Here's how I'd make sure there are no blank records in the file,
to get an accurate count of the records that have data in them:
(I've been doing a lot with regular expressions and file parsing lately)

local $/;             # Perl's record separator - default is \n unless...
$/ = undef;           # ... we make it undefined!
open(SESAME,$file);   # open the file
$f = (<SESAME>);      # the entire file, including all \n chars, is now sitting in a single scalar!
close(SESAME);
$f = s/\n+/\n/gs;     # regular expression replaces contiguous strings of \n chars with single \n chars.
$f = s/^\n//s;        # regex removes \n at the start of the string, if there is one.
$f = s/\n$//s;        # regex removes \n at the end of the string, if there is one.
@x = split(/\n/,$f);  # split string on the \n char, store split values in array @x
$x = @x;              # the number of elements in @x
print "There are $x records in file $file\n";
exit(0);

Comment on Re: I don't understand why I'm getting an unitialized value warning