tomcat-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Remy Maucherat" <r...@apache.org>
Subject Re: [4.1] Add additional events
Date Sat, 11 May 2002 00:16:14 GMT
> On Fri, 10 May 2002, Remy Maucherat wrote:
>
> > Date: Fri, 10 May 2002 15:51:26 -0700
> > From: Remy Maucherat <remm@apache.org>
> > Reply-To: Tomcat Developers List <tomcat-dev@jakarta.apache.org>
> > To: tomcat-dev@jakarta.apache.org
> > Subject: [4.1] Add additional events
> >
> > I plan to add additional events to the components start and stop
methods:
> > before_start, after_start, before_stop, after_stop. This has been
suggested
> > and discussed a while ago, but never actually implemented.
> >
>
> Are you thinking about adding new method calls to the Lifecycle interface,
> versus just some new event types?  If it's the former, a couple of
> thoughts:

I was only planning on new event types (4 of them, to be exact). I don't
remember who suggested that originally (maybe Bill). The two events used at
the moment (as well as their timing) will not be modified so that everything
remains backward compatible.

> * The initialize() method was added to o.a.c.Connector, primarily
>   to provide what I imagine you're thinking of as before_start --
>   I'd suggest either using this name or migrating it to before_start.

Actually, it was added to be able to call it through JNI, for the
port-80-without-root on unix.

> * There are quite a number of places that components implementing
>   Lifecycle are started and stopped dynamically -- dealing with
>   them all is a non-trivial amount of editing (although the changes
>   required are pretty straightforward).
>
> Adding just the extra event notifications makes sense, and is somewhat
> easier (although you still have to fire them from all the appropriate
> places), but doesn't give you quite as much flexibility in managing things
> since the listeners don't throw a checked exception.

Adding the extra calls would break compatibility, and I don't think we
should do that.

Remy


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


Mime
View raw message