commons-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Stephen Colebourne" <>
Subject Re: Proposal for the Commons Sandbox
Date Tue, 09 Mar 2004 22:24:50 GMT
A quick glance at your homepage, and a read of this discussion are enough to
convince me to vote -1 to adding this to [lang].

The task [lang] has is to hit a very delicate and fine sweet spot of basic
and missing useful JDK-style functions. It should have no framework
tendencies, and no great cause to grow.

jestr breaks the goals of lang on grounds of:
- size
- dependencies
- potential to grow
- 'extensible'

(Lang includes ToStringBuilder mainly because HashCodeBuilder got written.
The other three classes then naturally followed. ToStringBuilder could do
with a small enhancement to enable subclasses to format objects being
output, but it certainly isn't a general stringifying mechanism.)

Now, having said all of that, jestr looks quite nice and organised and
flexible. It could form a separate commons project. It could provide
alternative implementations/plugins to the [convert] project.


----- Original Message -----
From: "David Gilliland" <>
> >What's in IO/Collections that you can't do without for Jestr?
> >
> Looking at it again, I'm not using those as much as I had thought.  In
Collections I make one use of DualHashBidiMap, a few uses of IteratorChain,
and in IO I use NullOutputStream.  I do use Logging quite heavily though.
> But going forward I'd like to incorporate the JXPath stuff for predicates.
I think it would be highly cool to be able to give an XPath expression and
say "for all nodes in my object graph that match this expression, I want
such-and-such stringification style".  To reference JXPath from Lang would
obviously not be acceptable.

To unsubscribe, e-mail:
For additional commands, e-mail:

View raw message