Can someone help me with the correct use of the split function in perl
Here is my input list called @input_lines:
google.com/test
yahoo.com/test
##############
somethingelse.com/test
##############
12345
my(@first_array,@second_array,@rand_no) = split(/^\#+/, @input_lines);
I'll make a guess on what you really mean:
At first, you have probably a text input file input.txt w/following content:
google.com/test
yahoo.com/test
##############
somethingelse.com/test
##############
12345
Now, you are trying to separate records from the file, delimited by 14 '#'s. Therefore, you could just read the file with ############## as the input record separator and be done:
...
my $fn = 'input.txt'; # set the file name
open my $fh, '<', $fn or die $!; # open the file
$/="\n##############\n"; # set the input record separator
my @parts = <$fh>; # read the file record-wise
chomp @parts; # remove the record separator from data
close $fh # close the file
...
The elements of @parts now have the following content:
$parts[0]
google.com/test
yahoo.com/test
$parts[1]
somethingelse.com/test
$parts[2]
12345
If you need to look for #-separators of different size, you might achieve this in a very similar way by slurping the file in one read operation and splitting at the separators afterwards:
...
my $fn = 'input.txt';
open my $fh, '<', $fn or die $!;
undef $/; # remove the input record separator
my @parts = split /\n#+\n/, <$fh>; # read file as a block and split
close $fh;
...
with the same result.
Regards
rbo
$/ to slurp the file.If format of your @input_lines strings is the same, you can similar join all strings and then split it by parts. You should notice that use split /^#+/ is wrong in your case.
my $line = join ',', @input_lines;
my ($first_part, $second_part, $third_part) = split /\#+/, $line;
my @first_array = split ',', $first_part;
my @second_array = split ',', $second_part;
my @third_array = split ',', $third_part;
split operates on strings, not arrays. Also, you cannot assign to several arrays in the same assignment: the list on the right hand side gets flattened, so the first array takes all.
Update: This code works, though:
my (@first, @second, @rand);
for my $array (\@first, \@second, \@rand) {
my $line;
do {
push @$array, $line = shift @input_lines
} until $line =~ /^#+/ or ! @input_lines;
pop @$array if @input_lines; # Remove the separators
}
You can do something like this. There's an array ref each element of $output that represents one of your arrays.
use strict; use warnings;
use Data::Dumper;
my @input_lines = (
'google.com/test',
'yahoo.com/test',
'##############',
'somethingelse.com/test',
'##############',
'12345',
);
my $output = []; # array ref
my $rand_no;
my $i = 0;
foreach my $line (@input_lines) {
if ($line =~ m/^#+$/) {
# if it's the # we move to the next index
$i++;
next;
}
elsif ($line =~ m/^\d+$/) {
# this is the random numer
$rand_no = $line;
} else {
# everything else goes into the current index
push @{ $output->[$i] }, $line;
}
}
print Dumper $output, $rand_no;
Output:
$VAR1 = [
[
'google.com/test',
'yahoo.com/test'
],
[
'somethingelse.com/test'
]
];
$VAR2 = '12345';
Assuming your input lines are in $string (otherwise use join "\n", @input_lines), you can use split like this:
($first, $second, $rand_no) = split /\n#+\n/m, $string;
print "`", $_, "`\n" for (@fields)'
Script:
my @input_lines = <main::DATA>;
my $input_string = join /\n/, @input_lines;
my @split_lines = split(/\s*[#\n\r]+\s*/, $input_string);
print "$_\n" for @split_lines;
__DATA__
google.com/test
yahoo.com/test
##############
somethingelse.com/test
##############
12345
Output:
google.com/test
yahoo.com/test
somethingelse.com/test
12345
See and test the code here.
Script:
use Data::Dumper;
my @input_lines = <main::DATA>;
my $input_string = join /\n/, @input_lines;
my @blocks = split(/\s*#+\s*/, $input_string);
my @matches = ();
push @matches, [ split(/\s*[\n\r]+\s*/, $_) ] for @blocks;
print Dumper(@matches);
__DATA__
google.com/test
yahoo.com/test
##############
somethingelse.com/test
##############
12345
Output:
$VAR1 = [
'google.com/test',
'yahoo.com/test '
];
$VAR2 = [
'somethingelse.com/test '
];
$VAR3 = [
'12345'
];
See and test the code here.
