port simplejson upgrade from the trunk #4136

json also now works only with unicode strings

Patch by Antoine Pitrou; updated by me

12 files changed, 555 insertions, 529 deletions

diff --git a/Lib/json/__init__.py b/Lib/json/__init__.py
index 1ccde87..6a18518 100644
--- a/Lib/json/__init__.py
+++ b/Lib/json/__init__.py
@@ -1,11 +1,13 @@
-r"""A simple, fast, extensible JSON encoder and decoder
-
-JSON (JavaScript Object Notation) <http://json.org> is a subset of
+r"""JSON (JavaScript Object Notation) <http://json.org> is a subset of
 JavaScript syntax (ECMA-262 3rd edition) used as a lightweight data
 interchange format.
 
-json exposes an API familiar to uses of the standard library
-marshal and pickle modules.
+:mod:`json` exposes an API familiar to users of the standard library
+:mod:`marshal` and :mod:`pickle` modules. It is the externally maintained
+version of the :mod:`json` library contained in Python 2.6, but maintains
+compatibility with Python 2.4 and Python 2.5 and (currently) has
+significant performance advantages, even without using the optional C
+extension for speedups.
 
 Encoding basic Python object hierarchies::
 
@@ -32,23 +34,28 @@ Compact encoding::
     >>> json.dumps([1,2,3,{'4': 5, '6': 7}], separators=(',',':'))
     '[1,2,3,{"4":5,"6":7}]'
 
-Pretty printing (using repr() because of extraneous whitespace in the output)::
+Pretty printing::
 
     >>> import json
-    >>> print(repr(json.dumps({'4': 5, '6': 7}, sort_keys=True, indent=4)))
-    '{\n    "4": 5, \n    "6": 7\n}'
+    >>> s = json.dumps({'4': 5, '6': 7}, sort_keys=True, indent=4)
+    >>> print('\n'.join([l.rstrip() for l in  s.splitlines()]))
+    {
+        "4": 5,
+        "6": 7
+    }
 
 Decoding JSON::
 
     >>> import json
-    >>> json.loads('["foo", {"bar":["baz", null, 1.0, 2]}]')
-    ['foo', {'bar': ['baz', None, 1.0, 2]}]
-    >>> json.loads('"\\"foo\\bar"')
-    '"foo\x08ar'
+    >>> obj = ['foo', {'bar': ['baz', None, 1.0, 2]}]
+    >>> json.loads('["foo", {"bar":["baz", null, 1.0, 2]}]') == obj
+    True
+    >>> json.loads('"\\"foo\\bar"') == '"foo\x08ar'
+    True
     >>> from io import StringIO
     >>> io = StringIO('["streaming API"]')
-    >>> json.load(io)
-    ['streaming API']
+    >>> json.load(io)[0] == 'streaming API'
+    True
 
 Specializing JSON object decoding::
 
@@ -61,43 +68,36 @@ Specializing JSON object decoding::
     >>> json.loads('{"__complex__": true, "real": 1, "imag": 2}',
     ...     object_hook=as_complex)
     (1+2j)
-    >>> import decimal
-    >>> json.loads('1.1', parse_float=decimal.Decimal)
-    Decimal('1.1')
+    >>> from decimal import Decimal
+    >>> json.loads('1.1', parse_float=Decimal) == Decimal('1.1')
+    True
 
-Extending JSONEncoder::
+Specializing JSON object encoding::
 
     >>> import json
-    >>> class ComplexEncoder(json.JSONEncoder):
-    ...     def default(self, obj):
-    ...         if isinstance(obj, complex):
-    ...             return [obj.real, obj.imag]
-    ...         return json.JSONEncoder.default(self, obj)
+    >>> def encode_complex(obj):
+    ...     if isinstance(obj, complex):
+    ...         return [obj.real, obj.imag]
+    ...     raise TypeError(repr(o) + " is not JSON serializable")
     ...
-    >>> dumps(2 + 1j, cls=ComplexEncoder)
+    >>> json.dumps(2 + 1j, default=encode_complex)
+    '[2.0, 1.0]'
+    >>> json.JSONEncoder(default=encode_complex).encode(2 + 1j)
     '[2.0, 1.0]'
-    >>> ComplexEncoder().encode(2 + 1j)
+    >>> ''.join(json.JSONEncoder(default=encode_complex).iterencode(2 + 1j))
     '[2.0, 1.0]'
-    >>> list(ComplexEncoder().iterencode(2 + 1j))
-    ['[', '2.0', ', ', '1.0', ']']
 
 
-Using json.tool from the shell to validate and
-pretty-print::
+Using json.tool from the shell to validate and pretty-print::
 
-    $ echo '{"json":"obj"}' | python -mjson.tool
+    $ echo '{"json":"obj"}' | python -m json.tool
     {
         "json": "obj"
     }
-    $ echo '{ 1.2:3.4}' | python -mjson.tool
+    $ echo '{ 1.2:3.4}' | python -m json.tool
     Expecting property name: line 1 column 2 (char 2)
-
-Note that the JSON produced by this module's default settings
-is a subset of YAML, so it may be used as a serializer for that as well.
-
 """
-
-__version__ = '1.9'
+__version__ = '2.0.9'
 __all__ = [
     'dump', 'dumps', 'load', 'loads',
     'JSONDecoder', 'JSONEncoder',
@@ -115,45 +115,43 @@ _default_encoder = JSONEncoder(
     allow_nan=True,
     indent=None,
     separators=None,
-    encoding='utf-8',
     default=None,
 )
 
 def dump(obj, fp, skipkeys=False, ensure_ascii=True, check_circular=True,
         allow_nan=True, cls=None, indent=None, separators=None,
-        encoding='utf-8', default=None, **kw):
+        default=None, **kw):
     """Serialize ``obj`` as a JSON formatted stream to ``fp`` (a
     ``.write()``-supporting file-like object).
 
-    If ``skipkeys`` is ``True`` then ``dict`` keys that are not basic types
-    (``str``, ``unicode``, ``int``, ``long``, ``float``, ``bool``, ``None``)
-    will be skipped instead of raising a ``TypeError``.
+    If ``skipkeys`` is true then ``dict`` keys that are not basic types
+    (``str``, ``unicode``, ``int``, ``float``, ``bool``, ``None``) will be
+    skipped instead of raising a ``TypeError``.
 
-    If ``ensure_ascii`` is ``False``, then the some chunks written to ``fp``
+    If ``ensure_ascii`` is false, then the some chunks written to ``fp``
     may be ``unicode`` instances, subject to normal Python ``str`` to
     ``unicode`` coercion rules. Unless ``fp.write()`` explicitly
     understands ``unicode`` (as in ``codecs.getwriter()``) this is likely
     to cause an error.
 
-    If ``check_circular`` is ``False``, then the circular reference check
+    If ``check_circular`` is false, then the circular reference check
     for container types will be skipped and a circular reference will
     result in an ``OverflowError`` (or worse).
 
-    If ``allow_nan`` is ``False``, then it will be a ``ValueError`` to
+    If ``allow_nan`` is false, then it will be a ``ValueError`` to
     serialize out of range ``float`` values (``nan``, ``inf``, ``-inf``)
     in strict compliance of the JSON specification, instead of using the
     JavaScript equivalents (``NaN``, ``Infinity``, ``-Infinity``).
 
-    If ``indent`` is a non-negative integer, then JSON array elements and object
-    members will be pretty-printed with that indent level. An indent level
-    of 0 will only insert newlines. ``None`` is the most compact representation.
+    If ``indent`` is a non-negative integer, then JSON array elements and
+    object members will be pretty-printed with that indent level. An indent
+    level of 0 will only insert newlines. ``None`` is the most compact
+    representation.
 
     If ``separators`` is an ``(item_separator, dict_separator)`` tuple
     then it will be used instead of the default ``(', ', ': ')`` separators.
     ``(',', ':')`` is the most compact JSON representation.
 
-    ``encoding`` is the character encoding for str instances, default is UTF-8.
-
     ``default(obj)`` is a function that should return a serializable version
     of obj or raise TypeError. The default simply raises TypeError.
 
@@ -163,17 +161,17 @@ def dump(obj, fp, skipkeys=False, ensure_ascii=True, check_circular=True,
 
     """
     # cached encoder
-    if (skipkeys is False and ensure_ascii is True and
-        check_circular is True and allow_nan is True and
+    if (not skipkeys and ensure_ascii and
+        check_circular and allow_nan and
         cls is None and indent is None and separators is None and
-        encoding == 'utf-8' and default is None and not kw):
+        default is None and not kw):
         iterable = _default_encoder.iterencode(obj)
     else:
         if cls is None:
             cls = JSONEncoder
         iterable = cls(skipkeys=skipkeys, ensure_ascii=ensure_ascii,
             check_circular=check_circular, allow_nan=allow_nan, indent=indent,
-            separators=separators, encoding=encoding,
+            separators=separators,
             default=default, **kw).iterencode(obj)
     # could accelerate with writelines in some versions of Python, at
     # a debuggability cost
@@ -183,22 +181,22 @@ def dump(obj, fp, skipkeys=False, ensure_ascii=True, check_circular=True,
 
 def dumps(obj, skipkeys=False, ensure_ascii=True, check_circular=True,
         allow_nan=True, cls=None, indent=None, separators=None,
-        encoding='utf-8', default=None, **kw):
+        default=None, **kw):
     """Serialize ``obj`` to a JSON formatted ``str``.
 
