Hi there

I am wondering - have you ever encountered situation, when you want to do something more than simple CRUD ? For example joining some tables, counting sth and then put that in the databean ?

Cause as far as I know you cannot do that in simply and elegand manner in Hibernate. So will finish with two sets of DAOs- one wich is using hibernate and is used for simply gimmie all Apples, create Apple, etc, and one for more complicated queries like gimmie Apples that were in top 10 number of sells, with count of how many were sold and who was the most frequent buyer (yea, I know stupid, but whatever, it's simply and example if you don't like it think about some twisted finance/shipment report).

As far as I was told, I can do that in hibernate, but I will finish with executing plain SQL and then iterating over arrays of objects as a result.

So how are you handling those issues in your real-life apps ?

thank you in advance for any hints / help

regards
Leszek

I never really separate my manipulation statements from my query statements. They can all easily go in the same DAO. I just place the query with the DAO of the object that it returns. The queries can get as complicated as you want them to be.

However, I've seen some people name two DAOS something like ApplePersister and AppleRepository.

However, since store/delete/update is typically just 2 or maybe a few more methods in the DAO, it's usually not worth the separation since most of the DAO will be queries anyway.

And you are passing to 'normal' DAO methods Hibernate objects as a parameters (for connection, etc), or plan Connection ?

thanks,
Leszek

One of the main functions of the DAO pattern is to encapsulate the data store access, so you probably wouldn't want to require callers to supply things like JDBC connections or Hibernate Sessions. These data store resources should be supplied by a dependency injection framework like Picocontainer, Spring, or a home grown dependency injector of some sort.