The use of hash table in C++

The STL in C++ provides hash_map to implement the hash table function. Before introducing the use of hash_map, let's understand the hash table from the hash function and hash conflict.

1. Hash function

The so-called hash function is the mapping from the key (Key) to the value (Value): Value
$Value=The\; value\; of\; H\left(Key\right)$
reflects the storage address of the keyword.
1. Numerical analysis method
Select a few digits in the keyword as the value, and generally select a few digits with a relatively even distribution of numbers.
$H(k_1{k}_2{k}_3{k}_4{k}_5{k}_6{k}_7)=k_2{k}_3{k}_5$
2. Direct addressing method
Select a linear function as the hash function
$H(Key)=a\ast Key+b$
3. Folding method Divide
the keyword into several segments with the same number (the last segment may not be enough), and then add them to get the value.
$H(k_1{k}_2{k}_3{k}_4{k}_5{k}_6{k}_7)=(k_1+k_4+k_7)(k_2+k_5)(k_3+k_6)$
 $k_1{k}_2{k}_3$
 $k_4{k}_5{k}_6$
$+k_7$
4. Take the
middle of the square method. After the keyword is squared, select the middle digits as the value.
5. Divide the remainder method and
divide the keyword by $P$ remainder obtained as a value, where P is a prime number smaller than the maximum length of the hash table.
$H(Key)=KeyMODP$
6. Random number method
Select a random function as the hash function
$H\left(Key\right)=random(Key)$

Two, hash collision

For different keywords, the same value may be obtained through the mapping of the hash function, that is, a conflict has occurred. Here are several ways to solve the hash conflict
1. Public overflow area method
Create a new memory to store the conflicting data
2. Re-detection method
When a conflict occurs, find a new value for the keyword through a function:
$Value=(H(Key)+d_i)MODL$
where$L$ is the length of the hash table, according to$d_i$The value of can be subdivided into the following three methods:

• Linear detection method
  $d_i=1，2，\cdots ，L-1$

• Square detection method
  $d_i=1^2，-1^2，2^2，-2^2，\cdots$

• Random detection method at
  this time $d_i$Take a set of random numbers

3. Re-hashing
When a conflict occurs, perform the hashing function again. Of course, this hashing function can be the same as the previous one, or it can be different
$Value=RH\left(H\right(Key\left)\right)$
4. Chain address method The
chain address method is a method of combining an array with a linked list. This is also the most commonly used method. Different keywords with the same value are stored in a linked list. , Different linked lists, that is, different values ​​form an array.

Three, the use of unordered_map

1. Header file:

#include< unordered_map>

2. Define a hash table (we take Key and Value as int variables as an example):

unordered_map<int,int> Hash;

3. There are several ways to establish a hash table:

Hash[1]=3;
Hash.insert<make_pair(1,3)>;
Hash.insert({ {1,3},{2,4} });

4. Iterator :

unordered_map<int,int>::iterator it;

5. Use iterators to access variables:

it->first;
it->second;

6. Lookup of the hash table:

it=Hash.find(1);

If it is not found, it will return Hash.end();
7. Modify the hash table:

Hash[1] = 4;

8. Clear the hash table:

Hash.erase(1);
Hash.clear();

Above we have introduced several basic usages of hash tables. For detailed introduction, please visit the following website:
https://docs.microsoft.com/en-us/previous-versions/visualstudio/visual-studio-2008/bb982522( v=vs.90)