoid, non-ascii and common troublesome characters are also escaped.
b2d24c2898f9 encoding: backport paranoid escaping from templatefilters.jsonescape()
Yuya Nishihara <yuya@tcha.org>
parents: 28068
diff changeset
   436
    This is suitable for web output.
28068
9ece901f7a19 encoding: add option to escape non-ascii characters in JSON
Yuya Nishihara <yuya@tcha.org>
parents: 28067
diff changeset
   437
34131
0fa781320203 doctest: bulk-replace string literals with b'' for Python 3
Yuya Nishihara <yuya@tcha.org>
parents: 33928
diff changeset
   438
    >>> s = b'escape characters: \\0 \\x0b \\x7f'
33925
2c37f9dabc32 encoding: add fast path of jsonescape() (issue5533)
Yuya Nishihara <yuya@tcha.org>
parents: 33924
diff changeset
   439
    >>> assert jsonescape(s) == jsonescape(s, paranoid=True)
34131
0fa781320203 doctest: bulk-replace string literals with b'' for Python 3
Yuya Nishihara <yuya@tcha.org>
parents: 33928
diff changeset
   440
    >>> s = b'escape characters: \\b \\t \\n \\f \\r \\" \\\\'
33925
2c37f9dabc32 encoding: add fast path of jsonescape() (issue5533)
Yuya Nishihara <yuya@tcha.org>
parents: 33924
diff changeset
   441
    >>> assert jsonescape(s) == jsonescape(s, paranoid=True)
34131
0fa781320203 doctest: bulk-replace string literals with b'' for Python 3
Yuya Nishihara <yuya@tcha.org>
parents: 33928
diff changeset
   442
    >>> jsonescape(b'escape boundary: \\x7e \\x7f \\xc2\\x80', paranoid=True)
28068
9ece901f7a19 encoding: add option to escape non-ascii characters in JSON
Yuya Nishihara <yuya@tcha.org>
parents: 28067
diff changeset
   443
    'escape boundary: ~ \\\\u007f \\\\u0080'
34131
0fa781320203 doctest: bulk-replace string literals with b'' for Python 3
Yuya Nishihara <yuya@tcha.org>
parents: 33928
diff changeset
   444
    >>> jsonescape(b'a weird byte: \\xdd', paranoid=True)
28068
9ece901f7a19 encoding: add option to escape non-ascii characters in JSON
Yuya Nishihara <yuya@tcha.org>
parents: 28067
diff changeset
   445
    'a weird byte: \\\\udcdd'
34131
0fa781320203 doctest: bulk-replace string literals with b'' for Python 3
Yuya Nishihara <yuya@tcha.org>
parents: 33928
diff changeset
   446
    >>> jsonescape(b'utf-8: caf\\xc3\\xa9', paranoid=True)
28068
9ece901f7a19 encoding: add option to escape non-ascii characters in JSON
Yuya Nishihara <yuya@tcha.org>
parents: 28067
diff changeset
   447
    'utf-8: caf\\\\u00e9'
34131
0fa781320203 doctest: bulk-replace string literals with b'' for Python 3
Yuya Nishihara <yuya@tcha.org>
parents: 33928
diff changeset
   448
    >>> jsonescape(b'non-BMP: \\xf0\\x9d\\x84\\x9e', paranoid=True)
28068
9ece901f7a19 encoding: add option to escape non-ascii characters in JSON
Yuya Nishihara <yuya@tcha.org>
parents: 28067
diff changeset
   449
    'non-BMP: \\\\ud834\\\\udd1e'
34131
0fa781320203 doctest: bulk-replace string literals with b'' for Python 3
Yuya Nishihara <yuya@tcha.org>
parents: 33928
diff changeset
   450
    >>> jsonescape(b'<foo@example.org>', paranoid=True)
28069
b2d24c2898f9 encoding: backport paranoid escaping from templatefilters.jsonescape()
Yuya Nishihara <yuya@tcha.org>
parents: 28068
diff changeset
   451
    '\\\\u003cfoo@example.org\\\\u003e'
22426
f6b533e64ed6 encoding: add json escaping filter
Matt Mackall <mpm@selenic.com>
