Using sed to remove a column from a CSV

Question

I have a file where the fields are encapsulated with double quotes - general example of file:

"Internal ID", "External ID", "Name"
"123", "ABC", "ABC Incorporated"

Desired outcome, removing first column and saving the file back with the same name:

"External ID", "Name"
"ABC", "ABC Incorporated"

Can the above sed be modified to handle the "," delimiter? If so, how? Or are there better alternatives? Still a shell noob but I tried the following with no success:

    sed -i 's/[^'\"','\"']*,//' file.csv

Any help be greatly appreciated.

did you want to remove the first column? – Avinash Raj Aug 08 '14 at 15:20 — Avinash Raj, Aug 08 '14 at 15:20
Yes, the first column is to be removed. – lojkyelo Aug 08 '14 at 15:37 — lojkyelo, Aug 08 '14 at 15:37

score 3 · Answer 1 · edited May 23 '17 at 12:05

3

This is a job for cut.

 cut -d, -f 2- < file.csv

Use the shell to rename the output. Or use the script named inline at sed edit file in place and invoke cut as:

inline file.csv cut -d, -f 2-

edited May 23 '17 at 12:05

Community

1
1

answered Aug 08 '14 at 15:59

William Pursell

204,365
48
270
300

Avinash Raj · Accepted Answer · 2014-08-08T15:56:58.593

2

You could try the below sed command,

$ sed 's/^.*\(".*", ".*"\)$/\1/' file
"External ID", "Name"
"ABC", "ABC Incorporated"

It prints the last two values.

$ sed 's/^"[^"]*", //' file
"External ID", "Name"
"ABC", "ABC Incorporated"

It removes the first value in a comma separated string.

edited Aug 08 '14 at 15:56

answered Aug 08 '14 at 15:22

Avinash Raj

172,303
28
230
274

+1, but you don't need `g` when your regex is anchored -- it can only match once per line. – glenn jackman Aug 08 '14 at 15:50

score 0 · Answer 3 · answered Aug 08 '14 at 15:21

0

quick way:

sed 's/[^,]*,\s*//' file
"External ID", "Name"
"ABC", "ABC Incorporated"

Note this would fail if your field contains comma like "ABC,DEF","ABC Incor.."

answered Aug 08 '14 at 15:21

Kent

189,393
32
233
301

score 0 · Answer 4 · answered Aug 09 '14 at 06:07

0

Using Awk:

awk -F", " '{print substr($0, index($0,$2))}' file > tmp && mv tmp file

answered Aug 09 '14 at 06:07

John B

3,566
1
16
20

Using sed to remove a column from a CSV

4 Answers4