ea13f407a3
This does not affect code size or performance when debugging turned off. To address issue #420.
141 lines
4.6 KiB
C
141 lines
4.6 KiB
C
#include <stdlib.h>
|
|
#include <stdio.h>
|
|
#include <string.h>
|
|
#include <assert.h>
|
|
#include <math.h>
|
|
|
|
#include "nlr.h"
|
|
#include "misc.h"
|
|
#include "mpconfig.h"
|
|
#include "qstr.h"
|
|
#include "obj.h"
|
|
#include "parsenum.h"
|
|
#include "runtime0.h"
|
|
|
|
#if MICROPY_ENABLE_FLOAT
|
|
|
|
#if MICROPY_FLOAT_IMPL == MICROPY_FLOAT_IMPL_FLOAT
|
|
#include "formatfloat.h"
|
|
#endif
|
|
|
|
STATIC void float_print(void (*print)(void *env, const char *fmt, ...), void *env, mp_obj_t o_in, mp_print_kind_t kind) {
|
|
mp_obj_float_t *o = o_in;
|
|
#if MICROPY_FLOAT_IMPL == MICROPY_FLOAT_IMPL_FLOAT
|
|
char buf[32];
|
|
format_float(o->value, buf, sizeof(buf), 'g', 6, '\0');
|
|
print(env, "%s", buf);
|
|
#else
|
|
char buf[32];
|
|
sprintf(buf, "%.8g", (double) o->value);
|
|
print(env, buf);
|
|
if (strchr(buf, '.') == NULL) {
|
|
// Python floats always have decimal point
|
|
print(env, ".0");
|
|
}
|
|
#endif
|
|
}
|
|
|
|
STATIC mp_obj_t float_make_new(mp_obj_t type_in, uint n_args, uint n_kw, const mp_obj_t *args) {
|
|
// TODO check n_kw == 0
|
|
|
|
switch (n_args) {
|
|
case 0:
|
|
return mp_obj_new_float(0);
|
|
|
|
case 1:
|
|
if (MP_OBJ_IS_STR(args[0])) {
|
|
// a string, parse it
|
|
uint l;
|
|
const char *s = mp_obj_str_get_data(args[0], &l);
|
|
return mp_parse_num_decimal(s, l, false, false);
|
|
} else if (MP_OBJ_IS_TYPE(args[0], &mp_type_float)) {
|
|
// a float, just return it
|
|
return args[0];
|
|
} else {
|
|
// something else, try to cast it to a float
|
|
return mp_obj_new_float(mp_obj_get_float(args[0]));
|
|
}
|
|
|
|
default:
|
|
nlr_raise(mp_obj_new_exception_msg_varg(&mp_type_TypeError, "float takes at most 1 argument, %d given", n_args));
|
|
}
|
|
}
|
|
|
|
STATIC mp_obj_t float_unary_op(int op, mp_obj_t o_in) {
|
|
mp_obj_float_t *o = o_in;
|
|
switch (op) {
|
|
case MP_UNARY_OP_BOOL: return MP_BOOL(o->value != 0);
|
|
case MP_UNARY_OP_POSITIVE: return o_in;
|
|
case MP_UNARY_OP_NEGATIVE: return mp_obj_new_float(-o->value);
|
|
default: return NULL; // op not supported
|
|
}
|
|
}
|
|
|
|
STATIC mp_obj_t float_binary_op(int op, mp_obj_t lhs_in, mp_obj_t rhs_in) {
|
|
mp_obj_float_t *lhs = lhs_in;
|
|
if (MP_OBJ_IS_TYPE(rhs_in, &mp_type_complex)) {
|
|
return mp_obj_complex_binary_op(op, lhs->value, 0, rhs_in);
|
|
} else {
|
|
return mp_obj_float_binary_op(op, lhs->value, rhs_in);
|
|
}
|
|
}
|
|
|
|
const mp_obj_type_t mp_type_float = {
|
|
{ &mp_type_type },
|
|
.name = MP_QSTR_float,
|
|
.print = float_print,
|
|
.make_new = float_make_new,
|
|
.unary_op = float_unary_op,
|
|
.binary_op = float_binary_op,
|
|
};
|
|
|
|
mp_obj_t mp_obj_new_float(mp_float_t value) {
|
|
mp_obj_float_t *o = m_new(mp_obj_float_t, 1);
|
|
o->base.type = &mp_type_float;
|
|
o->value = value;
|
|
return (mp_obj_t)o;
|
|
}
|
|
|
|
mp_float_t mp_obj_float_get(mp_obj_t self_in) {
|
|
assert(MP_OBJ_IS_TYPE(self_in, &mp_type_float));
|
|
mp_obj_float_t *self = self_in;
|
|
return self->value;
|
|
}
|
|
|
|
mp_obj_t mp_obj_float_binary_op(int op, mp_float_t lhs_val, mp_obj_t rhs_in) {
|
|
mp_float_t rhs_val = mp_obj_get_float(rhs_in); // can be any type, this function will convert to float (if possible)
|
|
switch (op) {
|
|
case MP_BINARY_OP_ADD:
|
|
case MP_BINARY_OP_INPLACE_ADD: lhs_val += rhs_val; break;
|
|
case MP_BINARY_OP_SUBTRACT:
|
|
case MP_BINARY_OP_INPLACE_SUBTRACT: lhs_val -= rhs_val; break;
|
|
case MP_BINARY_OP_MULTIPLY:
|
|
case MP_BINARY_OP_INPLACE_MULTIPLY: lhs_val *= rhs_val; break;
|
|
// TODO: verify that C floor matches Python semantics
|
|
case MP_BINARY_OP_FLOOR_DIVIDE:
|
|
case MP_BINARY_OP_INPLACE_FLOOR_DIVIDE:
|
|
lhs_val = MICROPY_FLOAT_C_FUN(floor)(lhs_val / rhs_val);
|
|
goto check_zero_division;
|
|
case MP_BINARY_OP_TRUE_DIVIDE:
|
|
case MP_BINARY_OP_INPLACE_TRUE_DIVIDE:
|
|
lhs_val /= rhs_val;
|
|
check_zero_division:
|
|
if (isinf(lhs_val)){ // check for division by zero
|
|
nlr_raise(mp_obj_new_exception_msg(&mp_type_ZeroDivisionError, "float division by zero"));
|
|
}
|
|
break;
|
|
case MP_BINARY_OP_POWER:
|
|
case MP_BINARY_OP_INPLACE_POWER: lhs_val = MICROPY_FLOAT_C_FUN(pow)(lhs_val, rhs_val); break;
|
|
case MP_BINARY_OP_LESS: return MP_BOOL(lhs_val < rhs_val);
|
|
case MP_BINARY_OP_MORE: return MP_BOOL(lhs_val > rhs_val);
|
|
case MP_BINARY_OP_LESS_EQUAL: return MP_BOOL(lhs_val <= rhs_val);
|
|
case MP_BINARY_OP_MORE_EQUAL: return MP_BOOL(lhs_val >= rhs_val);
|
|
|
|
default:
|
|
return NULL; // op not supported
|
|
}
|
|
return mp_obj_new_float(lhs_val);
|
|
}
|
|
|
|
#endif // MICROPY_ENABLE_FLOAT
|