Am 13.11.19 um 19:25 schrieb Martin Ågren: > On Wed, 13 Nov 2019 at 18:19, René Scharfe <l.s.r@xxxxxx> wrote: >> ALLOC_ARRAY(ret, st_add3(a_len, b_len, 1)); >> - for (i = 0; i < a_len; i++) >> - ret[i] = a[i]; >> - for (i = 0; i < b_len; i++) >> - ret[a_len + i] = b[i]; >> + COPY_ARRAY(ret, a, a_len); >> + COPY_ARRAY(ret + a_len, b, b_len); >> ret[a_len + b_len] = b[b_len]; /* final OPTION_END */ > > Maybe include that last one in the COPY_ARRAY with something like this > on top? > > - COPY_ARRAY(ret + a_len, b, b_len); > - ret[a_len + b_len] = b[b_len]; /* final OPTION_END */ > + /* 1 more to include the final OPTION_END */ > + COPY_ARRAY(ret + a_len, b, st_add(b_len, 1)); I'd rather move towards allowing a and b to be NULL, for consistency and safety. Will send patches eventually, but preferably only once the original patch hits master. These follow-ups that I have in mind aren't terribly urgent. René
