Perhaps you could show us the code that you have at present and a very small sample of data that demonstrates the problem. The code below may be a good starting point (you'll have to fill in bits though):

use strict;
use warnings;

my $data = <<DATA;
this is a good \
nice day but not a good nice \
night
DATA

open my $IN, '<', \$data;

...

while (<$IN>) {
    ...
}

close $IN;
[download]

with the intent being that you fill in the ... bits to match your current code with the code above providing a nice framework for the sample.

    #!/usr/bin/perl -w

    use strict;

    $/ = '';

    while (<DATA>) {

        next unless /\S/;

        chomp;

        s{\\\n}{}g;

        print "Line: ", $_,"\n";
    }

    __DATA__

    this is a good \
    nice day but not a good nice \
    night


    This is a test line

    This is a another test \
    Line




    This is \ a final test line
[download]

    Line: this is a good nice day but not a good nice night
    Line: This is a test line
    Line: This is a another test Line
    Line: This is \ a final test line
[download]

--
John.

while (<>) { # read line
  while (s/\\$/ /) {
    # while the current line ends with \
    # replace it by blank and append a new line.
    chomp; # thanks to ww
    $_.= <>;
  }
[download]

Update: the chomp should do it. Thanks to ww

Close, but the output doesn't join the lines, as OP asked. I think a big part of the problem is rooted in the thought-process behind your first comment, "current line ends with \," which, strictly speaking, is not the case. The lines in question actually end with a "\" and a newline.

This helps (but does NOT solve; caveats below) if added after your second while :

while (s/\n/ /s) {          # remove the newlines
     my $out = $_;
     print $out;
  }
[download]

Tested, more-or-less using GrandFather's "framework" (below). And just BTW or nitpicking, your (Skeeve's) comment "append a new line" would perhaps be clearer or less ambiguous if written as "append the next line from DATA.

The problem (yep, this is the caveat) is that with multiple sets of data (which I infer from the OP's 'ignoring all the lines which are blank or contain just "space".') this still doesn't separate the sets, as does jmcnamara's, below, qv (and ++).

$_ = "[^\\]\n";
[download]