CODEWITHSUNDEEP
regexawk
Another.Chemist
Another.Chemist

Reputation: 2559

AWK, split content in files if pattern at specific column matchs

Good day,

I was wondering how to split a matrix into files if pattern at specific column match.

Pattern: when find zero at second column, split.

So far I have done:

cat file | awk -FS"\t" '$2==0 {close("result"f);f++}{print $0 > "result"f}'

Input

1   2
2   3
4   0
5   6
7   0

Expected output

File 1
1   2
2   3
4   0

File 2
5   6
7   0

Thank in advance for any clue

Upvotes: 2

Views: 163

Answers (2)

John1024
John1024

Reputation: 113844

awk -v n=1 '{print>("file-" n)} $2==0 {n++}' input

Explanation:

  • From your input data, there appears to be no need to set the field separator to a tab. By default, awk splits fields on any whitespace, tabs included.

  • -v n=1

    The name of the output file is determined by the variable n. We start it at 1.

  • {print>("file-" n)}

    This prints the current line to a file whose name depends on n.

  • $2==0 {n++}

    If the second column is zero, we increment n so that the next line goes to a new file.

Upvotes: 1

Etan Reisner
Etan Reisner

Reputation: 80931

-FS"\t" isn't doing what you think it is. awk sees that as -F 's\t' that is you are setting FS to s<tab>.

You want -F"\t" or -v FS="\t".

You also need to print out the current line before you close the old file.

Upvotes: 1

Related Questions