parents: 22425
diff changeset
   452
    '''
f6b533e64ed6 encoding: add json escaping filter
Matt Mackall <mpm@selenic.com>
parents: 22425
diff changeset
   453
28068
9ece901f7a19 encoding: add option to escape non-ascii characters in JSON
Yuya Nishihara <yuya@tcha.org>
parents: 28067
diff changeset
   454
    u8chars = toutf8b(s)
9ece901f7a19 encoding: add option to escape non-ascii characters in JSON
Yuya Nishihara <yuya@tcha.org>
parents: 28067
diff changeset
   455
    try:
33924
b9101467d88b encoding: extract stub for fast JSON escape
Yuya Nishihara <yuya@tcha.org>
parents: 33852
diff changeset
   456
        return _jsonescapeu8fast(u8chars, paranoid)
b9101467d88b encoding: extract stub for fast JSON escape
Yuya Nishihara <yuya@tcha.org>
parents: 33852
diff changeset
   457
    except ValueError:
28068
9ece901f7a19 encoding: add option to escape non-ascii characters in JSON
Yuya Nishihara <yuya@tcha.org>
parents: 28067
diff changeset
   458
        pass
33924
b9101467d88b encoding: extract stub for fast JSON escape
Yuya Nishihara <yuya@tcha.org>
parents: 33852
diff changeset
   459
    return charencodepure.jsonescapeu8fallback(u8chars, paranoid)
22426
f6b533e64ed6 encoding: add json escaping filter
Matt Mackall <mpm@selenic.com>
parents: 22425
diff changeset
   460
34218
aa877860d4d7 py3: use 'surrogatepass' error handler to process U+DCxx transparently
Yuya Nishihara <yuya@tcha.org>
parents: 34216
diff changeset
   461
# We need to decode/encode U+DCxx codes transparently since invalid UTF-8
aa877860d4d7 py3: use 'surrogatepass' error handler to process U+DCxx transparently
Yuya Nishihara <yuya@tcha.org>
parents: 34216
diff changeset
   462
# bytes are mapped to that range.
aa877860d4d7 py3: use 'surrogatepass' error handler to process U+DCxx transparently
Yuya Nishihara <yuya@tcha.org>
parents: 34216
diff changeset
   463
if pycompat.ispy3:
aa877860d4d7 py3: use 'surrogatepass' error handler to process U+DCxx transparently
Yuya Nishihara <yuya@tcha.org>
parents: 34216
diff changeset
   464
    _utf8strict = r'surrogatepass'
aa877860d4d7 py3: use 'surrogatepass' error handler to process U+DCxx transparently
Yuya Nishihara <yuya@tcha.org>
parents: 34216
diff changeset
   465
else:
aa877860d4d7 py3: use 'surrogatepass' error handler to process U+DCxx transparently
Yuya Nishihara <yuya@tcha.org>
parents: 34216
diff changeset
   466
    _utf8strict = r'strict'
aa877860d4d7 py3: use 'surrogatepass' error handler to process U+DCxx transparently
Yuya Nishihara <yuya@tcha.org>
parents: 34216
diff changeset
   467
26875
cf47bdb2183c encoding: add getutf8char helper
Matt Mackall <mpm@selenic.com>
parents: 25660
diff changeset
   468
_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
   469
cf47bdb2183c encoding: add getutf8char helper
Matt Mackall <mpm@selenic.com>
parents: 25660
diff changeset
   470
def getutf8char(s, pos):
cf47bdb2183c encoding: add getutf8char helper
Matt Mackall <mpm@selenic.com>
parents: 25660
diff changeset
   471
    '''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
   472
cf47bdb2183c encoding: add getutf8char helper
Matt Mackall <mpm@selenic.com>
parents: 25660
diff changeset
   473
    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
   474
    utf-8 character.
