# 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 io

import locale

import os

import unicodedata

from . import (

    error,

    policy,

    pycompat,

)

from .pure import (

    charencode as charencodepure,

)

charencode = policy.importmod(r'charencode')

isasciistr = charencode.isasciistr

asciilower = charencode.asciilower

asciiupper = charencode.asciiupper

_jsonescapeu8fast = charencode.jsonescapeu8fast

_sysstr = pycompat.sysstr

if pycompat.ispy3:

    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

assert all(i.startswith(("\xe2", "\xef")) for i in _ignore)

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

# encoding.environ is provided read-only, which may not be used to modify

# the process environment

_nativeenviron = (not pycompat.ispy3 or os.supports_bytes_environ)

if not pycompat.ispy3:

    environ = os.environ  # re-exports

elif _nativeenviron:

    environ = os.environb  # re-exports

else:

    # preferred encoding isn't known yet; use utf-8 to avoid unicode error

    # and recreate it once encoding is settled

    environ = dict((k.encode(u'utf-8'), v.encode(u'utf-8'))

                   for k, v in os.environ.items())  # re-exports

_encodingfixers = {

    '646': lambda: 'ascii',

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

}

try:

    encoding = environ.get("HGENCODING")

    if not encoding:

        encoding = locale.getpreferredencoding().encode('ascii') or 'ascii'

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

except locale.Error:

    encoding = 'ascii'

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

fallbackencoding = 'ISO-8859-1'

class localstr(bytes):

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

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

    def __new__(cls, u, l):

        s = bytes.__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 = b'foo: \\xc3\\xa4' # utf-8

    >>> l = tolocal(u)

    >>> l

    'foo: ?'

    >>> fromlocal(l)

    'foo: \\xc3\\xa4'

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

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

    >>> len(d) # no collision

    2

    >>> b'foo: ?' in d

    False

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

    >>> l = tolocal(l1)

    >>> l

    'foo: ?'

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

    'foo: \\xc3\\xa4'

    """

    if isasciistr(s):

        return s

    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(_sysstr(encoding), u"replace")

            if u == r.decode(_sysstr(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(_sysstr(fallbackencoding))

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

                if u == r.decode(_sysstr(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

                # can't round-trip

                return u.encode(_sysstr(encoding), u"replace")

    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

    if isasciistr(s):

        return s

    try:

        u = s.decode(_sysstr(encoding), _sysstr(encodingmode))

        return u.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")

def unitolocal(u):

    """Convert a unicode string to a byte string of local encoding"""

    return tolocal(u.encode('utf-8'))

def unifromlocal(s):

    """Convert a byte string of local encoding to a unicode string"""

    return fromlocal(s).decode('utf-8')

def unimethod(bytesfunc):

    """Create a proxy method that forwards __unicode__() and __str__() of

    Python 3 to __bytes__()"""

    def unifunc(obj):

        return unifromlocal(bytesfunc(obj))

    return unifunc

# converter functions between native str and byte string. use these if the

# character encoding is not aware (e.g. exception message) or is known to

# be locale dependent (e.g. date formatting.)

if pycompat.ispy3:

    strtolocal = unitolocal

    strfromlocal = unifromlocal

    strmethod = unimethod

else:

    strtolocal = pycompat.identity

    strfromlocal = pycompat.identity

    strmethod = pycompat.identity

if not _nativeenviron:

    # now encoding and helper functions are available, recreate the environ

    # dict to be exported to other modules

    environ = dict((tolocal(k.encode(u'utf-8')), tolocal(v.encode(u'utf-8')))

                   for k, v in os.environ.items())  # re-exports

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

_wide = _sysstr(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(_sysstr(encoding), u'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 = b'+++'

    >>> from . import encoding

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

    >>> t = b'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(pycompat.sysstr(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)

    +++

    >>> 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(_sysstr(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(_sysstr(encoding)))

    return ellipsis # no enough room for multi-column characters

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(_sysstr(encoding), _sysstr(encodingmode))

        lu = u.lower()

        if u == lu:

            return s # preserve localstring

        return lu.encode(_sysstr(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(_sysstr(encoding), _sysstr(encodingmode))

        uu = u.upper()

        if u == uu:

            return s # preserve localstring

        return uu.encode(_sysstr(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

def jsonescape(s, paranoid=False):

    '''returns a string suitable for JSON