🚀 go-pugleaf

In article <VA.000067fc.00f71953@trmphrst.demon.co.uk>,
Nikki Locke  <nikki@trmphrst.demon.co.uk> wrote:
> In article <19990714.001917.0AA80202.Q3aa@pierrot.co.uk>, Terrance Richard 
> Boyes wrote:
> > The other thing I was considering is that unless you make a 
> > habit of setting the newsstamp back a long way then reading
> > the newshistory file backwards into the hash could give some
> > performance improvement when you do actually find a duplicate.
> 
> If you use a hash table that stores a linked list of ID's with the same 
> hash in each node, then you can read the file forwards, but insert new 
> ID's at the front of each list. Reading the file forwards is obviously 
> easier, and adding something to the front of a singly linked list is 
> faster than adding it to the end. It's not often that you can make a 
> design decision that reduces code and increases efficiency at the same 
> time!

Just to be pedantic, it's a trivial mod to keep track of where the end
item of a linked list is, so appending to the end of a linked list is
really insignificantly more expensive. Nonetheless, your suggestion is
a good one.