Serhiy Storchaka added the comment:
> The 2.7 comments also apply to the 3.2 patch. Otherwise the 3.2 patch
> (without the _sre changes :) looks good to me.
Patches updated addressing new Stefan's comments. Unicode globals no longer
reinitialized in _PyUnicode_Init(). Note that I have added a consistency check
into the macro in 3.3+.
I hope Rietveld will accept this set.
----------
Added file: http://bugs.python.org/file28826/unicode_globals-2.7_3.patch
Added file: http://bugs.python.org/file28827/unicode_globals-3.2_3.patch
Added file: http://bugs.python.org/file28828/unicode_globals-3.3_3.patch
Added file: http://bugs.python.org/file28829/unicode_globals-3.4_3.patch
_______________________________________
Python tracker <rep...@bugs.python.org>
<http://bugs.python.org/issue10156>
_______________________________________
diff -r 864b9836dae6 Objects/unicodeobject.c
--- a/Objects/unicodeobject.c Fri Jan 25 17:55:39 2013 +0100
+++ b/Objects/unicodeobject.c Fri Jan 25 21:34:01 2013 +0200
@@ -82,8 +82,9 @@
/* --- Globals ------------------------------------------------------------
- The globals are initialized by the _PyUnicode_Init() API and should
- not be used before calling that API.
+NOTE: In the interpreter's initialization phase, some globals are currently
+ initialized dynamically as needed. In the process Unicode objects may
+ be created before the Unicode type is ready.
*/
@@ -93,15 +94,27 @@
#endif
/* Free list for Unicode objects */
-static PyUnicodeObject *free_list;
-static int numfree;
+static PyUnicodeObject *free_list = NULL;
+static int numfree = 0;
/* The empty Unicode object is shared to improve performance. */
-static PyUnicodeObject *unicode_empty;
+static PyUnicodeObject *unicode_empty = NULL;
+
+#define _Py_RETURN_UNICODE_EMPTY() \
+ do { \
+ if (unicode_empty != NULL) \
+ Py_INCREF(unicode_empty); \
+ else { \
+ unicode_empty = _PyUnicode_New(0); \
+ if (unicode_empty != NULL) \
+ Py_INCREF(unicode_empty); \
+ } \
+ return (PyObject *)unicode_empty; \
+ } while (0)
/* Single character Unicode strings in the Latin-1 range are being
shared as well. */
-static PyUnicodeObject *unicode_latin1[256];
+static PyUnicodeObject *unicode_latin1[256] = {NULL};
/* Default encoding to use and assume when NULL is passed as encoding
parameter; it is initialized by _PyUnicode_Init().
@@ -110,7 +123,7 @@
PyUnicode_GetDefaultEncoding() APIs to access this global.
*/
-static char unicode_default_encoding[100];
+static char unicode_default_encoding[100 + 1] = "ascii";
/* Fast detection of the most frequent whitespace characters */
const unsigned char _Py_ascii_whitespace[] = {
@@ -204,7 +217,7 @@
#define BLOOM_MASK unsigned long
-static BLOOM_MASK bloom_linebreak;
+static BLOOM_MASK bloom_linebreak = ~(BLOOM_MASK)0;
#define BLOOM_ADD(mask, ch) ((mask |= (1UL << ((ch) & (BLOOM_WIDTH - 1)))))
#define BLOOM(mask, ch) ((mask & (1UL << ((ch) & (BLOOM_WIDTH - 1)))))
@@ -448,10 +461,8 @@
if (u != NULL) {
/* Optimization for empty strings */
- if (size == 0 && unicode_empty != NULL) {
- Py_INCREF(unicode_empty);
- return (PyObject *)unicode_empty;
- }
+ if (size == 0)
+ _Py_RETURN_UNICODE_EMPTY();
/* Single character Unicode objects in the Latin-1 range are
shared when using this constructor */
@@ -497,10 +508,8 @@
if (u != NULL) {
/* Optimization for empty strings */
- if (size == 0 && unicode_empty != NULL) {
- Py_INCREF(unicode_empty);
- return (PyObject *)unicode_empty;
- }
+ if (size == 0)
+ _Py_RETURN_UNICODE_EMPTY();
/* Single characters are shared when using this constructor.
