If you do not have example data, and an understanding of what the output should look like, then you cannot test it. Code that only works in your head is not code.
Do you need to use awk? There is a utility command called paste that does this job.
Awk generally only reads one file at once (there is a method to work round this, but it is regarded as "advanced").
The standard method is to read the first file into an array, and then stitch together those array elements with the corresponding lines of the second file. You need to take special care if the files are not of the same length.
Maybe try something like that yourself, and post your effort so we can help fix it. It should only be about three lines of awk.
@oste91 , disappointed that you suddenly show some example data after saying you had none, however , remaining focused on the issue.
why not go away and think on the process needed to 'interleave' lines from these two files.
DO NOT think about awk/paste or any other program , just think about how to do this logically - step by step
write that process down on a piece of paper (or electronically) , review it, walk through it step by step as if you were actually doing it physically ... once you have an algorithm that you are happy with, post it back here for the team to review, then we can look at making suggestions on what technologies to use.
(a) Your original post shows awk -F, which implies CSV (comma separated value) file format. This data is white-space separated (any combination of tabs and spaces, although plain tab-separated would be more common as you can then have multi-word text fields). Can you clarify ?
(b) It is fairly unusual to interleave data rows with a different number of fields. When you said "combine them", I immediately assumed your first output would be more like:
12,180,0.144,2490,4900,4039,4847,42829,1039,3948
Are you sure your assignment does not intend this? Either is easy, but good to answer the problem exactly.
Your original post specifies "within loop for". Awk does not need a for loop for lines -- it has a built-in cycle that processes lines one-by-one anyway. A Bash solution would need a loop, though. Is awk stipulated in the assignment, or did you pick it because it seemed like the right tool ?
(a) You set the field separator as ,, and there is no comma field separator in the data. When you print $1, $2, $3, field 1 is all of 12 180 0.144, and fields 2 and 3 are empty. It is not doing what you think it is.
(b) The first two lines with BEGIN output two lines of data that do not even come from the input files.
(c) Where does ice.dat come into it ?
(d) You are running awk four times for every row of output.
(e) Sure, it runs for one line because one line cannot be out of sequence. The same logic can never work for two or more lines, because each new awk starts over at line 1 of each file. You have no mechanism to get the second (third, fourth, ...) line from each file, apart from embedding constant data in the script, instead of getting it from the file.
It would help if you explained exactly what tools and methods your course has covered so far, so we can come up with suggestions that conform to your stage of knowledge. No point in providing a solution that you do not learn from, or that your course supervisor knows you didn't write yourself.
Exactly why does NR==1{print $"1"} print seven fields, not one like you asked it to ?
Why does it throw a fatal error ?
Why does it output five lines, not two?
Why are there two extra (invisible) spaces after 12 180 0.144 ?
But mainly, why can't you see that nothing in your script will ever do anything with any data after line 1 of the input files?
This is a script that does exactly what you ask for. It runs a Bash loop for 100 items, and in each iteration it passes in the value of the line number which awk should take from the files this time around.
#! /bin/bash --
: > new.dat #.. Empty the file.
for (( Nr = 1; Nr <= 100; ++Nr )); do
awk -v Nr="${Nr}" 'NR == Nr { print; }' AAA.dat >> new.dat
awk -v Nr="${Nr}" 'NR == Nr { print; }' BBB.dat >> new.dat
done
However, this is a joke (I have to say this because all the mods will otherwise laugh at me more that usual).
awk iterates right through the file for every value of Nr. So it reads two 100-line files 100 times -- 20,000 reads to pick out the 200 that match NR at some point. That scales at about O(n-cubed), and starting 200 processes means a large constant multiplier. I hope your trainer gets the joke too.
As there are only two rows in each test data file, the last 98 iterations never find NR == Nr because Nr will be higher than 2, and awk only sets NR to 1 and 2 before it hits EOF. But they still try all 100 values of Nr. A sensible version would count the lines in AAA.dat first to make the loop do the right number of iterations.
I do not comprehend your posted OUTPUT file.
(a) The data goes to new.dat, but the +++ bash command log comes out on the terminal. So you did an edit to put it together.
(b) The text 42829 (MAX VALUE each BBB.dat row) has magically dropped from the sky. Nothing in your code can possibly generate that. That speaks of a requirement you have not yet mentioned.
