diff --git a/py/misc.h b/py/misc.h index e985bc5253..5b012d03dc 100644 --- a/py/misc.h +++ b/py/misc.h @@ -50,6 +50,8 @@ bool unichar_isdigit(unichar c); #define streq(s1, s2) (strcmp((s1), (s2)) == 0) */ +long strtonum(const char *restrict s, int base); + /** variable string *********************************************/ typedef struct _vstr_t { diff --git a/py/obj.h b/py/obj.h index 99d430f490..cbd12fe862 100644 --- a/py/obj.h +++ b/py/obj.h @@ -243,6 +243,7 @@ bool mp_obj_equal(mp_obj_t o1, mp_obj_t o2); bool mp_obj_less(mp_obj_t o1, mp_obj_t o2); machine_int_t mp_obj_get_int(mp_obj_t arg); +machine_int_t mp_obj_parse_int(mp_obj_t arg, mp_obj_t base_arg); #if MICROPY_ENABLE_FLOAT mp_float_t mp_obj_get_float(mp_obj_t self_in); void mp_obj_get_complex(mp_obj_t self_in, mp_float_t *real, mp_float_t *imag); diff --git a/py/objint.c b/py/objint.c index efec60030d..937bff7ae3 100644 --- a/py/objint.c +++ b/py/objint.c @@ -18,11 +18,17 @@ static mp_obj_t int_make_new(mp_obj_t type_in, int n_args, const mp_obj_t *args) return MP_OBJ_NEW_SMALL_INT(0); case 1: - // TODO allow string as arg and parse it - return mp_obj_new_int(mp_obj_get_int(args[0])); + if (MP_OBJ_IS_TYPE(args[0], &str_type)) { + // a string, parse it + return MP_OBJ_NEW_SMALL_INT(strtonum(qstr_str(mp_obj_get_qstr(args[0])), 0)); + } else { + return MP_OBJ_NEW_SMALL_INT(mp_obj_get_int(args[0])); + } - //case 2: - // TODO, parse with given base + case 2: + // should be a string, parse it + // TODO proper error checking of argument types + return MP_OBJ_NEW_SMALL_INT(strtonum(qstr_str(mp_obj_get_qstr(args[1])), mp_obj_get_int(args[0]))); default: nlr_jump(mp_obj_new_exception_msg_1_arg(MP_QSTR_TypeError, "int takes at most 2 arguments, %d given", (void*)(machine_int_t)n_args)); diff --git a/py/py.mk b/py/py.mk index e51b00f45f..61f14f7bb3 100644 --- a/py/py.mk +++ b/py/py.mk @@ -70,6 +70,7 @@ PY_O_BASENAME = \ emitinlinethumb.o \ runtime.o \ map.o \ + strtonum.o \ obj.o \ objbool.o \ objboundmeth.o \ diff --git a/py/strtonum.c b/py/strtonum.c new file mode 100644 index 0000000000..02c1506683 --- /dev/null +++ b/py/strtonum.c @@ -0,0 +1,96 @@ +#if defined(UNIX) + +#include +#include +#include + +#include "misc.h" +#include "mpconfig.h" +#include "mpqstr.h" +#include "nlr.h" +#include "obj.h" + +long strtonum(const char *restrict s, int base) { + int c, neg = 0; + const char *p = s; + char *num; + long found; + + // check radix base + if ((base != 0 && base < 2) || base > 36) { + nlr_jump(mp_obj_new_exception_msg(MP_QSTR_ValueError, "ValueError: int() arg 2 must be >=2 and <= 36")); + } + // skip surrounded whitespace + while (isspace((c = *(p++)))); + if (c == 0) { + goto value_error; + } + // preced sign + if (c == '+' || c == '-') { + neg = - (c == '-'); + c = *(p++); + } + + // find real radix base, and strip preced '0x', '0o' and '0b' + // TODO somehow merge with similar code in parse.c + if ((base == 0 || base == 16) && c == '0') { + c = *(p++); + if ((c | 32) == 'x') { + base = 16; + } else if (base == 0 && (c | 32) == 'o') { + base = 8; + } else if (base == 0 && (c | 32) == 'b') { + base = 2; + } else { + base = 10; + p -= 