Restrict to ASCII, since the input must be UTF-8. */
@@ -1162,13 +1171,10 @@
}
/* Convert to Unicode */
- if (len == 0) {
- Py_INCREF(unicode_empty);
- v = (PyObject *)unicode_empty;
- }
- else
- v = PyUnicode_Decode(s, len, encoding, errors);
-
+ if (len == 0)
+ _Py_RETURN_UNICODE_EMPTY();
+
+ v = PyUnicode_Decode(s, len, encoding, errors);
return v;
onError:
@@ -1381,7 +1387,7 @@
Py_DECREF(v);
strncpy(unicode_default_encoding,
encoding,
- sizeof(unicode_default_encoding));
+ sizeof(unicode_default_encoding) - 1);
return 0;
onError:
@@ -8850,8 +8856,6 @@
void _PyUnicode_Init(void)
{
- int i;
-
/* XXX - move this array to unicodectype.c ? */
Py_UNICODE linebreak[] = {
0x000A, /* LINE FEED */
@@ -8865,15 +8869,12 @@
};
/* Init the implementation */
- free_list = NULL;
- numfree = 0;
- unicode_empty = _PyUnicode_New(0);
- if (!unicode_empty)
- return;
-
- strcpy(unicode_default_encoding, "ascii");
- for (i = 0; i < 256; i++)
- unicode_latin1[i] = NULL;
+ if (!unicode_empty) {
+ unicode_empty = _PyUnicode_New(0);
+ if (!unicode_empty)
+ return;
+ }
+
if (PyType_Ready(&PyUnicode_Type) < 0)
Py_FatalError("Can't initialize 'unicode'");
@@ -8918,15 +8919,11 @@
{
int i;
- Py_XDECREF(unicode_empty);
- unicode_empty = NULL;
-
- for (i = 0; i < 256; i++) {
- if (unicode_latin1[i]) {
- Py_DECREF(unicode_latin1[i]);
- unicode_latin1[i] = NULL;
- }
- }
+ Py_CLEAR(unicode_empty);
+
+ for (i = 0; i < 256; i++)
+ Py_CLEAR(unicode_latin1[i]);
+
(void)PyUnicode_ClearFreeList();
}
diff -r ea9fd9c9c677 Objects/unicodeobject.c
--- a/Objects/unicodeobject.c Fri Jan 25 17:57:13 2013 +0100
+++ b/Objects/unicodeobject.c Fri Jan 25 21:34:11 2013 +0200
@@ -80,8 +80,9 @@
/* --- Globals ------------------------------------------------------------
- The globals are initialized by the _PyUnicode_Init() API and should
- not be used before calling that API.
+NOTE: In the interpreter's initialization phase, some globals are currently
+ initialized dynamically as needed. In the process Unicode objects may
+ be created before the Unicode type is ready.
*/
@@ -98,18 +99,30 @@
Another way to look at this is that to say that the actual reference
count of a string is: s->ob_refcnt + (s->state ? 2 : 0)
*/
-static PyObject *interned;
+static PyObject *interned = NULL;
/* Free list for Unicode objects */
-static PyUnicodeObject *free_list;
-static int numfree;
+static PyUnicodeObject *free_list = NULL;
+static int numfree = 0;
/* The empty Unicode object is shared to improve performance. */
-static PyUnicodeObject *unicode_empty;
+static PyUnicodeObject *unicode_empty = NULL;
+
+#define _Py_RETURN_UNICODE_EMPTY() \
+ do { \
+ if (unicode_empty != NULL) \
+ Py_INCREF(unicode_empty); \
+ else { \
+ unicode_empty = _PyUnicode_New(0); \
+ if (unicode_empty != NULL) \
+ Py_INCREF(unicode_empty); \
+ } \
+ return (PyObject *)unicode_empty; \
+ } while (0)
/* Single character Unicode strings in the Latin-1 range are being
shared as well. */
-static PyUnicodeObject *unicode_latin1[256];
+static PyUnicodeObject *unicode_latin1[256] = {NULL};
/* Fast detection of the most frequent whitespace characters */
const unsigned char _Py_ascii_whitespace[] = {
@@ -214,7 +227,7 @@
#define BLOOM_MASK unsigned long
-static BLOOM_MASK bloom_linebreak;
+static BLOOM_MASK bloom_linebreak = ~(BLOOM_MASK)0;
#define BLOOM_ADD(mask, ch) ((mask |= (1UL << ((ch) & (BLOOM_WIDTH - 1)))))
#define BLOOM(mask, ch) ((mask & (1UL << ((ch) & (BLOOM_WIDTH - 1)))))
@@ -479,10 +492,8 @@
if (u != NULL) {
/* Optimization for empty strings */
- if (size == 0 && unicode_empty != NULL) {
- Py_INCREF(unicode_empty);
- return (PyObject *)unicode_empty;
- }
+ if (size == 0)
+ _Py_RETURN_UNICODE_EMPTY();
/* Single character Unicode objects in the Latin-1 range are
shared when using this constructor */
@@ -528,10 +539,8 @@
if (u != NULL) {
/* Optimization for empty strings */
- if (size == 0 && unicode_empty != NULL) {
- Py_INCREF(unicode_empty);
- return (PyObject *)unicode_empty;
- }
+ if (size == 0)
+ _Py_RETURN_UNICODE_EMPTY();
/* Single characters are shared when using this constructor.
