Re: [clazz] New project?

[Stephen Colebourne <sc...@btopenworld.com>]

Its basically about hours in the day primarily. The Apache/Commons reorg
discussion has taken up far too much valuable time. Plus I am trying to get
a release of my Joda projects date and time code out. Anyway:

- so far the name seems to be [clazz]
- the proposal (attached) seemed to be accepted
- no design work has been done
- my original coding (done some time ago) doesn't contain much of use

However, I still believe in the idea, and that it will help [jxpath],
[betwixt], [digester] and my own Joda beans project. So I want to see it go
ahead.

The best way to proceed would seem to be to actually open the sandbox
project and commit whats on my hard drive.

I would obviously greatly appreciate support on this ;-)

Stephen

----- Original Message -----
From: "Dmitri Plotnikov" <dp...@yahoo.com>
> What was the resolution on this issue? In case you need more votes,
> here's mine:
>
> +1
>
> How far along are you?  Do you need any help? I am very much iterested
> in the outcome and could do a bunch of coding/testing for you.
>
>
> JXPath currently has its own introspection mechanism and I am eager to
> replace it with a generic one.  Some unique features of the
> JXPathIntrospector are:
>
> 1. From JXPath's perspective there exist so called "atomic" classes,
> which are not supposed to be drilled into.  Examples include all
> primitive types as well as String, Number, Date, arguably Class etc.
>
> 2. Some classes (or interfaces) are "dynamic" and are handled with the
> use of so called DynamicPropertyHandlers.  Examples of such "dynamic"
> classes are Map, ServletContext, etc.
>
> 4. JXPath has MethodLookup utilities that perform method lookups that
> allow type conversion and an additional hiddent argument
> (ExpressionContext).
>
> What bothers me about all this is that it is very ad hoc for JXPath's
> own needs, while conceptually it wants to be a specialization of a
> generic mechanism like what you are proposing for [clazz].
>
> I have been getting some new requirements from JXPath users.  For
> example the BeanInfo lookup mechanism has been repeatedly critisized -
> people want to see custom BeanInfo factories instead of or in addition
> to <beanclass>XBeanInfo classes.  I will need to address these
> requirements sooner or later.  Do you think I should continue growing
> introspection mechanisms inside JXPath for now, or bite the bullet and
> commit to integration with the [clazz] project?  I would personally
> prefer the latter.
>
> What is the timetable for this [clazz]?
>
> Thank you,
>
> - Dmitri Plotnikov
> dmitri@apache.org
>
>
> __________________________________________________
> Do you Yahoo!?
> Y! Web Hosting - Let the expert host your web site
> http://webhosting.yahoo.com/

Re: [clazz] New project?

[Stephen Colebourne <sc...@btopenworld.com>]

Great, I'm just tidying up what I have at the moment (not vast amounts). It
includes some code from Ola berg too.

I like the idea of getting together a bit more of a detailed proposal first.
We just need to ensure that we come back to the mailing list before too
long, so as not to exclude anybody ;-)

