commons-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Eric Pugh" <ep...@upstate.com>
Subject RE: [configuration] Refactor AbstractConfiguration(Configuration defaults)
Date Mon, 16 Feb 2004 14:48:38 GMT
True..  But, you could grab the propertiesConfiguration out of the composite
and save it..  I think the composite should be looked at as a read only
"thing" because saving has too many odd ramifications..

And, at least, the one user who mentioned the
AbstractConfiguration(Configuration conf) is actually using a
CompositeConfig anyway, and needs that exact same problem solved..

Eric


> -----Original Message-----
> From: Emmanuel Bourg [mailto:ebourg@micropole-univers.com]
> Sent: Monday, February 16, 2004 12:08 PM
> To: Jakarta Commons Developers List
> Subject: Re: [configuration] Refactor
> AbstractConfiguration(Configuration defaults)
>
>
> Hmm I'm not should we should remove this, even if a
> CompositeConfiguration looks cleaner, it also hides the
> specific methods
> of the underlying configuration. For example if I work with a
> PropertiesConfiguration, i can no longer call save() if it's
> embeeded in
> a CompositeConfiguration.
>
> Emmanuel Bourg
>
>
> Eric Pugh wrote:
> > Hi all,
> >
> > Based on the Bugzilla 26694 (BaseConfiguration: containsKey
> ignores default
> > properties) bug, I think that the constructor for
> > AbstractConfiguration(Configuration defaults) should be removed..
> >
> > Basically it is providing the equivalent of a
> CompositeConfiguration built
> > using the ConfigurationFactory's union versus override rules..  The
> > semantics are kinda weird to have these defaults..  If you
> want defualts
> > that are used when something else is missing, then just do:
> >
> > CompositeConfiguration cc = new CompositeConfiguration();
> > cc.addConfiguration(myDefaultConfig);
> > return cc;
> >
> > And that does the same thing, but cleaner!  Any opinons?
> Otherwise I am
> > going to delete it (after of course going and writing the
> testcase to prove
> > 26694 (argh!)).
> >
> > Eric
> >
> >
> >
> ---------------------------------------------------------------------
> > To unsubscribe, e-mail: commons-dev-unsubscribe@jakarta.apache.org
> > For additional commands, e-mail: commons-dev-help@jakarta.apache.org
> >
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: commons-dev-unsubscribe@jakarta.apache.org
> For additional commands, e-mail: commons-dev-help@jakarta.apache.org


---------------------------------------------------------------------
To unsubscribe, e-mail: commons-dev-unsubscribe@jakarta.apache.org
For additional commands, e-mail: commons-dev-help@jakarta.apache.org


Mime
View raw message