Merging Multiple ResultSet into Single ResultSet in Java

Question

Let say I have multiple ResultSet(each resultSet would refer to 1 row in database) (they are of same table.) . Now I want to create consolidated ResultSet which would intern have all other resultSet. So my primary goal is to create a combined ResultSet which would point to all rows which where previously pointed by individual resultSet.

I am using Java. Do any one know how pragmatically we can achieve this?

Edit : we are using java.sql.ResultSet.

Edit : To make it more clear :

Let say I have

List<ResultSet> someResults ; // each resultSet Would point to a single row in database.

I want to create a consolidated ResultSet finalResults;

psudo code :

List<ResultSet> resultSets = // poppulated from code
ResultSet rs  = convert(resultSets) // psude conver method

Solution 1

If you are talking about a java.sql.ResultSet, it's not possible as far as I know. I suggest you change your query instead.

Or you can retrieve the results into an java object and then combine all the objects into a Collection.

Solution 2

I would do it this way

class GatheringResultSet implements ResultSet {
    List<E> resultSets;
    ResultSet current;

    GatheringResultSet(List resultSets) {
        this.resultSets = new ArrayList(resultSets);
        current = resultSets.remove(0);
    }

    @Override
    public boolean next() throws SQLException {
        if (current.next()) {
            return true;
        }
        if (resultSets.isEmpty()) {
            return false;
        }
        current = resultSets.remove(0);
        return true;
    }

...

the rest of the methods just delegate call to current ResultSet

Solution 3

If the results are gonna be from the same table, why not use UNION/UNION ALL(depending on your needs) in your query itself.

Something like this:

select A, B
from C where Q = R
union
select A, B
from C where P = S

Or else, there is the hard way of iterating through each result set and populating POJOs and adding them to a List/Set(Based on your needs). This seems to be an overkill if there are a lot of result sets.