Commits

Anonymous committed 9249eac Merge

Fiz merge entre os branches. Eu tinha um rolando em paralelo não sei porque

  • Participants
  • Parent commits d6d333c, e08c1a7

Comments (0)

Files changed (28)

 import re
 import simplejson
 import time
+import cPickle
 from twitter import twitter
 
 twitter_username = 'fasassim'
 twitter_password = 'abc123'
 
+def get_friends_list(api, force = False):
+    try:
+        friends_list = cPickle.load(open('friends.pickle'))
+    except IOError:
+        force = True
+    if force:
+        friends = api.GetFriends()
+        friends_list = [f.screen_name for f in friends]
+        cPickle.dump(friends_list, open('friends.pickle','w'))
+    return friends_list
+
+def add_friend(api, screen_name):
+    try:
+        new_friend = api.CreateFriendship(screen_name)
+        print "twitter.fasassim.following %s" % new_friend.name
+    except:
+        print "twitter.fasassim.following %s - FAIL" % screen_name
+
+def get_full_feed():
+    feed = urllib.urlopen('http://search.twitter.com/search.json?q=comofas')
+    updates = simplejson.loads(feed.read())
+    cPickle.dump(updates, open('updates.pickle','w'))
+    return updates
+
 def get_last_feed():
-    feed = urllib.urlopen('http://search.twitter.com/search.json?q=comofas')
-    updates = feed.read()
-    updates = simplejson.loads(updates)
-
+    updates = get_full_feed()
     return updates['results'][0]['text'].lower(), updates['results'][0]['from_user']
 
 def clean_text(txt):
     # Lista de palavras que não ficam bem no fim da frase.
     dispensaveis = ['e','a','o','de','da','do','na','no','para','pra','mas','se','que','como','porque']
     a = raw_answer[:last].split()
-    while a[-1] in dispensaveis:
-        a.pop(-1)
+    if a:
+        while a[-1] in dispensaveis:
+            a.pop(-1)
     return "%s%s" % (msg, " ".join(a))
 
 
 
     api = twitter.Api(username=twitter_username, password=twitter_password)
     api.SetXTwitterHeaders('fasassim',"http://twitter.com/fasassim",'Ni!')
-
-    ultimos_usuarios={}
+    
+    friends_list = get_friends_list(api)   
+    print "Sigo %d amigos" % len(friends_list)
+    
+    
     ultima_pergunta=''
+    cycle_count = 0
     while (True):
-        time.sleep(random.randint(15,30)*2)
+        intervalo = random.randint(25,40)*2
+        print "Tempo: %d segundos..." % intervalo
+        time.sleep(intervalo)
+        feed = get_full_feed()
+        
+        if cycle_count >= 5:
+            #checa se no feed tem pessoas novas
+            new_people = False
+            for update in feed['results']:
+                if update['from_user'] not in friends_list:
+                    new_people = True
+                    add_friend(api, update['from_user'])
+            if new_people:
+                friends_list = get_friends_list(api, force=True)
+            cycle_count = 0
+        else:
+            cycle_count += 1
+        
+        #TODO Iterar do último item do feed para o primeiro, checando se já foi respondido.
+        #     Se não foi, responde e grava em algum lugar update_id e flag_respondida
+        #     Ignorar updates mais antigos do que 1 ou 2 horas
         f = get_last_feed()
         pergunta = f[0]
         usuario = f[1]
         follow_counter=0
 
-        ultimos_usuarios[usuario]=1
-
         if ultima_pergunta == pergunta:
             continue
         else:
                 #nao_entendi(usuario)
                 continue
 
-            follow_counter = follow_counter + 1
-
-            if follow_counter==5:
-                follow_counter = 0
-                added_friends = api.GetFriends()
-
-                for friend in added_friends:
-                    if not ultimos_usuarios.has_key(friend.name) and usuario != 'fasassim':
-                        api.CreateFriendship(friend)
-                        print "twitter.fasassim.following %s" % friend.name
-                        ultimos_usuarios.pop(friend.name)
-                ultimos_usuarios={}
 
             if questions:
                 for question in questions:
                         print '-' * 50
                         print "twitter.fasassim postUpdate: %s" % mensagem
                         status = api.PostUpdate(mensagem)
-                        ultimos_usuarios[usuario]=1
+
                     except UnicodeEncodeError:
                         continue
             else:

File friends.pickle

+(lp1
+Vjennytangerine
+p2
+aVgreevin
+p3
+aVsullyvan_blue
+p4
+aVrcarneiro
+p5
+aVcarolda
+p6
+aVrafo_
+p7
+aViana
+p8
+aVbany
+p9
+aVsobresonhos
+p10
+aValeJohnny
+p11
+aVjoaodocaminhao
+p12
+aVshingowatanabe
+p13
+aVrenatoelias
+p14
+aVausoleil
+p15
+aVgi_martins
+p16
+aVnanyrox
+p17
+aVCamilaCat
+p18
+aVforeversyoung
+p19
+aVolier
+p20
+aVlucah
+p21
+aVraquelalmeida
+p22
+aVjustcintia
+p23
+aVblogdoxavier
+p24
+aVagostini
+p25
+aVpaiva_thiago
+p26
+aVcisoxp
+p27
+aVfrangossauro
+p28
+aVpalmergroove
+p29
+aVfabiosantos99
+p30
+aVpushingdaisies
+p31
+aVlebravo
+p32
+aVcristalk
+p33
+aVnitroglicerina
+p34
+aVAlelex88
+p35
+aVdessalita
+p36
+aVserginhotavares
+p37
+aValeh
+p38
+aVdorlyneto
+p39
+aVfelipesimon
+p40
+aVPorquinho
+p41
+aVdotflux
+p42
+aVamanda_leal
+p43
+aVteka_avari
+p44
+aVdorafael
+p45
+aVdanilopopeye
+p46
+aVRodOgrO
+p47
+aVluishpenha
+p48
+aVrodrigomilfont
+p49
+aVnegapreta
+p50
+aVjonatasongs
+p51
+aVlorylooove
+p52
+aVmskulpas
+p53
+aVmborgomani
+p54
+aVmahrcius
+p55
+aVrafaelcamilo
+p56
+aVsilviogois
+p57
+aVnubibella
+p58
+aVCharmQGP
+p59
+aVfabriciodolci
+p60
+aVsamarinacorrea
+p61
+aVgaygeeksolteiro
+p62
+aVmr_biglia
+p63
+aVjvgfranco
+p64
+aVbrenomaciel
+p65
+aVedugards
+p66
+aVViniciusc
+p67
+aVrq_tata
+p68
+aVbarbaragondim
+p69
+aVIncrivelt
+p70
+aVvynn
+p71
+aVrubinhu
+p72
+aVtommarques
+p73
+aVtalita_pereira
+p74
+aVmarifomin
+p75
+aVseguraberenice
+p76
+aVjanainamoyses
+p77
+aVlosis
+p78
+aVjessicapassos
+p79
+aVrodrigojames
+p80
+aVtoscano
+p81
+aVdgmike
+p82
+aVuchizawa
+p83
+aVluanahermann
+p84
+aVpixoxo
+p85
+aVcorbata1982
+p86
+aVgustavocardoso
+p87
+aVstellabrunet
+p88
+aVkakah
+p89
+aV_hans
+p90
+aVbge
+p91
+aV_sukys
+p92
+aVbarone
+p93
+aVeduardovollmond
+p94
+aVjoaopapa
+p95
+aVmarcelohenrique
+p96
+aVlipeg
+p97
+aVsetzerknight
+p98
+aVwal_limoli
+p99
+aVthaismello
+p100
+aVAnanista
+p101
+a.