Stephen

----- Original Message -----
From: "Dmitri Plotnikov" <dp...@yahoo.com>
> I actually have some time I could dedicate to [clazz] right now.  Would
> you like to do this: you send me your source code. I look at that and
> write up a draft design proposal (English and/or JavaDoc).  I am
> intimately familiar with introspection in the JDK, [jxpath], [betwixt],
> [digester] and Rhino.  I will also take a closer look at Velocity's
> UberSpector or whatever that thing is called as well as other similar
> stuff I will come across.  I'll try to capture the variety in a
> separate document.  There has been some discussion on the mailing list
> as well - I'll try to summarize that.
>
> Once the first draft is done, we'll publish it and solicit input from
> everybody.  I think we just need to give this thing a jumpstart.
>
> BTW, I have done some work with metadata in the past: Rational Rose
> back end, MOF (CORBA Meta Object Facility), XMI (XML Metadata
> Interchange) and a bunch of other metadata frameworks.  So I understand
> how easy it is to overengineer one - and will do my best to keep things
> as simple as I can.
>
> Thanks,
>
> - Dmitri
>
>
> --- Stephen Colebourne <sc...@btopenworld.com> wrote:
> > Its basically about hours in the day primarily. The Apache/Commons
> > reorg
> > discussion has taken up far too much valuable time. Plus I am trying
> > to get
> > a release of my Joda projects date and time code out. Anyway:
> >
> > - so far the name seems to be [clazz]
> > - the proposal (attached) seemed to be accepted
> > - no design work has been done
> > - my original coding (done some time ago) doesn't contain much of use
> >
> > However, I still believe in the idea, and that it will help [jxpath],
> > [betwixt], [digester] and my own Joda beans project. So I want to see
> > it go
> > ahead.
> >
> > The best way to proceed would seem to be to actually open the sandbox
> > project and commit whats on my hard drive.
> >
> > I would obviously greatly appreciate support on this ;-)
> >
> > Stephen
> >
> > ----- Original Message -----
> > From: "Dmitri Plotnikov" <dp...@yahoo.com>
> > > What was the resolution on this issue? In case you need more votes,
> > > here's mine:
> > >
> > > +1
> > >
> > > How far along are you?  Do you need any help? I am very much
> > iterested
> > > in the outcome and could do a bunch of coding/testing for you.
> > >
> > >
> > > JXPath currently has its own introspection mechanism and I am eager
> > to
> > > replace it with a generic one.  Some unique features of the
> > > JXPathIntrospector are:
> > >
> > > 1. From JXPath's perspective there exist so called "atomic"
> > classes,
> > > which are not supposed to be drilled into.  Examples include all
> > > primitive types as well as String, Number, Date, arguably Class
> > etc.
> > >
> > > 2. Some classes (or interfaces) are "dynamic" and are handled with
> > the
> > > use of so called DynamicPropertyHandlers.  Examples of such
> > "dynamic"
> > > classes are Map, ServletContext, etc.
> > >
> > > 4. JXPath has MethodLookup utilities that perform method lookups
> > that
> > > allow type conversion and an additional hiddent argument
> > > (ExpressionContext).
> > >
> > > What bothers me about all this is that it is very ad hoc for
> > JXPath's
> > > own needs, while conceptually it wants to be a specialization of a
> > > generic mechanism like what you are proposing for [clazz].
> > >
> > > I have been getting some new requirements from JXPath users.  For
> > > example the BeanInfo lookup mechanism has been repeatedly
> > critisized -
> > > people want to see custom BeanInfo factories instead of or in
> > addition
> > > to <beanclass>XBeanInfo classes.  I will need to address these
> > > requirements sooner or later.  Do you think I should continue
> > growing
> > > introspection mechanisms inside JXPath for now, or bite the bullet
> > and
> > > commit to integration with the [clazz] project?  I would personally
> > > prefer the latter.
> > >
> > > What is the timetable for this [clazz]?
> > >
> > > Thank you,
> > >
> > > - Dmitri Plotnikov
> > > dmitri@apache.org
> > >
> > >
> > > __________________________________________________
> > > Do you Yahoo!?
> > > Y! Web Hosting - Let the expert host your web site
> > > http://webhosting.yahoo.com/
> >
> > > --
> > To unsubscribe, e-mail:
> > <ma...@jakarta.apache.org>
> > For additional commands, e-mail:
> <ma...@jakarta.apache.org>
>
>
> __________________________________________________
> Do you Yahoo!?
> Y! Web Hosting - Let the expert host your web site
> http://webhosting.yahoo.com/
>
> --
> To unsubscribe, e-mail:
<ma...@jakarta.apache.org>
> For additional commands, e-mail:
<ma...@jakarta.apache.org>
>


--
To unsubscribe, e-mail:   <ma...@jakarta.apache.org>
For additional commands, e-mail: <ma...@jakarta.apache.org>

Re: [clazz] New project?

