AppPkg/Applications/Python/Python-2.7.10: Initial Checkin part 4/5.

[mirror_edk2.git] / AppPkg / Applications / Python / Python-2.7.10 / Lib / json / decoder.py

"""Implementation of JSONDecoder\r
"""\r
import re\r
import sys\r
import struct\r
\r
from json import scanner\r
try:\r
    from _json import scanstring as c_scanstring\r
except ImportError:\r
    c_scanstring = None\r
\r
__all__ = ['JSONDecoder']\r
\r
FLAGS = re.VERBOSE | re.MULTILINE | re.DOTALL\r
\r
def _floatconstants():\r
    _BYTES = '7FF80000000000007FF0000000000000'.decode('hex')\r
    if sys.byteorder != 'big':\r
        _BYTES = _BYTES[:8][::-1] + _BYTES[8:][::-1]\r
    nan, inf = struct.unpack('dd', _BYTES)\r
    return nan, inf, -inf\r
\r
NaN, PosInf, NegInf = _floatconstants()\r
\r
\r
def linecol(doc, pos):\r
    lineno = doc.count('\n', 0, pos) + 1\r
    if lineno == 1:\r
        colno = pos + 1\r
    else:\r
        colno = pos - doc.rindex('\n', 0, pos)\r
    return lineno, colno\r
\r
\r
def errmsg(msg, doc, pos, end=None):\r
    # Note that this function is called from _json\r
    lineno, colno = linecol(doc, pos)\r
    if end is None:\r
        fmt = '{0}: line {1} column {2} (char {3})'\r
        return fmt.format(msg, lineno, colno, pos)\r
        #fmt = '%s: line %d column %d (char %d)'\r
        #return fmt % (msg, lineno, colno, pos)\r
    endlineno, endcolno = linecol(doc, end)\r
    fmt = '{0}: line {1} column {2} - line {3} column {4} (char {5} - {6})'\r
    return fmt.format(msg, lineno, colno, endlineno, endcolno, pos, end)\r
    #fmt = '%s: line %d column %d - line %d column %d (char %d - %d)'\r
    #return fmt % (msg, lineno, colno, endlineno, endcolno, pos, end)\r
\r
\r
_CONSTANTS = {\r
    '-Infinity': NegInf,\r
    'Infinity': PosInf,\r
    'NaN': NaN,\r
}\r
\r
STRINGCHUNK = re.compile(r'(.*?)(["\\\x00-\x1f])', FLAGS)\r
BACKSLASH = {\r
    '"': u'"', '\\': u'\\', '/': u'/',\r
    'b': u'\b', 'f': u'\f', 'n': u'\n', 'r': u'\r', 't': u'\t',\r
}\r
\r
DEFAULT_ENCODING = "utf-8"\r
\r
def _decode_uXXXX(s, pos):\r
    esc = s[pos + 1:pos + 5]\r
    if len(esc) == 4 and esc[1] not in 'xX':\r
        try:\r
            return int(esc, 16)\r
        except ValueError:\r
            pass\r
    msg = "Invalid \\uXXXX escape"\r
    raise ValueError(errmsg(msg, s, pos))\r
\r
def py_scanstring(s, end, encoding=None, strict=True,\r
        _b=BACKSLASH, _m=STRINGCHUNK.match):\r
    """Scan the string s for a JSON string. End is the index of the\r
    character in s after the quote that started the JSON string.\r
    Unescapes all valid JSON string escape sequences and raises ValueError\r
    on attempt to decode an invalid string. If strict is False then literal\r
    control characters are allowed in the string.\r
\r
    Returns a tuple of the decoded string and the index of the character in s\r
    after the end quote."""\r
    if encoding is None:\r
        encoding = DEFAULT_ENCODING\r
    chunks = []\r
    _append = chunks.append\r
    begin = end - 1\r
    while 1:\r
        chunk = _m(s, end)\r
        if chunk is None:\r
            raise ValueError(\r
                errmsg("Unterminated string starting at", s, begin))\r
        end = chunk.end()\r
        content, terminator = chunk.groups()\r
        # Content is contains zero or more unescaped string characters\r
        if content:\r
            if not isinstance(content, unicode):\r
                content = unicode(content, encoding)\r
            _append(content)\r
        # Terminator is the end of string, a literal control character,\r
        # or a backslash denoting that an escape sequence follows\r
        if terminator == '"':\r
            break\r
        elif terminator != '\\':\r
            if strict:\r
                #msg = "Invalid control character %r at" % (terminator,)\r
                msg = "Invalid control character {0!r} at".format(terminator)\r
                raise ValueError(errmsg(msg, s, end))\r
            else:\r
                _append(terminator)\r
                continue\r
        try:\r
            esc = s[end]\r
        except IndexError:\r
            raise ValueError(\r
                errmsg("Unterminated string starting at", s, begin))\r
        # If not a unicode escape sequence, must be in the lookup table\r
        if esc != 'u':\r
            try:\r
                char = _b[esc]\r
            except KeyError:\r
                msg = "Invalid \\escape: " + repr(esc)\r
                raise ValueError(errmsg(msg, s, end))\r
            end += 1\r
        else:\r
            # Unicode escape sequence\r
            uni = _decode_uXXXX(s, end)\r
            end += 5\r
            # Check for surrogate pair on UCS-4 systems\r
            if sys.maxunicode > 65535 and \\r
               0xd800 <= uni <= 0xdbff and s[end:end + 2] == '\\u':\r
                uni2 = _decode_uXXXX(s, end + 1)\r
                if 0xdc00 <= uni2 <= 0xdfff:\r
