How to delete a row in a file?

Shell Programming and Scripting
1

Does anyone know a way to delete rows 6-9 from the below output? I have searched the forum but did not find any thing helpful.

backups01laxint.liuc(s){jsandova}[~]<0>$ nsrjb -v | awk -F' ' '$3>=01{print $0}' | cut -b 1-79 | cat -n
setting verbosity level to `1'
1 slot volume used pool barcode
2 1: intwindows.0005 9% intwindows 010159L2
3 2: intunix.0009 19% intunix 010155L2
4 4: intrman.0002 2% intrman 010395L2
5 24: Cleaning Tape (50 uses left) 000002L1 000002L1
6 *not registered in the NetWorker media data base
7 7 registered volume(s), 7 volumes are less than 80% full.
8 1335 GB estimated capacity, 1278 GB remaining (4% full)
9 Default slot range(s) are 1-23

2

If you mean remove the 6th to 9th line, try this:

#!/usr/bin/perl
# You may need to change this path to /usr/local/bin/perl

$linecount = 0;

open (FILE,"input.txt");
@lines = <FILE>;
close (FILE);

open (FILE,">input.txt");
foreach $l (@lines) {
 $linecount++;
 if ($linecount ne 6 && $linecount ne 7 && $linecount ne 8 && $linecount ne 9) {
  print FILE $l;
 }
}
close (FILE);

exit;

If you mean the lines numbered 6 to 9, try:

#!/usr/bin/perl
# You may need to change this path to /usr/local/bin/perl

open (FILE,"input.txt");
@lines = <FILE>;
close (FILE);

open (FILE,">input.txt");
foreach $l (@lines) {
 if ($l !~ /^6/ && $l !~ /^7/ && $l !~ /^8/ && $l !~ /^9/) {
  print FILE $l;
 }
}
close (FILE);

exit;
3 
grep -v "^[6-9] " file

if you have PHP

<?php
$file = "file";
$lines = file($file);
foreach ( $lines as $k=>$v){
 $k=(int)$k;;
 if ( ( $k < 6 ) || ( $k > 9 )){
    echo "$v";
 }
}
?>
4

ghostdog74,
I tried that grep -v example however it didn't work. Is there a way to delete rows with awk or sed?

backups01laxint.liuc(s){jsandova}[~]<0>$ nsrjb -v | awk -F' ' '$3>=01{print $0}' | cut -b 1-79 | cat -n | grep -v "^[6-9] "
setting verbosity level to `1'
1 slot volume used pool barcode
2 1: intwindows.0005 9% intwindows 010159L2
3 2: intunix.0009 19% intunix 010155L2
4 4: intrman.0002 2% intrman 010395L2
5 24: Cleaning Tape (50 uses left) 000002L1 000002L1
6 *not registered in the NetWorker media data base
7 7 registered volume(s), 7 volumes are less than 80% full.
8 1335 GB estimated capacity, 1278 GB remaining (4% full)
9 Default slot range(s) are 1-23

5

it works for me

# grep -v "^[6-9] " file
backups01laxint.liuc(s){jsandova}[~]<0>$ nsrjb -v | awk -F' ' '$3>=01{print $0}' | cut -b 1-79 | cat -n
setting verbosity level to `1'
1 slot volume used pool barcode
2 1: intwindows.0005 9% intwindows 010159L2
3 2: intunix.0009 19% intunix 010155L2
4 4: intrman.0002 2% intrman 010395L2
5 24: Cleaning Tape (50 uses left) 000002L1 000002L1
6

It's not working for the original poster because the lines do not start with the numbers.

