Efficient way to Handle ResultSet in Java

Question

I'm using a ResultSet in Java, and am not sure how to properly close it. I'm considering using the ResultSet to construct a HashMap and then closing the ResultSet after that. Is this HashMap technique efficient, or are there more efficient ways of handling this situation? I need both keys and values, so using a HashMap seemed like a logical choice.

If using a HashMap is the most efficient method, how do I construct and use the HashMap in my code?

Here's what I've tried:

public HashMap resultSetToHashMap(ResultSet rs) throws SQLException {

  ResultSetMetaData md = rs.getMetaData();
  int columns = md.getColumnCount();
  HashMap row = new HashMap();
  while (rs.next()) {
     for (int i = 1; i <= columns; i++) {
       row.put(md.getColumnName(i), rs.getObject(i));
     }
  }
  return row;
}

Answer 1

1. Iterate over the ResultSet
2. Create a new Object for each row, to store the fields you need
3. Add this new object to ArrayList or Hashmap or whatever you fancy
4. Close the ResultSet, Statement and the DB connection

Done

EDIT: now that you have posted code, I have made a few changes to it.

public List resultSetToArrayList(ResultSet rs) throws SQLException{
  ResultSetMetaData md = rs.getMetaData();
  int columns = md.getColumnCount();
  ArrayList list = new ArrayList(50);
  while (rs.next()){
     HashMap row = new HashMap(columns);
     for(int i=1; i<=columns; ++i){           
      row.put(md.getColumnName(i),rs.getObject(i));
     }
      list.add(row);
  }

 return list;
}

Answer 2

I just cleaned up RHT's answer to eliminate some warnings and thought I would share. Eclipse did most of the work:

public List<HashMap<String,Object>> convertResultSetToList(ResultSet rs) throws SQLException {
    ResultSetMetaData md = rs.getMetaData();
    int columns = md.getColumnCount();
    List<HashMap<String,Object>> list = new ArrayList<HashMap<String,Object>>();

    while (rs.next()) {
        HashMap<String,Object> row = new HashMap<String, Object>(columns);
        for(int i=1; i<=columns; ++i) {
            row.put(md.getColumnName(i),rs.getObject(i));
        }
        list.add(row);
    }

    return list;
}

Answer 3

RHT pretty much has it. Or you could use a RowSetDynaClass and let someone else do all the work :)

Answer 4

A couple of things to enhance the other answers. First, you should never return a HashMap, which is a specific implementation. Return instead a plain old java.util.Map. But that's actually not right for this example, anyway. Your code only returns the last row of the ResultSet as a (Hash)Map. You instead want to return a List<Map<String,Object>>. Think about how you should modify your code to do that. (Or you could take Dave Newton's suggestion).

Answer 5

this is my alternative solution, instead of a List of Map, i'm using a Map of List. Tested on tables of 5000 elements, on a remote db, times are around 350ms for eiter method.

private Map<String, List<Object>> resultSetToArrayList(ResultSet rs) throws SQLException {
    ResultSetMetaData md = rs.getMetaData();
    int columns = md.getColumnCount();
    Map<String, List<Object>> map = new HashMap<>(columns);
    for (int i = 1; i <= columns; ++i) {
        map.put(md.getColumnName(i), new ArrayList<>());
    }
    while (rs.next()) {
        for (int i = 1; i <= columns; ++i) {
            map.get(md.getColumnName(i)).add(rs.getObject(i));
        }
    }

    return map;
}