Update SimpleJSON 3.13.2 (6ffddbe) → 3.16.0 (e2a54f7).

This commit is contained in:
JackDandy 2018-09-04 23:19:28 +01:00
parent b6ae9f529d
commit 29ae371a88
6 changed files with 202 additions and 213 deletions

View file

@ -11,6 +11,7 @@
* Update scandir module 1.6 (c3592ee) to 1.9.0 (9ab3d1f)
* Add urllib3 release 1.23 (7c216f4)
* Change if old scandir binary module is installed, fallback to slow Python module and inform user to upgrade binary
* Update SimpleJSON 3.13.2 (6ffddbe) to 3.16.0 (e2a54f7)
[develop changelog]

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__ = '3.13.2'
__version__ = '3.16.0'
__all__ = [
'dump', 'dumps', 'load', 'loads',
'JSONDecoder', 'JSONDecodeError', 'JSONEncoder',

View file

@ -5,36 +5,24 @@
#if PY_MAJOR_VERSION >= 3
#define PyInt_FromSsize_t PyLong_FromSsize_t
#define PyInt_AsSsize_t PyLong_AsSsize_t
#define PyString_Check PyBytes_Check
#define PyString_GET_SIZE PyBytes_GET_SIZE
#define PyString_AS_STRING PyBytes_AS_STRING
#define PyString_FromStringAndSize PyBytes_FromStringAndSize
#define PyInt_Check(obj) 0
#define PyInt_CheckExact(obj) 0
#define JSON_UNICHR Py_UCS4
#define JSON_InternFromString PyUnicode_InternFromString
#define JSON_Intern_GET_SIZE PyUnicode_GET_SIZE
#define JSON_ASCII_Check PyUnicode_Check
#define JSON_ASCII_AS_STRING PyUnicode_AsUTF8
#define PyInt_Type PyLong_Type
#define PyInt_FromString PyLong_FromString
#define PyString_GET_SIZE PyUnicode_GET_LENGTH
#define PY2_UNUSED
#define PY3_UNUSED UNUSED
#define JSON_NewEmptyUnicode() PyUnicode_New(0, 127)
#else /* PY_MAJOR_VERSION >= 3 */
#define PY2_UNUSED UNUSED
#define PY3_UNUSED
#define PyBytes_Check PyString_Check
#define PyUnicode_READY(obj) 0
#define PyUnicode_KIND(obj) (sizeof(Py_UNICODE))
#define PyUnicode_DATA(obj) ((void *)(PyUnicode_AS_UNICODE(obj)))
#define PyUnicode_READ(kind, data, index) ((JSON_UNICHR)((const Py_UNICODE *)(data))[(index)])
#define PyUnicode_GetLength PyUnicode_GET_SIZE
#define PyUnicode_GET_LENGTH PyUnicode_GET_SIZE
#define JSON_UNICHR Py_UNICODE
#define JSON_ASCII_Check PyString_Check
#define JSON_ASCII_AS_STRING PyString_AS_STRING
#define JSON_InternFromString PyString_InternFromString
#define JSON_Intern_GET_SIZE PyString_GET_SIZE
#define JSON_NewEmptyUnicode() PyUnicode_FromUnicode(NULL, 0)
#endif /* PY_MAJOR_VERSION < 3 */
#if PY_VERSION_HEX < 0x02070000
@ -84,6 +72,14 @@ json_PyOS_string_to_double(const char *s, char **endptr, PyObject *overflow_exce
#define JSON_ALLOW_NAN 1
#define JSON_IGNORE_NAN 2
static PyObject *JSON_Infinity = NULL;
static PyObject *JSON_NegInfinity = NULL;
static PyObject *JSON_NaN = NULL;
static PyObject *JSON_EmptyUnicode = NULL;
#if PY_MAJOR_VERSION < 3
static PyObject *JSON_EmptyStr = NULL;
#endif
static PyTypeObject PyScannerType;
static PyTypeObject PyEncoderType;
@ -188,8 +184,6 @@ join_list_unicode(PyObject *lst);
static PyObject *
JSON_ParseEncoding(PyObject *encoding);
static PyObject *
JSON_UnicodeFromChar(JSON_UNICHR c);
static PyObject *
maybe_quote_bigint(PyEncoderObject* s, PyObject *encoded, PyObject *obj);
static Py_ssize_t
ascii_char_size(JSON_UNICHR c);
@ -318,7 +312,7 @@ JSON_Accu_Accumulate(JSON_Accu *acc, PyObject *unicode)
#if PY_MAJOR_VERSION >= 3
assert(PyUnicode_Check(unicode));
#else /* PY_MAJOR_VERSION >= 3 */
assert(JSON_ASCII_Check(unicode) || PyUnicode_Check(unicode));
assert(PyString_Check(unicode) || PyUnicode_Check(unicode));
#endif /* PY_MAJOR_VERSION < 3 */
if (PyList_Append(acc->small_strings, unicode))
@ -368,19 +362,6 @@ IS_DIGIT(JSON_UNICHR c)
return c >= '0' && c <= '9';
}
static PyObject *
JSON_UnicodeFromChar(JSON_UNICHR c)
{
#if PY_MAJOR_VERSION >= 3
PyObject *rval = PyUnicode_New(1, c);
if (rval)