-    If ``skipkeys`` is ``True`` then ``dict`` keys that are not basic types
-    (``str``, ``unicode``, ``int``, ``long``, ``float``, ``bool``, ``None``)
-    will be skipped instead of raising a ``TypeError``.
+    If ``skipkeys`` is false then ``dict`` keys that are not basic types
+    (``str``, ``unicode``, ``int``, ``float``, ``bool``, ``None``) will be
+    skipped instead of raising a ``TypeError``.
 
-    If ``ensure_ascii`` is ``False``, then the return value will be a
+    If ``ensure_ascii`` is false, then the return value will be a
     ``unicode`` instance subject to normal Python ``str`` to ``unicode``
     coercion rules instead of being escaped to an ASCII ``str``.
 
-    If ``check_circular`` is ``False``, then the circular reference check
+    If ``check_circular`` is false, then the circular reference check
     for container types will be skipped and a circular reference will
     result in an ``OverflowError`` (or worse).
 
-    If ``allow_nan`` is ``False``, then it will be a ``ValueError`` to
+    If ``allow_nan`` is false, then it will be a ``ValueError`` to
     serialize out of range ``float`` values (``nan``, ``inf``, ``-inf``) in
     strict compliance of the JSON specification, instead of using the
     JavaScript equivalents (``NaN``, ``Infinity``, ``-Infinity``).
@@ -212,8 +210,6 @@ def dumps(obj, skipkeys=False, ensure_ascii=True, check_circular=True,
     then it will be used instead of the default ``(', ', ': ')`` separators.
     ``(',', ':')`` is the most compact JSON representation.
 
-    ``encoding`` is the character encoding for str instances, default is UTF-8.
-
     ``default(obj)`` is a function that should return a serializable version
     of obj or raise TypeError. The default simply raises TypeError.
 
@@ -223,35 +219,27 @@ def dumps(obj, skipkeys=False, ensure_ascii=True, check_circular=True,
 
     """
     # cached encoder
-    if (skipkeys is False and ensure_ascii is True and
-        check_circular is True and allow_nan is True and
+    if (not skipkeys and ensure_ascii and
+        check_circular and allow_nan and
         cls is None and indent is None and separators is None and
-        encoding == 'utf-8' and default is None and not kw):
+        default is None and not kw):
         return _default_encoder.encode(obj)
     if cls is None:
         cls = JSONEncoder
     return cls(
         skipkeys=skipkeys, ensure_ascii=ensure_ascii,
         check_circular=check_circular, allow_nan=allow_nan, indent=indent,
-        separators=separators, encoding=encoding, default=default,
+        separators=separators, default=default,
         **kw).encode(obj)
 
 
-_default_decoder = JSONDecoder(encoding=None, object_hook=None,
-                               object_pairs_hook=None)
+_default_decoder = JSONDecoder(object_hook=None, object_pairs_hook=None)
 
 
-def load(fp, encoding=None, cls=None, object_hook=None, parse_float=None,
+def load(fp, cls=None, object_hook=None, parse_float=None,
         parse_int=None, parse_constant=None, object_pairs_hook=None, **kw):
-    """Deserialize ``fp`` (a ``.read()``-supporting file-like object
-    containing a JSON document) to a Python object.
-
-    If the contents of ``fp`` is encoded with an ASCII based encoding other
-    than utf-8 (e.g. latin-1), then an appropriate ``encoding`` name must
-    be specified. Encodings that are not ASCII based (such as UCS-2) are
-    not allowed, and should be wrapped with
-    ``codecs.getreader(fp)(encoding)``, or simply decoded to a ``unicode``
-    object and passed to ``loads()``
+    """Deserialize ``fp`` (a ``.read()``-supporting file-like object containing
+    a JSON document) to a Python object.
 
     ``object_hook`` is an optional function that will be called with the
     result of any object literal decode (a ``dict``). The return value of
@@ -263,21 +251,16 @@ def load(fp, encoding=None, cls=None, object_hook=None, parse_float=None,
 
     """
     return loads(fp.read(),
-        encoding=encoding, cls=cls, object_hook=object_hook,
+        cls=cls, object_hook=object_hook,
         parse_float=parse_float, parse_int=parse_int,
         parse_constant=parse_constant, object_pairs_hook=object_pairs_hook, **kw)
 
 
 def loads(s, encoding=None, cls=None, object_hook=None, parse_float=None,
         parse_int=None, parse_constant=None, object_pairs_hook=None, **kw):
-    """Deserialize ``s`` (a ``str`` or ``unicode`` instance containing a JSON
+    """Deserialize ``s`` (a ``str`` instance containing a JSON
     document) to a Python object.
 
-    If ``s`` is a ``str`` instance and is encoded with an ASCII based encoding
-    other than utf-8 (e.g. latin-1) then an appropriate ``encoding`` name
-    must be specified. Encodings that are not ASCII based (such as UCS-2)
-    are not allowed and should be decoded to ``unicode`` first.
-
     ``object_hook`` is an optional function that will be called with the
     result of any object literal decode (a ``dict``). The return value of
     ``object_hook`` will be used instead of the ``dict``. This feature
@@ -302,7 +285,7 @@ def loads(s, encoding=None, cls=None, object_hook=None, parse_float=None,
     kwarg.
 
     """
-    if (cls is None and encoding is None and object_hook is None and
+    if (cls is None and object_hook is None and
             parse_int is None and parse_float is None and
             parse_constant is None and object_pairs_hook is None and not kw):
         return _default_decoder.decode(s)
@@ -318,4 +301,4 @@ def loads(s, encoding=None, cls=None, object_hook=None, parse_float=None,
         kw['parse_int'] = parse_int
     if parse_constant is not None:
         kw['parse_constant'] = parse_constant
-    return cls(encoding=encoding, **kw).decode(s)
+    return cls(**kw).decode(s)
diff --git a/Lib/json/decoder.py b/Lib/json/decoder.py
index b71215b..475b390 100644
--- a/Lib/json/decoder.py
+++ b/Lib/json/decoder.py
@@ -1,10 +1,11 @@
 """Implementation of JSONDecoder
 """
-
+import binascii
 import re
 import sys
+import struct
 
-from json.scanner import Scanner, pattern
+from json.scanner import make_scanner
 try:
     from _json import scanstring as c_scanstring
 except ImportError:
@@ -14,7 +15,14 @@ __all__ = ['JSONDecoder']
 
 FLAGS = re.VERBOSE | re.MULTILINE | re.DOTALL
 
-NaN, PosInf, NegInf = float('nan'), float('inf'), float('-inf')
+def _floatconstants():
+    _BYTES = binascii.unhexlify(b'7FF80000000000007FF0000000000000')
+    if sys.byteorder != 'big':
+        _BYTES = _BYTES[:8][::-1] + _BYTES[8:][::-1]
+    nan, inf = struct.unpack('dd', _BYTES)
+    return nan, inf, -inf
+
+NaN, PosInf, NegInf = _floatconstants()
 
 
 def linecol(doc, pos):
@@ -31,61 +39,43 @@ def linecol(doc, pos):
 
 
 def errmsg(msg, doc, pos, end=None):
+    # Note that this function is called from _json
     lineno, colno = linecol(doc, pos)
     if end is None:
         fmt = '{0}: line {1} column {2} (char {3})'
         return fmt.format(msg, lineno, colno, pos)
+        #fmt = '%s: line %d column %d (char %d)'
+        #return fmt % (msg, lineno, colno, pos)
     endlineno, endcolno = linecol(doc, end)
     fmt = '{0}: line {1} column {2} - line {3} column {4} (char {5} - {6})'
     return fmt.format(msg, lineno, colno, endlineno, endcolno, pos, end)
+    #fmt = '%s: line %d column %d - line %d column %d (char %d - %d)'
+    #return fmt % (msg, lineno, colno, endlineno, endcolno, pos, end)
 
 
 _CONSTANTS = {
     '-Infinity': NegInf,
     'Infinity': PosInf,
     'NaN': NaN,
-    'true': True,
-    'false': False,
-    'null': None,
 }
 
 
-def JSONConstant(match, context, c=_CONSTANTS):
-    s = match.group(0)
-    fn = getattr(context, 'parse_constant', None)
-    if fn is None:
-        rval = c[s]
-    else:
-        rval = fn(s)
-    return rval, None
-pattern('(-?Infinity|NaN|true|false|null)')(JSONConstant)
-
-
-def JSONNumber(match, context):
-    match = JSONNumber.regex.match(match.string, *match.span())
-    integer, frac, exp = match.groups()
-    if frac or exp:
-        fn = getattr(context, 'parse_float', None) or float
-        res = fn(integer + (frac or '') + (exp or ''))
-    else:
-        fn = getattr(context, 'parse_int', None) or int
-        res = fn(integer)
-    return res, None
-pattern(r'(-?(?:0|[1-9][0-9]*))(\.[0-9]+)?([eE][-+]?[0-9]+)?')(JSONNumber)
-
-
 STRINGCHUNK = re.compile(r'(.*?)(["\\\x00-\x1f])', FLAGS)
 BACKSLASH = {
     '"': '"', '\\': '\\', '/': '/',
     'b': '\b', 'f': '\f', 'n': '\n', 'r': '\r', 't': '\t',
 }
 
-DEFAULT_ENCODING = "utf-8"
-
+def py_scanstring(s, end, strict=True,
+        _b=BACKSLASH, _m=STRINGCHUNK.match):
+    """Scan the string s for a JSON string. End is the index of the
+    character in s after the quote that started the JSON string.
+    Unescapes all valid JSON string escape sequences and raises ValueError
+    on attempt to decode an invalid string. If strict is False then literal
+    control characters are allowed in the string.
 