File pynswers/Answers.py

+import urllib
+import urllib2
+from xml.dom import minidom
+def parseQuestion(ques):            
+    ques_ = {}
+    for node in ques.childNodes:
+        if node.nodeType == node.ELEMENT_NODE and not node.nodeName == 'Answers' :
+            ques_.update(__getData(node.toxml()))
+    return ques_
+
+def __getData(node):
+    i = node.find('>')
+    j = node.rfind('<')
+    k = node.find(' ')
+    return {node[1:i]:node[i+1:j]}    
+class Answers:
+    def __init__(self):
+        self.appid = 'YahooDemo'
+    
+    def questionSearch(self, params):
+        """Answers questionSearch wrapper"""
+        baseUrl = 'http://answers.yahooapis.com/AnswersService/V1/questionSearch?appid='+self.appid+'&'
+        finalUrl = baseUrl
+        for k, v in params.items():
+            finalUrl = finalUrl + urllib.quote(k) +'=' + urllib.quote(v) + '&'
+        finalUrl = finalUrl[:len(finalUrl) - 1]
+        site = urllib2.urlopen(finalUrl)
+        xmlDoc = minidom.parse(site)
+        questions = xmlDoc.getElementsByTagName('Question')
+        qList = []
+        for ques in questions:
+            qList.append(parseQuestion(ques))
+        answers = xmlDoc.getElementsByTagName('Answers')
+        for ans in answers:
+            qList.append(parseQuestion(ans))
+        return qList
+       
+    
+    def getByCategory(self, params):
+        """Answers getByCategory wrapper"""
+        baseUrl = 'http://answers.yahooapis.com/AnswersService/V1/getByCategory?appid='+self.appid+'&'
+        finalUrl = baseUrl
+        for k, v in params.items():
+            finalUrl = finalUrl + urllib.quote(k) +'=' + urllib.quote(v) + '&'
+        finalUrl = finalUrl[:len(finalUrl) - 1]
+        print finalUrl
+        site = urllib2.urlopen(finalUrl)
+        xmlDoc = minidom.parse(site)
+        questions = xmlDoc.getElementsByTagName('Question')
+        qList = []
+        for ques in questions:
+            qList.append(parseQuestion(ques))
+        return qList    
+    
+    def getByUser(self, params):
+        """Answers getByUser wrapper"""
+        baseUrl = 'http://answers.yahooapis.com/AnswersService/V1/getByUser?appid='+self.appid+'&'
+        finalUrl = baseUrl
+        for k, v in params.items():
+            finalUrl = finalUrl + urllib.quote(k) +'=' + urllib.quote(v) + '&'
+        finalUrl = finalUrl[:len(finalUrl) - 1]
+        print finalUrl
+        site = urllib2.urlopen(finalUrl)
+        xmlDoc = minidom.parse(site)
+        questions = xmlDoc.getElementsByTagName('Question')
+        qList = []
+        for ques in questions:
+            qList.append(parseQuestion(ques))
+        return qList      
+   
+   
+    def getQuestion(self, params):
+        """Answers getByUser wrapper"""
+        baseUrl = 'http://answers.yahooapis.com/AnswersService/V1/getQuestion?appid='+self.appid+'&'
+        finalUrl = baseUrl
+        for k, v in params.items():
+            finalUrl = finalUrl + urllib.quote(k) +'=' + urllib.quote(v) + '&'
+        finalUrl = finalUrl[:len(finalUrl) - 1]
+        print finalUrl
+        site = urllib2.urlopen(finalUrl)
+        xmlDoc = minidom.parse(site)
+        questions = xmlDoc.getElementsByTagName('Question')
+        qList = []
+        for ques in questions:
+            qList.append(parseQuestion(ques))
+        answers = xmlDoc.getElementsByTagName('Answer')
+        for ans in answers:
+            qList.append(parseQuestion(ans))        
+        return qList   
+
+            
+if __name__ == '__main__':
+    app = Answers()
+    #print app.questionSearch({'query':'cats', 'type':'open', 'region':'us'})
+    #print len(app.getByCategory({'category_id':'396546304'}))
+    #print app.getByUser({'user_id':'HPsA6va4aa'})
+    #20070331070321AAlyopp
+    a= app.getQuestion({'question_id':'20070331070321AAlyopp'})
+    print len(a)
+    print a

File pynswers/__init__.py

Empty file added.

File simplejson/__init__.py

+r"""A simple, fast, extensible JSON encoder and decoder
+
+JSON (JavaScript Object Notation) <http://json.org> is a subset of
+JavaScript syntax (ECMA-262 3rd edition) used as a lightweight data
+interchange format.
+
+simplejson exposes an API familiar to uses of the standard library
+marshal and pickle modules.
+
+Encoding basic Python object hierarchies::
+
+    >>> import simplejson
+    >>> simplejson.dumps(['foo', {'bar': ('baz', None, 1.0, 2)}])
+    '["foo", {"bar": ["baz", null, 1.0, 2]}]'
+    >>> print simplejson.dumps("\"foo\bar")
+    "\"foo\bar"
+    >>> print simplejson.dumps(u'\u1234')
+    "\u1234"
+    >>> print simplejson.dumps('\\')
+    "\\"
+    >>> print simplejson.dumps({"c": 0, "b": 0, "a": 0}, sort_keys=True)
+    {"a": 0, "b": 0, "c": 0}
+    >>> from StringIO import StringIO
+    >>> io = StringIO()
+    >>> simplejson.dump(['streaming API'], io)
+    >>> io.getvalue()
+    '["streaming API"]'
+
+Compact encoding::
+
+    >>> import simplejson
+    >>> compact = simplejson.dumps([1,2,3,{'4': 5, '6': 7}], separators=(',',':'))
+    >>> # Can't assume dict ordering
+    >>> compact in ('[1,2,3,{"4":5,"6":7}]', '[1,2,3,{"6":7,"4":5}]')
+    True
+
+Pretty printing (using repr() because of extraneous whitespace in the output)::
+
+    >>> import simplejson
+    >>> print repr(simplejson.dumps({'4': 5, '6': 7}, sort_keys=True, indent=4))
+    '{\n    "4": 5, \n    "6": 7\n}'
+
+Decoding JSON::
+
+    >>> import simplejson
+    >>> simplejson.loads('["foo", {"bar":["baz", null, 1.0, 2]}]') == ["foo", {"bar":["baz", None, 1.0, 2]}]
+    True
+    >>> simplejson.loads('"\\"foo\\bar"') == '"foo\x08ar'
+    True
+    >>> from StringIO import StringIO
+    >>> io = StringIO('["streaming API"]')
+    >>> simplejson.load(io) == ["streaming API"]
+    True
+
+Specializing JSON object decoding::
+
+    >>> import simplejson
+    >>> def as_complex(dct):
+    ...     if '__complex__' in dct:
+    ...         return complex(dct['real'], dct['imag'])
+    ...     return dct
+    ...
+    >>> simplejson.loads('{"__complex__": true, "real": 1, "imag": 2}',
+    ...     object_hook=as_complex)
+    (1+2j)
+    >>> from decimal import Decimal
+    >>> simplejson.loads('1.1', parse_float=Decimal) == Decimal("1.1")
+    True
+
+Extending JSONEncoder::
+
+    >>> import simplejson
+    >>> class ComplexEncoder(simplejson.JSONEncoder):
+    ...     def default(self, obj):
+    ...         if isinstance(obj, complex):
+    ...             return [obj.real, obj.imag]
+    ...         return simplejson.JSONEncoder.default(self, obj)
+    ...
+    >>> dumps(2 + 1j, cls=ComplexEncoder)
+    '[2.0, 1.0]'
+    >>> ComplexEncoder().encode(2 + 1j)
+    '[2.0, 1.0]'
+    >>> ''.join(ComplexEncoder().iterencode(2 + 1j))
+    '[2.0, 1.0]'
+
+
+Using simplejson from the shell to validate and
+pretty-print::
+
+    $ echo '{"json":"obj"}' | python -msimplejson.tool
+    {
+        "json": "obj"
+    }
+    $ echo '{ 1.2:3.4}' | python -msimplejson.tool
+    Expecting property name: line 1 column 2 (char 2)
+"""
+__version__ = '2.0.4'
+__all__ = [
+    'dump', 'dumps', 'load', 'loads',
+    'JSONDecoder', 'JSONEncoder',
+]
+
+from decoder import JSONDecoder
+from encoder import JSONEncoder
+
+_default_encoder = JSONEncoder(
+    skipkeys=False,
+    ensure_ascii=True,
+    check_circular=True,
+    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):
+    """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 ``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
+    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
+    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 ``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.
+
+    To use a custom ``JSONEncoder`` subclass (e.g. one that overrides the
+    ``.default()`` method to serialize additional types), specify it with
+    the ``cls`` kwarg.
+
+    """
+    # cached encoder
+    if (skipkeys is False and ensure_ascii is True and
+        check_circular is True and allow_nan is True and
+        cls is None and indent is None and separators is None and
+        encoding == 'utf-8' and 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,
+            default=default, **kw).iterencode(obj)
+    # could accelerate with writelines in some versions of Python, at
+    # a debuggability cost
+    for chunk in iterable:
+        fp.write(chunk)
+
+
+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):
+    """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 ``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
+    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
+    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 ``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.
+
+    To use a custom ``JSONEncoder`` subclass (e.g. one that overrides the
+    ``.default()`` method to serialize additional types), specify it with
+    the ``cls`` kwarg.
+
+    """
+    # cached encoder
+    if (skipkeys is False and ensure_ascii is True and
+        check_circular is True and allow_nan is True and
+        cls is None and indent is None and separators is None and
+        encoding == 'utf-8' and 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,
+        **kw).encode(obj)
+
+
+_default_decoder = JSONDecoder(encoding=None, object_hook=None)
+
+
+def load(fp, encoding=None, cls=None, object_hook=None, parse_float=None,
+        parse_int=None, parse_constant=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()``
+
+    ``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
+    can be used to implement custom decoders (e.g. JSON-RPC class hinting).
+
+    To use a custom ``JSONDecoder`` subclass, specify it with the ``cls``
+    kwarg.
+
+    """
+    return loads(fp.read(),
+        encoding=encoding, cls=cls, object_hook=object_hook,
+        parse_float=parse_float, parse_int=parse_int,
+        parse_constant=parse_constant, **kw)
+
+
+def loads(s, encoding=None, cls=None, object_hook=None, parse_float=None,
+        parse_int=None, parse_constant=None, **kw):
+    """Deserialize ``s`` (a ``str`` or ``unicode`` 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
+    can be used to implement custom decoders (e.g. JSON-RPC class hinting).
+
+    ``parse_float``, if specified, will be called with the string
+    of every JSON float to be decoded. By default this is equivalent to
+    float(num_str). This can be used to use another datatype or parser
+    for JSON floats (e.g. decimal.Decimal).
+
+    ``parse_int``, if specified, will be called with the string
+    of every JSON int to be decoded. By default this is equivalent to
+    int(num_str). This can be used to use another datatype or parser
+    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.
+    This can be used to raise an exception if invalid JSON numbers
+    are encountered.
+
+    To use a custom ``JSONDecoder`` subclass, specify it with the ``cls``
+    kwarg.
+
+    """
+    if (cls is None and encoding is None and object_hook is None and
+            parse_int is None and parse_float is None and
+            parse_constant is None and not kw):
+        return _default_decoder.decode(s)
+    if cls is None:
+        cls = JSONDecoder
+    if object_hook is not None:
+        kw['object_hook'] = object_hook
+    if parse_float is not None:
+        kw['parse_float'] = parse_float
+    if parse_int is not None:
+        kw['parse_int'] = parse_int
+    if parse_constant is not None:
+        kw['parse_constant'] = parse_constant
+    return cls(encoding=encoding, **kw).decode(s)

