--- comments: true --- # 6.3 Hash algorithms The previous two sections introduced the working principle of hash tables and the methods to handle hash collisions. However, both open addressing and chaining can **only ensure that the hash table functions normally when collisions occur, but cannot reduce the frequency of hash collisions**. If hash collisions occur too frequently, the performance of the hash table will deteriorate drastically. As shown in the Figure 6-8 , for a chaining hash table, in the ideal case, the key-value pairs are evenly distributed across the buckets, achieving optimal query efficiency; in the worst case, all key-value pairs are stored in the same bucket, degrading the time complexity to $O(n)$. ![Ideal and worst cases of hash collisions](hash_algorithm.assets/hash_collision_best_worst_condition.png){ class="animation-figure" }
Figure 6-8 Ideal and worst cases of hash collisions
**The distribution of key-value pairs is determined by the hash function**. Recalling the steps of calculating a hash function, first compute the hash value, then modulo it by the array length: ```shell index = hash(key) % capacity ``` Observing the above formula, when the hash table capacity `capacity` is fixed, **the hash algorithm `hash()` determines the output value**, thereby determining the distribution of key-value pairs in the hash table. This means that, to reduce the probability of hash collisions, we should focus on the design of the hash algorithm `hash()`. ## 6.3.1 Goals of hash algorithms To achieve a "fast and stable" hash table data structure, hash algorithms should have the following characteristics: - **Determinism**: For the same input, the hash algorithm should always produce the same output. Only then can the hash table be reliable. - **High efficiency**: The process of computing the hash value should be fast enough. The smaller the computational overhead, the more practical the hash table. - **Uniform distribution**: The hash algorithm should ensure that key-value pairs are evenly distributed in the hash table. The more uniform the distribution, the lower the probability of hash collisions. In fact, hash algorithms are not only used to implement hash tables but are also widely applied in other fields. - **Password storage**: To protect the security of user passwords, systems usually do not store the plaintext passwords but rather the hash values of the passwords. When a user enters a password, the system calculates the hash value of the input and compares it with the stored hash value. If they match, the password is considered correct. - **Data integrity check**: The data sender can calculate the hash value of the data and send it along; the receiver can recalculate the hash value of the received data and compare it with the received hash value. If they match, the data is considered intact. For cryptographic applications, to prevent reverse engineering such as deducing the original password from the hash value, hash algorithms need higher-level security features. - **Unidirectionality**: It should be impossible to deduce any information about the input data from the hash value. - **Collision resistance**: It should be extremely difficult to find two different inputs that produce the same hash value. - **Avalanche effect**: Minor changes in the input should lead to significant and unpredictable changes in the output. Note that **"Uniform Distribution" and "Collision Resistance" are two separate concepts**. Satisfying uniform distribution does not necessarily mean collision resistance. For example, under random input `key`, the hash function `key % 100` can produce a uniformly distributed output. However, this hash algorithm is too simple, and all `key` with the same last two digits will have the same output, making it easy to deduce a usable `key` from the hash value, thereby cracking the password. ## 6.3.2 Design of hash algorithms The design of hash algorithms is a complex issue that requires consideration of many factors. However, for some less demanding scenarios, we can also design some simple hash algorithms. - **Additive hash**: Add up the ASCII codes of each character in the input and use the total sum as the hash value. - **Multiplicative hash**: Utilize the non-correlation of multiplication, multiplying each round by a constant, accumulating the ASCII codes of each character into the