-def py_scanstring(s, end, encoding=None, strict=True, _b=BACKSLASH, _m=STRINGCHUNK.match):
-    if encoding is None:
-        encoding = DEFAULT_ENCODING
+    Returns a tuple of the decoded string and the index of the character in s
+    after the end quote."""
     chunks = []
     _append = chunks.append
     begin = end - 1
@@ -96,14 +86,16 @@ def py_scanstring(s, end, encoding=None, strict=True, _b=BACKSLASH, _m=STRINGCHU
                 errmsg("Unterminated string starting at", s, begin))
         end = chunk.end()
         content, terminator = chunk.groups()
+        # Content is contains zero or more unescaped string characters
         if content:
-            if not isinstance(content, str):
-                content = str(content, encoding)
             _append(content)
+        # Terminator is the end of string, a literal control character,
+        # or a backslash denoting that an escape sequence follows
         if terminator == '"':
             break
         elif terminator != '\\':
             if strict:
+                #msg = "Invalid control character %r at" % (terminator,)
                 msg = "Invalid control character {0!r} at".format(terminator)
                 raise ValueError(errmsg(msg, s, end))
             else:
@@ -114,9 +106,10 @@ def py_scanstring(s, end, encoding=None, strict=True, _b=BACKSLASH, _m=STRINGCHU
         except IndexError:
             raise ValueError(
                 errmsg("Unterminated string starting at", s, begin))
+        # If not a unicode escape sequence, must be in the lookup table
         if esc != 'u':
             try:
-                m = _b[esc]
+                char = _b[esc]
             except KeyError:
                 msg = "Invalid \\escape: {0!r}".format(esc)
                 raise ValueError(errmsg(msg, s, end))
@@ -124,131 +117,138 @@ def py_scanstring(s, end, encoding=None, strict=True, _b=BACKSLASH, _m=STRINGCHU
         else:
             esc = s[end + 1:end + 5]
             next_end = end + 5
-            msg = "Invalid \\uXXXX escape"
-            try:
-                if len(esc) != 4:
-                    raise ValueError
-                uni = int(esc, 16)
-                if 0xd800 <= uni <= 0xdbff and sys.maxunicode > 65535:
-                    msg = "Invalid \\uXXXX\\uXXXX surrogate pair"
-                    if not s[end + 5:end + 7] == '\\u':
-                        raise ValueError
-                    esc2 = s[end + 7:end + 11]
-                    if len(esc2) != 4:
-                        raise ValueError
-                    uni2 = int(esc2, 16)
-                    uni = 0x10000 + (((uni - 0xd800) << 10) | (uni2 - 0xdc00))
-                    next_end += 6
-                m = chr(uni)
-            except ValueError:
+            if len(esc) != 4:
+                msg = "Invalid \\uXXXX escape"
                 raise ValueError(errmsg(msg, s, end))
+            uni = int(esc, 16)
+            # Check for surrogate pair on UCS-4 systems
+            if 0xd800 <= uni <= 0xdbff and sys.maxunicode > 65535:
+                msg = "Invalid \\uXXXX\\uXXXX surrogate pair"
+                if not s[end + 5:end + 7] == '\\u':
+                    raise ValueError(errmsg(msg, s, end))
+                esc2 = s[end + 7:end + 11]
+                if len(esc2) != 4:
+                    raise ValueError(errmsg(msg, s, end))
+                uni2 = int(esc2, 16)
+                uni = 0x10000 + (((uni - 0xd800) << 10) | (uni2 - 0xdc00))
+                next_end += 6
+            char = chr(uni)
+
             end = next_end
-        _append(m)
+        _append(char)
     return ''.join(chunks), end
 
 
-# Use speedup
-if c_scanstring is not None:
-    scanstring = c_scanstring
-else:
-    scanstring = py_scanstring
-
-def JSONString(match, context):
-    encoding = getattr(context, 'encoding', None)
-    strict = getattr(context, 'strict', True)
-    return scanstring(match.string, match.end(), encoding, strict)
-pattern(r'"')(JSONString)
+# Use speedup if available
+scanstring = c_scanstring or py_scanstring
 
+WHITESPACE = re.compile(r'[ \t\n\r]*', FLAGS)
+WHITESPACE_STR = ' \t\n\r'
 
-WHITESPACE = re.compile(r'\s*', FLAGS)
 
-
-def JSONObject(match, context, _w=WHITESPACE.match):
+def JSONObject(s_and_end, strict, scan_once, object_hook, object_pairs_hook,
+        _w=WHITESPACE.match, _ws=WHITESPACE_STR):
+    s, end = s_and_end
     pairs = []
     pairs_append = pairs.append
-    s = match.string
-    end = _w(s, match.end()).end()
+    # Use a slice to prevent IndexError from being raised, the following
+    # check will raise a more specific ValueError if the string is empty
     nextchar = s[end:end + 1]
-    # Trivial empty object
-    if nextchar == '}':
-        return pairs, end + 1
+    # Normally we expect nextchar == '"'
     if nextchar != '"':
-        raise ValueError(errmsg("Expecting property name", s, end))
+        if nextchar in _ws:
+            end = _w(s, end).end()
+            nextchar = s[end:end + 1]
+        # Trivial empty object
+        if nextchar == '}':
+            return pairs, end + 1
+        elif nextchar != '"':
+            raise ValueError(errmsg("Expecting property name", s, end))
     end += 1
-    encoding = getattr(context, 'encoding', None)
-    strict = getattr(context, 'strict', True)
-    iterscan = JSONScanner.iterscan
     while True:
-        key, end = scanstring(s, end, encoding, strict)
-        end = _w(s, end).end()
+        key, end = scanstring(s, end, strict)
+        # To skip some function call overhead we optimize the fast paths where
+        # the JSON key separator is ": " or just ":".
         if s[end:end + 1] != ':':
-            raise ValueError(errmsg("Expecting : delimiter", s, end))
-        end = _w(s, end + 1).end()
+            end = _w(s, end).end()
+            if s[end:end + 1] != ':':
+                raise ValueError(errmsg("Expecting : delimiter", s, end))
+        end += 1
+
         try:
-            value, end = next(iterscan(s, idx=end, context=context))
+            if s[end] in _ws:
+                end += 1
+                if s[end] in _ws:
+                    end = _w(s, end + 1).end()
+        except IndexError:
+            pass
+
+        try:
+            value, end = scan_once(s, end)
         except StopIteration:
             raise ValueError(errmsg("Expecting object", s, end))
         pairs_append((key, value))
-        end = _w(s, end).end()
-        nextchar = s[end:end + 1]
+        try:
+            nextchar = s[end]
+            if nextchar in _ws:
+                end = _w(s, end + 1).end()
+                nextchar = s[end]
+        except IndexError:
+            nextchar = ''
         end += 1
+
         if nextchar == '}':
             break
-        if nextchar != ',':
+        elif nextchar != ',':
             raise ValueError(errmsg("Expecting , delimiter", s, end - 1))
         end = _w(s, end).end()
         nextchar = s[end:end + 1]
         end += 1
         if nextchar != '"':
             raise ValueError(errmsg("Expecting property name", s, end - 1))
-    object_pairs_hook = getattr(context, 'object_pairs_hook', None)
     if object_pairs_hook is not None:
         result = object_pairs_hook(pairs)
         return result, end
     pairs = dict(pairs)
-    object_hook = getattr(context, 'object_hook', None)
     if object_hook is not None:
         pairs = object_hook(pairs)
     return pairs, end
-pattern(r'{')(JSONObject)
-
 
