# encoding.py - character transcoding support for Mercurial

#

#  Copyright 2005-2009 Matt Mackall <mpm@selenic.com> and others

#

# This software may be used and distributed according to the terms of the

# GNU General Public License version 2 or any later version.

from __future__ import absolute_import

import array

import locale

import os

import sys

import unicodedata

from . import (

    error,

)

if sys.version_info[0] >= 3:

    unichr = chr

# These unicode characters are ignored by HFS+ (Apple Technote 1150,

# "Unicode Subtleties"), so we need to ignore them in some places for

# sanity.

_ignore = [unichr(int(x, 16)).encode("utf-8") for x in

           "200c 200d 200e 200f 202a 202b 202c 202d 202e "

           "206a 206b 206c 206d 206e 206f feff".split()]

# verify the next function will work

if sys.version_info[0] >= 3:

    assert set(i[0] for i in _ignore) == set([ord(b'\xe2'), ord(b'\xef')])

else:

    assert set(i[0] for i in _ignore) == set(["\xe2", "\xef"])

def hfsignoreclean(s):

    """Remove codepoints ignored by HFS+ from s.

    >>> hfsignoreclean(u'.h\u200cg'.encode('utf-8'))

    '.hg'

    >>> hfsignoreclean(u'.h\ufeffg'.encode('utf-8'))

    '.hg'

    """

    if "\xe2" in s or "\xef" in s:

        for c in _ignore:

            s = s.replace(c, '')

    return s

def _getpreferredencoding():

    '''

    On darwin, getpreferredencoding ignores the locale environment and

    always returns mac-roman. http://bugs.python.org/issue6202 fixes this

    for Python 2.7 and up. This is the same corrected code for earlier

    Python versions.

    However, we can't use a version check for this method, as some distributions

    patch Python to fix this. Instead, we use it as a 'fixer' for the mac-roman

    encoding, as it is unlikely that this encoding is the actually expected.

    '''

    try:

        locale.CODESET

    except AttributeError:

        # Fall back to parsing environment variables :-(

        return locale.getdefaultlocale()[1]

    oldloc = locale.setlocale(locale.LC_CTYPE)

    locale.setlocale(locale.LC_CTYPE, "")

    result = locale.nl_langinfo(locale.CODESET)

    locale.setlocale(locale.LC_CTYPE, oldloc)

    return result

_encodingfixers = {

    '646': lambda: 'ascii',

    'ANSI_X3.4-1968': lambda: 'ascii',

    'mac-roman': _getpreferredencoding

}

try:

    encoding = os.environ.get("HGENCODING")

    if not encoding:

        encoding = locale.getpreferredencoding() or 'ascii'

        encoding = _encodingfixers.get(encoding, lambda: encoding)()

except locale.Error:

    encoding = 'ascii'

encodingmode = os.environ.get("HGENCODINGMODE", "strict")

fallbackencoding = 'ISO-8859-1'

class localstr(str):

    '''This class allows strings that are unmodified to be

    round-tripped to the local encoding and back'''

    def __new__(cls, u, l):

        s = str.__new__(cls, l)

        s._utf8 = u

        return s

    def __hash__(self):

        return hash(self._utf8) # avoid collisions in local string space

def tolocal(s):

    """

    Convert a string from internal UTF-8 to local encoding

    All internal strings should be UTF-8 but some repos before the

    implementation of locale support may contain latin1 or possibly

    other character sets. We attempt to decode everything strictly

    using UTF-8, then Latin-1, and failing that, we use UTF-8 and

    replace unknown characters.

    The localstr class is used to cache the known UTF-8 encoding of

    strings next to their local representation to allow lossless

    round-trip conversion back to UTF-8.

    >>> u = 'foo: \\xc3\\xa4' # utf-8

    >>> l = tolocal(u)

    >>> l

    'foo: ?'

    >>> fromlocal(l)

    'foo: \\xc3\\xa4'

    >>> u2 = 'foo: \\xc3\\xa1'

    >>> d = { l: 1, tolocal(u2): 2 }

    >>> len(d) # no collision

    2

    >>> 'foo: ?' in d

    False

    >>> l1 = 'foo: \\xe4' # historical latin1 fallback

    >>> l = tolocal(l1)

    >>> l

    'foo: ?'