File simplejson/_speedups.c

+#include "Python.h"
+#include "structmember.h"
+#if PY_VERSION_HEX < 0x02060000 && !defined(Py_TYPE)
+#define Py_TYPE(ob)     (((PyObject*)(ob))->ob_type)
+#endif
+#if PY_VERSION_HEX < 0x02050000 && !defined(PY_SSIZE_T_MIN)
+typedef int Py_ssize_t;
+#define PY_SSIZE_T_MAX INT_MAX
+#define PY_SSIZE_T_MIN INT_MIN
+#define PyInt_FromSsize_t PyInt_FromLong
+#define PyInt_AsSsize_t PyInt_AsLong
+#endif
+#ifndef Py_IS_FINITE
+#define Py_IS_FINITE(X) (!Py_IS_INFINITY(X) && !Py_IS_NAN(X))
+#endif
+
+#ifdef __GNUC__
+#define UNUSED __attribute__((__unused__))
+#else
+#define UNUSED
+#endif
+
+#define DEFAULT_ENCODING "utf-8"
+
+#define PyScanner_Check(op) PyObject_TypeCheck(op, &PyScannerType)
+#define PyScanner_CheckExact(op) (Py_TYPE(op) == &PyScannerType)
+#define PyEncoder_Check(op) PyObject_TypeCheck(op, &PyEncoderType)
+#define PyEncoder_CheckExact(op) (Py_TYPE(op) == &PyEncoderType)
+
+static PyTypeObject PyScannerType;
+static PyTypeObject PyEncoderType;
+
+typedef struct _PyScannerObject {
+    PyObject_HEAD
+    PyObject *encoding;
+    PyObject *strict;
+    PyObject *object_hook;
+    PyObject *parse_float;
+    PyObject *parse_int;
+    PyObject *parse_constant;
+} PyScannerObject;
+
+static PyMemberDef scanner_members[] = {
+    {"encoding", T_OBJECT, offsetof(PyScannerObject, encoding), READONLY, "encoding"},
+    {"strict", T_OBJECT, offsetof(PyScannerObject, strict), READONLY, "strict"},
+    {"object_hook", T_OBJECT, offsetof(PyScannerObject, object_hook), READONLY, "object_hook"},
+    {"parse_float", T_OBJECT, offsetof(PyScannerObject, parse_float), READONLY, "parse_float"},
+    {"parse_int", T_OBJECT, offsetof(PyScannerObject, parse_int), READONLY, "parse_int"},
+    {"parse_constant", T_OBJECT, offsetof(PyScannerObject, parse_constant), READONLY, "parse_constant"},
+    {NULL}
+};
+
+typedef struct _PyEncoderObject {
+    PyObject_HEAD
+    PyObject *markers;
+    PyObject *defaultfn;
+    PyObject *encoder;
+    PyObject *indent;
+    PyObject *key_separator;
+    PyObject *item_separator;
+    PyObject *sort_keys;
+    PyObject *skipkeys;
+    int fast_encode;
+    int allow_nan;
+} PyEncoderObject;
+
+static PyMemberDef encoder_members[] = {
+    {"markers", T_OBJECT, offsetof(PyEncoderObject, markers), READONLY, "markers"},
+    {"default", T_OBJECT, offsetof(PyEncoderObject, defaultfn), READONLY, "default"},
+    {"encoder", T_OBJECT, offsetof(PyEncoderObject, encoder), READONLY, "encoder"},
+    {"indent", T_OBJECT, offsetof(PyEncoderObject, indent), READONLY, "indent"},
+    {"key_separator", T_OBJECT, offsetof(PyEncoderObject, key_separator), READONLY, "key_separator"},
+    {"item_separator", T_OBJECT, offsetof(PyEncoderObject, item_separator), READONLY, "item_separator"},
+    {"sort_keys", T_OBJECT, offsetof(PyEncoderObject, sort_keys), READONLY, "sort_keys"},
+    {"skipkeys", T_OBJECT, offsetof(PyEncoderObject, skipkeys), READONLY, "skipkeys"},
+    {NULL}
+};
+
+static Py_ssize_t
+ascii_escape_char(Py_UNICODE c, char *output, Py_ssize_t chars);
+static PyObject *
+ascii_escape_unicode(PyObject *pystr);
+static PyObject *
+ascii_escape_str(PyObject *pystr);
+static PyObject *
+py_encode_basestring_ascii(PyObject* self UNUSED, PyObject *pystr);
+void init_speedups(void);
+static PyObject *
+scan_once_str(PyScannerObject *s, PyObject *pystr, Py_ssize_t idx, Py_ssize_t *next_idx_ptr);
+static PyObject *
+scan_once_unicode(PyScannerObject *s, PyObject *pystr, Py_ssize_t idx, Py_ssize_t *next_idx_ptr);
+static PyObject *
+_build_rval_index_tuple(PyObject *rval, Py_ssize_t idx);
+static int
+scanner_init(PyObject *self, PyObject *args, PyObject *kwds);
+static void
+scanner_dealloc(PyObject *self);
+static int
+encoder_init(PyObject *self, PyObject *args, PyObject *kwds);
+static void
+encoder_dealloc(PyObject *self);
+static int
+encoder_listencode_list(PyEncoderObject *s, PyObject *rval, PyObject *seq, Py_ssize_t indent_level);
+static int
+encoder_listencode_obj(PyEncoderObject *s, PyObject *rval, PyObject *obj, Py_ssize_t indent_level);
+static int
+encoder_listencode_dict(PyEncoderObject *s, PyObject *rval, PyObject *dct, Py_ssize_t indent_level);
+static PyObject *
+_encoded_const(PyObject *const);
+static void
+raise_errmsg(char *msg, PyObject *s, Py_ssize_t end);
+static PyObject *
+encoder_encode_string(PyEncoderObject *s, PyObject *obj);
+static int
+_convertPyInt_AsSsize_t(PyObject *o, Py_ssize_t *size_ptr);
+static PyObject *
+_convertPyInt_FromSsize_t(Py_ssize_t *size_ptr);
+static PyObject *
+encoder_encode_float(PyEncoderObject *s, PyObject *obj);
+
+#define S_CHAR(c) (c >= ' ' && c <= '~' && c != '\\' && c != '"')
+#define IS_WHITESPACE(c) (((c) == ' ') || ((c) == '\t') || ((c) == '\n') || ((c) == '\r'))
+
+#define MIN_EXPANSION 6
+#ifdef Py_UNICODE_WIDE
+#define MAX_EXPANSION (2 * MIN_EXPANSION)
+#else
+#define MAX_EXPANSION MIN_EXPANSION
+#endif
+
+static int
+_convertPyInt_AsSsize_t(PyObject *o, Py_ssize_t *size_ptr)
+{
+    *size_ptr = PyInt_AsSsize_t(o);
+    if (*size_ptr == -1 && PyErr_Occurred());
+        return 1;
+    return 0;
+}
+
+static PyObject *
+_convertPyInt_FromSsize_t(Py_ssize_t *size_ptr)
+{
+    return PyInt_FromSsize_t(*size_ptr);
+}
+
+static Py_ssize_t
+ascii_escape_char(Py_UNICODE c, char *output, Py_ssize_t chars)
+{
+    output[chars++] = '\\';
+    switch (c) {
+        case '\\': output[chars++] = (char)c; break;
+        case '"': output[chars++] = (char)c; break;
+        case '\b': output[chars++] = 'b'; break;
+        case '\f': output[chars++] = 'f'; break;
+        case '\n': output[chars++] = 'n'; break;
+        case '\r': output[chars++] = 'r'; break;
+        case '\t': output[chars++] = 't'; break;
+        default:
+#ifdef Py_UNICODE_WIDE
+            if (c >= 0x10000) {
+                /* UTF-16 surrogate pair */
+                Py_UNICODE v = c - 0x10000;
+                c = 0xd800 | ((v >> 10) & 0x3ff);
+                output[chars++] = 'u';
+                output[chars++] = "0123456789abcdef"[(c >> 12) & 0xf];
+                output[chars++] = "0123456789abcdef"[(c >>  8) & 0xf];
+                output[chars++] = "0123456789abcdef"[(c >>  4) & 0xf];
+                output[chars++] = "0123456789abcdef"[(c      ) & 0xf];
+                c = 0xdc00 | (v & 0x3ff);
+                output[chars++] = '\\';
+            }
+#endif
+            output[chars++] = 'u';
+            output[chars++] = "0123456789abcdef"[(c >> 12) & 0xf];
+            output[chars++] = "0123456789abcdef"[(c >>  8) & 0xf];
+            output[chars++] = "0123456789abcdef"[(c >>  4) & 0xf];
+            output[chars++] = "0123456789abcdef"[(c      ) & 0xf];
+    }
+    return chars;
+}
+
+static PyObject *
+ascii_escape_unicode(PyObject *pystr)
+{
+    Py_ssize_t i;
+    Py_ssize_t input_chars;
+    Py_ssize_t output_size;
+    Py_ssize_t chars;
+    PyObject *rval;
+    char *output;
+    Py_UNICODE *input_unicode;
+
+    input_chars = PyUnicode_GET_SIZE(pystr);
+    input_unicode = PyUnicode_AS_UNICODE(pystr);
+
+    /* One char input can be up to 6 chars output, estimate 4 of these */
+    output_size = 2 + (MIN_EXPANSION * 4) + input_chars;
+    rval = PyString_FromStringAndSize(NULL, output_size);
+    if (rval == NULL) {
+        return NULL;
+    }
+    output = PyString_AS_STRING(rval);
+    chars = 0;
+    output[chars++] = '"';
+    for (i = 0; i < input_chars; i++) {
+        Py_UNICODE c = input_unicode[i];
+        if (S_CHAR(c)) {
+            output[chars++] = (char)c;
+        }
+        else {
+            chars = ascii_escape_char(c, output, chars);
+        }
+        if (output_size - chars < (1 + MAX_EXPANSION)) {
