double 类型的有效位

最新推荐文章于 2024-11-09 21:25:05 发布

致一

最新推荐文章于 2024-11-09 21:25:05 发布

阅读量7.3k

点赞数 1

分类专栏： C++

本文链接：https://blog.csdn.net/lcalqf/article/details/74301275

版权

C++ 专栏收录该内容

33 篇文章 1 订阅

订阅专栏

// double has 53 bit significant, therefore 2^53=9007199254740992
    // is the largest integer that can be represented. For a floating
    // point value, that consumes more than 16 characters may result
    // in precision loss during conversion. Note that we are being
    // pessimistic in that some floating point values that consume
    // 17 characters may be represented without conversion loss.
    if ( json->cfg->decode_big_numbers_as_strings &&
         ( endptr - json->ptr > 16 ||
           fabs(token->value.number) > 9007199254740991.0 ) )
    {
        token->type = T_STRING;
        strbuf_reset(json->tmp);
        for (; json->ptr != endptr; json->ptr++ ) {
            strbuf_append_char_unsafe(json->tmp, json->ptr[0]);
        }
        token->value.string = strbuf_string(json->tmp, &token->string_len);
        return;
    }

double类型有效位是16位，

当用double类型表示的数超过16位时，为了输出正确的值，可以转换为string类型输出