The DB init scripts initialize the database for products being configured, creating schemas and importing initial data required for some Oracle ATG Web Commerce products. Scripts are declared in the <ATG10dir>/module_name/cim/dbinit.xml file.
The dbinit.xml file has three top-level elements: schema, data-import, and server-instance-type.
The schema element identifies the database schema to be modified, and contains the tasks that do the work of running the scripts. These schema IDs are then specified for a particular server-instance-type and datasource. The IDs for server-instance-type and datasource come from the selected product.xml file. There is a special ID all that can be used for both server-instance-type and datasources. The order for processing is:
Server instance = “all”, Datasource = “all”
Server instance = “all”, Datasource = product selection datasource
Server instance = product selection server instance ID, Datasource = “all”
Server instance = product selection server instance ID, Datasource = product selection datasource
Which dbinit.xml files are used is based on the module list for the server-instance-type of the datasource. CIM looks at the full list. For every dbinit.xml found, it proceeds through the four steps to create a list of what to run. For example, if you create a custom module that requires DAS, CIM automatically runs the DAS table creation scripts (as long as they match the specified criteria).
The server-instance-type element identifies which server-instance-type (as defined in product.xml) is associated with this schema, since many of them have different database needs; for example, a reporting server needs certain utility tables that are not otherwise required by the platform.
Within the schema element you can specify SQL scripts and database import scripts to run. To specify a SQL script, use the structure shown in this example:
<create>sql/db_components/${database}/versioned_ddl.sql</create>The path is relative to the module the dbinit.xml file is in. For example, if the file is in DAS/CIM, the path is relative to the DAS module, and using the example above would end up as DAS/sql/database. The database value is passed in based on the user’s previous selections.
The data-import element contains database initialization scripts, and includes the following elements:
repository-path—The repository into which the data is imported.import-file-path—Path to the XML file that contains the data to import.user—Content Administration user for versioned imports.workspace—Content Administration workspace, for versioned imports.comment—Comment is required for versioned imports.requires-addon-id—During product selection, the user must select the specified add-ons, or the data import does not run. For example:<data-import start-up-module="MyModule">
<requires-addon-id id="MyAddOn" />required-datasources—Identifies data sources that must be created for this import script. The syntax forrequired-datasourcesis:<data-import start-up-module="DAS" required-
datasources="nonswitchingCore,switchingCore">When a required datasource is identified, CIM creates a properties file for that datasource, using the values configured in CIM and the property file name specified in the
<ATG10dir>/CIM/config/cim/DatasourcePropertyFileMapper.propertiesfile. The default configuration is:$class=atg.cim.database.dbsetup.DatasourcePropertyFileMapper
datasourcePropertyMap=\
nonswitchingCore=JTDataSource_production.properties,\
switchingCore=JTDataSource_production.properties,\
svcagent=JTDataSource_agent.propertiesAdd more mappings as needed. In this example, a properties file is created for the
nonswitchingCoredatasource, with the nameJTDataSource_production.propertiesthat point to the connection details fornonswitchingCore.
A detailed example follows. Note that data-import element paths are relative to the Oracle ATG Web Commerce installation location, while repository-loader paths are relative to the specified source module (src-module. The repository-loader element’s post-cleanup attribute is used to clean up data that is used during the installation, but which should not be included in a database used by a running Oracle ATG Web Commerce server.
<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE database-initializer SYSTEM "http://www.atg.com/dtds/dbinit_1_0.dtd">
<database-initializer>
<schema id="versionedSchema">
<sql>
<path>
<create>
sql/db_components/${database}/versioned_ddl.sql
</create>
<drop>
sql/uninstall/${database}/drop_versioned_ddl.sql
</drop>
</path>
</sql>
<data-import>
<repository-path>/atg/epub/file/PublishingFileRepository
</repository-path>
<import-file-path>/Publishing/base/install/epub-file-repository
-data.xml</import-file-path>
<user>publishing</user>
<workspace>workspace</workspace>
<comment>initial_data_import</comment>
</data-import>
<repository-loader post-cleanup="true">
<files>
<src-module>DSS</src-module>
<config-path>atg/registry/data/scenarios/DSS</config-path>
<file-pattern>*.sdl</file-pattern>
</files>
<files>
<src-module>DSS</src-module>
<config-path>atg/registry/data/scenarios/recorders</config-path>
<file-pattern>*.sdl</file-pattern>
</files>
<file-mapping>/atg/epub/file/typemappers/ScenarioTypeMapping
</file-mapping>
<folder-mapping>/atg/epub/file/typemappers/FileFolderTypeMapping
</folder-mapping>
</repository-loader>
</schema>
<server-instance-type id="management">
<datasource id="management">
<schema id="versionedSchema"/>
</datasource>
</server-instance-type>
<server-instance-type id="production">
<datasource id="all">
<schema id="nonVersionedSchema"/>
</datasource>
</server-instance-type>
</database-initializer>
