Upgrade simplejson to 3.6.0

This commit is contained in:
Matthew Jones
2014-08-06 15:37:54 -04:00
parent 15c2627968
commit 53a8e5c4c6
10 changed files with 228 additions and 123 deletions

View File

@@ -54,6 +54,6 @@ rax-default-network-flags-python-novaclient-ext==0.2.3 (rax_default_network_flag
rax-scheduled-images-python-novaclient-ext==0.2.1 (rax_scheduled_images_python_novaclient_ext/*)
requests==2.3.0 (requests/*)
setuptools==2.2 (setuptools/*, _markerlib/*, pkg_resources.py, easy_install.py, excluded bin/easy_install*)
simplejson==3.3.3 (simplejson/*, excluded simplejson/_speedups.so)
simplejson==3.6.0 (simplejson/*, excluded simplejson/_speedups.so)
six==1.6.1 (six.py)
South==0.8.4 (south/*)

View File

@@ -98,7 +98,7 @@ Using simplejson.tool from the shell to validate and pretty-print::
Expecting property name: line 1 column 3 (char 2)
"""
from __future__ import absolute_import
__version__ = '3.3.3'
__version__ = '3.6.0'
__all__ = [
'dump', 'dumps', 'load', 'loads',
'JSONDecoder', 'JSONDecodeError', 'JSONEncoder',
@@ -144,14 +144,15 @@ _default_encoder = JSONEncoder(
item_sort_key=None,
for_json=False,
ignore_nan=False,
int_as_string_bitcount=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, use_decimal=True,
namedtuple_as_object=True, tuple_as_array=True,
bigint_as_string=False, sort_keys=False, item_sort_key=None,
for_json=False, ignore_nan=False, **kw):
allow_nan=True, cls=None, indent=None, separators=None,
encoding='utf-8', default=None, use_decimal=True,
namedtuple_as_object=True, tuple_as_array=True,
bigint_as_string=False, sort_keys=False, item_sort_key=None,
for_json=False, ignore_nan=False, int_as_string_bitcount=None, **kw):
"""Serialize ``obj`` as a JSON formatted stream to ``fp`` (a
``.write()``-supporting file-like object).
@@ -209,6 +210,10 @@ def dump(obj, fp, skipkeys=False, ensure_ascii=True, check_circular=True,
lossy operation that will not round-trip correctly and should be used
sparingly.
If *int_as_string_bitcount* is a positive number (n), then int of size
greater than or equal to 2**n or lower than or equal to -2**n will be
encoded as strings.
If specified, *item_sort_key* is a callable used to sort the items in
each dictionary. This is useful if you want to sort items other than
in alphabetical order by key. This option takes precedence over
@@ -238,8 +243,8 @@ def dump(obj, fp, skipkeys=False, ensure_ascii=True, check_circular=True,
cls is None and indent is None and separators is None and
encoding == 'utf-8' and default is None and use_decimal
and namedtuple_as_object and tuple_as_array
and not bigint_as_string and not item_sort_key
and not for_json and not ignore_nan and not kw):
and not bigint_as_string and int_as_string_bitcount is None
and not item_sort_key and not for_json and not ignore_nan and not kw):
iterable = _default_encoder.iterencode(obj)
else:
if cls is None:
@@ -255,6 +260,7 @@ def dump(obj, fp, skipkeys=False, ensure_ascii=True, check_circular=True,
item_sort_key=item_sort_key,
for_json=for_json,
ignore_nan=ignore_nan,
int_as_string_bitcount=int_as_string_bitcount,
**kw).iterencode(obj)
# could accelerate with writelines in some versions of Python, at
# a debuggability cost
@@ -263,11 +269,11 @@ 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, use_decimal=True,
namedtuple_as_object=True, tuple_as_array=True,
bigint_as_string=False, sort_keys=False, item_sort_key=None,
for_json=False, ignore_nan=False, **kw):
allow_nan=True, cls=None, indent=None, separators=None,
encoding='utf-8', default=None, use_decimal=True,
namedtuple_as_object=True, tuple_as_array=True,
bigint_as_string=False, sort_keys=False, item_sort_key=None,
for_json=False, ignore_nan=False, int_as_string_bitcount=None, **kw):
"""Serialize ``obj`` to a JSON formatted ``str``.
If ``skipkeys`` is false then ``dict`` keys that are not basic types
@@ -319,6 +325,10 @@ def dumps(obj, skipkeys=False, ensure_ascii=True, check_circular=True,
or lower than -2**53 will be encoded as strings. This is to avoid the
rounding that happens in Javascript otherwise.
If *int_as_string_bitcount* is a positive number (n), then int of size
greater than or equal to 2**n or lower than or equal to -2**n will be
encoded as strings.
If specified, *item_sort_key* is a callable used to sort the items in
each dictionary. This is useful if you want to sort items other than
in alphabetical order by key. This option takes precendence over
@@ -343,14 +353,16 @@ def dumps(obj, skipkeys=False, ensure_ascii=True, check_circular=True,
"""
# cached encoder
if (not skipkeys and ensure_ascii 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 use_decimal
and namedtuple_as_object and tuple_as_array
and not bigint_as_string and not sort_keys
and not item_sort_key and not for_json
and not ignore_nan and not kw):
and not bigint_as_string and int_as_string_bitcount is None
and not sort_keys and not item_sort_key and not for_json
and not ignore_nan and not kw
):
return _default_encoder.encode(obj)
if cls is None:
cls = JSONEncoder
@@ -366,6 +378,7 @@ def dumps(obj, skipkeys=False, ensure_ascii=True, check_circular=True,
item_sort_key=item_sort_key,
for_json=for_json,
ignore_nan=ignore_nan,
int_as_string_bitcount=int_as_string_bitcount,
**kw).encode(obj)