PyUnicode_WRITE(PyUnicode_KIND(rval), PyUnicode_DATA(rval), 0, c);
return rval;
#else /* PY_MAJOR_VERSION >= 3 */
return PyUnicode_FromUnicode(&c, 1);
#endif /* PY_MAJOR_VERSION < 3 */
}
static PyObject *
maybe_quote_bigint(PyEncoderObject* s, PyObject *encoded, PyObject *obj)
{
@ -466,7 +447,7 @@ ascii_escape_char(JSON_UNICHR c, char *output, Py_ssize_t chars)
case '\r': output[chars++] = 'r'; break;
case '\t': output[chars++] = 't'; break;
default:
#if defined(Py_UNICODE_WIDE) || PY_MAJOR_VERSION >= 3
#if PY_MAJOR_VERSION >= 3 || defined(Py_UNICODE_WIDE)
if (c >= 0x10000) {
/* UTF-16 surrogate pair */
JSON_UNICHR v = c - 0x10000;
@ -505,7 +486,7 @@ ascii_char_size(JSON_UNICHR c)
c == '\t') {
return 2;
}
#if defined(Py_UNICODE_WIDE) || PY_MAJOR_VERSION >= 3
#if PY_MAJOR_VERSION >= 3 || defined(Py_UNICODE_WIDE)
else if (c >= 0x10000U) {
return 2 * MIN_EXPANSION;
}
@ -520,20 +501,14 @@ ascii_escape_unicode(PyObject *pystr)
{
/* Take a PyUnicode pystr and return a new ASCII-only escaped PyString */
Py_ssize_t i;
Py_ssize_t input_chars;
Py_ssize_t output_size;
Py_ssize_t input_chars = PyUnicode_GET_LENGTH(pystr);
Py_ssize_t output_size = 2;
Py_ssize_t chars;
PY2_UNUSED int kind;
void *data;
PY2_UNUSED int kind = PyUnicode_KIND(pystr);
void *data = PyUnicode_DATA(pystr);
PyObject *rval;
char *output;
if (PyUnicode_READY(pystr))
return NULL;
kind = PyUnicode_KIND(pystr);
data = PyUnicode_DATA(pystr);
input_chars = PyUnicode_GetLength(pystr);
output_size = 2;
for (i = 0; i < input_chars; i++) {
output_size += ascii_char_size(PyUnicode_READ(kind, data, i));
@ -568,7 +543,7 @@ static PyObject *
ascii_escape_str(PyObject *pystr)
{
PyObject *rval;
PyObject *input = PyUnicode_DecodeUTF8(PyString_AS_STRING(pystr), PyString_GET_SIZE(pystr), NULL);
PyObject *input = PyUnicode_DecodeUTF8(PyBytes_AS_STRING(pystr), PyBytes_GET_SIZE(pystr), NULL);
if (input == NULL)
return NULL;
rval = ascii_escape_unicode(input);
@ -634,21 +609,23 @@ encoder_stringify_key(PyEncoderObject *s, PyObject *key)
Py_INCREF(key);
return key;
}
else if (PyString_Check(key)) {
#if PY_MAJOR_VERSION >= 3
const char *encoding = JSON_ASCII_AS_STRING(s->encoding);
else if (PyBytes_Check(key) && s->encoding != NULL) {
const char *encoding = PyUnicode_AsUTF8(s->encoding);
if (encoding == NULL)
return NULL;
return PyUnicode_Decode(
PyString_AS_STRING(key),
PyString_GET_SIZE(key),
PyBytes_AS_STRING(key),
PyBytes_GET_SIZE(key),
encoding,
NULL);
}
#else /* PY_MAJOR_VERSION >= 3 */
else if (PyString_Check(key)) {
Py_INCREF(key);
return key;
#endif /* PY_MAJOR_VERSION < 3 */
}
#endif /* PY_MAJOR_VERSION < 3 */
else if (PyFloat_Check(key)) {
return encoder_encode_float(s, key);
}
@ -676,7 +653,7 @@ encoder_stringify_key(PyEncoderObject *s, PyObject *key)
else if (s->use_decimal && PyObject_TypeCheck(key, (PyTypeObject *)s->Decimal)) {
return PyObject_Str(key);
}
else if (s->skipkeys) {
if (s->skipkeys) {
Py_INCREF(Py_None);
return Py_None;
}
@ -799,18 +776,7 @@ static PyObject *
join_list_unicode(PyObject *lst)
{
/* return u''.join(lst) */
static PyObject *joinfn = NULL;
if (joinfn == NULL) {
PyObject *ustr = JSON_NewEmptyUnicode();
if (ustr == NULL)
return NULL;
joinfn = PyObject_GetAttrString(ustr, "join");
Py_DECREF(ustr);
if (joinfn == NULL)
return NULL;
}
return PyObject_CallFunctionObjArgs(joinfn, lst, NULL);
return PyUnicode_Join(JSON_EmptyUnicode, lst);
}
#if PY_MAJOR_VERSION >= 3
@ -822,12 +788,7 @@ join_list_string(PyObject *lst)
/* return ''.join(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);
joinfn = PyObject_GetAttrString(JSON_EmptyStr, "join");
if (joinfn == NULL)
return NULL;
}
@ -932,17 +893,6 @@ scanstring_str(PyObject *pystr, Py_ssize_t end, char *encoding, int strict, Py_s
