See: Description
| Interface | Description |
|---|---|
| TIntFunction |
Interface for functions that accept and return one int primitive.
|
| TIntHashingStrategy |
Interface to support pluggable hashing strategies in maps and sets.
|
| TIntObjectProcedure<T> |
Interface for procedures that take two parameters of type int and Object.
|
| TIntProcedure |
Interface for procedures with one int parameter.
|
| TLongFunction |
Interface for functions that accept and return one long primitive.
|
| TLongHashingStrategy |
Interface to support pluggable hashing strategies in maps and sets.
|
| TLongObjectProcedure<T> |
Interface for procedures that take two parameters of type long and Object.
|
| TLongProcedure |
Interface for procedures with one long parameter.
|
| TObjectFunction<T,R> |
Interface for functions that accept and return one Object reference.
|
| TObjectHashingStrategy<T> |
Interface to support pluggable hashing strategies in maps and sets.
|
| TObjectIntProcedure<K> |
Interface for procedures that take two parameters of type Object and int.
|
| TObjectLongProcedure<K> |
Interface for procedures that take two parameters of type Object and long.
|
| TObjectObjectProcedure<K,V> |
Interface for procedures that take two Object parameters.
|
| TObjectProcedure<T> |
Interface for procedures with one Object parameter.
|
| TObjectShortProcedure<K> |
Interface for procedures that take two parameters of type Object and short.
|
| TShortFunction |
Interface for functions that accept and return one short primitive.
|
| TShortHashingStrategy |
Interface to support pluggable hashing strategies in maps and sets.
|
| TShortLongProcedure |
Interface for procedures that take two parameters of type short and long.
|
| TShortObjectProcedure<T> |
Interface for procedures that take two parameters of type short and Object.
|
| TShortProcedure |
Interface for procedures with one short parameter.
|
| Class | Description |
|---|---|
| HashFunctions |
Provides various hash functions.
|
| PrimeFinder |
Used to keep hash table capacities prime numbers.
|
| THash |
Base class for hashtables that use open addressing to resolve collisions.
|
| THashMap<K,V> |
An implementation of the Map interface which uses an open addressed hash table to store its
contents.
|
| TIntArrayList |
A resizable, array-backed list of int primitives.
|
| TIntHash |
An open addressed hashing implementation for int primitives.
|
| TIntHashSet |
An open addressed set implementation for int primitives.
|
| TIntIterator |
Iterator for int collections.
|
| TIntObjectHashMap<V> |
An open addressed Map implementation for int keys and Object values.
|
| TIntObjectIterator<V> |
Iterator for maps of type int and Object.
|
| TLongHash |
An open addressed hashing implementation for long primitives.
|
| TLongObjectHashMap<V> |
An open addressed Map implementation for long keys and Object values.
|
| TLongObjectIterator<V> |
Iterator for maps of type long and Object.
|
| TObjectHash<T> |
An open addressed hashing implementation for Object types.
|
| TObjectIntHashMap<K> |
An open addressed Map implementation for Object keys and int values.
|
| TObjectIntIterator<K> |
Iterator for maps of type Object and int.
|
| TObjectLongHashMap<K> |
An open addressed Map implementation for Object keys and long values.
|
| TObjectLongIterator<K> |
Iterator for maps of type Object and long.
|
| TObjectShortHashMap<K> |
An open addressed Map implementation for Object keys and short values.
|
| TObjectShortIterator<K> |
Iterator for maps of type Object and short.
|
| TPrimitiveHash |
The base class for hashtables of primitive values.
|
| TShortArrayList |
A resizable, array-backed list of short primitives.
|
| TShortHash |
An open addressed hashing implementation for short primitives.
|
| TShortLongHashMap |
An open addressed Map implementation for short keys and long values.
|
| TShortLongIterator |
Iterator for maps of type short and long.
|
| TShortObjectHashMap<V> |
An open addressed Map implementation for short keys and Object values.
|
| TShortObjectIterator<V> |
Iterator for maps of type short and Object.
|
GNU Trove: High performance collections for Java.
The GNU Trove library has two objectives:
The Trove maps/sets use open addressing instead of the chaining approach taken by the JDK hashtables. This eliminates the need to create Map.Entry wrappper objects for every item in a table and so reduces the O (big-oh) in the performance of the hashtable algorithm. The size of the tables used in Trove's maps/sets is always a prime number, improving the probability of an optimal distribution of entries across the table, and so reducing the the likelihood of performance-degrading collisions. Trove sets are not backed by maps, and so using a THashSet does not result in the allocation of an unused "values" array.
Trove's maps/sets support the use of custom hashing strategies, allowing you to tune collections based on characteristics of the input data. This feature also allows you to define hash functions when it is not feasible to override Object.hashCode(). For example, the java.lang.String class is final, and its implementation of hashCode() takes O(n) time to complete. In some applications, however, it may be possible for a custom hashing function to save time by skipping portions of the string that are invariant.
Using java.util.HashMap, it is not possible to use Java language arrays as keys. For example, this code:
char[] foo, bar;
foo = new char[] {'a','b','c'};
bar = new char[] {'a','b','c'};
System.out.println(foo.hashCode() == bar.hashCode() ? "equal" : "not equal");
System.out.println(foo.equals(bar) ? "equal" : "not equal");
produces this output:
not equal
not equal
And so an entry stored in a java.util.HashMap with foo as a
key could not be retrieved with bar, since there is no way
to override hashCode() or equals() on language array objects.
In a gnu.trove.THashMap, however, you can implement a TObjectHashingStrategy to enable hashing on arrays:
class CharArrayStrategy implements TObjectHashingStrategy {
public int computeHashCode(Object o) {
char[] c = (char[])o;
// use the shift-add-xor class of string hashing functions
// cf. Ramakrishna and Zobel, "Performance in Practice of String Hashing Functions"
int h = 31; // seed chosen at random
for (int i = 0; i < c.length; i++) { // could skip invariants
h = h ^ ((h << 5) + (h >> 2) + c[i]); // L=5, R=2 works well for ASCII input
}
return h;
}
public boolean equals(Object o1, Object o2) {
char[] c1 = (char[])o1;
char[] c2 = (char[])o2;
if (c1.length != c2.length) { // could drop this check for fixed-length keys
return false;
}
for (int i = 0, len = c1.length; i < len; i++) { // could skip invariants
if (c1[i] != c2[i]) {
return false;
}
}
return true;
}
}
As of release 0.1.7, Trove's primitive mappings include access through Iterators as well as procedures and functions. The API documentation on those classes contains several examples showing how these can be used effectively and explaining why their semantics differ from those of java.util.Iterator.
N.B. using Map.entrySet on a Trove Map is supported, but not encouraged. The reason is that this API requires the creation of the Map.Entry Objects that all other parts of Trove manage to avoid. An alternative is to implement the appropriate Procedure interface and use it to invoke the Map's forEachEntry API. Map.keySet and Map.values are not similarly encumbered; nevertheless, the forEachKey, forEachValue, and transformValues APIs will yield slightly better performance at the cost of compatibility with the interface of java.util.Map.
Copyright © GigaSpaces.