Remove the CACHE_HASH and INTERN_STRINGS preprocessor symbols.

Author: tim-one

Include/stringobject.h

@@ -24,28 +24,18 @@ variant that assumes a zero-terminated string.  Note that none of the
 functions should be applied to nil objects.
 */
 
-/* Two speedup hacks.  Caching the hash saves recalculation of a
-   string's hash value.  Interning strings (which requires hash
-   caching) tries to ensure that only one string object with a given
-   value exists, so equality tests are one pointer comparison.
-   Together, these can speed the interpreter up by as much as 20%.
-   Each costs the size of a long or pointer per string object.  In
-   addition, interned strings live until the end of times.  If you are
-   concerned about memory footprint, simply comment the #define out
-   here (and rebuild everything!). */
-#define CACHE_HASH
-#ifdef CACHE_HASH
-#define INTERN_STRINGS
-#endif
+/* Caching the hash (ob_shash) saves recalculation of a string's hash value.
+   Interning strings (ob_sinterned) tries to ensure that only one string
+   object with a given value exists, so equality tests can be one pointer
+   comparison.  This is generally restricted to strings that "look like"
+   Python identifiers, although the intern() builtin can be used to force
+   interning of any string.
+   Together, these sped the interpreter by up to 20%. */
 
 typedef struct {
     PyObject_VAR_HEAD
-#ifdef CACHE_HASH
     long ob_shash;
-#endif
-#ifdef INTERN_STRINGS
     PyObject *ob_sinterned;
-#endif
     char ob_sval[1];
 } PyStringObject;
 
@@ -70,15 +60,9 @@ extern DL_IMPORT(PyObject *) PyString_Format(PyObject *, PyObject *);
 extern DL_IMPORT(PyObject *) _PyString_FormatLong(PyObject*, int, int,
 						  int, char**, int*);
 
-#ifdef INTERN_STRINGS
 extern DL_IMPORT(void) PyString_InternInPlace(PyObject **);
 extern DL_IMPORT(PyObject *) PyString_InternFromString(const char *);
 extern DL_IMPORT(void) _Py_ReleaseInternedStrings(void);
-#else
-#define PyString_InternInPlace(p)
-#define PyString_InternFromString(cp) PyString_FromString(cp)
-#define _Py_ReleaseInternedStrings()
-#endif
 
 /* Macro, trading safety for speed */
 #define PyString_AS_STRING(op) (((PyStringObject *)(op))->ob_sval)

Misc/NEWS

@@ -89,6 +89,10 @@ Tools/Demos
 
 Build
 
+- References to the CACHE_HASH and INTERN_STRINGS preprocessor symbols
+  were eliminated.  They were always defined, and the internal features
+  they enabled stopped being experimental long ago.
+
 C API
 
 - Objects allocated using the new PyMalloc_New and PyMalloc_NewVar

Objects/bufferobject.c

@@ -10,9 +10,7 @@ typedef struct {
 	void *b_ptr;
 	int b_size;
 	int b_readonly;
-#ifdef CACHE_HASH
 	long b_hash;
-#endif
 } PyBufferObject;
 
 
@@ -36,9 +34,7 @@ _PyBuffer_FromMemory(PyObject *base, void *ptr, int size, int readonly)
 	b->b_ptr = ptr;
 	b->b_size = size;
 	b->b_readonly = readonly;
-#ifdef CACHE_HASH
 	b->b_hash = -1;
-#endif
 
 	return (PyObject *) b;
 }
@@ -152,9 +148,7 @@ PyBuffer_New(int size)
 	b->b_ptr = (void *)(b + 1);
 	b->b_size = size;
 	b->b_readonly = 0;
-#ifdef CACHE_HASH
 	b->b_hash = -1;
-#endif
 
 	return o;
 }
@@ -211,10 +205,8 @@ buffer_hash(PyBufferObject *self)
 	register unsigned char *p;
 	register long x;
 
-#ifdef CACHE_HASH
 	if ( self->b_hash != -1 )
 		return self->b_hash;
-#endif
 
 	if ( !self->b_readonly )
 	{
@@ -231,9 +223,7 @@ buffer_hash(PyBufferObject *self)
 	x ^= self->b_size;
 	if (x == -1)
 		x = -2;
-#ifdef CACHE_HASH
 	self->b_hash = x;
-#endif
 	return x;
 }
 

