英文单词统计

---

前言

实现使用了状态机和环形缓冲区统计英文单词数量，状态机分为在单词外部和单词内部两个状态，‘ ’ ’, ‘ - ’这种连接符统计为一个单词，换行时使用的‘ - ’的连接符也统计为一个单词。

---

一、代码

代码如下（示例）：

#include <stdio.h>
#include <string.h>

#define ABS(a, b) ((a) > (b))?(((a) - (b)):((b) - (a)))
#define RW_OFFSET(r, w, size) (((w) >= (r))?((w)-(r)):((w)+(size)-(r)))
#define RW_IDX_INC(rw, size) (((rw) >= (size - 1))?(rw = 0):(rw++))
#define RW_IDX_NEXT(rw, cnt, size) (((rw + cnt%size) >= (size))?(cnt%size - 1):(rw + cnt%size))
#define RW_IDX_LAST(rw, cnt, size) (((int)(rw - cnt%size) < (0))?(size - cnt%size):(rw - cnt%size))

#define MAX_SIZE (10u)
typedef enum
{
	OUT_WORD = 0,
	IN_WORD = 1,
} STATE_MACH;

STATE_MACH state_mach = OUT_WORD;


/** 判断是否是字母
*** 输入字符 ch
*** return 0:不是字母，1：是字母
*/
int is_alphabet(char ch)
{
	int ret = 0;
	if ((('a' <= ch) && ('z' >= ch)) || (('A' <= ch) && ('Z' >= ch)))
	{
		ret = 1;
	}
	
	return ret;
}

int main(int argc, char* argv[])
{
	char ch;
	char ch_arr[MAX_SIZE];
	int r_idx = 0, w_idx = 0;
	int sum_word = 0;
	FILE *fp = fopen(argv[1], "r");
	if (NULL == fp)
	{
		printf("%s[%d] fopen file %s is failed\n", __FUNCTION__, __LINE__, argv[1]);
		return -1;
	}
	
	fgets(ch_arr, MAX_SIZE, fp);
	r_idx = 0;
	w_idx = strlen(ch_arr);
	
	while (EOF != ch_arr[r_idx])
	{
		while ((RW_OFFSET(r_idx, w_idx, MAX_SIZE) < (MAX_SIZE / 2))
			&& (EOF != ch_arr[RW_IDX_LAST(w_idx, 1, MAX_SIZE)]))    /**< 缓冲数据保持在一定数量 */
		{
			ch_arr[w_idx] = fgetc(fp);
			RW_IDX_INC(w_idx, MAX_SIZE);
		}
		
		switch (state_mach)
		{
			case OUT_WORD:
				if (1 == is_alphabet(ch_arr[r_idx]))
				{
					state_mach = IN_WORD;
					sum_word++;
				}
				break;
				
			case IN_WORD:
				if (0 == is_alphabet(ch_arr[r_idx]))
				{
					if (('\'' == ch_arr[r_idx]) || ('-' == ch_arr[r_idx]))
					{
						if ((RW_OFFSET(r_idx, w_idx, MAX_SIZE) > 0) && (is_alphabet(ch_arr[RW_IDX_NEXT(r_idx, 1, MAX_SIZE)]))) /**< 连接符 */
						{
							RW_IDX_INC(r_idx, MAX_SIZE);
							break;
						}
						else if (('-' == ch_arr[r_idx])
							  && (RW_OFFSET(r_idx, w_idx, MAX_SIZE) > 2) 
						      && (0x0d == ch_arr[RW_IDX_NEXT(r_idx, 1, MAX_SIZE)])
							  && (0x0a == ch_arr[RW_IDX_NEXT(r_idx, 2, MAX_SIZE)])
							  && (is_alphabet(ch_arr[RW_IDX_NEXT(r_idx, 3, MAX_SIZE)]))) /**< 行末尾连接符 */
						{
							RW_IDX_INC(r_idx, MAX_SIZE);
							RW_IDX_INC(r_idx, MAX_SIZE);
							RW_IDX_INC(r_idx, MAX_SIZE);
							break;
						}
					}
					state_mach = OUT_WORD;
				}
				break;
				
			default:
				break;
		}
		RW_IDX_INC(r_idx, MAX_SIZE);
	}
	
	printf("%s[%d] file %s word has %d\n", __FUNCTION__, __LINE__, argv[1], sum_word);
}