+            /* There's more than four, so let's resize by a lot */
+            output_size *= 2;
+            /* This is an upper bound */
+            if (output_size > 2 + (input_chars * MAX_EXPANSION)) {
+                output_size = 2 + (input_chars * MAX_EXPANSION);
+            }
+            if (_PyString_Resize(&rval, output_size) == -1) {
+                return NULL;
+            }
+            output = PyString_AS_STRING(rval);
+        }
+    }
+    output[chars++] = '"';
+    if (_PyString_Resize(&rval, chars) == -1) {
+        return NULL;
+    }
+    return rval;
+}
+
+static PyObject *
+ascii_escape_str(PyObject *pystr)
+{
+    Py_ssize_t i;
+    Py_ssize_t input_chars;
+    Py_ssize_t output_size;
+    Py_ssize_t chars;
+    PyObject *rval;
+    char *output;
+    char *input_str;
+
+    input_chars = PyString_GET_SIZE(pystr);
+    input_str = PyString_AS_STRING(pystr);
+
+    /* Fast path for a string that's already ASCII */
+    for (i = 0; i < input_chars; i++) {
+        Py_UNICODE c = (Py_UNICODE)(unsigned char)input_str[i];
+        if (!S_CHAR(c)) {
+            /* If we have to escape something, scan the string for unicode */
+            Py_ssize_t j;
+            for (j = i; j < input_chars; j++) {
+                c = (Py_UNICODE)(unsigned char)input_str[j];
+                if (c > 0x7f) {
+                    /* We hit a non-ASCII character, bail to unicode mode */
+                    PyObject *uni;
+                    uni = PyUnicode_DecodeUTF8(input_str, input_chars, "strict");
+                    if (uni == NULL) {
+                        return NULL;
+                    }
+                    rval = ascii_escape_unicode(uni);
+                    Py_DECREF(uni);
+                    return rval;
+                }
+            }
+            break;
+        }
+    }
+
+    if (i == input_chars) {
+        /* Input is already ASCII */
+        output_size = 2 + input_chars;
+    }
+    else {
+        /* One char input can be up to 6 chars output, estimate 4 of these */
+        output_size = 2 + (MIN_EXPANSION * 4) + input_chars;
+    }
+    rval = PyString_FromStringAndSize(NULL, output_size);
+    if (rval == NULL) {
+        return NULL;
+    }
+    output = PyString_AS_STRING(rval);
+    output[0] = '"';
+    
+    /* We know that everything up to i is ASCII already */
+    chars = i + 1;
+    memcpy(&output[1], input_str, i);
+
+    for (; i < input_chars; i++) {
+        Py_UNICODE c = (Py_UNICODE)(unsigned char)input_str[i];
+        if (S_CHAR(c)) {
+            output[chars++] = (char)c;
+        }
+        else {
+            chars = ascii_escape_char(c, output, chars);
+        }
+        /* An ASCII char can't possibly expand to a surrogate! */
+        if (output_size - chars < (1 + MIN_EXPANSION)) {
+            /* There's more than four, so let's resize by a lot */
+            output_size *= 2;
+            if (output_size > 2 + (input_chars * MIN_EXPANSION)) {
+                output_size = 2 + (input_chars * MIN_EXPANSION);
+            }
+            if (_PyString_Resize(&rval, output_size) == -1) {
+                return NULL;
+            }
+            output = PyString_AS_STRING(rval);
+        }
+    }
+    output[chars++] = '"';
+    if (_PyString_Resize(&rval, chars) == -1) {
+        return NULL;
+    }
+    return rval;
+}
+
+static void
+raise_errmsg(char *msg, PyObject *s, Py_ssize_t end)
+{
+    static PyObject *errmsg_fn = NULL;
+    PyObject *pymsg;
+    if (errmsg_fn == NULL) {
+        PyObject *decoder = PyImport_ImportModule("simplejson.decoder");
+        if (decoder == NULL)
+            return;
+        errmsg_fn = PyObject_GetAttrString(decoder, "errmsg");
+        Py_DECREF(decoder);
+        if (errmsg_fn == NULL)
+            return;
+    }
+    pymsg = PyObject_CallFunction(errmsg_fn, "(zOO&)", msg, s, _convertPyInt_FromSsize_t, &end);
+    if (pymsg) {
+        PyErr_SetObject(PyExc_ValueError, pymsg);
+        Py_DECREF(pymsg);
+    }
+}
+
+static PyObject *
+join_list_string(PyObject *lst)
+{
+    static PyObject *joinfn = NULL;
+    if (joinfn == NULL) {
+        PyObject *ustr = PyString_FromStringAndSize(NULL, 0);
+        if (ustr == NULL)
+            return NULL;
+        
+        joinfn = PyObject_GetAttrString(ustr, "join");
+        Py_DECREF(ustr);
+        if (joinfn == NULL)
+            return NULL;
+    }
+    return PyObject_CallFunctionObjArgs(joinfn, lst, NULL);
+}
+
+static PyObject *
+_build_rval_index_tuple(PyObject *rval, Py_ssize_t idx) {
+    PyObject *tpl;
+    PyObject *pyidx;
+    /*
+    steal a reference to rval, returns (rval, idx)
+    */
+    if (rval == NULL) {
+        return NULL;
+    }
+    pyidx = PyInt_FromSsize_t(idx);
+    if (pyidx == NULL) {
+        Py_DECREF(rval);
+        return NULL;
+    }
+    tpl = PyTuple_New(2);
+    if (tpl == NULL) {
+        Py_DECREF(pyidx);
+        Py_DECREF(rval);
+        return NULL;
+    }
+    PyTuple_SET_ITEM(tpl, 0, rval);
+    PyTuple_SET_ITEM(tpl, 1, pyidx);
+    return tpl;
+}
+
+static PyObject *
+scanstring_str(PyObject *pystr, Py_ssize_t end, char *encoding, int strict, Py_ssize_t *next_end_ptr)
+{
+    PyObject *rval;
+    Py_ssize_t len = PyString_GET_SIZE(pystr);
+    Py_ssize_t begin = end - 1;
+    Py_ssize_t next = begin;
+    int has_unicode = 0;
+    char *buf = PyString_AS_STRING(pystr);
+    PyObject *chunks = PyList_New(0);
+    if (chunks == NULL) {
+        goto bail;
+    }
+    if (end < 0 || len <= end) {
+        PyErr_SetString(PyExc_ValueError, "end is out of bounds");
+        goto bail;
+    }
+    while (1) {
+        /* Find the end of the string or the next escape */
+        Py_UNICODE c = 0;
+        PyObject *chunk = NULL;
+        for (next = end; next < len; next++) {
+            c = (unsigned char)buf[next];
+            if (c == '"' || c == '\\') {
+                break;
+            }
+            else if (strict && c <= 0x1f) {
+                raise_errmsg("Invalid control character at", pystr, next);
+                goto bail;
+            }
+            else if (c > 0x7f) {
+                has_unicode = 1;
+            }
+        }
+        if (!(c == '"' || c == '\\')) {
+            raise_errmsg("Unterminated string starting at", pystr, begin);
+            goto bail;
+        }
+        /* Pick up this chunk if it's not zero length */
+        if (next != end) {
+            PyObject *strchunk = PyString_FromStringAndSize(&buf[end], next - end);
+            if (strchunk == NULL) {
+                goto bail;
+            }
+            if (has_unicode) {
+                chunk = PyUnicode_FromEncodedObject(strchunk, encoding, NULL);
+                Py_DECREF(strchunk);
+                if (chunk == NULL) {
+                    goto bail;
+                }
+            }
+            else {
+                chunk = strchunk;
+            }
+            if (PyList_Append(chunks, chunk)) {
+                Py_DECREF(chunk);
+                goto bail;
+            }
+            Py_DECREF(chunk);
+        }
+        next++;
+        if (c == '"') {
+            end = next;
+            break;
+        }
+        if (next == len) {
+            raise_errmsg("Unterminated string starting at", pystr, begin);
+            goto bail;
+        }
+        c = buf[next];
+        if (c != 'u') {
+            /* Non-unicode backslash escapes */
+            end = next + 1;
+            switch (c) {
+                case '"': break;
+                case '\\': break;
+                case '/': break;
+                case 'b': c = '\b'; break;
+                case 'f': c = '\f'; break;
+                case 'n': c = '\n'; break;
