std::num_get::get

std::num_get::get, std::num_get::do_get

1) Public member function, calls the member function do_get of the most derived class.

2) Reads characters from the input iterator in and generates the value of the type of v, taking into account IO stream formatting flags from str.flags(), character classification rules from std::use_facet<std::ctype<charT>>(str.getloc()), and numeric punctuation characters from std::use_facet<std::numpunct<charT>>(str.getloc()). This function is called by all formatted input stream operators such as std::cin >> n;.

Conversion occurs in three stages.

Stage 1: conversion specifier selection

• I/O format flags are obtained, as if by

fmtflags basefield =(str.flags()&std::ios_base::basefieldfmtflags boolalpha =(str.flags()&std::ios_base::boolalpha

• If the type of v is an integer type, the the first applicable choice of the following five is selected:

If basefield == oct, will use conversion specifier %o If basefield == hex, will use conversion specifier %X If basefield == 0, will use conversion specifier %i If the type of v is signed, will use conversion specifier %d If the type of v is unsigned, will use conversion specifier %u

• For integer types, length modifier is added to the conversion specification if necessary: h for short and unsigned short, l for long and unsigned long, ll for long long and unsigned long long

Stage 2: character extraction

• If in==end, Stage 2 is terminated immediately, no further characters are extracted

Stage 3: conversion and storage

• The sequence of chars accumulated in Stage 2 is converted to a numeric value

• If the conversion function fails to convert the entire field, the value ​0​ is stored in v

(since C++11)

• In any case, if the conversion function fails std::ios_base::failbit is assigned to err

Return value

in.

Notes

In C++98/C++03, if an error occurs, v is left unchanged. In C++11, it is set to a value as described above.

The result of converting a negative number string into an unsigned integer was specified to produce zero until C++17, although some implementations followed the protocol of std::strtoull which negates in the target type, giving LLONG_MAX for "-1", and so produce the largest value of the target type instead. As of C++17, strictly following std::strtoull is the correct behavior.

Because stage 2 filters out characters such as 'p', 'N' or 'i', the hexadecimal floating-point numbers such as "0x1.23p-10" and the strings "NaN" or "inf" may be rejected by do_get(double) even if they are valid input to strtod: this is LWG #2381.

Example

An implementation of operator>> for a user-defined type.

​

#include <iostream>
#include <iterator>
#include <locale>
 
struct base { long x; };
 
template <class CharT, class Traits>
std::basic_istream<CharT, Traits>&
    operator >>(std::basic_istream<CharT, Traits>& is,
                base& b)
{
    std::ios_base::iostate err = std::ios_base::goodbit;
    try // setting err could throw
    {
        typename std::basic_istream<CharT, Traits>::sentry s(is
 
        if (s) // if stream is ready for input
        {
            std::istreambuf_iterator<CharT, Traits> end;
            std::use_facet<std::num_get<CharT>>(is.getloc()).get(is, end, is, err, b.x
        }
    } catch(std::ios_base::failure& error)
    {
        // handle the exception
    }
    return is;
}
 
int main()
{
    base b;
 
    std::cin >> b;
}

​

See also

 © cppreference.com

Licensed under the Creative Commons Attribution-ShareAlike Unported License v3.0.

http://en.cppreference.com/w/cpp/locale/num_get/get