Why do I need a prime read in c++?

Question

I have been back and forth with my professor regarding why a prime read is or is not needed in c++.

#include <iostream>
#include <fstream>

int addFunct(int& total, int& rNum);

int subFunct(int& total, int& rNum);


int main() {

    char op1;

    int rNum = 0;

    int total = 0;


    std::ifstream inFile;

    inFile.open("C:\\Users\\Administrator\\Desktop\\c++ files\\input.txt");
    if (inFile.fail()) {

        std::cout << "no read";
    }


    do {

        inFile >> op1 >> rNum;
        if (op1 == '+') {

            addFunct(total, rNum);

        }

        else if (op1 == '-') {

            subFunct(total, rNum);
        }

    }

    while (!inFile.eof());     //super while loop


    inFile.close();

    std::cout << "El total es: " << total;

    return 0;


}

int addFunct(int& total, int& rNum) {

    total = total + rNum;


    return total;

}

int subFunct(int& total, int& rNum) {

    total = total - rNum;

    return total;

}

If I place a prime read above my DO statement (e.g. inFile >> op1 >> rNum;) then the last value in the file will not be read. However, my professor argues that priming a read is important and necessary. I am wondering why this is and how I can resolve the issue of not performing operations on the last read in number.

And by *prime read* you mean... any reading operation executed as the first one in regard to a certain stream? Anyway, the safest and most reliable way of doing that is `while (inFile >> op1 >> rNum) { ... }`. Both you and your professor should read [why `f.eof()` is almost always wrong](https://stackoverflow.com/questions/5431941/why-is-while-feoffile-always-wrong). — Fureeish, Jul 17 '19 at 13:35
Besides, "*However, my professor argues that priming a read is important and necessary.*" - have you considered *asking them **why*** he claims that? — Fureeish, Jul 17 '19 at 13:37

score 3 · Accepted Answer · edited Jul 18 '19 at 15:49

Why do I need a prime read in c++?

You don't.

However, my professor argues that priming a read is important and necessary.

Did you ask them for an explanation? Or they just stated that and refused to explain?

To elaborate - both your and your professor's approaches are wrong. Both because they use std::fstream::eof as the loop condition, which is almost always wrong.

What you should do is move the reading to the loop condition, like so:

while (inFile >> op1 >> rNum) {
    if (op1 == '+') {
        addFunct(total, rNum);
    }
    else if (op1 == '-') {
        subFunct(total, rNum);
    }
}

This way you will not miss any data from the file. You will also correctly not enter the loop as soon as the reading fails, either due to end-of-file or inability to parse file content to char followed by an int.

This works because if a >> read fails, it sets certain failbits of the associated stream. Then, the >> operator returns the stream itself (in this case inFile) which is implicitly convertible to bool. It evaluates to true if there are no failbits set and to false otherwise.

Professor could not offer an explanation which puzzles me. I pressed for an answer and even received points off for not using a prime read, however, even after much research, I never found an answer to satisfy needing a prime read. Thank you for your explanation (especially the in-depth part) and for linking the other stackoverflow question. Your expertise is much appreciated! — wiregh0st, Jul 17 '19 at 14:01

score 0 · Answer 2 · answered Jun 11 '21 at 19:25

0

Priming read just loads the while loop for the first time it tests the condition. I think the professor thought it was important because it is a standard way to setup while loops for structured programing and Flow charting.

answered Jun 11 '21 at 19:25

Rick

1

Why do I need a prime read in c++?

2 Answers2