>
> The second point is most important here. This whole thread exists
> because people consider the existing ports system to be too slow. How
> is using XML going to help with that at all?
>
But which part? The /var half of the equation - well that depends on the
operation -
Lookup? E.g. testing for the existence of another port?
Update? E.g. Updating a dependency (Implicit Lookup)
Delete? E.g. Removing (Implicit Update and Lookup)
Install and so on
Lookup and update can be optimized but for what install base? E.g.
Do we know how many ports the typical system has? A simple solution
- to the lookup and update - could be to have a master dependencies matrix
N x N where each dimension is a port and a dependency - if the typical
install base is say 500 ports that only has to be 500x500 bits - and
so on.
The /usr/ports/INDEX side is another issue totally - and the primary problem
is maintaining the file without having to visit all directories
- well - a simple hack is only to update changed records based upon mtime -
it's still nasty - because all dependencies has
to be changed as well.
_______________________________________________
freebsd-hackers@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-hackers
To unsubscribe, send any mail to "[EMAIL PROTECTED]"
