During a complete installation of Oracle Database, Sample Schemas can be installed automatically with the seed database. If the seed database is removed from your system, then you will need to reinstall Sample Schemas before you can duplicate the examples you find in Oracle documentation and training materials.
This chapter describes how to install Sample Schemas. It contains the following sections:
|
Caution: By installing any of the Oracle Database Sample Schemas, you will destroy any previously installed schemas that use any of the following user names:
Data contained in any of these schemas will be lost if you run any of the installation scripts described in this section. You should not use Oracle Database Sample Schemas for your personal or business data and applications. They are meant to be used for demonstration purposes only. |
When you install Oracle with the Oracle Universal Installer, the sample schemas are installed by default when you select the Basic Installation option. If you have installed your database with the Advanced Option and did not choose to install the sample schemas, you can install the Sample Schemas with the Database Configuration Assistant (DBCA). When you run DBCA, selecting the Sample Schemas option installs all five schemas (HR, OE, PM, IX, SH) in your database.
At the end of the installation process, a window displays the accounts that have been created and their lock status. All Sample Schemas are locked by default. You can unlock the accounts at this point in the installation process. Alternatively, after the installation is complete, you can unlock the schemas with an ALTER USER ... ACCOUNT UNLOCK statement.
The sample schemas available to you depend on the edition of Oracle that you have installed and its configuration. Refer to the following table:
| Schema | Oracle Database Personal edition | Oracle Database Standard edition | Oracle Database Enterprise edition |
|---|---|---|---|
| HR | OK | OK | OK |
| OE | OK | OK | OK |
| PM | OK | OK | OK |
| IX | OK | OK | OK |
| SH | Not available | Not available | Needs Partitioning Option installed |
You can also create Sample Schemas manually by running SQL scripts, rather than using DBCA. The scripts are included in the companion directory on the installation medium.
Various dependencies have been established among the schemas. Therefore, when you create the schemas manually, you must create them in the following order: HR, OE, PM, IX, and SH.
Use this sequence to create the schemas:
Create the HR schema.
Create the OE schema: The HR schema is already present, and you must know the password for the HR schema to grant HR object privileges to OE. Some HR tables are visible to the OE user by using private synonyms. In addition, some OE tables have foreign key relationships to HR tables.
|
Note: TheOE schema requires the database to be enabled for spatial data. You can accomplish this during installation or later using the Database Configuration Assistant. |
Create the PM schema: Foreign key relationships require that the OE schema already exist when the PM schema is created. You need to know the password for OE to grant to PM the right to establish and use these foreign keys.
|
Note: ThePM schema requires the database to be enabled for the Java Virtual Machine (JVM) and interMedia. You can accomplish this during installation or later using the Database Configuration Assistant. |
Create the IX schema: The information exchange schema IX is based on order entry data in OE. Again, foreign key relationships require that the OE schema already be present when the IX schema is created. You need to know the password for OE to grant to IX the right to establish and use the foreign keys.
Create the SH schema. The SH schema logically depends on the OE schema, though you can create this schema without creating the other four schemas.
All scripts necessary to create Human Resource (HR) schema reside in $ORACLE_HOME/demo/schema/human_resources.
You need to call only one script, hr_main.sql, to create all the objects and load the data. Running hr_main.sql accomplishes the following tasks:
Prompts for passwords and tablespace names used within the scripts
Removes any previously installed HR schema
Creates the user HR and grants the necessary privileges
Connects as HR
Calls the scripts that create and populate the schema objects
For a complete listing of the scripts and their functions, refer to Table 4-1.
A pair of optional scripts, hr_dn_c.sql and hr_dn_d.sql, is provided as a schema extension. To prepare the HR schema for use with the directory capabilities of Oracle Internet Directory, run the hr_dn_c.sql script. If you want to return to the initial setup of the HR schema, then use the hr_dn_d.sql script to undo the effects of the hr_dn_c.sql script.
The hr_drop.sql script is used to drop the HR schema.
All scripts necessary to create the Order Entry (OE) schema and its Online Catalog (OC) subschema reside in $ORACLE_HOME/demo/schema/order_entry.
You need to call only one script, oe_main.sql, to create all the objects and load the data. Running oe_main.sql accomplishes the following tasks:
Prompts for passwords and tablespace names used within the scripts
Removes any previously installed OE schema
Creates the user OE and grants the necessary privileges
Connects as OE
Calls the scripts that create and populate the schema objects
For a complete listing of the scripts and their functions, refer to Table 4-2, respectively.
The oe_drop.sql and oc_drop.sql scripts are used to drop the OE schema and OC subschema.
