\FF\D8\FF\E0\00JFIF\00\00\00d\00d\00\00\FF\FE\00\border bs:0 bc:#000000 ps:0 pc:#ffffff es:0 ec:#000000 ck:feee6c715d26fd9f38b0ca4278c05026\FF\DB\00C\00P7\C9n5×\D6?\BDê\9Ds\EBp\9F[`8m\B7)o\B5\E8\E6I\99\FE3]]A2\BA\8Cw\D6E\93\\DEv\C8\009\F2\F1NI?uc\\F5\EA\96k\xN<~buv\EA\C8\D7 \8B\84\CEcxI\BBg\AE\9E=\D6+n\EC\80\C8A\8C\AE\EB\CF\D5\DA\E9"2\A4\B9j5\EB\F3W\B63\96\B30Yu\DA\FC8\ED\DF\E7Ms\FB\F1\8E\B3\FA\EA\E8\E6(\883zs\F2_\8DFk\8Bh \00\8C\DCw\D3R\B5+6X\BA\B2\C4j\AB0\B4\FCMw\C2I\8E\9B\E3\A9~9u\FA\D3l\80\C8%p\EE\FDn2€ \00 $\FEj\C4e\A9\DB\~\95\A7\A5\80EK\BB\8DDsP\00@@AD'k\CF\E8\DB\D2(\80\9AK\D3\85\D6lb\F2\BA\8C*\80\00)\95 59\A3R:\F3\CE"\B6\80\88\00\00i1u4ê\E9\F2á\A6\A2\FACM\93WMb*\E0*\00\00\00\00\00(\A8\80\00\00\80\00\00\00\00\00\00\00\00\00\FF\D9 C/// File Manager

File Manager

Path: /opt/cloudlinux/venv/lib/python3.11/site-packages/future/utils/

Viewing File: surrogateescape.py

"""
This is Victor Stinner's pure-Python implementation of PEP 383: the "surrogateescape" error
handler of Python 3.

Source: misc/python/surrogateescape.py in https://bitbucket.org/haypo/misc
"""

# This code is released under the Python license and the BSD 2-clause license

import codecs
import sys

from future import utils


FS_ERRORS = 'surrogateescape'

#     # -- Python 2/3 compatibility -------------------------------------
#     FS_ERRORS = 'my_surrogateescape'

def u(text):
    if utils.PY3:
        return text
    else:
        return text.decode('unicode_escape')

def b(data):
    if utils.PY3:
        return data.encode('latin1')
    else:
        return data

if utils.PY3:
    _unichr = chr
    bytes_chr = lambda code: bytes((code,))
else:
    _unichr = unichr
    bytes_chr = chr

def surrogateescape_handler(exc):
    """
    Pure Python implementation of the PEP 383: the "surrogateescape" error
    handler of Python 3. Undecodable bytes will be replaced by a Unicode
    character U+DCxx on decoding, and these are translated into the
    original bytes on encoding.
    """
    mystring = exc.object[exc.start:exc.end]

    try:
        if isinstance(exc, UnicodeDecodeError):
            # mystring is a byte-string in this case
            decoded = replace_surrogate_decode(mystring)
        elif isinstance(exc, UnicodeEncodeError):
            # In the case of u'\udcc3'.encode('ascii',
            # 'this_surrogateescape_handler'), both Python 2.x and 3.x raise an
            # exception anyway after this function is called, even though I think
            # it's doing what it should. It seems that the strict encoder is called
            # to encode the unicode string that this function returns ...
            decoded = replace_surrogate_encode(mystring)
        else:
            raise exc
    except NotASurrogateError:
        raise exc
    return (decoded, exc.end)


class NotASurrogateError(Exception):
    pass


def replace_surrogate_encode(mystring):
    """
    Returns a (unicode) string, not the more logical bytes, because the codecs
    register_error functionality expects this.
    """
    decoded = []
    for ch in mystring:
        # if utils.PY3:
        #     code = ch
        # else:
        code = ord(ch)

