On Sep 23, 2010, at 14:14, Paul Berry <[email protected]> wrote: On Thu, Sep 23, 2010 at 11:57 AM, Markus Roberts <[email protected]>wrote:
> Paul wrote: > > It sounds to me like we have a decent consensus for Proposed Solution II > > > (restrict the contents of autoloaded files) with the caveat that if we're > going > > to make the changes in version n, it would be nice to issue deprecation > > warnings in version n-1 for instances where the classes inside a file > don't > > fit the naming requirements. > > The consensus position also had option I (rebuild the TypeCollection for > each compile) as an adjunct. Good point, I forgot about that. I believe the only argument for option I as an adjunct was for performance reasons (this came from Brice), is that correct? When Markus and I initially wrote up the RFC I think we may have misled people into thinking that option I would be more performant than option II. Actually, option I will be slower, since it does extra work on every compile. Option II should be the same performance as the existing code. Were there other reasons for including option I as an adjunct other than performance? Not that I know of, but those performance differences are likely gigantic - at least doubling compile times in some cases. -- http://puppetlabs.com/ | +1-615-594-8199 | @puppetmasterd -- You received this message because you are subscribed to the Google Groups "Puppet Developers" group. To post to this group, send email to [email protected]. To unsubscribe from this group, send email to [email protected]. For more options, visit this group at http://groups.google.com/group/puppet-dev?hl=en.