Restrict to ASCII, since the input must be UTF-8. */
@@ -1393,15 +1402,11 @@
/* Decoding bytes objects is the most common case and should be fast */
if (PyBytes_Check(obj)) {
- if (PyBytes_GET_SIZE(obj) == 0) {
- Py_INCREF(unicode_empty);
- v = (PyObject *) unicode_empty;
- }
- else {
- v = PyUnicode_Decode(
- PyBytes_AS_STRING(obj), PyBytes_GET_SIZE(obj),
- encoding, errors);
- }
+ if (PyBytes_GET_SIZE(obj) == 0)
+ _Py_RETURN_UNICODE_EMPTY();
+ v = PyUnicode_Decode(
+ PyBytes_AS_STRING(obj), PyBytes_GET_SIZE(obj),
+ encoding, errors);
return v;
}
@@ -1421,12 +1426,11 @@
}
if (buffer.len == 0) {
- Py_INCREF(unicode_empty);
- v = (PyObject *) unicode_empty;
- }
- else
- v = PyUnicode_Decode((char*) buffer.buf, buffer.len, encoding, errors);
-
+ PyBuffer_Release(&buffer);
+ _Py_RETURN_UNICODE_EMPTY();
+ }
+
+ v = PyUnicode_Decode((char*) buffer.buf, buffer.len, encoding, errors);
PyBuffer_Release(&buffer);
return v;
}
@@ -8323,10 +8327,8 @@
Py_ssize_t nchars;
size_t nbytes;
- if (len < 1) {
- Py_INCREF(unicode_empty);
- return (PyObject *)unicode_empty;
- }
+ if (len < 1)
+ _Py_RETURN_UNICODE_EMPTY();
if (len == 1 && PyUnicode_CheckExact(str)) {
/* no repeat, return original string */
@@ -10056,8 +10058,6 @@
void _PyUnicode_Init(void)
{
- int i;
-
/* XXX - move this array to unicodectype.c ? */
Py_UNICODE linebreak[] = {
0x000A, /* LINE FEED */
@@ -10071,14 +10071,12 @@
};
/* Init the implementation */
- free_list = NULL;
- numfree = 0;
- unicode_empty = _PyUnicode_New(0);
- if (!unicode_empty)
- return;
-
- for (i = 0; i < 256; i++)
- unicode_latin1[i] = NULL;
+ if (!unicode_empty) {
+ unicode_empty = _PyUnicode_New(0);
+ if (!unicode_empty)
+ return;
+ }
+
if (PyType_Ready(&PyUnicode_Type) < 0)
Py_FatalError("Can't initialize 'unicode'");
@@ -10123,15 +10121,11 @@
{
int i;
- Py_XDECREF(unicode_empty);
- unicode_empty = NULL;
-
- for (i = 0; i < 256; i++) {
- if (unicode_latin1[i]) {
- Py_DECREF(unicode_latin1[i]);
- unicode_latin1[i] = NULL;
- }
- }
+ Py_CLEAR(unicode_empty);
+
+ for (i = 0; i < 256; i++)
+ Py_CLEAR(unicode_latin1[i]);
+
(void)PyUnicode_ClearFreeList();
}
@@ -10250,8 +10244,7 @@
"mortal/immortal\n", mortal_size, immortal_size);
Py_DECREF(keys);
PyDict_Clear(interned);
- Py_DECREF(interned);
- interned = NULL;
+ Py_CLEAR(interned);
}
diff -r ff5cf42136b3 Objects/unicodeobject.c
--- a/Objects/unicodeobject.c Fri Jan 25 18:45:12 2013 +0100
+++ b/Objects/unicodeobject.c Fri Jan 25 21:36:29 2013 +0200
@@ -57,8 +57,9 @@
/* --- Globals ------------------------------------------------------------
- The globals are initialized by the _PyUnicode_Init() API and should
- not be used before calling that API.
+NOTE: In the interpreter's initialization phase, some globals are currently
+ initialized dynamically as needed. In the process Unicode objects may
+ be created before the Unicode type is ready.
