[ZODB-Dev] Re: Reloading product in 2.8

Michel Pelletier michel at dialnetwork.com
Sun Jul 17 01:59:29 EDT 2005


On Sat, 2005-07-16 at 12:00 -0400, zodb-dev-request at zope.org wrote:

> | 
> | But otherwise, refresh saved me *lots* of time...
> 
> Same here. If you know the gotchas and you program properly it can
> work flawleslly. Hell, I've even been able to refresh Archetypes
> products, what can be more prone to fail than that?

I agree with Dieter and Sidnei that refresh is useful, I use it, but I'm
using it less and using Selenium more for development iteration.  It's
even *slower* but the discipline enforced by a testing framework is a
better win than the seconds saved.  And I can drink tea while the tests
run.

And while someone in the thread mentioned that restarting instead of
refreshing Zope gave you a "fresh" Zope, it's not much more fresh: most
of Zope's data, and therefore data error, is persisted.  A testing
framework run against demo storages is the *real* way to be fresh.  This
is the mode I use in Selenium the most.

I still think refresh should be kept. It has drawbacks that can be
explained in more technical detail on the tab.  People should be told in
big bold letters to verify they are not using reload before reporting
any possible bugs they discover.  Maybe I missed something in the thread
where there's something wrong with it, but folks like Dieter and Sidnei
use it and obviously are aware of its drawbacks, then it should be kept.

-Michel



More information about the ZODB-Dev mailing list