CODEWITHSUNDEEP
awk
bserra
bserra

Reputation: 540

Deleting field,keeping header using command line

I've got a bunch of files that have more values (unnecessary values) than fields name and on top of it I've got a header I would like to keep.

For example, with a test_awk.txt file containing:

My header is here
it can have several lines
data1 data2 data3
1,2,3,4
2,3,4,5

What I want to have is the following:

My header is here
it can have several lines
data1,data2,data3
1,2,3
2,3,4

I tried with a simple awk command, but can only suppress the column for the whole file. Thus deleting my header, and most important, the last field name:

awk 'BEGIN{FS=OFS=","}NR>2{NF--;print}' test_awk.txt

gives:

data1,data2
1,2,3
2,3,4

Upvotes: 2

Views: 49

Answers (2)

RavinderSingh13
RavinderSingh13

Reputation: 133458

Following awk may help you in same.

awk -F' |,' '/^data/{val=NF;} /^[0-9]/ && NF>val{NF=val} 1' OFS=,   Input_file

Output will be as follows.

My header is here
it can have several lines
data1 data2 data3
1,2,3
2,3,4

Explanation: Adding non-one liner form with explanation too here:

awk -F' |,' '         ##Making field separator as space or comma for each line of Input_file here.
/^data/{              ##Checking condition here if a line is starting from string data, if yes then do following:
  val=NF;             ##Creating variable named val and its value is value of the number of fields on current line of Input_file.
}
/^[0-9]/ && NF>val{   ##Checking condition here if any line starts from digits and value of current NF is greater than variable val then do following:
  NF=val              ##Assigning the value of NF to variable named val here.
}
1                     ##Mentioning 1 here will make sure I we are making condition TRUE here and not mentioning any action here so by default print of current line will happen as an action.
' OFS=, Input_file    ##Setting OFS(output field separator) as comma here and mentioning Input_file here.

Upvotes: 2

Sundeep
Sundeep

Reputation: 23667

NR>3{NF--;print} means if NR>3, change NF and print it.. so this misses out printing lines NR<=3

$ awk 'BEGIN{FS=OFS=","} NR>3{NF--} 1' test_awk.txt
My header is here
it can have several lines
data1 data2 data3
1,2,3
2,3,4
  • NR>3{NF--} change number of fields only for line numbers > 3
  • 1 idiomatic way to print contents of $0


You can also use sed

$ sed '4,$ s/,[^,]*$//' test_awk.txt
My header is here
it can have several lines
data1 data2 data3
1,2,3
2,3,4
  • 4,$ substitution will apply to only these lines - i.e line numbers > 3
  • s/,[^,]*$// delete last field

Upvotes: 3

Related Questions