Antwort: Re: mini-gmp: mpz_init_set_str fails on leading zeroes

[Axel Miller]

Wouldn't that break mpz_sgn(v)
if i use mpz_set_str(v, "00000000000000000000000000000000000000000000", 
10) ?

mpn_set_str_other would then return a value greater than zero for v, even 
if the value is zero.
Thus v->_mp_size would be greater than zero:

4179       rn = mpn_set_str_other (rp, dp, sn, base, &info);
4180     }
4181   assert (rn <= alloc);
4182   gmp_free (dp);
4183 
4184   r->_mp_size = sign ? - rn : rn;

As a consequence, mpz_sgn(v) would return 1.

Kind regards
Axel



Von:    Austyn Krutsinger <akrutsinger at gmail.com>
An:     Torbjörn Granlund <tg at gmplib.org>
Kopie:  Axel Miller <axel.miller at ppi.de>, gmp-bugs at gmplib.org
Datum:  21.07.2016 10:53
Betreff:        Re: mini-gmp: mpz_init_set_str fails on leading zeroes




On Thu, Jul 21, 2016 at 1:43 AM, Torbjörn Granlund <tg at gmplib.org> wrote:
Austyn Krutsinger <akrutsinger at gmail.com> writes:

  My initial though was to just skip the leading zero's in the mpz_set_str
  function by something like this:

    while (isspace( (unsigned char) *sp) || (*sp == '0'))
      sp++;

  Only problems is that this doesn't work for negative numbers that still
  have a bunch of leading zeros;

I think we should not accept strings like 00000-0000017.

Absolutely agree with you, kind of a silly proposal in retrospect.

We can fix this in the mpn_set_str_other function by changing the 
comparison in line 1321. If we accept that w can be > or = to 0, then 
there is no issue if the number has leading zeros. So line 1321 in 
mpn_set_str_other becomes:

1321   for (rn = (w >= 0); j < sn;)
1322     {
1323       mp_limb_t cy;
1324
1325       w = sp[j++];
1326       for (k = 1; k < info->exp; k++)
1327     w = w * b + sp[j++];
1328
1329       cy = mpn_mul_1 (rp, rp, rn, info->bb);
1330       cy += mpn_add_1 (rp, rp, rn, w);
1331       if (cy > 0)
1332     rp[rn++] = cy;
1333     }
1334   assert (j == sn);
1321   for (rn = (w > 0); j < sn;)

Regards,
Austyn