View File

@@ -384,6 +384,17 @@ class JSONDecoder(object):
have extraneous data at the end.
"""
if idx < 0:
# Ensure that raw_decode bails on negative indexes, the regex
# would otherwise mask this behavior. #98
raise JSONDecodeError('Expecting value', s, idx)
if _PY3 and not isinstance(s, text_type):
raise TypeError("Input string must be text, not bytes")
# strip UTF-8 bom
if len(s) > idx:
ord0 = ord(s[idx])
if ord0 == 0xfeff:
idx += 1
elif ord0 == 0xef and s[idx:idx + 3] == '\xef\xbb\xbf':
idx += 3
return self.scan_once(s, idx=_w(s, idx).end())

View File

@@ -116,12 +116,14 @@ class JSONEncoder(object):
"""
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,
use_decimal=True, namedtuple_as_object=True,
tuple_as_array=True, bigint_as_string=False,
item_sort_key=None, for_json=False, ignore_nan=False):
check_circular=True, allow_nan=True, sort_keys=False,
indent=None, separators=None, encoding='utf-8', default=None,
use_decimal=True, namedtuple_as_object=True,
tuple_as_array=True, bigint_as_string=False,
item_sort_key=None, for_json=False, ignore_nan=False,
int_as_string_bitcount=None):
"""Constructor for JSONEncoder, with sensible defaults.
If skipkeys is false, then it is a TypeError to attempt
@@ -180,6 +182,10 @@ class JSONEncoder(object):
or lower than -2**53 will be encoded as strings. This is to avoid the
rounding that happens in Javascript otherwise.
If int_as_string_bitcount is a positive number (n), then int of size
greater than or equal to 2**n or lower than or equal to -2**n will be
encoded as strings.
If specified, item_sort_key is a callable used to sort the items in
each dictionary. This is useful if you want to sort items other than
in alphabetical order by key.
@@ -207,6 +213,7 @@ class JSONEncoder(object):
self.item_sort_key = item_sort_key
self.for_json = for_json
self.ignore_nan = ignore_nan
self.int_as_string_bitcount = int_as_string_bitcount
if indent is not None and not isinstance(indent, string_types):
indent = indent * ' '
self.indent = indent
@@ -315,8 +322,9 @@ class JSONEncoder(object):
return text
key_memo = {}
int_as_string_bitcount = (
53 if self.bigint_as_string else self.int_as_string_bitcount)
if (_one_shot and c_make_encoder is not None
and self.indent is None):
_iterencode = c_make_encoder(
@@ -324,17 +332,17 @@ class JSONEncoder(object):
self.key_separator, self.item_separator, self.sort_keys,
self.skipkeys, self.allow_nan, key_memo, self.use_decimal,
self.namedtuple_as_object, self.tuple_as_array,
self.bigint_as_string, self.item_sort_key,
self.encoding, self.for_json, self.ignore_nan,
Decimal)
int_as_string_bitcount,
self.item_sort_key, self.encoding, self.for_json,
self.ignore_nan, Decimal)
else:
_iterencode = _make_iterencode(
markers, self.default, _encoder, self.indent, floatstr,
self.key_separator, self.item_separator, self.sort_keys,
self.skipkeys, _one_shot, self.use_decimal,
self.namedtuple_as_object, self.tuple_as_array,
