CODEWITHSUNDEEP
unixunix-timestamp
par181
par181

Reputation: 401

Formatting issue of a file in UNIX

I have some data in the format below:

Pin|add
jtyjg
Kolk|aaa||
Kawr|wht u
Disnce
Djhdb|bbb||

I want to convert this to the format below:

Pin|add jtyjg Kolk|aaa||
Kawr|wht u Disnce Djhdb|bbb||

How can I do this?

Upvotes: 1

Views: 153

Answers (4)

Chris
Chris

Reputation: 342

It's not quite clear what you want. This one-liner should work your example, though:

tr -d '\n' < oldfile | sed 's/||/||\n/g' > newfile

Depending on your system, you may need to use literal newlines for sed substitution like so:

tr -d '\n' < oldfile | sed 's/||/||\<RETURN>/g' > newfile

Upvotes: 5

marklark
marklark

Reputation: 897

I've made the assumption that the original file doesn't have blanks before the end-of-line characters...

This is fairly basic Perl and works for v5.8.9

#!/usr/bin/perl

open( IN, '<', 'text.txt' );    # the input file
open( OUT, '>', 'text2.txt' );  # the output file

while( <IN> ) {
        chomp;          # get rid of the end-of-line characters
        $out .= $_;     # add the current input to the output string
        if ( /\|\|/ ) { # does this contain the output signal characters "||"?
                print( OUT "$out\n" );  # output the built string
                $out = '';              # clear the output string
        }
        else {
                $out = $out . ' ';      # append a space to the end
        }
}
print( OUT $out );                      # output anything left over...

Upvotes: 2

Debaditya
Debaditya

Reputation: 2497

Try this..

Input.txt

Pin|add
jtyjg
Kolk|aaa||
Kawr|wht u
Disnce
Djhdb|bbb||

Code

cat Input.txt | tr '\n' ' ' | sed 's/|| ./||~~/g' | tr '~~' '\n'| sed '/^$/d' > Output.txt

Output.txt

Pin|add jtyjg Kolk|aaa||
awr|wht u Disnce Djhdb|bbb||

Upvotes: 2

Jonathan Leffler
Jonathan Leffler

Reputation: 753625

On the face of it, you want groups of three input lines to be combined into one with spaces in place of the original newlines. Given that the question does not restrict the tool-set, then a Perl solution is moderately apposite:

#!/usr/bin/env perl
use strict;
use warnings;

my($l1, $l2, $l3);
while (defined($l1 = <>) && defined($l2 = <>) && defined($l3 = <>))
{
    chomp($l1, $l2);
    print "$l1 $l2 $l3";
}

If the number of lines in the input is not a multiple of three, the extra lines are omitted. The code does not handle each input file separately; it just combines them all together. For the given input data, the output is:

Pin|add jtyjg Kolk|aaa||
Kawr|wht u Disnce Djhdb|bbb||

This seems to be correct.

Upvotes: 1

Related Questions