hash value. - **XOR hash**: Accumulate the hash value by XORing each element of the input data. - **Rotating hash**: Accumulate the ASCII code of each character into a hash value, performing a rotation operation on the hash value before each accumulation. === "Python" ```python title="simple_hash.py" def add_hash(key: str) -> int: """加法哈希""" hash = 0 modulus = 1000000007 for c in key: hash += ord(c) return hash % modulus def mul_hash(key: str) -> int: """乘法哈希""" hash = 0 modulus = 1000000007 for c in key: hash = 31 * hash + ord(c) return hash % modulus def xor_hash(key: str) -> int: """异或哈希""" hash = 0 modulus = 1000000007 for c in key: hash ^= ord(c) return hash % modulus def rot_hash(key: str) -> int: """旋转哈希""" hash = 0 modulus = 1000000007 for c in key: hash = (hash << 4) ^ (hash >> 28) ^ ord(c) return hash % modulus ``` === "C++" ```cpp title="simple_hash.cpp" /* 加法哈希 */ int addHash(string key) { long long hash = 0; const int MODULUS = 1000000007; for (unsigned char c : key) { hash = (hash + (int)c) % MODULUS; } return (int)hash; } /* 乘法哈希 */ int mulHash(string key) { long long hash = 0; const int MODULUS = 1000000007; for (unsigned char c : key) { hash = (31 * hash + (int)c) % MODULUS; } return (int)hash; } /* 异或哈希 */ int xorHash(string key) { int hash = 0; const int MODULUS = 1000000007; for (unsigned char c : key) { hash ^= (int)c; } return hash & MODULUS; } /* 旋转哈希 */ int rotHash(string key) { long long hash = 0; const int MODULUS = 1000000007; for (unsigned char c : key) { hash = ((hash << 4) ^ (hash >> 28) ^ (int)c) % MODULUS; } return (int)hash; } ``` === "Java" ```java title="simple_hash.java" /* 加法哈希 */ int addHash(String key) { long hash = 0; final int MODULUS = 1000000007; for (char c : key.toCharArray()) { hash = (hash + (int) c) % MODULUS; } return (int) hash; } /* 乘法哈希 */ int mulHash(String key) { long hash = 0; final int MODULUS = 1000000007; for (char c : key.toCharArray()) { hash = (31 * hash + (int) c) % MODULUS; } return (int) hash; } /* 异或哈希 */ int xorHash(String key) { int hash = 0; final int MODULUS = 1000000007; for (char c : key.toCharArray()) { hash ^= (int) c; } return hash & MODULUS; } /* 旋转哈希 */ int rotHash(String key) { long hash = 0; final int MODULUS = 1000000007; for (char c : key.toCharArray()) { hash = ((hash << 4) ^ (hash >> 28) ^ (int) c) % MODULUS; } return (int) hash; } ``` === "C#" ```csharp title="simple_hash.cs" /* 加法哈希 */ int AddHash(string key) { long hash = 0; const int MODULUS = 1000000007; foreach (char c in key) { hash = (hash + c) % MODULUS; } return (int)hash; } /* 乘法哈希 */ int MulHash(string key) { long hash = 0; const int MODULUS = 1000000007; foreach (char c in key) { hash = (31 * hash + c) % MODULUS; } return (int)hash; } /* 异或哈希 */ int XorHash(string key) { int hash = 0; const int MODULUS = 1000000007; foreach (char c in key) { hash ^= c; } return hash & MODULUS; } /* 旋转哈希 */ int RotHash(string key) { long hash = 0; const int MODULUS = 1000000007; foreach (char c in key) { hash = ((hash << 4) ^ (hash >> 28) ^ c) % MODULUS; } return (int)hash; } ``` === "Go" ```go title="simple_hash.go" /* 加法哈希 */ func addHash(key string) int { var hash int64 var modulus int64 modulus = 1000000007 for _, b := range []byte(key) { hash = (hash + int64(b)) % modulus } return int(hash) } /* 乘法哈希 */ func mulHash(key string) int { var hash int64 var modulus int64 modulus = 1000000007 for _, b := range []byte(key) { hash = (31*hash + int64(b)) % modulus } return int(hash) } /* 异或哈希 */ func xorHash(key string) int { hash := 0 modulus := 1000000007 for _, b := range []byte(key) { fmt.Println(int(b)) hash ^= int(b) hash = (31*hash + int(b)) % modulus } return hash & modulus } /* 旋转哈希 */ func rotHash(key string) int { var hash int64 var modulus int64 modulus = 1000000007 for _, b := range []byte(key) { hash = ((hash << 4) ^ (hash >> 28) ^ int64(b)) % modulus } return int(hash) } ``` === "Swift" ```swift title="simple_hash.swift" /* 加法哈希 */ func addHash(key: String) -> Int { var hash = 0 let MODULUS = 1_000_000_007 for c in key { for scalar in c.unicodeScalars { hash = (hash + Int(scalar.value)) % MODULUS } } return hash } /* 乘法哈希 */ func mulHash(key: String) -> Int { var hash = 0 let MODULUS = 1_000_000_007 for c in key { for scalar in c.unicodeScalars { hash = (31 * hash + Int(scalar.value)) % MODULUS } } return hash } /* 异或哈希 */ func xorHash(key: String) -> Int { var hash = 0 let