hex.idiv¶
if b==0: goto div0
q = a/b (signed division)
r = a%b (signed modulo - signed according to rem_opt:)
if rem_opt is 0: sign(r)==sign(b) // Most programming languages implement this option.
if rem_opt is 1: sign(r)==sign(a)
if rem_opt is 2: sign(r) is always positive
(On any other rem_opt value, will jump to "div0". This check is guaranteed to be after the b==0 check).
Anyway, a == q*b + r.
@NOTE: a,b are SIGNED numbers. If you want a division with unsigned ints, use the div macro.
q,a are hex[:n], while r,b are hex[:nb]. div0 is the bit-address this function will jump to in-case b is zero.
Signature¶
def idiv n, nb, q, r, a, b, div0, rem_opt @ set_negative_a, test_b, set_negative_b, update_rem_opt_0, update_rem_opt_2, add_b, sub_b, negative_a, negative_b, one_negative, do_div, neg_b_2, neg_ans, end, fix_rem { ... }
Defined in hex/div.fj — lines 86–150 (view on GitHub).
Complexity¶
Time:
n^2(2@+8) + n*nb(34@+92) so if nb==n: n^2(36@+100)Space:
n(8.5@+132) + nb(21.5@+309) so if nb==n: n(30@+441)
See the complexity glossary for what @, w, dw, dbit, n mean.
Requires init¶
The following must be initialised before this macro is invoked:
hex.sub.init & hex.cmp.init (or hex.init)