Re: [Zope-dev] [ZConfig] wishlist

2004-01-01 Thread Fred L. Drake, Jr.
Chris McDonough writes: > In a perfect world, it would be possible to extend or change the main > section of a schema (or any section in a schema) with altnerate keys or > sections, I think. I have no idea how to do it, though. Phillip Eby and I discussed the override issue a bit, and decided

Re: [Zope-dev] [ZConfig] wishlist

2004-01-01 Thread Chris McDonough
On Thu, 2004-01-01 at 22:41, Fred L. Drake, Jr. wrote: > Dieter Maurer writes: > > It uses top level keys for configuration. > > Sigh. We should change that. In a perfect world, it would be possible to extend or change the main section of a schema (or any section in a schema) with altnerate key

Re: [Zope-dev] [ZConfig] wishlist

2004-01-01 Thread Fred L. Drake, Jr.
Dieter Maurer writes: > It uses top level keys for configuration. Sigh. We should change that. > But anyway (even if session configuration were > encapsulated in its own section), > I would need to change the Zope schema (if I understand > "ZConfig" correctly, which is not yet sure). That

Re: [Zope-dev] more persistent mapping misery

2004-01-01 Thread Chris McDonough
On Thu, 2004-01-01 at 14:16, Dieter Maurer wrote: > There is also the "__module_aliases__" directive in > product initialization code. It is used (e.g.) for > "PythonScripts". > > It does things very similar to what you did (it is slightly more > careful as it does not override an existing module)

Re: [Zope-dev] more persistent mapping misery

2004-01-01 Thread Dieter Maurer
Chris McDonough wrote at 2003-12-30 20:13 -0500: >On the HEAD recently, a bunch of changes were made to the Persistence >package. Notably, the Persistence.PersistentMapping module was removed >in favor of a Persistence.mapping module. Unfortunately, there are >instance pickles lying around in any

Re: [Zope-dev] [ZConfig] wishlist

2004-01-01 Thread Dieter Maurer
Hi Fred, Fred L. Drake, Jr. wrote at 2003-12-30 15:08 -0500: > ... >Dieter Maurer writes: > > ... > > I feel the configuration process is not yet as modular > > as it should be? > >There's certainly room for improvement. > > > Let's explain what extensions I have: > > > > 1. an alternate "Transi