+                case 'r': c = '\r'; break;
+                case 't': c = '\t'; break;
+                default: c = 0;
+            }
+            if (c == 0) {
+                raise_errmsg("Invalid \\escape", pystr, end - 2);
+                goto bail;
+            }
+        }
+        else {
+            c = 0;
+            next++;
+            end = next + 4;
+            if (end >= len) {
+                raise_errmsg("Invalid \\uXXXX escape", pystr, next - 1);
+                goto bail;
+            }
+            /* Decode 4 hex digits */
+            for (; next < end; next++) {
+                Py_UNICODE digit = buf[next];
+                c <<= 4;
+                switch (digit) {
+                    case '0': case '1': case '2': case '3': case '4':
+                    case '5': case '6': case '7': case '8': case '9':
+                        c |= (digit - '0'); break;
+                    case 'a': case 'b': case 'c': case 'd': case 'e':
+                    case 'f':
+                        c |= (digit - 'a' + 10); break;
+                    case 'A': case 'B': case 'C': case 'D': case 'E':
+                    case 'F':
+                        c |= (digit - 'A' + 10); break;
+                    default:
+                        raise_errmsg("Invalid \\uXXXX escape", pystr, end - 5);
+                        goto bail;
+                }
+            }
+#ifdef Py_UNICODE_WIDE
+            /* Surrogate pair */
+            if ((c & 0xfc00) == 0xd800) {
+                Py_UNICODE c2 = 0;
+                if (end + 6 >= len) {
+                    raise_errmsg("Unpaired high surrogate", pystr, end - 5);
+                    goto bail;
+                }
+                if (buf[next++] != '\\' || buf[next++] != 'u') {
+                    raise_errmsg("Unpaired high surrogate", pystr, end - 5);
+                    goto bail;
+                }
+                end += 6;
+                /* Decode 4 hex digits */
+                for (; next < end; next++) {
+                    c2 <<= 4;
+                    Py_UNICODE digit = buf[next];
+                    switch (digit) {
+                        case '0': case '1': case '2': case '3': case '4':
+                        case '5': case '6': case '7': case '8': case '9':
+                            c2 |= (digit - '0'); break;
+                        case 'a': case 'b': case 'c': case 'd': case 'e':
+                        case 'f':
+                            c2 |= (digit - 'a' + 10); break;
+                        case 'A': case 'B': case 'C': case 'D': case 'E':
+                        case 'F':
+                            c2 |= (digit - 'A' + 10); break;
+                        default:
+                            raise_errmsg("Invalid \\uXXXX escape", pystr, end - 5);
+                            goto bail;
+                    }
+                }
+                if ((c2 & 0xfc00) != 0xdc00) {
+                    raise_errmsg("Unpaired high surrogate", pystr, end - 5);
+                    goto bail;
+                }
+                c = 0x10000 + (((c - 0xd800) << 10) | (c2 - 0xdc00));
+            }
+            else if ((c & 0xfc00) == 0xdc00) {
+                raise_errmsg("Unpaired low surrogate", pystr, end - 5);
+                goto bail;
+            }
+#endif
+        }
+        if (c > 0x7f) {
+            has_unicode = 1;
+        }
+        if (has_unicode) {
+            chunk = PyUnicode_FromUnicode(&c, 1);
+            if (chunk == NULL) {
+                goto bail;
+            }
+        }
+        else {
+            char c_char = Py_CHARMASK(c);
+            chunk = PyString_FromStringAndSize(&c_char, 1);
+            if (chunk == NULL) {
+                goto bail;
+            }
+        }
+        if (PyList_Append(chunks, chunk)) {
+            Py_DECREF(chunk);
+            goto bail;
+        }
+        Py_DECREF(chunk);
+    }
+
+    rval = join_list_string(chunks);
+    if (rval == NULL) {
+        goto bail;
+    }
+    Py_CLEAR(chunks);
+    *next_end_ptr = end;
+    return rval;
+bail:
+    *next_end_ptr = -1;
+    Py_XDECREF(chunks);
+    return NULL;
+}
+
+
+static PyObject *
+scanstring_unicode(PyObject *pystr, Py_ssize_t end, int strict, Py_ssize_t *next_end_ptr)
+{
+    PyObject *rval;
+    Py_ssize_t len = PyUnicode_GET_SIZE(pystr);
+    Py_ssize_t begin = end - 1;
+    Py_ssize_t next = begin;
+    const Py_UNICODE *buf = PyUnicode_AS_UNICODE(pystr);
+    PyObject *chunks = PyList_New(0);
+    if (chunks == NULL) {
+        goto bail;
+    }
+    if (end < 0 || len <= end) {
+        PyErr_SetString(PyExc_ValueError, "end is out of bounds");
+        goto bail;
+    }
+    while (1) {
+        /* Find the end of the string or the next escape */
+        Py_UNICODE c = 0;
+        PyObject *chunk = NULL;
+        for (next = end; next < len; next++) {
+            c = buf[next];
+            if (c == '"' || c == '\\') {
+                break;
+            }
+            else if (strict && c <= 0x1f) {
+                raise_errmsg("Invalid control character at", pystr, next);
+                goto bail;
+            }
+        }
+        if (!(c == '"' || c == '\\')) {
+            raise_errmsg("Unterminated string starting at", pystr, begin);
+            goto bail;
+        }
+        /* Pick up this chunk if it's not zero length */
+        if (next != end) {
+            chunk = PyUnicode_FromUnicode(&buf[end], next - end);
+            if (chunk == NULL) {
+                goto bail;
+            }
+            if (PyList_Append(chunks, chunk)) {
+                Py_DECREF(chunk);
+                goto bail;
+            }
+            Py_DECREF(chunk);
+        }
+        next++;
+        if (c == '"') {
+            end = next;
+            break;
+        }
+        if (next == len) {
+            raise_errmsg("Unterminated string starting at", pystr, begin);
+            goto bail;
+        }
+        c = buf[next];
+        if (c != 'u') {
+            /* Non-unicode backslash escapes */
+            end = next + 1;
+            switch (c) {
+                case '"': break;
+                case '\\': break;
+                case '/': break;
+                case 'b': c = '\b'; break;
+                case 'f': c = '\f'; break;
+                case 'n': c = '\n'; break;
+                case 'r': c = '\r'; break;
+                case 't': c = '\t'; break;
+                default: c = 0;
+            }
+            if (c == 0) {
+                raise_errmsg("Invalid \\escape", pystr, end - 2);
+                goto bail;
+            }
+        }
+        else {
+            c = 0;
+            next++;
+            end = next + 4;
+            if (end >= len) {
+                raise_errmsg("Invalid \\uXXXX escape", pystr, next - 1);
+                goto bail;
+            }
+            /* Decode 4 hex digits */
+            for (; next < end; next++) {
+                Py_UNICODE digit = buf[next];
+                c <<= 4;
+                switch (digit) {
+                    case '0': case '1': case '2': case '3': case '4':
+                    case '5': case '6': case '7': case '8': case '9':
+                        c |= (digit - '0'); break;
+                    case 'a': case 'b': case 'c': case 'd': case 'e':
+                    case 'f':
+                        c |= (digit - 'a' + 10); break;
+                    case 'A': case 'B': case 'C': case 'D': case 'E':
+                    case 'F':
+                        c |= (digit - 'A' + 10); break;
+                    default:
+                        raise_errmsg("Invalid \\uXXXX escape", pystr, end - 5);