*/
@@ -179,17 +180,36 @@
Another way to look at this is that to say that the actual reference
count of a string is: s->ob_refcnt + (s->state ? 2 : 0)
*/
-static PyObject *interned;
+static PyObject *interned = NULL;
/* The empty Unicode object is shared to improve performance. */
-static PyObject *unicode_empty;
+static PyObject *unicode_empty = NULL;
+
+#define _Py_INCREF_UNICODE_EMPTY() \
+ do { \
+ if (unicode_empty != NULL) \
+ Py_INCREF(unicode_empty); \
+ else { \
+ unicode_empty = PyUnicode_New(0, 0); \
+ if (unicode_empty != NULL) { \
+ Py_INCREF(unicode_empty); \
+ assert(_PyUnicode_CheckConsistency(unicode_empty, 1)); \
+ } \
+ } \
+ } while (0)
+
+#define _Py_RETURN_UNICODE_EMPTY() \
+ do { \
+ _Py_INCREF_UNICODE_EMPTY(); \
+ return unicode_empty; \
+ } while (0)
/* List of static strings. */
-static _Py_Identifier *static_strings;
+static _Py_Identifier *static_strings = NULL;
/* Single character Unicode strings in the Latin-1 range are being
shared as well. */
-static PyObject *unicode_latin1[256];
+static PyObject *unicode_latin1[256] = {NULL};
/* Fast detection of the most frequent whitespace characters */
const unsigned char _Py_ascii_whitespace[] = {
@@ -416,9 +436,8 @@
len = _PyUnicode_WSTR_LENGTH(unicode);
if (len == 0) {
- Py_INCREF(unicode_empty);
Py_DECREF(unicode);
- return unicode_empty;
+ _Py_RETURN_UNICODE_EMPTY();
}
if (len == 1) {
@@ -450,8 +469,8 @@
length = PyUnicode_GET_LENGTH(unicode);
if (length == 0) {
if (unicode != unicode_empty) {
- Py_INCREF(unicode_empty);
Py_DECREF(unicode);
+ _Py_RETURN_UNICODE_EMPTY();
}
return unicode_empty;
}
@@ -528,7 +547,7 @@
#define BLOOM_MASK unsigned long
-static BLOOM_MASK bloom_linebreak;
+static BLOOM_MASK bloom_linebreak = ~(BLOOM_MASK)0;
#define BLOOM_ADD(mask, ch) ((mask |= (1UL << ((ch) & (BLOOM_WIDTH - 1)))))
#define BLOOM(mask, ch) ((mask & (1UL << ((ch) & (BLOOM_WIDTH - 1)))))
@@ -1582,9 +1601,11 @@
return 0;
if (length == 0) {
+ _Py_INCREF_UNICODE_EMPTY();
+ if (!unicode_empty)
+ return -1;
Py_DECREF(*p_unicode);
*p_unicode = unicode_empty;
- Py_INCREF(*p_unicode);
return 0;
}
@@ -1731,10 +1752,8 @@
some optimizations which share commonly used objects. */
/* Optimization for empty strings */
- if (size == 0 && unicode_empty != NULL) {
- Py_INCREF(unicode_empty);
- return unicode_empty;
- }
+ if (size == 0)
+ _Py_RETURN_UNICODE_EMPTY();
/* Single character Unicode objects in the Latin-1 range are
shared when using this constructor */
@@ -1893,10 +1912,8 @@
PyObject *res;
unsigned char max_char;
- if (size == 0) {
- Py_INCREF(unicode_empty);
- return unicode_empty;
- }
+ if (size == 0)
+ _Py_RETURN_UNICODE_EMPTY();
assert(size > 0);
if (size == 1)
return get_latin1_char(u[0]);
@@ -1916,10 +1933,8 @@
PyObject *res;
Py_UCS2 max_char;
- if (size == 0) {
- Py_INCREF(unicode_empty);
- return unicode_empty;
- }
+ if (size == 0)
+ _Py_RETURN_UNICODE_EMPTY();
assert(size > 0);
if (size == 1) {
Py_UCS4 ch = u[0];
@@ -1954,10 +1969,8 @@
PyObject *res;
Py_UCS4 max_char;
- if (size == 0) {
- Py_INCREF(unicode_empty);
- return unicode_empty;
- }
+ if (size == 0)
+ _Py_RETURN_UNICODE_EMPTY();
assert(size > 0);
if (size == 1) {
Py_UCS4 ch = u[0];
@@ -2249,10 +2262,8 @@
PyUnicode_FromWideChar(register const wchar_t *w, Py_ssize_t size)
{
if (w == NULL) {
- if (size == 0) {
- Py_INCREF(unicode_empty);
- return unicode_empty;
- }
+ if (size == 0)
+ _Py_RETURN_UNICODE_EMPTY();
PyErr_BadInternalCall();
return NULL;
}
@@ -3007,15 +3018,11 @@
/* Decoding bytes objects is the most common case and should be fast */
if (PyBytes_Check(obj)) {
- if (PyBytes_GET_SIZE(obj) == 0) {
- Py_INCREF(unicode_empty);
- v = unicode_empty;
- }
- else {
- v = PyUnicode_Decode(
- PyBytes_AS_STRING(obj), PyBytes_GET_SIZE(obj),
- encoding, errors);
- }
+ if (PyBytes_GET_SIZE(obj) == 0)
+ _Py_RETURN_UNICODE_EMPTY();
+ v = PyUnicode_Decode(
+ PyBytes_AS_STRING(obj), PyBytes_GET_SIZE(obj),
+ encoding, errors);
return v;
}
@@ -3035,12 +3042,11 @@
}
if (buffer.len == 0) {
- Py_INCREF(unicode_empty);
- v = unicode_empty;
- }
- else
- v = PyUnicode_Decode((char*) buffer.buf, buffer.len, encoding, errors);
-
+ PyBuffer_Release(&buffer);
+ _Py_RETURN_UNICODE_EMPTY();
+ }
+
+ v = PyUnicode_Decode((char*) buffer.buf, buffer.len, encoding, errors);