-def JSONArray(match, context, _w=WHITESPACE.match):
+def JSONArray(s_and_end, scan_once, context, _w=WHITESPACE.match):
+    s, end = s_and_end
     values = []
-    s = match.string
-    end = _w(s, match.end()).end()
-    # Look-ahead for trivial empty array
     nextchar = s[end:end + 1]
+    if nextchar in _ws:
+        end = _w(s, end + 1).end()
+        nextchar = s[end:end + 1]
+    # Look-ahead for trivial empty array
     if nextchar == ']':
         return values, end + 1
-    iterscan = JSONScanner.iterscan
+    _append = values.append
     while True:
         try:
-            value, end = next(iterscan(s, idx=end, context=context))
+            value, end = scan_once(s, end)
         except StopIteration:
             raise ValueError(errmsg("Expecting object", s, end))
-        values.append(value)
-        end = _w(s, end).end()
+        _append(value)
         nextchar = s[end:end + 1]
+        if nextchar in _ws:
+            end = _w(s, end + 1).end()
+            nextchar = s[end:end + 1]
         end += 1
         if nextchar == ']':
             break
-        if nextchar != ',':
+        elif nextchar != ',':
             raise ValueError(errmsg("Expecting , delimiter", s, end))
-        end = _w(s, end).end()
-    return values, end
-pattern(r'\[')(JSONArray)
-
-
-ANYTHING = [
-    JSONObject,
-    JSONArray,
-    JSONString,
-    JSONConstant,
-    JSONNumber,
-]
+        try:
+            if s[end] in _ws:
+                end += 1
+                if s[end] in _ws:
+                    end = _w(s, end + 1).end()
+        except IndexError:
+            pass
 
-JSONScanner = Scanner(ANYTHING)
+    return values, end
 
 
 class JSONDecoder(object):
@@ -278,23 +278,14 @@ class JSONDecoder(object):
 
     It also understands ``NaN``, ``Infinity``, and ``-Infinity`` as
     their corresponding ``float`` values, which is outside the JSON spec.
-    """
 
-    _scanner = Scanner(ANYTHING)
-    __all__ = ['__init__', 'decode', 'raw_decode']
+    """
 
-    def __init__(self, encoding=None, object_hook=None, parse_float=None,
+    def __init__(self, object_hook=None, parse_float=None,
             parse_int=None, parse_constant=None, strict=True,
             object_pairs_hook=None):
-        """``encoding`` determines the encoding used to interpret any ``str``
-        objects decoded by this instance (utf-8 by default).  It has no
-        effect when decoding ``unicode`` objects.
-
-        Note that currently only encodings that are a superset of ASCII work,
-        strings of other encodings should be passed in as ``unicode``.
-
-        ``object_hook``, if specified, will be called with the result of
-        every JSON object decoded and its return value will be used in
+        """``object_hook``, if specified, will be called with the result
+        of every JSON object decoded and its return value will be used in
         place of the given ``dict``.  This can be used to provide custom
         deserializations (e.g. to support JSON-RPC class hinting).
 
@@ -309,22 +300,25 @@ class JSONDecoder(object):
         for JSON integers (e.g. float).
 
         ``parse_constant``, if specified, will be called with one of the
-        following strings: -Infinity, Infinity, NaN, null, true, false.
+        following strings: -Infinity, Infinity, NaN.
         This can be used to raise an exception if invalid JSON numbers
         are encountered.
 
         """
-        self.encoding = encoding
         self.object_hook = object_hook
-        self.object_pairs_hook = object_pairs_hook
-        self.parse_float = parse_float
-        self.parse_int = parse_int
-        self.parse_constant = parse_constant
+        self.parse_float = parse_float or float
+        self.parse_int = parse_int or int
+        self.parse_constant = parse_constant or _CONSTANTS.__getitem__
         self.strict = strict
+        self.object_pairs_hook = object_pairs_hook
+        self.parse_object = JSONObject
+        self.parse_array = JSONArray
+        self.parse_string = scanstring
+        self.scan_once = make_scanner(self)
+
 
     def decode(self, s, _w=WHITESPACE.match):
-        """
-        Return the Python representation of ``s`` (a ``str`` or ``unicode``
+        """Return the Python representation of ``s`` (a ``str`` or ``unicode``
         instance containing a JSON document)
 
         """
@@ -334,18 +328,17 @@ class JSONDecoder(object):
             raise ValueError(errmsg("Extra data", s, end, len(s)))
         return obj
 
-    def raw_decode(self, s, **kw):
-        """Decode a JSON document from ``s`` (a ``str`` or ``unicode`` beginning
-        with a JSON document) and return a 2-tuple of the Python
+    def raw_decode(self, s, idx=0):
+        """Decode a JSON document from ``s`` (a ``str`` or ``unicode``
+        beginning with a JSON document) and return a 2-tuple of the Python
         representation and the index in ``s`` where the document ended.
 
         This can be used to decode a JSON document from a string that may
         have extraneous data at the end.
 
         """
-        kw.setdefault('context', self)
         try:
-            obj, end = next(self._scanner.iterscan(s, **kw))
+            obj, end = self.scan_once(s, idx)
         except StopIteration:
             raise ValueError("No JSON object could be decoded")
         return obj, end
diff --git a/Lib/json/encoder.py b/Lib/json/encoder.py
index fa744c9..3c306ab 100644
--- a/Lib/json/encoder.py
+++ b/Lib/json/encoder.py
@@ -1,19 +1,19 @@
 """Implementation of JSONEncoder
 """
-
 import re
-import math
 
 try:
     from _json import encode_basestring_ascii as c_encode_basestring_ascii
 except ImportError:
     c_encode_basestring_ascii = None
-
-__all__ = ['JSONEncoder']
+try:
+    from _json import make_encoder as c_make_encoder
+except ImportError:
+    c_make_encoder = None
 
 ESCAPE = re.compile(r'[\x00-\x1f\\"\b\f\n\r\t]')
 ESCAPE_ASCII = re.compile(r'([\\"]|[^\ -~])')
-HAS_UTF8 = re.compile(r'[\x80-\xff]')
+HAS_UTF8 = re.compile(b'[\x80-\xff]')
 ESCAPE_DCT = {
     '\\': '\\\\',
     '"': '\\"',
@@ -25,30 +25,12 @@ ESCAPE_DCT = {
 }
 for i in range(0x20):
     ESCAPE_DCT.setdefault(chr(i), '\\u{0:04x}'.format(i))
+    #ESCAPE_DCT.setdefault(chr(i), '\\u%04x' % (i,))
 
+# Assume this produces an infinity on all machines (probably not guaranteed)
+INFINITY = float('1e66666')
 FLOAT_REPR = repr
 
-def floatstr(o, allow_nan=True):
-    # Check for specials.  Note that this type of test is processor- and/or
-    # platform-specific, so do tests which don't depend on the internals.
-
-    if math.isnan(o):
-        text = 'NaN'
-    elif math.isinf(o):
-        if math.copysign(1., o) == 1.:
-            text = 'Infinity'
-        else:
-            text = '-Infinity'
-    else:
-        return FLOAT_REPR(o)
-
-    if not allow_nan:
-        msg = "Out of range float values are not JSON compliant: " + repr(o)
-        raise ValueError(msg)
-
-    return text
-
-
 def encode_basestring(s):
     """Return a JSON representation of a Python string
 
@@ -59,8 +41,9 @@ def encode_basestring(s):
 
 
 def py_encode_basestring_ascii(s):
-    if isinstance(s, bytes): # and HAS_UTF8.search(s) is not None:
-        s = s.decode('utf-8')
+    """Return an ASCII-only JSON representation of a Python string
+
+    """
     def replace(match):
         s = match.group(0)
         try:
@@ -69,20 +52,18 @@ def py_encode_basestring_ascii(s):
             n = ord(s)
             if n < 0x10000:
                 return '\\u{0:04x}'.format(n)
+                #return '\\u%04x' % (n,)
             else:
                 # surrogate pair
                 n -= 0x10000
                 s1 = 0xd800 | ((n >> 10) & 0x3ff)
                 s2 = 0xdc00 | (n & 0x3ff)
                 return '\\u{0:04x}\\u{1:04x}'.format(s1, s2)
-    return '"' + (ESCAPE_ASCII.sub(replace, s)) + '"'
+    return '"' + ESCAPE_ASCII.sub(replace, s) + '"'
 
 
-if c_encode_basestring_ascii is not None:
-    encode_basestring_ascii = c_encode_basestring_ascii
-else:
-    encode_basestring_ascii = py_encode_basestring_ascii
-
+encode_basestring_ascii = (
+    c_encode_basestring_ascii or py_encode_basestring_ascii)
 
 class JSONEncoder(object):
     """Extensible JSON <http://json.org> encoder for Python data structures.
@@ -113,33 +94,32 @@ class JSONEncoder(object):
     implementation (to raise ``TypeError``).
 
     """
-    __all__ = ['__init__', 'default', 'encode', 'iterencode']
     item_separator = ', '
     key_separator = ': '
     def __init__(self, skipkeys=False, ensure_ascii=True,
             check_circular=True, allow_nan=True, sort_keys=False,
-            indent=None, separators=None, encoding='utf-8', default=None):
+            indent=None, separators=None, default=None):
         """Constructor for JSONEncoder, with sensible defaults.
 
-        If skipkeys is False, then it is a TypeError to attempt
+        If skipkeys is false, then it is a TypeError to attempt
         encoding of keys that are not str, int, long, float or None.  If
         skipkeys is True, such items are simply skipped.
 