cf47bdb2183c encoding: add getutf8char helper
Matt Mackall <mpm@selenic.com>
parents: 25660
diff changeset
   475
    '''
cf47bdb2183c encoding: add getutf8char helper
Matt Mackall <mpm@selenic.com>
parents: 25660
diff changeset
   476
cf47bdb2183c encoding: add getutf8char helper
Matt Mackall <mpm@selenic.com>
parents: 25660
diff changeset
   477
    # find how many bytes to attempt decoding from first nibble
34200
112f118ecb00 encoding: ensure getutf8char always returns a bytestr, never an int
Augie Fackler <raf@durin42.com>
parents: 34137
diff changeset
   478
    l = _utf8len[ord(s[pos:pos + 1]) >> 4]
26875
cf47bdb2183c encoding: add getutf8char helper
Matt Mackall <mpm@selenic.com>
parents: 25660
diff changeset
   479
    if not l: # ascii
34200
112f118ecb00 encoding: ensure getutf8char always returns a bytestr, never an int
Augie Fackler <raf@durin42.com>
parents: 34137
diff changeset
   480
        return s[pos:pos + 1]
26875
cf47bdb2183c encoding: add getutf8char helper
Matt Mackall <mpm@selenic.com>
parents: 25660
diff changeset
   481
cf47bdb2183c encoding: add getutf8char helper
Matt Mackall <mpm@selenic.com>
parents: 25660
diff changeset
   482
    c = s[pos:pos + l]
cf47bdb2183c encoding: add getutf8char helper
Matt Mackall <mpm@selenic.com>
parents: 25660
diff changeset
   483
    # validate with attempted decode
34218
aa877860d4d7 py3: use 'surrogatepass' error handler to process U+DCxx transparently
Yuya Nishihara <yuya@tcha.org>
parents: 34216
diff changeset
   484
    c.decode("utf-8", _utf8strict)
26875
cf47bdb2183c encoding: add getutf8char helper
Matt Mackall <mpm@selenic.com>
parents: 25660
diff changeset
   485
    return c
cf47bdb2183c encoding: add getutf8char helper
Matt Mackall <mpm@selenic.com>
parents: 25660
diff changeset
   486
16133
84c58da3a1f8 encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents: 15769
diff changeset
   487
def toutf8b(s):
84c58da3a1f8 encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents: 15769
diff changeset
   488
    '''convert a local, possibly-binary string into UTF-8b
84c58da3a1f8 encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents: 15769
diff changeset
   489
84c58da3a1f8 encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents: 15769
diff changeset
   490
    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
   491
    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
   492
    arbitrary byte strings. As Mercurial often doesn't know
84c58da3a1f8 encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents: 15769
diff changeset
   493
    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
   494
84c58da3a1f8 encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents: 15769
diff changeset
   495
    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
   496
    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
   497
    uDC00-uDCFF.
84c58da3a1f8 encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents: 15769
diff changeset
   498
84c58da3a1f8 encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents: 15769
diff changeset
   499
    Principles of operation:
84c58da3a1f8 encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents: 15769
diff changeset
   500
17424
e7cfe3587ea4 fix trivial spelling errors
Mads Kiilerich <mads@kiilerich.com>
parents: 17236
diff changeset
   501
    - 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
   502
      by Unicode-oriented clients
84c58da3a1f8 encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents: 15769
diff changeset
   503
    - 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
   504
      be round-tripped or recovered by clueful clients
84c58da3a1f8 encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents: 15769
diff changeset
   505
    - 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
   506
      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
   507
      Unicode data they want
37947
3ea3c96ada54 encoding: introduce tagging type for non-lossy non-ASCII string
Yuya Nishihara <yuya@tcha.org>
parents: 37946
diff changeset
   508
    - non-lossy local strings (aka safelocalstr) get sent as UTF-8 as well
16133
84c58da3a1f8 encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents: 15769
diff changeset
   509
    - 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
   510
      filenames, metadata can't be roundtripped without help
84c58da3a1f8 encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents: 15769
diff changeset
   511
84c58da3a1f8 encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents: 15769
diff changeset
   512
    (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
   513
    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
   514
    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
   515
    internal surrogate encoding as a UTF-8 string.)