/* Pick up this chunk if it's not zero length */
if (next != end) {
APPEND_OLD_CHUNK
#if PY_MAJOR_VERSION >= 3
if (!has_unicode) {
chunk = PyUnicode_DecodeASCII(&buf[end], next - end, NULL);
}
else {
chunk = PyUnicode_Decode(&buf[end], next - end, encoding, NULL);
}
if (chunk == NULL) {
goto bail;
}
#else /* PY_MAJOR_VERSION >= 3 */
strchunk = PyString_FromStringAndSize(&buf[end], next - end);
if (strchunk == NULL) {
goto bail;
@ -957,7 +907,6 @@ scanstring_str(PyObject *pystr, Py_ssize_t end, char *encoding, int strict, Py_s
else {
chunk = strchunk;
}
#endif /* PY_MAJOR_VERSION < 3 */
}
next++;
if (c == '"') {
@ -1015,7 +964,7 @@ scanstring_str(PyObject *pystr, Py_ssize_t end, char *encoding, int strict, Py_s
goto bail;
}
}
#if (PY_MAJOR_VERSION >= 3 || defined(Py_UNICODE_WIDE))
#if defined(Py_UNICODE_WIDE)
/* Surrogate pair */
if ((c & 0xfc00) == 0xd800) {
if (end + 6 < len && buf[next] == '\\' && buf[next+1] == 'u') {
@ -1050,20 +999,14 @@ scanstring_str(PyObject *pystr, Py_ssize_t end, char *encoding, int strict, Py_s
}
}
}
#endif /* PY_MAJOR_VERSION >= 3 || Py_UNICODE_WIDE */
#endif /* Py_UNICODE_WIDE */
}
if (c > 0x7f) {
has_unicode = 1;
}
APPEND_OLD_CHUNK
#if PY_MAJOR_VERSION >= 3
chunk = JSON_UnicodeFromChar(c);
if (chunk == NULL) {
goto bail;
}
#else /* PY_MAJOR_VERSION >= 3 */
if (has_unicode) {
chunk = JSON_UnicodeFromChar(c);
chunk = PyUnicode_FromOrdinal(c);
if (chunk == NULL) {
goto bail;
}
@ -1075,14 +1018,15 @@ scanstring_str(PyObject *pystr, Py_ssize_t end, char *encoding, int strict, Py_s
goto bail;
}
}
#endif
}
if (chunks == NULL) {
if (chunk != NULL)
rval = chunk;
else
rval = JSON_NewEmptyUnicode();
else {
rval = JSON_EmptyStr;
Py_INCREF(rval);
}
}
else {
APPEND_OLD_CHUNK
@ -1118,7 +1062,7 @@ scanstring_unicode(PyObject *pystr, Py_ssize_t end, int strict, Py_ssize_t *next
Py_ssize_t begin = end - 1;
Py_ssize_t next = begin;
PY2_UNUSED int kind = PyUnicode_KIND(pystr);
Py_ssize_t len = PyUnicode_GetLength(pystr);
Py_ssize_t len = PyUnicode_GET_LENGTH(pystr);
void *buf = PyUnicode_DATA(pystr);
PyObject *chunks = NULL;
PyObject *chunk = NULL;
@ -1256,7 +1200,7 @@ scanstring_unicode(PyObject *pystr, Py_ssize_t end, int strict, Py_ssize_t *next
#endif
}
APPEND_OLD_CHUNK
chunk = JSON_UnicodeFromChar(c);
chunk = PyUnicode_FromOrdinal(c);
if (chunk == NULL) {
goto bail;
}
@ -1265,8 +1209,10 @@ scanstring_unicode(PyObject *pystr, Py_ssize_t end, int strict, Py_ssize_t *next
if (chunks == NULL) {
if (chunk != NULL)
rval = chunk;
else
rval = JSON_NewEmptyUnicode();
else {
rval = JSON_EmptyUnicode;
Py_INCREF(rval);
}
}
else {
APPEND_OLD_CHUNK
@ -1314,6 +1260,8 @@ py_scanstring(PyObject* self UNUSED, PyObject *args)
encoding = DEFAULT_ENCODING;
}
if (PyUnicode_Check(pystr)) {
if (PyUnicode_READY(pystr))
return NULL;
rval = scanstring_unicode(pystr, end, strict, &next_end);
}
#if PY_MAJOR_VERSION < 3
@ -1343,10 +1291,12 @@ py_encode_basestring_ascii(PyObject* self UNUSED, PyObject *pystr)
{
/* Return an ASCII-only JSON representation of a Python string */
/* METH_O */
if (PyString_Check(pystr)) {
if (PyBytes_Check(pystr)) {
return ascii_escape_str(pystr);
}
else if (PyUnicode_Check(pystr)) {
if (PyUnicode_READY(pystr))
return NULL;
return ascii_escape_unicode(pystr);
}
else {
@ -1419,7 +1369,7 @@ _parse_object_str(PyScannerObject *s, PyObject *pystr, Py_ssize_t idx, Py_ssize_
PyObject *item;
PyObject *key = NULL;
PyObject *val = NULL;
char *encoding = JSON_ASCII_AS_STRING(s->encoding);
char *encoding = PyString_AS_STRING(s->encoding);
int has_pairs_hook = (s->pairs_hook != Py_None);
int did_parse = 0;
Py_ssize_t next_idx;
@ -1573,7 +1523,7 @@ _parse_object_unicode(PyScannerObject *s, PyObject *pystr, Py_ssize_t idx, Py_ss