                    uni = 0x10000 + (((uni - 0xd800) << 10) | (uni2 - 0xdc00))\r
                    end += 6\r
            char = unichr(uni)\r
        # Append the unescaped character\r
        _append(char)\r
    return u''.join(chunks), end\r
\r
\r
# Use speedup if available\r
scanstring = c_scanstring or py_scanstring\r
\r
WHITESPACE = re.compile(r'[ \t\n\r]*', FLAGS)\r
WHITESPACE_STR = ' \t\n\r'\r
\r
def JSONObject(s_and_end, encoding, strict, scan_once, object_hook,\r
               object_pairs_hook, _w=WHITESPACE.match, _ws=WHITESPACE_STR):\r
    s, end = s_and_end\r
    pairs = []\r
    pairs_append = pairs.append\r
    # Use a slice to prevent IndexError from being raised, the following\r
    # check will raise a more specific ValueError if the string is empty\r
    nextchar = s[end:end + 1]\r
    # Normally we expect nextchar == '"'\r
    if nextchar != '"':\r
        if nextchar in _ws:\r
            end = _w(s, end).end()\r
            nextchar = s[end:end + 1]\r
        # Trivial empty object\r
        if nextchar == '}':\r
            if object_pairs_hook is not None:\r
                result = object_pairs_hook(pairs)\r
                return result, end + 1\r
            pairs = {}\r
            if object_hook is not None:\r
                pairs = object_hook(pairs)\r
            return pairs, end + 1\r
        elif nextchar != '"':\r
            raise ValueError(errmsg(\r
                "Expecting property name enclosed in double quotes", s, end))\r
    end += 1\r
    while True:\r
        key, end = scanstring(s, end, encoding, strict)\r
\r
        # To skip some function call overhead we optimize the fast paths where\r
        # the JSON key separator is ": " or just ":".\r
        if s[end:end + 1] != ':':\r
            end = _w(s, end).end()\r
            if s[end:end + 1] != ':':\r
                raise ValueError(errmsg("Expecting ':' delimiter", s, end))\r
        end += 1\r
\r
        try:\r
            if s[end] in _ws:\r
                end += 1\r
                if s[end] in _ws:\r
                    end = _w(s, end + 1).end()\r
        except IndexError:\r
            pass\r
\r
        try:\r
            value, end = scan_once(s, end)\r
        except StopIteration:\r
            raise ValueError(errmsg("Expecting object", s, end))\r
        pairs_append((key, value))\r
\r
        try:\r
            nextchar = s[end]\r
            if nextchar in _ws:\r
                end = _w(s, end + 1).end()\r
                nextchar = s[end]\r
        except IndexError:\r
            nextchar = ''\r
        end += 1\r
\r
        if nextchar == '}':\r
            break\r
        elif nextchar != ',':\r
            raise ValueError(errmsg("Expecting ',' delimiter", s, end - 1))\r
\r
        try:\r
            nextchar = s[end]\r
            if nextchar in _ws:\r
                end += 1\r
                nextchar = s[end]\r
                if nextchar in _ws:\r
                    end = _w(s, end + 1).end()\r
                    nextchar = s[end]\r
        except IndexError:\r
            nextchar = ''\r
\r
        end += 1\r
        if nextchar != '"':\r
            raise ValueError(errmsg(\r
                "Expecting property name enclosed in double quotes", s, end - 1))\r
    if object_pairs_hook is not None:\r
        result = object_pairs_hook(pairs)\r
        return result, end\r
    pairs = dict(pairs)\r
    if object_hook is not None:\r
        pairs = object_hook(pairs)\r
    return pairs, end\r
\r
def JSONArray(s_and_end, scan_once, _w=WHITESPACE.match, _ws=WHITESPACE_STR):\r
    s, end = s_and_end\r
    values = []\r
    nextchar = s[end:end + 1]\r
    if nextchar in _ws:\r
        end = _w(s, end + 1).end()\r
        nextchar = s[end:end + 1]\r
    # Look-ahead for trivial empty array\r
    if nextchar == ']':\r
        return values, end + 1\r
    _append = values.append\r
    while True:\r
        try:\r
            value, end = scan_once(s, end)\r
        except StopIteration:\r
            raise ValueError(errmsg("Expecting object", s, end))\r
        _append(value)\r
        nextchar = s[end:end + 1]\r
        if nextchar in _ws:\r
            end = _w(s, end + 1).end()\r
            nextchar = s[end:end + 1]\r
        end += 1\r
        if nextchar == ']':\r
            break\r
        elif nextchar != ',':\r
            raise ValueError(errmsg("Expecting ',' delimiter", s, end))\r
        try:\r
            if s[end] in _ws:\r
                end += 1\r
                if s[end] in _ws:\r
                    end = _w(s, end + 1).end()\r
        except IndexError:\r
            pass\r
\r
    return values, end\r
\r
class JSONDecoder(object):\r
    """Simple JSON <http://json.org> decoder\r
\r
    Performs the following translations in decoding by default:\r
\r
    +---------------+-------------------+\r
    | JSON          | Python            |\r