84c58da3a1f8 encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents: 15769
diff changeset
   516
    '''
84c58da3a1f8 encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents: 15769
diff changeset
   517
37946
57b0c7221dba encoding: fix toutf8b() to resurrect lossy characters even if "\xed" in it
Yuya Nishihara <yuya@tcha.org>
parents: 36797
diff changeset
   518
    if isinstance(s, localstr):
57b0c7221dba encoding: fix toutf8b() to resurrect lossy characters even if "\xed" in it
Yuya Nishihara <yuya@tcha.org>
parents: 36797
diff changeset
   519
        # assume that the original UTF-8 sequence would never contain
57b0c7221dba encoding: fix toutf8b() to resurrect lossy characters even if "\xed" in it
Yuya Nishihara <yuya@tcha.org>
parents: 36797
diff changeset
   520
        # invalid characters in U+DCxx range
57b0c7221dba encoding: fix toutf8b() to resurrect lossy characters even if "\xed" in it
Yuya Nishihara <yuya@tcha.org>
parents: 36797
diff changeset
   521
        return s._utf8
37947
3ea3c96ada54 encoding: introduce tagging type for non-lossy non-ASCII string
Yuya Nishihara <yuya@tcha.org>
parents: 37946
diff changeset
   522
    elif isinstance(s, safelocalstr):
3ea3c96ada54 encoding: introduce tagging type for non-lossy non-ASCII string
Yuya Nishihara <yuya@tcha.org>
parents: 37946
diff changeset
   523
        # already verified that s is non-lossy in legacy encoding, which
3ea3c96ada54 encoding: introduce tagging type for non-lossy non-ASCII string
Yuya Nishihara <yuya@tcha.org>
parents: 37946
diff changeset
   524
        # shouldn't contain characters in U+DCxx range
3ea3c96ada54 encoding: introduce tagging type for non-lossy non-ASCII string
Yuya Nishihara <yuya@tcha.org>
parents: 37946
diff changeset
   525
        return fromlocal(s)
37946
57b0c7221dba encoding: fix toutf8b() to resurrect lossy characters even if "\xed" in it
Yuya Nishihara <yuya@tcha.org>
parents: 36797
diff changeset
   526
    elif isasciistr(s):
33928
6c119dbfd0c0 encoding: add fast path of from/toutf8b() for ASCII strings
Yuya Nishihara <yuya@tcha.org>
parents: 33927
diff changeset
   527
        return s
26879
a24b98f4e03c encoding: re-escape U+DCxx characters in toutf8b input (issue4927)
Matt Mackall <mpm@selenic.com>
parents: 26878
diff changeset
   528
    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
   529
        try:
34218
aa877860d4d7 py3: use 'surrogatepass' error handler to process U+DCxx transparently
Yuya Nishihara <yuya@tcha.org>
parents: 34216
diff changeset
   530
            s.decode('utf-8', _utf8strict)
26879
a24b98f4e03c encoding: re-escape U+DCxx characters in toutf8b input (issue4927)
Matt Mackall <mpm@selenic.com>
parents: 26878
diff changeset
   531
            return s
a24b98f4e03c encoding: re-escape U+DCxx characters in toutf8b input (issue4927)
Matt Mackall <mpm@selenic.com>
parents: 26878
diff changeset
   532
        except UnicodeDecodeError:
a24b98f4e03c encoding: re-escape U+DCxx characters in toutf8b input (issue4927)
Matt Mackall <mpm@selenic.com>
parents: 26878
diff changeset
   533
            pass
26878
d7e83f106459 encoding: use getutf8char in toutf8b
Matt Mackall <mpm@selenic.com>
parents: 26877
diff changeset
   534
34216
1c601df9894c py3: wrap bytes in encoding.from/toutf8b() with bytestr
Yuya Nishihara <yuya@tcha.org>
parents: 34200
diff changeset
   535
    s = pycompat.bytestr(s)
26878
d7e83f106459 encoding: use getutf8char in toutf8b
Matt Mackall <mpm@selenic.com>
parents: 26877
diff changeset
   536
    r = ""
d7e83f106459 encoding: use getutf8char in toutf8b
Matt Mackall <mpm@selenic.com>
parents: 26877
diff changeset
   537
    pos = 0
d7e83f106459 encoding: use getutf8char in toutf8b
Matt Mackall <mpm@selenic.com>
parents: 26877
diff changeset
   538
    l = len(s)
d7e83f106459 encoding: use getutf8char in toutf8b
Matt Mackall <mpm@selenic.com>
parents: 26877
diff changeset
   539
    while pos < l:
d7e83f106459 encoding: use getutf8char in toutf8b
Matt Mackall <mpm@selenic.com>
parents: 26877
diff changeset
   540
        try:
d7e83f106459 encoding: use getutf8char in toutf8b
Matt Mackall <mpm@selenic.com>
parents: 26877
diff changeset
   541
            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
   542
            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
   543
