Entering and Executing Commands

Entering Commands

Your computer's cursor, or pointer (typically an underline, a rectangular block, or a slash), appears after the command prompt. The cursor indicates the place where the next character you type will appear on your screen.

To tell SQL*Plus what to do, simply type the command you wish to use. Usually, you separate the words in a command from each other by a space or tab. You can use additional spaces or tabs between words, if you wish, to make your commands more readable.

Note: You will see examples of spacing and indentation throughout this guide. When you enter the commands in the exercises, you do not have to space them as shown, but you may find them clearer to read if you do.

Case sensitivity is operating system specific. For the sake of clarity, all table names, column names, and commands in this guide appear in capital letters.

You can enter three kinds of commands at the command prompt:

• SQL commands, for working with information in the database

• PL/SQL blocks, also for working with information in the database

• SQL*Plus commands, for formatting query results, setting options, and editing and storing SQL commands and PL/SQL blocks

The manner in which you continue a command on additional lines, end a command, or execute a command differs depending on the type of command you wish to enter and run. Examples of how to run and execute these types of commands are found on the following pages.

The SQL Buffer

The area where SQL*Plus stores your most recently entered SQL command or PL/SQL block is called the SQL buffer. The command or block remains there until you enter another. If you want to edit or re-run the current SQL command or PL/SQL block, you may do so without re-entering it. For more information about editing or re-running a command or block stored in the buffer see the section "Running Command Files" in Chapter 3.

SQL*Plus does not store the semicolon or the slash you type to execute a command in the SQL buffer.

Note: SQL*Plus commands are not stored in the SQL buffer.

Getting Help

The database administrator creates the SQL*Plus help tables and populates them with SQL*Plus help data. Before you can install SQL*Plus help, ensure that:

• SQL*Plus is installed, otherwise, you cannot create and load the help tables.

• The default tablespace for the SYSTEM user is large enough to accommodate the help system. You must have at least 128K of free space.

• The SQL*Plus help script files are available in

  $ORACLE_HOME/SQLPLUS/ADMIN/HELP/ 
  

The help script files are:

• HLPBLD.SQL - to drop and create new help tables.

• HELPUS.SQL - to populate the help tables with the help data.

• HELPDROP.SQL - to drop existing SQL*Plus help tables.

To install SQL*Plus help:

1. Run SQL*Plus as the SYSTEM user with:

   
   SQLPLUS SYSTEM/PASSWORD

   where PASSWORD is the password you have defined for the SYSTEM user (MANAGER by default).

2. Run the SQL script, HLPBLD.SQL, from SQL*Plus with:

   
   @$ORACLE_HOME/SQLPLUS/ADMIN/HELP/HLPBLD.SQL HELPUS.SQL

To get online help for SQL*Plus commands, type HELP at the command prompt followed by the name of the command, for example:

HELP ACCEPT

If you get a response indicating that help is not available, consult your database administrator. For more details about the help system, see "Getting Help" in Chapter 7, and the HELP command in Chapter 8.

Executing Commands

After you enter the command and direct SQL*Plus to execute it, SQL*Plus processes the command and re-displays the command prompt, indicating that you can enter another command.

Running SQL Commands

The SQL command language enables you to manipulate data in the database. See your Oracle9i SQL Reference for information on individual SQL commands.

Example 2-3 Entering a SQL Command

In this example, you will enter and execute a SQL command to display the employee number, name, job, and salary of each employee in the EMP_DETAILS_VIEW view.

1. At the command prompt, enter the first line of the command:

   
   SELECT EMPLOYEE_ID, LAST_NAME, JOB_ID, SALARY

   If you make a mistake, use Backspace to erase it and re-enter. When you are done, press Return to move to the next line.

2. SQL*Plus will display a "2", the prompt for the second line. Enter the second line of the command:

   
   FROM EMP_DETAILS_VIEW WHERE SALARY > 12000;

   The semicolon (;) means that this is the end of the command. Press Return. SQL*Plus processes the command and displays the results on the screen:

   
   EMPLOYEE_ID LAST_NAME                 JOB_ID             SALARY
   ----------- ------------------------- ---------- --------------
           100 King                      AD_PRES           $24,000
           101 Kochhar                   AD_VP             $17,000
           102 De Haan                   AD_VP             $17,000
           145 Russell                   SA_MAN            $14,000
           146 Partners                  SA_MAN            $13,500
           201 Hartstein                 MK_MAN            $13,000
   
   6 rows selected.

   After displaying the results and the number of rows retrieved, SQL*Plus displays the command prompt again. If you made a mistake and therefore did not get the results shown above, simply re-enter the command.

