What Is a Schema?
A schema is a collection of database objects (as far as this hour is concerned—tables) associated with one particular database username. This username is called the schema owner, or the owner of the related group of objects. You may have one or multiple schemas in a database. Basically, any user who creates an object has just created his or her own schema. A schema can consist of a single table and has no limits to the number of objects that it may contain, unless restricted by a specific database implementation.
Say you have been issued a database username and password by the database administrator. Your username is USER1. Suppose you log on to the database and then create a table called EMPLOYEE_TBL. Your table's actual name is USER1.EMPLOYEE_TBL. The schema name for that table is USER1, which is also the owner of that table. You have just created the first table of a schema.
The good thing about schemas is that when you access a table that you own (in your own schema), you do not have to refer to the schema name. For instance, you could refer to your table as either one of the following:
EMPLOYEE_TBL USER1.EMPLOYEE_TBL
The first option is preferred because it requires fewer keystrokes. If another user were to query one of your tables, the user would have to specify the schema, as follows:
USER1.EMPLOYEE_TBL
In Hour 20, "Creating and Using Views and Synonyms," you learn about the distribution of permissions so that other users can access your tables. You also learn about synonyms, which allow you to give a table another name so you do not have to specify the schema name when accessing a table. Figure 3.1 illustrates two schemas in a relational database.
Figure 3.1. Schemas in a database.
There are, in Figure 3.1, two user accounts in the database that own tables: USER1 and USER2. Each user account has its own schema. Some examples for how the two users can access their own tables and tables owned by the other user follow:
| USER1 accesses own table1: | TABLE1 |
| USER1 accesses own test: | TEST |
| USER1 accesses USER2's table10: | USER2.TABLE10 |
| USER1 accesses USER2's test: | USER2.TEST |
Both users have a table called TEST. Tables can have the same names in a database as long as they belong to different schemas. If you look at it this way, table names are always unique in a database, because the schema owner is actually part of the table name. For instance, USER1.TEST is different than USER2.TEST. If you do not specify a schema with the table name when accessing tables in a database, the database server looks for a table that you own by default. That is, if USER1 tries to access TEST, the database server looks for a USER1-owned table named TEST before it looks for other objects owned by USER1, such as synonyms to tables in another schema. Hour 21, "Working with the System Catalog," helps you fully understand how synonyms work .
Note
Every database server has rules concerning how you can name objects and elements of objects, such as field names. You must check your particular implementation for the exact naming conventions or rules.