mirrors/cpython
1
0
Fork 0
mirror of https://github.com/python/cpython.git synced 2025-08-03 08:34:29 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 6

Optimize unicode_hash() by not calling

Browse source 
_PyUnicode_AsDefaultEncodedString() at all -- this saves two object
allocations (three block allocations!)  and lots of redundant work.
By using the same hash algorithm as string_hash(), we maintain the
invariant that the hash of an ASCII string is the same whether
represented as a PyString or a PyUnicode.
This commit is contained in:
Guido van Rossum 2007-09-18 19:42:40 +00:00
parent e4a9e788d3
commit c2504931ee
1 changed files with 18 additions and 12 deletions
Download patch file Download diff file Expand all files Collapse all files

30
Objects/unicodeobject.c
View file

@ -6587,21 +6587,27 @@ unicode_getitem(PyUnicodeObject *self, Py_ssize_t index)
return (PyObject*) PyUnicode_FromUnicode(&self->str[index], 1);
}
/* Believe it or not, this produces the same value for ASCII strings
as string_hash(). */
static long
unicode_hash(PyUnicodeObject *self)
{
if (self->hash != -1) {
return self->hash;
}
else {
/* Since Unicode objects compare equal to their UTF-8 string
counterparts, we hash the UTF-8 string. */
PyObject *v = _PyUnicode_AsDefaultEncodedString((PyObject*)self, NULL);
if (v == NULL)
return -1;
assert(PyString_CheckExact(v));
return self->hash = v->ob_type->tp_hash(v);
}
Py_ssize_t len;
Py_UNICODE *p;
long x;
if (self->hash != -1)
return self->hash;
len = Py_Size(self);
p = self->str;
x = *p << 7;
while (--len >= 0)
x = (1000003*x) ^ *p++;
x ^= Py_Size(self);
if (x == -1)
x = -2;
self->hash = x;
return x;
}
PyDoc_STRVAR(index__doc__,