Over a million developers have joined DZone.
{{announcement.body}}
{{announcement.title}}

Watch Out For Recursion in Java 8’s [Primitive]Stream.iterate()

DZone's Guide to

Watch Out For Recursion in Java 8’s [Primitive]Stream.iterate()

Why two code snippets, very similar to each other, don't work as intended, with one printing and the other creating a stack overflow error.

· Java Zone
Free Resource

Build vs Buy a Data Quality Solution: Which is Best for You? Gain insights on a hybrid approach. Download white paper now!

An interesting question by Tagir Valeev on Stack Overflow has recently caught my attention. To keep things short (read the question for details), while the following code works:

public static Stream<Long> longs() {
    return Stream.iterate(1L, i ->
        1L + longs().skip(i - 1L)
                    .findFirst()
                    .get());
}

longs().limit(5).forEach(System.out::println);

Printing:





The following, similar code won’t work:

public static LongStream longs() {
    return LongStream.iterate(1L, i ->
        1L + longs().skip(i - 1L)
                    .findFirst()
                    .getAsLong());
}

Causing a StackOverflowError.

Sure, this kind of recursive iteration is not optimal. It wasn’t prior to Java 8 and it certainly isn’t with the new APIs either. But one might think it should at least work, right? The reason why it doesn’t work is because of a subtle implementation difference between the two iterate() methods in Java 8. While the reference type stream’s Iterator first returns the seed and only then proceeds with iterating by applying the iteration function on the previous value:

final Iterator<T> iterator = new Iterator<T>() {
    @SuppressWarnings("unchecked")
    T t = (T) Streams.NONE;

    @Override
    public boolean hasNext() {
        return true;
    }

    @Override
    public T next() {
        return t = (t == Streams.NONE) ? seed : f.apply(t);
    }
};

This is not the case for the LongStream.iterate() version (and other primitive streams):

final PrimitiveIterator.OfLong iterator = new PrimitiveIterator.OfLong() {
    long t = seed;

    @Override
    public boolean hasNext() {
        return true;
    }

    @Override
    public long nextLong() {
        long v = t;
        t = f.applyAsLong(t);
        return v;
    }
};

The iteration function is already pre-fetched one value in advance. This is usually not a problem, but can lead to:

  1. Optimization issues when the iteration function is expensive
  2. Infinite recursions when the iterator is used recursively

As a workaround, it might be best to simply avoid recursion with this method in primitive type streams. Luckily, a fix in JDK 9 is already on its way (as a side effect for a feature enhancement):
https://bugs.openjdk.java.net/browse/JDK-8072727

Build vs Buy a Data Quality Solution: Which is Best for You? Maintaining high quality data is essential for operational efficiency, meaningful analytics and good long-term customer relationships. But, when dealing with multiple sources of data, data quality becomes complex, so you need to know when you should build a custom data quality tools effort over canned solutions. Download our whitepaper for more insights into a hybrid approach.

Topics:
java 8 ,stack overflow ,error

Published at DZone with permission of Lukas Eder, DZone MVB. See the original article here.

Opinions expressed by DZone contributors are their own.

{{ parent.title || parent.header.title}}

{{ parent.tldr }}

{{ parent.urlSource.name }}