Shell Programming and Scripting

I wish I could explain it better. I will try out your suggestion in a bit. I apreciate the effort.

Mike

I agree. I wasn't familiar with stacking in this context, and, as you noticed, there was an error with a missing header in the original post. I also agree that your second example is clear.

The following is a more generalized version of vgersh99's approach:

Code:

awk '

NR == 1 {
    split($0, labels)
    pr_rec(h1 FS h2)
}

NR > 1 {
    for (i=a; i<=b; i++)
        pr_rec(labels[i] FS $i)
}

function pr_rec(s, r, i) {
    for (i=1; i<a; i++)
        r = r $i FS
    r = r s
    for (i=b+1; i<=NF; i++)
        r = r FS $i
    print r
}

' FS=, a=2 b=4 h1=' Label' h2=' Value' filename

a: 1-based index of first column to stack
b: 1-based index of last column to stack
h1: Name of first new column
h2: Name of second new column

The values assigned in the code correspond to your second example's data.

If I were going to deploy this code, in the NR==1 section I would also add a sanity check to confirm that a is positive and not greater than b, and that b is less than NF (the number of columns in the first line). If any of the checks fail, print a diagnostic to stderr and abort.

Regards,
Alister

I think the following example the poster provided is clear enough.

Get attribute names from first input row. Print special invariant first output row. For "Bob" input row, print separate output line for each attribute, with format: Bob, Attribute_Name, Attribute_Value, ID. Then, do same thing for "Jill" input row. Keep going until no more input rows.

If I'm wrong about the logic, please correct me. But it seems straightforward format to me.

Code:

Name, Sex, Height, Weight, ID 
Bob, M, 75, 220, 21345 
Jill, F, 61, 120, 43563 
Frank, M, 61, 300, 57864

Code:

Name, Label, Value, ID
Bob, Sex, M, 21345
Bob, Height, 75, 21345
Bob, Weight, 220, 21345
Jill, Sex, F, 43563
Jill, Height, 61, 43563 
Jill, Weight, 120, 43563 
Frank, Sex, M, 57864 
Frank, Height, 61, 57864 
Frank, Weight, 300, 57864

Sorry I don't have a solution. But this looks like a setup for awk.

Alister, it works great, It works on a huge database file in a fraction of a second.

I had no idea you would pass an array like that--nice!

I am trying to understand a couple things:
1. Why does r = r $i FS not result in a leading comma at the beginning of the line when the initial value of r passed to the function is ,?
2. What is the difference between the i passed to the function and the i in the loops? I don't think they can be the same variable for my understanding of the code to function.

Mike