I have a file containing numeric data. Each line has a varying number of columns. I want to Split the file in to multiple files based on number of columns in a Line. Each Line may have columns varying from 1-10
Below is a sample Input
file.txt
23 53;
34;
31 45 67;
46 78 95;
34 17;
19;
37 65 83;
Target Output
file_1column.txt
34;
19;
file_2column.txt
23 53;
34 17;
file_3column.txt
31 45 67;
46 78 95;
37 65 83;
With awk, creating the desired filenames by concatenating fixed strings "file_" and"column.txt" with the internal variable NF (which contains the number of fields - or columns - in each input record - or line):
awk '{print > "file_" NF "column.txt"}' file.txt
Result:
$ head file_?column.txt
==> file_1column.txt <==
34;
19;
==> file_2column.txt <==
23 53;
34 17;
==> file_3column.txt <==
31 45 67;
46 78 95;
37 65 83;
awk is processing the file line by line, as usual, and for each line it is executing a program that reads "print the current line to the file that is named after the current line's number of columns (fields)". Think about it as if you were sorting a deck of cards - reading the number printed on each one an putting it on the corresponding pile. (Sorry for the invasion, steeldriver).
we can with below command too
command
split -l 3 file.txt
it will split into 3 files with xaa,xab,xcc now we renaming xaa==: file_xaa and so on
ls -ltr xa*| awk '{print "mv" " " $NF " " "file_"$NF}'| sh
n@praveen:~$ cat file_xaa
23 53;
34;
praveen@praveen:~$ cat file_xab
31 45 67;
46 78 95;
34 17;
praveen@praveen:~$ cat file_xac
19;
37 65 83;