    >>> fromlocal(l) # magically in utf-8

    'foo: \\xc3\\xa4'

    """

    try:

        try:

            # make sure string is actually stored in UTF-8

            u = s.decode('UTF-8')

            if encoding == 'UTF-8':

                # fast path

                return s

            r = u.encode(encoding, "replace")

            if u == r.decode(encoding):

                # r is a safe, non-lossy encoding of s

                return r

            return localstr(s, r)

        except UnicodeDecodeError:

            # we should only get here if we're looking at an ancient changeset

            try:

                u = s.decode(fallbackencoding)

                r = u.encode(encoding, "replace")

                if u == r.decode(encoding):

                    # r is a safe, non-lossy encoding of s

                    return r

                return localstr(u.encode('UTF-8'), r)

            except UnicodeDecodeError:

                u = s.decode("utf-8", "replace") # last ditch

                return u.encode(encoding, "replace") # can't round-trip

    except LookupError as k:

        raise error.Abort(k, hint="please check your locale settings")

def fromlocal(s):

    """

    Convert a string from the local character encoding to UTF-8

    We attempt to decode strings using the encoding mode set by

    HGENCODINGMODE, which defaults to 'strict'. In this mode, unknown

    characters will cause an error message. Other modes include

    'replace', which replaces unknown characters with a special

    Unicode character, and 'ignore', which drops the character.

    """

    # can we do a lossless round-trip?

    if isinstance(s, localstr):

        return s._utf8

    try:

        return s.decode(encoding, encodingmode).encode("utf-8")

    except UnicodeDecodeError as inst:

        sub = s[max(0, inst.start - 10):inst.start + 10]

        raise error.Abort("decoding near '%s': %s!" % (sub, inst))

    except LookupError as k:

        raise error.Abort(k, hint="please check your locale settings")

# How to treat ambiguous-width characters. Set to 'wide' to treat as wide.

wide = (os.environ.get("HGENCODINGAMBIGUOUS", "narrow") == "wide"

        and "WFA" or "WF")

def colwidth(s):

    "Find the column width of a string for display in the local encoding"

    return ucolwidth(s.decode(encoding, 'replace'))

def ucolwidth(d):

    "Find the column width of a Unicode string for display"

    eaw = getattr(unicodedata, 'east_asian_width', None)

    if eaw is not None:

        return sum([eaw(c) in wide and 2 or 1 for c in d])

    return len(d)

def getcols(s, start, c):

    '''Use colwidth to find a c-column substring of s starting at byte

    index start'''

    for x in xrange(start + c, len(s)):

        t = s[start:x]

        if colwidth(t) == c:

            return t

def trim(s, width, ellipsis='', leftside=False):

    """Trim string 's' to at most 'width' columns (including 'ellipsis').

    If 'leftside' is True, left side of string 's' is trimmed.

    'ellipsis' is always placed at trimmed side.

    >>> ellipsis = '+++'

    >>> from . import encoding

    >>> encoding.encoding = 'utf-8'

    >>> t= '1234567890'

    >>> print trim(t, 12, ellipsis=ellipsis)

    1234567890

    >>> print trim(t, 10, ellipsis=ellipsis)

    1234567890

    >>> print trim(t, 8, ellipsis=ellipsis)

    12345+++

    >>> print trim(t, 8, ellipsis=ellipsis, leftside=True)

    +++67890

    >>> print trim(t, 8)

    12345678

    >>> print trim(t, 8, leftside=True)

    34567890

    >>> print trim(t, 3, ellipsis=ellipsis)

    +++

    >>> print trim(t, 1, ellipsis=ellipsis)

    +

    >>> u = u'\u3042\u3044\u3046\u3048\u304a' # 2 x 5 = 10 columns

    >>> t = u.encode(encoding.encoding)

    >>> print trim(t, 12, ellipsis=ellipsis)

    \xe3\x81\x82\xe3\x81\x84\xe3\x81\x86\xe3\x81\x88\xe3\x81\x8a

    >>> print trim(t, 10, ellipsis=ellipsis)

    \xe3\x81\x82\xe3\x81\x84\xe3\x81\x86\xe3\x81\x88\xe3\x81\x8a

    >>> print trim(t, 8, ellipsis=ellipsis)

    \xe3\x81\x82\xe3\x81\x84+++

    >>> print trim(t, 8, ellipsis=ellipsis, leftside=True)

    +++\xe3\x81\x88\xe3\x81\x8a

    >>> print trim(t, 5)

    \xe3\x81\x82\xe3\x81\x84

    >>> print trim(t, 5, leftside=True)

    \xe3\x81\x88\xe3\x81\x8a

    >>> print trim(t, 4, ellipsis=ellipsis)

    +++

    >>> print trim(t, 4, ellipsis=ellipsis, leftside=True)

    +++

    >>> t = '\x11\x22\x33\x44\x55\x66\x77\x88\x99\xaa' # invalid byte sequence

    >>> print trim(t, 12, ellipsis=ellipsis)

    \x11\x22\x33\x44\x55\x66\x77\x88\x99\xaa

    >>> print trim(t, 10, ellipsis=ellipsis)

    \x11\x22\x33\x44\x55\x66\x77\x88\x99\xaa

    >>> print trim(t, 8, ellipsis=ellipsis)

