Reading from file

[Asif]

Hi everyone!

I was playing with some code from a tutorial and was a bit perplexed with the result. Can someone please explain what's going on?

The following data are in a file:
-------------------
This is some test data
"x-values","y-values"
0.0,1.4
1.0,4.3
2.0,6.2
3.0,5.1
4.0,6.4
5.0,7.7
6.0,7.9
7.0,8.4
8.0,7.3
9.0,6.2
10.0,9.1
=======================================
'Following code reads those data from the file

Dim xvals(100) As Single
Dim yvals(100) As Single
Dim npts As Integer

Sub GetDataFromDisk()
Open ThisWorkbook.Path + "\Task2.2.dat" For Input As #1
Line Input #1, Title$
Cells(1, 1) = Title$
Input #1, xtit$, ytit$
Cells(2, 1) = xtit$
Cells(2, 2) = ytit$
npts = 0
While Not EOF(1)
npts = npts + 1
Input #1, xvals(npts), yvals(npts)
Cells(2 + npts, 1) = xvals(npts)
Cells(2 + npts, 2) = yvals(npts)
Wend
Close #1
End Sub
========================

And this is what I got (look at the y-values):
x-values y-values
0 1.39999997615814
1 4.30000019073486
2 6.19999980926513
3 5.09999990463256
4 6.40000009536743
5 7.69999980926513
6 7.90000009536743
7 8.39999961853027
8 7.30000019073486
9 6.19999980926513
10 9.10000038146972

 

[Dan E]

Dim xvals(100) As Single
Dim yvals(100) As Single
Dim npts As Integer
' Dimension variables used in the code, due to the dimensioning of
' xvals and yvals as array's with 101 items, this code will work for text
' files with up to 101 data points.


Sub GetDataFromDisk()
' Open a text file named "Task2.2.dat" from the same directory as
' the current excel file is in...
Open ThisWorkbook.Path + "\Task2.2.dat" For Input As #1
' Get the first line (assumed to be a title)
Line Input #1, Title$
' Put the title in cell A1
Cells(1, 1) = Title$
' Get the second line (assumed to be column headers)
Input #1, xtit$, ytit$
' Put the column headers in A2 and B2
Cells(2, 1) = xtit$
Cells(2, 2) = ytit$
' Set the number of data points to 0
npts = 0
' While their is still data in the file do . . .
While Not EOF(1)
' Increment points counter
npts = npts + 1
' Get an x and y
Input #1, xvals(npts), yvals(npts)
' Put them in cells A(2+counter) and B(2+counter)
Cells(2 + npts, 1) = xvals(npts)
Cells(2 + npts, 2) = yvals(npts)
' Loop
Wend
' Close the open file
Close #1
End Sub
========================

And this is what I got (look at the y-values):
x-valuesy-values
0 1.39999997615814
1 4.30000019073486
2 6.19999980926513
3 5.09999990463256
4 6.40000009536743
5 7.69999980926513
6 7.90000009536743
7 8.39999961853027
8 7.30000019073486
9 6.19999980926513
10 9.10000038146972


Note: I've seen much better code than this!

Dan E

 

[Dan E]

Provide an example of the differences and I can try, though their
are people on this board who know more about data types, etc...
than me.

Dan E

 

[Asif]

Here's the original data:

"x-values","y-values"
0.0,1.4
1.0,4.3
2.0,6.2
3.0,5.1
4.0,6.4
5.0,7.7
6.0,7.9
7.0,8.4
8.0,7.3
9.0,6.2
10.0,9.1


And here's the output:

x-values y-values
0 1.39999997615814
1 4.30000019073486
2 6.19999980926513
3 5.09999990463256
4 6.40000009536743
5 7.69999980926513
6 7.90000009536743
7 8.39999961853027
8 7.30000019073486
9 6.19999980926513
10 9.10000038146972

 

[Dan E]

The error is a result of assigning the values to single precision
data types. This topic has been discussed here previously, here
are some threads on that subject

http://www.google.com/groups?hl=en&...ail.com&rnum=34&prev=/groups?q=excel+single%2
Bprecision%2Berror%26start%3D30%26hl%3Den%26lr%3D%26ie%3DUTF-8%26selm%3D3DF01491.8090500%2540no_e-mail.com%26rnum%3D34

http://www.google.com/groups?hl=en&...tCHA.2636@TK2MSFTNGP11&rnum=1&prev=/&frame=on

http://www.google.com/groups?hl=en&[email protected]&rnum=1&prev=/groups?hl=
en%26lr%3D%26ie%3DUTF-8%26selm%3D131ac1c1.0308060311.df78cc2%2540posting.google.com

watch out for word wrap on the above links you may need to
cut and paste them into your browser.

Dan E

 

[Tom Ogilvy]

Fix you clock - it isn't december yet.

 

[Asif]

Thanks very much, Dan. You explained it very well. But can you explain why
the y-values are slightly different from what they were originally (in the
file)?


Dan E wrote in message ...