Rand Stats



Configuration refinements

Build Status AppVeyor Build Status License


The following piece of code

use Config::DataLang::Refine;

my Config::DataLang::Refine $c .= new(:config-name<myConfig.toml>);

my Hash $hp1 = $c.refine(<options plugin1 test>);
my Hash $hp2 = $c.refine(<options plugin2 deploy>);

With the following config file in myConfig.toml

  key1 = 'val1'

  key2 = 'val2'

  key1 = false
  key2 = 'val3'

  key3 = 'val3'

Will get you the following as if $hp* is set like

$hp1 = { key2 => 'val3'};
$hp2 = { key1 => 'val1', key3 => 'val3'};

A better example might be from the MongoDB project to test several server setups, the config again in TOML format;

  journal = false
  fork = true
  smallfiles = true
  oplogSize = 128
  logappend = true

# Configuration for Server 1
  logpath = './Sandbox/Server1/m.log'
  pidfilepath = './Sandbox/Server1/m.pid'
  dbpath = './Sandbox/Server1/m.data'
  port = 65010

  replSet = 'first_replicate'

  replSet = 'second_replicate'

  auth = true

Now, to get run options to start server 1 one does the following;

my Config::DataLang::Refine $c .= new(:config-name<mongoservers.toml>);
my Array $opts = $c.refine-str( <mongod s1 replicate1>, :C-UNIX-OPTS-T2);

# Output
# --nojournal, --fork, --smallfiles, --oplogSize=128, --logappend,
# --logpath='./Sandbox/Server1/m.log', --pidfilepath='./Sandbox/Server1/m.pid',
# --dbpath='./Sandbox/Server1/m.data', --port=65010, --replSet=first_replicate

Easy to run the server now;

my Proc $proc = shell(('/usr/bin/mongod', |@$opts).join(' '));


The Config::DataLang::Refine class adds facilities to use a configuration file and gather the key value pairs by searching top down a list of keys thereby refining the resulting set of keys. Boolean values are used to add a key without a value when True or to cancel a previously found key out when False.


Look for documentation and other information at