Returns a new PyObject (usually a dict, but object_hook can change that)
*/
void *str = PyUnicode_DATA(pystr);
Py_ssize_t end_idx = PyUnicode_GetLength(pystr) - 1;
Py_ssize_t end_idx = PyUnicode_GET_LENGTH(pystr) - 1;
PY2_UNUSED int kind = PyUnicode_KIND(pystr);
PyObject *rval = NULL;
PyObject *pairs = NULL;
@ -1818,7 +1768,7 @@ _parse_array_unicode(PyScannerObject *s, PyObject *pystr, Py_ssize_t idx, Py_ssi
*/
PY2_UNUSED int kind = PyUnicode_KIND(pystr);
void *str = PyUnicode_DATA(pystr);
Py_ssize_t end_idx = PyUnicode_GetLength(pystr) - 1;
Py_ssize_t end_idx = PyUnicode_GET_LENGTH(pystr) - 1;
PyObject *val = NULL;
PyObject *rval = PyList_New(0);
Py_ssize_t next_idx;
@ -1887,10 +1837,10 @@ bail:
}
static PyObject *
_parse_constant(PyScannerObject *s, char *constant, Py_ssize_t idx, Py_ssize_t *next_idx_ptr)
_parse_constant(PyScannerObject *s, PyObject *constant, Py_ssize_t idx, Py_ssize_t *next_idx_ptr)
{
/* Read a JSON constant from PyString pystr.
constant is the constant string that was found
constant is the Python string that was found
("NaN", "Infinity", "-Infinity").
idx is the index of the first character of the constant
*next_idx_ptr is a return-by-reference index to the first character after
@ -1898,17 +1848,11 @@ _parse_constant(PyScannerObject *s, char *constant, Py_ssize_t idx, Py_ssize_t *
Returns the result of parse_constant
*/
PyObject *cstr;
PyObject *rval;
/* constant is "NaN", "Infinity", or "-Infinity" */
cstr = JSON_InternFromString(constant);
if (cstr == NULL)
return NULL;
/* rval = parse_constant(constant) */
rval = PyObject_CallFunctionObjArgs(s->parse_constant, cstr, NULL);
idx += JSON_Intern_GET_SIZE(cstr);
Py_DECREF(cstr);
rval = PyObject_CallFunctionObjArgs(s->parse_constant, constant, NULL);
idx += PyString_GET_SIZE(constant);
*next_idx_ptr = idx;
return rval;
}
@ -2033,7 +1977,7 @@ _match_number_unicode(PyScannerObject *s, PyObject *pystr, Py_ssize_t start, Py_
*/
PY2_UNUSED int kind = PyUnicode_KIND(pystr);
void *str = PyUnicode_DATA(pystr);
Py_ssize_t end_idx = PyUnicode_GetLength(pystr) - 1;
Py_ssize_t end_idx = PyUnicode_GET_LENGTH(pystr) - 1;
Py_ssize_t idx = start;
int is_float = 0;
JSON_UNICHR c;
@ -2153,7 +2097,7 @@ scan_once_str(PyScannerObject *s, PyObject *pystr, Py_ssize_t idx, Py_ssize_t *n
case '"':
/* string */
rval = scanstring_str(pystr, idx + 1,
JSON_ASCII_AS_STRING(s->encoding),
PyString_AS_STRING(s->encoding),
s->strict,
next_idx_ptr);
break;
@ -2206,7 +2150,7 @@ scan_once_str(PyScannerObject *s, PyObject *pystr, Py_ssize_t idx, Py_ssize_t *n
case 'N':
/* NaN */
if ((idx + 2 < length) && str[idx + 1] == 'a' && str[idx + 2] == 'N') {
rval = _parse_constant(s, "NaN", idx, next_idx_ptr);
rval = _parse_constant(s, JSON_NaN, idx, next_idx_ptr);
}
else
fallthrough = 1;
@ -2214,7 +2158,7 @@ scan_once_str(PyScannerObject *s, PyObject *pystr, Py_ssize_t idx, Py_ssize_t *n
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') {
rval = _parse_constant(s, "Infinity", idx, next_idx_ptr);
rval = _parse_constant(s, JSON_Infinity, idx, next_idx_ptr);
}
else
fallthrough = 1;
@ -2222,7 +2166,7 @@ scan_once_str(PyScannerObject *s, PyObject *pystr, Py_ssize_t idx, Py_ssize_t *n
case '-':
/* -Infinity */
if ((idx + 8 < length) && str[idx + 1] == 'I' && str[idx + 2] == 'n' && str[idx + 3] == 'f' && str[idx + 4] == 'i' && str[idx + 5] == 'n' && str[idx + 6] == 'i' && str[idx + 7] == 't' && str[idx + 8] == 'y') {
rval = _parse_constant(s, "-Infinity", idx, next_idx_ptr);
rval = _parse_constant(s, JSON_NegInfinity, idx, next_idx_ptr);
}
else
fallthrough = 1;
@ -2250,7 +2194,7 @@ scan_once_unicode(PyScannerObject *s, PyObject *pystr, Py_ssize_t idx, Py_ssize_
*/
PY2_UNUSED int kind = PyUnicode_KIND(pystr);