   The headings may be repeated in your output, depending on the setting of a system variable called PAGESIZE. Sometimes, the result from a query will not fit the available page width. You will need to adjust a system variable called LINESIZE, which sets the width of the output in characters, see "Setting Page Dimensions". Typically, in the examples in this guide this is set to 70 characters. You may need to SET LINESIZE to 70 so the query output appears the same as in this guide. Whether you see the message concerning the number of records retrieved depends on the setting of a system variable called FEEDBACK. You will learn more about system variables later in this chapter in the section "Variables that Affect Running Commands". To save space, the number of records selected will not be shown in the rest of the examples in this guide.

Understanding SQL Command Syntax

Just as spoken language has syntax rules that govern the way we assemble words into sentences, SQL*Plus has syntax rules that govern how you assemble words into commands. You must follow these rules if you want SQL*Plus to accept and execute your commands.

Dividing a SQL Command into Separate Lines

You can divide your SQL command into separate lines at any points you wish, as long as individual words are not split between lines. Thus, you can enter the query you entered in Example 2-3 on three lines:

SELECT EMPLOYEE_ID, LAST_NAME, JOB_ID
FROM EMP_DETAILS_VIEW
WHERE SALARY>12000;

In this guide, you will find most SQL commands divided into clauses, one clause on each line. In Example 2-3, for instance, the SELECT and FROM clauses were placed on separate lines. Many people find this clearly visible structure helpful, but you may choose whatever line division makes commands most readable to you.

Ending a SQL Command

You can end a SQL command in one of three ways:

• with a semicolon (;)

• with a slash (/) on a line by itself

• with a blank line

A semicolon (;) tells SQL*Plus that you want to run the command. Type the semicolon at the end of the last line of the command, as shown in Example 2-3, and press Return. SQL*Plus will process the command and store it in the SQL buffer (see the section "The SQL Buffer" for details). If you mistakenly press Return before typing the semicolon, SQL*Plus prompts you with a line number for the next line of your command. Type the semicolon and press Return again to run the command.

Note: You cannot enter a comment on the same line after a semicolon. For more information about placing comments, see "Placing Comments in Command Files" in Chapter 3.

A slash (/) on a line by itself also tells SQL*Plus that you wish to run the command. Press Return at the end of the last line of the command. SQL*Plus prompts you with another line number. Type a slash and press Return again. SQL*Plus executes the command and stores it in the buffer (see the section "The SQL Buffer" for details).

A blank line in a SQL statement or script tells SQL*Plus that you have finished entering the command, but do not want to run it yet. Press Return at the end of the last line of the command. SQL*Plus prompts you with another line number.

Note: You can change the way blank lines appear and behave in SQL statements using the SET SQLBLANKLINES command. For more information about changing blank line behavior, see the SET command in Chapter 8.

Press Return again; SQL*Plus now prompts you with the SQL*Plus command prompt. SQL*Plus does not execute the command, but stores it in the SQL buffer (see the section "The SQL Buffer" for details). If you subsequently enter another SQL command, SQL*Plus overwrites the previous command in the buffer.

Creating Stored Procedures

Stored procedures are PL/SQL functions, packages, or procedures. To create stored procedures, you use SQL CREATE commands. The following SQL CREATE commands are used to create stored procedures:

• CREATE FUNCTION

• CREATE LIBRARY

• CREATE PACKAGE

• CREATE PACKAGE BODY

• CREATE PROCEDURE

• CREATE TRIGGER

• CREATE TYPE

Entering any of these commands places you in PL/SQL mode, where you can enter your PL/SQL subprogram. For more information, see the section "Running PL/SQL Blocks" in this chapter). When you are done typing your PL/SQL subprogram, enter a period (.) on a line by itself to terminate PL/SQL mode. To run the SQL command and create the stored procedure, you must enter RUN or slash (/). A semicolon (;) will not execute these CREATE commands.

When you use CREATE to create a stored procedure, a message appears if there are compilation errors. To view these errors, you use SHOW ERRORS. For example:

SHOW ERRORS PROCEDURE ASSIGNVL

For more information about the SHOW command, see Chapter 8, "Command Reference".

To execute a PL/SQL statement that references a stored procedure, you can use the EXECUTE command. EXECUTE runs the PL/SQL statement that you enter immediately after the command. For example:

EXECUTE :ID := EMPLOYEE_MANAGEMENT.GET_ID('BLAKE')

For more information about the EXECUTE command, see Chapter 8, "Command Reference".

Executing the Current SQL Command or PL/SQL Block from the Command Prompt

You can run (or re-run) the current SQL command or PL/SQL block by entering the RUN command or the slash (/) command at the command prompt. The RUN command lists the SQL command or PL/SQL block in the buffer before executing the command or block; the slash (/) command simply runs the SQL command or PL/SQL block.

Running PL/SQL Blocks

You can also use PL/SQL subprograms (called blocks) to manipulate data in the database. See your PL/SQL User's Guide and Reference for information on individual PL/SQL statements.