Objects/dictobject.c

@@ -480,10 +480,8 @@ PyDict_GetItem(PyObject *op, PyObject *key)
 	if (!PyDict_Check(op)) {
 		return NULL;
 	}
-#ifdef CACHE_HASH
 	if (!PyString_CheckExact(key) ||
 	    (hash = ((PyStringObject *) key)->ob_shash) == -1)
-#endif
 	{
 		hash = PyObject_Hash(key);
 		if (hash == -1) {
@@ -512,24 +510,18 @@ PyDict_SetItem(register PyObject *op, PyObject *key, PyObject *value)
 		return -1;
 	}
 	mp = (dictobject *)op;
-#ifdef CACHE_HASH
 	if (PyString_CheckExact(key)) {
-#ifdef INTERN_STRINGS
 		if (((PyStringObject *)key)->ob_sinterned != NULL) {
 			key = ((PyStringObject *)key)->ob_sinterned;
 			hash = ((PyStringObject *)key)->ob_shash;
 		}
-		else
-#endif
-		{
+		else {
 			hash = ((PyStringObject *)key)->ob_shash;
 			if (hash == -1)
 				hash = PyObject_Hash(key);
 		}
 	}
-	else
-#endif
-	{
+	else {
 		hash = PyObject_Hash(key);
 		if (hash == -1)
 			return -1;
@@ -564,11 +556,8 @@ PyDict_DelItem(PyObject *op, PyObject *key)
 		PyErr_BadInternalCall();
 		return -1;
 	}
-#ifdef CACHE_HASH
 	if (!PyString_CheckExact(key) ||
-	    (hash = ((PyStringObject *) key)->ob_shash) == -1)
-#endif
-	{
+	    (hash = ((PyStringObject *) key)->ob_shash) == -1) {
 		hash = PyObject_Hash(key);
 		if (hash == -1)
 			return -1;
@@ -844,11 +833,8 @@ dict_subscript(dictobject *mp, register PyObject *key)
 	PyObject *v;
 	long hash;
 	assert(mp->ma_table != NULL);
-#ifdef CACHE_HASH
 	if (!PyString_CheckExact(key) ||
-	    (hash = ((PyStringObject *) key)->ob_shash) == -1)
-#endif
-	{
+	    (hash = ((PyStringObject *) key)->ob_shash) == -1) {
 		hash = PyObject_Hash(key);
 		if (hash == -1)
 			return NULL;
@@ -1436,11 +1422,8 @@ dict_has_key(register dictobject *mp, PyObject *key)
 {
 	long hash;
 	register long ok;
-#ifdef CACHE_HASH
 	if (!PyString_CheckExact(key) ||
-	    (hash = ((PyStringObject *) key)->ob_shash) == -1)
-#endif
-	{
+	    (hash = ((PyStringObject *) key)->ob_shash) == -1) {
 		hash = PyObject_Hash(key);
 		if (hash == -1)
 			return NULL;
@@ -1460,11 +1443,8 @@ dict_get(register dictobject *mp, PyObject *args)
 	if (!PyArg_ParseTuple(args, "O|O:get", &key, &failobj))
 		return NULL;
 
-#ifdef CACHE_HASH
 	if (!PyString_CheckExact(key) ||
-	    (hash = ((PyStringObject *) key)->ob_shash) == -1)
-#endif
-	{
+	    (hash = ((PyStringObject *) key)->ob_shash) == -1) {
 		hash = PyObject_Hash(key);
 		if (hash == -1)
 			return NULL;
@@ -1489,11 +1469,8 @@ dict_setdefault(register dictobject *mp, PyObject *args)
 	if (!PyArg_ParseTuple(args, "O|O:setdefault", &key, &failobj))
 		return NULL;
 
-#ifdef CACHE_HASH
 	if (!PyString_CheckExact(key) ||
-	    (hash = ((PyStringObject *) key)->ob_shash) == -1)
-#endif
-	{
+	    (hash = ((PyStringObject *) key)->ob_shash) == -1) {
 		hash = PyObject_Hash(key);
 		if (hash == -1)
 			return NULL;
@@ -1725,11 +1702,8 @@ dict_contains(dictobject *mp, PyObject *key)
 {
 	long hash;
 
-#ifdef CACHE_HASH
 	if (!PyString_CheckExact(key) ||
-	    (hash = ((PyStringObject *) key)->ob_shash) == -1)
-#endif
-	{
+	    (hash = ((PyStringObject *) key)->ob_shash) == -1) {
 		hash = PyObject_Hash(key);
 		if (hash == -1)
 			return -1;

