Library

Product

Contents

Index

Accessing SQL Databases

Connecting to the Default Database
Connecting to a Remote Database
Copying Data from One Database to Another
Copying Data between Tables on One Database

This chapter explains how to access databases through SQL*Plus, and discusses the following topics:

connecting to the default database

connecting to a remote database

copying data from one database to another

copying data between tables on one database

Read this chapter while sitting at your computer and try out the example shown. Before beginning, make sure you have access to the sample tables described

Connecting to the Default Database

In order to access data in a given database, you must first connect to the database. When you start SQL*Plus, you normally connect to your default Oracle database under the username and password you enter while starting. Once you have logged in, you can connect under a different username with the SQL*Plus CONNECT command. The username and password must be valid for the database.

For example, to connect the username TODD to the default database using the password FOX, you could enter

SQL> CONNECT TODD/FOX

If you omit the username and password, SQL*Plus prompts you for them. You also have the option of typing only the username following CONNECT and omitting the password (SQL*Plus then prompts for the password). Because CONNECT first disconnects you from your current database, you will be left unconnected to any database if you use an invalid username and password in your CONNECT command.

If you log on or connect as a user whose account has expired, SQL*Plus prompts you to change your password before you can connect.

If an account is locked, a message is displayed and connection as this user is not permitted until the account is unlocked by your DBA.

You can disconnect the username currently connected to Oracle without leaving SQL*Plus by entering the SQL*Plus command DISCONNECT at the SQL*Plus command prompt.

The default database is configured at an operating system level by setting operating system environment variables, symbols or, possibly, by editing an Oracle specific configuration file. Refer to your Oracle documentation for your operating system for more information.

Connecting to a Remote Database

Many large installations run Oracle on more than one computer. Such computers are often connected in a network, which permits programs on different computers to exchange data rapidly and efficiently. Networked computers can be physically near each other, or can be separated by large distances and connected by telecommunication links.

Databases on other computers or databases on your host computer other than your default database are called remote databases. You can access remote databases if the desired database has SQL*Net and both databases have compatible network drivers.

You can connect to a remote database in one of two ways:

from within SQL*Plus, using the CONNECT command

as you start SQL*Plus, using the SQLPLUS command

**Connecting to a Remote Database from within SQL*Plus**

To connect to a remote database using CONNECT , include a SQL*Net database specification in the CONNECT command in one of the following forms (the username and password you enter must be valid for the database to which you wish to connect):

CONNECT SCOTT@database_specification

CONNECT SCOTT/TIGER@database_specification

SQL*Plus prompts you for a password as needed, and connects you to the specified database. This database becomes the default database until you CONNECT again to another database, DISCONNECT, or leave SQL*Plus.

If you log on or connect as a user whose account has expired, SQL*Plus prompts you to change your password before you can connect.

If an account is locked, a message is displayed and connection as this user is not permitted until the account is unlocked by your DBA.

When you connect to a remote database in this manner, you can use the complete range of SQL and SQL*Plus commands and PL/SQL blocks on the database.

The exact string you enter for the database specification depends upon the SQL*Net protocol your computer uses. For more information, see CONNECT and the SQL*Net manual appropriate for your protocol, or contact your DBA.

**Connecting to a Remote Database as You Start SQL*Plus**

To connect to a remote database when you start SQL*Plus, include the SQL*Net database specification in your SQLPLUS command in one of the following forms:

SQLPLUS SCOTT@database_specification

SQLPLUS SCOTT/TIGER@database_specification

You must use a username and password valid for the remote database and substitute the appropriate database specification for the remote database. SQL*Plus prompts you for username and password as needed, starts SQL*Plus, and connects you to the specified database. This database becomes the default database until you CONNECT to another database, DISCONNECT, or leave SQL*Plus.

If you log on or connect as a user whose account has expired, SQL*Plus prompts you to change your password before you can connect.

If an account is locked, a message is displayed and connection as this user is not permitted until the account is unlocked by your DBA.

Once again, you can manipulate tables in the remote database directly after you connect in this manner.

Note: Do not confuse the @ symbol of the connect string with the @ command used to run a command file.

Copying Data from One Database to Another

Use the SQL*Plus COPY command to copy data between databases and between tables on the same database. With the COPY command, you can copy data between databases in the following ways:

copy data from a remote database to your local database

copy data from your local (default) database to a remote database (on most systems)

copy data from one remote database to another remote database (on most systems)

Note: In general, the COPY command was designed to be used for copying data between Oracle and non-Oracle databases. You should use SQL commands (CREATE TABLE AS and INSERT) to copy data between Oracle databases.

Understanding COPY Command Syntax

You enter the COPY command in the following form:

COPY FROM database TO database action -
  destination_table (column_name, column_name, -
  column_name ...) USING query

Here is a sample COPY command:

COPY FROM SCOTT/TIGER@BOSTONDB -
  TO TODD/FOX@CHICAGODB -
  CREATE NEWDEPT (DNUMBER, DNAME, CITY)-
  USING SELECT * FROM DEPT

To specify a database in the FROM or TO clause, you must have a valid username and password for the local and remote database(s) and know the appropriate database specification(s). COPY obeys Oracle security, so the username you specify must have been granted access to tables for you to have access to tables. For information on what databases are available to you, contact your DBA.

When you copy to your local database from a remote database, you can omit the TO clause. When you copy to a remote database from your local database, you can omit the FROM clause. When you copy between remote databases, you must include both clauses.

The COPY command behaves differently based on whether the destination table already exists and on the action clause you enter (CREATE in the example above). See "Controlling Treatment of the Destination Table" later in this chapter.

