The application configuration file

Using the configuration file

The serve administrative command needs a proper configuration file to launch an application.

You can directly give the configuration file to the serve command, as:

<NAGARE_HOME>/bin/nagare-admin serve /path/to/the/config.cfg

Or, if the application was registered to the framework (as described in EntryPoints), then you can launch it with:

<NAGARE_HOME>/bin/nagare-admin serve <application>

In this case, the framework reads the configuration file named <application>.cfg in the directory <application>*/conf under <NAGARE_HOME>/lib/python2.5/site-packages.

Note

If an application is launched in debug mode, changes in the configuration file is immediatly reflected. Else, you need to stop then re-launch the application.

Structure

For the boolean parameters, a value of true, yes, on or 1 means True and a value of false, no, off or 0 mean False.

You can use the $here variable which contains the path to the directory where the configuration file is located.

Comments, starting with the # character can be added to a configuration file.

[application] section

Name	Mandatory	Default value	Description
path	Yes	No default value	Reference to the root component factory of the application (see ObjectReferences)
name	Yes	No default value	/name will be the URL of the application
static	No	static directory	Filesystem path to the static contents of the application. By default, it's the static directory under the application installation directory.
always_html	No	yes	If this parameter is false, the framework will send XHTML to the browsers that accept XHTML, else HTML. If this parameter is true, HTML is always generated
debug	No	no	Display the web debug page when an exception occurs. The nagare[debug] extra must be installed.

[database] section

Name	Mandatory	Default value	Description
activated	No	off	If not activated, the framework will not read the following parameters
uri	Yes	No default value	URI or connection string to the database, as described in http://www.sqlalchemy.org/docs/core/engines.html#supported-databases
metadata	Yes	No default value	Reference to the SQLAlchemy metadata object (see ObjectReferences)
populate	No	No default value	Reference to an optional function, called after the table creation to populate them with some initial data (see ObjectReferences)
debug	No	off	Display the generated SQL requests

All other parameters, if present, are passed as keywords to the SQLALchemy create_engine() call (see http://www.sqlalchemy.org/docs/core/engines.html#engine-creation-api)

If an application needs to work with several database, several subsections can be embedded into the main [database] section:

[database]

[[database1]]      # The name of a subsection is irrelevant but must be unique
activated = on
uri = ...
metadata = ...

[[database2]]
activated = on
uri = ...
metadata = ...