Objects/stringobject.c

@@ -72,12 +72,8 @@ PyString_FromStringAndSize(const char *str, int size)
 	if (op == NULL)
 		return PyErr_NoMemory();
 	PyObject_INIT_VAR(op, &PyString_Type, size);
-#ifdef CACHE_HASH
 	op->ob_shash = -1;
-#endif
-#ifdef INTERN_STRINGS
 	op->ob_sinterned = NULL;
-#endif
 	if (str != NULL)
 		memcpy(op->ob_sval, str, size);
 	op->ob_sval[size] = '\0';
@@ -135,12 +131,8 @@ PyString_FromString(const char *str)
 	if (op == NULL)
 		return PyErr_NoMemory();
 	PyObject_INIT_VAR(op, &PyString_Type, size);
-#ifdef CACHE_HASH
 	op->ob_shash = -1;
-#endif
-#ifdef INTERN_STRINGS
 	op->ob_sinterned = NULL;
-#endif
 	memcpy(op->ob_sval, str, size+1);
 #ifndef DONT_SHARE_SHORT_STRINGS
 	if (size == 0) {
@@ -737,12 +729,8 @@ string_concat(register PyStringObject *a, register PyObject *bb)
 	if (op == NULL)
 		return PyErr_NoMemory();
 	PyObject_INIT_VAR(op, &PyString_Type, size);
-#ifdef CACHE_HASH
 	op->ob_shash = -1;
-#endif
-#ifdef INTERN_STRINGS
 	op->ob_sinterned = NULL;
-#endif
 	memcpy(op->ob_sval, a->ob_sval, (int) a->ob_size);
 	memcpy(op->ob_sval + a->ob_size, b->ob_sval, (int) b->ob_size);
 	op->ob_sval[size] = '\0';
@@ -784,12 +772,8 @@ string_repeat(register PyStringObject *a, register int n)
 	if (op == NULL)
 		return PyErr_NoMemory();
 	PyObject_INIT_VAR(op, &PyString_Type, size);
-#ifdef CACHE_HASH
 	op->ob_shash = -1;
-#endif
-#ifdef INTERN_STRINGS
 	op->ob_sinterned = NULL;
-#endif
 	for (i = 0; i < size; i += a->ob_size)
 		memcpy(op->ob_sval+i, a->ob_sval, (int) a->ob_size);
 	op->ob_sval[size] = '\0';
@@ -945,15 +929,11 @@ string_hash(PyStringObject *a)
 	register unsigned char *p;
 	register long x;
 
-#ifdef CACHE_HASH
 	if (a->ob_shash != -1)
 		return a->ob_shash;
-#ifdef INTERN_STRINGS
 	if (a->ob_sinterned != NULL)
 		return (a->ob_shash =
 			((PyStringObject *)(a->ob_sinterned))->ob_shash);
-#endif
-#endif
 	len = a->ob_size;
 	p = (unsigned char *) a->ob_sval;
 	x = *p << 7;
@@ -962,9 +942,7 @@ string_hash(PyStringObject *a)
 	x ^= a->ob_size;
 	if (x == -1)
 		x = -2;
-#ifdef CACHE_HASH
 	a->ob_shash = x;
-#endif
 	return x;
 }
 
@@ -2730,14 +2708,10 @@ str_subtype_new(PyTypeObject *type, PyObject *args, PyObject *kwds)
 	pnew = type->tp_alloc(type, n);
 	if (pnew != NULL) {
 		memcpy(PyString_AS_STRING(pnew), PyString_AS_STRING(tmp), n+1);
-#ifdef CACHE_HASH
 		((PyStringObject *)pnew)->ob_shash =
 			((PyStringObject *)tmp)->ob_shash;
-#endif
-#ifdef INTERN_STRINGS
 		((PyStringObject *)pnew)->ob_sinterned =
 			((PyStringObject *)tmp)->ob_sinterned;
-#endif
 	}
 	Py_DECREF(tmp);
 	return pnew;
@@ -3579,7 +3553,6 @@ PyString_Format(PyObject *format, PyObject *args)
 }
 
 
-#ifdef INTERN_STRINGS
 
 /* This dictionary will leak at PyString_Fini() time.  That's acceptable
  * because PyString_Fini() specifically frees interned strings that are
@@ -3656,8 +3629,6 @@ PyString_InternFromString(const char *cp)
 	return s;
 }
 
-#endif
-
 void
 PyString_Fini(void)
 {
@@ -3670,7 +3641,6 @@ PyString_Fini(void)
 	Py_XDECREF(nullstring);
 	nullstring = NULL;
 #endif
-#ifdef INTERN_STRINGS
 	if (interned) {
 		int pos, changed;
 		PyObject *key, *value;
@@ -3685,10 +3655,8 @@ PyString_Fini(void)
 			}
 		} while (changed);
 	}
-#endif
 }
 
-#ifdef INTERN_STRINGS
 void _Py_ReleaseInternedStrings(void)
 {
 	if (interned) {
@@ -3698,4 +3666,3 @@ void _Py_ReleaseInternedStrings(void)
 		interned = NULL;
 	}
 }
-#endif /* INTERN_STRINGS */

Python/import.c

@@ -978,7 +978,6 @@ find_module(char *realname, PyObject *path, char *buf, size_t buflen,
 		if (strlen(buf) != len)
 			continue; /* v contains '\0' */
 #ifdef macintosh
-#ifdef INTERN_STRINGS
 		/*
 		** Speedup: each sys.path item is interned, and
 		** FindResourceModule remembers which items refer to
@@ -987,7 +986,6 @@ find_module(char *realname, PyObject *path, char *buf, size_t buflen,
 		*/
 		PyString_InternInPlace(&PyList_GET_ITEM(path, i));
 		v = PyList_GET_ITEM(path, i);
-#endif
 		if (PyMac_FindResourceModule((PyStringObject *)v, name, buf)) {
 			static struct filedescr resfiledescr =
 				{"", "", PY_RESOURCE};