By default, the copied columns have the same names in the destination table that they have in the source table. If you want to give new names to the columns in the destination table, enter the new names in parentheses after the destination table name. If you enter any column names, you must enter a name for every column you are copying.

Note: To enable the copying of data between Oracle and non-Oracle databases, NUMBER columns are changed to DECIMAL columns in the destination table. Hence, if you are copying between Oracle databases, a NUMBER column with no precision will be changed to a DECIMAL(38) column. When copying between Oracle databases, you should use SQL commands (CREATE TABLE AS and INSERT) or you should ensure that your columns have a precision specified.

The USING clause specifies a query that names the source table and specifies the data that COPY copies to the destination table. You can use any form of the SQL SELECT command to select the data that the COPY command copies.

Here is an example of a COPY command that copies only two columns from the source table, and copies only those rows in which the value of DEPTNO is 30:

SQL> COPY FROM SCOTT/TIGER@BOSTONDB -
>  REPLACE EMPCOPY2 -
>  USING SELECT ENAME, SAL -
>  FROM EMPCOPY -
>  WHERE DEPTNO = 30

You may find it easier to enter and edit long COPY commands in command files rather than trying to enter them directly at the command prompt.

Controlling Treatment of the Destination Table

You control the treatment of the destination table by entering one of four control clauses--REPLACE, CREATE, INSERT, or APPEND.

The REPLACE clause names the table to be created in the destination database and specifies the following actions:

If the destination table already exists, COPY drops the existing table and replaces it with a table containing the copied data.

If the destination table does not already exist, COPY creates it using the copied data.

You can use the CREATE clause to avoid accidentally writing over an existing table. CREATE specifies the following actions:

If the destination table already exists, COPY reports an error and stops.

If the destination table does not already exist, COPY creates the table using the copied data.

Use INSERT to insert data into an existing table. INSERT specifies the following actions:

If the destination table already exists, COPY inserts the copied data in the destination table.

If the destination table does not already exist, COPY reports an error and stops.

Use APPEND when you want to insert data in an existing table, or create a new table if the destination table does not exist. APPEND specifies the following actions:

If the destination table already exists, COPY inserts the copied data in the destination table.

If the table does not already exist, COPY creates the table and then inserts the copied data in it.

Example 5-1 Copying from a Remote Database to Your Local Database Using CREATE

To copy EMP from a remote database into a table called EMPCOPY on your own database, enter the following command:

Note: See your DBA for an appropriate username , password , and database specification for a remote computer that contains a copy of EMP.

SQL> COPY FROM SCOTT/TIGER@BOSTONDB -
>  CREATE EMPCOPY -
>  USING SELECT * FROM EMP

SQL*Plus displays the following messages:

Array fetch/bind size is 20. (arraysize is 20)
Will commit when done. (copycommit is 0)
Maximum long size is 80. (long is 80)

SQL*Plus then creates the table EMPCOPY, copies the rows, and displays the following additional messages:

Table EMPCOPY created.
 14 rows selected from SCOTT@BOSTONDB.
 14 rows inserted into EMPCOPY.
 14 rows committed into EMPCOPY at DEFAULT HOST connection.

In this COPY command, the FROM clause directs COPY to connect you to the database with the specification D:BOSTON-MFG as SCOTT, with the password TIGER.

Notice that you do not need a semicolon at the end of the command; COPY is a SQL*Plus command, not a SQL command, even though it contains a query. Since most COPY commands are longer than one line, you must use a hyphen (-), optionally preceded by a space, at the end of each line except the last.

Interpreting the Messages that COPY Displays

The first three messages displayed by COPY show the values of SET command variables that affect the COPY operation. The most important one is LONG , which limits the length of a LONG column's value. (LONG is a datatype, similar to CHAR.) If the source table contains a LONG column, COPY truncates values in that column to the length specified by the system variable LONG.

The variable ARRAYSIZE limits the number of rows that SQL*Plus fetches from the database at one time. This number of rows makes up a batch. The variable COPYCOMMIT sets the number of batches after which COPY commits changes to the database. (If you set COPYCOMMIT to zero, COPY commits changes only after all batches are copied.) For more information on the variables of the SET command, including how to change their settings, see the SET command.

After listing the three system variables and their values, COPY tells you if a table was dropped, created, or updated during the copy. Then COPY lists the number of rows selected, inserted, and committed.

Specifying Another User's Table

You can refer to another user's table in a COPY command by qualifying the table name with the username, just as you would in your local database, or in a query with a database link.

For example, to make a local copy of a table named DEPT, owned by the username ADAMS on the database associated with the SQL*Net connect string BOSTONDB, you would enter

SQL> COPY FROM SCOTT/TIGER@BOSTONDB -
>  CREATE EMPCOPY2 -
>  USING SELECT * FROM ADAMS.DEPT

Of course, you could get the same result by instructing COPY to log in to the remote database as ADAMS. You cannot do that, however, unless you know the password associated with the username ADAMS.

Copying Data between Tables on One Database

You can copy data from one table to another in a single database (local or remote). To copy between tables in your local database, specify your own username and password and the database specification for your local database in either a FROM or a TO clause (omit the other clause):

SQL> COPY FROM SCOTT/TIGER@MYDATABASE -
>  INSERT EMPCOPY2 -
>  USING SELECT * FROM EMP

To copy between tables on a remote database, include the same username, password, and database specification in the FROM and TO clauses:

SQL> COPY FROM SCOTT/TIGER@BOSTONDB -
>  TO SCOTT/TIGER@BOSTONDB -
>  INSERT EMPCOPY2 -
>  USING SELECT * FROM EMP

Prev Next

Library

Product

Contents

Index