Hash table structure
+ Hash table is a linked list consisting of an array, there is a first array, each position in the array are used to store a list, the list of base node: [hash value, key value, value value, next], when stored when a key-value pair, the first call hashcode () method hashCode key is obtained, and then a hash value is calculated by the algorithm, when a different key to get the same hash value as a key behind a node connected to the same hash value to the previous the node key.
Algorithm hash value
Worst algorithm: hashcode / hashcode all the elements will be stored in an array subscript, it has actually reduced to an array
The algorithm used is : hashcode% array length
This algorithm obtained by [0, array length -1], i.e. hash values. This algorithm can be made as uniform as possible distribution of the respective key in the array. This table is also called a "hash (hash translation) table."
Optimization algorithms : the same is take the remainder, but is optimized. First agreed length of the array is an integer power of 2 , then quickly attain the hash value calculation by bit: hashcode & (array length -1)
Below about checking algorithm:
public class Test {
public static void main(String[] args) {
Test t = new Test();
@ Algorithm: hashcode% the length of the array, assuming the array length (2 to the power 4) 16
System.out.println(t.hashCode()%16);
System.out.println(t.hashCode()&(16-1));
}
}
result:
2
2
Followed by another optimized once, the hash algorithm is currently used (I do not understand):
get the underlying method
When calling the get method it will be passed by a key value, call hashCode () method to obtain a value corresponding hashCode, and then obtain the same hash value, and the bit bucket array (hash table is an array) by calculating a position corresponding to the list one by one equals comparison, the same as the return value.
equals hashCode method of comparison is a value, if a.equals (b) is true, then a and b has a value equal hashCode.
Expansion of the hash table
HashMap bit bucket array, the initial size of 16. When the bit array in the tub reaches an element (0.75 * array length), to re-adjust the size of the array is 2 times the original.
But expansion is very time-consuming.
When the bit bucket array list length is greater than 8, the list will be converted to a red-black tree, which will greatly improve search efficiency.