Could this be considered a Statistics module? In which case a name might
something like Statistics::GroupBy (the SQL clause just occurred to me
while I was thinking it over).
Huh, and now looking at CPAN I see there is an Array::GroupBy module that
has a small overlap with what you are doing. So it seems close, even if
you don't think it belongs in Statistics.
-john
On Thu, November 21, 2013 7:43 am, Philippe Bruhat (BooK) wrote:
> Hi,
>
> I'm having trouble naming a small module.
>
> This module provides a scoring function based on a given ruleset.
> It is used to produce a single status out of an aggregate of statuses.
>
> The rules for deciding what is the result status can be stated in a very
> simple mini-language. The parser reads some source code written in that
> domain-specific language, and returns a code reference. That coderef
> takes a hash reference with statuses as keys and counts as values,
> and returns a single string (the "score" or summary).
>
> It's not an aggregation module. The hashref is actually the aggregation
> from some dataset, for which we want to produce a single, summary
> value. We can then create several rulesets that are applied to various
> aggregations of our Nagios data, to provide more synthetic indicators.
>
> The use case for us is to be able to monitor at a service level, rather
> than a host level. 5% of servers down from a pool might not be critical,
> where 10% would. The module is just one of the components used to provide
> this service level monitoring.
>
> * * *
>
> The module is currently named "Fleur", but that's more of a working title,
> as it's going to make it very difficult to find on CPAN.
>
> Here's he SYNOPSIS of the module:
>
> # create a parser
> my $parser = Fleur->new(
> input => [qw( OK WARN CRIT UNKN )],
> output => [qw( OK WARN CRIT UNKN )], # default: same as input
> default => 'OK', # default: first item in
> output
> );
>
> # generate a scoring function
> my $score = $parser->parse( << 'CODE' );
> %OK > 90%, %CRIT < .1: OK;
> %CRIT > 50%: CRIT;
> %CRIT + %WARN > 25%: WARN;
> UNKN;
> CODE
>
> # OK
> $status = $score->( { OK => 19, CRIT => 2 } );
>
> # WARN
> $status = $score->( { OK => 14, CRIT => 6, WARN => 1 } );
>
> # CRIT
> $status = $score->( { OK => 8, CRIT => 11, WARN => 1, UNKN => 1 } );
>
> # UNKN
> $status = $score->( { OK => 18, CRIT => 2, WARN => 1 } );
>
> The $score subroutine would be compiled from the following source code
> (generated from the lines above):
>
> sub {
> my %count = %{ shift() };
> my $total = 0;
> $total += $_ for values %count;
> my %percent
> = $total
> ? map +( $_ => $count{$_} / $total ), keys %count
> : map +( $_ => 0 ), keys %count;
> return 'OK'
> if $percent{OK} > 0.9
> && $percent{CRIT} < .1;
> return 'CRIT'
> if $percent{CRIT} > 0.5;
> return 'WARN'
> if $percent{CRIT} > 0.25;
> return 'UNKN';
> }
>
> Without the %-prefix, names are translated to absolute values from %count.
>
> Thanks for your time and ideas,
>
> --
> Philippe Bruhat (BooK)
>
> Food and life were both meant to be shared with others.
> (Moral from Groo The Wanderer #119
> (Epic))
>
>