-        If ensure_ascii is True, the output is guaranteed to be str
+        If ensure_ascii is true, the output is guaranteed to be str
         objects with all incoming unicode characters escaped.  If
         ensure_ascii is false, the output will be unicode object.
 
-        If check_circular is True, then lists, dicts, and custom encoded
+        If check_circular is true, then lists, dicts, and custom encoded
         objects will be checked for circular references during encoding to
         prevent an infinite recursion (which would cause an OverflowError).
         Otherwise, no such check takes place.
 
-        If allow_nan is True, then NaN, Infinity, and -Infinity will be
+        If allow_nan is true, then NaN, Infinity, and -Infinity will be
         encoded as such.  This behavior is not JSON specification compliant,
         but is consistent with most JavaScript based encoders and decoders.
         Otherwise, it will be a ValueError to encode such floats.
 
-        If sort_keys is True, then the output of dictionaries will be
+        If sort_keys is true, then the output of dictionaries will be
         sorted by key; this is useful for regression tests to ensure
         that JSON serializations can be compared on a day-to-day basis.
 
@@ -156,28 +136,130 @@ class JSONEncoder(object):
         that can't otherwise be serialized.  It should return a JSON encodable
         version of the object or raise a ``TypeError``.
 
-        If encoding is not None, then all input strings will be
-        transformed into unicode using that encoding prior to JSON-encoding.
-        The default is UTF-8.
-
         """
+
         self.skipkeys = skipkeys
         self.ensure_ascii = ensure_ascii
         self.check_circular = check_circular
         self.allow_nan = allow_nan
         self.sort_keys = sort_keys
         self.indent = indent
-        self.current_indent_level = 0
         if separators is not None:
             self.item_separator, self.key_separator = separators
         if default is not None:
             self.default = default
-        self.encoding = encoding
 
-    def _newline_indent(self):
-        return '\n' + (' ' * (self.indent * self.current_indent_level))
+    def default(self, o):
+        """Implement this method in a subclass such that it returns
+        a serializable object for ``o``, or calls the base implementation
+        (to raise a ``TypeError``).
+
+        For example, to support arbitrary iterators, you could
+        implement default like this::
+
+            def default(self, o):
+                try:
+                    iterable = iter(o)
+                except TypeError:
+                    pass
+                else:
+                    return list(iterable)
+                return JSONEncoder.default(self, o)
+
+        """
+        raise TypeError(repr(o) + " is not JSON serializable")
+
+    def encode(self, o):
+        """Return a JSON string representation of a Python data structure.
+
+        >>> JSONEncoder().encode({"foo": ["bar", "baz"]})
+        '{"foo": ["bar", "baz"]}'
+
+        """
+        # This is for extremely simple cases and benchmarks.
+        if isinstance(o, str):
+            if self.ensure_ascii:
+                return encode_basestring_ascii(o)
+            else:
+                return encode_basestring(o)
+        # This doesn't pass the iterator directly to ''.join() because the
+        # exceptions aren't as detailed.  The list call should be roughly
+        # equivalent to the PySequence_Fast that ''.join() would do.
+        chunks = self.iterencode(o, _one_shot=True)
+        if not isinstance(chunks, (list, tuple)):
+            chunks = list(chunks)
+        return ''.join(chunks)
+
+    def iterencode(self, o, _one_shot=False):
+        """Encode the given object and yield each string
+        representation as available.
+
+        For example::
+
+            for chunk in JSONEncoder().iterencode(bigobject):
+                mysocket.write(chunk)
+
+        """
+        if self.check_circular:
+            markers = {}
+        else:
+            markers = None
+        if self.ensure_ascii:
+            _encoder = encode_basestring_ascii
+        else:
+            _encoder = encode_basestring
+
+        def floatstr(o, allow_nan=self.allow_nan,
+                _repr=FLOAT_REPR, _inf=INFINITY, _neginf=-INFINITY):
+            # Check for specials.  Note that this type of test is processor
+            # and/or platform-specific, so do tests which don't depend on the
+            # internals.
+
+            if o != o:
+                text = 'NaN'
+            elif o == _inf:
+                text = 'Infinity'
+            elif o == _neginf:
+                text = '-Infinity'
+            else:
+                return _repr(o)
+
+            if not allow_nan:
+                raise ValueError(
+                    "Out of range float values are not JSON compliant: " +
+                    repr(o))
+
+            return text
 
-    def _iterencode_list(self, lst, markers=None):
+
+        if (_one_shot and c_make_encoder is not None
+                and not self.indent and not self.sort_keys):
+            _iterencode = c_make_encoder(
+                markers, self.default, _encoder, self.indent,
+                self.key_separator, self.item_separator, self.sort_keys,
+                self.skipkeys, self.allow_nan)
+        else:
+            _iterencode = _make_iterencode(
+                markers, self.default, _encoder, self.indent, floatstr,
+                self.key_separator, self.item_separator, self.sort_keys,
+                self.skipkeys, _one_shot)
+        return _iterencode(o, 0)
+
+def _make_iterencode(markers, _default, _encoder, _indent, _floatstr,
+        _key_separator, _item_separator, _sort_keys, _skipkeys, _one_shot,
+        ## HACK: hand-optimized bytecode; turn globals into locals
+        ValueError=ValueError,
+        dict=dict,
+        float=float,
+        id=id,
+        int=int,
+        isinstance=isinstance,
+        list=list,
+        str=str,
+        tuple=tuple,
+    ):
+
+    def _iterencode_list(lst, _current_indent_level):
         if not lst:
             yield '[]'
             return
@@ -186,31 +268,51 @@ class JSONEncoder(object):
             if markerid in markers:
                 raise ValueError("Circular reference detected")
             markers[markerid] = lst
-        yield '['
-        if self.indent is not None:
-            self.current_indent_level += 1
-            newline_indent = self._newline_indent()
-            separator = self.item_separator + newline_indent
-            yield newline_indent
+        buf = '['
+        if _indent is not None:
+            _current_indent_level += 1
+            newline_indent = '\n' + (' ' * (_indent * _current_indent_level))
+            separator = _item_separator + newline_indent
+            buf += newline_indent
         else:
             newline_indent = None
-            separator = self.item_separator
+            separator = _item_separator
         first = True
         for value in lst:
             if first:
                 first = False
             else:
-                yield separator
-            for chunk in self._iterencode(value, markers):
-                yield chunk
+                buf = separator
+            if isinstance(value, str):
+                yield buf + _encoder(value)
+            elif value is None:
+                yield buf + 'null'
+            elif value is True:
+                yield buf + 'true'
+            elif value is False:
+                yield buf + 'false'
+            elif isinstance(value, int):
+                yield buf + str(value)
+            elif isinstance(value, float):
+                yield buf + _floatstr(value)
+            else:
+                yield buf
+                if isinstance(value, (list, tuple)):
+                    chunks = _iterencode_list(value, _current_indent_level)
+                elif isinstance(value, dict):
+                    chunks = _iterencode_dict(value, _current_indent_level)
+                else:
+                    chunks = _iterencode(value, _current_indent_level)
+                for chunk in chunks:
+                    yield chunk
         if newline_indent is not None:
-            self.current_indent_level -= 1
-            yield self._newline_indent()
+            _current_indent_level -= 1
+            yield '\n' + (' ' * (_indent * _current_indent_level))
         yield ']'
         if markers is not None:
             del markers[markerid]
 
