How can I detect which line I'm processing in Perl?

Question

Got a text file that looks like:

    200.0     250.0     300.0     350.0     400.0  
162:02:10 017:01:56 017:08:18 011:16:22 008:40:18
    580.0     600.0     620.0     640.0     660.0   
004:04:01 001:47:27 007:25:29 017:44:53 003:07:34

Trying to parse out lines 1 & 3 as "values", and lines 2 & 4 as "times".

My code:

if($line =~ /^\d[^:]*\d/){
     my @values = split(/\s/,$line);
        }
elsif($line =~/^\d+:\d+:\d+/){
     my @time = split(/\s/,$line);
}

Problem: Always matches first regex. My understanding of regex #1 is it will match a line that starts with a digit, followed by any value that is not a ':' any number of times, followed by another digit.

Answer 1

Lines 1 and 3 satisfy the following regex:

(?m)^(?:\s*\d+\.\d+\s*)+$

Try this:

open(FILE, 'yourfile.txt') or die("Could not open file!");
foreach $line (<FILE>) {
  if($line =~ /(?m)^(?:\s*\d+\.\d+\s*)+$/) {
    print $line; 
  }
}

Answer 2

Simply change the order of checks:

if($line =~/^\d+:\d+:\d+/){
     ...
        }
elsif($line =~ /^\d[^:]*\d/){
    ... 
}

Answer 3

if($line =~/\^d+:\d+:\d+/){
  my @values = split /\s+/, $line;
}else{
  my @time = split /\s+/, $line;
}

Answer 4

It happens because lines 2 and 4 really contain the first regex's pattern.

Maybe you can simply check, whether a line has a colon sign, and it will be sufficient? Like this:

my @time;
my @values;
if($line =~ /:/){
     @time = split(/\s+/,$line);
}
else{
     @values = split(/\s+/,$line);
}

Answer 5

The reason your first regex matches every time is that it is detecting any string that starts with a digit can have any number of characters (other than :) and then has another digit. This means line 2 will match on the first three characters before the colon.

You may wish to match on the end of the line as well, or do something more simple like just match against the colon.

Answer 6

The other answers have all focused on regexes. But there is another way to tell where you are in a file.

If you are certain that the lines always alternate, and will be in the same order you can use $. to get the line number you are processing.

This only works if values are always on odd lines, and times are always on even lines.

my @times_and_values;
my $values; 
while(  my $line = <DATA> ) {

    if( $. % 2 ) {
        $values = parse_values($line);
    }
    else {
        my $times = parse_times($line);

        push @times_and_values, [$times, $values]
            if defined $values and defined $times;

    }

}

Your parsing functions can then handle validation and decomposition of the lines. Use regexes tailored to each to reject incorrect values and do any parsing. You can either throw a fatal error or warn. The above code will skip time/value pairs where either part of the pair fails to parse.