awk + multiple spaces as the delimiter. 2 [duplicate] - awk

Question

I have file with multiple spaces as the delimiter.
> cat file1.csv
col1 col2 col3 col4
col1 col2 col3
col1 col2
col1
col1 col2 col3 col4 col5
This is the output I want but with no empty new lines (which makes me wonder if my -F' {2,}' is working)
> awk -F' {2,}' 'NR==1{print $0}' file1.csv | tr " " "\n"
col1
col2
col3
col4
>
But I was hoping to do it with AWK and using the OFS, but not sure if doing it right
> awk -F' {2,}' 'BEGIN{OFS=","} NR==1{print $0}' file1.csv
col1 col2 col3 col4
Other work in this space for my ref
I want to do something like this but with
>cat file.csv
col1,col2,col3,col4
col1,col2,col3
col1,col2
col1
col1,col2,col3,col4,col5
> awk -F, 'NR==1{print $0}' file.csv
col1,col2,col3,col4
> awk -F, 'NR==1{print $0}' file.csv | tr "," "\n"
col1
col2
col3
col4
> awk -F, 'NR==1{print $0}' file.csv | tr "," "\n" | cat -n
1 col1
2 col2
3 col3
4 col4
I can just use sed to remove blank lines but I want to use awk OFS as above:
> awk -F' {2,}' 'NR==1{print $0}' file1.csv | tr " " "\n" | sed '/^[[:space:]]*$/d'
col1
col2
col3
col4
>

Answer 1

Assumptions:
we're only interested in the 1st line of the file
each field/column of the (1st) line is to be printed on a separate line
Sample inputs:
$ head file?.csv
==> file1.csv <==
col1 col2 col3 col4
col1 col2 col3
col1 col2
col1
col1 col2 col3 col4 col5
==> file2.csv <==
col1,col2,col3,col4
col1,col2,col3
col1,col2
col1
col1,col2,col3,col4,col5
For the 1st file (file1.csv) we can use the default input field delimiter (ie, white space):
$ awk '{for (i=1;i<=NF;i++) print $i; exit}' file1.csv
col1
col2
col3
col4
For the 2nd file (file2.csv) we use an input field delimiter of a comma (,):
$ awk -F',' '{for (i=1;i<=NF;i++) print $i; exit}' file2.csv
col1
col2
col3
col4
NOTE: in neither of these cases do we need to worry setting the output field separator (OFS)
If we absolutely, positively need to set, and use, a non-default OFS:
$ awk 'BEGIN{OFS="\n"} {$1=$1; print; exit}' file1.csv
col1
col2
col3
col4
$ awk -F',' 'BEGIN{OFS="\n"} {$1=$1; print; exit}' file2.csv
col1
col2
col3
col4

Answer 2

Given:
$ head file?.csv
==> file1.csv <==
col1 col2 col3 col4
col1 col2 col3
col1 col2
col1
col1 col2 col3 col4 col5
==> file2.csv <==
col1,col2,col3,col4
col1,col2,col3
col1,col2
col1
col1,col2,col3,col4,col5
You can use head and sed.
For a space / tab separated file:
$ head -1 file1.csv | sed -E 's/[[:blank:]]{1,}/\n/g'
col1
col2
col3
col4
Comma separated:
$ head -1 file2.csv | sed -E 's/,/\n/g'
col1
col2
col3
col4
You can also skip head altogether and use sed to 1) find line n, 2) do the replacement; 3) quit. Here for the fifth line:
sed -nE '5{s/([[:blank:]]{1,})/\n/g; p; q; }' file1.csv
col1
col2
col3
col4
col5
Just use 1 for the first line.
Or similarly with awk:
$ awk -v ln=5 -F"[[:blank:]]{1,}" 'FNR==ln{for(i=1;i<=NF;i++) print $i; exit}' file1.csv

Answer 3

if you just wanna de-dupe it all down to col1 to col5, try :
1 col1 col2 col3 col4
2 col1 col2 col3
3 col1 col2
4 col1
5 col1 col2 col3 col4 col5
6 col1,col2,col3,col4
7 col1,col2,col3
8 col1,col2
9 col1
10 col1,col2,col3,col4,col5
{m,g,n}awk '!__[$0]++' RS='[,[:space:]]+'
col1
col2
col3
col4
col5