How do I join two files side by side?

[esther]

I want to create a file XYZ from two other files, XY and Z.

XYZ should have three columns of data. The first two columns come from XY and the third column comes from Z.

Originally posted as a Categorized Question.

[davorg]

print while <>;

And you call it as:

./joinfiles x y > xy

Originally posted as a Categorized Answer.

[jdporter]

There's a standard UNIX tool for this called paste.
The Perl Power Tools project includes a Perl implementation of paste.

[blakem]

how about:

% perl -pe1 xy.txt z.txt > xyz.txt

-Blake

[blakem]

Or even the slightly silly:

% perl -perl xy.txt z.txt > xyz.txt

-Blake

[Fletch]

Of course being this is perlmonks, not catmonks . . .

open( OUT, ">xyzfile" ) or die "Can't create xyzfile: $!\n";
foreach( qw( xfile yfile zfile ) ){
  local( *IN );
  open( IN, $_ ) or die "Can't open $_: $!\n";
  print OUT while <IN>;
  close( IN );
}

Originally posted as a Categorized Answer.

[runrig]

One possibility:

use File::Copy;

open FH, "> xyz" or die "Write xyz - $!";
copy( $_, \*FH ) for qw( xy z );
close FH;

Originally posted as a Categorized Answer.

[TomK32]

a much shorter shell version:

cat xy z > xyz

Originally posted as a Categorized Answer.

[Anonymous Monk]

open INF,  "< z.txt"    or die; # open for reading
open OUTF, ">> xyz.txt" or die; # open for appending 
print OUTF <INF>;
close INF;
close OUTF;

Originally posted as a Categorized Answer.

[belg4mit]

lam does this. It can join any number of files "side by side".

[Beatnik]

in your shell (UNIX) : cat x y > xy Similar will work on Windows. type x y > xyaltho I might be wrong there. It's been quite a while since I did THAT :)

[esther]

Hi! Thanks for answering so quickly. The problem is that the outfile should contain three columns. My first file contains two data columns, the second file one column. Unfortunately cat only puts one file after the other and not next to the other. Is there any other way?

[Beatnik]

Well, that's more than just a simple join. I doubt that any of the listed solutions provide that kind of functionality. I suggest you post a question in SOPW and provide some data format skeleton.

Greetz
Beatnik
... Quidquid perl dictum sit, altum viditur.

[Mago]

#!/usr/bin/perl -w
use strict;

my ($string1, $string2);

$/ = undef;

open (FILE1, "< x") || die "Cannot open file : $!";
open (FILE2, "< y") || die "Cannot open file : $!";
open (FILE3, "> z") || die "Cannot open file : $!";

$string1 = <FILE1>;
$string2 = <FILE2>;

print FILE3 join('', $string1, $string2);

close (FILE1);
close (FILE2);
close (FILE3);

[Joost]

That code will in fact only cat the first line of file x and the first line of file y.

I wouldn't even use perl for this, but if you must:

#!/usr/bin/perl -w
use strict;

my ($string1, $string2);

# read files in chunks instead of lines
# remove if you realy want to read textfiles
# line by line
local $/ = \32768; # "block size"

open (FILE1, "< x") || die "Cannot open file : $!";
open (FILE2, "< y") || die "Cannot open file : $!";
open (FILE3, "> z") || die "Cannot open file : $!";

# set binmode, so binary files don't get messed
# up on systems with text/binary mode files
# (like windows or MS-DOS)
binmode (FILE1);
binmode (FILE2);
binmode (FILE3);

print FILE3 <FILE1>,<FILE2> or die "Error concatenating: $!";

close FILE1 or die "Error closing x: $!";
close FILE2 or die "Error closing y: $!";
close FILE3 or die "Error closing z: $!";

update: fixed copy/paste error

[Mago]

Copy and Paste Error !

Correct Code:



#!/usr/bin/perl -w
use strict;

my ($string1, $string2);

$/ = undef;

open (FILE1, "< x") || die "Cannot open file : $!";
open (FILE2, "< y") || die "Cannot open file : $!";
open (FILE3, "> z") || die "Cannot open file : $!";

$string1 = <FILE1>;
$string2 = <FILE2>;

print FILE3 join('', $string1, $string2);

close (FILE1);
close (FILE2);
close (FILE3);

[ysth]

Updated your answer; does it look ok now? Please /msg QandAEditors if you need changes made.