                # have to re-escape existing U+DCxx characters
34218
aa877860d4d7 py3: use 'surrogatepass' error handler to process U+DCxx transparently
Yuya Nishihara <yuya@tcha.org>
parents: 34216
diff changeset
   544
                c = unichr(0xdc00 + ord(s[pos])).encode('utf-8', _utf8strict)
26879
a24b98f4e03c encoding: re-escape U+DCxx characters in toutf8b input (issue4927)
Matt Mackall <mpm@selenic.com>
parents: 26878
diff changeset
   545
                pos += 1
a24b98f4e03c encoding: re-escape U+DCxx characters in toutf8b input (issue4927)
Matt Mackall <mpm@selenic.com>
parents: 26878
diff changeset
   546
            else:
a24b98f4e03c encoding: re-escape U+DCxx characters in toutf8b input (issue4927)
Matt Mackall <mpm@selenic.com>
parents: 26878
diff changeset
   547
                pos += len(c)
26878
d7e83f106459 encoding: use getutf8char in toutf8b
Matt Mackall <mpm@selenic.com>
parents: 26877
diff changeset
   548
        except UnicodeDecodeError:
34218
aa877860d4d7 py3: use 'surrogatepass' error handler to process U+DCxx transparently
Yuya Nishihara <yuya@tcha.org>
parents: 34216
diff changeset
   549
            c = unichr(0xdc00 + ord(s[pos])).encode('utf-8', _utf8strict)
26878
d7e83f106459 encoding: use getutf8char in toutf8b
Matt Mackall <mpm@selenic.com>
parents: 26877
diff changeset
   550
            pos += 1
d7e83f106459 encoding: use getutf8char in toutf8b
Matt Mackall <mpm@selenic.com>
parents: 26877
diff changeset
   551
        r += c
d7e83f106459 encoding: use getutf8char in toutf8b
Matt Mackall <mpm@selenic.com>
parents: 26877
diff changeset
   552
    return r
16133
84c58da3a1f8 encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents: 15769
diff changeset
   553
84c58da3a1f8 encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents: 15769
diff changeset
   554
def fromutf8b(s):
84c58da3a1f8 encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents: 15769
diff changeset
   555
    '''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
   556
84c58da3a1f8 encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents: 15769
diff changeset
   557
    return the original binary string. This
84c58da3a1f8 encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents: 15769
diff changeset
   558
    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
   559
    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
   560
26963
de5ae97ce9f4 encoding: extend test cases for utf8b
Matt Mackall <mpm@selenic.com>
parents: 26879
diff changeset
   561
    >>> roundtrip = lambda x: fromutf8b(toutf8b(x)) == x
34131
0fa781320203 doctest: bulk-replace string literals with b'' for Python 3
Yuya Nishihara <yuya@tcha.org>
parents: 33928
diff changeset
   562
    >>> m = b"\\xc3\\xa9\\x99abcd"
26963
de5ae97ce9f4 encoding: extend test cases for utf8b
Matt Mackall <mpm@selenic.com>
parents: 26879
diff changeset
   563
    >>> toutf8b(m)
16133
84c58da3a1f8 encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents: 15769
diff changeset
   564
    '\\xc3\\xa9\\xed\\xb2\\x99abcd'
26963
de5ae97ce9f4 encoding: extend test cases for utf8b
Matt Mackall <mpm@selenic.com>
parents: 26879
diff changeset
   565
    >>> roundtrip(m)
de5ae97ce9f4 encoding: extend test cases for utf8b
Matt Mackall <mpm@selenic.com>
parents: 26879
diff changeset
   566
    True
34131
0fa781320203 doctest: bulk-replace string literals with b'' for Python 3
Yuya Nishihara <yuya@tcha.org>
parents: 33928
diff changeset
   567
    >>> roundtrip(b"\\xc2\\xc2\\x80")
26963
de5ae97ce9f4 encoding: extend test cases for utf8b
Matt Mackall <mpm@selenic.com>
parents: 26879
diff changeset
   568
    True
34131
0fa781320203 doctest: bulk-replace string literals with b'' for Python 3
Yuya Nishihara <yuya@tcha.org>
parents: 33928
diff changeset
   569
    >>> roundtrip(b"\\xef\\xbf\\xbd")
26963
de5ae97ce9f4 encoding: extend test cases for utf8b
Matt Mackall <mpm@selenic.com>
parents: 26879
diff changeset
   570
    True
34131
0fa781320203 doctest: bulk-replace string literals with b'' for Python 3
Yuya Nishihara <yuya@tcha.org>
parents: 33928
diff changeset
   571
    >>> roundtrip(b"\\xef\\xef\\xbf\\xbd")
16133
84c58da3a1f8 encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents: 15769
diff changeset
   572
    True
34131
0fa781320203 doctest: bulk-replace string literals with b'' for Python 3
Yuya Nishihara <yuya@tcha.org>
