Loss of integer

[Sgreen]

Hello, I am practicing a problem in Ch3 of Friedman c++ book, where functions are introduced. The problem is:

Write a function that calculates the elapsed time in minutes between a start time and an end time expressed as integers on a 24-hourclock (8:30 P.M = 2030). You need to deal only with end times occurring later on the same day as the start time. Also write a driver program to test your function.

Certain inputs result in a loss of a minute. For example 1144 to 1215, where 1145 or 1146 and 1143work fine. Why does this happen and how would I fix it?

#include <cstdlib>

#include <iostream>

#include <cmath>

using namespace std;

int calculations(int, int);

int main()
{
    int start,
    end;
    
    cout << "Please input your start time in military time(example: 8:3=2030): ";
    cin >> start;
    cout << "Please input your end time: ";
    cin >> end;
    
    int calculations(int start, int end);


    cout << "Your elapsed time is " << calculations(start, end) << " minutes.";

    return 0;
}

int calculations(int s, int e)         //calculations to determine the difference in time
{
    float startHr,                  //separation of hours and minutes
            endHr,
            startDec,               //hours and minutes in decimal form 
            endDec,
            startMin,             //minutes when separated from hours
            endMin;
    
    startHr = s/100;                //calculations to separate the hour to integer form
    floor(startHr);
    startMin = s%100;              //calculations to find remainder in minutes
    startDec = startMin/60 + startHr;   //converting minutes to decimal form and adding to hours
    
    endHr = e/100;                    //same calculations for end time
    floor(endHr);
    endMin = e%100;
    endDec = endMin/60 + endHr;
    
    return (endDec - startDec)*60;  //return of final calculations in decimal form multiplied by 60 to convert to minutes
}

Any other criticism is welcome as I would like to look as professional/perfect as possible. "Title" comments are excluded obviously, but I am here to learn! Thanks!

[wildblue]

Just a couple of things off the bat.

You don't need line 21.

line 39 and 44 - these return values but you're not saving them anywhere. Did you mean to update the values of startHr and endHr there?

[Chervil]

Looks like you have integer division, which will discard the fractional part.
e.g. startHr = s/100;
both s and 100 are integers.
You could use s/100.0f to specify that 100 is of type float, not int.

[cire]

Using floating point variables when it isn't needed should be avoided.

floor returns a value. It does not change the variable fed to it, so the correct way to use it is: variable = floor(variable);

Consider the following:

#include <iostream>

using namespace std;

unsigned difference_in_minutes(unsigned, unsigned);

int main()
{
    unsigned start, end;

    cout << "Please input your start time in military time(example: 8:3=2030): ";
    cin >> start;

    cout << "Please input your end time: ";
    cin >> end;

    cout << "Your elapsed time is " << difference_in_minutes(start, end) << " minutes.";

    return 0;
}

unsigned minutes_since_midnight(unsigned military_time)
{
    const unsigned hours = military_time / 100;
    const unsigned minutes = military_time % 100;

    return hours * 60 + minutes;
}

unsigned difference_in_minutes(unsigned begin_time, unsigned end_time)
{
    const unsigned beg = minutes_since_midnight(begin_time);
    const unsigned end = minutes_since_midnight(end_time);

    return end - beg;
}

http://ideone.com/jJN4IS

[Irrelevant Elephant]

at the end, when you return, it is being converted to an integer, so rounding is not what you would want:
(input) 1144 1215
the program would return:
30.99...
which will be converted to an integer, so it will return 30 minutes. A good way of rounding it would be:
float r = ...
int rounded = (int)(r + 0.5f);
so it will be rounded, which will return 31 minutes rather than 30:

return (int) ( ( (endDec - startDec) * 60.0f) + 0.5f);
Note that the brackets are important, because (int) converts the expression to its right to an integer, and this operation is done before any addition, etc, so it will convert + 0.5f to an int, which will be 0.

[Sgreen]

Thanks much for such quick responses! Now bear with me as I try to interpret with my nooby brain.
First, I see my obvious line 21 and floor mistake.
Next, half of what was said had no been learned yet (adding .0f or unsigned), but I did some research to figure out what those mean.
But - trying to avoid possibly using more advanced techniques as the book states to use float (unsigned, even though it is far less messy) and using what I have, I've been trying to figure out how to use this answer:

Looks like you have integer division, which will discard the fractional part.
e.g. startHr = s/100;
both s and 100 are integers.
You could use s/100.0f to specify that 100 is of type float, not int.
Here is my modification with error:

#include <cstdlib>

#include <iostream>

#include <cmath>

using namespace std;

int calculations(int, int);

int main()
{
    int start,
    end;
    
    cout << "Please input your start time in military time(example: 8:30=2030): ";
    cin >> start;
    cout << "Please input your end time: ";
    cin >> end;
    


    cout << "Your elapsed time is " << calculations(start, end) << " minutes.";

    return 0;
}

int calculations(int s, int e)         //calculations to determine the difference in time
{
    float startHr,                  //separation of hours and minutes
            endHr,
            startDec,               //hours and minutes in decimal form 
            endDec,
            startMin,             //minutes when separated from hours
            endMin;
    
    startHr = s/100.0f;                //calculations to separate the hour to integer form
    startHr=floor(startHr);
    startMin = s%100.0f;              //calculations to find remainder in minutes
    startDec = startMin/60.0f + startHr;   //converting minutes to decimal form and adding to hours
    
    endHr = e/100.0f;                    //same calculations for end time
    endHr=floor(endHr);
    endMin = e%100.0f;
    endDec = endMin/60.0f + endHr;
    
    return (endDec - startDec)*60.0f;  //return of final calculations in decimal form multiplied by 60 to convert to minutes
}