void *str = PyUnicode_DATA(pystr);
Py_ssize_t length = PyUnicode_GetLength(pystr);
Py_ssize_t length = PyUnicode_GET_LENGTH(pystr);
PyObject *rval = NULL;
int fallthrough = 0;
if (idx < 0 || idx >= length) {
@ -2325,7 +2269,7 @@ scan_once_unicode(PyScannerObject *s, PyObject *pystr, Py_ssize_t idx, Py_ssize_
if ((idx + 2 < length) &&
PyUnicode_READ(kind, str, idx + 1) == 'a' &&
PyUnicode_READ(kind, str, idx + 2) == 'N') {
rval = _parse_constant(s, "NaN", idx, next_idx_ptr);
rval = _parse_constant(s, JSON_NaN, idx, next_idx_ptr);
}
else
fallthrough = 1;
@ -2340,7 +2284,7 @@ scan_once_unicode(PyScannerObject *s, PyObject *pystr, Py_ssize_t idx, Py_ssize_
PyUnicode_READ(kind, str, idx + 5) == 'i' &&
PyUnicode_READ(kind, str, idx + 6) == 't' &&
PyUnicode_READ(kind, str, idx + 7) == 'y') {
rval = _parse_constant(s, "Infinity", idx, next_idx_ptr);
rval = _parse_constant(s, JSON_Infinity, idx, next_idx_ptr);
}
else
fallthrough = 1;
@ -2356,7 +2300,7 @@ scan_once_unicode(PyScannerObject *s, PyObject *pystr, Py_ssize_t idx, Py_ssize_
PyUnicode_READ(kind, str, idx + 6) == 'i' &&
PyUnicode_READ(kind, str, idx + 7) == 't' &&
PyUnicode_READ(kind, str, idx + 8) == 'y') {
rval = _parse_constant(s, "-Infinity", idx, next_idx_ptr);
rval = _parse_constant(s, JSON_NegInfinity, idx, next_idx_ptr);
}
else
fallthrough = 1;
@ -2386,6 +2330,8 @@ scanner_call(PyObject *self, PyObject *args, PyObject *kwds)
return NULL;
if (PyUnicode_Check(pystr)) {
if (PyUnicode_READY(pystr))
return NULL;
rval = scan_once_unicode(s, pystr, idx, &next_idx);
}
#if PY_MAJOR_VERSION < 3
@ -2406,18 +2352,24 @@ scanner_call(PyObject *self, PyObject *args, PyObject *kwds)
static PyObject *
JSON_ParseEncoding(PyObject *encoding)
{
if (encoding == NULL)
return NULL;
if (encoding == Py_None)
return JSON_InternFromString(DEFAULT_ENCODING);
#if PY_MAJOR_VERSION < 3
if (PyUnicode_Check(encoding))
return PyUnicode_AsEncodedString(encoding, NULL, NULL);
#endif
if (JSON_ASCII_Check(encoding)) {
#if PY_MAJOR_VERSION >= 3
if (PyUnicode_Check(encoding)) {
if (PyUnicode_AsUTF8(encoding) == NULL) {
return NULL;
}
Py_INCREF(encoding);
return encoding;
}
#else /* PY_MAJOR_VERSION >= 3 */
if (PyString_Check(encoding)) {
Py_INCREF(encoding);
return encoding;
}
if (PyUnicode_Check(encoding))
return PyUnicode_AsEncodedString(encoding, NULL, NULL);
#endif /* PY_MAJOR_VERSION >= 3 */
PyErr_SetString(PyExc_TypeError, "encoding must be a string");
return NULL;
}
@ -2444,8 +2396,9 @@ scanner_new(PyTypeObject *type, PyObject *args, PyObject *kwds)
goto bail;
}
/* JSON_ASCII_AS_STRING is used on encoding */
encoding = PyObject_GetAttrString(ctx, "encoding");
if (encoding == NULL)
goto bail;
s->encoding = JSON_ParseEncoding(encoding);
Py_XDECREF(encoding);
if (s->encoding == NULL)
@ -2578,11 +2531,17 @@ encoder_new(PyTypeObject *type, PyObject *args, PyObject *kwds)
s->defaultfn = defaultfn;
Py_INCREF(encoder);
s->encoder = encoder;
#if PY_MAJOR_VERSION >= 3
if (encoding == Py_None) {
s->encoding = NULL;
}
else
#endif /* PY_MAJOR_VERSION >= 3 */
{
s->encoding = JSON_ParseEncoding(encoding);
if (s->encoding == NULL)
goto bail;
if (JSON_ASCII_AS_STRING(s->encoding) == NULL)
goto bail;
}
Py_INCREF(indent);
s->indent = indent;
Py_INCREF(key_separator);
@ -2768,28 +2727,16 @@ encoder_encode_float(PyEncoderObject *s, PyObject *obj)
}
/* JSON_ALLOW_NAN is set */
else if (i > 0) {
static PyObject *sInfinity = NULL;
if (sInfinity == NULL)
sInfinity = JSON_InternFromString("Infinity");
if (sInfinity)
Py_INCREF(sInfinity);
return sInfinity;
Py_INCREF(JSON_Infinity);
return JSON_Infinity;
}
else if (i < 0) {
static PyObject *sNegInfinity = NULL;
if (sNegInfinity == NULL)
sNegInfinity = JSON_InternFromString("-Infinity");
if (sNegInfinity)