self.bigint_as_string, self.item_sort_key,
self.encoding, self.for_json,
int_as_string_bitcount,
self.item_sort_key, self.encoding, self.for_json,
Decimal=Decimal)
try:
return _iterencode(o, 0)
@@ -372,7 +380,8 @@ class JSONEncoderForHTML(JSONEncoder):
def _make_iterencode(markers, _default, _encoder, _indent, _floatstr,
_key_separator, _item_separator, _sort_keys, _skipkeys, _one_shot,
_use_decimal, _namedtuple_as_object, _tuple_as_array,
_bigint_as_string, _item_sort_key, _encoding, _for_json,
_int_as_string_bitcount, _item_sort_key,
_encoding,_for_json,
## HACK: hand-optimized bytecode; turn globals into locals
_PY3=PY3,
ValueError=ValueError,
@@ -392,6 +401,26 @@ def _make_iterencode(markers, _default, _encoder, _indent, _floatstr,
elif _sort_keys and not _item_sort_key:
_item_sort_key = itemgetter(0)
if (_int_as_string_bitcount is not None and
(_int_as_string_bitcount <= 0 or
not isinstance(_int_as_string_bitcount, integer_types))):
raise TypeError("int_as_string_bitcount must be a positive integer")
def _encode_int(value):
skip_quoting = (
_int_as_string_bitcount is None
or
_int_as_string_bitcount < 1
)
if (
skip_quoting or
(-1 << _int_as_string_bitcount)
< value <
(1 << _int_as_string_bitcount)
):
return str(value)
return '"' + str(value) + '"'
def _iterencode_list(lst, _current_indent_level):
if not lst:
yield '[]'
@@ -426,10 +455,7 @@ def _make_iterencode(markers, _default, _encoder, _indent, _floatstr,
elif value is False:
yield buf + 'false'
elif isinstance(value, integer_types):
yield ((buf + str(value))
if (not _bigint_as_string or
(-1 << 53) < value < (1 << 53))
else (buf + '"' + str(value) + '"'))
yield buf + _encode_int(value)
elif isinstance(value, float):
yield buf + _floatstr(value)
elif _use_decimal and isinstance(value, Decimal):
@@ -540,10 +566,7 @@ def _make_iterencode(markers, _default, _encoder, _indent, _floatstr,
elif value is False:
yield 'false'
elif isinstance(value, integer_types):
yield (str(value)
if (not _bigint_as_string or
(-1 << 53) < value < (1 << 53))
else ('"' + str(value) + '"'))
yield _encode_int(value)
elif isinstance(value, float):
yield _floatstr(value)
elif _use_decimal and isinstance(value, Decimal):
@@ -585,10 +608,7 @@ def _make_iterencode(markers, _default, _encoder, _indent, _floatstr,
elif o is False:
yield 'false'
elif isinstance(o, integer_types):
yield (str(o)
if (not _bigint_as_string or
(-1 << 53) < o < (1 << 53))
else ('"' + str(o) + '"'))
yield _encode_int(o)
elif isinstance(o, float):
yield _floatstr(o)
else:

View File

@@ -118,6 +118,11 @@ def py_make_scanner(context):
raise JSONDecodeError(errmsg, string, idx)
def scan_once(string, idx):
if idx < 0:
# Ensure the same behavior as the C speedup, otherwise
# this would work for *some* negative string indices due
# to the behavior of __getitem__ for strings. #98
raise JSONDecodeError('Expecting value', string, idx)
try:
return _scan_once(string, idx)
finally:

