https://www.jianshu.com/p/24090fb63968
从内存利用和CPU利用开始了解List和Tuple的优缺点
List:动态数组,元素可变,可改变大小(append,resize)
Tuple:静态数组,不可变,数据一旦创建后不可改变
M = (N » 3) + (N <9 ? 3 : 6) + 1 **
我们来看Python3.6.1的列表resize过程,源代码位于Objects/listobject.c中的list_resize函数:
static int
list_resize(PyListObject *self, Py_ssize_t newsize)
{
PyObject **items;
size_t new_allocated;
Py_ssize_t allocated = self->allocated;
/* Bypass realloc() when a previous overallocation is large enough
to accommodate the newsize. If the newsize falls lower than half
the allocated size, then proceed with the realloc() to shrink the list.
*/
if (allocated >= newsize && newsize >= (allocated >> 1)) {
assert(self->ob_item != NULL || newsize == 0);
Py_SIZE(self) = newsize;
return 0;
}
/* This over-allocates proportional to the list size, making room
* for additional growth. The over-allocation is mild, but is
* enough to give linear-time amortized behavior over a long
* sequence of appends() in the presence of a poorly-performing
* system realloc().
* The growth pattern is: 0, 4, 8, 16, 25, 35, 46, 58, 72, 88, ...
*/
new_allocated = (newsize >> 3) + (newsize < 9 ? 3 : 6);
/* check for integer overflow */
if (new_allocated > SIZE_MAX - newsize) {
PyErr_NoMemory();
return -1;
} else {
new_allocated += newsize;
}
if (newsize == 0)
new_allocated = 0;
items = self->ob_item;
if (new_allocated <= (SIZE_MAX / sizeof(PyObject *)))
PyMem_RESIZE(items, PyObject *, new_allocated);
else
items = NULL;
if (items == NULL) {
PyErr_NoMemory();
return -1;
}
self->ob_item = items;
Py_SIZE(self) = newsize;
self->allocated = new_allocated;
return 0;
}
结合C源码我们来举个例子,当一个list长度为8时,发生append操作后:
1)new_size = 原有的size + append一个对象 = 8 + 1 = 9
2)newsize为9,二进制是 1001,9 » 3 = 1
3)new_allocated = 9 » 3 + 6 = 7
4)new_allocated += new_size,为9 + 7 = 16
4)列表的最终大小为Py_SIZE = 16
我们可以通过Python源码看到上面的结论,代码位于Objects/tupleobject.c,我们可以清楚的看到tuple的粘贴过程:
static PyObject *
tupleconcat(PyTupleObject *a, PyObject *bb)
{
Py_ssize_t size;
Py_ssize_t i;
PyObject **src, **dest;
PyTupleObject *np;
if (!PyTuple_Check(bb)) {
PyErr_Format(PyExc_TypeError,
"can only concatenate tuple (not \"%.200s\") to tuple",
Py_TYPE(bb)->tp_name);
return NULL;
}
#define b ((PyTupleObject *)bb)
if (Py_SIZE(a) > PY_SSIZE_T_MAX - Py_SIZE(b))
return PyErr_NoMemory();
size = Py_SIZE(a) + Py_SIZE(b);
np = (PyTupleObject *) PyTuple_New(size);
if (np == NULL) {
return NULL;
}
src = a->ob_item;
dest = np->ob_item;
for (i = 0; i < Py_SIZE(a); i++) {
PyObject *v = src[i];
Py_INCREF(v);
dest[i] = v;
}
src = b->ob_item;
dest = np->ob_item + Py_SIZE(a);
for (i = 0; i < Py_SIZE(b); i++) {
PyObject *v = src[i];
Py_INCREF(v);
dest[i] = v;
}
return (PyObject *)np;
#undef b
}
PyObject *
PyTuple_New(Py_ssize_t size)
{
PyTupleObject *op;
Py_ssize_t i;
if (size < 0) {
PyErr_BadInternalCall();
return NULL;
}
#if PyTuple_MAXSAVESIZE > 0
if (size == 0 && free_list[0]) {
op = free_list[0];
Py_INCREF(op);
#ifdef COUNT_ALLOCS
tuple_zero_allocs++;
#endif
return (PyObject *) op;
}
if (size < PyTuple_MAXSAVESIZE && (op = free_list[size]) != NULL) {
free_list[size] = (PyTupleObject *) op->ob_item[0];
numfree[size]--;
#ifdef COUNT_ALLOCS
fast_tuple_allocs++;
#endif
/* Inline PyObject_InitVar */
#ifdef Py_TRACE_REFS
Py_SIZE(op) = size;
Py_TYPE(op) = &PyTuple_Type;
#endif
_Py_NewReference((PyObject *)op);
}
else
#endif
{
/* Check for overflow */
if ((size_t)size > ((size_t)PY_SSIZE_T_MAX - sizeof(PyTupleObject) -
sizeof(PyObject *)) / sizeof(PyObject *)) {
return PyErr_NoMemory();
}
op = PyObject_GC_NewVar(PyTupleObject, &PyTuple_Type, size);
if (op == NULL)
return NULL;
}
for (i=0; i < size; i++)
op->ob_item[i] = NULL;
#if PyTuple_MAXSAVESIZE > 0
if (size == 0) {
free_list[0] = op;
++numfree[0];
Py_INCREF(op); /* extra INCREF so that this is never freed */
}
#endif
#ifdef SHOW_TRACK_COUNT
count_tracked++;
#endif
_PyObject_GC_TRACK(op);
return (PyObject *) op;
}