To enter a PL/SQL subprogram in SQL*Plus, you need to be in PL/SQL mode. You are placed in PL/SQL mode when

• You type DECLARE or BEGIN at the SQL*Plus command prompt. After you enter PL/SQL mode in this way, type the remainder of your PL/SQL subprogram.

• You type a SQL command (such as CREATE FUNCTION) that creates a stored procedure. After you enter PL/SQL mode in this way, type the stored procedure you want to create.

SQL*Plus treats PL/SQL subprograms in the same manner as SQL commands, except that a semicolon (;) or a blank line does not terminate and execute a block. Terminate PL/SQL subprograms by entering a period (.) by itself on a new line. You can also terminate and execute a PL/SQL subprogram by entering a slash (/) by itself on a new line.

SQL*Plus stores the subprograms you enter at the SQL*Plus command prompt in the SQL buffer. Execute the current subprogram by issuing a RUN or slash (/) command. Likewise, to execute a SQL CREATE command that creates a stored procedure, you must also enter RUN or slash (/). A semicolon (;) will not execute these SQL commands as it does other SQL commands.

SQL*Plus sends the complete PL/SQL subprogram to Oracle for processing (as it does SQL commands). See your PL/SQL User's Guide and Reference for more information.

You might enter and execute a PL/SQL subprogram as follows:

DECLARE
    x   NUMBER := 100;
    BEGIN
        FOR i IN 1..10 LOOP
           IF MOD (i, 2) = 0 THEN    --i is even
             INSERT INTO temp VALUES (i, x, 'i is even');
           ELSE
             INSERT INTO temp VALUES (i, x, 'i is odd');
           END IF;
           x := x + 100;
       END LOOP;
   END;
   .
/

When you run a subprogram, the SQL commands within the subprogram may behave somewhat differently than they would outside the subprogram. See your PL/SQL User's Guide and Reference for detailed information on the PL/SQL language.

Running SQL*Plus Commands

You can use SQL*Plus commands to manipulate SQL commands and PL/SQL blocks and to format and print query results. SQL*Plus treats SQL*Plus commands differently than SQL commands or PL/SQL blocks. For information on individual SQL*Plus commands, see Chapter 8, "Command Reference".

To speed up command entry, you can abbreviate many SQL*Plus commands to one or a few letters. Abbreviations for some SQL*Plus commands are described along with the commands in Chapter 3, Chapter 4, and Chapter 6. For abbreviations of all SQL*Plus commands, see Chapter 8, "Command Reference".

Example 2-4 Entering a SQL*Plus Command

This example shows how you might enter a SQL*Plus command to change the format used to display the column SALARY of the sample view, EMP_DETAILS_VIEW.

1. On the command line, enter this SQL*Plus command:

   
   COLUMN SALARY FORMAT $99,999 HEADING 'MONTHLY SALARY'

   If you make a mistake, use Backspace to erase it and re-enter. When you have entered the line, press Return. SQL*Plus notes the new format and displays the SQL*Plus command prompt again, ready for a new command.

2. Enter the RUN command to re-run the most recent query (from Example 2-3):

   
   RUN
   
   
   
   EMPLOYEE_ID LAST_NAME                 JOB_ID     MONTHLY SALARY
   ----------- ------------------------- ---------- --------------
           100 King                      AD_PRES           $24,000
           101 Kochhar                   AD_VP             $17,000
           102 De Haan                   AD_VP             $17,000
           145 Russell                   SA_MAN            $14,000
           146 Partners                  SA_MAN            $13,500
           201 Hartstein                 MK_MAN            $13,000
   
   6 rows selected.

The COLUMN command formatted the column SALARY with a dollar sign ($) and a comma (,) and gave it a new heading. The RUN command then re-ran the query of Example 2-3, which was stored in the buffer. SQL*Plus does not store SQL*Plus commands in the SQL buffer.

Understanding SQL*Plus Command Syntax

SQL*Plus commands have a different syntax from SQL commands or PL/SQL blocks.

Continuing a Long SQL*Plus Command on Additional Lines

You can continue a long SQL*Plus command by typing a hyphen at the end of the line and pressing Return. If you wish, you can type a space before typing the hyphen. SQL*Plus displays a right angle-bracket (>) as a prompt for each additional line.

For example:

COLUMN SALARY FORMAT $99,999 -
HEADING 'MONTHLY SALARY'

Since SQL*Plus identifies the hyphen as a continuation character, entering a hyphen within a SQL statement is ignored by SQL*Plus. SQL*Plus does not identify the statement as a SQL statement until after the input processing has joined the lines together and removed the hyphen. For example, entering the following:

SELECT 200 -
100 FROM DUAL;

returns the error:

SELECT 200 100 FROM DUAL
           *
ERROR at line 1:
ORA-00923: FROM keyword not found where expected

To ensure that the statement is interpreted correctly, reposition the hyphen from the end of the first line to the beginning of the second line.

