c++ - Which way to search for elements in a container is more efficient

Question

I'm just iterating/subscripting for values inside a container or string to see if those values are what I want. I just want to know which way (between numbers1 and numbers2) is more efficient.

bool isnumber (const char character)
{
    return(character>=48 && character<=57);             
}

template <typename T> bool numbers1(const T &str)
{
    if(!str.empty())
    {
        for (auto it=str.cbegin(); it!=str.cend(); ++it) if (!isnumber(*it)) return false;

        return true;
    }
    else return false;
}

template <typename T> bool numbers2(const T &str)
{
    if(!str.empty())
    {
        for (auto &i : str) if(!isnumber(i)) return false;

        return true;
    }
    else return false;
}

Just use [std::find](http://en.cppreference.com/w/cpp/algorithm/find) — Jesper Juhl, Nov 29 '16 at 06:54
And use `isdigit()` instead of implementing it yourself. But if you do implement it yourself, use `'0'` instead of 48. — John Zwinck, Nov 29 '16 at 13:35
@JohnZwinck I don't like to use isdigit() simply because it throws an exception every time I send a character that is not in range (-1 to 255), for example Ñ. — Yaz, Nov 29 '16 at 13:42

score 0 · Answer 1 · edited May 23 '17 at 12:30

0

Range-based for loops in C++ are essentially just shorthand for a loop that uses begin() and end() generally with iterators. In numbers2, the compiler expands the loop to use str.begin and str.end. The only real difference on a compiled level is that your handwritten iterator loop uses cbegin() and cend() instead of begin() and end().

Just further reading: If you make a class that implements begin() and end() then you will be able to perform range-based for loops on that class: How to make my custom type to work with "range-based for loops"?

edited May 23 '17 at 12:30

Community

1
1

answered Nov 29 '16 at 07:01

Baladash

1
1

"The only real difference on a compiled level is that your handwritten iterator loop uses cbegin() and cend() instead of begin() and end()" - and the fact that range-for only checks the end condition once whereas a classic for loop tests it on each iteration. – Jesper Juhl Nov 29 '16 at 14:57
So that means that range-for loops should not alter the container due to the fact than end() can be invalidated because of it? – Yaz Nov 29 '16 at 19:33

c++ - Which way to search for elements in a container is more efficient

1 Answers1