diff options
Diffstat (limited to 'strings/decimal.c')
-rw-r--r-- | strings/decimal.c | 100 |
1 files changed, 60 insertions, 40 deletions
diff --git a/strings/decimal.c b/strings/decimal.c index 9af95511f6d..1e62333ee66 100644 --- a/strings/decimal.c +++ b/strings/decimal.c @@ -274,20 +274,20 @@ static dec1 *remove_leading_zeroes(decimal_t *from, int *intg_result) /* - Remove ending 0 digits from fraction part + Count actual length of fraction part (without ending zeroes) SYNOPSIS - decimal_optimize_fraction() + decimal_actual_fraction() from number for processing */ -void decimal_optimize_fraction(decimal_t *from) +int decimal_actual_fraction(decimal_t *from) { int frac= from->frac, i; dec1 *buf0= from->buf + ROUND_UP(from->intg) + ROUND_UP(frac) - 1; if (frac == 0) - return; + return 0; i= ((frac - 1) % DIG_PER_DEC1 + 1); while (frac > 0 && *buf0 == 0) @@ -302,7 +302,7 @@ void decimal_optimize_fraction(decimal_t *from) *buf0 % powers10[i++] == 0; frac--); } - from->frac= frac; + return frac; } @@ -332,23 +332,15 @@ int decimal2string(decimal_t *from, char *to, int *to_len, int fixed_precision, int fixed_decimals, char filler) { - int len, intg, frac=from->frac, i, intg_len, frac_len, fill; + int len, intg, frac= from->frac, i, intg_len, frac_len, fill; /* number digits before decimal point */ int fixed_intg= (fixed_precision ? - (fixed_precision - - (from->sign ? 1 : 0) - - (fixed_decimals ? 1 : 0) - - fixed_decimals) : - 0); + (fixed_precision - fixed_decimals) : 0); int error=E_DEC_OK; char *s=to; dec1 *buf, *buf0=from->buf, tmp; DBUG_ASSERT(*to_len >= 2+from->sign); - DBUG_ASSERT(fixed_precision == 0 || - (fixed_precision < *to_len && - fixed_precision > ((from->sign ? 1 : 0) + - (fixed_decimals ? 1 : 0)))); /* removing leading zeroes */ buf0= remove_leading_zeroes(from, &intg); @@ -1170,6 +1162,8 @@ int decimal2bin(decimal_t *from, char *to, int precision, int frac) isize0=intg0*sizeof(dec1)+dig2bytes[intg0x], fsize0=frac0*sizeof(dec1)+dig2bytes[frac0x], fsize1=frac1*sizeof(dec1)+dig2bytes[frac1x]; + const int orig_isize0= isize0; + const int orig_fsize0= fsize0; char *orig_to= to; buf1= remove_leading_zeroes(from, &from_intg); @@ -1260,10 +1254,15 @@ int decimal2bin(decimal_t *from, char *to, int precision, int frac) } if (fsize0 > fsize1) { - while (fsize0-- > fsize1) + char *to_end= orig_to + orig_fsize0 + orig_isize0; + + while (fsize0-- > fsize1 && to < to_end) *to++=(uchar)mask; } orig_to[0]^= 0x80; + + /* Check that we have written the whole decimal and nothing more */ + DBUG_ASSERT(to == orig_to + orig_fsize0 + orig_isize0); return error; } @@ -1620,13 +1619,19 @@ static int do_add(decimal_t *from1, decimal_t *from2, decimal_t *to) x=intg1 > intg2 ? from1->buf[0] : intg2 > intg1 ? from2->buf[0] : from1->buf[0] + from2->buf[0] ; - if (unlikely(x > DIG_MASK*9)) /* yes, there is */ + if (unlikely(x > DIG_MAX-1)) /* yes, there is */ { intg0++; to->buf[0]=0; /* safety */ } FIX_INTG_FRAC_ERROR(to->len, intg0, frac0, error); + if (unlikely(error == E_DEC_OVERFLOW)) + { + max_decimal(to->len * DIG_PER_DEC1, 0, to); + return error; + } + buf0=to->buf+intg0+frac0; to->sign=from1->sign; @@ -1711,19 +1716,23 @@ static int do_sub(decimal_t *from1, decimal_t *from2, decimal_t *to) carry=1; else if (intg2 == intg1) { - while (unlikely(stop1[frac1-1] == 0)) - frac1--; - while (unlikely(stop2[frac2-1] == 0)) - frac2--; - while (buf1 < stop1+frac1 && buf2 < stop2+frac2 && *buf1 == *buf2) + dec1 *end1= stop1 + (frac1 - 1); + dec1 *end2= stop2 + (frac2 - 1); + while (unlikely((buf1 <= end1) && (*end1 == 0))) + end1--; + while (unlikely((buf2 <= end2) && (*end2 == 0))) + end2--; + frac1= (end1 - stop1) + 1; + frac2= (end2 - stop2) + 1; + while (buf1 <=end1 && buf2 <= end2 && *buf1 == *buf2) buf1++, buf2++; - if (buf1 < stop1+frac1) - if (buf2 < stop2+frac2) + if (buf1 <= end1) + if (buf2 <= end2) carry= *buf2 > *buf1; else carry= 0; else - if (buf2 < stop2+frac2) + if (buf2 <= end2) carry=1; else /* short-circuit everything: from1 == from2 */ { @@ -1922,6 +1931,17 @@ int decimal_mul(decimal_t *from1, decimal_t *from2, decimal_t *to) for (; carry; buf0--) ADD(*buf0, *buf0, 0, carry); } + + /* Now we have to check for -0.000 case */ + if (to->sign) + { + dec1 *buf= to->buf; + dec1 *end= to->buf + intg0 + frac0; + for (; (buf<end) && !*buf; buf++); + if (buf == end) + /* So we got decimal zero */ + decimal_make_zero(to); + } return error; } @@ -1953,6 +1973,18 @@ static int do_div_mod(decimal_t *from1, decimal_t *from2, sanity(to); /* removing all the leading zeroes */ + i= ((prec2 - 1) % DIG_PER_DEC1) + 1; + while (prec2 > 0 && *buf2 == 0) + { + prec2-= i; + i= DIG_PER_DEC1; + buf2++; + } + if (prec2 <= 0) /* short-circuit everything: from2 == 0 */ + return E_DEC_DIV_ZERO; + for (i= (prec2 - 1) % DIG_PER_DEC1; *buf2 < powers10[i--]; prec2--) ; + DBUG_ASSERT(prec2 > 0); + i=((prec1-1) % DIG_PER_DEC1)+1; while (prec1 > 0 && *buf1 == 0) { @@ -1968,19 +2000,6 @@ static int do_div_mod(decimal_t *from1, decimal_t *from2, for (i=(prec1-1) % DIG_PER_DEC1; *buf1 < powers10[i--]; prec1--) ; DBUG_ASSERT(prec1 > 0); - i=((prec2-1) % DIG_PER_DEC1)+1; - while (prec2 > 0 && *buf2 == 0) - { - prec2-=i; - i=DIG_PER_DEC1; - buf2++; - } - if (prec2 <= 0) /* short-circuit everything: from2 == 0 */ - return E_DEC_DIV_ZERO; - - for (i=(prec2-1) % DIG_PER_DEC1; *buf2 < powers10[i--]; prec2--) ; - DBUG_ASSERT(prec2 > 0); - /* let's fix scale_incr, taking into account frac1,frac2 increase */ if ((scale_incr-= frac1 - from1->frac + frac2 - from2->frac) < 0) scale_incr=0; @@ -2609,7 +2628,7 @@ void test_fr(const char *s1, const char *orig) printf("%-40s => ", s); end= strend(s1); string2decimal(s1, &a, &end); - decimal_optimize_fraction(&a); + a.frac= decimal_actual_fraction(&a); print_decimal(&a, orig, 0, 0); printf("\n"); } @@ -2731,6 +2750,7 @@ int main() test_dv("123", "0.01","12300.000000000", 0); test_dv("120", "100000000000.00000","0.000000001200000000", 0); test_dv("123", "0","", 4); + test_dv("0", "0", "", 4); test_dv("-12193185.1853376", "98765.4321","-123.456000000000000000", 0); test_dv("121931851853376", "987654321","123456.000000000", 0); test_dv("0", "987","0", 0); @@ -2947,7 +2967,7 @@ int main() test_sh("123456789.987654321", 0, "123456789.987654321", 0); a.len= sizeof(buf1)/sizeof(dec1); - printf("==== decimal_optimize_fraction ====\n"); + printf("==== decimal_actual_fraction ====\n"); test_fr("1.123456789000000000", "1.123456789"); test_fr("1.12345678000000000", "1.12345678"); test_fr("1.1234567000000000", "1.1234567"); |