Alan DeKok writes: > > "Vic Abell" <[EMAIL PROTECTED]> wrote: > > > *last = t; > > > last = &t->next > > > > If t == NULL, won't &t->next attempt to de-reference a > > NULL pointer? > > Yes, but I didn't show all of the code I commited. > > Check the CVS version of files.c > > > Isn't last (currently) supposed to point to the end > > of the chain whose head is contained in pl? > > Yes, it does, by design. 'last = &pl' means that it points to the > "end" of the chain, when 'pl == NULL' > > When it's updated, and one new entry is added, we do: > > *last = new_entry > last = &(new_entry->next) > > And we never need to touch 'pl' again. > > > I thought last was used to help extend pl's chain. I don't see how > > the chain whose pointer is returned in t is linked to the end of > > pl's chain via last->next. > > Because it never looks at 'last->next'. last is a ptr to a ptr, not > a ptr. > > > Perhaps I'm not seeing the full context of your change. > > Go read the code. > > Alan DeKok.
Unfortunately it's not possible for me to read the code, because the CVS web interface at: http://www.freeradius.org/cgi-bin/cvsweb.cgi/radiusd/ returns "Internal Server Error." Is there another method to access the CVS archives? Vic Abell - List info/subscribe/unsubscribe? See http://www.freeradius.org/list/users.html