[Dmitri Plotnikov <dp...@yahoo.com>]

Stephen,

I actually have some time I could dedicate to [clazz] right now.  Would
you like to do this: you send me your source code. I look at that and
write up a draft design proposal (English and/or JavaDoc).  I am
intimately familiar with introspection in the JDK, [jxpath], [betwixt],
[digester] and Rhino.  I will also take a closer look at Velocity's
UberSpector or whatever that thing is called as well as other similar
stuff I will come across.  I'll try to capture the variety in a
separate document.  There has been some discussion on the mailing list
as well - I'll try to summarize that.

Once the first draft is done, we'll publish it and solicit input from
everybody.  I think we just need to give this thing a jumpstart.

BTW, I have done some work with metadata in the past: Rational Rose
back end, MOF (CORBA Meta Object Facility), XMI (XML Metadata
Interchange) and a bunch of other metadata frameworks.  So I understand
how easy it is to overengineer one - and will do my best to keep things
as simple as I can.

Thanks,

- Dmitri


--- Stephen Colebourne <sc...@btopenworld.com> wrote:
> Its basically about hours in the day primarily. The Apache/Commons
> reorg
> discussion has taken up far too much valuable time. Plus I am trying
> to get
> a release of my Joda projects date and time code out. Anyway:
> 
> - so far the name seems to be [clazz]
> - the proposal (attached) seemed to be accepted
> - no design work has been done
> - my original coding (done some time ago) doesn't contain much of use
> 
> However, I still believe in the idea, and that it will help [jxpath],
> [betwixt], [digester] and my own Joda beans project. So I want to see
> it go
> ahead.
> 
> The best way to proceed would seem to be to actually open the sandbox
> project and commit whats on my hard drive.
> 
> I would obviously greatly appreciate support on this ;-)
> 
> Stephen
> 
> ----- Original Message -----
> From: "Dmitri Plotnikov" <dp...@yahoo.com>
> > What was the resolution on this issue? In case you need more votes,
> > here's mine:
> >
> > +1
> >
> > How far along are you?  Do you need any help? I am very much
> iterested
> > in the outcome and could do a bunch of coding/testing for you.
> >
> >
> > JXPath currently has its own introspection mechanism and I am eager
> to
> > replace it with a generic one.  Some unique features of the
> > JXPathIntrospector are:
> >
> > 1. From JXPath's perspective there exist so called "atomic"
> classes,
> > which are not supposed to be drilled into.  Examples include all
> > primitive types as well as String, Number, Date, arguably Class
> etc.
> >
> > 2. Some classes (or interfaces) are "dynamic" and are handled with
> the
> > use of so called DynamicPropertyHandlers.  Examples of such
> "dynamic"
> > classes are Map, ServletContext, etc.
> >
> > 4. JXPath has MethodLookup utilities that perform method lookups
> that
> > allow type conversion and an additional hiddent argument
> > (ExpressionContext).
> >
> > What bothers me about all this is that it is very ad hoc for
> JXPath's
> > own needs, while conceptually it wants to be a specialization of a
> > generic mechanism like what you are proposing for [clazz].
> >
> > I have been getting some new requirements from JXPath users.  For
> > example the BeanInfo lookup mechanism has been repeatedly
> critisized -
> > people want to see custom BeanInfo factories instead of or in
> addition
> > to <beanclass>XBeanInfo classes.  I will need to address these
> > requirements sooner or later.  Do you think I should continue
> growing
> > introspection mechanisms inside JXPath for now, or bite the bullet
> and
> > commit to integration with the [clazz] project?  I would personally
> > prefer the latter.
> >
> > What is the timetable for this [clazz]?
> >
> > Thank you,
> >
> > - Dmitri Plotnikov
> > dmitri@apache.org
> >
> >
> > __________________________________________________
> > Do you Yahoo!?
> > Y! Web Hosting - Let the expert host your web site
> > http://webhosting.yahoo.com/
> 
> > --
> To unsubscribe, e-mail:  
> <ma...@jakarta.apache.org>
> For additional commands, e-mail:
<ma...@jakarta.apache.org>


__________________________________________________
Do you Yahoo!?
Y! Web Hosting - Let the expert host your web site
http://webhosting.yahoo.com/

--
To unsubscribe, e-mail:   <ma...@jakarta.apache.org>
For additional commands, e-mail: <ma...@jakarta.apache.org>