Hello Eric, I have a similar model where objects are a set of key-value paires. Each object has its metadata represented with metadataId and a set of propertyIds. This is how the type of object is defined.
Another entity bean stores a property of some object. Its fields are metadataId, propertyId and property's value. Then to fetch an object I use dynamic QL. General method call looks like this: HashMap getObject(String metadataId, List propertyIds) Note, I can fetch the whole object's properties or its subset. Indeed, it's mapping of objects not horizonatlly (to rows in a table) but vertically. It's even possible to try to achieve some object inheritence with EJB and QL here. further inlined... Sunday, December 15, 2002, 7:23:40 PM, you wrote: EK> We have an architecture in which a majority of our domain objects are EK> modelled as key-value pairs. It's a financial application. So, for EK> example, we have a Position object that represents a holding, and it may EK> have things in it like "Amount owned", "Associated Asset", etc. The system EK> is completely extensible. A given customer describes their own set of EK> "Position" data fields, along with meta data about each one of these fields, EK> and these values are used throughout the suite of applications which we have EK> built (e.g. reporting, etc.). We use ejb to model their persistent data. EK> These ejbs are customer specific, tailored to each individual installations EK> database, etc. We then have proxies that interface to these custom EK> components and do things like pull back all positions in a given portfolio EK> on a given date. These are returned from the proxy as a Collection of Maps EK> (essentially). EK> The question is follows. We currently use BMP entity beans for which the EK> code is autogenerated. The bulk loads are actually done is direct jdbc and EK> the data is mapped to specific keys in the map on the way in based upon a EK> bean definition file which the customer writes. We would like to move to EK> CMP but have two questions. EK> 1. I've only seen cmp beans specify mapping to individual data members EK> (instance variables), not to keys in a map. Is there a technology available EK> to do such a mapping? AFAIK, no, but I thought about it too. EK> 2. I asked this before, but I want to confirm. If we go to cmp, can we do EK> away with our bulk jdbc loads, which essentially bypass any cacheing the EK> container might do for us? In the past, going through the entity beans with EK> BMP to load large amounts of data was expensive. Can CMP handle a bulk load EK> efficiently, or must it select the pks and then go back for the data one by EK> one? I think CMP is the right way to go. I extensively use read-ahead and caching facilities and sutisfied though there are still many things to improve. HTH, alex EK> Thoughts would be appreciated. EK> Regards EK> Eric Kaplan EK> Armanta, Inc. ------------------------------------------------------- This sf.net email is sponsored by: With Great Power, Comes Great Responsibility Learn to use your power at OSDN's High Performance Computing Channel http://hpc.devchannel.org/ _______________________________________________ JBoss-user mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/jboss-user
