> > > rt.repos = reallocarray(rt.repos,
> > > rt.reposz + 1, sizeof(struct repo));
> >
> > This line could be unwrapped. The code could also be simplified by using
> > recallocarray() (it looks like the -portable update.sh is prepared for
> > that).
>
> I leave that for later. There are a bunch of reallocarrays around the code
> that are used in the same way. Let's change them all.
Sure.
> > > + /* Look up in repository table. (Lookup should actually fail here) */
> > > + for (i = 0; i < rt.reposz; i++) {
> > > + if (rt.repos[i].repouri != NULL ||
> > > + strcmp(rt.repos[i].local, local))
> >
> > While I think this is correct, it looks odd. Couldn't this check for
> > rt.repos[i].local != NULL?
>
> No. The rt.repos[i].repouri check is there to skip repo's added by
> repo_lookup. In repo_lookup the same check is there but the other way
> around. local on the other hand can't be NULL, it will always be set.
Ugh, of course. I misread or misunderstood. Probably both.
> I think the repouri check could be removed since there is no way that have
> conflicting local repo paths. I will just remove the check.
Yes, this makes sense. It is confusing.