View File

@@ -3,19 +3,16 @@ import unittest
import doctest
import sys
class OptionalExtensionTestSuite(unittest.TestSuite):
class NoExtensionTestSuite(unittest.TestSuite):
def run(self, result):
import simplejson
run = unittest.TestSuite.run
run(self, result)
if simplejson._import_c_make_encoder() is None:
TestMissingSpeedups().run(result)
else:
simplejson._toggle_speedups(False)
run(self, result)
simplejson._toggle_speedups(True)
simplejson._toggle_speedups(False)
result = unittest.TestSuite.run(self, result)
simplejson._toggle_speedups(True)
return result
class TestMissingSpeedups(unittest.TestCase):
def runTest(self):
if hasattr(sys, 'pypy_translation_info'):
@@ -23,6 +20,7 @@ class TestMissingSpeedups(unittest.TestCase):
elif hasattr(self, 'skipTest'):
self.skipTest('_speedups.so is missing!')
def additional_tests(suite=None):
import simplejson
import simplejson.encoder
@@ -36,34 +34,45 @@ def additional_tests(suite=None):
def all_tests_suite():
suite = unittest.TestLoader().loadTestsFromNames([
'simplejson.tests.test_bigint_as_string',
'simplejson.tests.test_check_circular',
'simplejson.tests.test_decode',
'simplejson.tests.test_default',
'simplejson.tests.test_dump',
'simplejson.tests.test_encode_basestring_ascii',
'simplejson.tests.test_encode_for_html',
'simplejson.tests.test_errors',
'simplejson.tests.test_fail',
'simplejson.tests.test_float',
'simplejson.tests.test_indent',
'simplejson.tests.test_pass1',
'simplejson.tests.test_pass2',
'simplejson.tests.test_pass3',
'simplejson.tests.test_recursion',
'simplejson.tests.test_scanstring',
'simplejson.tests.test_separators',
'simplejson.tests.test_speedups',
'simplejson.tests.test_unicode',
'simplejson.tests.test_decimal',
'simplejson.tests.test_tuple',
'simplejson.tests.test_namedtuple',
'simplejson.tests.test_tool',
'simplejson.tests.test_for_json',
])
suite = additional_tests(suite)
return OptionalExtensionTestSuite([suite])
def get_suite():
return additional_tests(
unittest.TestLoader().loadTestsFromNames([
'simplejson.tests.test_bitsize_int_as_string',
'simplejson.tests.test_bigint_as_string',
'simplejson.tests.test_check_circular',
'simplejson.tests.test_decode',
'simplejson.tests.test_default',
'simplejson.tests.test_dump',
'simplejson.tests.test_encode_basestring_ascii',
'simplejson.tests.test_encode_for_html',
'simplejson.tests.test_errors',
'simplejson.tests.test_fail',
'simplejson.tests.test_float',
'simplejson.tests.test_indent',
'simplejson.tests.test_pass1',
'simplejson.tests.test_pass2',
'simplejson.tests.test_pass3',
'simplejson.tests.test_recursion',
'simplejson.tests.test_scanstring',
'simplejson.tests.test_separators',
'simplejson.tests.test_speedups',
'simplejson.tests.test_unicode',
'simplejson.tests.test_decimal',
'simplejson.tests.test_tuple',
'simplejson.tests.test_namedtuple',
'simplejson.tests.test_tool',
'simplejson.tests.test_for_json',
]))
suite = get_suite()
import simplejson
if simplejson._import_c_make_encoder() is None:
suite.addTest(TestMissingSpeedups())
else:
suite = unittest.TestSuite([
suite,
NoExtensionTestSuite([get_suite()]),
])
return suite
def main():