parents: 33928
diff changeset
   573
    >>> roundtrip(b"\\xf1\\x80\\x80\\x80\\x80")
27699
c8d3392f76e1 encoding: handle UTF-16 internal limit with fromutf8b (issue5031)
Matt Mackall <mpm@selenic.com>
parents: 27356
diff changeset
   574
    True
16133
84c58da3a1f8 encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents: 15769
diff changeset
   575
    '''
84c58da3a1f8 encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents: 15769
diff changeset
   576
33928
6c119dbfd0c0 encoding: add fast path of from/toutf8b() for ASCII strings
Yuya Nishihara <yuya@tcha.org>
parents: 33927
diff changeset
   577
    if isasciistr(s):
6c119dbfd0c0 encoding: add fast path of from/toutf8b() for ASCII strings
Yuya Nishihara <yuya@tcha.org>
parents: 33927
diff changeset
   578
        return s
16133
84c58da3a1f8 encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents: 15769
diff changeset
   579
    # fast path - look for uDxxx prefixes in s
84c58da3a1f8 encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents: 15769
diff changeset
   580
    if "\xed" not in s:
84c58da3a1f8 encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents: 15769
diff changeset
   581
        return s
84c58da3a1f8 encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents: 15769
diff changeset
   582
27699
c8d3392f76e1 encoding: handle UTF-16 internal limit with fromutf8b (issue5031)
Matt Mackall <mpm@selenic.com>
parents: 27356
diff changeset
   583
    # We could do this with the unicode type but some Python builds
c8d3392f76e1 encoding: handle UTF-16 internal limit with fromutf8b (issue5031)
Matt Mackall <mpm@selenic.com>
parents: 27356
diff changeset
   584
    # use UTF-16 internally (issue5031) which causes non-BMP code
c8d3392f76e1 encoding: handle UTF-16 internal limit with fromutf8b (issue5031)
Matt Mackall <mpm@selenic.com>
parents: 27356
diff changeset
   585
    # points to be escaped. Instead, we use our handy getutf8char
c8d3392f76e1 encoding: handle UTF-16 internal limit with fromutf8b (issue5031)
Matt Mackall <mpm@selenic.com>
parents: 27356
diff changeset
   586
    # helper again to walk the string without "decoding" it.
c8d3392f76e1 encoding: handle UTF-16 internal limit with fromutf8b (issue5031)
Matt Mackall <mpm@selenic.com>
parents: 27356
diff changeset
   587
34216
1c601df9894c py3: wrap bytes in encoding.from/toutf8b() with bytestr
Yuya Nishihara <yuya@tcha.org>
parents: 34200
diff changeset
   588
    s = pycompat.bytestr(s)
16133
84c58da3a1f8 encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents: 15769
diff changeset
   589
    r = ""
27699
c8d3392f76e1 encoding: handle UTF-16 internal limit with fromutf8b (issue5031)
Matt Mackall <mpm@selenic.com>
parents: 27356
diff changeset
   590
    pos = 0
c8d3392f76e1 encoding: handle UTF-16 internal limit with fromutf8b (issue5031)
Matt Mackall <mpm@selenic.com>
parents: 27356
diff changeset
   591
    l = len(s)
c8d3392f76e1 encoding: handle UTF-16 internal limit with fromutf8b (issue5031)
Matt Mackall <mpm@selenic.com>
parents: 27356
diff changeset
   592
    while pos < l:
c8d3392f76e1 encoding: handle UTF-16 internal limit with fromutf8b (issue5031)
Matt Mackall <mpm@selenic.com>
parents: 27356
diff changeset
   593
        c = getutf8char(s, pos)
c8d3392f76e1 encoding: handle UTF-16 internal limit with fromutf8b (issue5031)
Matt Mackall <mpm@selenic.com>
parents: 27356
diff changeset
   594
        pos += len(c)
c8d3392f76e1 encoding: handle UTF-16 internal limit with fromutf8b (issue5031)
Matt Mackall <mpm@selenic.com>
parents: 27356
diff changeset
   595
        # unescape U+DCxx characters
c8d3392f76e1 encoding: handle UTF-16 internal limit with fromutf8b (issue5031)
Matt Mackall <mpm@selenic.com>
parents: 27356
diff changeset
   596
        if "\xed\xb0\x80" <= c <= "\xed\xb3\xbf":
34218
aa877860d4d7 py3: use 'surrogatepass' error handler to process U+DCxx transparently
Yuya Nishihara <yuya@tcha.org>
parents: 34216
diff changeset
   597
            c = pycompat.bytechr(ord(c.decode("utf-8", _utf8strict)) & 0xff)
27699
c8d3392f76e1 encoding: handle UTF-16 internal limit with fromutf8b (issue5031)
Matt Mackall <mpm@selenic.com>
parents: 27356
diff changeset
   598
        r += c
16133
84c58da3a1f8 encoding: introduce utf8-b helpers
Matt Mackall <mpm@selenic.com>
parents: 15769
diff changeset
   599
    return r