Re: problem/question with getIteratorByQuery()

2003-11-03 Thread sclark
As has been suggested, an RsIterator is no longer valid once its underlying broker has been closed. I can't find the code at the moment, so maybe I'm dreaming, but I'm pretty sure I wrote a BrokerClosingIterator once, which could be used like so: Iterator iter = null; try { broker = Persisten

RE: problem/question with getIteratorByQuery()

2003-10-30 Thread Andy Czerwonka
You need it for the lazy instantiation via the iterator On Thu, 2003-10-30 at 09:06, [EMAIL PROTECTED] wrote: > > get rid of the transaction... why are you doing a > > broker.close() before > > you starting iterating? Not sure what you are trying to do. > > I'm doing a broker.close() to be sure

RE: problem/question with getIteratorByQuery()

2003-10-30 Thread Sylvain.Thevoz
> get rid of the transaction... why are you doing a > broker.close() before > you starting iterating? Not sure what you are trying to do. I'm doing a broker.close() to be sure to put the broker instance back to the connection pool. Sylvain -

Re: problem/question with getIteratorByQuery()

2003-10-30 Thread Andy Czerwonka
get rid of the transaction... why are you doing a broker.close() before you starting iterating? Not sure what you are trying to do. getIteratorByQuery() is a lazy version of getCollectionByQuery(). If does a count when creating the iterator and a select on each iter.next(). your code should loo