mirror of
https://github.com/python-pillow/Pillow.git
synced 2024-11-16 06:36:49 +03:00
295 lines
7.4 KiB
C
295 lines
7.4 KiB
C
/*
|
|
* The Python Imaging Library
|
|
*
|
|
* a simple math add-on for the Python Imaging Library
|
|
*
|
|
* history:
|
|
* 1999-02-15 fl Created
|
|
* 2005-05-05 fl Simplified and cleaned up for PIL 1.1.6
|
|
*
|
|
* Copyright (c) 1999-2005 by Secret Labs AB
|
|
* Copyright (c) 2005 by Fredrik Lundh
|
|
*
|
|
* See the README file for information on usage and redistribution.
|
|
*/
|
|
|
|
#include "Python.h"
|
|
|
|
#include "libImaging/Imaging.h"
|
|
|
|
#include "math.h"
|
|
#include "float.h"
|
|
|
|
#define MAX_INT32 2147483647.0
|
|
#define MIN_INT32 -2147483648.0
|
|
|
|
#define UNOP(name, op, type) \
|
|
void name(Imaging out, Imaging im1) { \
|
|
int x, y; \
|
|
for (y = 0; y < out->ysize; y++) { \
|
|
type *p0 = (type *)out->image[y]; \
|
|
type *p1 = (type *)im1->image[y]; \
|
|
for (x = 0; x < out->xsize; x++) { \
|
|
*p0 = op(type, *p1); \
|
|
p0++; \
|
|
p1++; \
|
|
} \
|
|
} \
|
|
}
|
|
|
|
#define BINOP(name, op, type) \
|
|
void name(Imaging out, Imaging im1, Imaging im2) { \
|
|
int x, y; \
|
|
for (y = 0; y < out->ysize; y++) { \
|
|
type *p0 = (type *)out->image[y]; \
|
|
type *p1 = (type *)im1->image[y]; \
|
|
type *p2 = (type *)im2->image[y]; \
|
|
for (x = 0; x < out->xsize; x++) { \
|
|
*p0 = op(type, *p1, *p2); \
|
|
p0++; \
|
|
p1++; \
|
|
p2++; \
|
|
} \
|
|
} \
|
|
}
|
|
|
|
#define NEG(type, v1) -(v1)
|
|
#define INVERT(type, v1) ~(v1)
|
|
|
|
#define ADD(type, v1, v2) (v1) + (v2)
|
|
#define SUB(type, v1, v2) (v1) - (v2)
|
|
#define MUL(type, v1, v2) (v1) * (v2)
|
|
|
|
#define MIN(type, v1, v2) ((v1) < (v2)) ? (v1) : (v2)
|
|
#define MAX(type, v1, v2) ((v1) > (v2)) ? (v1) : (v2)
|
|
|
|
#define AND(type, v1, v2) (v1) & (v2)
|
|
#define OR(type, v1, v2) (v1) | (v2)
|
|
#define XOR(type, v1, v2) (v1) ^ (v2)
|
|
#define LSHIFT(type, v1, v2) (v1) << (v2)
|
|
#define RSHIFT(type, v1, v2) (v1) >> (v2)
|
|
|
|
#define ABS_I(type, v1) abs((v1))
|
|
#define ABS_F(type, v1) fabs((v1))
|
|
|
|
/* --------------------------------------------------------------------
|
|
* some day, we should add FPE protection mechanisms. see pyfpe.h for
|
|
* details.
