Hello,

I would like to check whether an SQL query has any results, but I don't need the returned entities/tuples. A simple solution would be:

SQLQuery query = session.createSQLQuery("...");
boolean hasResults = !query.list().isEmpty();

But this is very inefficient since list() creates all the entities (or in this case tuples) if I'm not mistaken.

Is there any way to check whether the query's ResultSet contains rows without actually creating any entities?

Thanks,
Christoph

Don't think I've ever seen anything with that functionality built in. I know it's not the same query, but could you modify the query to just do select count(*) ... as the query?

Maybe projecting the results into nothingness? I don't know, just guessing :)

hi cbrehm,
Integer count = (Integer) session.createQuery("select count(*) from tablename").uniqueResult();

by using this you can count the number of query results without actually returning them.



nirav
(dont forget to rate if this helps)

have a generic method as shown below, make note of count(1) rather than count(*), as count(*) considers all the columns in each row, while calculation, count(1) gives more efficient results compared to count(*)



int queryStatus(String query){

String newquery = "select count(1) "+ query; // assumption, your query ll be "from tablename where crieteria=1"

Integer count = (Integer) session.createQuery(newquery).uniqueResult();

return count.intVal();

}

Also this depends on what you have in your where criteria of the query

Thanks for all your replies. But I am looking for a generic solution that assumes no knowledge of the query string and just tests whether the underlying ResultSet contains any rows, without causing Hibernate to transform all those rows into entities or tuples.

count(*) considers number of rows rite.Can u plz explain how the number of columns come into the picture.

I want to know how select count(1) is more efficient than select count(*) ....

Plz help

Thanks a lot in advance