位运算 详解 技巧

原文来源于网络，已经无法考证出处，我将原文进行了一个理解性的翻译和总结：

• Set union A | B 并集
• Set intersection A & B 交集
• Set subtraction A & ~B 求差集
• Set negation ~A 取反
• Set bit A |= 1 << bit 将某一位置1
• Clear bit A &= ~(1 << bit) 将某一位置0
• Test bit (A & 1 << bit) != 0 测试某一位是否为1
• Remove last bit A&(A-1) 清除最右边的一个1
• Extract last bit A&-A or A&~(A-1) or x^(x&(x-1)) 提取出来最右边的一个1
• Get all 1-bits ~0 拿到所有位全是1的值

Examples

Count the number of ones in the binary representation of the given number 求1的个数

int count_one(int n) {
    while(n) {
        n = n&(n-1);
        count++;
    }
    return count;
}

Is power of four (actually map-checking, iterative and recursive methods can do the same) 求是否是4的次方数

bool isPowerOfFour(int n) {
    return !(n&(n-1)) && (n&0x55555555);
    //check the 1-bit location;
}

^ tricks

Sum of Two Integers

int getSum(int a, int b) {
    return b==0? a:getSum(a^b, (a&b)<<1); //be careful about the terminating condition;
}

Missing Number

Given an array containing n distinct numbers taken from 0, 1, 2, …, n, find the one that is missing from the array. For example, Given nums = [0, 1, 3] return 2. (Of course, you can do this by math.)

int missingNumber(vector<int>& nums) {
    int ret = 0;
    for(int i = 0; i < nums.size(); ++i) {
        ret ^= i;
        ret ^= nums[i];
    }
    return ret^=nums.size();
}

| tricks

Keep as many 1-bits as possible
求<=给定的数的2的幂
Find the largest power of 2 (most significant bit in binary form), which is less than or equal to the given number N.
将连续的两位置1，将连续的四位置1，直到将连续的32位置1。最后结果右移。

long largest_power(long N) {
    //changing all right side bits to 1.
    N = N | (N>>1);
    N = N | (N>>2);
    N = N | (N>>4);
    N = N | (N>>8);
    N = N | (N>>16);
    return (N+1)>>1;
}

Reverse Bits

Reverse bits of a given 32 bits unsigned integer.

Solution
利用中间变量来记录每一位是不是1

uint32_t reverseBits(uint32_t n) {
    unsigned int mask = 1<<31, res = 0;
    for(int i = 0; i < 32; ++i) {
        if(n & 1) res |= mask;
        mask >>= 1;
        n >>= 1;
    }
    return res;
}

& tricks

Just selecting certain bits

Reversing the bits in integer
同样是翻转一个数，利用了与的办法，将每两位翻转，每四位翻转，以此类推。

x = ((x & 0xaaaaaaaa) >> 1) | ((x & 0x55555555) << 1);
x = ((x & 0xcccccccc) >> 2) | ((x & 0x33333333) << 2);
x = ((x & 0xf0f0f0f0) >> 4) | ((x & 0x0f0f0f0f) << 4);
x = ((x & 0xff00ff00) >> 8) | ((x & 0x00ff00ff) << 8);
x = ((x & 0xffff0000) >> 16) | ((x & 0x0000ffff) << 16);

Bitwise AND of Numbers Range

Given a range [m, n] where 0 <= m <= n <= 2147483647, return the bitwise AND of all numbers in this range, inclusive. For example, given the range [5, 7], you should return 4.

Solution

int rangeBitwiseAnd(int m, int n) {
    int a = 0;
    while(m != n) {
        m >>= 1;
        n >>= 1;
        a++;
    }
    return m<<a; 
}

Number of 1 Bits

Write a function that takes an unsigned integer and returns the number of ’1’ bits it has (also known as the Hamming weight).

Solution

int hammingWeight(uint32_t n) {
    int count = 0;
    while(n) {
        n = n&(n-1);
        count++;
    }
    return count;
}

Application

Repeated DNA Sequences

All DNA is composed of a series of nucleotides abbreviated as A, C, G, and T, for example: “ACGAATTCCG”. When studying DNA, it is sometimes useful to identify repeated sequences within the DNA. Write a function to find all the 10-letter-long sequences (substrings) that occur more than once in a DNA molecule.
For example,
Given s = “AAAAACCCCCAAAAACCCCCCAAAAAGGGTTT”,
Return: [“AAAAACCCCC”, “CCCCCAAAAA”].
给了一个DNA序列，问这个序列中有哪些长度为10的子串多次出现过？
Solution
每两位保存一个hash值，一共20位保存10个字符的hash值，然后利用keymap比较是否多次出现。

class Solution {
public:
    vector<string> findRepeatedDnaSequences(string s) {
        int sLen = s.length();
        vector<string> v;
        if(sLen < 11) return v;
        char keyMap[1<<21]{0};
        int hashKey = 0;
        for(int i = 0; i < 9; ++i) hashKey = (hashKey<<2) | (s[i]-'A'+1)%5;
        for(int i = 9; i < sLen; ++i) {
            if(keyMap[hashKey = ((hashKey<<2)|(s[i]-'A'+1)%5)&0xfffff]++ == 1)
                v.push_back(s.substr(i-9, 10));
        }
        return v;
    }
};

