Reading Excel Row Cells A through J

hate to bug you prelude but you seem to understand what im trying to do. now im not asking you to do this for me but you've been great at explaining this to me today. so anyhoo. when i first ran the program it would calculate the mean and display only the first line. now when i added the following code

Code:

	//verify open was sucessful
	if(!inFile.fail())	//ifopen did not fail
	{
		//read cell a1 through j1
		inFile >> cell1 >> cell2 >> cell3 >> cell4 >> cell5 >> cell6 >> cell7 >> cell8
               >> cell9 >> cell10;
		while(!inFile.eof()){

        //calculate the mean for ROW and place in mean variable
		mean = (cell1+cell2+cell3+cell4+cell5+cell6+cell7+cell8+cell9+cell10)/10;
		inFile >> cell1 >> cell2 >> cell3 >> cell4 >> cell5 >> cell6 >> cell7 >> cell8
               >> cell9 >> cell10;
		}
	}

...it only calculates the mean for the very last set of floating point numbers and displays it. How in the world do i get it to display all the rows 1-2000???

well i got it to spit out 2000 rows but whats wierd is the calculations are all wrong for every row now. hmmmm maybe some pringles and soda pop will help illeviate this problem

Assuming your file is formatted correctly (it should be from what I saw), the following is something like what you want:

Code:

#include <fstream.h>
#include <iostream.h>

int
main()
{
  float    cell[10];
  ifstream in("healthy.txt");

  // Check for success :-)

  while (in>> cell[0]) {
    // If we get one, then assume the rest are there
    for (int i = 1; i < 10; i++) {
      in>> cell[i];
    }
    float sum;
    for (int i = 0; i < 10; i++) {
      sum += cell[i];
    }
    out<< sum / 10 <<endl;
  }

  return 0;
}

im going to take a break and try it out after my snack attack. but it seems your code will get the first row of 10 foating point numbers. Then my program needs to do that for the next 1999 rows which i dont see just yet but dont tell me unless your code wouldnt work for that. thanks again and hopefully i can figure this out tonight

>but it seems your code will get the first row of 10 foating point numbers.
My loop doesn't work with a "row" per se because the >> operator treats all whitespace alike. Just read values in groups of 10 and as long as all rows have 10 values, everything will work fine because if the first value isn't there, the condition will fail (on eof we hope) and terminate the loop at the right time.

okay this is what ive done today and it calculates the mean for the first row in the healthy.txt attachment i added earliear. I need it to do the rest of the 1999 rows as well...

Code:

#include<iostream.h>
#include<fstream.h>

void main()
{

	//might need variables to store in the mean
	float mean = 0.0;
	float cell1= 0.0;
	float cell2= 0.0;
	float cell3= 0.0;
	float cell4= 0.0;
	float cell5= 0.0;
	float cell6= 0.0;
	float cell7= 0.0;
	float cell8= 0.0;
	float cell9= 0.0;
	float cell10= 0.0;



	//open up file
	ifstream inFile;
	inFile.open("healthy.txt");


	//verify open was sucessful
	if(!inFile.fail())	//ifopen did not fail
	{
		//read cell a1 through j1
		inFile >> cell1 >> cell2 >> cell3 >> cell4 >> cell5 >> cell6 >> cell7 >> cell8
               >> cell9 >> cell10;
	
		
        //calculate the mean for ROW and place in mean variable
		mean = (cell1+cell2+cell3+cell4+cell5+cell6+cell7+cell8+cell9+cell10)/10;
		cout<<mean<<endl;
		
	}//end if


	
	else //open failed
		cout<<"Error opening file."<<endl;



}//end main

now when i added in the EOF function like so...

Code:

//verify open was sucessful
	if(!inFile.fail())	//ifopen did not fail
	{
		//read cell a1 through j1
		inFile >> cell1 >> cell2 >> cell3 >> cell4 >> cell5 >> cell6 >> cell7 >> cell8
               >> cell9 >> cell10;
		while(!inFile.eof()){

        //calculate the mean for ROW and place in mean variable
		mean = (cell1+cell2+cell3+cell4+cell5+cell6+cell7+cell8+c  ell9+cell10)/10;
                cout<<mean<<endl;
		inFile >> cell1 >> cell2 >> cell3 >> cell4 >> cell5 >> cell6 >> cell7 >> cell8
               >> cell9 >> cell10;
		}
	}

it calculates all 2000 rows and displays them but it calculates them wrong. I checked with my trusty rusty calculator several times. Do you see anything wrong with my logic on getting the mean for each row?

oh btw prelude when i tried your code i got errors and im such a noob i tried fixing them on my compiler but i just didnt see how. I hope this thread isnt breaking any forum rules

Well the code actually works but its starting halfway down the list and i need it to start at the beginning. any ideas. maybe im saving it wrong but all i did was copy all the cells i needed and pasted them into a .txt file. I tried saving it as a tab deliminated file in excel as well but it just saves and empty txt file without the needed data. this is going to be the longest thread ever!

