attrbagger Cookbook

Get attributes from data bags.

Somewhat fancy.

Requirements

Attrbagger does not depend on any other cookbooks. It works with both
chef-solo and chef-client by using data_bag_item.

Attributes

attrbagger::default

<table>
<tr>
<th>Key</th><th>Type</th><th>Description</th><th>Default</th>
</tr>
<tr>
<td><tt>node['attrbagger']['autoload']</tt></td>
<td><tt>Boolean</tt></td>
<td>
Used when including the default recipe to decide whether or
not to autoload attributes.
</td>
<td><tt>true</tt></td>
</tr>
<tr>
<td><tt>node['attrbagger']['precedence_level']</tt></td>
<td><tt>Boolean</tt></td>
<td>
The default precedence level to use when not specified in each
<tt>Hash</tt> of <tt>node['attrbagger']['configs']</tt>.
</td>
<td><tt>:override</tt></td>
</tr>
<tr>
<td><tt>node['attrbagger']['configs']</tt></td>
<td><tt>Hash</tt></td>
<td>
The <tt>Hash</tt> structure requires unique string keys representing the
targets of each merged "data bag cascade" in a <tt>::</tt>-separated
format to denote nesting, e.g. <tt>first::second</tt>. The value of
each key must be a <tt>Hash</tt> with an optional <tt>precedence_level</tt>
key (default <tt>override</tt>) and a required <tt>bag_cascade</tt> key.
The value of the <tt>bag_cascade</tt> key must be an <tt>Array</tt> containing
strings representing the cascade of attributes that will be
loaded. The only type of resource currently supported is
<tt>data_bag_item</tt>, e.g.
<tt>'bag_cascade' => ['data_bag_item[a::b]', 'data_bag_item[c::d]']</tt>.
</td>
<td><tt>{}</tt></td>
</tr>
</table>

Usage

Attrbagger may be used either as a library or by including the
recipe[attrbagger::default] in one's run_list.

When used as a library, the Attrbagger class may be used via its
.autoload! class method, or by creating a new instance and using the
#load_config instance method. Both require injection of the
run_context, which is available in any recipe as self.run_context.
See the Attrbagger tests for more usage examples.

Inclusion of recipe[attrbagger::default] in the run_list will result
in the node['attrbagger']['configs'] hash being walked and merged into
the node's attribute precedence level specified via each hash's
precedence_level attribute (defaulting to
node['attrbagger']['precedence_level'].

An example role using attrbagger to load application-specific
configuration from a base, custom, and dynamic environment-specific data
bag, as well as loading mail service configuration might look like this:

    name 'attrbagger_example'
    description 'Attrbagger example role'

    default_attributes(
      'deployment_env' => 'demo',
      'attrbagger' => {
        'configs' => {
          'example_app' => {
            'precedence_level' => 'override',
            'bag_cascade' => [
              'data_bag_item[applications::base::example_app]',
              'data_bag_item[applications::example_app]',
              "data_bag_item[applications::config_<%= node['deployment_env'] %>::example_app]"
            ]
          },
          'services::mail' => {
            'precedence_level' => 'default',
            'bag_cascade' => [
              'data_bag_item[services::mail]'
            ]
          }
        }
      },
      'example_app' => {
        'awesome' => true
      },
      'mail' => {
        'host' => 'localhost',
        'port' => 25
      }
    )

    run_list(
      'recipe[attrbagger]',
      # ... other stuff
    )

This would result in the following actions:

• Load data_bag_item('applications', 'base') and merge its example_app hash with node.default['example_app'], then assign the result to node.override['example_app']
• Load data_bag_item('applications', 'example_app') and merge the entire data bag contents (except for builtin attributes like id and chef_type) with node.default['example_app'], then assign the result to node.override['example_app'].
• Load data_bag_item('applications', 'config_demo') and merge its example_app hash with node.default['example_app'], then assign the result to node.override['example_app']
• Load data_bag_item('services', 'mail') and merge the entire data bag contents (except for builtin attributes like id and chef_type) with node.default['services']['mail'], then assign the result to node.default['services']['mail'].

Contributing

See CONTRIBUTING.md

License

See LICENSE.md

Authors

See AUTHORS.md