|
|
*
|
|
* PyFPE_START_PROTECT("Error in foobar", return 0)
|
|
* PyFPE_END_PROTECT(result)
|
|
*/
|
|
|
|
#define DIV_I(type, v1, v2) ((v2) != 0) ? (v1) / (v2) : 0
|
|
#define DIV_F(type, v1, v2) ((v2) != 0.0F) ? (v1) / (v2) : 0.0F
|
|
|
|
#define MOD_I(type, v1, v2) ((v2) != 0) ? (v1) % (v2) : 0
|
|
#define MOD_F(type, v1, v2) ((v2) != 0.0F) ? fmod((v1), (v2)) : 0.0F
|
|
|
|
static int
|
|
powi(int x, int y) {
|
|
double v = pow(x, y) + 0.5;
|
|
if (errno == EDOM) {
|
|
return 0;
|
|
}
|
|
if (v < MIN_INT32) {
|
|
v = MIN_INT32;
|
|
} else if (v > MAX_INT32) {
|
|
v = MAX_INT32;
|
|
}
|
|
return (int)v;
|
|
}
|
|
|
|
#define POW_I(type, v1, v2) powi(v1, v2)
|
|
#define POW_F(type, v1, v2) powf(v1, v2) /* FIXME: EDOM handling */
|
|
|
|
#define DIFF_I(type, v1, v2) abs((v1) - (v2))
|
|
#define DIFF_F(type, v1, v2) fabs((v1) - (v2))
|
|
|
|
#define EQ(type, v1, v2) (v1) == (v2)
|
|
#define NE(type, v1, v2) (v1) != (v2)
|
|
#define LT(type, v1, v2) (v1) < (v2)
|
|
#define LE(type, v1, v2) (v1) <= (v2)
|
|
#define GT(type, v1, v2) (v1) > (v2)
|
|
#define GE(type, v1, v2) (v1) >= (v2)
|
|
|
|
UNOP(abs_I, ABS_I, INT32)
|
|
UNOP(neg_I, NEG, INT32)
|
|
|
|
BINOP(add_I, ADD, INT32)
|
|
BINOP(sub_I, SUB, INT32)
|
|
BINOP(mul_I, MUL, INT32)
|
|
BINOP(div_I, DIV_I, INT32)
|
|
BINOP(mod_I, MOD_I, INT32)
|
|
BINOP(pow_I, POW_I, INT32)
|
|
BINOP(diff_I, DIFF_I, INT32)
|
|
|
|
UNOP(invert_I, INVERT, INT32)
|
|
BINOP(and_I, AND, INT32)
|
|
BINOP(or_I, OR, INT32)
|
|
BINOP(xor_I, XOR, INT32)
|
|
BINOP(lshift_I, LSHIFT, INT32)
|
|
BINOP(rshift_I, RSHIFT, INT32)
|
|
|
|
BINOP(min_I, MIN, INT32)
|
|
BINOP(max_I, MAX, INT32)
|
|
|
|
BINOP(eq_I, EQ, INT32)
|
|
BINOP(ne_I, NE, INT32)
|
|
BINOP(lt_I, LT, INT32)
|
|
BINOP(le_I, LE, INT32)
|
|
BINOP(gt_I, GT, INT32)
|
|
BINOP(ge_I, GE, INT32)
|
|
|
|
UNOP(abs_F, ABS_F, FLOAT32)
|
|
UNOP(neg_F, NEG, FLOAT32)
|
|
|
|
BINOP(add_F, ADD, FLOAT32)
|
|
BINOP(sub_F, SUB, FLOAT32)
|
|
BINOP(mul_F, MUL, FLOAT32)
|
|
BINOP(div_F, DIV_F, FLOAT32)
|
|
BINOP(mod_F, MOD_F, FLOAT32)
|
|
BINOP(pow_F, POW_F, FLOAT32)
|
|
BINOP(diff_F, DIFF_F, FLOAT32)
|
|
|
|
BINOP(min_F, MIN, FLOAT32)
|
|
BINOP(max_F, MAX, FLOAT32)
|
|
|
|
BINOP(eq_F, EQ, FLOAT32)
|
|
BINOP(ne_F, NE, FLOAT32)
|
|
BINOP(lt_F, LT, FLOAT32)
|
|
BINOP(le_F, LE, FLOAT32)
|
|
BINOP(gt_F, GT, FLOAT32)
|
|
BINOP(ge_F, GE, FLOAT32)
|
|
|
|
static PyObject *
|
|
_unop(PyObject *self, PyObject *args) {
|
|
Imaging out;
|
|
Imaging im1;
|
|
void (*unop)(Imaging, Imaging);
|
|
|
|
Py_ssize_t op, i0, i1;
|
|
if (!PyArg_ParseTuple(args, "nnn", &op, &i0, &i1)) {
|
|
return NULL;
|
|
}
|
|
|
|
out = (Imaging)i0;
|
|
im1 = (Imaging)i1;
|
|
|
|
unop = (void *)op;
|
|
