Stowage/cpython
2
0
Fork 0
mirror of https://github.com/python/cpython.git synced 2025-11-01 14:11:41 +00:00
Code Issues Projects Releases Packages Wiki Activity

needforspeed: refactored the replace code slightly; special-case

Browse source 
constant-length changes; use fastsearch to locate the first match.
This commit is contained in:
Fredrik Lundh 2006-05-24 16:35:18 +00:00
parent eb62127842
commit 347ee277aa
1 changed files with 86 additions and 84 deletions
Download patch file Download diff file Expand all files Collapse all files

170
Objects/unicodeobject.c
View file

@ -2020,9 +2020,20 @@ PyObject *PyUnicode_DecodeUnicodeEscape(const char *s,
*/ */
static const Py_UNICODE *findchar(const Py_UNICODE *s, LOCAL(const Py_UNICODE *) findchar(const Py_UNICODE *s,
Py_ssize_t size, Py_ssize_t size,
Py_UNICODE ch); Py_UNICODE ch)
{
/* like wcschr, but doesn't stop at NULL characters */
while (size-- > 0) {
if (*s == ch)
return s;
s++;
}
return NULL;
}
static static
PyObject *unicodeescape_string(const Py_UNICODE *s, PyObject *unicodeescape_string(const Py_UNICODE *s,
@ -4141,22 +4152,6 @@ Py_ssize_t PyUnicode_Tailmatch(PyObject *str,
return result; return result;
} }
static
const Py_UNICODE *findchar(const Py_UNICODE *s,
Py_ssize_t size,
Py_UNICODE ch)
{
/* like wcschr, but doesn't stop at NULL characters */
while (size-- > 0) {
if (*s == ch)
return s;
s++;
}
return NULL;
}
/* Apply fixfct filter to the Unicode object self and return a /* Apply fixfct filter to the Unicode object self and return a
reference to the modified object */ reference to the modified object */
@ -4825,36 +4820,47 @@ PyObject *replace(PyUnicodeObject *self,
if (maxcount < 0) if (maxcount < 0)
maxcount = PY_SSIZE_T_MAX; maxcount = PY_SSIZE_T_MAX;
if (str1->length == 1 && str2->length == 1) { if (str1->length == str2->length) {
/* same length */
Py_ssize_t i; Py_ssize_t i;
if (str1->length == 1) {
/* replace characters */ /* replace characters */
if (!findchar(self->str, self->length, str1->str[0]) && Py_UNICODE u1, u2;
PyUnicode_CheckExact(self)) { if (!findchar(self->str, self->length, str1->str[0]))
/* nothing to replace, return original string */ goto nothing;
Py_INCREF(self); u = (PyUnicodeObject*) PyUnicode_FromUnicode(NULL, self->length);
u = self; if (!u)
return NULL;
Py_UNICODE_COPY(u->str, self->str, self->length);
u1 = str1->str[0];
u2 = str2->str[0];
for (i = 0; i < u->length; i++)
if (u->str[i] == u1) {
if (--maxcount < 0)
break;
u->str[i] = u2;
}
} else { } else {
Py_UNICODE u1 = str1->str[0]; i = fastsearch(
Py_UNICODE u2 = str2->str[0]; self->str, self->length, str1->str, str1->length, FAST_SEARCH
u = (PyUnicodeObject*) PyUnicode_FromUnicode(
NULL,
self->length
); );
if (u != NULL) { if (i < 0)
Py_UNICODE_COPY(u->str, self->str, goto nothing;
self->length); u = (PyUnicodeObject*) PyUnicode_FromUnicode(NULL, self->length);
for (i = 0; i < u->length; i++) if (!u)
if (u->str[i] == u1) { return NULL;
if (--maxcount < 0) Py_UNICODE_COPY(u->str, self->str, self->length);
break; while (i <= self->length - str1->length)
u->str[i] = u2; if (Py_UNICODE_MATCH(self, i, str1)) {
} if (--maxcount < 0)
break;
Py_UNICODE_COPY(u->str+i, str2->str, str2->length);
i += str1->length;
} else
i++;
} }
}
} else { } else {
Py_ssize_t n, i; Py_ssize_t n, i;
Py_UNICODE *p; Py_UNICODE *p;
@ -4862,51 +4868,47 @@ PyObject *replace(PyUnicodeObject *self,
n = count(self, 0, self->length, str1); n = count(self, 0, self->length, str1);
if (n > maxcount) if (n > maxcount)
n = maxcount; n = maxcount;
if (n == 0) { if (n == 0)
/* nothing to replace, return original string */ goto nothing;
if (PyUnicode_CheckExact(self)) { u = _PyUnicode_New(self->length + n * (str2->length - str1->length));
Py_INCREF(self); if (!u)
u = self; return NULL;
} i = 0;
else { p = u->str;
u = (PyUnicodeObject *) if (str1->length > 0) {
PyUnicode_FromUnicode(self->str, self->length); while (i <= self->length - str1->length)
} if (Py_UNICODE_MATCH(self, i, str1)) {
} else { /* replace string segment */
u = _PyUnicode_New( Py_UNICODE_COPY(p, str2->str, str2->length);
self->length + n * (str2->length - str1->length)); p += str2->length;
if (u) { i += str1->length;
i = 0; if (--n <= 0) {
p = u->str; /* copy remaining part */
if (str1->length > 0) { Py_UNICODE_COPY(p, self->str+i, self->length-i);
while (i <= self->length - str1->length) break;
if (Py_UNICODE_MATCH(self, i, str1)) {
/* replace string segment */
Py_UNICODE_COPY(p, str2->str, str2->length);
p += str2->length;
i += str1->length;
if (--n <= 0) {
/* copy remaining part */
Py_UNICODE_COPY(p, self->str+i, self->length-i);
break;
}
} else
*p++ = self->str[i++];
} else {
while (n > 0) {
Py_UNICODE_COPY(p, str2->str, str2->length);
p += str2->length;
if (--n <= 0)
break;
*p++ = self->str[i++];
} }
Py_UNICODE_COPY(p, self->str+i, self->length-i); } else
} *p++ = self->str[i++];
} else {
while (n > 0) {
Py_UNICODE_COPY(p, str2->str, str2->length);
p += str2->length;
if (--n <= 0)
break;
*p++ = self->str[i++];
} }
Py_UNICODE_COPY(p, self->str+i, self->length-i);
} }
} }
return (PyObject *) u; return (PyObject *) u;
nothing:
/* nothing to replace; return original string (when possible) */
if (PyUnicode_CheckExact(self)) {
Py_INCREF(self);
return (PyObject *) self;
}
return PyUnicode_FromUnicode(self->str, self->length);
} }
/* --- Unicode Object Methods --------------------------------------------- */ /* --- Unicode Object Methods --------------------------------------------- */