PyBuffer_Release(&buffer);
return v;
}
@@ -4720,8 +4726,7 @@
if (size == 0) {
if (consumed)
*consumed = 0;
- Py_INCREF(unicode_empty);
- return unicode_empty;
+ _Py_RETURN_UNICODE_EMPTY();
}
/* ASCII is equivalent to the first 128 ordinals in Unicode. */
@@ -5232,8 +5237,7 @@
if (q == e) {
if (consumed)
*consumed = size;
- Py_INCREF(unicode_empty);
- return unicode_empty;
+ _Py_RETURN_UNICODE_EMPTY();
}
#ifdef BYTEORDER_IS_LITTLE_ENDIAN
@@ -6558,10 +6562,8 @@
PyObject *errorHandler = NULL;
PyObject *exc = NULL;
- if (size == 0) {
- Py_INCREF(unicode_empty);
- return unicode_empty;
- }
+ if (size == 0)
+ _Py_RETURN_UNICODE_EMPTY();
/* ASCII is equivalent to the first 128 ordinals in Unicode. */
if (size == 1 && (unsigned char)s[0] < 128)
@@ -6940,8 +6942,7 @@
if (chunk_size == 0 && done) {
if (v != NULL)
break;
- Py_INCREF(unicode_empty);
- return unicode_empty;
+ _Py_RETURN_UNICODE_EMPTY();
}
@@ -9503,9 +9504,7 @@
/* If empty sequence, return u"". */
if (seqlen == 0) {
Py_DECREF(fseq);
- Py_INCREF(unicode_empty);
- res = unicode_empty;
- return res;
+ _Py_RETURN_UNICODE_EMPTY();
}
/* If singleton sequence with an exact Unicode, return that. */
@@ -10205,7 +10204,9 @@
}
new_size = slen + n * (len2 - len1);
if (new_size == 0) {
- Py_INCREF(unicode_empty);
+ _Py_INCREF_UNICODE_EMPTY();
+ if (!unicode_empty)
+ goto error;
u = unicode_empty;
goto done;
}
@@ -11672,10 +11673,8 @@
PyErr_SetString(PyExc_IndexError, "string index out of range");
return NULL;
}
- if (start >= length || end < start) {
- Py_INCREF(unicode_empty);
- return unicode_empty;
- }
+ if (start >= length || end < start)
+ _Py_RETURN_UNICODE_EMPTY();
length = end - start;
if (PyUnicode_IS_ASCII(self)) {
@@ -11802,10 +11801,8 @@
PyObject *u;
Py_ssize_t nchars, n;
- if (len < 1) {
- Py_INCREF(unicode_empty);
- return unicode_empty;
- }
+ if (len < 1)
+ _Py_RETURN_UNICODE_EMPTY();
/* no repeat, return original string */
if (len == 1)
@@ -12924,8 +12921,7 @@
{
if (writer->pos == 0) {
Py_XDECREF(writer->buffer);
- Py_INCREF(unicode_empty);
- return unicode_empty;
+ _Py_RETURN_UNICODE_EMPTY();
}
if (writer->readonly) {
assert(PyUnicode_GET_LENGTH(writer->buffer) == writer->pos);
@@ -13143,8 +13139,7 @@
}
if (slicelength <= 0) {
- Py_INCREF(unicode_empty);
- return unicode_empty;
+ _Py_RETURN_UNICODE_EMPTY();
} else if (start == 0 && step == 1 &&
slicelength == PyUnicode_GET_LENGTH(self)) {
return unicode_result_unchanged(self);
@@ -13974,10 +13969,8 @@
if (!PyArg_ParseTupleAndKeywords(args, kwds, "|Oss:str",
kwlist, &x, &encoding, &errors))
return NULL;
- if (x == NULL) {
- Py_INCREF(unicode_empty);
- return unicode_empty;
- }
+ if (x == NULL)
+ _Py_RETURN_UNICODE_EMPTY();
if (encoding == NULL && errors == NULL)
return PyObject_Str(x);
else
@@ -14146,8 +14139,6 @@
int _PyUnicode_Init(void)
{
- int i;
-
/* XXX - move this array to unicodectype.c ? */
Py_UCS2 linebreak[] = {
0x000A, /* LINE FEED */
@@ -14161,13 +14152,11 @@
};
/* Init the implementation */
- unicode_empty = PyUnicode_New(0, 0);
+ _Py_INCREF_UNICODE_EMPTY();
if (!unicode_empty)
Py_FatalError("Can't create empty string");
- assert(_PyUnicode_CheckConsistency(unicode_empty, 1));
-
- for (i = 0; i < 256; i++)
- unicode_latin1[i] = NULL;
+ Py_DECREF(unicode_empty);
+
if (PyType_Ready(&PyUnicode_Type) < 0)
Py_FatalError("Can't initialize 'unicode'");
@@ -14207,15 +14196,10 @@
{
int i;
- Py_XDECREF(unicode_empty);
- unicode_empty = NULL;
-
- for (i = 0; i < 256; i++) {
- if (unicode_latin1[i]) {
- Py_DECREF(unicode_latin1[i]);
- unicode_latin1[i] = NULL;
- }
- }
+ Py_CLEAR(unicode_empty);
+
+ for (i = 0; i < 256; i++)
+ Py_CLEAR(unicode_latin1[i]);
_PyUnicode_ClearStaticStrings();
(void)PyUnicode_ClearFreeList();
}
@@ -14344,8 +14328,7 @@
"mortal/immortal\n", mortal_size, immortal_size);
Py_DECREF(keys);
PyDict_Clear(interned);
- Py_DECREF(interned);
- interned = NULL;
+ Py_CLEAR(interned);
}
diff -r 1942987921e9 Objects/unicodeobject.c
--- a/Objects/unicodeobject.c Fri Jan 25 18:45:41 2013 +0100
+++ b/Objects/unicodeobject.c Fri Jan 25 21:36:35 2013 +0200
@@ -49,8 +49,9 @@
/* --- Globals ------------------------------------------------------------
- The globals are initialized by the _PyUnicode_Init() API and should
- not be used before calling that API.