But the above solution can be invalid when repeated sequence appears too many times, in which case we should use unordered_map< int, int> keyMap to replace char keyMap[1<<21]{0}here.
当数据太多的时候推荐使用unordered_map，这个底层是一个哈希表，用这个效率和空间都会好很多。
————————分割线————未完待续———
Majority Element

Given an array of size n, find the majority element. The majority element is the element that appears more than ⌊ n/2 ⌋ times. (bit-counting as a usual way, but here we actually also can adopt sorting and Moore Voting Algorithm)

Solution

int majorityElement(vector<int>& nums) {
    int len = sizeof(int)*8, size = nums.size();
    int count = 0, mask = 1, ret = 0;
    for(int i = 0; i < len; ++i) {
        count = 0;
        for(int j = 0; j < size; ++j)
            if(mask & nums[j]) count++;
        if(count > size/2) ret |= mask;
        mask <<= 1;
    }
    return ret;
}

Single Number III

Given an array of integers, every element appears three times except for one. Find that single one. (Still this type can be solved by bit-counting easily.) But we are going to solve it by digital logic design

Solution

//inspired by logical circuit design and boolean algebra;
//counter - unit of 3;
//current   incoming  next
//a b            c    a b
//0 0            0    0 0
//0 1            0    0 1
//1 0            0    1 0
//0 0            1    0 1
//0 1            1    1 0
//1 0            1    0 0
//a = a&~b&~c + ~a&b&c;
//b = ~a&b&~c + ~a&~b&c;
//return a|b since the single number can appear once or twice;
int singleNumber(vector<int>& nums) {
    int t = 0, a = 0, b = 0;
    for(int i = 0; i < nums.size(); ++i) {
        t = (a&~b&~nums[i]) | (~a&b&nums[i]);
        b = (~a&b&~nums[i]) | (~a&~b&nums[i]);
        a = t;
    }
    return a | b;
}

Maximum Product of Word Lengths

Given a string array words, find the maximum value of length(word[i]) * length(word[j]) where the two words do not share common letters. You may assume that each word will contain only lower case letters. If no such two words exist, return 0.

Example 1:
Given [“abcw”, “baz”, “foo”, “bar”, “xtfn”, “abcdef”]
Return 16
The two words can be “abcw”, “xtfn”.
Example 2:
Given [“a”, “ab”, “abc”, “d”, “cd”, “bcd”, “abcd”]
Return 4
The two words can be “ab”, “cd”.
Example 3:
Given [“a”, “aa”, “aaa”, “aaaa”]
Return 0
No such pair of words.
Solution

Since we are going to use the length of the word very frequently and we are to compare the letters of two words checking whether they have some letters in common:

using an array of int to pre-store the length of each word reducing the frequently measuring process;
since int has 4 bytes, a 32-bit type, and there are only 26 different letters, so we can just use one bit to indicate the existence of the letter in a word.

int maxProduct(vector<string>& words) {
    vector<int> mask(words.size());
    vector<int> lens(words.size());
    for(int i = 0; i < words.size(); ++i) lens[i] = words[i].length();
    int result = 0;
    for (int i=0; i<words.size(); ++i) {
        for (char c : words[i])
            mask[i] |= 1 << (c - 'a');
        for (int j=0; j<i; ++j)
            if (!(mask[i] & mask[j]))
                result = max(result, lens[i]*lens[j]);
    }
    return result;
}

Attention

result after shifting left(or right) too much is undefined
right shifting operations on negative values are undefined
right operand in shifting should be non-negative, otherwise the result is undefined
The & and | operators have lower precedence than comparison operators
Sets
All the subsets
A big advantage of bit manipulation is that it is trivial to iterate over all the subsets of an N-element set: every N-bit value represents some subset. Even better, if A is a subset of B then the number representing A is less than that representing B, which is convenient for some dynamic programming solutions.

It is also possible to iterate over all the subsets of a particular subset (represented by a bit pattern), provided that you don’t mind visiting them in reverse order (if this is problematic, put them in a list as they’re generated, then walk the list backwards). The trick is similar to that for finding the lowest bit in a number. If we subtract 1 from a subset, then the lowest set element is cleared, and every lower element is set. However, we only want to set those lower elements that are in the superset. So the iteration step is just i = (i - 1) & superset.

vector

// bitset::count
#include <iostream>       // std::cout
#include <string>         // std::string
#include <bitset>         // std::bitset

int main () {
  std::bitset<8> foo (std::string("10110011"));
  std::cout << foo << " has ";
  std::cout << foo.count() << " ones and ";
  std::cout << (foo.size()-foo.count()) << " zeros.\n";
  return 0;
}