You didn't show us what should NOT be matched. As far as we know, m/./ might be all that's needed to match the correct non-empty line. Or maybe only lines that give exactly those numbers and no other possible numbers should be accepted. Or if negative numbers or decimal points or fractions might be encountered. We'll just have to guess something out of thin air.

for (@filedata)
{
    next if not m/\d+\.\.\d+/;
    @spans = m/(\d+\.\.\d+)/g;
    print "@spans\n";
}
[download]