+NOTE: In the interpreter's initialization phase, some globals are currently
+ initialized dynamically as needed. In the process Unicode objects may
+ be created before the Unicode type is ready.
*/
@@ -171,17 +172,36 @@
Another way to look at this is that to say that the actual reference
count of a string is: s->ob_refcnt + (s->state ? 2 : 0)
*/
-static PyObject *interned;
+static PyObject *interned = NULL;
/* The empty Unicode object is shared to improve performance. */
-static PyObject *unicode_empty;
+static PyObject *unicode_empty = NULL;
+
+#define _Py_INCREF_UNICODE_EMPTY() \
+ do { \
+ if (unicode_empty != NULL) \
+ Py_INCREF(unicode_empty); \
+ else { \
+ unicode_empty = PyUnicode_New(0, 0); \
+ if (unicode_empty != NULL) { \
+ Py_INCREF(unicode_empty); \
+ assert(_PyUnicode_CheckConsistency(unicode_empty, 1)); \
+ } \
+ } \
+ } while (0)
+
+#define _Py_RETURN_UNICODE_EMPTY() \
+ do { \
+ _Py_INCREF_UNICODE_EMPTY(); \
+ return unicode_empty; \
+ } while (0)
/* List of static strings. */
-static _Py_Identifier *static_strings;
+static _Py_Identifier *static_strings = NULL;
/* Single character Unicode strings in the Latin-1 range are being
shared as well. */
-static PyObject *unicode_latin1[256];
+static PyObject *unicode_latin1[256] = {NULL};
/* Fast detection of the most frequent whitespace characters */
const unsigned char _Py_ascii_whitespace[] = {
@@ -406,9 +426,8 @@
len = _PyUnicode_WSTR_LENGTH(unicode);
if (len == 0) {
- Py_INCREF(unicode_empty);
Py_DECREF(unicode);
- return unicode_empty;
+ _Py_RETURN_UNICODE_EMPTY();
}
if (len == 1) {
@@ -442,8 +461,8 @@
length = PyUnicode_GET_LENGTH(unicode);
if (length == 0) {
if (unicode != unicode_empty) {
- Py_INCREF(unicode_empty);
Py_DECREF(unicode);
+ _Py_RETURN_UNICODE_EMPTY();
}
return unicode_empty;
}
@@ -520,7 +539,7 @@
#define BLOOM_MASK unsigned long
-static BLOOM_MASK bloom_linebreak;
+static BLOOM_MASK bloom_linebreak = ~(BLOOM_MASK)0;
#define BLOOM_ADD(mask, ch) ((mask |= (1UL << ((ch) & (BLOOM_WIDTH - 1)))))
#define BLOOM(mask, ch) ((mask & (1UL << ((ch) & (BLOOM_WIDTH - 1)))))
@@ -1602,9 +1621,11 @@
return 0;
if (length == 0) {
+ _Py_INCREF_UNICODE_EMPTY();
+ if (!unicode_empty)
+ return -1;
Py_DECREF(*p_unicode);
*p_unicode = unicode_empty;
- Py_INCREF(*p_unicode);
return 0;
}
@@ -1727,10 +1748,8 @@
some optimizations which share commonly used objects. */
/* Optimization for empty strings */
- if (size == 0 && unicode_empty != NULL) {
- Py_INCREF(unicode_empty);
- return unicode_empty;
- }
+ if (size == 0)
+ _Py_RETURN_UNICODE_EMPTY();
/* Single character Unicode objects in the Latin-1 range are
shared when using this constructor */
@@ -1889,10 +1908,8 @@
PyObject *res;
unsigned char max_char;
- if (size == 0) {
- Py_INCREF(unicode_empty);
- return unicode_empty;
- }
+ if (size == 0)
