-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Thu, Apr 17, 2008 at 9:53 AM, Alvaro Herrera wrote: > > Well, consider that there are going to be 1 or 2 entries in the arrays > in most cases anyway :-) Well, as far as footers go anyway ... I just > realized that in all other cases it will certainly be the wrong thing to > do :-) Still, perhaps a integer count is better? >
I oscillated between using an integer and a pointer for ->header and - ->cell while I was writing the code. In the end I went with pointer simply because it makes the iterations nicer looking, and there's a symmetry in having the "first item" and "last item" pointers be the same type. If there's some technical reason that integers would be better, I'll all ears, but from an aesthetic/code readability perspective I like the pointer. > > Brendan Jurd escribió: > > What is it about the extra fields that makes you unhappy? > > I don't know if "unnecessarity" is a word, but I hope you get what I > mean :-) Since the footers list is usually pretty short, you could make an argument for dropping the "last footer" pointer and just iterating through the list to find the last element every time you want to do something with footers. And to do that, you'd have to declare a temporary pointer inside the AddFooter function anyway, so you're not getting rid of the pointer so much as making it slightly more transient. All that having been said, C isn't my native language. So if the pointer is indeed wasteful I'm happy to cede to the wisdom of more experienced C coders. Cheers, BJ -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.7 (GNU/Linux) Comment: http://getfiregpg.org iD8DBQFIBw7m5YBsbHkuyV0RAv4IAJ0cKrziZpNWkVV7LxFhlV/V5L0pJACfUtZ+ cVbcjL1e89j21JDZJBVdBqw= =Nzr+ -----END PGP SIGNATURE----- -- Sent via pgsql-patches mailing list (pgsql-patches@postgresql.org) To make changes to your subscription: http://www.postgresql.org/mailpref/pgsql-patches