Py_INCREF(sNegInfinity);
return sNegInfinity;
Py_INCREF(JSON_NegInfinity);
return JSON_NegInfinity;
}
else {
static PyObject *sNaN = NULL;
if (sNaN == NULL)
sNaN = JSON_InternFromString("NaN");
if (sNaN)
Py_INCREF(sNaN);
return sNaN;
Py_INCREF(JSON_NaN);
return JSON_NaN;
}
}
/* Use a better float format here? */
@ -2821,7 +2768,7 @@ encoder_encode_string(PyEncoderObject *s, PyObject *obj)
encoded = PyObject_CallFunctionObjArgs(s->encoder, obj, NULL);
if (encoded != NULL &&
#if PY_MAJOR_VERSION < 3
!JSON_ASCII_Check(encoded) &&
!PyString_Check(encoded) &&
#endif /* PY_MAJOR_VERSION < 3 */
!PyUnicode_Check(encoded))
{
@ -2854,18 +2801,13 @@ encoder_listencode_obj(PyEncoderObject *s, JSON_Accu *rval, PyObject *obj, Py_ss
if (cstr != NULL)
rv = _steal_accumulate(rval, cstr);
}
else if (PyString_Check(obj) || PyUnicode_Check(obj))
else if ((PyBytes_Check(obj) && s->encoding != NULL) ||
PyUnicode_Check(obj))
{
PyObject *encoded = encoder_encode_string(s, obj);
if (encoded != NULL)
rv = _steal_accumulate(rval, encoded);
}
else if (is_raw_json(obj))
{
PyObject *encoded = PyObject_GetAttrString(obj, "encoded_json");
if (encoded != NULL)
rv = _steal_accumulate(rval, encoded);
}
else if (PyInt_Check(obj) || PyLong_Check(obj)) {
PyObject *encoded;
if (PyInt_CheckExact(obj) || PyLong_CheckExact(obj)) {
@ -2933,6 +2875,12 @@ encoder_listencode_obj(PyEncoderObject *s, JSON_Accu *rval, PyObject *obj, Py_ss
if (encoded != NULL)
rv = _steal_accumulate(rval, encoded);
}
else if (is_raw_json(obj))
{
PyObject *encoded = PyObject_GetAttrString(obj, "encoded_json");
if (encoded != NULL)
rv = _steal_accumulate(rval, encoded);
}
else {
PyObject *ident = NULL;
PyObject *newobj;
@ -3366,6 +3314,32 @@ import_dependency(char *module_name, char *attr_name)
return rval;
}
static int
init_constants(void)
{
JSON_NaN = JSON_InternFromString("NaN");
if (JSON_NaN == NULL)
return 0;
JSON_Infinity = JSON_InternFromString("Infinity");
if (JSON_Infinity == NULL)
return 0;
JSON_NegInfinity = JSON_InternFromString("-Infinity");
if (JSON_NegInfinity == NULL)
return 0;
#if PY_MAJOR_VERSION >= 3
JSON_EmptyUnicode = PyUnicode_New(0, 127);
#else /* PY_MAJOR_VERSION >= 3 */
JSON_EmptyStr = PyString_FromString("");
if (JSON_EmptyStr == NULL)
return 0;
JSON_EmptyUnicode = PyUnicode_FromUnicode(NULL, 0);
#endif /* PY_MAJOR_VERSION >= 3 */
if (JSON_EmptyUnicode == NULL)
return 0;
return 1;
}
static PyObject *
moduleinit(void)
{
@ -3374,6 +3348,8 @@ moduleinit(void)
return NULL;
if (PyType_Ready(&PyEncoderType) < 0)
return NULL;
if (!init_constants())
return NULL;
#if PY_MAJOR_VERSION >= 3
m = PyModule_Create(&moduledef);

View file

@ -5,10 +5,11 @@ if sys.version_info[0] < 3:
PY3 = False
def b(s):
return s
def u(s):
return unicode(s, 'unicode_escape')
import cStringIO as StringIO
StringIO = BytesIO = StringIO.StringIO
try:
from cStringIO import StringIO
except ImportError:
from StringIO import StringIO
BytesIO = StringIO
text_type = unicode
binary_type = str
string_types = (basestring,)
@ -21,20 +22,13 @@ else:
from importlib import reload as reload_module
else:
from imp import reload as reload_module
import codecs
def b(s):
return codecs.latin_1_encode(s)[0]
def u(s):
return s
import io
StringIO = io.StringIO
BytesIO = io.BytesIO
return bytes(s, 'latin1')
from io import StringIO, BytesIO
text_type = str
binary_type = bytes
string_types = (str,)
integer_types = (int,)
def unichr(s):
return u(chr(s))
unichr = chr
long_type = integer_types[-1]

View file

@ -4,7 +4,7 @@ from __future__ import absolute_import
import re
import sys
import struct
from .compat import u, text_type, binary_type, PY3, unichr
from .compat import PY3, unichr
from .scanner import make_scanner, JSONDecodeError