+ _Py_RETURN_UNICODE_EMPTY();
assert(size > 0);
if (size == 1)
return get_latin1_char(u[0]);
@@ -1912,10 +1929,8 @@
PyObject *res;
Py_UCS2 max_char;
- if (size == 0) {
- Py_INCREF(unicode_empty);
- return unicode_empty;
- }
+ if (size == 0)
+ _Py_RETURN_UNICODE_EMPTY();
assert(size > 0);
if (size == 1) {
Py_UCS4 ch = u[0];
@@ -1950,10 +1965,8 @@
PyObject *res;
Py_UCS4 max_char;
- if (size == 0) {
- Py_INCREF(unicode_empty);
- return unicode_empty;
- }
+ if (size == 0)
+ _Py_RETURN_UNICODE_EMPTY();
assert(size > 0);
if (size == 1) {
Py_UCS4 ch = u[0];
@@ -2245,10 +2258,8 @@
PyUnicode_FromWideChar(register const wchar_t *w, Py_ssize_t size)
{
if (w == NULL) {
- if (size == 0) {
- Py_INCREF(unicode_empty);
- return unicode_empty;
- }
+ if (size == 0)
+ _Py_RETURN_UNICODE_EMPTY();
PyErr_BadInternalCall();
return NULL;
}
@@ -2825,15 +2836,11 @@
/* Decoding bytes objects is the most common case and should be fast */
if (PyBytes_Check(obj)) {
- if (PyBytes_GET_SIZE(obj) == 0) {
- Py_INCREF(unicode_empty);
- v = unicode_empty;
- }
- else {
- v = PyUnicode_Decode(
- PyBytes_AS_STRING(obj), PyBytes_GET_SIZE(obj),
- encoding, errors);
- }
+ if (PyBytes_GET_SIZE(obj) == 0)
+ _Py_RETURN_UNICODE_EMPTY();
+ v = PyUnicode_Decode(
+ PyBytes_AS_STRING(obj), PyBytes_GET_SIZE(obj),
+ encoding, errors);
return v;
}
@@ -2853,12 +2860,11 @@
}
if (buffer.len == 0) {
- Py_INCREF(unicode_empty);
- v = unicode_empty;
- }
- else
- v = PyUnicode_Decode((char*) buffer.buf, buffer.len, encoding, errors);
-
+ PyBuffer_Release(&buffer);
+ _Py_RETURN_UNICODE_EMPTY();
+ }
+
+ v = PyUnicode_Decode((char*) buffer.buf, buffer.len, encoding, errors);
PyBuffer_Release(&buffer);
return v;
}
@@ -4201,8 +4207,7 @@
if (size == 0) {
if (consumed)
*consumed = 0;
- Py_INCREF(unicode_empty);
- return unicode_empty;
+ _Py_RETURN_UNICODE_EMPTY();
}
/* Start off assuming it's all ASCII. Widen later as necessary. */
@@ -4609,8 +4614,7 @@
if (size == 0) {
if (consumed)
*consumed = 0;
- Py_INCREF(unicode_empty);
- return unicode_empty;
+ _Py_RETURN_UNICODE_EMPTY();
}
/* ASCII is equivalent to the first 128 ordinals in Unicode. */
@@ -4868,8 +4872,7 @@
if (q == e) {
if (consumed)
*consumed = size;
- Py_INCREF(unicode_empty);
- return unicode_empty;
+ _Py_RETURN_UNICODE_EMPTY();
}
#ifdef WORDS_BIGENDIAN
@@ -5108,8 +5111,7 @@
if (q == e) {
if (consumed)
*consumed = size;
- Py_INCREF(unicode_empty);
- return unicode_empty;
+ _Py_RETURN_UNICODE_EMPTY();
}
#if PY_LITTLE_ENDIAN
@@ -5386,10 +5388,8 @@
Py_ssize_t len;
len = length_of_escaped_ascii_string(s, size);
- if (len == 0) {
- Py_INCREF(unicode_empty);
- return unicode_empty;
- }
+ if (len == 0)
+ _Py_RETURN_UNICODE_EMPTY();
/* After length_of_escaped_ascii_string() there are two alternatives,
either the string is pure ASCII with named escapes like \n, etc.
