redis源码之util

         util.c里面是一些常见的辅助处理算法，下面看一下。

/* Convert a string representing an amount of memory into the number of
 * bytes, so for instance memtoll("1Gb") will return 1073741824 that is
 * (1024*1024*1024).
 *
 * On parsing error, if *err is not NULL, it's set to 1, otherwise it's
 * set to 0. On error the function return value is 0, regardless of the
 * fact 'err' is NULL or not. */
long long memtoll(const char *p, int *err) {//将表示内存量的字符串转换为字节
    const char *u;
    char buf[128];
    long mul; /* unit multiplier */  //单位乘数
    long long val;
    unsigned int digits;

    if (err) *err = 0;

    /* Search the first non digit character. */ //搜索第一个非数字字符
    u = p;
    if (*u == '-') u++;
	
    while(*u && isdigit(*u)) u++;    //isdigit检查其参数是否为十进制数字字符
    
    if (*u == '\0' || !strcasecmp(u,"b")) { //strcasecmp比较两个字符串是否相等，且不区分大小写
        mul = 1;
    } else if (!strcasecmp(u,"k")) {
        mul = 1000;
    } else if (!strcasecmp(u,"kb")) {
        mul = 1024;
    } else if (!strcasecmp(u,"m")) {
        mul = 1000*1000;
    } else if (!strcasecmp(u,"mb")) {
        mul = 1024*1024;
    } else if (!strcasecmp(u,"g")) {
        mul = 1000L*1000*1000;
    } else if (!strcasecmp(u,"gb")) {
        mul = 1024L*1024*1024;
    } else {
        if (err) *err = 1;
        return 0;
    }

    /* Copy the digits into a buffer, we'll use strtoll() to convert
     * the digit (without the unit) into a number. */
    digits = u-p;
    if (digits >= sizeof(buf)) {
        if (err) *err = 1;
        return 0;
    }
    memcpy(buf,p,digits);
    buf[digits] = '\0';

    char *endptr;
    errno = 0;
    val = strtoll(buf,&endptr,10); //strtoll将字符串转换为长整数
    if ((val == 0 && errno == EINVAL) || *endptr != '\0') {
        if (err) *err = 1;
        return 0;
    }
    return val*mul;
}

此函数可以来将对应单位转换为多少字节，先找出对应单位，然后把值存储在字符串中并将字符串转成long long类型，最后返回乘积。

/* Convert a long long into a string. Returns the number of
 * characters needed to represent the number.
 * If the buffer is not big enough to store the string, 0 is returned.
 *
 * Based on the following article (that apparently does not provide a
 * novel approach but only publicizes an already used technique):
 *
 * https://www.facebook.com/notes/facebook-engineering/three-optimization-tips-for-c/10151361643253920
 *
 * Modified in order to handle signed integers since the original code was
 * designed for unsigned integers. */
int ll2string(char *dst, size_t dstlen, long long svalue) {//将long long转换为字符串
    static const char digits[201] =   //数组大小是200，下标是0~199
        "0001020304050607080910111213141516171819"         
        "2021222324252627282930313233343536373839"
        "4041424344454647484950515253545556575859"
        "6061626364656667686970717273747576777879"
        "8081828384858687888990919293949596979899";
    int negative;   //负
    unsigned long long value;

    /* The main loop works with 64bit unsigned integers for simplicity, so
     * we convert the number here and remember if it is negative. */
    //LLONG_MIN = -9223372036854775808     LLONG_MAX = 9223372036854775807
    if (svalue < 0) {
        if (svalue != LLONG_MIN) {
            value = -svalue;
        } else {
            value = ((unsigned long long) LLONG_MAX)+1;
        }
        negative = 1;
    } else {
        value = svalue;
        negative = 0;
    }

    /* Check length. */
    uint32_t const length = digits10(value)+negative;
    if (length >= dstlen) return 0;

    /* Null term. */
    uint32_t next = length;
    dst[next] = '\0';
    next--;
    while (value >= 100) {
        int const i = (value % 100) * 2;
        value /= 100;
        dst[next] = digits[i + 1];  //每次处理2个，这样可以少循环一倍
        dst[next - 1] = digits[i];
        next -= 2;
    }

    /* Handle last 1-2 digits. */
    if (value < 10) {
        dst[next] = '0' + (uint32_t) value;
    } else {
        int i = (uint32_t) value * 2;
        dst[next] = digits[i + 1];
        dst[next - 1] = digits[i];
    }

    /* Add sign. */
    if (negative) dst[0] = '-';//若为负数，在字符串开头加上负号标志
    return length;
}

这里借用了字符串数组，每次处理2个，少循环了一倍。

/* Convert a string into a long long. Returns 1 if the string could be parsed
 * into a (non-overflowing) long long, 0 otherwise. The value will be set to
 * the parsed value when appropriate.
 *
 * Note that this function demands that the string strictly represents
 * a long long: no spaces or other characters before or after the string
 * representing the number are accepted, nor zeroes at the start if not
 * for the string "0" representing the zero number.
 *
 * Because of its strictness, it is safe to use this function to check if
 * you can convert a string into a long long, and obtain back the string
 * from the number without any loss in the string representation. */
int string2ll(const char *s, size_t slen, long long *value) { //字符串转换成long long
    const char *p = s;
    size_t plen = 0;
    int negative = 0;
    unsigned long long v;

    /* A zero length string is not a valid number. */
    if (plen == slen)
        return 0;

    /* Special case: first and only digit is 0. */
    if (slen == 1 && p[0] == '0') {
        if (value != NULL) *value = 0;
        return 1;
    }