+                        goto bail;
+                }
+            }
+#ifdef Py_UNICODE_WIDE
+            /* Surrogate pair */
+            if ((c & 0xfc00) == 0xd800) {
+                Py_UNICODE c2 = 0;
+                if (end + 6 >= len) {
+                    raise_errmsg("Unpaired high surrogate", pystr, end - 5);
+                    goto bail;
+                }
+                if (buf[next++] != '\\' || buf[next++] != 'u') {
+                    raise_errmsg("Unpaired high surrogate", pystr, end - 5);
+                    goto bail;
+                }
+                end += 6;
+                /* Decode 4 hex digits */
+                for (; next < end; next++) {
+                    c2 <<= 4;
+                    Py_UNICODE digit = buf[next];
+                    switch (digit) {
+                        case '0': case '1': case '2': case '3': case '4':
+                        case '5': case '6': case '7': case '8': case '9':
+                            c2 |= (digit - '0'); break;
+                        case 'a': case 'b': case 'c': case 'd': case 'e':
+                        case 'f':
+                            c2 |= (digit - 'a' + 10); break;
+                        case 'A': case 'B': case 'C': case 'D': case 'E':
+                        case 'F':
+                            c2 |= (digit - 'A' + 10); break;
+                        default:
+                            raise_errmsg("Invalid \\uXXXX escape", pystr, end - 5);
+                            goto bail;
+                    }
+                }
+                if ((c2 & 0xfc00) != 0xdc00) {
+                    raise_errmsg("Unpaired high surrogate", pystr, end - 5);
+                    goto bail;
+                }
+                c = 0x10000 + (((c - 0xd800) << 10) | (c2 - 0xdc00));
+            }
+            else if ((c & 0xfc00) == 0xdc00) {
+                raise_errmsg("Unpaired low surrogate", pystr, end - 5);
+                goto bail;
+            }
+#endif
+        }
+        chunk = PyUnicode_FromUnicode(&c, 1);
+        if (chunk == NULL) {
+            goto bail;
+        }
+        if (PyList_Append(chunks, chunk)) {
+            Py_DECREF(chunk);
+            goto bail;
+        }
+        Py_DECREF(chunk);
+    }
+
+    rval = join_list_string(chunks);
+    if (rval == NULL) {
+        goto bail;
+    }
+    Py_DECREF(chunks);
+    *next_end_ptr = end;
+    return rval;
+bail:
+    *next_end_ptr = -1;
+    Py_XDECREF(chunks);
+    return NULL;
+}
+
+PyDoc_STRVAR(pydoc_scanstring,
+    "scanstring(basestring, end, encoding) -> (str, end)\n"
+    "\n"
+    "..."
+);
+
+static PyObject *
+py_scanstring(PyObject* self UNUSED, PyObject *args)
+{
+    PyObject *pystr;
+    PyObject *rval;
+    Py_ssize_t end;
+    Py_ssize_t next_end = -1;
+    char *encoding = NULL;
+    int strict = 0;
+    if (!PyArg_ParseTuple(args, "OO&|zi:scanstring", &pystr, _convertPyInt_AsSsize_t, &end, &encoding, &strict)) {
+        return NULL;
+    }
+    if (encoding == NULL) {
+        encoding = DEFAULT_ENCODING;
+    }
+    if (PyString_Check(pystr)) {
+        rval = scanstring_str(pystr, end, encoding, strict, &next_end);
+    }
+    else if (PyUnicode_Check(pystr)) {
+        rval = scanstring_unicode(pystr, end, strict, &next_end);
+    }
+    else {
+        PyErr_Format(PyExc_TypeError,
+                     "first argument must be a string, not %.80s",
+                     Py_TYPE(pystr)->tp_name);
+        return NULL;
+    }
+    return _build_rval_index_tuple(rval, next_end);
+}
+
+PyDoc_STRVAR(pydoc_encode_basestring_ascii,
+    "encode_basestring_ascii(basestring) -> str\n"
+    "\n"
+    "..."
+);
+
+static PyObject *
+py_encode_basestring_ascii(PyObject* self UNUSED, PyObject *pystr)
+{
+    /* METH_O */
+    if (PyString_Check(pystr)) {
+        return ascii_escape_str(pystr);
+    }
+    else if (PyUnicode_Check(pystr)) {
+        return ascii_escape_unicode(pystr);
+    }
+    else {
+        PyErr_Format(PyExc_TypeError,
+                     "first argument must be a string, not %.80s",
+                     Py_TYPE(pystr)->tp_name);
+        return NULL;
+    }
+}
+
+static void
+scanner_dealloc(PyObject *self)
+{
+    PyScannerObject *s;
+    assert(PyScanner_Check(self));
+    s = (PyScannerObject *)self;
+    Py_CLEAR(s->encoding);
+    Py_CLEAR(s->strict);
+    Py_CLEAR(s->object_hook);
+    Py_CLEAR(s->parse_float);
+    Py_CLEAR(s->parse_int);
+    Py_CLEAR(s->parse_constant);
+    self->ob_type->tp_free(self);
+}
+
+static PyObject *
+_parse_object_str(PyScannerObject *s, PyObject *pystr, Py_ssize_t idx, Py_ssize_t *next_idx_ptr) {
+    char *str = PyString_AS_STRING(pystr);
+    Py_ssize_t end_idx = PyString_GET_SIZE(pystr) - 1;
+    PyObject *rval = PyDict_New();
+    PyObject *key = NULL;
+    PyObject *val = NULL;
+    char *encoding = PyString_AS_STRING(s->encoding);
+    int strict = PyObject_IsTrue(s->strict);
+    Py_ssize_t next_idx;
+    if (rval == NULL)
+        return NULL;
+
+    /* skip whitespace after { */
+    while (idx <= end_idx && IS_WHITESPACE(str[idx])) idx++;
+
+    /* only loop if the object is non-empty */
+    if (idx <= end_idx && str[idx] != '}') {
+        while (idx <= end_idx) {
+            /* read key */
+            if (str[idx] != '"') {
+                raise_errmsg("Expecting property name", pystr, idx);
+                goto bail;
+            }
+            key = scanstring_str(pystr, idx + 1, encoding, strict, &next_idx);
+            if (key == NULL)
+                goto bail;
+            idx = next_idx;
+            
+            /* skip whitespace between key and : delimiter, read :, skip whitespace */
+            while (idx <= end_idx && IS_WHITESPACE(str[idx])) idx++;
+            if (idx > end_idx || str[idx] != ':') {
+                raise_errmsg("Expecting : delimiter", pystr, idx);
+                goto bail;
+            }
+            idx++;
+            while (idx <= end_idx && IS_WHITESPACE(str[idx])) idx++;
+            
+            /* read any JSON data type */
+            val = scan_once_str(s, pystr, idx, &next_idx);
+            if (val == NULL)
+                goto bail;
+
+            if (PyDict_SetItem(rval, key, val) == -1)
+                goto bail;
+
+            Py_CLEAR(key);
+            Py_CLEAR(val);
+            idx = next_idx;
+            
+            /* skip whitespace before } or , */
+            while (idx <= end_idx && IS_WHITESPACE(str[idx])) idx++;
+
+            /* bail if the object is closed or we didn't get the , delimiter */
+            if (idx > end_idx) break;
+            if (str[idx] == '}') {
+                break;
+            }
+            else if (str[idx] != ',') {
+                raise_errmsg("Expecting , delimiter", pystr, idx);
+                goto bail;
+            }
+            idx++;
+
+            /* skip whitespace after , delimiter */
+            while (idx <= end_idx && IS_WHITESPACE(str[idx])) idx++;
+        }
+    }
+    /* verify that idx < end_idx, str[idx] should be '}' */
+    if (idx > end_idx || str[idx] != '}') {
+        raise_errmsg("Expecting object", pystr, end_idx);
+        goto bail;
+    }
+    /* if object_hook is not None: rval = object_hook(rval) */
+    if (s->object_hook != Py_None) {
+        val = PyObject_CallFunctionObjArgs(s->object_hook, rval, NULL);
+        if (val == NULL)
+            goto bail;
+        Py_DECREF(rval);
+        rval = val;
+        val = NULL;
+    }
+    *next_idx_ptr = idx + 1;
+    return rval;
+bail:
+    Py_XDECREF(key);
+    Py_XDECREF(val);
+    Py_DECREF(rval);
+    return NULL;    
+}
+
+static PyObject *
