commons-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Simon Kitching <si...@ecnetwork.co.nz>
Subject Re: [Digester] Digesting into a database
Date Mon, 29 Nov 2004 07:42:09 GMT
On Sun, 2004-11-07 at 12:08, Michael Schuerig wrote:
> I have written a couple of rules and support classes for Commons 
> Digester that help to abuse it in an interesting way. Instead of 
> creating application objects, now the digester inserts rows into a JDBC 
> database.
> 
> At the current stage, the code _works for me_(TM). If you look at it too 
> sharply, you may cut it into pieces. As yet it works with HSQLDB. 
> Adaptation to another DBMS requires implementing an interface, which is 
> no big deal presumably.
> 
> I'm not sure about the future of this thing. I don't promise to develop 
> it any further. As is, the functionality has proven useful to me, 
> possibly its useful to others, too.
> 
> With all these caveats in place, have a look at the (minimal) javadocs 
> and get the source+binary zip at
> 
> http://www.schuerig.de/michael/java/dbdigester
> 

Hi Michael,

On Sun, 2004-11-07 at 12:08, Michael Schuerig wrote: 
> I have written a couple of rules and support classes for Commons 
> Digester that help to abuse it in an interesting way. Instead of 
> creating application objects, now the digester inserts rows into a JDBC 
> database.
> 
> At the current stage, the code _works for me_(TM). If you look at it too 
> sharply, you may cut it into pieces. As yet it works with HSQLDB. 
> Adaptation to another DBMS requires implementing an interface, which is 
> no big deal presumably.
> 
> I'm not sure about the future of this thing. I don't promise to develop 
> it any further. As is, the functionality has proven useful to me, 
> possibly its useful to others, too.
> 
> With all these caveats in place, have a look at the (minimal) javadocs 
> and get the source+binary zip at
> http://www.schuerig.de/michael/java/dbdigester
> 

I finally got around to having a look at your code this evening; sorry it tooks so long.

Unfortunately I feel that what you have created is too complicated to be
an 'example' of Digester usage. I think that eighteen classes (with some
non-trivial interaction between them) is rather a lot to inflict on some
poor soul trying to figure out whether Digester is for them :-).

That's not to say that there is anything wrong with your code; I just
don't know quite how the Digester project could make use of it. We do
have one largish "worked example" currently in CVS, being the RSS
parser. But I'm not wild on having even that; keeping it functional
between releases is non-trivial.

And I don't believe that database-insertion functionality would fit well
in the digester core. The core really should be as generic as possible;
the database stuff is more an application of the digester than a
framework for others to build upon.

It might be best for this code to reside on your site as currently; I
would be happy to have a Wiki entry point to it, if you wish, under the
titles "more advanced worked examples" or similar. You probably need to
add some more documentation to what you currently have, though, to make
it useful to people following that link.

Anyway, this is just my personal opinion. If you disagree, feel free to
say so and see if other commons developers have alternative viewpoints.

Regardless, I see you are a regular on the commons lists, and hope to
see other contributions from you in the future..

Regards,

Simon


---------------------------------------------------------------------
To unsubscribe, e-mail: commons-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: commons-user-help@jakarta.apache.org


Mime
View raw message