2; + } + } else if (base == 8 && c == '0') { + c = *(p++); + if ((c | 32) != 'o') { + p -= 2; + } + } else if (base == 2 && c == '0') { + c = *(p++); + if ((c | 32) != 'b') { + p -= 2; + } + } else { + if (base == 0) base = 10; + p--; + } + + found = strtol(p, &num, base); + if (errno) { + goto value_error; + } else if (found && *(num) == 0) { + goto done; + } else if (found || num != p) { + goto check_tail_space; + } else { + goto value_error; + } + +check_tail_space: + if (*(num) != 0) { + while (isspace((c = *(num++)))); + if (c != 0) { + goto value_error; + } + } + +done: + return (found ^ neg) - neg; + +value_error: + nlr_jump(mp_obj_new_exception_msg_2_args(MP_QSTR_ValueError, "invalid literal for int() with base %d: '%s'", (void*)(machine_uint_t)base, s)); +} + +#else /* defined(UNIX) */ + +long strtonum(const char *restrict s, int base) { + // TODO port strtol to stm + return 0; +} + +#endif /* defined(UNIX) */ diff --git a/tests/basics/tests/int1.py b/tests/basics/tests/int1.py new file mode 100644 index 0000000000..4ce0f9e70c --- /dev/null +++ b/tests/basics/tests/int1.py @@ -0,0 +1,74 @@ +print(int(0)) +print(int(1)) +print(int(+1)) +print(int(-1)) + +print(int('0')) +print(int('+0')) +print(int('-0')) +print(int('1')) +print(int('+1')) +print(int('-1')) +print(int('01')) +print(int('9')) +print(int('10')) +print(int('+10')) +print(int('-10')) +print(int('12')) +print(int('-12')) +print(int('99')) +print(int('100')) +print(int('314')) +print(int(' 314')) +print(int('314 ')) +print(int(' \t\t 314 \t\t ')) +print(int(' 1 ')) +print(int(' -3 ')) + +print(int('0', 10)) +print(int('1', 10)) +print(int(' \t 1 \t ', 10)) +print(int('11', 10)) +print(int('11', 16)) +print(int('11', 8)) +print(int('11', 2)) +print(int('11', 36)) +print(int('0o123', 0)) +print(int('8388607')) +print(int('0x123', 16)) +print(int('0X123', 16)) +print(int('0o123', 8)) +print(int('0O123', 8)) +print(int('0123', 8)) +print(int('0b100', 2)) +print(int('0B100', 2)) +print(int('0100', 2)) +print(int(' \t 0o12', 8)) +print(int('0o12 \t ', 8)) + + +def test(value, base): + try: + print(int(value, base)) + except ValueError: + print('ValueError') + + +test('x', 0) +test('1x', 0) +test(' 1x', 0) +test(' 1' + chr(2) + ' ', 0) +test('', 0) +test(' ', 0) +test(' \t\t ', 0) +test("\u0200", 0) +test('0x', 16) +test('0x', 0) +test('0o', 8) +test('0o', 0) +test('0b', 2) +test('0b', 0) +test('0b2', 2) +test('0o8', 8) +test('0xg', 16) +test('1 1', 16) diff --git a/unix-cpy/Makefile b/unix-cpy/Makefile index e59a06b7ad..4955ea0c50 100644 --- a/unix-cpy/Makefile +++ b/unix-cpy/Makefile @@ -11,7 +11,7 @@ ECHO = @echo # compiler settings CC = gcc -CFLAGS = -I. -I$(PY_SRC) -Wall -Werror -ansi -std=gnu99 +CFLAGS = -I. -I$(PY_SRC) -Wall -Werror -ansi -std=gnu99 -DUNIX LDFLAGS = -lm #Debugging/Optimization diff --git a/unix/Makefile b/unix/Makefile index 2c54cf2d4c..9055909fb4 100644 --- a/unix/Makefile +++ b/unix/Makefile @@ -11,7 +11,7 @@ ECHO = @echo # compiler settings CC = gcc -CFLAGS = -I. -I$(PY_SRC) -Wall -Werror -ansi -std=gnu99 +CFLAGS = -I. -I$(PY_SRC) -Wall -Werror -ansi -std=gnu99 -DUNIX LDFLAGS = -lm #Debugging/Optimization