I've got a set of rows in a database, and I'd like to provide an interface to spin through them like this:
def findAll: Iterable[MyObject]
Where we don't require having all the instances in memory at once. In C# you can easily create generators like this using yield, the compiler takes care of converting code that loops through the recordset into an iterator (sort of inverting it).
My current code looks like this:
def findAll: List[MyObject] = {
val rs = getRs
val values = new ListBuffer[MyObject]
while ( rs.next() )
values += new valueFromResultSet(rs)
values.toList
}
Is there a way I could convert this to not store the entire set in memory? Perhaps I could use a for comprehension?
I came across the same problem and based on the ideas above I created the following solution by simply writing an adapter class:
class RsIterator(rs: ResultSet) extends Iterator[ResultSet] {
def hasNext: Boolean = rs.next()
def next(): ResultSet = rs
}
With this you can e.g. perform map operations on the result set - which was my personal intention:
val x = new RsIterator(resultSet).map(x => {
(x.getString("column1"), x.getInt("column2"))
})
Append a .toList
in order to force evaluation. This is useful if the database connection is closed before you use the values. Otherwise you will get an error saying that you cannot access the ResultSet after the connection was closed.
Try extending Iterator instead. I haven't tested it, but something like this:
def findAll: Iterator[MyObject] = new Iterator[MyObject] {
val rs = getRs
override def hasNext = rs.hasNext
override def next = new valueFromResultSet(rs.next)
}
This should store rs when it's called, and otherwise just be a light wrapper to calls to rs.
If you want to save the values that you traverse, check out Stream.
If you love us? You can donate to us via Paypal or buy me a coffee so we can maintain and grow! Thank you!
Donate Us With