On Thu, 2003-06-26 at 17:20, Henrik Nordstrom wrote: > > Same thing should probably start out separately. Once the database > (tries is just another form of databases) is proven we can consider a > more close integration with Squid.
A trie is much less than a database, but certainly is often a key component of a database. I'm sure that a separate implementation could be done, but it would be a lot harder, and require much of the same squid alterations as an in-squid implementation. Also, most external implementations would end up reproducing large chunks of squid (the aio / diskd / pthread+ufs io logic for starters) which are not currently separate. Regardless of the path, any implementation would need to be done in a devel branch initially, and only considered for merging as part/all is ready. Rob -- GPG key available at: <http://members.aardvark.net.au/lifeless/keys.txt>.
signature.asc
Description: This is a digitally signed message part