    \x11\x22\x33\x44\x55+++

    >>> print trim(t, 8, ellipsis=ellipsis, leftside=True)

    +++\x66\x77\x88\x99\xaa

    >>> print trim(t, 8)

    \x11\x22\x33\x44\x55\x66\x77\x88

    >>> print trim(t, 8, leftside=True)

    \x33\x44\x55\x66\x77\x88\x99\xaa

    >>> print trim(t, 3, ellipsis=ellipsis)

    +++

    >>> print trim(t, 1, ellipsis=ellipsis)

    +

    """

    try:

        u = s.decode(encoding)

    except UnicodeDecodeError:

        if len(s) <= width: # trimming is not needed

            return s

        width -= len(ellipsis)

        if width <= 0: # no enough room even for ellipsis

            return ellipsis[:width + len(ellipsis)]

        if leftside:

            return ellipsis + s[-width:]

        return s[:width] + ellipsis

    if ucolwidth(u) <= width: # trimming is not needed

        return s

    width -= len(ellipsis)

    if width <= 0: # no enough room even for ellipsis

        return ellipsis[:width + len(ellipsis)]

    if leftside:

        uslice = lambda i: u[i:]

        concat = lambda s: ellipsis + s

    else:

        uslice = lambda i: u[:-i]

        concat = lambda s: s + ellipsis

    for i in xrange(1, len(u)):

        usub = uslice(i)

        if ucolwidth(usub) <= width:

            return concat(usub.encode(encoding))

    return ellipsis # no enough room for multi-column characters

def _asciilower(s):

    '''convert a string to lowercase if ASCII

    Raises UnicodeDecodeError if non-ASCII characters are found.'''

    s.decode('ascii')

    return s.lower()

def asciilower(s):

    # delay importing avoids cyclic dependency around "parsers" in

    # pure Python build (util => i18n => encoding => parsers => util)

    from . import parsers

    impl = getattr(parsers, 'asciilower', _asciilower)

    global asciilower

    asciilower = impl

    return impl(s)

def _asciiupper(s):

    '''convert a string to uppercase if ASCII

    Raises UnicodeDecodeError if non-ASCII characters are found.'''

    s.decode('ascii')

    return s.upper()

def asciiupper(s):

    # delay importing avoids cyclic dependency around "parsers" in

    # pure Python build (util => i18n => encoding => parsers => util)

    from . import parsers

    impl = getattr(parsers, 'asciiupper', _asciiupper)

    global asciiupper

    asciiupper = impl

    return impl(s)

def lower(s):

    "best-effort encoding-aware case-folding of local string s"

    try:

        return asciilower(s)

    except UnicodeDecodeError:

        pass

    try:

        if isinstance(s, localstr):

            u = s._utf8.decode("utf-8")

        else:

            u = s.decode(encoding, encodingmode)

        lu = u.lower()

        if u == lu:

            return s # preserve localstring

        return lu.encode(encoding)

    except UnicodeError:

        return s.lower() # we don't know how to fold this except in ASCII

    except LookupError as k:

        raise error.Abort(k, hint="please check your locale settings")

def upper(s):

    "best-effort encoding-aware case-folding of local string s"

    try:

        return asciiupper(s)

    except UnicodeDecodeError:

        return upperfallback(s)

def upperfallback(s):

    try:

        if isinstance(s, localstr):

            u = s._utf8.decode("utf-8")

        else:

            u = s.decode(encoding, encodingmode)

        uu = u.upper()

        if u == uu:

            return s # preserve localstring

        return uu.encode(encoding)

    except UnicodeError:

        return s.upper() # we don't know how to fold this except in ASCII

    except LookupError as k:

        raise error.Abort(k, hint="please check your locale settings")

class normcasespecs(object):

    '''what a platform's normcase does to ASCII strings

    This is specified per platform, and should be consistent with what normcase

    on that platform actually does.

    lower: normcase lowercases ASCII strings

    upper: normcase uppercases ASCII strings

    other: the fallback function should always be called

    This should be kept in sync with normcase_spec in util.h.'''

    lower = -1

    upper = 1

    other = 0

_jsonmap = []

_jsonmap.extend("\\u%04x" % x for x in range(32))

_jsonmap.extend(chr(x) for x in range(32, 127))

_jsonmap.append('\\u007f')

_jsonmap[0x09] = '\\t'

_jsonmap[0x0a] = '\\n'

_jsonmap[0x22] = '\\"'

_jsonmap[0x5c] = '\\\\'

_jsonmap[0x08] = '\\b'

_jsonmap[0x0c] = '\\f'

_jsonmap[0x0d] = '\\r'

_paranoidjsonmap = _jsonmap[:]

_paranoidjsonmap[0x3c] = '\\u003c'  # '<' (e.g. escape "</script>")