        # The following magic comes from Py3.3's Python/codecs.c file:
        if not 0xD800 <= code <= 0xDCFF:
            # Not a surrogate. Fail with the original exception.
            raise NotASurrogateError
        # mybytes = [0xe0 | (code >> 12),
        #            0x80 | ((code >> 6) & 0x3f),
        #            0x80 | (code & 0x3f)]
        # Is this a good idea?
        if 0xDC00 <= code <= 0xDC7F:
            decoded.append(_unichr(code - 0xDC00))
        elif code <= 0xDCFF:
            decoded.append(_unichr(code - 0xDC00))
        else:
            raise NotASurrogateError
    return str().join(decoded)


def replace_surrogate_decode(mybytes):
    """
    Returns a (unicode) string
    """
    decoded = []
    for ch in mybytes:
        # We may be parsing newbytes (in which case ch is an int) or a native
        # str on Py2
        if isinstance(ch, int):
            code = ch
        else:
            code = ord(ch)
        if 0x80 <= code <= 0xFF:
            decoded.append(_unichr(0xDC00 + code))
        elif code <= 0x7F:
            decoded.append(_unichr(code))
        else:
            # # It may be a bad byte
            # # Try swallowing it.
            # continue
            # print("RAISE!")
            raise NotASurrogateError
    return str().join(decoded)


def encodefilename(fn):
    if FS_ENCODING == 'ascii':
        # ASCII encoder of Python 2 expects that the error handler returns a
        # Unicode string encodable to ASCII, whereas our surrogateescape error
        # handler has to return bytes in 0x80-0xFF range.
        encoded = []
        for index, ch in enumerate(fn):
            code = ord(ch)
            if code < 128:
                ch = bytes_chr(code)
            elif 0xDC80 <= code <= 0xDCFF:
                ch = bytes_chr(code - 0xDC00)
            else:
                raise UnicodeEncodeError(FS_ENCODING,
                    fn, index, index+1,
                    'ordinal not in range(128)')
            encoded.append(ch)
        return bytes().join(encoded)
    elif FS_ENCODING == 'utf-8':
        # UTF-8 encoder of Python 2 encodes surrogates, so U+DC80-U+DCFF
        # doesn't go through our error handler
        encoded = []
        for index, ch in enumerate(fn):
            code = ord(ch)
            if 0xD800 <= code <= 0xDFFF:
                if 0xDC80 <= code <= 0xDCFF:
                    ch = bytes_chr(code - 0xDC00)
                    encoded.append(ch)
                else:
                    raise UnicodeEncodeError(
                        FS_ENCODING,
                        fn, index, index+1, 'surrogates not allowed')
            else:
                ch_utf8 = ch.encode('utf-8')
                encoded.append(ch_utf8)
        return bytes().join(encoded)
    else:
        return fn.encode(FS_ENCODING, FS_ERRORS)

def decodefilename(fn):
    return fn.decode(FS_ENCODING, FS_ERRORS)

FS_ENCODING = 'ascii'; fn = b('[abc\xff]'); encoded = u('[abc\udcff]')
# FS_ENCODING = 'cp932'; fn = b('[abc\x81\x00]'); encoded = u('[abc\udc81\x00]')
# FS_ENCODING = 'UTF-8'; fn = b('[abc\xff]'); encoded = u('[abc\udcff]')


# normalize the filesystem encoding name.
# For example, we expect "utf-8", not "UTF8".
FS_ENCODING = codecs.lookup(FS_ENCODING).name


def register_surrogateescape():
    """
    Registers the surrogateescape error handler on Python 2 (only)
    """
    if utils.PY3:
        return
    try:
        codecs.lookup_error(FS_ERRORS)
    except LookupError:
        codecs.register_error(FS_ERRORS, surrogateescape_handler)


if __name__ == '__main__':
    pass
    # # Tests:
    # register_surrogateescape()

    # b = decodefilename(fn)
    # assert b == encoded, "%r != %r" % (b, encoded)
    # c = encodefilename(b)
    # assert c == fn, '%r != %r' % (c, fn)
    # # print("ok")