Here is what he's actually starting with.

backups01laxint.liuc(s){jsandova}[~]<0>$ nsrjb -v |  awk -F' ' '$3>=01{print $0}'  | cut -b 1-79 | cat -n | grep -v "^[6-9] "
setting verbosity level to `1'
     1    slot  volume                                     used  pool        barcode
     2       1: intwindows.0005                              9%  intwindows  010159L2
     3       2: intunix.0009                                19%  intunix     010155L2
     4       4: intrman.0002                                 2%  intrman     010395L2
     5      24: Cleaning Tape (50 uses left) 000002L1                            000002L1
     6          *not registered in the NetWorker media data base
     7        7 registered volume(s), 7 volumes are less than 80% full.
     8     1335 GB estimated capacity, 1278 GB remaining (4% full)
     9          Default slot range(s) are 1-23

Even my 2nd Perl version won't work. Need this:

#!/usr/bin/perl
# You may need to change this path to /usr/local/bin/perl

open (FILE,"input.txt");
@lines = <FILE>;
close (FILE);

open (FILE,">input.txt");
foreach $l (@lines) {
 if ($l !~ /^(\s|\t)*[6-9]/) {
  print FILE $l;
 }
}
close (FILE);

exit;
7

Here is ghostdog74's version adjusted:

grep -v "^(\s|\t)*[6-9] " file
8

Possibly there is are spaces at the beginning

Try this sample.

sed 's/^  *//g' b | grep -v "^[6-9] "

( certainly, not an efficient way of doing it )

9

Thanks for the input PWSwebmaster however that grep-v you adjusted still didn't work. The closest I can get is doing the following but it's still doesn't delete all the lines I want.

backups01laxint.liuc(s){jsandova}[~]<0>$ nsrjb -v | awk -F' ' '$3>=01{print $0}' | cut -b 1-79 | grep -v "("
setting verbosity level to `1'
slot volume used pool barcode
1: intwindows.0005 9% intwindows 010159L2
2: intunix.0009 19% intunix 010155L2
4: intrman.0002 2% intrman 010395L2
*not registered in the NetWorker media data base

10

Making slight modifications

$l =~ s/^  *//g;
if (!( $l >= 6 && $l <= 9 )) {
  print FILE $l;
}
11

Try escaping the round brackets.

grep -v "^\(\s|\t\)*[6-9] " file
12

That'll work too, but don't need the global option since anchored matches can only match once. :wink:

13 
 # grep  -v "^ *[6-9]" file
backups01laxint.liuc(s){jsandova}[~]<0>$ nsrjb -v |  awk -F' ' '$3>=01{print $0}'  | cut -b 1-79 | cat -n | grep -v "^[6-9] "
setting verbosity level to `1'
     1    slot  volume                                     used  pool        barcode
     2       1: intwindows.0005                              9%  intwindows  010159L2
     3       2: intunix.0009                                19%  intunix     010155L2
     4       4: intrman.0002                                 2%  intrman     010395L2
     5      24: Cleaning Tape (50 uses left) 000002L1
14

You are right !

$l =~ s/^  *//;
print FILE $1 if (!( $l >= 6 && $l <= 9 ));
15

What about this solution in general:

..... | sed 6,9d 
..... | awk 'NR~"[6-9]"{next}1'

or for the OP request in special:

.... | sed 5q
.... | awk 'NR==6{exit}1'

... and don't make your life complicated :slight_smile:

16

danmero,
That is exactly what I was look for. Simple and short, I however have not used sed much so was not sure on how to implement it. Thanks for the great tip!

backups01laxint.liuc(s){jsandova}[~]<0>$ nsrjb -v | awk -F' ' '$3>=01{print $0}' | cut -b 1-79 | sed 5q
setting verbosity level to `1'
slot volume used pool barcode
1: intwindows.0005 9% intwindows 010159L2
2: intunix.0009 20% intunix 010155L2
4: intrman.0002 2% intrman 010395L2
24: Cleaning Tape (50 uses left) 000002L1 000002L1

17

Oops,
it should be:

awk '!(NR~/^[6-9]$/)'

or:

awk 'NR~/^[6-9]$/{next}1'
18

I like this one :slight_smile: