commons-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Dmitri Plotnikov <dmi...@apache.org>
Subject Re: [reflect] Proposals:
Date Sun, 16 Jun 2002 23:25:15 GMT
How about this:  we make this thing explicitly multi-level.  For example, we
could have

    org.apache.commons.reflect

for the convenience stuff Ola is describing.

Then, separately, we could have

    org.apache.commons.reflect.introspect

for the framework that builds on ...reflect and provides similar
conveniences for all kinds of introspection.  Perhaps that's where Steve's
rules belong.

<dmitri:nag>
I still think we'd be better of with a more catchy name:

org.apache.commons.jin.reflect
org.apache.commons.jin.introspect
</dmitri:nag>

- Dmitri

----- Original Message -----
From: "Ola Berg" <ola.berg@arkitema.se>
To: <commons-dev@jakarta.apache.org>
Sent: Sunday, June 16, 2002 3:51 PM
Subject: [reflect] Proposals:


> Dimitri wrote:
> >Perhaps I misunderstood what the proposal was all about. My understanding
> >was that it was after introspection rather than pure reflection. When I
> >hear the word \"reflection\", the list of features that comes to my mind
is:
>
> No, I don\'t think you misunderstood. Hey, we are trying to make something
new here, the limits aren\'t set, borders ain\'t drawn.
>
> Reflection in JVM is basically set and very low level. What people already
do with reflection/introspection is useful and often very high level. I
challenged your proposal by trying to draw a sketchy level for this new
package, placing it somewhere in the middle: make reflection utility for
often-needed mechanisms, thus supporting (as opposed to implementing) a lot
of high level stuff.
>
> Using Java reflection is straight forward, once you have the right Method
objects. I believe that the low level operations in the package should deal
with these two issues:
>
> 1) Means to easily select the right set of Methods (hopefully
dynamically), and defining groups or categories of Methods (just like
setters are one category of Methods) with some kind of Method category
definition (a Rule, a Predicate, whatever)
>
> 2) Caching of the reflection look-up results, for performance issues.
>
> These two are very low-level issues with java.lang.reflect, but I believe
important ones.
>
> The second question is how high this package should go (in terms of
\"levels\"), still maintaining a good separation of concerns. Maybe the
ideas we come up with will result in different packages at different levels,
who knows?
>
> <ola:self-criticism>
> Anyway: don\'t limit your thoughts based on anything I come up with. My
personal signal-to-noise ratio is low, I admit that...
> </ola:self-criticism>
>
> /O
>
> --------------------
> ola.berg@arkitema.se
> 0733 - 99 99 17
>
> --
> To unsubscribe, e-mail:
<mailto:commons-dev-unsubscribe@jakarta.apache.org>
> For additional commands, e-mail:
<mailto:commons-dev-help@jakarta.apache.org>
>
>


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


Mime
View raw message