Hi Bryan Looks great stuff! Also a nice demo of Simper and Struts in action - am looking forward to seeing this in the sandbox.
Funnily enough I've been meaning to post a mail for a while now to see if anyone fancied using DynaBeans to make a lightweight SQL persistence framework, hopefully attempting to work with the Torque folk, in particular to reuse their SQL schema XML document and SQL DDL generation which is cool. Their code generated 'peers' seem an alternative to DynaBeans - both have their pros and cons - maybe they could be integrated into the same framework? For example on Simper you might want to reuse Torque's XML document for the SQL schema (maybe extending it with anything else you need); then you could write a single reusable Digester to load the tables, queries, keys, relations into Simper and as an alternative to hand coding it in the IInit implementation (e.g. the DemoInit class in your sample). Then users would just need to write their single XML document to describe their SQL tables & relations, all the DDL gets autogenerated by Torque (to drop & recreate tables etc) and Simper would autoconfigure itself from this document. This XML document could also be reverse engineered from an existing database (I'm not sure if thats in Torque already). Though take these Torque comments with a pinch of salt - I've had a look around the code, liked what I saw, but am not coming from a Turbine background. I guess its time for me to lurk awhile on Torque's list... ;-) James ----- Original Message ----- From: "Bryan Field-Elliot" <[EMAIL PROTECTED]> To: "Jakarta Commons Developers List" <[EMAIL PROTECTED]> Sent: Saturday, February 02, 2002 3:54 PM Subject: Re: Bean storage in database > Hi Juozas, > > Before starting your project this weekend, I recommend you have a look > at my Simper framework, which does roughly what you describe already. > I'm in the process of working with Ted Husted to get it submitted into > the sandbox here. In the meantime, take a look at: > > http://www.netmeme.org/simper > > Feature highlights include: > - Wrapping of table rows using the new DynaBean > - Automatic change detection and writing back to database, like EJB CMP > - Automatic use of Servlet 2.3 filters to demark transactions (also like > EJB entity beans) > - Support for relations between table, in a querying capacity (not > cascading updates or deletes) > - < 1,500 lines of code - simple to understand, yet packs a punch > - README describes rationale behind it's design and the tradeoffs made > (coming from an EJB CMP perspective) > > Thanks, > > Bryan > > > On Sat, 2002-02-02 at 02:03, Juozas Baliuka wrote: > > Hi, > this work will be started today. It will be simplestore samples. > I will try to clear my code, it is very dirty and have a lot of bugs at > this time. > I decided to implement this for my current project, but I see it can be > useful in some more common situations. > I think it is more example for simplestore usage, not any kind of framework. > It will be very limited, but it always possible to enhance. I will use this > for "readonly" data. > like : > CREATE VIEW MY_STAT AS ( SELECT ID, SUM( SOMETHING ) , > COUNT(SOMETHING), MAX(SOMETHING) FROM MY_TABLE, > ...... WHERE ........ GROUP BY ID HAVING ..... ORDER BY . ) > I have plans to support "Stored Procedures" , Transactions, > Finders, Relations .......... . > But it is not very trivial to implement this stuff on this weekend :) > > > _________________________________________________________ Do You Yahoo!? Get your free @yahoo.com address at http://mail.yahoo.com -- To unsubscribe, e-mail: <mailto:[EMAIL PROTECTED]> For additional commands, e-mail: <mailto:[EMAIL PROTECTED]>