    /* Handle negative numbers: just set a flag and continue like if it
     * was a positive number. Later convert into negative. */
    if (p[0] == '-') {
        negative = 1;
        p++; plen++;

        /* Abort on only a negative sign. */
        if (plen == slen)
            return 0;
    }

    /* First digit should be 1-9, otherwise the string should just be 0. */
    if (p[0] >= '1' && p[0] <= '9') {
        v = p[0]-'0';  //字符 - '0' 就为数字
        p++; plen++;
    } else {
        return 0;
    }

    /* Parse all the other digits, checking for overflow at every step. */
    while (plen < slen && p[0] >= '0' && p[0] <= '9') {
        if (v > (ULLONG_MAX / 10)) /* Overflow. */
            return 0;
        v *= 10;

        if (v > (ULLONG_MAX - (p[0]-'0'))) /* Overflow. */
            return 0;
        v += p[0]-'0';

        p++; plen++;
    }

    /* Return if not all bytes were used. */
    if (plen < slen)
        return 0;

    /* Convert to negative if needed, and do the final overflow check when
     * converting from unsigned long long to long long. */
    if (negative) {
        if (v > ((unsigned long long)(-(LLONG_MIN+1))+1)) /* Overflow. */
            return 0;
        if (value != NULL) *value = -v;
    } else {
        if (v > LLONG_MAX) /* Overflow. */
            return 0;
        if (value != NULL) *value = v;
    }
    return 1;
}

注意long long类型值溢出的判断。

/* Get random bytes, attempts to get an initial seed from /dev/urandom and
 * the uses a one way hash function in counter mode to generate a random
 * stream. However if /dev/urandom is not available, a weaker seed is used.
 *
 * This function is not thread safe, since the state is global. */
void getRandomBytes(unsigned char *p, size_t len) {  //获取随机数种子
    /* Global state. */
    static int seed_initialized = 0;
    static unsigned char seed[20]; /* The SHA1 seed, from /dev/urandom. */
    static uint64_t counter = 0; /* The counter we hash with the seed. */

    if (!seed_initialized) {
        /* Initialize a seed and use SHA1 in counter mode, where we hash
         * the same seed with a progressive counter. For the goals of this
         * function we just need non-colliding strings, there are no
         * cryptographic security needs. */
        FILE *fp = fopen("/dev/urandom","r");
        if (fp == NULL || fread(seed,sizeof(seed),1,fp) != 1) { //fread从给定输入流stream读取最多count个对象到数组buffer中
            /* Revert to a weaker seed, and in this case reseed again
             * at every call.*/            
            for (unsigned int j = 0; j < sizeof(seed); j++) {   //fopen读取失败，使用弱种子
                struct timeval tv;
                gettimeofday(&tv,NULL);
                pid_t pid = getpid();
                seed[j] = tv.tv_sec ^ tv.tv_usec ^ pid ^ (long)fp;
            }
        } else {
            seed_initialized = 1;
        }
        if (fp) fclose(fp);
    }

    //sha1加密
    while(len) {   
        unsigned char digest[20];
        SHA1_CTX ctx;
        unsigned int copylen = len > 20 ? 20 : len;

        SHA1Init(&ctx);
        SHA1Update(&ctx, seed, sizeof(seed));//将新数据（原始数据、填充位、长度）依次附加到context->buffer[]
        SHA1Update(&ctx, (unsigned char*)&counter,sizeof(counter));
        SHA1Final(digest, &ctx);
        counter++;

        memcpy(p,digest,copylen);
        len -= copylen;
        p += copylen;
    }
}

获取随机数种子。

/* Given the filename, return the absolute path as an SDS string, or NULL
 * if it fails for some reason. Note that "filename" may be an absolute path
 * already, this will be detected and handled correctly.
 *
 * The function does not try to normalize everything, but only the obvious
 * case of one or more "../" appearing at the start of "filename"
 * relative path. */
sds getAbsolutePath(char *filename) { //给定文件名，将绝对路径作为SDS字符串返回
    char cwd[1024];
    sds abspath;
    sds relpath = sdsnew(filename);
	printf("relpath =%s\n",relpath);

    relpath = sdstrim(relpath," \r\n\t"); // \t 的意思是 横向跳到下一制表符位置
    if (relpath[0] == '/') return relpath; /* Path is already absolute. */

    /* If path is relative, join cwd and relative path. */
    if (getcwd(cwd,sizeof(cwd)) == NULL) { //getcwd()会将当前工作目录的绝对路径复制到参数buffer所指的内存空间中
        sdsfree(relpath);
        return NULL;
    }
    abspath = sdsnew(cwd);
	printf("abspath =%s\n",abspath);
    if (sdslen(abspath) && abspath[sdslen(abspath)-1] != '/')
        abspath = sdscat(abspath,"/");

    /* At this point we have the current path always ending with "/", and
     * the trimmed relative path. Try to normalize the obvious case of
     * trailing ../ elements at the start of the path.
     *
     * For every "../" we find in the filename, we remove it and also remove
     * the last element of the cwd, unless the current cwd is "/". */
    while (sdslen(relpath) >= 3 &&
           relpath[0] == '.' && relpath[1] == '.' && relpath[2] == '/')//排除../对应的目录
    {
        sdsrange(relpath,3,-1);
        if (sdslen(abspath) > 1) {
            char *p = abspath + sdslen(abspath)-2;
            int trimlen = 1;

            while(*p != '/') {
                p--;
                trimlen++;
            }
            sdsrange(abspath,0,-(trimlen+1));
        }
    }

    /* Finally glue the two parts together. */
    abspath = sdscatsds(abspath,relpath);
    sdsfree(relpath);
    return abspath;
}

注意对../目录的处理。