-    def _iterencode_dict(self, dct, markers=None):
+    def _iterencode_dict(dct, _current_indent_level):
         if not dct:
             yield '{}'
             return
@@ -220,78 +322,75 @@ class JSONEncoder(object):
                 raise ValueError("Circular reference detected")
             markers[markerid] = dct
         yield '{'
-        key_separator = self.key_separator
-        if self.indent is not None:
-            self.current_indent_level += 1
-            newline_indent = self._newline_indent()
-            item_separator = self.item_separator + newline_indent
+        if _indent is not None:
+            _current_indent_level += 1
+            newline_indent = '\n' + (' ' * (_indent * _current_indent_level))
+            item_separator = _item_separator + newline_indent
             yield newline_indent
         else:
             newline_indent = None
-            item_separator = self.item_separator
+            item_separator = _item_separator
         first = True
-        if self.ensure_ascii:
-            encoder = encode_basestring_ascii
-        else:
-            encoder = encode_basestring
-        allow_nan = self.allow_nan
-        if self.sort_keys:
-            keys = list(dct.keys())
-            keys.sort()
-            items = [(k, dct[k]) for k in keys]
+        if _sort_keys:
+            items = sorted(dct.items(), key=lambda kv: kv[0])
         else:
-            items = iter(dct.items())
-        _encoding = self.encoding
-        _do_decode = (_encoding is not None
-            and not (_encoding == 'utf-8'))
+            items = dct.items()
         for key, value in items:
             if isinstance(key, str):
-                if _do_decode:
-                    key = key.decode(_encoding)
-            elif isinstance(key, str):
                 pass
             # JavaScript is weakly typed for these, so it makes sense to
             # also allow them.  Many encoders seem to do something like this.
             elif isinstance(key, float):
-                key = floatstr(key, allow_nan)
-            elif isinstance(key, (int, int)):
-                key = str(key)
+                key = _floatstr(key)
             elif key is True:
                 key = 'true'
             elif key is False:
                 key = 'false'
             elif key is None:
                 key = 'null'
-            elif self.skipkeys:
+            elif isinstance(key, int):
+                key = str(key)
+            elif _skipkeys:
                 continue
             else:
-                raise TypeError("key {0!r} is not a string".format(key))
+                raise TypeError("key " + repr(key) + " is not a string")
             if first:
                 first = False
             else:
                 yield item_separator
-            yield encoder(key)
-            yield key_separator
-            for chunk in self._iterencode(value, markers):
-                yield chunk
+            yield _encoder(key)
+            yield _key_separator
+            if isinstance(value, str):
+                yield _encoder(value)
+            elif value is None:
+                yield 'null'
+            elif value is True:
+                yield 'true'
+            elif value is False:
+                yield 'false'
+            elif isinstance(value, int):
+                yield str(value)
+            elif isinstance(value, float):
+                yield _floatstr(value)
+            else:
+                if isinstance(value, (list, tuple)):
+                    chunks = _iterencode_list(value, _current_indent_level)
+                elif isinstance(value, dict):
+                    chunks = _iterencode_dict(value, _current_indent_level)
+                else:
+                    chunks = _iterencode(value, _current_indent_level)
+                for chunk in chunks:
+                    yield chunk
         if newline_indent is not None:
-            self.current_indent_level -= 1
-            yield self._newline_indent()
+            _current_indent_level -= 1
+            yield '\n' + (' ' * (_indent * _current_indent_level))
         yield '}'
         if markers is not None:
             del markers[markerid]
 
-    def _iterencode(self, o, markers=None):
+    def _iterencode(o, _current_indent_level):
         if isinstance(o, str):
-            if self.ensure_ascii:
-                encoder = encode_basestring_ascii
-            else:
-                encoder = encode_basestring
-            _encoding = self.encoding
-            if (_encoding is not None and isinstance(o, str)
-                    and not (_encoding == 'utf-8')):
-                o = o.decode(_encoding)
-            yield encoder(o)
+            yield _encoder(o)
         elif o is None:
             yield 'null'
         elif o is True:
@@ -301,12 +400,12 @@ class JSONEncoder(object):
         elif isinstance(o, (int, int)):
             yield str(o)
         elif isinstance(o, float):
-            yield floatstr(o, self.allow_nan)
+            yield _floatstr(o)
         elif isinstance(o, (list, tuple)):
-            for chunk in self._iterencode_list(o, markers):
+            for chunk in _iterencode_list(o, _current_indent_level):
                 yield chunk
         elif isinstance(o, dict):
-            for chunk in self._iterencode_dict(o, markers):
+            for chunk in _iterencode_dict(o, _current_indent_level):
                 yield chunk
         else:
             if markers is not None:
@@ -314,71 +413,9 @@ class JSONEncoder(object):
                 if markerid in markers:
                     raise ValueError("Circular reference detected")
                 markers[markerid] = o
-            for chunk in self._iterencode_default(o, markers):
+            o = _default(o)
+            for chunk in _iterencode(o, _current_indent_level):
                 yield chunk
             if markers is not None:
                 del markers[markerid]
-
-    def _iterencode_default(self, o, markers=None):
-        newobj = self.default(o)
-        return self._iterencode(newobj, markers)
-
-    def default(self, o):
-        """Implement this method in a subclass such that it returns a serializable
-        object for ``o``, or calls the base implementation (to raise a
-        ``TypeError``).
-
-        For example, to support arbitrary iterators, you could implement
-        default like this::
-
-            def default(self, o):
-                try:
-                    iterable = iter(o)
-                except TypeError:
-                    pass
-                else:
-                    return list(iterable)
-                return JSONEncoder.default(self, o)
-
-        """
-        raise TypeError(repr(o) + " is not JSON serializable")
-
-    def encode(self, o):
-        """Return a JSON string representation of a Python data structure.
-
-        >>> JSONEncoder().encode({"foo": ["bar", "baz"]})
-        '{"foo": ["bar", "baz"]}'
-
-        """
-        # This is for extremely simple cases and benchmarks.
-        if isinstance(o, (str, bytes)):
-            if isinstance(o, bytes):
-                _encoding = self.encoding
-                if (_encoding is not None
-                        and not (_encoding == 'utf-8')):
-                    o = o.decode(_encoding)
-            if self.ensure_ascii:
-                return encode_basestring_ascii(o)
-            else:
-                return encode_basestring(o)
-        # This doesn't pass the iterator directly to ''.join() because the
-        # exceptions aren't as detailed.  The list call should be roughly
-        # equivalent to the PySequence_Fast that ''.join() would do.
-        chunks = list(self.iterencode(o))
-        return ''.join(chunks)
-
-    def iterencode(self, o):
-        """Encode the given object and yield each string representation as
-        available.
-
-        For example::
-
-            for chunk in JSONEncoder().iterencode(bigobject):
-                mysocket.write(chunk)
-
-        """
-        if self.check_circular:
-            markers = {}
-        else:
-            markers = None
-        return self._iterencode(o, markers)
+    return _iterencode
diff --git a/Lib/json/scanner.py b/Lib/json/scanner.py
index 4b065ab..b4f3561 100644
--- a/Lib/json/scanner.py
+++ b/Lib/json/scanner.py
@@ -1,69 +1,65 @@
-"""Iterator based sre token scanner
-
+"""JSON token scanner
 """
-
 import re
-import sre_parse
-import sre_compile
-import sre_constants
-
-from re import VERBOSE, MULTILINE, DOTALL
-from sre_constants import BRANCH, SUBPATTERN
+try:
+    from _json import make_scanner as c_make_scanner
+except ImportError:
+    c_make_scanner = None
 
-__all__ = ['Scanner', 'pattern']
+__all__ = ['make_scanner']
 
-FLAGS = (VERBOSE | MULTILINE | DOTALL)
+NUMBER_RE = re.compile(
+    r'(-?(?:0|[1-9]\d*))(\.\d+)?([eE][-+]?\d+)?',
+    (re.VERBOSE | re.MULTILINE | re.DOTALL))
 
-class Scanner(object):
-    def __init__(self, lexicon, flags=FLAGS):
-        self.actions = [None]
-        # Combine phrases into a compound pattern
-        s = sre_parse.Pattern()
-        s.flags = flags
-        p = []
-        for idx, token in enumerate(lexicon):
-            phrase = token.pattern
-            try:
-                subpattern = sre_parse.SubPattern(s,
-                    [(SUBPATTERN, (idx + 1, sre_parse.parse(phrase, flags)))])
-            except sre_constants.error:
-                raise
-            p.append(subpattern)
-            self.actions.append(token)
+def py_make_scanner(context):
+    parse_object = context.parse_object
+    parse_array = context.parse_array
+    parse_string = context.parse_string
+    match_number = NUMBER_RE.match
+    strict = context.strict
+    parse_float = context.parse_float
+    parse_int = context.parse_int
+    parse_constant = context.parse_constant
+    object_hook = context.object_hook
 
-        s.groups = len(p) + 1 # NOTE(guido): Added to make SRE validation work
-        p = sre_parse.SubPattern(s, [(BRANCH, (None, p))])
-        self.scanner = sre_compile.compile(p)
+    def _scan_once(string, idx):
+        try:
+            nextchar = string[idx]
+        except IndexError:
+            raise StopIteration
 
-    def iterscan(self, string, idx=0, context=None):
-        """Yield match, end_idx for each match
+        if nextchar == '"':
+            return parse_string(string, idx + 1, strict)
+        elif nextchar == '{':
+            return parse_object((string, idx + 1), strict,
+                _scan_once, object_hook, object_pairs_hook)
+        elif nextchar == '[':
+            return parse_array((string, idx + 1), _scan_once)
+        elif nextchar == 'n' and string[idx:idx + 4] == 'null':
+            return None, idx + 4
+        elif nextchar == 't' and string[idx:idx + 4] == 'true':
+            return True, idx + 4
+        elif nextchar == 'f' and string[idx:idx + 5] == 'false':
+            return False, idx + 5
 
-        """
-        match = self.scanner.scanner(string, idx).match
-        actions = self.actions
-        lastend = idx
-        end = len(string)
-        while True:
-            m = match()
-            if m is None:
-                break
-            matchbegin, matchend = m.span()
-            if lastend == matchend:
-                break
-            action = actions[m.lastindex]
-            if action is not None:
-                rval, next_pos = action(m, context)
-                if next_pos is not None and next_pos != matchend:
-                    # "fast forward" the scanner
-                    matchend = next_pos
-                    match = self.scanner.scanner(string, matchend).match
-                yield rval, matchend
-            lastend = matchend
+        m = match_number(string, idx)
+        if m is not None:
+            integer, frac, exp = m.groups()
+            if frac or exp:
+                res = parse_float(integer + (frac or '') + (exp or ''))
+            else:
+                res = parse_int(integer)
+            return res, m.end()
+        elif nextchar == 'N' and string[idx:idx + 3] == 'NaN':
+            return parse_constant('NaN'), idx + 3
+        elif nextchar == 'I' and string[idx:idx + 8] == 'Infinity':
+            return parse_constant('Infinity'), idx + 8
+        elif nextchar == '-' and string[idx:idx + 9] == '-Infinity':
+            return parse_constant('-Infinity'), idx + 9
+        else:
+            raise StopIteration
 