+_parse_object_unicode(PyScannerObject *s, PyObject *pystr, Py_ssize_t idx, Py_ssize_t *next_idx_ptr) {
+    Py_UNICODE *str = PyUnicode_AS_UNICODE(pystr);
+    Py_ssize_t end_idx = PyUnicode_GET_SIZE(pystr) - 1;
+    PyObject *val = NULL;
+    PyObject *rval = PyDict_New();
+    PyObject *key = NULL;
+    int strict = PyObject_IsTrue(s->strict);
+    Py_ssize_t next_idx;
+    if (rval == NULL)
+        return NULL;
+    
+    /* skip whitespace after { */
+    while (idx <= end_idx && IS_WHITESPACE(str[idx])) idx++;
+
+    /* only loop if the object is non-empty */
+    if (idx <= end_idx && str[idx] != '}') {
+        while (idx <= end_idx) {
+            /* read key */
+            if (str[idx] != '"') {
+                raise_errmsg("Expecting property name", pystr, idx);
+                goto bail;
+            }
+            key = scanstring_unicode(pystr, idx + 1, strict, &next_idx);
+            if (key == NULL)
+                goto bail;
+            idx = next_idx;
+
+            /* skip whitespace between key and : delimiter, read :, skip whitespace */
+            while (idx <= end_idx && IS_WHITESPACE(str[idx])) idx++;
+            if (idx > end_idx || str[idx] != ':') {
+                raise_errmsg("Expecting : delimiter", pystr, idx);
+                goto bail;
+            }
+            idx++;
+            while (idx <= end_idx && IS_WHITESPACE(str[idx])) idx++;
+            
+            /* read any JSON term */
+            val = scan_once_unicode(s, pystr, idx, &next_idx);
+            if (val == NULL)
+                goto bail;
+
+            if (PyDict_SetItem(rval, key, val) == -1)
+                goto bail;
+
+            Py_CLEAR(key);
+            Py_CLEAR(val);
+            idx = next_idx;
+
+            /* skip whitespace before } or , */
+            while (idx <= end_idx && IS_WHITESPACE(str[idx])) idx++;
+
+            /* bail if the object is closed or we didn't get the , delimiter */
+            if (idx > end_idx) break;
+            if (str[idx] == '}') {
+                break;
+            }
+            else if (str[idx] != ',') {
+                raise_errmsg("Expecting , delimiter", pystr, idx);
+                goto bail;
+            }
+            idx++;
+
+            /* skip whitespace after , delimiter */
+            while (idx <= end_idx && IS_WHITESPACE(str[idx])) idx++;
+        }
+    }
+
+    /* verify that idx < end_idx, str[idx] should be '}' */
+    if (idx > end_idx || str[idx] != '}') {
+        raise_errmsg("Expecting object", pystr, end_idx);
+        goto bail;
+    }
+
+    /* if object_hook is not None: rval = object_hook(rval) */
+    if (s->object_hook != Py_None) {
+        val = PyObject_CallFunctionObjArgs(s->object_hook, rval, NULL);
+        if (val == NULL)
+            goto bail;
+        Py_DECREF(rval);
+        rval = val;
+        val = NULL;
+    }
+    *next_idx_ptr = idx + 1;
+    return rval;
+bail:
+    Py_XDECREF(key);
+    Py_XDECREF(val);
+    Py_DECREF(rval);
+    return NULL;
+}
+
+static PyObject *
+_parse_array_str(PyScannerObject *s, PyObject *pystr, Py_ssize_t idx, Py_ssize_t *next_idx_ptr) {
+    char *str = PyString_AS_STRING(pystr);
+    Py_ssize_t end_idx = PyString_GET_SIZE(pystr) - 1;
+    PyObject *val = NULL;
+    PyObject *rval = PyList_New(0);
+    Py_ssize_t next_idx;
+    if (rval == NULL)
+        return NULL;
+
+    /* skip whitespace after [ */
+    while (idx <= end_idx && IS_WHITESPACE(str[idx])) idx++;
+
+    /* only loop if the array is non-empty */
+    if (idx <= end_idx && str[idx] != ']') {
+        while (idx <= end_idx) {
+
+            /* read any JSON term and de-tuplefy the (rval, idx) */
+            val = scan_once_str(s, pystr, idx, &next_idx);
+            if (val == NULL)
+                goto bail;
+
+            if (PyList_Append(rval, val) == -1)
+                goto bail;
+
+            Py_CLEAR(val);
+            idx = next_idx;
+            
+            /* skip whitespace between term and , */
+            while (idx <= end_idx && IS_WHITESPACE(str[idx])) idx++;
+
+            /* bail if the array is closed or we didn't get the , delimiter */
+            if (idx > end_idx) break;
+            if (str[idx] == ']') {
+                break;
+            }
+            else if (str[idx] != ',') {
+                raise_errmsg("Expecting , delimiter", pystr, idx);
+                goto bail;
+            }
+            idx++;
+            
+            /* skip whitespace after , */
+            while (idx <= end_idx && IS_WHITESPACE(str[idx])) idx++;
+        }
+    }
+
+    /* verify that idx < end_idx, str[idx] should be ']' */
+    if (idx > end_idx || str[idx] != ']') {
+        raise_errmsg("Expecting object", pystr, end_idx);
+        goto bail;
+    }
+    *next_idx_ptr = idx + 1;
+    return rval;
+bail:
+    Py_XDECREF(val);
+    Py_DECREF(rval);
+    return NULL;
+}
+
+static PyObject *
+_parse_array_unicode(PyScannerObject *s, PyObject *pystr, Py_ssize_t idx, Py_ssize_t *next_idx_ptr) {
+    Py_UNICODE *str = PyUnicode_AS_UNICODE(pystr);
+    Py_ssize_t end_idx = PyUnicode_GET_SIZE(pystr) - 1;
+    PyObject *val = NULL;
+    PyObject *rval = PyList_New(0);
+    Py_ssize_t next_idx;
+    if (rval == NULL)
+        return NULL;
+
+    /* skip whitespace after [ */
+    while (idx <= end_idx && IS_WHITESPACE(str[idx])) idx++;
+
+    /* only loop if the array is non-empty */
+    if (idx <= end_idx && str[idx] != ']') {
+        while (idx <= end_idx) {
+
+            /* read any JSON term  */
+            val = scan_once_unicode(s, pystr, idx, &next_idx);
+            if (val == NULL)
+                goto bail;
+
+            if (PyList_Append(rval, val) == -1)
+                goto bail;
+
+            Py_CLEAR(val);
+            idx = next_idx;
+
+            /* skip whitespace between term and , */
+            while (idx <= end_idx && IS_WHITESPACE(str[idx])) idx++;
+
+            /* bail if the array is closed or we didn't get the , delimiter */
+            if (idx > end_idx) break;
+            if (str[idx] == ']') {
+                break;
+            }
+            else if (str[idx] != ',') {
+                raise_errmsg("Expecting , delimiter", pystr, idx);
+                goto bail;
+            }
+            idx++;
+
+            /* skip whitespace after , */
+            while (idx <= end_idx && IS_WHITESPACE(str[idx])) idx++;
+        }
+    }
+
+    /* verify that idx < end_idx, str[idx] should be ']' */
+    if (idx > end_idx || str[idx] != ']') {
+        raise_errmsg("Expecting object", pystr, end_idx);
+        goto bail;
+    }
+    *next_idx_ptr = idx + 1;
+    return rval;
+bail:
+    Py_XDECREF(val);
+    Py_DECREF(rval);
+    return NULL;
+}
+
+static PyObject *
+_parse_constant(PyScannerObject *s, char *constant, Py_ssize_t idx, Py_ssize_t *next_idx_ptr) {
+    PyObject *cstr;
+    PyObject *rval;
+    /* constant is "NaN", "Infinity", or "-Infinity" */
+    cstr = PyString_InternFromString(constant);
+    if (cstr == NULL)
+        return NULL;
+
+    /* rval = parse_constant(constant) */
+    rval = PyObject_CallFunctionObjArgs(s->parse_constant, cstr, NULL);
+    idx += PyString_GET_SIZE(cstr);
+    Py_DECREF(cstr);
+    *next_idx_ptr = idx;
+    return rval;
+}
+
+static PyObject *
+_match_number_str(PyScannerObject *s, PyObject *pystr, Py_ssize_t start, Py_ssize_t *next_idx_ptr) {
+    char *str = PyString_AS_STRING(pystr);
+    Py_ssize_t end_idx = PyString_GET_SIZE(pystr) - 1;
+    Py_ssize_t idx = start;
+    int is_float = 0;
+    PyObject *rval;
+    PyObject *numstr;
+    
+    /* read a sign if it's there, make sure it's not the end of the string */