All files necessary to create Product Media (PM) schema reside in $ORACLE_HOME/demo/schema/product_media.
You need to call only one script, pm_main.sql, to create all the objects and load the data. Running pm_main.sql accomplishes the following tasks:
Prompts for passwords and tablespace names used within the scripts, as well as datafile and log file directories
Removes any previously installed PM schema
Creates the user PM and grants the necessary privileges
Connects as PM
Calls the following scripts that create and populate the schema objects
For a complete listing of the scripts and their functions, refer to Table 4-3 .
The pm_drop.sql script is used to drop the PM schema.
|
Note: The SQL*Loader data filepm_p_lob.dat contains hard-coded absolute path names that have been set during installation. Before attempting to load the data in a different environment, you should first edit the path names in this file. |
To install the Information Exchange (IX) schema, you need to call only one script, ix_main.sql, to create all the objects and load the data. Running ix_main.sql accomplishes the following tasks:
Prompts for passwords and tablespace names used within the scripts
Removes any previously installed IX schema
Creates the user IX and grants the necessary privileges
Connects as IX
Calls the scripts that create and populate the schema objects
For a complete listing of the scripts and their functions, refer to Table 4-4.
The ix_drop.sql script is used for dropping the IX schema.
All files necessary to create the Sales History (SH) schema reside in $ORACLE_HOME/demo/schema/sales_history.
You need to call only one script, sh_main.sql, to create all the objects and load the data. Running sh_main.sql accomplishes the following tasks:
Prompts for passwords and tablespace names used within the scripts, as well as datafile and log file directories
Removes any previously installed SH schema
Creates the user SH and grants the necessary privileges
Connects as SH
Calls the scripts that create and populate the schema objects
For a complete listing of the scripts and their functions, refer to Table 4-5.
|
Note: The dimension tablesPROMOTIONS, CUSTOMERS, PRODUCTS and the fact table SALES are loaded by SQL*Loader, after which directory paths are created inside the database to point to the load and log file locations. This allows the loading of the COSTS table by using the external table sales_transactions_ext. |
A pair of optional scripts, sh_olp_c.sql and sh_olp_d.sql, is provided as a schema extension. To prepare the SH schema for use with the advanced analytical capabilities of OLAP Services, run the sh_olp_c.sql create script. If you want to return to the initial setup of the SH schema, then use the script sh_olp_d.sql to erase the effects of sh_olp_c.sql and reinstate dimensions as they were before.
The file used to drop the SH schema is sh_drop.sql.
To reset Sample Schemas to their initial state, use the following syntax from the SQL*Plus command-line interface:
@?/demo/schema/mksample systempwd syspwd hrpwd oepwd pmpwd ixpwd shpwd bipwd default_tablespace temp_tablespace log_file_directory/
The mksample script expects 11 parameters. Provide the password for SYSTEM and SYS, and for the HR, OE, PM, IX, SH, and BI schemas. Specify a temporary and a default tablespace, and make sure to end the name of the log file directory with a trailing slash.
The mksample script produces several log files:
mkverify.log is the Sample Schema creation log file.
hr_main.log is the HR schema creation log file.
oe_oc_main.log is the OE schema creation log file.
pm_main.log is the PM schema creation log file.
pm_p_lob.log is the SQL*Loader log file for PM.PRINT_MEDIA.
ix_main.log is the IX schema creation log file.
sh_main.log is the SH schema creation log file.
cust.log is the SQL*Loader log file for SH.CUSTOMERS.
prod.log is the SQL*Loader log file for SH.PRODUCTS.
promo.log is the SQL*Loader log file for SH.PROMOTIONS.
sales.log is the SQL*Loader log file for SH.SALES.
sales_ext.log is the external table log file for SH.COSTS.
In most situations, there is no difference between installing a Sample Schema for the first time or reinstalling it over a previously installed version. The *_main.sql scripts drop the schema users and all their objects.
In some cases, complex interobject relationships in the OE or IX schemas prevent the DROP USER ... CASCADE operations from completing normally. To correct these rare cases, use one of the following procedures:
For the OC catalog subschema of the OE schema:
Connect as the user OE.
Run the oc_drop.sql. script
Connect as SYSTEM.
Ensure that no user is connected as OE:
SELECT username FROM v$session;
Drop the user:
DROP USER oe CASCADE;
For the IX schemas:
Connect as SYSTEM.
Ensure that no user is connected as an IX user:
SELECT username FROM v$session WHERE username like 'IX%';
Drop the schemas by running the dix.sql. script. You will be prompted for passwords for the individual users.