Ending a SQL*Plus Command

You do not need to end a SQL*Plus command with a semicolon. When you finish entering the command, you can just press Return. If you wish, however, you can enter a semicolon at the end of a SQL*Plus command.

Variables that Affect Running Commands

The SQL*Plus command SET controls many variables--called system variables--the settings of which affect the way SQL*Plus runs your commands. System variables control a variety of conditions within SQL*Plus, including default column widths for your output, whether SQL*Plus displays the number of records selected by a command, and your page size. System variables are also called SET command variables.

The examples in this guide are based on running SQL*Plus with the system variables at their default settings. Depending on the settings of your system variables, your output may appear slightly different than the output shown in the examples. (Your settings might differ from the default settings if you have a SQL*Plus LOGIN file on your computer.)

For more information on system variables and their default settings, see the SET command in Chapter 8. For details on the SQL*Plus LOGIN file, refer to the section "Setting Up Your SQL*Plus Environment" under "Saving Commands for Later Use" in Chapter 3 and to the SQLPLUS command in Chapter 7.

To list the current setting of a SET command variable, enter SHOW followed by the variable name at the command prompt. See the SHOW command in Chapter 8 for information on other items you can list with SHOW.

Saving Changes to the Database Automatically

Through the SQL DML commands UPDATE, INSERT, and DELETE--which can be used independently or within a PL/SQL block--specify changes you wish to make to the information stored in the database. These changes are not made permanent until you enter a SQL COMMIT command or a SQL DCL or DDL command (such as CREATE TABLE), or use the autocommit feature. The SQL*Plus autocommit feature causes pending changes to be committed after a specified number of successful SQL DML transactions. (A SQL DML transaction is either an UPDATE, INSERT, or DELETE command, or a PL/SQL block.)

You control the autocommit feature with the SQL*Plus SET command's AUTOCOMMIT variable.

Example 2-5 Turning Autocommit On

To turn the autocommit feature on, enter

SET AUTOCOMMIT ON

Alternatively, you can enter the following to turn the autocommit feature on:

SET AUTOCOMMIT IMMEDIATE

Until you change the setting of AUTOCOMMIT, SQL*Plus automatically commits changes from each SQL DML command that specifies changes to the database. After each autocommit, SQL*Plus displays the following message:

COMMIT COMPLETE

When the autocommit feature is turned on, you cannot roll back changes to the database.

To commit changes to the database after a number of SQL DML commands, for example, 10, enter

SET AUTOCOMMIT 10

SQL*Plus counts SQL DML commands as they are executed and commits the changes after each 10th SQL DML command.

Note: For this feature, a PL/SQL block is regarded as one transaction, regardless of the actual number of SQL commands contained within it.

To turn the autocommit feature off again, enter the following command:

SET AUTOCOMMIT OFF

To confirm that AUTOCOMMIT is now set to OFF, enter the following SHOW command:

SHOW AUTOCOMMIT



AUTOCOMMIT OFF

For more information, see the AUTOCOMMIT variable of the SET command in Chapter 8.

Stopping a Command while it is Running

Suppose you have displayed the first page of a 50 page report and decide you do not need to see the rest of it. Press Cancel, the system's interrupt character, which is usually CTRL+C. SQL*Plus stops the display and returns to the command prompt.

Note: Pressing Cancel does not stop the printing of a file that you have sent to a printer with the OUT clause of the SQL*Plus SPOOL command. (You will learn about printing query results in Chapter 4.) You can stop the printing of a file through your operating system. For more information, see your operating system's installation and user's guide.

Collecting Timing Statistics on Commands You Run

Use the SQL*Plus TIMING command to collect and display data on the amount of computer resources used to run one or more commands or blocks. TIMING collects data for an elapsed period of time, saving the data on commands run during the period in a timer.

See TIMING in Chapter 8 and the Oracle installation and user's guide provided for your operating system for more information. See also "Tracing Statements" in Chapter 3 for information about using AUTOTRACE to collect statistics.

To delete all timers, enter CLEAR TIMING at the command prompt.

Running Host Operating System Commands

You can execute a host operating system command from the SQL*Plus command prompt. This is useful when you want to perform a task such as listing existing host operating system files.

To run a host operating system command, enter the SQL*Plus command HOST followed by the host operating system command. For example, this SQL*Plus command runs a host command, DIRECTORY *.SQL:

HOST DIRECTORY *.SQL

When the host command finishes running, the SQL*Plus command prompt appears again.

Note: Operating system commands entered from a SQL*Plus session using the HOST command do not effect the current SQL*Plus session. For example, setting an operating system environment variable does not effect the current SQL*Plus session, but may effect SQL*Plus sessions started subsequently. You can suppress access to the HOST command. For more information about suppressing the HOST command see Appendix E, "Security".