To remove lines using perl

[Calvin Tan Tjin Wei]

Hi all,

I am trying to remove line from a while when i have search for a particular sting.
i.e.

&123&123&12314215&&abc123
&123&123&12314215&hour&abc112
&123&123&12314215&&abc115
&123&123&12314215&hour&abc113
&123&123&12314215&&abc115
&123&123&12314215&&abc115

i would want to remove the lines which contains the whole string which contains hour and the results would be

&123&123&12314215&&abc123
&123&123&12314215&&abc115
&123&123&12314215&&abc115
&123&123&12314215&&abc115

[James R. Ferguson]

Hi Calvin:

# perl -ne 'print unless /&hour/' file

...or in a 'while' loop:

# perl -e 'while (<>) {print unless /&hour/}' file

Regards!

...JRF...

[H.Merijn Brand (procura]

though the suggested

$ perl -ne'/&hour/ or print' file.dta

is fine, it is much easier with grep

$ fgrep -v '&hour' file.dta

There is a right tool for every problem. In many an occasion that would be perl, but imho, here it is not

Enjoy, Have FUN! H.Merijn

[Doug O'Leary]

Hey;

Both previous posts are correct. I would probably use perl if I wanted to change original file or a 'grep -v' if I was piping the output to something else.

perl -i -nle 'print unless (/&hour/)' ${file}

would edit the file in place.

grep -v '&hour' | while IFS=& read ...

if you were going to do something with the output...

That's the nice thing about UNIX; there's usually several 'right' answers - just depends on what you want to do.

Doug O'Leary

[James R. Ferguson]

HI (again):

It occurs to me (since you problem description is so nebulous and abstract) that perhaps you want to perform some processing in the loop but skip over some records:

...
while (<>) {
...
next if /&hour/; # resume reading next record, if any
...
}

Regards!

...JRF...