def _import_c_scanstring():
@ -40,14 +40,14 @@ _CONSTANTS = {
STRINGCHUNK = re.compile(r'(.*?)(["\\\x00-\x1f])', FLAGS)
BACKSLASH = {
'"': u('"'), '\\': u('\u005c'), '/': u('/'),
'b': u('\b'), 'f': u('\f'), 'n': u('\n'), 'r': u('\r'), 't': u('\t'),
'"': u'"', '\\': u'\\', '/': u'/',
'b': u'\b', 'f': u'\f', 'n': u'\n', 'r': u'\r', 't': u'\t',
}
DEFAULT_ENCODING = "utf-8"
def py_scanstring(s, end, encoding=None, strict=True,
_b=BACKSLASH, _m=STRINGCHUNK.match, _join=u('').join,
_b=BACKSLASH, _m=STRINGCHUNK.match, _join=u''.join,
_PY3=PY3, _maxunicode=sys.maxunicode):
"""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.
@ -71,8 +71,8 @@ def py_scanstring(s, end, encoding=None, strict=True,
content, terminator = chunk.groups()
# Content is contains zero or more unescaped string characters
if content:
if not _PY3 and not isinstance(content, text_type):
content = text_type(content, encoding)
if not _PY3 and not isinstance(content, unicode):
content = unicode(content, encoding)
_append(content)
# Terminator is the end of string, a literal control character,
# or a backslash denoting that an escape sequence follows
@ -365,8 +365,8 @@ class JSONDecoder(object):
instance containing a JSON document)
"""
if _PY3 and isinstance(s, binary_type):
s = s.decode(self.encoding)
if _PY3 and isinstance(s, bytes):
s = str(s, self.encoding)
obj, end = self.raw_decode(s)
end = _w(s, end).end()
if end != len(s):
@ -388,7 +388,7 @@ class JSONDecoder(object):
# 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):
if _PY3 and not isinstance(s, str):
raise TypeError("Input string must be text, not bytes")
# strip UTF-8 bom
if len(s) > idx:

View file

@ -5,7 +5,7 @@ import re
from operator import itemgetter
# Do not import Decimal directly to avoid reload issues
import decimal
from .compat import u, unichr, binary_type, text_type, string_types, integer_types, PY3
from .compat import unichr, binary_type, text_type, string_types, integer_types, PY3
def _import_speedups():
try:
from . import _speedups
@ -17,10 +17,7 @@ c_encode_basestring_ascii, c_make_encoder = _import_speedups()
from .decoder import PosInf
from .raw_json import RawJSON
#ESCAPE = re.compile(ur'[\x00-\x1f\\"\b\f\n\r\t\u2028\u2029]')
# This is required because u() will mangle the string and ur'' isn't valid
# python3 syntax
ESCAPE = re.compile(u'[\\x00-\\x1f\\\\"\\b\\f\\n\\r\\t\u2028\u2029]')
ESCAPE = re.compile(r'[\x00-\x1f\\"]')
ESCAPE_ASCII = re.compile(r'([\\"]|[^\ -~])')
HAS_UTF8 = re.compile(r'[\x80-\xff]')
ESCAPE_DCT = {
@ -35,24 +32,27 @@ 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,))
for i in [0x2028, 0x2029]:
ESCAPE_DCT.setdefault(unichr(i), '\\u%04x' % (i,))
FLOAT_REPR = repr
def encode_basestring(s, _PY3=PY3, _q=u('"')):
def encode_basestring(s, _PY3=PY3, _q=u'"'):
"""Return a JSON representation of a Python string
"""
if _PY3:
if isinstance(s, binary_type):
s = s.decode('utf-8')
if type(s) is not text_type:
s = text_type.__str__(s)
if isinstance(s, bytes):
s = str(s, 'utf-8')
elif type(s) is not str:
# convert an str subclass instance to exact str
# raise a TypeError otherwise
s = str.__str__(s)
else:
if isinstance(s, str) and HAS_UTF8.search(s) is not None:
s = s.decode('utf-8')
if type(s) not in string_types:
s = unicode(s, 'utf-8')
elif type(s) not in (str, unicode):
# convert an str subclass instance to exact str
# convert a unicode subclass instance to exact unicode
# raise a TypeError otherwise
if isinstance(s, str):
s = str.__str__(s)
else:
@ -67,14 +67,19 @@ def py_encode_basestring_ascii(s, _PY3=PY3):
"""
if _PY3:
if isinstance(s, binary_type):
s = s.decode('utf-8')
if type(s) is not text_type:
s = text_type.__str__(s)
if isinstance(s, bytes):
s = str(s, 'utf-8')
