[英]showbase and leading zero for stringstream
Here is the code: 这是代码:
std::wstringstream wss;
wss.setf(std::ios_base::hex, std::ios_base::basefield);
wss.setf(std::ios_base::showbase);
// wss << std::showbase;
// wss.width(2);
wss.fill('0');
wss << std::setw(2) << 7;
// wss << std::noshowbase;
wss.unsetf(std::ios_base::showbase);
wss << std::setw(2) << 0;
wss << std::setw(2) << 20;
wss << std::setw(2) << 6;
auto ret = wss.str();
If I set showbase
for the stream, I always get below output: 如果我为流设置showbase
,则总是得到以下输出:
0x7001406
, instead of 0x07001406
0x7001406
,而不是0x07001406
How can I get this zero before 7? 我怎样才能在7之前获得零? I don't want to set the prefix 0x
manually like wss << "0x"
. 我不想像wss << "0x"
那样手动设置前缀0x
。
Thanks in advance!! 提前致谢!!
Problem is: the prefix is part of the output width! 问题是:前缀是输出宽度的一部分! Try wss << std::setw(4) << 7;
尝试wss << std::setw(4) << 7;
for comparison (you now get 00x7
, which still is unwanted...). 进行比较(您现在得到00x7
,但仍然不需要...)。
Unfortunately, you cannot use precision
for integers to get a behaviour equivalent to printf("%#.2x\\n", 7);
不幸的是,您不能对整数使用precision
来获得与printf("%#.2x\\n", 7);
等效的行为printf("%#.2x\\n", 7);
, which obviously is what you intend... ,这显然是您打算的...
My personal variant is having my own converter: 我的个人变体是拥有自己的转换器:
template <typename T>
struct Hex
{
Hex(T value, size_t width) : value(value), width(width) { }
private:
T value;
size_t width;
template <typename Stream>
friend Stream& operator<<(Stream& s, Hex h)
{
auto fill = s.fill();
auto flags = s.flags();
s.fill('0');
s << "0x" << std::noshowbase << std::hex << std::setw(h.width) << h.value;
s.fill(fill);
s.flags(flags);
return s;
}
};
template <typename T>
auto hex(T t, size_t width = sizeof(T) * 2) { return Hex<T>(t, width); }
You now can use it as: 您现在可以将其用作:
wss << hex(7, 2);
getting even shorter than having wss << std::setw(2) << 7;
变得比wss << std::setw(2) << 7;
更短wss << std::setw(2) << 7;
and coming with nice default appropriate to size of the type... 并带有适合该类型大小的默认值...
One minor drawback yet: you need specialisations or overloads for signed and unsigned char, as for these, the character representation ( 0x0s
) is output instead of numerical value ( 0x73
). 还有一个小缺点:带符号的和无符号的char需要专门化或重载,因为对于这些字符,输出的是字符表示形式( 0x0s
)而不是数值( 0x73
)。
auto hex(char t, size_t width = sizeof(char) * 2)
{ return Hex<unsigned int>(t, width); }
auto hex(signed char t, size_t width = sizeof(signed char) * 2)
{ return Hex<signed int>(t, width); }
auto hex(unsigned char t, size_t width = sizeof(unsigned char) * 2)
{ return Hex<unsigned int>(t, width); }
You might replace 2
with CHAR_BIT / 4
in the default, depending on your needs/desires, might cover systems having eg CHAR_BIT == 16
better... 默认情况下,您可以将CHAR_BIT / 4
替换为2
,具体取决于您的需求/愿望,可能覆盖具有CHAR_BIT == 16
更好的系统...
Thanks @Aconcagua for your clue! 感谢@Aconcagua的提示!
I think we can use std::ios_base::adjustfield and std::ios_base::internal to do it like this: 我认为我们可以使用std :: ios_base :: adjustfield和std :: ios_base :: internal来做到这一点:
wss.setf(std::ios_base::hex, std::ios_base::basefield);
int oldFlag = wss.setf(std::ios_base::internal, std::ios_base::adjustfield);
Then 然后
wss.setf(std::ios_base::showbase);
wss.fill('0');
wss << std::setw(4) << 7;
wss.unsetf(std::ios_base::showbase);
// wss.setf(oldFlag);
wss << std::setw(2) << 0;
wss << std::setw(2) << 20;
wss << std::setw(2) << 6;
Then I get 0x07001406. 然后我得到0x07001406。 Correct me if I cannot do it like this, thanks! 如果我不能这样纠正我,谢谢!
声明:本站的技术帖子网页,遵循CC BY-SA 4.0协议,如果您需要转载,请注明本站网址或者原文地址。任何问题请咨询:yoyou2525@163.com.