I receive an error:
error: invalid operands of types ‘int’ and ‘float’ to binary ‘operator%’
endMin = e%100.0f;
(Using Netbeans btw)
I tried to just add .0f to each the s/100 and e/100 but the result was the same.
Trying to add the .0f to every other constant obviously doesn't work, why is that? How could I make that work? Is the rounding solution the only other way? (as stated by Irrelevant Elephant)

[Chervil]

I think my previous reply may not have been appropriate here, looks like i didn't read the question properly.

[keskiverto]

invalid operands of types ‘int’ and ‘float’ to binary ‘operator%’

Modulus (%) expects integral operands, not floats, and returns an integral.

Overall, you should not use floats at all. Convert the times to minutes for the computation.
For example:
1144 -> 704
1215 -> 735
Diff == 31

[Sgreen]

Ah of course, as cire pointed out >.>.
Thanks a lot guys, you have educated me! :D

[keskiverto]

the book states to use float

Where does it state so? Not in your OP.

[Sgreen]

the book states to use float

Where does it state so? Not in your OP.

I apolagize as I went back and even misread what the book meant, which was to use float rather than double or long double, so even that update was incorrect.

I have run into a new issue as I have been working on this problem all day :(

Assume your friend has solved the previous programming exercise. Write another function that uses your friend’s function to calculate the speed (km/h) one must average to reach a certain desitnation by a designated time. Function inputs include same-day departure and arrival times as integers on a 24-hour clock and the distance to the destination in kilometers. Also write a driver program to test your function.

Updated code:

#include <iostream>

using namespace std;

int timeDifference(int, int);
float calcSpeed(float);

int main()
{
    int start,
    end,
            distance;
    
    cout << "Please input your start time in military time(example: 8:30=2030): ";  //request for user to input start time
    cin >> start;
    cout << "Please input your end time: ";                                      //request for user to input end time
    cin >> end;
    

    cout << "Your elapsed time is " << timeDifference(start, end) << " minutes."<<endl;   //result of calculations displayed
    
    cout << "How far must you travel in km?: ";
    cin >> distance;
    
    cout << "Your velocity needed is: " << calcSpeed(distance) << " km/hr.";

    return 0;
}

int minutesSinceMidnight(int time)         //function to convert time to total minutes
{
    
    int hours = time/100;        //separation of time to hours
    int minutes = time%100;      //separation of minutes
    return hours*60 + minutes;         //calculation to return total minutes of hours and minutes

}    
int timeDifference(int s, int e)             //function to subtract start and end time
{
    int begin = minutesSinceMidnight(s);  //declaration of start and end times
    int end = minutesSinceMidnight(e);
    return end - begin;                   //calculation to return minute difference
}

float calcSpeed(float x)
{
    float time = timeDifference(start, end);
    float velocity;
    velocity = x/time/60;
    return velocity;
    
}

errors:
54:33: error: ‘start’ was not declared in this scope
float time = timeDifference(start, end);
54:40: error: ‘end’ was not declared in this scope
float time = timeDifference(start, end);

I believe there are other methods to solving the problem, but I believe the issue I have here is taking the solution from the timeDifference calculation, and inserting it into the calcSpeed function. Is there a way to do that? Thanks!

Edit: Upon reflection, I was wondering why even use int or float when it seemed unsigned could have been used as a universal data type?

[Maharaja]

#include <iostream>

using namespace std;

int timeDifference(int, int);
float calcSpeed(float, int);

int main()
{
    int start, end, distance;

    cout << "Please input your start time in military time(example: 8:30=2030): ";  //request for user to input start time
    cin >> start;
    cout << "Please input your end time: ";                                      //request for user to input end time
    cin >> end;


    cout << "Your elapsed time is " << timeDifference(start, end) << " minutes."<<endl;   //result of calculations displayed

    cout << "How far must you travel in km?: ";
    cin >> distance;

    cout << "Your velocity needed is: " << calcSpeed(distance, timeDifference(start, end)) << " km/hr.";

    return 0;
}

int minutesSinceMidnight(int time)         //function to convert time to total minutes
{

    int hours = time/100;        //separation of time to hours
    int minutes = time%100;      //separation of minutes
    return hours*60 + minutes;         //calculation to return total minutes of hours and minutes

}
int timeDifference(int s, int e)             //function to subtract start and end time
{
    int begin = minutesSinceMidnight(s);  //declaration of start and end times
    int end = minutesSinceMidnight(e);
    return end - begin;                   //calculation to return minute difference
}

float calcSpeed(float x, int diff)
{
    float time = diff;
    float velocity;
    velocity = x/time/60;
    return velocity;

}

should take care of the scope problem...I think