axis-java-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Sanjiva Weerawarana" <sanj...@watson.ibm.com>
Subject Re: TCK issue: beans with full constructors?
Date Fri, 14 Jun 2002 03:26:31 GMT
While it may not be explicit, its definitely required. Otherwise
the lifecycle model of design-time and run-time doesn't work,
for example. Same goes for Beans.instantiate() .. (see section
10.3 of the Beans spec ftp://ftp.javasoft.com/docs/beans/beans.101.pdf).

Since Eduardo co-wrote the "how to be a good bean" doc that Greg
pointed to I imagine there will be no opposition to removing this!

Sanjiva.

----- Original Message -----
From: <gtruty@us.ibm.com>
To: <axis-dev@xml.apache.org>
Sent: Friday, June 14, 2002 8:50 AM
Subject: Re: TCK issue: beans with full constructors?


> Well,
>
> I think it's more good programming practice as there is nothing
explicitely
> in the java bean spec about this... However... there is also the following
> reference...
>       http://java.sun.com/products/javabeans/docs/goodbean.pdf
> which states...
>
> Two rules implied by the JavaBeans architecture but that are sometimes
> missed
> include:
> - the Bean class must provide zero-argument constructors so it can be
> created
> using Beans.instantiate(), and
> - the Bean must support persistence, by implementing either Serializable
or
> Externalizable.
>
> Regards... Greg
>
>
>
>
>                       Sam
>                       Ruby/Raleigh/IBM         To:
axis-dev@xml.apache.org
>                       @IBMUS                   cc:
>                                                Subject: Re: TCK issue:
beans with full constructors?
>                       06/13/2002 08:39
>                       PM
>                       Please respond
>                       to axis-dev
>
>
>
>
>
>
> Sanjiva Weerawarana wrote:
>
> >+1 .. beans can only have no-args constructors; otherwise it breaks
> >the entire programming model they espouse.
> >
> >
> Do you have a definative reference for this?  I see nothing in the spec
> that requires this, and conflicting information out on the web.
>
> In particular, I find the following very informative:
> http://developer.java.sun.
> com/developer/community/chat/OfficeHours/1999/oh0812.html
>
> - Sam Ruby
>
>
>


Mime
View raw message