3排序

  这个是紧接着的下一道题.发现了,其实,随着数据种类多起来,分块也就多了,这也正是快速排序的来历吧.还有呢,就是,随着数据多起来,也就是会重复的情况下,也就会出现array[i] == array[j] == array[pivot]的时候,那样我这段代码的主循环也就无法运行了.于是,在median3 () 函数中先要将两头排好序, 以便在主循环中安全地使用前缀自增运算符.

  这本书里的练习题感觉不错,只是算法分析的题做着比较吃力,相信随着数学的学习,情况会好起来吧!

  贴上代码,呵呵.

---

/* 7-32-11-26-23.19.c -- 第七章第三十二题 */ #include <stdio.h> #define SIZE 22 #define FALSE 1 #define MAYBE 2 #define TRUE 3 #define PROBABLYFALSE 4 #define PROBABLYTRUE 5 int main (void) ; void print_array (const int * const array, const int size) ; void swap (int * const vala, int * const valb) ; int main (void) { int array[SIZE] = {TRUE, FALSE, FALSE, FALSE, TRUE, TRUE, TRUE, TRUE, TRUE, MAYBE, MAYBE, MAYBE, MAYBE, FALSE, FALSE, FALSE, FALSE, FALSE, MAYBE, MAYBE, MAYBE, FALSE} ; int i, j, size, index, probablyfalse, probablytrue ; probablyfalse = PROBABLYFALSE ; probablytrue = PROBABLYTRUE ; size = SIZE ; index = (size - 1) / 2 ; i = 0 ; j = size - 1 ; print_array (array, size) ; /* First step */ swap (array + index, &probablytrue) ; while (1) { while (FALSE == array[i] || MAYBE == array[i]) i++ ; while (TRUE == array[j]) j-- ; if (i < j) swap (array + i, array + j) ; else break ; } swap (array + i, &probablytrue) ; /* Second step */ if (TRUE == array[i]) i -= 1 ; index = i / 2 ; j = i ; i = 0 ; swap (array + index, &probablyfalse) ; while (1) { while (FALSE == array[i]) i++ ; while (MAYBE == array[j]) j-- ; if (i < j) swap (array + i, array + j) ; else break ; } swap (array + i, &probablyfalse) ; print_array (array, size) ; return 0 ; } void print_array (const int * const array, const int size) { int i ; for (i = 0; i < size; i++) printf ("%s/n", FALSE == array[i] ? "FALSE" : (MAYBE == array[i] ? "MAYBE" : "TRUE")) ; putchar ('/n') ; } void swap (int * const vala, int * const valb) { int temp ; temp = *vala ; *vala = *valb ; *valb = temp ; }