+    return _scan_once
 
-def pattern(pattern, flags=FLAGS):
-    def decorator(fn):
-        fn.pattern = pattern
-        fn.regex = re.compile(pattern, flags)
-        return fn
-    return decorator
+make_scanner = c_make_scanner or py_make_scanner
diff --git a/Lib/json/tests/test_decode.py b/Lib/json/tests/test_decode.py
index 67c4c29..78e7e07 100644
--- a/Lib/json/tests/test_decode.py
+++ b/Lib/json/tests/test_decode.py
@@ -32,3 +32,10 @@ class TestDecode(TestCase):
                                     object_pairs_hook = OrderedDict,
                                     object_hook = lambda x: None),
                          OrderedDict(p))
+
+    def test_decoder_optimizations(self):
+        # Several optimizations were made that skip over calls to
+        # the whitespace regex, so this test is designed to try and
+        # exercise the uncommon cases. The array cases are already covered.
+        rval = json.loads('{   "key"    :    "value"    ,  "k":"v"    }')
+        self.assertEquals(rval, {"key":"value", "k":"v"})
diff --git a/Lib/json/tests/test_dump.py b/Lib/json/tests/test_dump.py
index 06a56d7..54e218f 100644
--- a/Lib/json/tests/test_dump.py
+++ b/Lib/json/tests/test_dump.py
@@ -11,3 +11,11 @@ class TestDump(TestCase):
 
     def test_dumps(self):
         self.assertEquals(json.dumps({}), '{}')
+
+    def test_encode_truefalse(self):
+        self.assertEquals(json.dumps(
+                 {True: False, False: True}, sort_keys=True),
+                 '{"false": true, "true": false}')
+        self.assertEquals(json.dumps(
+                {2: 3.0, 4.0: 5, False: 1, 6: True}, sort_keys=True),
+                '{"false": 1, "2": 3.0, "4.0": 5, "6": true}')
diff --git a/Lib/json/tests/test_encode_basestring_ascii.py b/Lib/json/tests/test_encode_basestring_ascii.py
index 4517ae9..68ce7ef 100644
--- a/Lib/json/tests/test_encode_basestring_ascii.py
+++ b/Lib/json/tests/test_encode_basestring_ascii.py
@@ -3,22 +3,20 @@ from unittest import TestCase
 import json.encoder
 
 CASES = [
-    ('/\\"\ucafe\ubabe\uab98\ufcde\ubcda\uef4a\x08\x0c\n\r\t`1~!@#$%^&*()_+-=[]{}|;:\',./<>?', b'"/\\\\\\"\\ucafe\\ubabe\\uab98\\ufcde\\ubcda\\uef4a\\b\\f\\n\\r\\t`1~!@#$%^&*()_+-=[]{}|;:\',./<>?"'),
-    ('\u0123\u4567\u89ab\ucdef\uabcd\uef4a', b'"\\u0123\\u4567\\u89ab\\ucdef\\uabcd\\uef4a"'),
-    ('controls', b'"controls"'),
-    ('\x08\x0c\n\r\t', b'"\\b\\f\\n\\r\\t"'),
-    ('{"object with 1 member":["array with 1 element"]}', b'"{\\"object with 1 member\\":[\\"array with 1 element\\"]}"'),
-    (' s p a c e d ', b'" s p a c e d "'),
-    ('\U0001d120', b'"\\ud834\\udd20"'),
-    ('\u03b1\u03a9', b'"\\u03b1\\u03a9"'),
-    (b'\xce\xb1\xce\xa9', b'"\\u03b1\\u03a9"'),
-    ('\u03b1\u03a9', b'"\\u03b1\\u03a9"'),
-    (b'\xce\xb1\xce\xa9', b'"\\u03b1\\u03a9"'),
-    ('\u03b1\u03a9', b'"\\u03b1\\u03a9"'),
-    ('\u03b1\u03a9', b'"\\u03b1\\u03a9"'),
-    ("`1~!@#$%^&*()_+-={':[,]}|;.</>?", b'"`1~!@#$%^&*()_+-={\':[,]}|;.</>?"'),
-    ('\x08\x0c\n\r\t', b'"\\b\\f\\n\\r\\t"'),
-    ('\u0123\u4567\u89ab\ucdef\uabcd\uef4a', b'"\\u0123\\u4567\\u89ab\\ucdef\\uabcd\\uef4a"'),
+    ('/\\"\ucafe\ubabe\uab98\ufcde\ubcda\uef4a\x08\x0c\n\r\t`1~!@#$%^&*()_+-=[]{}|;:\',./<>?', '"/\\\\\\"\\ucafe\\ubabe\\uab98\\ufcde\\ubcda\\uef4a\\b\\f\\n\\r\\t`1~!@#$%^&*()_+-=[]{}|;:\',./<>?"'),
+    ('\u0123\u4567\u89ab\ucdef\uabcd\uef4a', '"\\u0123\\u4567\\u89ab\\ucdef\\uabcd\\uef4a"'),
+    ('controls', '"controls"'),
+    ('\x08\x0c\n\r\t', '"\\b\\f\\n\\r\\t"'),
+    ('{"object with 1 member":["array with 1 element"]}', '"{\\"object with 1 member\\":[\\"array with 1 element\\"]}"'),
+    (' s p a c e d ', '" s p a c e d "'),
+    ('\U0001d120', '"\\ud834\\udd20"'),
+    ('\u03b1\u03a9', '"\\u03b1\\u03a9"'),
+    ('\u03b1\u03a9', '"\\u03b1\\u03a9"'),
+    ('\u03b1\u03a9', '"\\u03b1\\u03a9"'),
+    ('\u03b1\u03a9', '"\\u03b1\\u03a9"'),
+    ("`1~!@#$%^&*()_+-={':[,]}|;.</>?", '"`1~!@#$%^&*()_+-={\':[,]}|;.</>?"'),
+    ('\x08\x0c\n\r\t', '"\\b\\f\\n\\r\\t"'),
+    ('\u0123\u4567\u89ab\ucdef\uabcd\uef4a', '"\\u0123\\u4567\\u89ab\\ucdef\\uabcd\\uef4a"'),
 ]
 
 class TestEncodeBaseStringAscii(TestCase):
@@ -26,12 +24,14 @@ class TestEncodeBaseStringAscii(TestCase):
         self._test_encode_basestring_ascii(json.encoder.py_encode_basestring_ascii)
 
     def test_c_encode_basestring_ascii(self):
-        if json.encoder.c_encode_basestring_ascii is not None:
-            self._test_encode_basestring_ascii(json.encoder.c_encode_basestring_ascii)
+        if not json.encoder.c_encode_basestring_ascii:
+            return
+        self._test_encode_basestring_ascii(json.encoder.c_encode_basestring_ascii)
 
     def _test_encode_basestring_ascii(self, encode_basestring_ascii):
         fname = encode_basestring_ascii.__name__
         for input_string, expect in CASES:
             result = encode_basestring_ascii(input_string)
-            result = result.encode("ascii")
-            self.assertEquals(result, expect)
+            self.assertEquals(result, expect,
+                '{0!r} != {1!r} for {2}({3!r})'.format(
+                    result, expect, fname, input_string))
diff --git a/Lib/json/tests/test_fail.py b/Lib/json/tests/test_fail.py
index 252ebc4..94e01e6 100644
--- a/Lib/json/tests/test_fail.py
+++ b/Lib/json/tests/test_fail.py
@@ -73,4 +73,4 @@ class TestFail(TestCase):
             except ValueError:
                 pass
             else:
-                self.fail("Expected failure for fail%d.json: %r" % (idx, doc))
+                self.fail("Expected failure for fail{0}.json: {1!r}".format(idx, doc))
diff --git a/Lib/json/tests/test_float.py b/Lib/json/tests/test_float.py
index 9df6d1ee..2d4aea2 100644
--- a/Lib/json/tests/test_float.py
+++ b/Lib/json/tests/test_float.py
@@ -5,5 +5,11 @@ import json
 