This works just fine on my end:

Code:

#include <iostream.h>
#include <fstream.h>

int main()
{
  //might need variables to store in the mean
  float mean = 0.0;
  float cell1= 0.0;
  float cell2= 0.0;
  float cell3= 0.0;
  float cell4= 0.0;
  float cell5= 0.0;
  float cell6= 0.0;
  float cell7= 0.0;
  float cell8= 0.0;
  float cell9= 0.0;
  float cell10= 0.0;

  //open up file
  ifstream inFile;
  inFile.open("healthy.txt");
  //verify open was sucessful
  if(!inFile.fail())	//ifopen did not fail
  {
    //read cell a1 through j1
    while (inFile >> cell1 >> cell2 >> cell3 >> cell4 >> cell5 >> cell6 >> cell7 >> cell8 >> cell9 >> cell10)
    {
      //calculate the mean for ROW and place in mean variable
      mean = (cell1+cell2+cell3+cell4+cell5+cell6+cell7+cell8+cell9+cell10)/10;
      cout<<mean<<endl;
    }
  }//end if
  else //open failed
    cout<<"Error opening file."<<endl;
}//end main

>void main()
This isn't valid C++. The main function always returns int.

>while(!inFile.eof()){
stream.eof() wasn't designed to work as a loop condition. The eof flag only gets set after you fail to read from the stream due to end-of-file. That means that the last line of the file will usually be processed twice.

thank you prelude. When I desk check the calculations though I have found that both our codes are skipping rows for some reason. The final output is giving the results for only 299 rows. These seem to be random rows when i check with my calculator but the last two results are correct. The first one is not correct and I havent quite pin pointed which row from the healthy.txt my program is starting at. any suggestions on making it start at row1 and go through row2000? im mucho confusedo on this problemo

>any suggestions on making it start at row1 and go through row2000?
All of the code I've given to you works properly for all of the rows. Any errors in the mean are likely to be floating-point rounding errors that are present in computer science due to the imprecise nature of representing floating-point values in binary. Sadly, if I can't reproduce your problem then I can't help you with it. I would guess that your output code is the problem, not the input code.

i was beginning to think that. so you do get all 2000 rows? thank you so much for you help prelude

Why wait to print out all the rows? If you already have the data in readable form so you can use your calculator as a check on the computer calculations then you can check the first 10 or 20 rows one at a time. Once you are sure the first x number of rows calculate correctly you can display every tenth row or every 100th to check on them. All it takes is a strategically placed display sequence and a call for input to a dummy variable to hold your place so you can check the results if you are sure that something is going haywire someplace along the way. (I believe a debugger can help you do this if you understand how to use one. I do it manually, however, so I can't help you with a debugger.) If you can consistently reproduce the problem and you can pinpoint where/when it happens, then someone may be able to help you figure it out, even if it isn't reproduceable up front on their machine.

great idea elad. Only thing is the books I learned from only deal with sequential access files. It says in my textbooks that random and binary are beyond the scope of the books unfortunately. Im guessing that I should go learn about random access files with I/O #fstream. thanks for the help

you can do it with any type filehere's an example to check every line one at atime, which in this case is every 10 items

Code:

#include <fstream.h>
#include <iostream.h>
intmain()
{ 
  float cell[10]; 
  ifstream in("healthy.txt"); 
  char ch; int lineNum = 1; 
  // Check for success :-) 
  while (in >> cell[0]) 
  { 
	// If we get one, then assume the rest are there 
	for (int i = 1; i < 10; i++) 
	{ 
	  in>> cell[i]; 
	} 
 
	float sum = 0.0; 
	for (int i = 0; i < 10; i++) 
	{ 
	  sum += cell[i]; 
	} 
	 //display average for each row one at a time 
	 //to check for error before moving to next row 
	 cout << "average for row " << lineNum << " = "; 
	 cout<< sum / 10 <<endl; 
	 cin >> ch; 
	 ++lineNum; 
  } 
  return 0;
}