|
|
unop(out, im1);
|
|
|
|
Py_INCREF(Py_None);
|
|
return Py_None;
|
|
}
|
|
|
|
static PyObject *
|
|
_binop(PyObject *self, PyObject *args) {
|
|
Imaging out;
|
|
Imaging im1;
|
|
Imaging im2;
|
|
void (*binop)(Imaging, Imaging, Imaging);
|
|
|
|
Py_ssize_t op, i0, i1, i2;
|
|
if (!PyArg_ParseTuple(args, "nnnn", &op, &i0, &i1, &i2)) {
|
|
return NULL;
|
|
}
|
|
|
|
out = (Imaging)i0;
|
|
im1 = (Imaging)i1;
|
|
im2 = (Imaging)i2;
|
|
|
|
binop = (void *)op;
|
|
|
|
binop(out, im1, im2);
|
|
|
|
Py_INCREF(Py_None);
|
|
return Py_None;
|
|
}
|
|
|
|
static PyMethodDef _functions[] = {
|
|
{"unop", _unop, 1}, {"binop", _binop, 1}, {NULL, NULL}};
|
|
|
|
static void
|
|
install(PyObject *d, char *name, void *value) {
|
|
PyObject *v = PyLong_FromSsize_t((Py_ssize_t)value);
|
|
if (!v || PyDict_SetItemString(d, name, v)) {
|
|
PyErr_Clear();
|
|
}
|
|
Py_XDECREF(v);
|
|
}
|
|
|
|
static int
|
|
setup_module(PyObject *m) {
|
|
PyObject *d = PyModule_GetDict(m);
|
|
|
|
install(d, "abs_I", abs_I);
|
|
install(d, "neg_I", neg_I);
|
|
install(d, "add_I", add_I);
|
|
install(d, "sub_I", sub_I);
|
|
install(d, "diff_I", diff_I);
|
|
install(d, "mul_I", mul_I);
|
|
install(d, "div_I", div_I);
|
|
install(d, "mod_I", mod_I);
|
|
install(d, "min_I", min_I);
|
|
install(d, "max_I", max_I);
|
|
install(d, "pow_I", pow_I);
|
|
|
|
install(d, "invert_I", invert_I);
|
|
install(d, "and_I", and_I);
|
|
install(d, "or_I", or_I);
|
|
install(d, "xor_I", xor_I);
|
|
install(d, "lshift_I", lshift_I);
|
|
install(d, "rshift_I", rshift_I);
|
|
|
|
install(d, "eq_I", eq_I);
|
|
install(d, "ne_I", ne_I);
|
|
install(d, "lt_I", lt_I);
|
|
install(d, "le_I", le_I);
|
|
install(d, "gt_I", gt_I);
|
|
install(d, "ge_I", ge_I);
|
|
|
|
install(d, "abs_F", abs_F);
|
|
install(d, "neg_F", neg_F);
|
|
install(d, "add_F", add_F);
|
|
install(d, "sub_F", sub_F);
|
|
install(d, "diff_F", diff_F);
|
|
install(d, "mul_F", mul_F);
|
|
install(d, "div_F", div_F);
|
|
install(d, "mod_F", mod_F);
|
|
install(d, "min_F", min_F);
|
|
install(d, "max_F", max_F);
|
|
install(d, "pow_F", pow_F);
|
|
|
|
install(d, "eq_F", eq_F);
|
|
install(d, "ne_F", ne_F);
|
|
install(d, "lt_F", lt_F);
|
|
install(d, "le_F", le_F);
|
|
install(d, "gt_F", gt_F);
|
|
install(d, "ge_F", ge_F);
|
|
|
|
return 0;
|
|
}
|
|
|
|
PyMODINIT_FUNC
|
|
PyInit__imagingmath(void) {
|
|
PyObject *m;
|
|
|
|
static PyModuleDef module_def = {
|
|
PyModuleDef_HEAD_INIT,
|
|
"_imagingmath", /* m_name */
|
|
NULL, /* m_doc */
|
|
-1, /* m_size */
|
|
_functions, /* m_methods */
|
|
};
|
|
|
|
m = PyModule_Create(&module_def);
|
|
|
|
if (setup_module(m) < 0) {
|
|
return NULL;
|
|
}
|
|
|
|
return m;
|
|
}
|