    +===============+===================+\r
    | object        | dict              |\r
    +---------------+-------------------+\r
    | array         | list              |\r
    +---------------+-------------------+\r
    | string        | unicode           |\r
    +---------------+-------------------+\r
    | number (int)  | int, long         |\r
    +---------------+-------------------+\r
    | number (real) | float             |\r
    +---------------+-------------------+\r
    | true          | True              |\r
    +---------------+-------------------+\r
    | false         | False             |\r
    +---------------+-------------------+\r
    | null          | None              |\r
    +---------------+-------------------+\r
\r
    It also understands ``NaN``, ``Infinity``, and ``-Infinity`` as\r
    their corresponding ``float`` values, which is outside the JSON spec.\r
\r
    """\r
\r
    def __init__(self, encoding=None, object_hook=None, parse_float=None,\r
            parse_int=None, parse_constant=None, strict=True,\r
            object_pairs_hook=None):\r
        """``encoding`` determines the encoding used to interpret any ``str``\r
        objects decoded by this instance (utf-8 by default).  It has no\r
        effect when decoding ``unicode`` objects.\r
\r
        Note that currently only encodings that are a superset of ASCII work,\r
        strings of other encodings should be passed in as ``unicode``.\r
\r
        ``object_hook``, if specified, will be called with the result\r
        of every JSON object decoded and its return value will be used in\r
        place of the given ``dict``.  This can be used to provide custom\r
        deserializations (e.g. to support JSON-RPC class hinting).\r
\r
        ``object_pairs_hook``, if specified will be called with the result of\r
        every JSON object decoded with an ordered list of pairs.  The return\r
        value of ``object_pairs_hook`` will be used instead of the ``dict``.\r
        This feature can be used to implement custom decoders that rely on the\r
        order that the key and value pairs are decoded (for example,\r
        collections.OrderedDict will remember the order of insertion). If\r
        ``object_hook`` is also defined, the ``object_pairs_hook`` takes\r
        priority.\r
\r
        ``parse_float``, if specified, will be called with the string\r
        of every JSON float to be decoded. By default this is equivalent to\r
        float(num_str). This can be used to use another datatype or parser\r
        for JSON floats (e.g. decimal.Decimal).\r
\r
        ``parse_int``, if specified, will be called with the string\r
        of every JSON int to be decoded. By default this is equivalent to\r
        int(num_str). This can be used to use another datatype or parser\r
        for JSON integers (e.g. float).\r
\r
        ``parse_constant``, if specified, will be called with one of the\r
        following strings: -Infinity, Infinity, NaN.\r
        This can be used to raise an exception if invalid JSON numbers\r
        are encountered.\r
\r
        If ``strict`` is false (true is the default), then control\r
        characters will be allowed inside strings.  Control characters in\r
        this context are those with character codes in the 0-31 range,\r
        including ``'\\t'`` (tab), ``'\\n'``, ``'\\r'`` and ``'\\0'``.\r
\r
        """\r
        self.encoding = encoding\r
        self.object_hook = object_hook\r
        self.object_pairs_hook = object_pairs_hook\r
        self.parse_float = parse_float or float\r
        self.parse_int = parse_int or int\r
        self.parse_constant = parse_constant or _CONSTANTS.__getitem__\r
        self.strict = strict\r
        self.parse_object = JSONObject\r
        self.parse_array = JSONArray\r
        self.parse_string = scanstring\r
        self.scan_once = scanner.make_scanner(self)\r
\r
    def decode(self, s, _w=WHITESPACE.match):\r
        """Return the Python representation of ``s`` (a ``str`` or ``unicode``\r
        instance containing a JSON document)\r
\r
        """\r
        obj, end = self.raw_decode(s, idx=_w(s, 0).end())\r
        end = _w(s, end).end()\r
        if end != len(s):\r
            raise ValueError(errmsg("Extra data", s, end, len(s)))\r
        return obj\r
\r
    def raw_decode(self, s, idx=0):\r
        """Decode a JSON document from ``s`` (a ``str`` or ``unicode``\r
        beginning with a JSON document) and return a 2-tuple of the Python\r
        representation and the index in ``s`` where the document ended.\r
\r
        This can be used to decode a JSON document from a string that may\r
        have extraneous data at the end.\r
\r
        """\r
        try:\r
            obj, end = self.scan_once(s, idx)\r
        except StopIteration:\r
            raise ValueError("No JSON object could be decoded")\r
        return obj, end\r