Gantry::Conf::FAQ - Frequently Asked Questions regarding Gantry::Conf
There are many reasons why we feel Gantry::Conf is helpful both during development and after deployment. The next two entries should hopefully answer this quesiton for you as they outline a few common scenarios programmers and system administrators often face.
Often programmers have a separate development environment from their production environment. By using <shared> blocks and dev instances you can avoid spending any serious time setting up your application in the development environment. Take this configuration example:
<shared dev> dbuser nobody dbpass secret dbconn "dbi:Pg:dbname=dev" </shared> <shared production> dbuser apache dbpass secret2 dbconn "dbi:Pg:dbname=production" </shared> <instance app1> ConfigureVia FlatFile Config::General /etc/apps/app1.conf use production </instance> <instance app1-dev> ConfigureVia FlatFile Config::General /etc/apps/app1.conf use dev </instance>
By separating out our production and dev database information into shared blocks we can essentially switch between our production and dev environments by simply changing the instance we are using. If you were working on a script this would be a simple matter of running:
$ script.pl --instance=app1-dev
instead of:
$ script.pl --instance=app1
Gantry::Conf has several advantages in a production environment. First, it provides a single place for all config information, if you commit to it. Even if you don't commit to it for all apps, it still provides control to the installing admin over how and where conf information is stored. For instance, the admin could put the config information directly into /etc/gantry.conf, or into a separate file in /etc/gantry.d. She could even set up a secure web server where all boxes would go to get their conf.
The short answer is, Gantry::Conf is flexible and production environments benefit from flexibility.
There are many possible ways to do this a few of which are:
If your application accepts arguments on the command line we suggest adding an --instance option to pass in the instance's name.
--instance
In a mod_perl environment you could use a PerlSetVar, possibly named GantryConfInstance, to pull in this value for your application.
GantryConfInstance
Again in a mod_perl environment, another option would be to use ModPerl::ParamBuilder to pass the instance name.
We include this for the sake of completeness, but advise against it. You could always simply hard code your instance information into your application, but this will greatly reduce the flexibility you have.
Place your provider module in the Gantry::Conf::Provider::Method::* namespace. Make sure your public API matches the existing providers which use the same method. For instance the flat file providers all implement a config method which is called as a class method and receives a file name.
If none of the existing provider methods will do, you need to work in Gantry::Conf. In particular, you need to augment the dispatch hash with the name of your provisioning method and a sub name which will handle it. Then you need to implement the method you put in %dispatch.
%dispatch
Gantry(3), Gantry::Conf(3), Gantry::Conf::Tutorial(3)
Frank Wiles <frank@revsys.com>
Copyright (c) 2006, Frank Wiles.
This library is free software; you can redistribute it and/or modify it under the same terms as Perl itself, either Perl version 5.8.6 or, at your option, any later version of Perl 5 you may have available.
To install Gantry, copy and paste the appropriate command in to your terminal.
cpanm
cpanm Gantry
CPAN shell
perl -MCPAN -e shell install Gantry
For more information on module installation, please visit the detailed CPAN module installation guide.