Triming spaces few columns

Shell Programming and Scripting
1

File contains 10 columns,i want trim the 2,4,10 columns using single unix command.
Please suggest me how to do?

Thanks

2

Neither title nor the description clears your requirement (at least to me).
Please post sample data and required output.

How the columns are separated?

3

Excluding the header row,only data.

4

You need to specify how each column is separated. Is it space or comma or pipe?
Assuming default field separator:

awk '{$2=$4=$10=""; print}' input
5

Sample data file test1.txt

CCNI | data564_input1 | 264 | CCNI | data564_input1 |264
ccni| data564_input1    
CORO1A|155

i tried with above data not triming the spaces.

6

Hi bmk,

Your file seems irregular. All lines have different number of columns. What about trimming spaces from fourth field if only exists two?

Regards,
Birei.

7

Just i want trim the space specific columns,it not necessary to check the 4 th column is there or not.
Just trim....

8

You can always post the exact requirement at the very first beginning.
This saves lots of time.

Try

sed 's/[ ][ ]*//g' file
9

Just i used this command it's trimming all columns,i want trim specific column as like 3 or 8,not mandatory for the file columns...

sed 's/[ ][ ]*//g' test1.txt

---------- Post updated at 06:37 AM ---------- Previous update was at 06:26 AM ----------

Just i tried it's trimming all columns ,i want trim specefic column as like 2 and 4...

sed 's/[ ][ ]*//g' test1.txt
10

One way using 'perl':

$ cat infile
CCNI | data564_input1 | 264 | CCNI | data564_input1 |264
ccni| data564_input1    
CORO1A|155
$ cat script.pl
use warnings;
use strict;

while ( <> ) {
    chomp;
    my @f = split /\|/;
    for ( my $i = 0; $i <= $#f; $i++ ) {
                if ( $i == 1 || $i == 3 || $i == 9 ) {
                        $f[ $i ] =~ s/\A\s*//; 
                        $f[ $i ] =~ s/\s*\Z//;
                }
        }
        printf qq[%s\n], join qq[|], @f;
}
$ perl script.pl infile
CCNI |data564_input1| 264 |CCNI| data564_input1 |264
ccni|data564_input1
CORO1A|155

Regards,
Birei

11

Thanks for quick reply. Is there any simple way using AWK command.

13

Hi Bmk,

Try the following, it will work

tr -cd '[:alnum:],|,\012' < filename

Note: I have tried this in LINUX(RedhatE4).

14

Hey ,i know this command,it's trimming all columns. Check my privioues comments.

---------- Post updated at 06:53 AM ---------- Previous update was at 06:51 AM ----------

@sdebasis,it's not working

15

Using awk:

$ awk 'BEGIN { FS = OFS = "|" } { for ( i = 1; i <= NF; i++ ) { if ( i == 2 || i == 4 || i == 10 ) { sub( /^\s*/, "", $i ); sub( /\s*$/, "", $i ) } } { print } }' infile
CCNI |data564_input1| 264 |CCNI| data564_input1 |264
ccni|data564_input1
CORO1A|155

Regards,
Birei

16

ok, got it.

It will trim all the spaces between the columns.

17

i tried with same data and same command,not trimming the spaces. Help me

18

I followed the below post.

19

@Birei,it tried with your command,not trimming 2,4 and 10..
Help me.

20

Ok. Why don't we do things right, then?

1.- I put two scripts, one in 'perl' and other in 'awk'. Which of those don't work?
2.- I tested my own scripts and they seem to work with the data provided by you. Are you testing with similar data or not?
3.- Paste your input data you are testing, put the script you are running and the output it gives, or the error description.

I know a little about programming but can not read minds for the moment. Some users gave a solution and none of them seem to work, so think if we all are mistaken or perhaps you are not explaining your problem correctly , so try to help us a little bit, please. Thank you.

Regards,
Birei.

21

@Birei, Please below one... Perl script is working fine....

-bash-3.2$ awk 'BEGIN { FS = OFS = "|" } { for ( i = 1; i <= NF; i++ ) { if ( i == 2 || i == 4 || i == 10 ) { sub( /^\s*/, "", $i ); sub( /\s*$/, "", $i ) } } { print } }' test1.txt
CCNI | data564_input1 | 264 |CCNI| data564_input1 |264
ccni| data564_input1
CORO1A| 155