Help with my regex

cspctec has asked for the wisdom of the Perl Monks concerning the following question:

I have the following data:

Controller Status:          Optimal
NCQ status:                 Enabled
Status of logical device:   Optimal

Power State:                RPM
Supported Power State:      RPM, Powered off
[download]

And I need a regex that will match "Controller Status", "Status of logical device", and "Power State", but not "NCQ Status" or "Supported Power State".

I have my regex set up like this:

my $match = 'status|power';
foreach (@data) {
  if (/$match/i) {
    print $_;
  }
}
[download]

That is matching everything. I tried something like this as well:

my $match = '[^ncq]status|[^supported]power';
[download]

I thought the ^ncq meant do not match.

Comment on Help with my regex Select or Download Code

Replies are listed 'Best First'.
Re: Help with my regex by toolic (Bishop) on Jul 31, 2013 at 20:44 UTC
I thought the ^ncq meant do not match. No, `[^ncq]` is a character class (see perlretut). Tip #9 from the Basic debugging checklist (YAPE::Regex::Explain) `[^ncq] any character except: 'n', 'c', 'q'` [download]	[reply] [d/l] [select]
Re: Help with my regex by SuicideJunkie (Vicar) on Jul 31, 2013 at 20:54 UTC
Every line there except for the blank one, does indeed have either "power" or "status" somewhere in it. If your set is as small as the example, you might be best off listing all of the options in a long enough form to make them unique. This specifies exactly what should be found before the first colon on the line. `# Maybe make the values here a sub reference defining what to call whe +n you see the line in question. my %wantedInfoHash = ('Controller Status' => undef, 'Status of logical device' => undef, ...); while ($line = <>) { if ($line =~ /([^:]):/) { # grabbed everything up to the first colon into $1 if (exists $wantedInfoHash{$1}) { # Found something we wanted, so do something with it. } } }` [download]	[reply] [d/l]
Re^2: Help with my regex by cspctec (Sexton) on Jul 31, 2013 at 21:13 UTC
I'm not sure how to get "Power State" only without also grabbing "Supported Power State" since they both have the same string in them. I just need the shorter one.	[reply]
Re^3: Help with my regex by Preceptor (Deacon) on Jul 31, 2013 at 21:25 UTC
`m/^Power State:/` [download] The ^ is an anchor to 'beginning of line'.	[reply] [d/l]
Re^3: Help with my regex by SuicideJunkie (Vicar) on Jul 31, 2013 at 21:36 UTC
The match in that code above grabs everything up to the first colon. Which means it will get "Supported Power State" on that line, which will then not exist in the hash of things you want. Later it will get "Power State" on the shorter line, and that will exist in the hash of things you want.	[reply]
Re: Help with my regex by Laurent_R (Canon) on Jul 31, 2013 at 21:51 UTC
You could also exclude lines that you don't want: `my $match = 'status\|power'; foreach (@data) { next if /NCQ/ or /Supported/; print $_ if /$match/i; }` [download] Or even simpler: `foreach (@data) { print unless /NCQ/ or /Supported/ or /^$/; }` [download]	[reply] [d/l] [select]
Re: Help with my regex by moritz (Cardinal) on Aug 01, 2013 at 05:38 UTC
How about just parsing the whole thing, and afterward using those fields you're interested in, and simply ignore the rest? `use strict; use warnings; use 5.010; my %states; while (<DATA>) { chomp; if (/^ ([^:]+) : \s+ (.*)/x) { $states{$1} = $2; } } use Data::Dumper; print Dumper \%states; __DATA__ Controller Status: Optimal NCQ status: Enabled Status of logical device: Optimal Power State: RPM Supported Power State: RPM, Powered off` [download] Prints `$VAR1 = { 'Supported Power State' => 'RPM, Powered off', 'Power State' => 'RPM', 'NCQ status' => 'Enabled', 'Status of logical device' => 'Optimal', 'Controller Status' => 'Optimal' };` [download] So all the interesting (and some of the uninteresting) data is in `%states`. Perl 6 - the future is here, just unevenly distributed	[reply] [d/l] [select]