Re: [Zope-dev] PANIC!!! That old refresh bug again.

2002-09-18 Thread Dieter Maurer
Florent Guillaume writes: > Shane Hathaway <[EMAIL PROTECTED]> wrote: > > - Making one product depend on another using import statements > > I'm doing that, and I observe the problem Lennart is describing (objects > turning into None). Any idea for a workaround? First refresh the "base" pro

Re: [Zope-dev] PANIC!!! That old refresh bug again.

2002-09-18 Thread Lennart Regebro
From: "Shane Hathaway" <[EMAIL PROTECTED]> > I assume you mean ProductA here. Yes. > It might. In general, it's not a good thing to have circular imports in > your code anyway, since it's brittle. They are not circular on a module-level, just on a product level, and indeed, it seems to help to

Re: [Zope-dev] PANIC!!! That old refresh bug again.

2002-09-18 Thread Leonardo Rochael Almeida
On Wed, 2002-09-18 at 13:09, Toby Dickenson wrote: > On Wednesday 18 Sep 2002 4:23 pm, Lennart Regebro wrote: > > > Having to restart each time I make a change is slowing down my development > > to a complete halt. > > Something is wrong here. You should be able to restart Zope within a few >

Re: [Zope-dev] PANIC!!! That old refresh bug again.

2002-09-18 Thread Shane Hathaway
Lennart Regebro wrote: > A new idea: > > Could the problem appear when you have "circular" product imports, i.e.: > > In Product.ProductA.ProductA.py: > from Product.ProductB.SupportModule import Supportmodule > > In Product.ProductB.ProductB.py: > from Product.ProductB.AnotherModule im

Re: [Zope-dev] PANIC!!! That old refresh bug again.

2002-09-18 Thread Lennart Regebro
A new idea: Could the problem appear when you have "circular" product imports, i.e.: In Product.ProductA.ProductA.py: from Product.ProductB.SupportModule import Supportmodule In Product.ProductB.ProductB.py: from Product.ProductB.AnotherModule import AnotherModule Maybe the refresh get

Re: [Zope-dev] PANIC!!! That old refresh bug again.

2002-09-18 Thread Lennart Regebro
From: "Shane Hathaway" <[EMAIL PROTECTED]> > The first thing you need to make sure you do is set up dependencies. If > FooProduct imports EasyPublisher, set FooProduct to be auto-refreshable, > then visit EasyPublisher and specify the dependency. Then, every time > EasyPublisher gets refreshed,

Re: [Zope-dev] PANIC!!! That old refresh bug again.

2002-09-18 Thread Toby Dickenson
On Wednesday 18 Sep 2002 4:23 pm, Lennart Regebro wrote: > Having to restart each time I make a change is slowing down my development > to a complete halt. Something is wrong here. You should be able to restart Zope within a few seconds. Are you using FileStorage? I guess you are not shutting

Re: [Zope-dev] PANIC!!! That old refresh bug again.

2002-09-18 Thread Shane Hathaway
Lennart Regebro wrote: > From: "Shane Hathaway" <[EMAIL PROTECTED]> > >>You're likely to run into this if your product does >>anything more than simply define and register classes that derive from >>SimpleItem. > > > Strangely enough we do this all the time, but we only get problems with the >

Re: [Zope-dev] PANIC!!! That old refresh bug again.

2002-09-18 Thread Lennart Regebro
From: "Shane Hathaway" <[EMAIL PROTECTED]> > You're likely to run into this if your product does > anything more than simply define and register classes that derive from > SimpleItem. Strangely enough we do this all the time, but we only get problems with the core EasyPublisher products. > Which

Re: [Zope-dev] PANIC!!! That old refresh bug again.

2002-09-18 Thread Florent Guillaume
Shane Hathaway <[EMAIL PROTECTED]> wrote: > - Making one product depend on another using import statements I'm doing that, and I observe the problem Lennart is describing (objects turning into None). Any idea for a workaround? Florent -- Florent Guillaume, Nuxeo (Paris, France) +33 1 40 33 79

Re: [Zope-dev] PANIC!!! That old refresh bug again.

2002-09-18 Thread Shane Hathaway
Lennart Regebro wrote: > Once again I have the problem that all references to imported > modules/objects/whatever dissapear when refreshing. > > It seems like sometimes some modules are not refreshed, even though they > really are. Could it be that not all modules are properly purged from > memor