 class TestFloat(TestCase):
     def test_floats(self):
-        for num in [1617161771.7650001, math.pi, math.pi**100, math.pi**-100]:
+        for num in [1617161771.7650001, math.pi, math.pi**100, math.pi**-100, 3.1]:
             self.assertEquals(float(json.dumps(num)), num)
+            self.assertEquals(json.loads(json.dumps(num)), num)
+
+    def test_ints(self):
+        for num in [1, 1<<32, 1<<64]:
+            self.assertEquals(json.dumps(num), str(num))
+            self.assertEquals(int(json.dumps(num)), num)
diff --git a/Lib/json/tests/test_scanstring.py b/Lib/json/tests/test_scanstring.py
index 025d15d..2d55672 100644
--- a/Lib/json/tests/test_scanstring.py
+++ b/Lib/json/tests/test_scanstring.py
@@ -15,96 +15,90 @@ class TestScanString(TestCase):
 
     def _test_scanstring(self, scanstring):
         self.assertEquals(
-            scanstring('"z\\ud834\\udd20x"', 1, None, True),
+            scanstring('"z\\ud834\\udd20x"', 1, True),
             ('z\U0001d120x', 16))
 
         if sys.maxunicode == 65535:
             self.assertEquals(
-                scanstring('"z\U0001d120x"', 1, None, True),
+                scanstring('"z\U0001d120x"', 1, True),
                 ('z\U0001d120x', 6))
         else:
             self.assertEquals(
-                scanstring('"z\U0001d120x"', 1, None, True),
+                scanstring('"z\U0001d120x"', 1, True),
                 ('z\U0001d120x', 5))
 
         self.assertEquals(
-            scanstring('"\\u007b"', 1, None, True),
+            scanstring('"\\u007b"', 1, True),
             ('{', 8))
 
         self.assertEquals(
-            scanstring('"A JSON payload should be an object or array, not a string."', 1, None, True),
+            scanstring('"A JSON payload should be an object or array, not a string."', 1, True),
             ('A JSON payload should be an object or array, not a string.', 60))
 
         self.assertEquals(
-            scanstring('["Unclosed array"', 2, None, True),
+            scanstring('["Unclosed array"', 2, True),
             ('Unclosed array', 17))
 
         self.assertEquals(
-            scanstring('["extra comma",]', 2, None, True),
+            scanstring('["extra comma",]', 2, True),
             ('extra comma', 14))
 
         self.assertEquals(
-            scanstring('["double extra comma",,]', 2, None, True),
+            scanstring('["double extra comma",,]', 2, True),
             ('double extra comma', 21))
 
         self.assertEquals(
-            scanstring('["Comma after the close"],', 2, None, True),
+            scanstring('["Comma after the close"],', 2, True),
             ('Comma after the close', 24))
 
         self.assertEquals(
-            scanstring('["Extra close"]]', 2, None, True),
+            scanstring('["Extra close"]]', 2, True),
             ('Extra close', 14))
 
         self.assertEquals(
-            scanstring('{"Extra comma": true,}', 2, None, True),
+            scanstring('{"Extra comma": true,}', 2, True),
             ('Extra comma', 14))
 
         self.assertEquals(
-            scanstring('{"Extra value after close": true} "misplaced quoted value"', 2, None, True),
+            scanstring('{"Extra value after close": true} "misplaced quoted value"', 2, True),
             ('Extra value after close', 26))
 
         self.assertEquals(
-            scanstring('{"Illegal expression": 1 + 2}', 2, None, True),
+            scanstring('{"Illegal expression": 1 + 2}', 2, True),
             ('Illegal expression', 21))
 
         self.assertEquals(
-            scanstring('{"Illegal invocation": alert()}', 2, None, True),
+            scanstring('{"Illegal invocation": alert()}', 2, True),
             ('Illegal invocation', 21))
 
         self.assertEquals(
-            scanstring('{"Numbers cannot have leading zeroes": 013}', 2, None, True),
+            scanstring('{"Numbers cannot have leading zeroes": 013}', 2, True),
             ('Numbers cannot have leading zeroes', 37))
 
         self.assertEquals(
-            scanstring('{"Numbers cannot be hex": 0x14}', 2, None, True),
+            scanstring('{"Numbers cannot be hex": 0x14}', 2, True),
             ('Numbers cannot be hex', 24))
 
         self.assertEquals(
-            scanstring('[[[[[[[[[[[[[[[[[[[["Too deep"]]]]]]]]]]]]]]]]]]]]', 21, None, True),
+            scanstring('[[[[[[[[[[[[[[[[[[[["Too deep"]]]]]]]]]]]]]]]]]]]]', 21, True),
             ('Too deep', 30))
 
         self.assertEquals(
-            scanstring('{"Missing colon" null}', 2, None, True),
+            scanstring('{"Missing colon" null}', 2, True),
             ('Missing colon', 16))
 
         self.assertEquals(
-            scanstring('{"Double colon":: null}', 2, None, True),
+            scanstring('{"Double colon":: null}', 2, True),
             ('Double colon', 15))
 
         self.assertEquals(
-            scanstring('{"Comma instead of colon", null}', 2, None, True),
+            scanstring('{"Comma instead of colon", null}', 2, True),
             ('Comma instead of colon', 25))
 
         self.assertEquals(
-            scanstring('["Colon instead of comma": false]', 2, None, True),
+            scanstring('["Colon instead of comma": false]', 2, True),
             ('Colon instead of comma', 25))
 
         self.assertEquals(
-            scanstring('["Bad value", truth]', 2, None, True),
+            scanstring('["Bad value", truth]', 2, True),
             ('Bad value', 12))
-
-    def test_issue3623(self):
-        self.assertRaises(ValueError, json.decoder.scanstring, b"xxx", 1,
-                          "xxx")
-        self.assertRaises(UnicodeDecodeError,
-                          json.encoder.encode_basestring_ascii, b"xx\xff")
diff --git a/Lib/json/tests/test_unicode.py b/Lib/json/tests/test_unicode.py
index 00bf58e..12de83c 100644
--- a/Lib/json/tests/test_unicode.py
+++ b/Lib/json/tests/test_unicode.py
@@ -4,20 +4,8 @@ import json
 from collections import OrderedDict
 
 class TestUnicode(TestCase):
-    def test_encoding1(self):
-        encoder = json.JSONEncoder(encoding='utf-8')
-        u = '\N{GREEK SMALL LETTER ALPHA}\N{GREEK CAPITAL LETTER OMEGA}'
-        s = u.encode('utf-8')
-        ju = encoder.encode(u)
-        js = encoder.encode(s)
-        self.assertEquals(ju, js)
-
-    def test_encoding2(self):
-        u = '\N{GREEK SMALL LETTER ALPHA}\N{GREEK CAPITAL LETTER OMEGA}'
-        s = u.encode('utf-8')
-        ju = json.dumps(u, encoding='utf-8')
-        js = json.dumps(s, encoding='utf-8')
-        self.assertEquals(ju, js)
+    # test_encoding1 and test_encoding2 from 2.x are irrelevant (only str
+    # is supported as input, not bytes).
 
     def test_encoding3(self):
         u = '\N{GREEK SMALL LETTER ALPHA}\N{GREEK CAPITAL LETTER OMEGA}'
@@ -52,8 +40,22 @@ class TestUnicode(TestCase):
     def test_unicode_decode(self):
         for i in range(0, 0xd7ff):
             u = chr(i)
-            js = '"\\u{0:04x}"'.format(i)
-            self.assertEquals(json.loads(js), u)
+            s = '"\\u{0:04x}"'.format(i)
+            self.assertEquals(json.loads(s), u)
+
+    def test_unicode_preservation(self):
+        self.assertEquals(type(json.loads('""')), str)
+        self.assertEquals(type(json.loads('"a"')), str)
+        self.assertEquals(type(json.loads('["a"]')[0]), str)
+
+    def test_bytes_encode(self):
+        self.assertRaises(TypeError, json.dumps, b"hi")
+        self.assertRaises(TypeError, json.dumps, [b"hi"])
+
+    def test_bytes_decode(self):
+        self.assertRaises(TypeError, json.loads, b'"hi"')
+        self.assertRaises(TypeError, json.loads, b'["hi"]')
+
 
     def test_object_pairs_hook_with_unicode(self):
         s = '{"xkd":1, "kcw":2, "art":3, "hxm":4, "qrt":5, "pad":6, "hoy":7}'
diff --git a/Lib/json/tool.py b/Lib/json/tool.py
index cec29b6..6d6558a 100644
--- a/Lib/json/tool.py
+++ b/Lib/json/tool.py
@@ -2,11 +2,11 @@ r"""Command-line tool to validate and pretty-print JSON
 
 Usage::
 
-    $ echo '{"json":"obj"}' | python -mjson.tool
+    $ echo '{"json":"obj"}' | python -m json.tool
     {
         "json": "obj"
     }
-    $ echo '{ 1.2:3.4}' | python -mjson.tool
+    $ echo '{ 1.2:3.4}' | python -m json.tool
     Expecting property name: line 1 column 2 (char 2)
 
 """
@@ -24,7 +24,7 @@ def main():
         infile = open(sys.argv[1], 'rb')
         outfile = open(sys.argv[2], 'wb')
     else:
-        raise SystemExit("{0} [infile [outfile]]".format(sys.argv[0]))
+        raise SystemExit(sys.argv[0] + " [infile [outfile]]")
     try:
         obj = json.load(infile)
     except ValueError as e: