RE: [picocontainer-dev] While looking at the streamlining of NanoWar ...

"Michael Rimov" <[email protected]>
Newsgroups gmane.comp.java.picocontainer.devel
Organization Centerline Computers, Inc
Message-ID <013e01c8bf3e$bd2f4b50$378de1f0$@com>
> At themoment I'm quite silent, but I follow
> the discussion closely.
> 
> Thread local is really dangerous - webservers
> do reuse threads. I got bitten in my ass more than
> once.

Konstantin,

Don't webservers only reuse a thread once one servlet request is complete??  

Currently, if I read the code right, the storage is only taking place during the
servlet request and gets destroyed as it comes back through the filter.


> BTW, do we need nanowar at all?  It's kind of spring
> way to try to provide everything and kitchen sink.
> 3-Tier container hierarchy is a good thing, and
> WW/S2 can profit from it.
> 
> But nanowar itself offers really poor functionality
> compared to XW / WW / S2 - or doyou like to
> reimplement parameter handling / conversion /
> propagation,
> validation, i18n, templating etc.?

Personally, I think that NanoWAR has something worthwhile to offer and in fact,
should be the jumping off point to provide integration with other frameworks.
For example, I think that  Pico -> FrameworkX integration should really be
(under the hood), NanoWAR -> FrameworkX integration.

In essence it provides a way to bind PicoContainer to a servlet environment.
Now NanoWEB on the other hand I think serves more as simple example to show how
integration could be done.  I never used it, but I did study it to figure out
how everything fit together.


Paul,  I've really been giving a lot of thought to the whole component
instance/lifecycle issue, and I guess I believe that the simplification gained
by having the restriction of Instance Registrations being only valid for the
Webapp-level services outweighs the negatives.  

As far as I can see, it finally unifies the application/session/request
containers into a single configuration script without ugly if/else blocks, and
that alone (IMO) solves the A-number-one ugly facet of NanoWAR 1.

							-Mike



---------------------------------------------------------------------
To unsubscribe from this list, please visit:

    http://xircles.codehaus.org/manage_email
lmpx.com only provides a reader for public news (NNTP) servers. It is not affiliated with the servers or forums shown here and is not responsible for the content of articles, which is written by their respective authors.