@@ -5781,10 +5781,8 @@
PyObject *errorHandler = NULL;
PyObject *exc = NULL;
- if (size == 0) {
- Py_INCREF(unicode_empty);
- return unicode_empty;
- }
+ if (size == 0)
+ _Py_RETURN_UNICODE_EMPTY();
/* Escaped strings will always be longer than the resulting
Unicode string, so we start with size here and then reduce the
@@ -5988,10 +5986,8 @@
1))
return NULL;
- if (size == 0) {
- Py_INCREF(unicode_empty);
- return unicode_empty;
- }
+ if (size == 0)
+ _Py_RETURN_UNICODE_EMPTY();
/* XXX overflow detection missing */
_PyUnicodeWriter_Init(&writer, 0);
@@ -6439,10 +6435,8 @@
PyObject *errorHandler = NULL;
PyObject *exc = NULL;
- if (size == 0) {
- Py_INCREF(unicode_empty);
- return unicode_empty;
- }
+ if (size == 0)
+ _Py_RETURN_UNICODE_EMPTY();
/* ASCII is equivalent to the first 128 ordinals in Unicode. */
if (size == 1 && (unsigned char)s[0] < 128)
@@ -6820,8 +6814,7 @@
if (chunk_size == 0 && done) {
if (v != NULL)
break;
- Py_INCREF(unicode_empty);
- return unicode_empty;
+ _Py_RETURN_UNICODE_EMPTY();
}
@@ -7298,10 +7291,8 @@
if (mapping == NULL)
return PyUnicode_DecodeLatin1(s, size, errors);
- if (size == 0) {
- Py_INCREF(unicode_empty);
- return unicode_empty;
- }
+ if (size == 0)
+ _Py_RETURN_UNICODE_EMPTY();
_PyUnicodeWriter_Init(&writer, 0);
if (_PyUnicodeWriter_Prepare(&writer, size, 127) == -1)
goto onError;
@@ -9354,9 +9345,7 @@
/* If empty sequence, return u"". */
if (seqlen == 0) {
Py_DECREF(fseq);
- Py_INCREF(unicode_empty);
- res = unicode_empty;
- return res;
+ _Py_RETURN_UNICODE_EMPTY();
}
/* If singleton sequence with an exact Unicode, return that. */
@@ -10056,7 +10045,9 @@
}
new_size = slen + n * (len2 - len1);
if (new_size == 0) {
- Py_INCREF(unicode_empty);
+ _Py_INCREF_UNICODE_EMPTY();
+ if (!unicode_empty)
+ goto error;
u = unicode_empty;
goto done;
}
@@ -11559,10 +11550,8 @@
PyErr_SetString(PyExc_IndexError, "string index out of range");
return NULL;
}
- if (start >= length || end < start) {
- Py_INCREF(unicode_empty);
- return unicode_empty;
- }
+ if (start >= length || end < start)
+ _Py_RETURN_UNICODE_EMPTY();
length = end - start;
if (PyUnicode_IS_ASCII(self)) {
@@ -11689,10 +11678,8 @@
PyObject *u;
Py_ssize_t nchars, n;
- if (len < 1) {
- Py_INCREF(unicode_empty);
- return unicode_empty;
- }
+ if (len < 1)
+ _Py_RETURN_UNICODE_EMPTY();
/* no repeat, return original string */
if (len == 1)
@@ -12832,8 +12819,7 @@
{
if (writer->pos == 0) {
Py_XDECREF(writer->buffer);
- Py_INCREF(unicode_empty);
- return unicode_empty;
+ _Py_RETURN_UNICODE_EMPTY();
}
if (writer->readonly) {
assert(PyUnicode_GET_LENGTH(writer->buffer) == writer->pos);
@@ -13051,8 +13037,7 @@
}
if (slicelength <= 0) {
- Py_INCREF(unicode_empty);
- return unicode_empty;
+ _Py_RETURN_UNICODE_EMPTY();
} else if (start == 0 && step == 1 &&
slicelength == PyUnicode_GET_LENGTH(self)) {
return unicode_result_unchanged(self);
@@ -14056,10 +14041,8 @@
if (!PyArg_ParseTupleAndKeywords(args, kwds, "|Oss:str",
kwlist, &x, &encoding, &errors))
return NULL;
- if (x == NULL) {
- Py_INCREF(unicode_empty);
- return unicode_empty;
- }
+ if (x == NULL)
+ _Py_RETURN_UNICODE_EMPTY();
if (encoding == NULL && errors == NULL)
return PyObject_Str(x);
else
@@ -14228,8 +14211,6 @@
int _PyUnicode_Init(void)
{
- int i;
-
/* XXX - move this array to unicodectype.c ? */
Py_UCS2 linebreak[] = {
0x000A, /* LINE FEED */
@@ -14243,13 +14224,11 @@
};
/* Init the implementation */
- unicode_empty = PyUnicode_New(0, 0);
+ _Py_INCREF_UNICODE_EMPTY();
if (!unicode_empty)
Py_FatalError("Can't create empty string");
- assert(_PyUnicode_CheckConsistency(unicode_empty, 1));
-
- for (i = 0; i < 256; i++)
- unicode_latin1[i] = NULL;
+ Py_DECREF(unicode_empty);
+
if (PyType_Ready(&PyUnicode_Type) < 0)
Py_FatalError("Can't initialize 'unicode'");
@@ -14289,15 +14268,10 @@
{
int i;
- Py_XDECREF(unicode_empty);
- unicode_empty = NULL;
-
- for (i = 0; i < 256; i++) {
- if (unicode_latin1[i]) {
- Py_DECREF(unicode_latin1[i]);
- unicode_latin1[i] = NULL;
- }
- }
+ Py_CLEAR(unicode_empty);
+
+ for (i = 0; i < 256; i++)
+ Py_CLEAR(unicode_latin1[i]);
_PyUnicode_ClearStaticStrings();
(void)PyUnicode_ClearFreeList();
}
@@ -14426,8 +14400,7 @@
"mortal/immortal\n", mortal_size, immortal_size);
Py_DECREF(keys);
PyDict_Clear(interned);
- Py_DECREF(interned);
- interned = NULL;
+ Py_CLEAR(interned);
}
_______________________________________________
Python-bugs-list mailing list
Unsubscribe:
http://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com