MODULUS = 1_000_000_007 for c in key { for scalar in c.unicodeScalars { hash ^= Int(scalar.value) } } return hash & MODULUS } /* 旋转哈希 */ func rotHash(key: String) -> Int { var hash = 0 let MODULUS = 1_000_000_007 for c in key { for scalar in c.unicodeScalars { hash = ((hash << 4) ^ (hash >> 28) ^ Int(scalar.value)) % MODULUS } } return hash } ``` === "JS" ```javascript title="simple_hash.js" /* 加法哈希 */ function addHash(key) { let hash = 0; const MODULUS = 1000000007; for (const c of key) { hash = (hash + c.charCodeAt(0)) % MODULUS; } return hash; } /* 乘法哈希 */ function mulHash(key) { let hash = 0; const MODULUS = 1000000007; for (const c of key) { hash = (31 * hash + c.charCodeAt(0)) % MODULUS; } return hash; } /* 异或哈希 */ function xorHash(key) { let hash = 0; const MODULUS = 1000000007; for (const c of key) { hash ^= c.charCodeAt(0); } return hash & MODULUS; } /* 旋转哈希 */ function rotHash(key) { let hash = 0; const MODULUS = 1000000007; for (const c of key) { hash = ((hash << 4) ^ (hash >> 28) ^ c.charCodeAt(0)) % MODULUS; } return hash; } ``` === "TS" ```typescript title="simple_hash.ts" /* 加法哈希 */ function addHash(key: string): number { let hash = 0; const MODULUS = 1000000007; for (const c of key) { hash = (hash + c.charCodeAt(0)) % MODULUS; } return hash; } /* 乘法哈希 */ function mulHash(key: string): number { let hash = 0; const MODULUS = 1000000007; for (const c of key) { hash = (31 * hash + c.charCodeAt(0)) % MODULUS; } return hash; } /* 异或哈希 */ function xorHash(key: string): number { let hash = 0; const MODULUS = 1000000007; for (const c of key) { hash ^= c.charCodeAt(0); } return hash & MODULUS; } /* 旋转哈希 */ function rotHash(key: string): number { let hash = 0; const MODULUS = 1000000007; for (const c of key) { hash = ((hash << 4) ^ (hash >> 28) ^ c.charCodeAt(0)) % MODULUS; } return hash; } ``` === "Dart" ```dart title="simple_hash.dart" /* 加法哈希 */ int addHash(String key) { int hash = 0; final int MODULUS = 1000000007; for (int i = 0; i < key.length; i++) { hash = (hash + key.codeUnitAt(i)) % MODULUS; } return hash; } /* 乘法哈希 */ int mulHash(String key) { int hash = 0; final int MODULUS = 1000000007; for (int i = 0; i < key.length; i++) { hash = (31 * hash + key.codeUnitAt(i)) % MODULUS; } return hash; } /* 异或哈希 */ int xorHash(String key) { int hash = 0; final int MODULUS = 1000000007; for (int i = 0; i < key.length; i++) { hash ^= key.codeUnitAt(i); } return hash & MODULUS; } /* 旋转哈希 */ int rotHash(String key) { int hash = 0; final int MODULUS = 1000000007; for (int i = 0; i < key.length; i++) { hash = ((hash << 4) ^ (hash >> 28) ^ key.codeUnitAt(i)) % MODULUS; } return hash; } ``` === "Rust" ```rust title="simple_hash.rs" /* 加法哈希 */ fn add_hash(key: &str) -> i32 { let mut hash = 0_i64; const MODULUS: i64 = 1000000007; for c in key.chars() { hash = (hash + c as i64) % MODULUS; } hash as i32 } /* 乘法哈希 */ fn mul_hash(key: &str) -> i32 { let mut hash = 0_i64; const MODULUS: i64 = 1000000007; for c in key.chars() { hash = (31 * hash + c as i64) % MODULUS; } hash as i32 } /* 异或哈希 */ fn xor_hash(key: &str) -> i32 { let mut hash = 0_i64; const MODULUS: i64 = 1000000007; for c in key.chars() { hash ^= c as i64; } (hash & MODULUS) as i32 } /* 旋转哈希 */ fn rot_hash(key: &str) -> i32 { let mut hash = 0_i64; const MODULUS: i64 = 1000000007; for c in key.chars() { hash = ((hash << 4) ^ (hash >> 28) ^ c as i64) % MODULUS; } hash as i32 } ``` === "C" ```c title="simple_hash.c" /* 加法哈希 */ int addHash(char *key) { long long hash = 0; const int MODULUS = 1000000007; for (int i = 0; i < strlen(key); i++) { hash = (hash + (unsigned char)key[i]) % MODULUS; } return (int)hash; } /* 乘法哈希 */ int mulHash(char *key) { long long hash = 0; const int MODULUS = 1000000007; for (int i = 0; i < strlen(key); i++) { hash = (31 * hash + (unsigned char)key[i]) % MODULUS; } return (int)hash; } /* 异或哈希 */ int xorHash(char *key) { int hash = 0; const int MODULUS = 1000000007; for (int i = 0; i < strlen(key); i++) { hash ^= (unsigned char)key[i]; } return hash & MODULUS; } /* 旋转哈希 */ int rotHash(char *key) { long long hash = 0; const int MODULUS = 1000000007; for (int i = 0; i < strlen(key); i++) { hash = ((hash << 4) ^ (hash >> 28) ^ (unsigned char)key[i]) % MODULUS; } return (int)hash; } ``` === "Kotlin" ```kotlin title="simple_hash.kt" /* 加法哈希 */ fun addHash(key: String): Int { var hash = 0L for (c in key.toCharArray()) { hash = (hash + c.code) % MODULUS } return hash.toInt() } /* 乘法哈希 */ fun mulHash(key: String): Int { var hash = 0L for (c in key.toCharArray()) { hash = (31 * hash + c.code) % MODULUS } return hash.toInt() } /* 异或哈希 */ fun xorHash(key: String): Int { var hash = 0 for (c in key.toCharArray()) { hash = hash xor c.code } return hash and MODULUS } /* 旋转哈希 */ fun rotHash(key: String): Int { var hash = 0L for (c in key.toCharArray()) { hash = ((hash shl 4) xor (hash shr 28) xor c.code.toLong()) % MODULUS } return hash.toInt() } ``` === "Ruby" ```ruby title="simple_hash.rb" [class]{}-[func]{add_hash} [class]{}-[func]{mul_hash} [class]{}-[func]{xor_hash} [class]{}-[func]{rot_hash} ``` === "Zig" ```zig title="simple_hash.zig" [class]{}-[func]{addHash} [class]{}-[func]{mulHash} [class]{}-[func]{xorHash} [class]{}-[func]{rotHash} ``` ??? pythontutor "Code Visualization" It is observed that the last step of each hash algorithm is to take the modulus of the large prime number $1000000007$ to ensure that the hash value is within an appropriate range. It is worth pondering why emphasis is placed on modulo a prime number, or what are the disadvantages of modulo a composite number? This is an interesting question. To conclude: **Using a large prime number as the modulus can maximize the uniform distribution of hash values**. Since a prime number does not share common factors with other numbers, it can reduce the periodic patterns caused by the modulo operation, thus avoiding hash collisions. For example, suppose we choose the composite number $9$ as the modulus, which can be divided by $3$, then all `key` divisible by $3$ will be mapped to hash values $0$, $3$, $6$. $$ \begin{aligned} \text{modulus} & = 9 \newline \text{key} & = \{ 0, 3, 6, 9, 12, 15, 18, 21, 24, 27, 30, 33, \dots \} \newline \text{hash} & = \{ 0, 3, 6, 0, 3, 6, 0, 3, 6, 0, 3, 6,\dots \} \end{aligned} $$ If the input `key` happens to have this kind of arithmetic sequence distribution, then the hash values will cluster, thereby exacerbating hash collisions. Now, suppose we replace `modulus` with the prime number $13$, since there are no common factors between `key` and `modulus`, the uniformity of the output hash values will be significantly improved. $$ \begin{aligned} \text{modulus} & = 13 \newline \text{key} & = \{ 0, 3, 6, 9, 12, 15, 18, 21, 24, 27, 30, 33, \dots \} \newline \text{hash} & = \{ 0, 3, 6, 9, 12, 2, 5, 8, 11, 1, 4, 7, \dots \} \end{aligned} $$ It is worth noting that if the `key` is guaranteed to be randomly and uniformly distributed, then choosing a prime number or a composite number as the modulus can both produce uniformly distributed hash values. However, when the distribution of `key` has some periodicity, modulo a composite number is more likely to result in clustering. In summary, we usually choose a prime number as the modulus, and this prime number should be large enough to eliminate periodic patterns as much as possible, enhancing the robustness of the hash algorithm. ## 6.3.3 Common hash algorithms It is not hard to see that the simple hash algorithms mentioned above are quite "fragile" and far from reaching the design goals of hash algorithms. For example, since addition and XOR obey the commutative law, additive hash and XOR hash cannot distinguish strings with the same content but in different order, which may exacerbate hash collisions and cause security issues. In practice, we usually use some standard hash algorithms, such as MD5, SHA-1, SHA-2, and SHA-3. They can map input data of any length to a fixed-length hash value. Over the past century, hash algorithms have been in a continuous process of upgrading and optimization. Some researchers strive to improve the performance of hash algorithms, while others, including hackers, are dedicated to finding security issues in hash algorithms. The Table 6-2 shows hash algorithms commonly used in practical applications. - MD5 and SHA-1 have been successfully attacked multiple times and are thus abandoned in various security applications. - SHA-2 series, especially SHA-256, is one of the most secure hash algorithms to date, with no successful attacks reported, hence commonly used in various security applications and protocols. - SHA-3 has lower implementation costs and higher computational efficiency compared to SHA-2, but its current usage coverage is not as extensive as the SHA-2 series.Table 6-2 Common hash algorithms