[prev in list] [next in list] [prev in thread] [next in thread] 

List:       groovy-dev
Subject:    Re: [groovy-dev] modify classpath at runtime
From:       Jason Dillon <jason () planet57 ! com>
Date:       2004-06-22 2:55:53
Message-ID: AD3922A2-C3F7-11D8-8735-000A9566A360 () planet57 ! com
[Download RAW message or body]


Why don't you just expose an API to access the ClassWorld which was 
used to load Groovy with?  Then you can simply get a realm and add 
constituents to it.

--jason


On Jun 21, 2004, at 7:24 PM, Chris Poirier wrote:

> Hi Stepan,
>
>> There is no easy way to add any jars to Groovy from the Groovy script
>> itself, right? If that way exists, please stop reading and point me
>> where I missed :-)
>
> This brings up a question that is rather core to what Groovy will 
> become
> -- should we break with some platform-independence for the sake of
> convenience?
>
> Embedding paths in a Java program may be A Bad Thing, but, for a
> scripting language, where many problems are one-offs, it does seem a
> reasonable thing to do.  Along similar lines, I'm sure lots of people
> would like environment variables to be imported, if we could figure out
> a way to do it.
>
> OTOH, classpath does seem a hairy issue, in general, and maybe we 
> should
> consider a more fundamental way to solve it.  A "load()" function, in
> isolation, only hides the problem.
>
> James: was classworlds to be Groovy's saviour on the classpath issue?
> If so, should we perhaps do something there to make Groovy friendlier
> for the scripter?
>
> We've had some "yes" votes on the idea of a "load()" function -- does
> anyone want to defend a "no" position?
>
> Thanks,
>    Chris.
>

["PGP.sig" (application/pgp-signature)]

[prev in list] [next in list] [prev in thread] [next in thread]