+    if (str[idx] == '-') {
+        idx++;
+        if (idx > end_idx) {
+            PyErr_SetNone(PyExc_StopIteration);
+            return NULL;
+        }
+    }
+
+    /* read as many integer digits as we find as long as it doesn't start with 0 */
+    if (str[idx] >= '1' && str[idx] <= '9') {
+        idx++;
+        while (idx <= end_idx && str[idx] >= '0' && str[idx] <= '9') idx++;
+    }
+    /* if it starts with 0 we only expect one integer digit */
+    else if (str[idx] == '0') {
+        idx++;
+    }
+    /* no integer digits, error */
+    else {
+        PyErr_SetNone(PyExc_StopIteration);
+        return NULL;
+    }
+    
+    /* if the next char is '.' followed by a digit then read all float digits */
+    if (idx < end_idx && str[idx] == '.' && str[idx + 1] >= '0' && str[idx + 1] <= '9') {
+        is_float = 1;
+        idx += 2;
+        while (idx <= end_idx && str[idx] >= '0' && str[idx] <= '9') idx++;
+    }
+
+    /* if the next char is 'e' or 'E' then maybe read the exponent (or backtrack) */
+    if (idx < end_idx && (str[idx] == 'e' || str[idx] == 'E')) {
+
+        /* save the index of the 'e' or 'E' just in case we need to backtrack */
+        Py_ssize_t e_start = idx;
+        idx++;
+
+        /* read an exponent sign if present */
+        if (idx < end_idx && (str[idx] == '-' || str[idx] == '+')) idx++;
+
+        /* read all digits */
+        while (idx <= end_idx && str[idx] >= '0' && str[idx] <= '9') idx++;
+
+        /* if we got a digit, then parse as float. if not, backtrack */
+        if (str[idx - 1] >= '0' && str[idx - 1] <= '9') {
+            is_float = 1;
+        }
+        else {
+            idx = e_start;
+        }
+    }
+    
+    /* copy the section we determined to be a number */
+    numstr = PyString_FromStringAndSize(&str[start], idx - start);
+    if (numstr == NULL)
+        return NULL;
+    if (is_float) {
+        /* parse as a float using a fast path if available, otherwise call user defined method */
+        if (s->parse_float != (PyObject *)&PyFloat_Type) {
+            rval = PyObject_CallFunctionObjArgs(s->parse_float, numstr, NULL);
+        }
+        else {
+            rval = PyFloat_FromDouble(PyOS_ascii_atof(PyString_AS_STRING(numstr)));
+        }
+    }
+    else {
+        /* parse as an int using a fast path if available, otherwise call user defined method */
+        if (s->parse_int != (PyObject *)&PyInt_Type) {
+            rval = PyObject_CallFunctionObjArgs(s->parse_int, numstr, NULL);
+        }
+        else {
+            rval = PyInt_FromString(PyString_AS_STRING(numstr), NULL, 10);
+        }
+    }
+    Py_DECREF(numstr);
+    *next_idx_ptr = idx;
+    return rval;
+}
+
+static PyObject *
+_match_number_unicode(PyScannerObject *s, PyObject *pystr, Py_ssize_t start, Py_ssize_t *next_idx_ptr) {
+    Py_UNICODE *str = PyUnicode_AS_UNICODE(pystr);
+    Py_ssize_t end_idx = PyUnicode_GET_SIZE(pystr) - 1;
+    Py_ssize_t idx = start;
+    int is_float = 0;
+    PyObject *rval;
+    PyObject *numstr;
+
+    /* read a sign if it's there, make sure it's not the end of the string */
+    if (str[idx] == '-') {
+        idx++;
+        if (idx > end_idx) {
+            PyErr_SetNone(PyExc_StopIteration);
+            return NULL;
+        }
+    }
+
+    /* read as many integer digits as we find as long as it doesn't start with 0 */
+    if (str[idx] >= '1' && str[idx] <= '9') {
+        idx++;
+        while (idx <= end_idx && str[idx] >= '0' && str[idx] <= '9') idx++;
+    }
+    /* if it starts with 0 we only expect one integer digit */
+    else if (str[idx] == '0') {
+        idx++;
+    }
+    /* no integer digits, error */
+    else {
+        PyErr_SetNone(PyExc_StopIteration);
+        return NULL;
+    }
+
+    /* if the next char is '.' followed by a digit then read all float digits */
+    if (idx < end_idx && str[idx] == '.' && str[idx + 1] >= '0' && str[idx + 1] <= '9') {
+        is_float = 1;
+        idx += 2;
+        while (idx < end_idx && str[idx] >= '0' && str[idx] <= '9') idx++;
+    }
+
+    /* if the next char is 'e' or 'E' then maybe read the exponent (or backtrack) */
+    if (idx < end_idx && (str[idx] == 'e' || str[idx] == 'E')) {
+        Py_ssize_t e_start = idx;
+        idx++;
+
+        /* read an exponent sign if present */
+        if (idx < end_idx && (str[idx] == '-' || str[idx] == '+')) idx++;
+
+        /* read all digits */
+        while (idx <= end_idx && str[idx] >= '0' && str[idx] <= '9') idx++;
+
+        /* if we got a digit, then parse as float. if not, backtrack */
+        if (str[idx - 1] >= '0' && str[idx - 1] <= '9') {
+            is_float = 1;
+        }
+        else {
+            idx = e_start;
+        }
+    }
+
+    /* copy the section we determined to be a number */
+    numstr = PyUnicode_FromUnicode(&str[start], idx - start);
+    if (numstr == NULL)
+        return NULL;
+    if (is_float) {
+        /* parse as a float using a fast path if available, otherwise call user defined method */
+        if (s->parse_float != (PyObject *)&PyFloat_Type) {
+            rval = PyObject_CallFunctionObjArgs(s->parse_float, numstr, NULL);
+        }
+        else {
+            rval = PyFloat_FromString(numstr, NULL);
+        }
+    }
+    else {
+        /* no fast path for unicode -> int, just call */
+        rval = PyObject_CallFunctionObjArgs(s->parse_int, numstr, NULL);
+    }
+    Py_DECREF(numstr);
+    *next_idx_ptr = idx;
+    return rval;
+}
+
+static PyObject *
+scan_once_str(PyScannerObject *s, PyObject *pystr, Py_ssize_t idx, Py_ssize_t *next_idx_ptr)
+{
+    char *str = PyString_AS_STRING(pystr);
+    Py_ssize_t length = PyString_GET_SIZE(pystr);
+    if (idx >= length) {
+        PyErr_SetNone(PyExc_StopIteration);
+        return NULL;
+    }
+    switch (str[idx]) {
+        case '"':
+            /* string */
+            return scanstring_str(pystr, idx + 1,
+                PyString_AS_STRING(s->encoding),
+                PyObject_IsTrue(s->strict),
+                next_idx_ptr);
+        case '{':
+            /* object */
+            return _parse_object_str(s, pystr, idx + 1, next_idx_ptr);
+        case '[':
+            /* array */
+            return _parse_array_str(s, pystr, idx + 1, next_idx_ptr);
+        case 'n':
+            /* null */
+            if ((idx + 3 < length) && str[idx + 1] == 'u' && str[idx + 2] == 'l' && str[idx + 3] == 'l') {
+                Py_INCREF(Py_None);
+                *next_idx_ptr = idx + 4;
+                return Py_None;
+            }
+            break;
+        case 't':
+            /* true */
+            if ((idx + 3 < length) && str[idx + 1] == 'r' && str[idx + 2] == 'u' && str[idx + 3] == 'e') {
+                Py_INCREF(Py_True);
+                *next_idx_ptr = idx + 4;
+                return Py_True;
+            }
+            break;
+        case 'f':
+            /* false */
+            if ((idx + 4 < length) && str[idx + 1] == 'a' && str[idx + 2] == 'l' && str[idx + 3] == 's' && str[idx + 4] == 'e') {
+                Py_INCREF(Py_False);
+                *next_idx_ptr = idx + 5;
+                return Py_False;
+            }
+            break;
+        case 'N':
+            /* NaN */
+            if ((idx + 2 < length) && str[idx + 1] == 'a' && str[idx + 2] == 'N') {
+                return _parse_constant(s, "NaN", idx, next_idx_ptr);
+            }
+            break;
+        case 'I':
+            /* Infinity */
+            if ((idx + 7 < length) && str[idx + 1] == 'n' && str[idx + 2] == 'f' && str[idx + 3] == 'i' && str[idx + 4] == 'n' && str[idx + 5] == 'i' && str[idx + 6] == 't' && str[idx + 7] == 'y') {