FlatMap in Guava
by
This is a short post about a method I recently discovered in Guava.
The Issue
I had a situation at work where I was working with objects structured something like this:
public class Outer {
String outerId;
List<Inner> innerList;
.......
}
public class Inner {
String innerId;
Date timestamp;
}
public class Merged {
String outerId;
String innerId;
Date timestamp;
}
My task was flatten a list Outer
objects (along with the list of Inner
objects) into a list of Merged
objects. Since I’m working with Java 7, using streams is not an option.
The First Solution
Instead I turn to the FluentIterable class from Guava. My first instinct is to go with the FluentIterable.transform method (which is essentially a map function):
List<Outer> originalList = getListOfObjects();
Function<Outer,List<Merged>> flattenFunction //Details left out for clarity
//returns an Iterable of Lists!
Iterable<List<Merged>> mergedObjects = FluentIterable.from(originalList).tranform(flattenFunction);
But I really want a single collection of Merged
objects, not an iterable of lists! The missing ingredient here is a flatMap function. Since I’m not using Scala, Clojure or Java 8, I feel that I’m out of luck.
A Better Solution
I decide to take a closer look at the FluentIterable
class and I discover the FluentIterable.transformAndConcat method. The transformAndConcat method applies a function to each element of the fluent iterable and appends the results into a single iterable instance. I have my flatMap function in Guava! Now my solution looks like this:
List<Outer> originalList = getListOfObjects();
Function<Outer,List<Merged>> flattenFunction //Details left out for clarity
Iterable<Merged> mergedObjects = FluentIterable.from(originalList).transformAndConcat(flattenFunction);
Conclusion
While this is a very short post, it goes to show how useful the Guava library is and how functional programming concepts can make our code more concise.
Subscribe via RSS