elif type(s) is not str:
# convert an str subclass instance to exact str
# raise a TypeError otherwise
s = str.__str__(s)
else:
if isinstance(s, str) and HAS_UTF8.search(s) is not None:
s = s.decode('utf-8')
if type(s) not in string_types:
s = unicode(s, 'utf-8')
elif type(s) not in (str, unicode):
# convert an str subclass instance to exact str
# convert a unicode subclass instance to exact unicode
# raise a TypeError otherwise
if isinstance(s, str):
s = str.__str__(s)
else:
@ -279,7 +284,7 @@ class JSONEncoder(object):
if isinstance(o, binary_type):
_encoding = self.encoding
if (_encoding is not None and not (_encoding == 'utf-8')):
o = o.decode(_encoding)
o = text_type(o, _encoding)
if isinstance(o, string_types):
if self.ensure_ascii:
return encode_basestring_ascii(o)
@ -314,10 +319,10 @@ class JSONEncoder(object):
_encoder = encode_basestring_ascii
else:
_encoder = encode_basestring
if self.encoding != 'utf-8':
if self.encoding != 'utf-8' and self.encoding is not None:
def _encoder(o, _orig_encoder=_encoder, _encoding=self.encoding):
if isinstance(o, binary_type):
o = o.decode(_encoding)
o = text_type(o, _encoding)
return _orig_encoder(o)
def floatstr(o, allow_nan=self.allow_nan, ignore_nan=self.ignore_nan,
@ -382,6 +387,11 @@ class JSONEncoderForHTML(JSONEncoder):
characters &, < and > should be escaped. They cannot be escaped
with the usual entities (e.g. &amp;) because they are not expanded
within <script> tags.
This class also escapes the line separator and paragraph separator
characters U+2028 and U+2029, irrespective of the ensure_ascii setting,
as these characters are not valid in JavaScript strings (see
http://timelessrepo.com/json-isnt-a-javascript-subset).
"""
def encode(self, o):
@ -399,6 +409,11 @@ class JSONEncoderForHTML(JSONEncoder):
chunk = chunk.replace('&', '\\u0026')
chunk = chunk.replace('<', '\\u003c')
chunk = chunk.replace('>', '\\u003e')
if not self.ensure_ascii:
chunk = chunk.replace(u'\u2028', '\\u2028')
chunk = chunk.replace(u'\u2029', '\\u2029')
yield chunk
@ -477,8 +492,9 @@ def _make_iterencode(markers, _default, _encoder, _indent, _floatstr,
first = False
else:
buf = separator
if (isinstance(value, string_types) or
(_PY3 and isinstance(value, binary_type))):
if isinstance(value, string_types):
yield buf + _encoder(value)
elif _PY3 and isinstance(value, bytes) and _encoding is not None:
yield buf + _encoder(value)
elif isinstance(value, RawJSON):
yield buf + value.encoded_json
@ -528,8 +544,8 @@ def _make_iterencode(markers, _default, _encoder, _indent, _floatstr,
def _stringify_key(key):
if isinstance(key, string_types): # pragma: no cover
pass
elif isinstance(key, binary_type):
key = key.decode(_encoding)
elif _PY3 and isinstance(key, bytes) and _encoding is not None:
key = str(key, _encoding)
elif isinstance(key, float):
key = _floatstr(key)
elif key is True:
@ -598,8 +614,9 @@ def _make_iterencode(markers, _default, _encoder, _indent, _floatstr,
yield item_separator
yield _encoder(key)
yield _key_separator
if (isinstance(value, string_types) or
(_PY3 and isinstance(value, binary_type))):
if isinstance(value, string_types):
yield _encoder(value)
elif _PY3 and isinstance(value, bytes) and _encoding is not None:
yield _encoder(value)
elif isinstance(value, RawJSON):
yield value.encoded_json
@ -642,8 +659,9 @@ def _make_iterencode(markers, _default, _encoder, _indent, _floatstr,
del markers[markerid]
def _iterencode(o, _current_indent_level):
if (isinstance(o, string_types) or
(_PY3 and isinstance(o, binary_type))):
if isinstance(o, string_types):
yield _encoder(o)
elif _PY3 and isinstance(o, bytes) and _encoding is not None:
yield _encoder(o)
elif isinstance(o, RawJSON):
yield o.encoded_json