View File

@@ -1,7 +1,7 @@
from unittest import TestCase
import simplejson as json
from simplejson.compat import long_type
class TestBigintAsString(TestCase):
# Python 2.5, at least the one that ships on Mac OS X, calculates
@@ -15,44 +15,53 @@ class TestBigintAsString(TestCase):
((-1 << 53) - 1, '-9007199254740993'),
((-1 << 53) + 1, -9007199254740991)]
options = (
{"bigint_as_string": True},
{"int_as_string_bitcount": 53}
)
def test_ints(self):
for val, expect in self.values:
self.assertEqual(
val,
json.loads(json.dumps(val)))
self.assertEqual(
expect,
json.loads(json.dumps(val, bigint_as_string=True)))
for opts in self.options:
for val, expect in self.values:
self.assertEqual(
val,
json.loads(json.dumps(val)))
self.assertEqual(
expect,
json.loads(json.dumps(val, **opts)))
def test_lists(self):
for val, expect in self.values:
val = [val, val]
expect = [expect, expect]
self.assertEqual(
val,
json.loads(json.dumps(val)))
self.assertEqual(
expect,
json.loads(json.dumps(val, bigint_as_string=True)))
for opts in self.options:
for val, expect in self.values:
val = [val, val]
expect = [expect, expect]
self.assertEqual(
val,
json.loads(json.dumps(val)))
self.assertEqual(
expect,
json.loads(json.dumps(val, **opts)))
def test_dicts(self):
for val, expect in self.values:
val = {'k': val}
expect = {'k': expect}
self.assertEqual(
val,
json.loads(json.dumps(val)))
self.assertEqual(
expect,
json.loads(json.dumps(val, bigint_as_string=True)))
for opts in self.options:
for val, expect in self.values:
val = {'k': val}
expect = {'k': expect}
self.assertEqual(
val,
json.loads(json.dumps(val)))
self.assertEqual(
expect,
json.loads(json.dumps(val, **opts)))
def test_dict_keys(self):
for val, _ in self.values:
expect = {str(val): 'value'}
val = {val: 'value'}
self.assertEqual(
expect,
json.loads(json.dumps(val)))
self.assertEqual(
expect,
json.loads(json.dumps(val, bigint_as_string=True)))
for opts in self.options:
for val, _ in self.values:
expect = {str(val): 'value'}
val = {val: 'value'}
self.assertEqual(
expect,
json.loads(json.dumps(val)))
self.assertEqual(
expect,
json.loads(json.dumps(val, **opts)))

View File

@@ -86,3 +86,14 @@ class TestDecode(TestCase):
self.assertEqual(
({'a': {}}, 11),
cls().raw_decode(" \n{\"a\": {}}"))
def test_bounds_checking(self):
# https://github.com/simplejson/simplejson/issues/98
j = json.decoder.JSONDecoder()
for i in [4, 5, 6, -1, -2, -3, -4, -5, -6]:
self.assertRaises(ValueError, j.scan_once, '1234', i)
self.assertRaises(ValueError, j.raw_decode, '1234', i)
x, y = sorted(['128931233', '472389423'], key=id)
diff = id(x) - id(y)
self.assertRaises(ValueError, j.scan_once, y, diff)
self.assertRaises(ValueError, j.raw_decode, y, i)

View File

@@ -1,20 +1,39 @@
import sys
import unittest
from unittest import TestCase
from simplejson import encoder, scanner
def has_speedups():
return encoder.c_make_encoder is not None
class TestDecode(TestCase):
def test_make_scanner(self):
def skip_if_speedups_missing(func):
def wrapper(*args, **kwargs):
if not has_speedups():
return
if hasattr(unittest, 'SkipTest'):
raise unittest.SkipTest("C Extension not available")
else:
sys.stdout.write("C Extension not available")
return
return func(*args, **kwargs)
return wrapper
class TestDecode(TestCase):
@skip_if_speedups_missing
def test_make_scanner(self):
self.assertRaises(AttributeError, scanner.c_make_scanner, 1)
@skip_if_speedups_missing
def test_make_encoder(self):
if not has_speedups():
return
self.assertRaises(TypeError, encoder.c_make_encoder,
self.assertRaises(
TypeError,
encoder.c_make_encoder,
None,
"\xCD\x7D\x3D\x4E\x12\x4C\xF9\x79\xD7\x52\xBA\x82\xF2\x27\x4A\x7D\xA0\xCA\x75",
None)
("\xCD\x7D\x3D\x4E\x12\x4C\xF9\x79\xD7"
"\x52\xBA\x82\xF2\x27\x4A\x7D\xA0\xCA\x75"),
None
)

View File

@@ -1,8 +1,9 @@
import sys
import codecs
from unittest import TestCase
import simplejson as json
from simplejson.compat import unichr, text_type, b, u
from simplejson.compat import unichr, text_type, b, u, BytesIO
class TestUnicode(TestCase):
def test_encoding1(self):
@@ -143,3 +144,10 @@ class TestUnicode(TestCase):
self.assertEqual(
json.dumps(c, ensure_ascii=False),
'"' + c + '"')
def test_strip_bom(self):
content = u"\u3053\u3093\u306b\u3061\u308f"
json_doc = codecs.BOM_UTF8 + b(json.dumps(content))
self.assertEqual(json.load(BytesIO(json_doc)), content)
for doc in json_doc, json_doc.decode('utf8'):
self.assertEqual(json.loads(doc), content)