Print all pairs of values in column of tsv file via bash - linux

Question

I want to write a small shell script to make a labeling task easier for myself. Alas, I'm still lacking the skills to do so.
I've got a news.tsv file which looks like this:
id foo bar text
1 a b lorem
2 c d ipsum
...
50 e f muspi
Where the actual entries in the text column are lengthy news articles. I want to print 2 of these texts at once, until all possible pairs in the column have been printed.
Searching for a solution, I found that awk might be the right tool for the task. I know how to print two specific entries in the text column, e.g,
awk -F '\t' 'NR==2 {print $4} NR==3 {print $4}' news.tsv
will print lorem and ipsum. For getting all pairs, I think I'll need a nested for-loop, but I fail at implementing it with awk.
My spaghetti-try looks like this:
awk -F '\t' '{for (i=0; i<50; i++){for (j=i+1; j<50; j++) if(i!=j){NR==i {print $4} NR==j {print $4}}}}' news.tsv
I'm open for other tools as well.

Answer 1

Is this what you want?
awk 'NR>1 {print $4; next; print $4}'

Answer 2

Here's an attempt based on your own script ... assuming that there are no embedded TABs, ever.
$ cat awkward
NR>1{
a[NR-1]=$4
}
END{
for(i=1;i<=50;i++){
for(j=1;j<=50;j++){
if(i!=j){
print a[i],a[j]
}
}
}
}
Invoked like so:
$ awk -f awkward test.tsv
Lorem ipsum
Lorem dolor
Lorem sit
Lorem amet,
Lorem consectetur
Lorem adipiscing
Lorem elit,
Lorem sed
Lorem do
Lorem eiusmod
.
.
[over 2000 lines stripped]
.
.
fugiat dolor
fugiat in
fugiat reprehenderit
fugiat in
fugiat voluptate
fugiat velit
fugiat esse
fugiat cillum
fugiat dolore
fugiat eu

Answer 3

If the total amount of the texts is not so huge, it will be efficient
store the whole texts in an array. Otherwise you need to read the input file
multiple times (the number of combination).
Then how about:
awk -F '\t' '
NR>1 {texts[++n]=$4}
END {
for (i=1; i<n; i++) {
for (j=i+1; j<=n; j++) {
print texts[i] " " texts[j]
}
}
}' news.tsv
Hope this helps.