check if the value is a numerical or a character

[rshen98]

I am new in this board and a new user. Here is my naive question:

I have some data file which are mixed with numerical values and characters. I want only numerical values. How can I check if a value is numerical or charcter? My data is looked like

HISTORIC PRECIPITATION [IN] STATION: 897

Annual Data Display: Summary for period ending 1929/ 9/30

Day Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug Sep

1 1.980 0.000 0.231 0.605 0.682 0.242 0.000 0.000 0.814 0.000 0.000 0.110
2 1.089 0.000 0.000 0.880 0.220 0.605 0.000 1.012 0.000 0.044 0.000 0.011
3 0.165 0.044 0.000 0.000 0.000 0.088 0.044 0.088 0.000 0.000 0.088 0.000
4 0.880 0.231 0.000 0.000 0.000 0.000 0.275 0.110 0.000 0.000 0.000 0.000
5 0.869 0.132 0.000 0.000 0.000 0.935 0.000 0.220 0.000 0.000 0.000 0.000

[TimP]

You could read each line as a character string. If you know what the acceptable formats are, and can distinguish the strings containing numerical data, you can use internal read to process the strings containing numeric fields after identifying them.

[rshen98]

Could you giv me an example how to do it?

[TimP]

character(len=96)instring,fname

open(unit=infile, file=fname)

read(infile,'(A)')instring

if(instring(1:22)='HISTORIC PRECIPITATION')then

[TimP]

character(len=96)instring,fname

open(unit=infile, file=fname)

read(infile,'(A)')instring

if(instring(1:22)='HISTORIC PRECIPITATION')then

[TimP]

Message Edited by tim18 on 08-25-2004 03:58 PM

[Steven_L_Intel1]

Whatever you do, don't do a read with format * (list-directed) in an attempt to "validate" the input. You will find it accepting far more than you want.

[TimP]

Sorry about the single = signs, where == is required.

[larsm]

Here is a solution that I extracted from my software. The basic procedures do some more checking (to get away from some of the things that list-directed format passes, as Steve warned about) and are more general, but this solution works well, especially if you can assert that the lines you are interested in processing have the same number of numbers in them and that they occur in the beginning of the line. I have found that this is often the case.

I attach the function, "get_dirtfile", and a driver for your example, that I have put in the "example.txt" file.

Best wishes

Lars

[anthonyrichards]

Why bother?

Why not paste the text into Excel and use Data menu to convert text to columns? Then you can pick out the numbers you want.

[rshen98]

Thank all of you for the help. I finally solve the problem. BTW, Excel doesn't help in this case because the file is large,has more rows taht excel can handle.